CDISC Pilot replication w/ gtsummary

CDISCPILOT01

Brief Description of Case Study

This study was a prospective, randomized, multi-center, double-blind, placebo-controlled, parallel-group study. The objectives of the study were to evaluate the efficacy and safety of transdermal xanomeline, 50 cm² and 75 cm² , and placebo in subjects with mild to moderate Alzheimer’s disease.

CDISC Pilot Replication

Submissions Pilot to FDA

Report

Tables

Tables 14-1

Table 14-1.01

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adsl_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt')

# UPDATE DATA
adsl <- adsl_orig %>% 
   mutate(TRT01P = factor(TRT01P, 
                          levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                          labels = c('Placebo', 'Low Dose','High Dose')))

# TABLE
tbl_summary(
   data = adsl,
   by = TRT01P,
   include = c(ITTFL, SAFFL, EFFFL, COMP24FL, DCDECOD),
   value = list(everything() ~ 'Y',
                DCDECOD ~ 'COMPLETED'),
   label = list( ITTFL ~ 'Intent-To-Treat (ITT)', 
                 SAFFL ~ 'Safety', 
                 EFFFL ~ 'Efficacy', 
                 COMP24FL ~ 'Complete Week 24',
                 DCDECOD ~ 'Complete Study')) %>% 
   modify_header(
      label = '',
      all_stat_cols(stat_0 = FALSE) ~ '**{level}<br>(N={n})**') %>% 
   add_overall(
      last = TRUE,
      col_label =  '**Total<br>(N={N})**') %>% 
   modify_spanning_header(
      c(stat_2, stat_3) ~ '**Xanomeline**') %>% 
   modify_footnote(
      all_stat_cols() ~ 'N in column headers represents number of subjects entered in study (i.e., signed informed consent).') %>%
   modify_table_styling(
      columns = label,
      rows = label == "Intent-To-Treat (ITT)",
      footnote = "The ITT population includes all subjects randomized.") %>% 
   modify_table_styling(
      columns = label,
      rows = label == "Safety",
      footnote = "The Safety population includes all randomized subjects known to have taken at least one dose of randomized study drug.") %>% 
   modify_table_styling(
      columns = label,
      rows = label == "Efficacy",
      footnote = "The Efficacy population includes all subjects in the safety population who also have at least one post-baseline ADAS-Cog and CIBIC+ assessment.") %>% 
   modify_caption(
      '**Table 14-1.01 <br> Summary of Populations**')

Table 14-1.01
Summary of Populations

	Placebo (N=86)1	Xanomeline		Total (N=254)1
		Low Dose (N=84)1	High Dose (N=84)1
Intent-To-Treat (ITT)2	86 (100%)	84 (100%)	84 (100%)	254 (100%)
Safety3	86 (100%)	84 (100%)	84 (100%)	254 (100%)
Efficacy4	79 (92%)	81 (96%)	74 (88%)	234 (92%)
Complete Week 24	60 (70%)	28 (33%)	30 (36%)	118 (46%)
Complete Study	58 (67%)	25 (30%)	27 (32%)	110 (43%)
1 N in column headers represents number of subjects entered in study (i.e., signed informed consent).
2 The ITT population includes all subjects randomized.
3 The Safety population includes all randomized subjects known to have taken at least one dose of randomized study drug.
4 The Efficacy population includes all subjects in the safety population who also have at least one post-baseline ADAS-Cog and CIBIC+ assessment.

Table 14-1.02

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adsl_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt')

# UPDATE DATA
adsl <- adsl_orig %>% 
   mutate(TRT01P = factor(TRT01P, 
                          levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                          labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(DCDECOD = case_when(
      COMP24FL == 'Y' & COMP24FL != 'Complete' ~ 'Completed',
      .default = DCDECOD)) %>% 
   mutate(DCDECOD = factor(DCDECOD) %>% 
             str_to_title() %>% 
             fct_recode(`Subject decided to withdraw` = 'Withdrawal By Subject') %>% 
             fct_relevel('Adverse Event',
                         'Death',
                         'Lack Of Efficacy',
                         'Lost To Follow-Up',
                         'Subject decided to withdraw')) %>% 
   mutate(COMP24FL = factor(COMP24FL, 
                            levels = c('Y','N'), 
                            labels = c('Completed Week 24','Early Termination (prior to Week 24)')) )

# TEST
p_loefl <- adsl %>%
   mutate(loefl = ifelse(DCREASCD %in% 'Lack of Efficacy', 1, 0)) %>%
   with(.,table(TRT01P, loefl)) %>% 
   fisher.test() %>% 
   broom::tidy() %>% 
   pull(p.value)

# TABLE
tbl_summary(
   data = adsl,
   by = TRT01P,
   include = c(COMP24FL, DCDECOD),
   label = list(COMP24FL = 'Completion Status',
                DCDECOD = 'Reason for Early Termination (prior to Week 24)'),
   missing = 'always',
   missing_text = 'Missing') %>% 
   modify_header(
      label = '',
      all_stat_cols(stat_0 = FALSE) ~ '**{level}<br>(N={n})**') %>% 
   add_overall(
      last = TRUE,
      col_label =  '**Total<br>(N={N})**') %>% 
   remove_row_type(
      variables = DCDECOD,
      type = 'level',
      level_value = 'Completed') %>% 
   add_p(
      test = list(COMP24FL ~ 'fisher.test',
                  DCDECOD  ~ 'chisq.test'),
      pvalue_fun = function(x) style_pvalue(x, digits = 3)) %>%
   modify_footnote(
      all_stat_cols() ~ NA,
      p.value ~ "Fisher's exact test.") %>% 
   modify_spanning_header(
      c(stat_2, stat_3) ~ '**Xanomeline**') %>% 
   modify_table_styling(
      columns = label,
      rows = label == 'Lack Of Efficacy',
      footnote = 'Based on either patient/caregiver perception or physician perception.') %>% 
   modify_table_body(
      ~.x %>% 
         mutate(label  = case_match(
            label,
            'Physician Decision'          ~ 'Physician decided to withdraw subject',
            'Study Terminated By Sponsor' ~ 'Sponsor decision',
            .default = label)) %>% 
         mutate(p.value  = ifelse(variable == 'DCDECOD' & label == 'Lack Of Efficacy', p_loefl, p.value )) %>% 
         mutate(across(all_stat_cols(), ~gsub('^0.*', '0 (0%)', .))) ) %>%  
   modify_caption(
      '**Table 14-1.02 <br> Summary of End of Study Data**') 

Table 14-1.02
Summary of End of Study Data

	Placebo (N=86)	Xanomeline		Total (N=254)	p-value1
		Low Dose (N=84)	High Dose (N=84)
Completion Status					<0.001
Completed Week 24	60 (70%)	28 (33%)	30 (36%)	118 (46%)
Early Termination (prior to Week 24)	26 (30%)	56 (67%)	54 (64%)	136 (54%)
Missing	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Reason for Early Termination (prior to Week 24)					<0.001
Adverse Event	8 (9.3%)	44 (52%)	39 (46%)	91 (36%)
Death	1 (1.2%)	1 (1.2%)	0 (0%)	2 (0.8%)
Lack Of Efficacy2	3 (3.5%)	0 (0%)	1 (1.2%)	4 (1.6%)	0.328
Lost To Follow-Up	1 (1.2%)	0 (0%)	0 (0%)	1 (0.4%)
Subject decided to withdraw	9 (10%)	8 (9.5%)	8 (9.5%)	25 (9.8%)
Physician decided to withdraw subject	1 (1.2%)	0 (0%)	2 (2.4%)	3 (1.2%)
Protocol Violation	2 (2.3%)	1 (1.2%)	3 (3.6%)	6 (2.4%)
Sponsor decision	1 (1.2%)	2 (2.4%)	1 (1.2%)	4 (1.6%)
Missing	0 (0%)	0 (0%)	0 (0%)	0 (0%)
1 Fisher’s exact test.
2 Based on either patient/caregiver perception or physician perception.

Table 14-1.03

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adsl_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt')

# LONG
adsl_l <- adsl_orig %>% 
   select(SUBJID, SITEGR1, SITEID, TRT01P, ITTFL, EFFFL, COMP24FL ) %>% 
   pivot_longer(cols = -c(SUBJID, SITEGR1, SITEID, TRT01P),
                names_to = 'DESC',
                values_to = 'YN') %>%
   mutate(YN = ifelse(YN == 'Y', 1, 0)) %>% 
   filter(YN == 1) %>% 
   mutate(DESC = factor(DESC, 
                        levels = c('ITTFL','EFFFL','COMP24FL'),
                        labels = c('ITT','Eff','Com'))) %>% 
   mutate(TRT01P = factor(TRT01P, 
                          levels = c('Placebo', 'Xanomeline Low Dose', 'Xanomeline High Dose'),
                          labels = c('Placebo', 'Xanomeline Low Dose', 'Xanomeline High Dose'))) %>% 
   mutate(SITE = str_glue('{SITEGR1} \U2014 {SITEID}'))

# TABLE
tbl_merge( 
   tbls = list(
      adsl_l %>% tbl_cross(row = SITE,
                           col = DESC,
                           margin = 'row'),
      
      adsl_l %>% filter(TRT01P == 'Placebo') %>% 
         tbl_cross(row = SITE,
                   col = DESC,
                   margin = 'row'),
      
      adsl_l %>% filter(TRT01P == 'Xanomeline Low Dose') %>% 
         tbl_cross(row = SITE,
                   col = DESC,
                   margin = 'row'),
      
      adsl_l %>% filter(TRT01P == 'Xanomeline High Dose') %>% 
         tbl_cross(row = SITE,
                   col = DESC,
                   margin = 'row')),
   tab_spanner = c('**<br>&nbsp;Total<br>(N=254)**', '**Xanomeline<br>Low Dose<br>(N=84)**', '**Xanomeline<br>High Dose<br>(N=84)**', '**<br>&nbsp;Placebo<br>(N=86)**')) %>% 
   modify_table_body(
      ~.x %>% 
         filter(!(variable == 'SITE' & row_type == 'label')) %>% 
         mutate(across(all_stat_cols(), ~ifelse(.x %in% c(0, NA), '0', .x) ) ) %>% 
         relocate(c(stat_1_1, stat_2_1, stat_3_1), .after = last_col())) %>% 
   modify_header(
      label           ~ '**Pooled \U2014 Site ID**',
      all_stat_cols() ~ '**{level}**') %>% 
   modify_footnote(
      starts_with('stat_1_') ~ 'ITT: Number of subjects in the ITT population',
      starts_with('stat_2_') ~ 'Eff: Number of subjects in the Efficacy population',
      starts_with('stat_3_') ~ 'Com: Number of subjects completing Week 24.') %>%
   modify_column_alignment(
      columns = all_stat_cols() , align = 'right') %>% 
   modify_caption(
      '**Table 14-1.03<br>Summary of Number of Subjects By Site**') 

Table 14-1.03
Summary of Number of Subjects By Site

Pooled — Site ID	Xanomeline Low Dose (N=84)			Xanomeline High Dose (N=84)			Placebo (N=86)			Total (N=254)
	ITT1	Eff2	Com3	ITT1	Eff2	Com3	ITT1	Eff2	Com3	ITT1	Eff2	Com3
701 — 701	14	14	11	13	13	5	14	14	7	41	41	23
703 — 703	6	5	4	6	5	1	6	5	2	18	15	7
704 — 704	9	9	5	8	7	3	8	8	0	25	24	8
705 — 705	5	3	2	5	5	3	6	4	1	16	12	6
708 — 708	9	9	7	8	8	2	8	5	2	25	22	11
709 — 709	7	7	5	7	6	2	7	7	3	21	20	10
710 — 710	11	8	6	10	10	2	10	8	5	31	26	13
713 — 713	3	3	3	3	3	2	3	2	2	9	8	7
716 — 716	8	8	7	8	8	3	8	8	3	24	24	13
718 — 718	4	4	3	5	5	1	4	4	1	13	13	5
900 — 702	0	0	0	1	1	0	0	0	0	1	1	0
900 — 706	1	1	1	1	1	0	1	1	0	3	3	1
900 — 707	1	1	1	1	1	0	0	0	0	2	2	1
900 — 711	1	1	1	1	1	0	2	1	0	4	3	1
900 — 714	2	2	2	2	2	1	2	2	1	6	6	4
900 — 715	3	2	2	3	3	1	2	2	0	8	7	3
900 — 717	2	2	0	2	2	2	3	3	3	7	7	5
Total	86	79	60	84	81	28	84	74	30	254	234	118
1 ITT: Number of subjects in the ITT population
2 Eff: Number of subjects in the Efficacy population
3 Com: Number of subjects completing Week 24.

Tables 14-2

Table 14-2.01

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adsl_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt')

# DATA
adsl <- adsl_orig %>% 
   mutate(TRT01P = factor(TRT01P, 
                          levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                          labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(SEX = factor(SEX, labels = c('Female','Male')) %>% fct_rev(),
          AGEGR1 = factor(AGEGR1, 
                          level = c('<65','65-80','>80'),
                          labels = c('<65 yrs','65-80 yrs','>80 yrs')),
          RACE = factor(RACE) %>% str_to_title(),
          BMIBLGR1 = factor(BMIBLGR1, levels = c('<25','25-<30','>=30') ),
          DURDSGR1 = factor(DURDSGR1, labels = c('<12 months', '>=12 months')))

# TABLE
tbl_summary(
   data = adsl,
   by = TRT01P,
   include = c(AGE, AGEGR1, SEX, RACE, MMSETOT, DURDIS, DURDSGR1, EDUCLVL, WEIGHTBL, HEIGHTBL, BMIBL, BMIBLGR1),
   type = list(all_continuous() ~ 'continuous2'),
   statistic = list( all_continuous() ~ c("{N_nonmiss}", 
                                          "{mean}",
                                          "{sd}",
                                          "{median}",
                                          "{min}",
                                          "{max}"),
                     all_categorical() ~ c("{n} ({p}%)")),
   label = list( AGE    ~ 'Age (y)',
                 AGEGR1 ~ '',
                 SEX    ~ 'Sex',
                 RACE   ~ 'Race (Origin)',
                 MMSETOT~ 'MMSE',
                 DURDIS ~ 'Duration of disease',
                 DURDSGR1 ~ '',
                 BMIBL  ~ 'Baseline BMI',
                 BMIBLGR1 ~ ''),
   digits = list(all_continuous() ~ c(0, 1, 2, 1, 1, 1),
                 all_categorical() ~ c(0, 0) ),
   sort = list(RACE ~ 'frequency'),
   missing = 'no',
   missing_text = 'Missing') %>% 
   modify_header(
      label = '',
      all_stat_cols(stat_0 = FALSE) ~ '**{level}<br>(N={n})**'
   ) %>% 
   add_overall(
      last = TRUE,
      col_label =  '**Total<br>(N={N})**') %>%
   add_p(
      test = list(all_continuous() ~ 'aov', 
                  all_categorical() ~ 'chisq.test'),
      pvalue_fun = function(x) style_pvalue(x, digits = 3)) %>% 
   modify_spanning_header(
      c(stat_2, stat_3) ~ '**Xanomeline**') %>% 
   modify_footnote(
      all_stat_cols() ~ NA,
      p.value ~ 'P-values are results of ANOVA treatment group comparison for continuous variable and Pearsons chisquare test for categorical variables.') %>% 
   modify_table_styling(
      columns = label,
      rows =  variable == 'DURDIS' & row_type  == 'label',
      footnote = "Duration of disease is computed as months between date of enrollment and date of onset of the first definite symptoms of Alzheimers disease.") %>% 
   modify_column_alignment(
      all_stat_cols(), align = 'right') %>% 
   modify_table_body(
      ~.x %>% 
         mutate(label = case_match(
            label,
            "Minimum" ~ "Min",
            "Maximum" ~ "Max",
            "N"       ~ "n",
            .default = label)) ) %>% 
   modify_caption(
      '**Table 14-2.01<br>Summary of Demographic and Baseline Characteristics**')

Table 14-2.01
Summary of Demographic and Baseline Characteristics

	Placebo (N=86)	Xanomeline		Total (N=254)	p-value1
		Low Dose (N=84)	High Dose (N=84)
Age (y)					0.593
n	86	84	84	254
Mean	75.2	75.7	74.4	75.1
SD	8.59	8.29	7.89	8.25
Median	76.0	77.5	76.0	77.0
Min	52.0	51.0	56.0	51.0
Max	89.0	88.0	88.0	89.0
					0.144
<65 yrs	14 (16%)	8 (10%)	11 (13%)	33 (13%)
65-80 yrs	42 (49%)	47 (56%)	55 (65%)	144 (57%)
>80 yrs	30 (35%)	29 (35%)	18 (21%)	77 (30%)
Sex					0.141
Male	33 (38%)	34 (40%)	44 (52%)	111 (44%)
Female	53 (62%)	50 (60%)	40 (48%)	143 (56%)
Race (Origin)					0.604
White	78 (91%)	78 (93%)	74 (88%)	230 (91%)
Black Or African American	8 (9%)	6 (7%)	9 (11%)	23 (9%)
American Indian Or Alaska Native	0 (0%)	0 (0%)	1 (1%)	1 (0%)
MMSE					0.595
n	86	84	84	254
Mean	18.0	17.9	18.5	18.1
SD	4.27	4.22	4.16	4.21
Median	19.5	18.0	20.0	19.0
Min	10.0	10.0	10.0	10.0
Max	23.0	24.0	24.0	24.0
Duration of disease2					0.153
n	86	84	84	254
Mean	42.7	48.7	40.5	43.9
SD	30.24	29.58	24.69	28.40
Median	35.3	40.3	36.0	36.3
Min	7.2	7.8	2.2	2.2
Max	183.1	130.8	135.0	183.1
					0.789
<12 months	5 (6%)	3 (4%)	4 (5%)	12 (5%)
>=12 months	81 (94%)	81 (96%)	80 (95%)	242 (95%)
Years of Education					0.388
n	86	84	84	254
Mean	12.6	13.2	12.5	12.8
SD	2.95	4.15	2.92	3.38
Median	12.0	12.0	12.0	12.0
Min	6.0	3.0	6.0	3.0
Max	21.0	24.0	20.0	24.0
Baseline Weight (kg)					0.003
n	86	83	84	253
Mean	62.8	67.3	70.0	66.6
SD	12.77	14.12	14.65	14.13
Median	60.6	64.9	69.2	66.7
Min	34.0	45.4	41.7	34.0
Max	86.2	106.1	108.0	108.0
Baseline Height (cm)					0.126
n	86	84	84	254
Mean	162.6	163.4	165.8	163.9
SD	11.52	10.42	10.13	10.76
Median	162.6	162.6	165.1	162.9
Min	137.2	135.9	146.1	135.9
Max	185.4	195.6	190.5	195.6
Baseline BMI					0.013
n	86	83	84	253
Mean	23.6	25.1	25.3	24.7
SD	3.67	4.27	4.16	4.09
Median	23.4	24.3	24.8	24.2
Min	15.1	17.7	13.7	13.7
Max	33.3	40.1	34.5	40.1
					0.233
<25	59 (69%)	47 (56%)	44 (52%)	150 (59%)
25-<30	21 (24%)	27 (32%)	28 (33%)	76 (30%)
>=30	6 (7%)	10 (12%)	12 (14%)	28 (11%)
1 P-values are results of ANOVA treatment group comparison for continuous variable and Pearsons chisquare test for categorical variables.
2 Duration of disease is computed as months between date of enrollment and date of onset of the first definite symptoms of Alzheimers disease.

Tables 14-3

Table 14-3.01

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# SUBSET
adqsadas <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y') %>% 
   filter(AVISITN %in% c(0, 24)) %>% 
   filter(!is.na(CHG)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) 

# REGRESSION
data       <- adqsadas
variable   <- 'CHG'
by         <- 'TRTP'
adj.vars   <- c('BASE','SITEGR1')

lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   
   b
}
#lm_pairwise(adqsadas, 'CHG','TRTP', c('BASE','SITEGR1'))

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   # l <- lm(as.formula(f), data = data %>% mutate(!!sym(by) := as.numeric(!!sym(by))) )
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}
#lm_trend(adqsadas, 'CHG','TRTP', c('BASE','SITEGR1'))

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsadas,
   by = TRTP,
   include = c(BASE, AVAL, CHG),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Week 24',
                CHG ~ 'Change from Baseline')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = CHG ~ lm_trend ) %>% 
   add_stat(fns = CHG ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.01<br>Primary Endpoint Analysis: ADAS Cog (11) - Change from Baseline to Week 24 - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.01
Primary Endpoint Analysis: ADAS Cog (11) - Change from Baseline to Week 24 - LOCF

	Placebo (N=79)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=74)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	79 24.1 (12.19) 21.0 (5;61)	81 24.4 (12.92) 21.0 (5;57)	74 21.3 (11.74) 18.0 (3;57)
Week 24, n Mean (SD) Median (Range)	79 26.7 (13.79) 24.0 (5;62)	81 26.4 (13.18) 25.0 (6;62)	74 22.8 (12.48) 20.0 (3;62)
Change from Baseline, n Mean (SD) Median (Range)	79 2.5 (5.80) 2.0 (-11;16)	81 2.0 (5.55) 2.0 (-11;17)	74 1.5 (4.26) 1.0 (-7;13)	0.245	0.569 -0.5 (0.82) (-2.1;1.1)	0.233 -1.0 (0.84) (-2.7;0.7)	0.520 -0.5 (0.84) (-2.2;1.1)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons

Warning

I updated the table layout because gtsummary does not accommodate stacking rows with different formats well.

Caution

Here, we need to apply gt::fmt_markdown to format the line breaks within the table.

Table 14-3.02

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqscibc_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqscibc.xpt')

# SUBSET
adqscibc <- adqscibc_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'CIBICVAL' & ANL01FL == 'Y') %>% 
   filter(AVISITN == 24) %>% 
   filter(!is.na(AVAL)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) 

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   b
   }

lm_trend  <- function(data, variable, by, adj.vars = c('SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()

   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqscibc,
   by = TRTP,
   include = c(AVAL),
   type = list(AVAL ~ 'continuous'),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(AVAL ~ 'Week 24')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = AVAL ~ lm_trend ) %>% 
   add_stat(fns = AVAL ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ ' Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons.',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.02<br>Primary Endpoint Analysis: CIBIC+ - Summary at Week 24 - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.02
Primary Endpoint Analysis: CIBIC+ - Summary at Week 24 - LOCF

	Placebo (N=79)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=74)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Week 24, n Mean (SD) Median (Range)	79 4.3 (0.77) 4.0 (2;6)	81 4.2 (0.79) 4.0 (2;6)	74 4.3 (0.81) 4.0 (3;6)	0.960	0.489 -0.1 (0.13) (-0.3;0.2)	0.799 0.0 (0.13) (-0.2;0.3)	0.349 0.1 (0.13) (-0.1;0.4)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons.

Table 14-3.03

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# SUBSET
adqsadas <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y') %>% 
   filter(AVISITN %in% c(0, 8)) %>% 
   filter(!is.na(CHG)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose')))

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   
   b
}

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   # l <- lm(as.formula(f), data = data %>% mutate(!!sym(by) := as.numeric(!!sym(by))) )
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsadas,
   by = TRTP,
   include = c(BASE, AVAL, CHG),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Week 8',
                CHG ~ 'Change from Baseline')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = CHG ~ lm_trend ) %>% 
   add_stat(fns = CHG ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.03<br>ADAS Cog (11) - Change from Baseline to Week 8 - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.03
ADAS Cog (11) - Change from Baseline to Week 8 - LOCF

	Placebo (N=79)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=74)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	79 24.1 (12.19) 21.0 (5;61)	81 24.4 (12.92) 21.0 (5;57)	74 21.3 (11.74) 18.0 (3;57)
Week 8, n Mean (SD) Median (Range)	79 25.0 (13.10) 22.0 (5;62)	81 26.2 (12.98) 25.0 (5;62)	74 22.3 (12.41) 19.0 (2;62)
Change from Baseline, n Mean (SD) Median (Range)	79 0.8 (4.81) 1.0 (-12;16)	81 1.8 (4.14) 2.0 (-12;14)	74 1.0 (3.62) 1.0 (-8;13)	0.497	0.099 1.1 (0.65) (-0.2;2.4)	0.751 0.2 (0.67) (-1.1;1.5)	0.195 -0.9 (0.66) (-2.2;0.4)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons

Table 14-3.04

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqscibc_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqscibc.xpt')

# SUBSET
adqscibc <- adqscibc_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'CIBICVAL' & ANL01FL == 'Y') %>% 
   filter(AVISITN == 8) %>% 
   filter(!is.na(AVAL)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) 

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   b
   }

lm_trend  <- function(data, variable, by, adj.vars = c('SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()

   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqscibc,
   by = TRTP,
   include = c(AVAL),
   type = list(AVAL ~ 'continuous'),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(AVAL ~ 'Week 8')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = AVAL ~ lm_trend ) %>% 
   add_stat(fns = AVAL ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ ' Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons.',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.04<br>CIBIC+ - Summary at Week 8 - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.04
CIBIC+ - Summary at Week 8 - LOCF

	Placebo (N=77)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=73)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Week 8, n Mean (SD) Median (Range)	77 3.9 (0.73) 4.0 (2;6)	81 4.0 (0.72) 4.0 (2;6)	73 4.1 (0.75) 4.0 (2;6)	0.167	0.754 0.0 (0.12) (-0.2;0.3)	0.128 0.2 (0.12) (-0.1;0.4)	0.218 0.1 (0.12) (-0.1;0.4)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons.

Table 14-3.05

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# SUBSET
adqsadas <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y') %>% 
   filter(AVISITN %in% c(0, 16)) %>% 
   filter(!is.na(CHG)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose')))

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   b
}

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsadas,
   by = TRTP,
   include = c(BASE, AVAL, CHG),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Week 16',
                CHG ~ 'Change from Baseline')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = CHG ~ lm_trend ) %>% 
   add_stat(fns = CHG ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.05<br>ADAS Cog (11) - Change from Baseline to Week 16 - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.05
ADAS Cog (11) - Change from Baseline to Week 16 - LOCF

	Placebo (N=79)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=74)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	79 24.1 (12.19) 21.0 (5;61)	81 24.4 (12.92) 21.0 (5;57)	74 21.3 (11.74) 18.0 (3;57)
Week 16, n Mean (SD) Median (Range)	79 26.1 (14.16) 23.0 (5;63)	81 26.0 (13.05) 25.0 (5;62)	74 22.5 (12.33) 20.0 (4;62)
Change from Baseline, n Mean (SD) Median (Range)	79 2.0 (5.89) 2.0 (-17;23)	81 1.6 (4.10) 2.0 (-9;14)	74 1.2 (4.33) 1.0 (-11;13)	0.412	0.724 -0.3 (0.77) (-1.8;1.2)	0.392 -0.7 (0.79) (-2.2;0.9)	0.606 -0.4 (0.78) (-1.9;1.1)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons

Table 14-3.06

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE)

# IMPORT DATA
adqscibc_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqscibc.xpt')

# SUBSET
adqscibc <- adqscibc_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'CIBICVAL' & ANL01FL == 'Y') %>% 
   filter(AVISITN == 16) %>% 
   filter(!is.na(AVAL)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) 

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   b
   }

lm_trend  <- function(data, variable, by, adj.vars = c('SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()

   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqscibc,
   by = TRTP,
   include = c(AVAL),
   type = list(AVAL ~ 'continuous'),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(AVAL ~ 'Week 16')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = AVAL ~ lm_trend ) %>% 
   add_stat(fns = AVAL ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ ' Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons.',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.06<br>CIBIC+ - Summary at Week 16 - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.06
CIBIC+ - Summary at Week 16 - LOCF

	Placebo (N=79)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=74)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Week 16, n Mean (SD) Median (Range)	79 4.2 (0.70) 4.0 (3;6)	81 4.0 (0.77) 4.0 (2;6)	74 4.0 (0.75) 4.0 (2;5)	0.214	0.219 -0.1 (0.12) (-0.4;0.1)	0.272 -0.1 (0.12) (-0.4;0.1)	0.916 0.0 (0.12) (-0.2;0.2)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons.

Table 14-3.07

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# SUBSET
adqsadas <- adqsadas_orig %>%
   filter(COMP24FL == "Y" & EFFFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y' & DTYPE != 'LOCF') %>% 
   filter(AVISITN %in% c(0, 24)) %>% 
   filter(!is.na(CHG)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose')))

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   
   b
}

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   # l <- lm(as.formula(f), data = data %>% mutate(!!sym(by) := as.numeric(!!sym(by))) )
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsadas,
   by = TRTP,
   include = c(BASE, AVAL, CHG),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Week 24',
                CHG ~ 'Change from Baseline')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = CHG ~ lm_trend ) %>% 
   add_stat(fns = CHG ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons<br>
                                          Note that only assessments falling within the assessment window are included in the summary for a visit.',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.07<br>ADAS Cog (11) - Change from Baseline to Week 24 - Completers at Wk 24-Observed Cases-Windowed**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.07
ADAS Cog (11) - Change from Baseline to Week 24 - Completers at Wk 24-Observed Cases-Windowed

	Placebo (N=59)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=27)	High Dose (N=30)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	59 23.2 (11.74) 21.0 (5;51)	27 24.0 (13.89) 20.0 (5;57)	30 20.5 (11.50) 18.0 (3;49)
Week 24, n Mean (SD) Median (Range)	59 25.3 (13.32) 23.0 (5;58)	27 24.1 (11.87) 22.0 (8;51)	30 21.8 (12.60) 18.5 (3;44)
Change from Baseline, n Mean (SD) Median (Range)	59 2.1 (5.89) 2.0 (-11;16)	27 0.1 (5.86) 1.0 (-11;12)	30 1.3 (4.51) 1.0 (-7;13)	0.234	0.105 -2.1 (1.26) (-4.6;0.4)	0.461 -0.9 (1.22) (-3.3;1.5)	0.430 1.2 (1.47) (-1.8;4.1)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons Note that only assessments falling within the assessment window are included in the summary for a visit.

Table 14-3.08

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# SUBSET
adqsadas <- adqsadas_orig %>%
   filter(EFFFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y' & SEX == "M") %>% 
   filter(AVISITN %in% c(0, 24)) %>% 
   filter(!is.na(CHG)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) 

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   
   b
}

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   # l <- lm(as.formula(f), data = data %>% mutate(!!sym(by) := as.numeric(!!sym(by))) )
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsadas,
   by = TRTP,
   include = c(BASE, AVAL, CHG),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Week 24',
                CHG ~ 'Change from Baseline')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = CHG ~ lm_trend ) %>% 
   add_stat(fns = CHG ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.08<br>ADAS Cog (11) - Change from Baseline to Week 24 in Male Subjects - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.08
ADAS Cog (11) - Change from Baseline to Week 24 in Male Subjects - LOCF

	Placebo (N=33)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=34)	High Dose (N=39)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	33 22.8 (13.45) 19.0 (5;61)	34 23.3 (14.03) 21.5 (7;57)	39 20.8 (11.11) 17.0 (3;51)
Week 24, n Mean (SD) Median (Range)	33 24.7 (13.89) 20.0 (5;62)	34 25.7 (14.72) 24.0 (6;57)	39 22.7 (12.32) 21.0 (3;51)
Change from Baseline, n Mean (SD) Median (Range)	33 1.9 (6.14) 1.0 (-11;16)	34 2.5 (5.61) 1.0 (-6;14)	39 1.8 (3.77) 1.0 (-7;13)	0.873	0.712 0.5 (1.30) (-2.1;3.1)	0.915 0.1 (1.25) (-2.4;2.6)	0.783 -0.3 (1.26) (-2.9;2.2)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons

Table 14-3.09

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# SUBSET
adqsadas <- adqsadas_orig %>%
   filter(EFFFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y' & SEX == "F") %>% 
   filter(AVISITN %in% c(0, 24)) %>% 
   filter(!is.na(CHG)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) 

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   
   b
}

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   # l <- lm(as.formula(f), data = data %>% mutate(!!sym(by) := as.numeric(!!sym(by))) )
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsadas,
   by = TRTP,
   include = c(BASE, AVAL, CHG),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Week 24',
                CHG ~ 'Change from Baseline')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = CHG ~ lm_trend ) %>% 
   add_stat(fns = CHG ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.09<br>ADAS Cog (11) - Change from Baseline to Week 24 in Female Subjects - LOCF**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.09
ADAS Cog (11) - Change from Baseline to Week 24 in Female Subjects - LOCF

	Placebo (N=46)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=47)	High Dose (N=35)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	46 25.1 (11.25) 23.5 (5;51)	47 25.2 (12.15) 21.0 (5;55)	35 21.8 (12.54) 18.0 (5;57)
Week 24, n Mean (SD) Median (Range)	46 28.1 (13.70) 24.0 (8;59)	47 26.9 (12.09) 25.0 (8;62)	35 22.9 (12.84) 19.0 (4;62)
Change from Baseline, n Mean (SD) Median (Range)	46 3.0 (5.57) 3.0 (-8;16)	47 1.7 (5.54) 2.0 (-11;17)	35 1.1 (4.77) 0.0 (-7;13)	0.094	0.160 -1.6 (1.10) (-3.7;0.6)	0.135 -1.8 (1.20) (-4.2;0.6)	0.843 -0.2 (1.21) (-2.6;2.2)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons

Table 14-3.10

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# DATA LOCF
adqsadas_locf <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y') %>% 
   filter(AVISITN %in% c(0, 8, 16, 24)) %>% 
   select(USUBJID, TRTP, AVISIT, AVISITN, AVAL, BASE, CHG)

# DATA WINDOWED
adqsadas_win <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & ITTFL=='Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y') %>% 
   filter(AVISITN %in% c(0, 8, 16, 24)) %>% 
   filter(AVISITN != 0 & DTYPE != 'LOCF') %>% 
   select(USUBJID, TRTP, AVISIT, AVISITN, AVAL, BASE, CHG)

# LONG
adqsadas_l <- 
   bind_rows(
      LOCF     = adqsadas_locf,
      Windowed = adqsadas_win,
      .id = 'SET' ) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose')) ) %>% 
   mutate(AVISIT_SET = case_when(
      AVISIT == 'Baseline' ~ AVISIT,
      .default = str_glue('{AVISIT} ({SET})'))) %>% 
   mutate(AVISIT_SET = factor(AVISIT_SET) %>% 
             fct_relevel('Baseline',
                         'Week 8 (Windowed)', 'Week 16 (Windowed)', 'Week 24 (Windowed)',
                         'Week 8 (LOCF)', 'Week 16 (LOCF)', 'Week 24 (LOCF)') ) %>% 
   pivot_longer(cols = c(AVAL, BASE, CHG),
                names_to = 'VAR',
                values_to = 'AVAL',
                values_drop_na = TRUE)

# NEST
adqsadas_n <- adqsadas_l %>% 
   droplevels() %>% 
   nest_by(SET, TRTP, AVISITN, AVISIT, AVISIT_SET, VAR)

# TABLE BY STAT WIDE
stats <- c('N'    = '{length}', 
           'Mean' = '{mean}',
           'Std'  = '{sd}',
           'Med.' = '{median}',
           'Min.' = '{min}',
           'Max.' = '{max}')

adqsadas_t <- adqsadas_n %>% 
   mutate(t = list(
      purrr::imap(
         stats,
         ~data %>%
            tbl_summary(include = 'AVAL', 
                        missing = 'no', 
                        digits = list(AVAL ~ 5),
                        label = list(AVAL ~ str_glue('{AVISIT_SET}')),
                        statistic = ~.x)  %>%
            modify_header(all_stat_cols() ~ stringr::str_glue('**{.y}**'))
         ) %>%
         tbl_merge(tab_spanner = FALSE) %>%
         modify_footnote(~NA))
   )

# TRANS to WIDE
adqsadas_w <- adqsadas_t %>% 
   select(-data) %>% 
   pivot_wider(names_from = VAR,
               values_from = t) %>%
   rowwise() %>% 
   mutate(t_m = ifelse( is.null(CHG) ,
                        list( tbl_merge(tbls = list(AVAL, BASE) ) ),
                        list( tbl_merge(tbls = list(AVAL, BASE, CHG)) ) ) )

# TABLES STACK
adqsadas_s <- adqsadas_w %>% 
   group_by(AVISIT_SET, TRTP) %>% 
   summarise(t_s = list( 
      tbl_stack(t_m) ) )

# PRINT TABLE
with(adqsadas_s,
     tbl_stack( tbls = t_s,
                group_header = TRTP)) %>% 
   modify_fmt_fun(
      update = list(
         starts_with('stat_0_1_') ~ function(x) style_number(as.numeric(x), digits = 0),
         starts_with('stat_0_2_') ~ function(x) style_number(as.numeric(x), digits = 1),
         starts_with('stat_0_3_') ~ function(x) style_number(as.numeric(x), digits = 2),
         starts_with('stat_0_4_') ~ function(x) style_number(as.numeric(x), digits = 1),
         starts_with('stat_0_5_') ~ function(x) style_number(as.numeric(x), digits = 0),
         starts_with('stat_0_6_') ~ function(x) style_number(as.numeric(x), digits = 0)),
      rows = !is.na(stat_0_2_1) ) %>% 
   modify_column_hide(
      columns = c(stat_0_1_2,
                  stat_0_4_2,
                  stat_0_5_2,
                  stat_0_6_2,
                  
                  stat_0_1_3) ) %>% 
   modify_header(
      label ~ '') %>% 
   modify_spanning_header(
      ends_with('_1') ~ '&nbsp;',
      ends_with('_2') ~ '**Baseline**',
      ends_with('_3') ~ '**Change from baseline**')%>% 
   modify_caption(
      '**Table 14-3.10<br>ADAS Cog (11) - Mean and Mean Change from Baseline over Time**') %>% 
   bstfun::bold_italicize_group_labels(bold = TRUE)

Table 14-3.10
ADAS Cog (11) - Mean and Mean Change from Baseline over Time

							Baseline		Change from baseline
	N	Mean	Std	Med.	Min.	Max.	Mean	Std	Mean	Std	Med.	Min.	Max.
Placebo
Baseline	79	24.1	12.19	21.0	5	61	24.1	12.19
Week 8 (Windowed)	79	25.0	13.10	22.0	5	62	24.1	12.19	0.8	4.81	1.0	-12	16
Week 16 (Windowed)	68	25.1	13.42	21.0	5	63	23.4	11.32	1.7	5.92	2.0	-17	23
Week 24 (Windowed)	65	25.7	13.90	23.0	5	59	23.6	12.13	2.1	5.99	2.0	-11	16
Week 8 (LOCF)	79	25.0	13.10	22.0	5	62	24.1	12.19	0.8	4.81	1.0	-12	16
Week 16 (LOCF)	79	26.1	14.16	23.0	5	63	24.1	12.19	2.0	5.89	2.0	-17	23
Week 24 (LOCF)	79	26.7	13.79	24.0	5	62	24.1	12.19	2.5	5.80	2.0	-11	16
Low Dose
Baseline	81	24.4	12.92	21.0	5	57	24.4	12.92
Week 8 (Windowed)	81	26.2	12.98	25.0	5	62	24.4	12.92	1.8	4.14	2.0	-12	14
Week 16 (Windowed)	42	26.2	12.23	25.0	8	53	25.0	12.52	1.2	4.33	1.0	-8	13
Week 24 (Windowed)	49	25.6	13.81	24.0	7	57	24.4	13.76	1.3	6.05	1.0	-11	17
Week 8 (LOCF)	81	26.2	12.98	25.0	5	62	24.4	12.92	1.8	4.14	2.0	-12	14
Week 16 (LOCF)	81	26.0	13.05	25.0	5	62	24.4	12.92	1.6	4.10	2.0	-9	14
Week 24 (LOCF)	81	26.4	13.18	25.0	6	62	24.4	12.92	2.0	5.55	2.0	-11	17
High Dose
Baseline	74	21.3	11.74	18.0	3	57	21.3	11.74
Week 8 (Windowed)	74	22.3	12.41	19.0	2	62	21.3	11.74	1.0	3.62	1.0	-8	13
Week 16 (Windowed)	40	21.9	12.39	19.5	4	49	21.1	11.79	0.8	4.92	1.0	-11	10
Week 24 (Windowed)	41	21.8	12.38	19.0	3	45	20.1	11.13	1.7	4.74	1.0	-7	13
Week 8 (LOCF)	74	22.3	12.41	19.0	2	62	21.3	11.74	1.0	3.62	1.0	-8	13
Week 16 (LOCF)	74	22.5	12.33	20.0	4	62	21.3	11.74	1.2	4.33	1.0	-11	13
Week 24 (LOCF)	74	22.8	12.48	20.0	3	62	21.3	11.74	1.5	4.26	1.0	-7	13

Table 14-3.11

Original

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtreg, gtsummary)
# MMRM MODEL
pacman::p_load(emmeans, mmrm)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# DATA LOCF
adqsadas <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y' & DTYPE != 'LOCF' & AVISITN > 0) %>% 
   select(USUBJID, TRTP, SITEGR1, AVISIT, AVISITN, AVAL, BASE, CHG) %>% 
   mutate(USUBJID = factor(USUBJID)) %>% 
   mutate(across(c(USUBJID, SITEGR1, AVISIT), factor)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(AVISIT = factor(AVISIT,
                          levels = c( 'Week 8', 'Week 16', 'Week 24') ) )

# EMMEANS & BIND DATA
m <- mmrm(
   CHG ~ SITEGR1 + BASE * AVISIT + TRTP * AVISIT + us(AVISIT | USUBJID),
   data = adqsadas,
   reml = TRUE,
   control = mmrm_control(
      method = 'Kenward-Roger') )

e <- emmeans(m, ~TRTP, calc = c(n = '.wgt.'), infer = TRUE)
c <- contrast(e, method = 'revpairwise') %>% summary(infer = TRUE, adjust = 'none')

d <- bind_rows(
   e %>% data.frame() %>% rename(estimate = emmean),
   c %>% data.frame() %>% rename(TRTP = contrast)) %>% 
   select(-n, -df, -t.ratio)

# WIDE DATA
d_l <- d %>% 
   pivot_wider(values_from = -TRTP,
               names_from = TRTP) %>% 
   janitor::clean_names()

# TABLE
tbl_listing(
   data = d_l ) %>% 
   modify_column_merge('{estimate_placebo} ({se_placebo})') %>% 
   modify_column_merge('{estimate_low_dose} ({se_low_dose})') %>% 
   modify_column_merge('{estimate_high_dose} ({se_high_dose})') %>%
   modify_column_merge('{p_value_low_dose_placebo}<br>{estimate_low_dose_placebo}   ({se_low_dose_placebo})<br>({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})') %>% 
   modify_column_merge('{p_value_high_dose_placebo}<br>{estimate_high_dose_placebo} ({se_high_dose_placebo})<br>({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})') %>%
   modify_column_merge('{p_value_high_dose_low_dose}<br>{estimate_high_dose_low_dose}   ({se_high_dose_low_dose})<br>({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})') %>%
   modify_column_hide(
      c('upper_cl_placebo','upper_cl_low_dose','upper_cl_high_dose',
        'lower_cl_placebo','lower_cl_low_dose','lower_cl_high_dose',
        'p_value_placebo','p_value_low_dose','p_value_high_dose')) %>% 
   modify_fmt_fun(
      list(
         starts_with( c('estimate_', 'upper_', 'lower_') ) ~ function(x) style_number(x, digits = 1),
         starts_with( 'se_') ~ function(x) style_number(x, digits = 2),
         starts_with( 'p_value_') ~ function(x) style_pvalue(x, digits = 3)),
      rows = !is.na(estimate_placebo)) %>% 
   modify_spanning_header(
      c(estimate_low_dose, estimate_high_dose) ~ '**Xanomeline**',
      c(p_value_low_dose_placebo:p_value_high_dose_low_dose) ~ '**p-value / Diff (SE) / (95% CI)**') %>% 
   modify_header(
      estimate_placebo ~ '**Placebo**',
      estimate_low_dose ~ '**Low Dose**',
      estimate_high_dose ~ '**High Dose**',
      p_value_low_dose_placebo  ~ '**Low Dose - <br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose - <br>Placebo**',
      p_value_high_dose_low_dose    ~ '**High Dose - <br>Low Dose**') %>% 
   modify_column_alignment(
      columns = starts_with(c('est','p_val')), 'center') %>% 
   modify_footnote(
      update = list(
         c(p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ '
         Note: The change from baseline is calculated as the post-baseline score minus the baseline score. <br>The
         covariates included in the MMRM model are treatment, site group, time and treatment by time interaction,<br>
         baseline ADAS-Cog (11) score, and baseline ADAS-Cog (11) score by time interaction.' ))%>% 
   modify_caption(
      '**Table 14-3.11<br>ADAS Cog (11) - Repeated Measures Analysis of Change from Baseline to Week 24**')

Table 14-3.11
ADAS Cog (11) - Repeated Measures Analysis of Change from Baseline to Week 24

Placebo	Xanomeline		p-value / Diff (SE) / (95% CI)
	Low Dose	High Dose	Low Dose - Placebo1	High Dose - Placebo1	High Dose - Low Dose1
1.6 (0.49)	1.5 (0.52)	1.1 (0.55)	0.954 0.0 (0.70) (-1.4;1.3)	0.555 -0.4 (0.72) (-1.8;1.0)	0.604 -0.4 (0.74) (-1.9;1.1)
1 Note: The change from baseline is calculated as the post-baseline score minus the baseline score. The covariates included in the MMRM model are treatment, site group, time and treatment by time interaction, baseline ADAS-Cog (11) score, and baseline ADAS-Cog (11) score by time interaction.

By Week

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtreg, gtsummary)
# MMRM MODEL
pacman::p_load(emmeans, mmrm)

# THEME
theme_gtsummary_compact()

# IMPORT DATA
adqsadas_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsadas.xpt')

# DATA LOCF
adqsadas <- adqsadas_orig %>%
   filter(EFFFL == 'Y' & PARAMCD == 'ACTOT' & ANL01FL == 'Y' & DTYPE != 'LOCF' & AVISITN > 0) %>% 
   select(USUBJID, TRTP, SITEGR1, AVISIT, AVISITN, AVAL, BASE, CHG) %>% 
   mutate(USUBJID = factor(USUBJID)) %>% 
   mutate(across(c(USUBJID, SITEGR1, AVISIT), factor)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(AVISIT = factor(AVISIT,
                          levels = c( 'Week 8', 'Week 16', 'Week 24') ) )

# EMMEANS & BIND DATA
m <- mmrm(
   CHG ~ SITEGR1 + BASE + AVISIT + TRTP * AVISIT + us(AVISIT | USUBJID),
   data = adqsadas,
   reml = TRUE,
   control = mmrm_control(
      method = 'Kenward-Roger'))

e <- emmeans(m, ~TRTP|AVISIT, calc = c(n = '.wgt.'), infer = TRUE)
c <- contrast(e, method = 'revpairwise') %>% summary(infer = TRUE, adjust = 'none')

d <- bind_rows(
   e %>% data.frame() %>% rename(estimate = emmean),
   c %>% data.frame() %>% rename(TRTP = contrast)) %>% 
   select(-n, -df, -t.ratio)

# WIDE DATA
d_l <- d %>% 
   pivot_wider(id_cols = AVISIT,
               values_from = -c(TRTP, AVISIT),
               names_from = TRTP) %>% 
   janitor::clean_names()

# TABLE
tbl_listing(
   data = d_l ) %>% 
   modify_column_merge('{estimate_placebo} ({se_placebo})') %>% 
   modify_column_merge('{estimate_low_dose} ({se_low_dose})') %>% 
   modify_column_merge('{estimate_high_dose} ({se_high_dose})') %>%
   modify_column_merge('{p_value_low_dose_placebo}<br>{estimate_low_dose_placebo}   ({se_low_dose_placebo})<br>({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})') %>% 
   modify_column_merge('{p_value_high_dose_placebo}<br>{estimate_high_dose_placebo} ({se_high_dose_placebo})<br>({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})') %>%
   modify_column_merge('{p_value_high_dose_low_dose}<br>{estimate_high_dose_low_dose}   ({se_high_dose_low_dose})<br>({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})') %>%
   modify_column_hide(
      c('upper_cl_placebo','upper_cl_low_dose','upper_cl_high_dose',
        'lower_cl_placebo','lower_cl_low_dose','lower_cl_high_dose',
        'p_value_placebo','p_value_low_dose','p_value_high_dose')) %>% 
   modify_fmt_fun(
      list(
         starts_with( c('estimate_', 'upper_', 'lower_') ) ~ function(x) style_number(x, digits = 1),
         starts_with( 'se_') ~ function(x) style_number(x, digits = 2),
         starts_with( 'p_value_') ~ function(x) style_pvalue(x, digits = 3)),
      rows = !is.na(estimate_placebo)) %>% 
   modify_header( avisit ~ 'Week') %>% 
   modify_spanning_header(
      c(estimate_low_dose, estimate_high_dose) ~ '**Xanomeline**',
      c(p_value_low_dose_placebo:p_value_high_dose_low_dose) ~ '**p-value / Diff (SE) / (95% CI)**') %>% 
   modify_header(
      estimate_placebo ~ '**Placebo**',
      estimate_low_dose ~ '**Low Dose**',
      estimate_high_dose ~ '**High Dose**',
      p_value_low_dose_placebo  ~ '**Low Dose - <br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose - <br>Placebo**',
      p_value_high_dose_low_dose    ~ '**High Dose - <br>Low Dose**') %>% 
   modify_column_alignment(
      columns = starts_with(c('est','p_val')), 'center') %>% 
   modify_footnote(
      update = list(
         c(p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ '
         Note: The change from baseline is calculated as the post-baseline score minus the baseline score. <br>The
         covariates included in the MMRM model are treatment, site group, time and treatment by time interaction,<br>
         baseline ADAS-Cog (11) score, and baseline ADAS-Cog (11) score by time interaction.' ))%>% 
   modify_caption(
      '**Table 14-3.11<br>ADAS Cog (11) - Repeated Measures Analysis of Change from Baseline to Week 8, 16 and 24**')

Table 14-3.11
ADAS Cog (11) - Repeated Measures Analysis of Change from Baseline to Week 8, 16 and 24

Week	Placebo	Xanomeline		p-value / Diff (SE) / (95% CI)
		Low Dose	High Dose	Low Dose - Placebo1	High Dose - Placebo1	High Dose - Low Dose1
Week 8	0.6 (0.48)	1.6 (0.47)	0.8 (0.49)	0.107 1.0 (0.65) (-0.2;2.3)	0.757 0.2 (0.67) (-1.1;1.5)	0.205 -0.8 (0.66) (-2.2;0.5)
Week 16	1.8 (0.64)	1.2 (0.76)	1.1 (0.79)	0.587 -0.5 (0.98) (-2.5;1.4)	0.488 -0.7 (1.00) (-2.7;1.3)	0.882 -0.2 (1.09) (-2.3;2.0)
Week 24	2.3 (0.68)	1.7 (0.76)	1.5 (0.82)	0.550 -0.6 (1.01) (-2.6;1.4)	0.441 -0.8 (1.06) (-2.9;1.3)	0.848 -0.2 (1.11) (-2.4;2.0)
1 Note: The change from baseline is calculated as the post-baseline score minus the baseline score. The covariates included in the MMRM model are treatment, site group, time and treatment by time interaction, baseline ADAS-Cog (11) score, and baseline ADAS-Cog (11) score by time interaction.

Warning

I successfully reproduced the results, unlike the attempt by atorus-research, by using the newer mmrm package.

Tip

The title of the table suggests a ‘… change from baseline to week 24’; however, as slophaven pointed out, the numbers presented do not correspond to changes up to week 24 but rather represent the average changes observed at weeks 8, 16, and 24. This discrepancy suggests that the intended focus of this analysis was on the interaction, which we have accordingly replicated here.

Table 14-3.12

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gt, gtsummary)
pacman::p_load(emmeans)

# THEME
theme_gtsummary_compact()

# OPTIONS
theme_gtsummary_language(
   language = 'en',
   decimal.mark = '.',
   ci.sep = '; ',
   iqr.sep = '; ',
   set_theme = TRUE
)

# IMPORT DATA
adqsnpix_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqsnpix.xpt')

# SUBSET
adqsnpix <- adqsnpix_orig %>%
   filter(EFFFL == 'Y' & ITTFL == 'Y' & PARAMCD == 'NPTOTMN') %>%
   filter(AVISITN %in% c(0, 98)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   select(USUBJID, SITEGR1, TRTP, AVISITN, AVAL)

# TRANSFORM
adqsnpix <- adqsnpix %>% 
   mutate(AVISITN = factor(AVISITN, labels = c('BASE','AVAL'))) %>% 
   pivot_wider(names_from = AVISITN,
               values_from = AVAL)

# REGRESSION
lm_pairwise <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'), ...) {
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), data = data)
   e <- l %>% 
      emmeans::emmeans(specs = t, calc = c(n = '.wgt.')) %>% 
      emmeans::contrast(method = 'revpairwise') %>% 
      summary(infer = TRUE, adjust = 'none')
   b <- e %>% 
      select( -df, -t.ratio) %>% 
      rename(label = contrast) %>% 
      pivot_wider(names_from = label,
                  values_from = -1) %>% 
      janitor::clean_names()
   b
}

lm_trend  <- function(data, variable, by, adj.vars = c('BASE','SITEGR1'),...){
   f <- str_glue('{variable} ~ {paste(adj.vars,  collapse = " + ")} + {by}')
   t <- str_glue('~ {by}') %>% as.formula()
   l <- lm(as.formula(f), 
           data = data %>% 
              mutate(!!sym(by) := factor(!!sym(by), labels = c(0, 54, 81)) %>% 
                        as.character() %>% 
                        as.numeric() ) )
   b <- l %>% broom::tidy(conf.int = TRUE) %>% slice(n())
   b %>% select(p.value)
}

# SUMMARY w/ PAIRWISE
tbl_summary(
   data = adqsnpix,
   by = TRTP,
   include = c(BASE, AVAL),
   statistic = list(everything() ~ c('{N_nonmiss}<br>{mean} ({sd})<br>{median} ({min};{max})')),
   digit = list(everything() ~ c(0, 1, 2, 1, 0, 0)),
   missing = 'no',
   label = list(BASE ~ 'Baseline',
                AVAL ~ 'Mean of Weeks 4-24')) %>% 
   add_stat_label(
      label = all_continuous() ~ c('n<br>Mean (SD)<br>Median (Range)') ) %>% 
   add_stat(fns = AVAL ~ lm_trend ) %>% 
   add_stat(fns = AVAL ~ lm_pairwise) %>% 
   modify_fmt_fun( list(
      starts_with('p.value')    ~ function(x) style_pvalue(x, digits = 3) ,
      starts_with('estimate_')  ~ function(x) style_number(x, digits = 1),
      starts_with('se_')        ~ function(x) style_number(x, digits = 2),
      starts_with('lower_cl')   ~ function(x) style_number(x, digits = 1),
      starts_with('upper_cl')   ~ function(x) style_number(x, digits = 1))) %>% 
   modify_column_merge(
      pattern = '{p_value_low_dose_placebo}<br>
      {estimate_low_dose_placebo} ({se_low_dose_placebo})<br>
      ({lower_cl_low_dose_placebo};{upper_cl_low_dose_placebo})<br>',
      rows = !is.na(estimate_low_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_placebo}<br>
      {estimate_high_dose_placebo} ({se_high_dose_placebo})<br>
      ({lower_cl_high_dose_placebo};{upper_cl_high_dose_placebo})<br>',
      rows = !is.na(estimate_high_dose_placebo)) %>% 
   modify_column_merge(
      pattern = '{p_value_high_dose_low_dose}<br>
      {estimate_high_dose_low_dose} ({se_high_dose_low_dose})<br>
      ({lower_cl_high_dose_low_dose};{upper_cl_high_dose_low_dose})<br>',
      rows = !is.na(estimate_high_dose_low_dose)) %>% 
   modify_header(
      label = '',
      all_stat_cols() ~ '**{level}<br>(N={n})**',
      p.value ~ '**Trend Test**',
      p_value_low_dose_placebo ~ '**Low Dose -<br>Placebo**', 
      p_value_high_dose_placebo ~ '**High Dose -<br>Placebo**',
      p_value_high_dose_low_dose ~ '**High Dose -<br>Low Dose**') %>%    
   modify_spanning_header(
      c(stat_2, stat_3) ~ '<br><br>**Xanomeline**',
                p.value ~ '<br><br>**p-value**',
      starts_with('p_value_') ~ '**p-value<br>Diff of LS Means (SE)<br>95% CI**') %>% 
   modify_footnote(
      update = list(
         c(p.value, 
           p_value_low_dose_placebo,
           p_value_high_dose_placebo,
           p_value_high_dose_low_dose) ~ 'Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate.<br>
                                          Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons<br>
                                          Note that only assessments falling within the assessment window are included in the summary for a visit.',
         c(p.value)                    ~ 'Test for a non-zero coefficient for treatment (dose) as a continuous variable.') ) %>% 
   modify_caption(
      '**Table 14-3.12<br>Mean NPI-X Total Score from Week 4 through Week 24 - Windowed**') %>% 
   as_gt() %>% 
   gt::fmt_markdown(columns = -c(p.value))

Table 14-3.12
Mean NPI-X Total Score from Week 4 through Week 24 - Windowed

	Placebo (N=79)	Xanomeline		p-value	p-value Diff of LS Means (SE) 95% CI
		Low Dose (N=81)	High Dose (N=74)	Trend Test1	Low Dose - Placebo2	High Dose - Placebo2	High Dose - Low Dose2
Baseline, n Mean (SD) Median (Range)	79 9.5 (12.10) 5.0 (0;66)	81 8.7 (9.82) 4.0 (0;32)	74 11.9 (13.70) 8.0 (0;61)
Mean of Weeks 4-24, n Mean (SD) Median (Range)	76 9.1 (11.82) 4.2 (0;67)	69 7.7 (10.09) 3.3 (0;49)	65 7.8 (10.54) 2.8 (0;42)	0.142	0.594 -0.6 (1.14) (-2.9;1.6)	0.119 -1.8 (1.16) (-4.1;0.5)	0.311 -1.2 (1.19) (-3.5;1.1)
1 Test for a non-zero coefficient for treatment (dose) as a continuous variable.
2 Based on Analysis of covariance (ANCOVA) model with treatment and site group as factors and baseline value as a covariate. Pairwise comparison with treatment as a categorical variable: p-values without adjustment for multiple comparisons Note that only assessments falling within the assessment window are included in the summary for a visit.

Warning

Table 14-3.12. Difference in values of Mean of Weeks 4-24. This was programmed using the derived NPTOTMN variable. The ARM in the original CDISC Pilot Define.xml was followed as best as possible to determine the subset. This means that the counts are a discrepancy with the original CDISC Pilot analysis data, which is no longer available, therefore we were not able to investigate the discrepancy with the current PHUSE CDISC Pilot replication data. The subsequent statistical summaries therefore also have differences.

Table 14-3.13

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_compact()

# DATA LOCF
adqscibc_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adqscibc.xpt') 

adqscibc <- adqscibc_orig %>% 
   filter(EFFFL == 'Y' & ITTFL == 'Y', AVISITN %in% c(8, 16, 24) & ANL01FL=='Y') %>%
   select(USUBJID, TRTP, SITEGR1, AVISIT, AVISITN, AVAL) %>% 
   mutate(AVALC = factor(AVAL, levels = c(1:7) ) %>% 
             fct_recode('Marked improvement'   = '1',
                        'Moderate improvement' = '2',
                        'Minimal improvement'  = '3',
                        'No Change'            = '4',
                        'Minimal worsening'    = '5',
                        'Moderate worsening'   = '6',
                        'Marked worsening'     = '7') ) %>% 
   mutate(across(c(USUBJID, SITEGR1, AVISIT), factor)) %>% 
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(AVISIT = factor(AVISIT,
                          levels = c( 'Week 8', 'Week 16', 'Week 24') ) )

# cMH TEST
mntl_test <- function (data, variable, by, strata = 'SITEGR1', ...) {
   expr(
      stats::mantelhaen.test(
         !!data[[variable]], 
         as.factor(!!data[[by]]),
         as.factor(!!data[[strata]]) )) %>% 
      eval() %>% 
      broom::tidy() %>% 
      select(p.value, method)
}
# mntl_test(adqscibc %>% filter(AVISIT == 'Week 8'), 'AVAL', 'TRTP', strata = 'SITEGR1')

# SUMMARY
t_summary <- adqscibc %>% 
   tbl_strata2(
      strata = AVISIT,
      .tbl_fun = 
         ~ .x %>% 
         tbl_summary(
            include = AVALC,
            by = TRTP,
            label = list(AVALC = .y)) %>% 
         add_p( 
            test = AVALC ~ 'mntl_test',
            pvalue_fun = \(.) style_pvalue(., digits = 3)),
      .combine_with = 'tbl_stack',
      .combine_args = list(group_header = NULL),
      .quiet = TRUE)

# CROSS
adqscibc %>% 
   tbl_strata2(
      strata = AVISIT,
      .tbl_fun = 
         ~ .x %>% 
         tbl_cross(
            label = list(AVALC = .y ),
            row = AVALC,
            col = TRTP, 
            digits = c(0, 0),
            percent = 'column',
            margin = 'row',
            margin_text = 'n') %>%
         add_p( pvalue_fun = \(.) style_pvalue(., digits = 3)),
      .combine_with = 'tbl_stack',
      .combine_args = list(group_header = NULL)) %>%
   modify_table_body(
      ~ .x %>%
         mutate(across(starts_with('stat_'), ~ifelse(var_label=='n', str_replace(.x, '\\(100%\\)', ''), .x) ) ) %>% 
         mutate(across(starts_with('stat_'), ~ifelse(var_label!='n', str_replace(.x, '\\(0%\\)', ''), .x) ) ) %>% 
         rows_update(t_summary$table_body %>% filter(!is.na(p.value)),
                     by = c('tbl_id1', 'var_label', 'row_type', 'label') ) %>%
         slice(c(1,n(),2:8), .by = tbl_id1)) %>%
   modify_header( label ~ '**Assessment**',
                  all_stat_cols() ~ '**{level}**',
                  p.value         ~ '**p-value**') %>% 
   modify_spanning_header(
      stat_1            ~ '&nbsp;',
      c(stat_2, stat_3) ~ '**Xanomeline**' ) %>%
   modify_column_alignment(all_stat_cols(), 'left') %>% 
   # modify_column_indent(columns = label, undo = TRUE) %>% 
   modify_footnote(
      p.value ~ 'Overall comparison of treatments using CMH test (Pearson Chi-Square), controlling for site group.') %>% 
   modify_caption(
      '**Table 14-3.13<br>CIBIC+ - Categorical Analysis - LOCF**')

Table 14-3.13
CIBIC+ - Categorical Analysis - LOCF

Assessment		Xanomeline		p-value1
	Placebo	Low Dose	High Dose
Week 8				0.478
n	77	81	73
Marked improvement	0	0	0
Moderate improvement	1 (1%)	2 (2%)	1 (1%)
Minimal improvement	19 (25%)	16 (20%)	13 (18%)
No Change	45 (58%)	48 (59%)	38 (52%)
Minimal worsening	10 (13%)	14 (17%)	20 (27%)
Moderate worsening	2 (3%)	1 (1%)	1 (1%)
Marked worsening	0	0	0
Week 16				0.770
n	79	81	74
Marked improvement	0	0	0
Moderate improvement	0	3 (4%)	2 (3%)
Minimal improvement	12 (15%)	12 (15%)	13 (18%)
No Change	41 (52%)	46 (57%)	39 (53%)
Minimal worsening	25 (32%)	19 (23%)	20 (27%)
Moderate worsening	1 (1%)	1 (1%)	0
Marked worsening	0	0	0
Week 24				0.868
n	79	81	74
Marked improvement	0	0	0
Moderate improvement	1 (1%)	1 (1%)	0
Minimal improvement	9 (11%)	14 (17%)	11 (15%)
No Change	38 (48%)	37 (46%)	33 (45%)
Minimal worsening	28 (35%)	27 (33%)	25 (34%)
Moderate worsening	3 (4%)	2 (2%)	5 (7%)
Marked worsening	0	0	0
1 Overall comparison of treatments using CMH test (Pearson Chi-Square), controlling for site group.

Important

I was unable to reproduce the Cochran-Mantel-Haenszel Chi-Squared Test for this table. However, by using the original data from atorus-research, I could replicate my findings, leading me to believe that there may be an error in the original table.

Table 14-4

Table 14-4-01

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# Read in ADSL
adsl_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt') 


# STACK
adsl_l <- bind_rows(
   list(
      COMP24FL =  adsl_orig %>% filter(COMP24FL == 'Y') , 
      SAFFL    =  adsl_orig %>% filter(SAFFL == 'Y')
   ),
   .id = 'POP') %>% 
   select(POP, USUBJID, TRT01P, AVGDD, CUMDOSE) %>% 
   mutate(TRT01P = factor(TRT01P, 
                          levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                          labels = c('Placebo', 'Xanomeline<br>Low Dose','Xanomeline<br>High Dose'))) %>% 
   mutate(POP = factor(POP,
                       labels = c('**Completers at Week 24** <br>',
                                  '**Safety Population** <br> <small>Includes completers and early terminators</small>') ))

# TABLE
adsl_l %>%
   tbl_strata(
      strata = POP,
      .tbl_fun =
         ~ .x %>%
         tbl_summary(
            by = TRT01P,
            include = c(AVGDD, CUMDOSE),
            label = list(AVGDD ~ 'Average daily dose (mg)',
                         CUMDOSE ~ 'Cumulative dose at end of study'),
            type = list(everything() ~ 'continuous2'),
            digits    = list(everything() ~ c(            0,        1,      2,          1,       1,       1)),
            statistic = list(everything() ~ c('{N_nonmiss}', '{mean}', '{sd}', '{median}', '{min}', '{max}'))) %>% 
         modify_header(
            label = '',
            all_stat_cols() ~ '**{level}<br>(n={n})**') %>% 
         modify_table_body(
            ~ .x %>%
               mutate(label = ifelse(row_type == 'level', str_to_lower(label), label) ) %>% 
               mutate(label = case_when(
                  label == 'minimum' ~ 'min',
                  label == 'maximum' ~ 'max',
                  .default = label))),
      .header = '**{strata}<br>(n={n})**',
      .combine_with = 'tbl_merge') %>% 
   modify_table_styling(
      columns = label,
      rows =  variable == 'CUMDOSE' & row_type  == 'label',
      footnote = 'End of Study refers to week 26/Early Termination.') %>% 
   modify_caption(
      '**Table 14-4.01<br> Summary of Planned Exposure to Study Drug, as of End of Study**' )

Table 14-4.01
Summary of Planned Exposure to Study Drug, as of End of Study

	Completers at Week 24 (n=118)			Safety Population Includes completers and early terminators (n=254)
	Placebo (n=60)	Xanomeline Low Dose (n=28)	Xanomeline High Dose (n=30)	Placebo (n=86)	Xanomeline Low Dose (n=84)	Xanomeline High Dose (n=84)
Average daily dose (mg)
n	60	28	30	86	84	84
mean	0.0	54.0	77.0	0.0	54.0	71.6
sd	0.00	0.00	0.58	0.00	0.00	8.11
median	0.0	54.0	76.9	0.0	54.0	75.1
min	0.0	54.0	76.1	0.0	54.0	54.0
max	0.0	54.0	78.6	0.0	54.0	78.6
Cumulative dose at end of study1
n	60	28	30	86	84	84
mean	0.0	9918.6	14089.5	0.0	5347.3	7551.0
sd	0.00	603.84	481.01	0.00	3680.35	5531.04
median	0.0	9936.0	14080.5	0.0	4455.0	5778.0
min	0.0	7884.0	12960.0	0.0	108.0	54.0
max	0.0	11448.0	15417.0	0.0	11448.0	15417.0
1 End of Study refers to week 26/Early Termination.

Table 14-5

Table 14-5-01

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary, gtreg)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ ADSL
adsl_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt') 

adsl <- adsl_orig %>% 
   filter(SAFFL == 'Y') %>% 
   rename(TRTA = TRT01A) %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose')))

# READ ADAE
adae_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adae.xpt') 

adae <- adae_orig %>% 
   filter(SAFFL == 'Y' & TRTEMFL == 'Y') %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(across(c(AEBODSYS, AETERM), ~str_to_title(.x))) 
   # %>% filter(AEBODSYS %in% c('Cardiac Disorders') )

# COMPLETE DATA AE SOC
cmpl_ <- .complete_ae_data(
   data = adae,
   id   = "USUBJID",
   ae   = "AETERM",
   soc  = "AEBODSYS",
   strata = 'TRTA',
   id_df = adsl) %>% 
   mutate(across(c(`..soc..`,`..ae..`), ~ factor(.x)) ) 

# COMPLETE DATA TT
cmpl_tt <- .complete_ae_data(
   data = adae %>% mutate(AETERM   = 'Any Body System',
                          AEBODSYS = 'Any Body System'),
   id   = "USUBJID",
   ae   = "AETERM",
   soc  = "AEBODSYS",
   strata = 'TRTA',
   id_df = adsl) %>% 
   mutate(across(c(`..soc..`,`..ae..`), ~ factor(.x)) ) 

# LONG w/ WITH TOTALS
cmpl_l <- 
   bind_rows(cmpl_,
             cmpl_tt) %>% 
   pivot_longer(cols = c(`..soc..`,`..ae..`)) %>% 
   mutate(ae = ifelse(name == '..soc..', str_glue('All {soc}'), str_glue('{ae}')) ) %>% 
   nest_by(name, soc, ae) %>% 
   filter(! (name == '..ae..' & ae == 'Any Body System')) %>% 
   arrange(soc, ae)


cmpl_tbl <- cmpl_l %>% 
   mutate(
      t_n  = list(
         tbl_summary(
            data = data,
            by = strata,
            include = `by`,
            value = list(`by` ~ 'Overall'),
            label = list(`by` ~ ae),
            statistic = list(everything() ~ '[{n}]')
         ) 
      ),
      t_3g = list( 
         tbl_summary(
            data = data %>% filter(value == TRUE),
            by = strata,
            value = list(`by` ~ 'Overall'),
            include = `by`,
            label = list(`by` ~ ae),
         ) 
      ),
      t_PvsL = list(
         tbl_summary(
            data = data %>% 
               filter(value == TRUE) %>% 
               filter(strata %in% c('Placebo','Low Dose')) %>% 
               droplevels(),
            by = strata,
            value = list(`by` ~ 'Overall'),
            include = `by`,
            label = list(`by` ~ ae),
         ) %>% 
            add_p(
               test = list(all_categorical() ~ 'fisher.test'),
               pvalue_fun = function(x) style_pvalue(x, digits = 3) 
            ) %>% 
            add_significance_stars(
               thresholds = c(0.15),
               hide_p = FALSE,
               pattern = "{p.value}{stars}"
            ) 
      ),
      t_PvsH = list(
         tbl_summary(
            data = data %>% 
               filter(value == TRUE) %>% 
               filter(strata %in% c('Placebo','High Dose')) %>% 
               droplevels(),
            by = strata,
            value = list(`by` ~ 'Overall'),
            include = `by`,
            label = list(`by` ~ ae),
         ) %>% 
            add_p(
               test = list(all_categorical() ~ 'fisher.test'),
               pvalue_fun = function(x) style_pvalue(x, digits = 3) 
            ) %>% 
            add_significance_stars(
               thresholds = c(0.15),
               hide_p = FALSE,
               pattern = "{p.value}{stars}"
            ) 
      ),
      t = list(
         tbl_merge( tbls = list(t_3g,
                                t_n,
                                t_PvsL %>% modify_column_hide(all_stat_cols()) ,
                                t_PvsH %>% modify_column_hide(all_stat_cols()) ),
                    tab_spanner = FALSE)
      )
      
   )

# STACK
with(cmpl_tbl,
     tbl_stack(tbls = t,
               quiet = TRUE) ) %>%
   modify_column_merge( "{stat_1_1} {stat_1_2}") %>% 
   modify_column_merge( "{stat_2_1} {stat_2_2}") %>% 
   modify_column_merge( "{stat_3_1} {stat_3_2}") %>% 
   modify_header(
      label = '',
      c(stat_1_1, 
        stat_2_1,
        stat_3_1) ~ '**{level}<br>(N = {n})**',
      p.value_3 ~ '**Placebo v<br>Low Dose**',
      p.value_4 ~ '**Placebo v<br>High Dose**') %>% 
   modify_spanning_header(
      c(stat_2_1, stat_3_1) ~ "**Xanomeline**",
      c(p.value_3, p.value_4) ~ "**Fisher's Exact p-values**") %>% 
   modify_table_body(
      ~ .x %>%
         mutate(stat_3_2_n =  str_extract(stat_3_2 , "\\d+") %>% as.numeric()) %>% 
         mutate(across(starts_with('stat_'), ~str_replace_all(.x, "\\[0\\]", ""))) %>%
         mutate(across(starts_with('stat_'), ~str_replace(., "0 \\(0%\\)", "0"))) %>% 
         slice(c(n(), 1:(n()-1)) ) ) %>% 
   modify_table_body(
      ~ .x %>%
         mutate(row_type = ifelse( label %in% label[!str_detect(label, "^All")], 'level','label') ) ) %>% 
   modify_column_alignment(columns = all_stat_cols(), align = "left") %>% 
   bold_labels() %>% 
   modify_footnote(
      all_stat_cols() ~ 
         "Treatment emergent events are defined as events which start on or after the start of treatment.<br>
      Adverse events are coded using MedDRA.<br>
      Percentages are based on the number of subjects in the safety population within each treatment group.<br>
      [AE] represents the total number of times an event was recorded.",
      starts_with('p.value') ~ 
   "P-values are based on Fisher's Exact test for the comparison of placebo versus each active treatment
      group. An asterisk is appended to p-values that are less than 0.15.") %>% 
   modify_caption('**Table 14-5.01<br>Incidence of Treatment Emergent Adverse Events by Treatment Group**')

Table 14-5.01
Incidence of Treatment Emergent Adverse Events by Treatment Group

	Placebo (N = 86)1	Xanomeline		Fisher’s Exact p-values
		Low Dose (N = 84)1	High Dose (N = 84)1	Placebo v Low Dose2	Placebo v High Dose2
All Any Body System	65 (76%) [281]	77 (92%) [412]	76 (90%) [433]	0.007*	0.014*
All Cardiac Disorders	12 (14%) [26]	13 (15%) [30]	15 (18%) [30]	0.831	0.534
Atrial Fibrillation	1 (1.2%) [1]	1 (1.2%) [1]	3 (3.6%) [5]	>0.999	0.365
Atrial Flutter	0	1 (1.2%) [1]	1 (1.2%) [2]	0.494	0.494
Atrial Hypertrophy	1 (1.2%) [2]	0	0	>0.999	>0.999
Atrioventricular Block First Degree	1 (1.2%) [1]	1 (1.2%) [1]	0	>0.999	>0.999
Atrioventricular Block Second Degree	1 (1.2%) [1]	0	0	>0.999	>0.999
Bradycardia	1 (1.2%) [4]	0	0	>0.999	>0.999
Bundle Branch Block Left	1 (1.2%) [1]	0	0	>0.999	>0.999
Bundle Branch Block Right	1 (1.2%) [2]	1 (1.2%) [1]	0	>0.999	>0.999
Cardiac Disorder	0	0	1 (1.2%) [1]		0.494
Cardiac Failure Congestive	1 (1.2%) [1]	0	0	>0.999	>0.999
Myocardial Infarction	4 (4.7%) [4]	2 (2.4%) [4]	4 (4.8%) [8]	0.682	>0.999
Palpitations	0	2 (2.4%) [2]	0	0.243
Sinus Arrhythmia	1 (1.2%) [2]	0	0	>0.999	>0.999
Sinus Bradycardia	2 (2.3%) [2]	7 (8.3%) [10]	8 (9.5%) [12]	0.097*	0.056*
Supraventricular Extrasystoles	1 (1.2%) [2]	1 (1.2%) [2]	1 (1.2%) [1]	>0.999	>0.999
Supraventricular Tachycardia	0	1 (1.2%) [2]	0	0.494
Tachycardia	1 (1.2%) [2]	0	0	>0.999	>0.999
Ventricular Extrasystoles	0	2 (2.4%) [4]	1 (1.2%) [1]	0.243	0.494
Ventricular Hypertrophy	1 (1.2%) [1]	0	0	>0.999	>0.999
Wolff-Parkinson-White Syndrome	0	1 (1.2%) [2]	0	0.494
All Congenital, Familial And Genetic Disorders	0	1 (1.2%) [1]	2 (2.4%) [2]	0.494	0.243
Ventricular Septal Defect	0	1 (1.2%) [1]	2 (2.4%) [2]	0.494	0.243
All Ear And Labyrinth Disorders	1 (1.2%) [2]	2 (2.4%) [2]	1 (1.2%) [1]	0.618	>0.999
Cerumen Impaction	0	1 (1.2%) [1]	0	0.494
Ear Pain	1 (1.2%) [2]	0	0	>0.999	>0.999
Vertigo	0	1 (1.2%) [1]	1 (1.2%) [1]	0.494	0.494
All Eye Disorders	2 (2.3%) [5]	2 (2.4%) [2]	1 (1.2%) [2]	>0.999	>0.999
Conjunctival Haemorrhage	0	1 (1.2%) [1]	0	0.494
Conjunctivitis	1 (1.2%) [2]	0	0	>0.999	>0.999
Eye Allergy	1 (1.2%) [1]	0	0	>0.999	>0.999
Eye Pruritus	1 (1.2%) [1]	0	0	>0.999	>0.999
Eye Swelling	1 (1.2%) [1]	0	0	>0.999	>0.999
Vision Blurred	0	1 (1.2%) [1]	1 (1.2%) [2]	0.494	0.494
Abdominal Discomfort	0	0	1 (1.2%) [1]		0.494
Abdominal Pain	1 (1.2%) [1]	3 (3.6%) [3]	1 (1.2%) [2]	0.365	>0.999
All Gastrointestinal Disorders	17 (20%) [26]	14 (17%) [22]	20 (24%) [36]	0.692	0.580
Constipation	1 (1.2%) [1]	0	0	>0.999	>0.999
Diarrhoea	9 (10%) [10]	4 (4.8%) [5]	4 (4.8%) [4]	0.248	0.248
Dyspepsia	1 (1.2%) [2]	1 (1.2%) [2]	0	>0.999	>0.999
Dysphagia	0	1 (1.2%) [1]	0	0.494
Flatulence	1 (1.2%) [2]	0	0	>0.999	>0.999
Gastrointestinal Haemorrhage	0	0	1 (1.2%) [1]		0.494
Gastrooesophageal Reflux Disease	1 (1.2%) [1]	0	0	>0.999	>0.999
Glossitis	1 (1.2%) [1]	0	0	>0.999	>0.999
Hiatus Hernia	1 (1.2%) [2]	0	0	>0.999	>0.999
Nausea	3 (3.5%) [3]	3 (3.6%) [5]	6 (7.1%) [13]	>0.999	0.326
Rectal Haemorrhage	0	1 (1.2%) [2]	0	0.494
Salivary Hypersecretion	0	0	4 (4.8%) [5]		0.057*
Stomach Discomfort	0	0	1 (1.2%) [1]		0.494
Vomiting	3 (3.5%) [3]	3 (3.6%) [4]	7 (8.3%) [9]	>0.999	0.208
All General Disorders And Administration Site Conditions	21 (24%) [46]	47 (56%) [118]	40 (48%) [124]	<0.001*	0.002*
Application Site Bleeding	0	1 (1.2%) [1]	0	0.494
Application Site Dermatitis	5 (5.8%) [9]	9 (11%) [15]	7 (8.3%) [12]	0.277	0.563
Application Site Desquamation	0	1 (1.2%) [1]	0	0.494
Application Site Discharge	0	0	1 (1.2%) [1]		0.494
Application Site Discolouration	0	1 (1.2%) [1]	0	0.494
Application Site Erythema	3 (3.5%) [3]	12 (14%) [20]	15 (18%) [23]	0.015*	0.002*
Application Site Induration	1 (1.2%) [1]	0	0	>0.999	>0.999
Application Site Irritation	3 (3.5%) [7]	9 (11%) [18]	9 (11%) [16]	0.078*	0.078*
Application Site Pain	0	0	2 (2.4%) [2]		0.243
Application Site Perspiration	0	0	2 (2.4%) [3]		0.243
Application Site Pruritus	6 (7.0%) [10]	22 (26%) [32]	22 (26%) [35]	<0.001*	<0.001*
Application Site Reaction	1 (1.2%) [2]	0	1 (1.2%) [1]	>0.999	>0.999
Application Site Swelling	0	1 (1.2%) [1]	2 (2.4%) [3]	0.494	0.243
Application Site Urticaria	0	2 (2.4%) [2]	1 (1.2%) [1]	0.243	0.494
Application Site Vesicles	1 (1.2%) [2]	4 (4.8%) [5]	6 (7.1%) [6]	0.208	0.062*
Application Site Warmth	0	1 (1.2%) [2]	0	0.494
Asthenia	1 (1.2%) [2]	0	1 (1.2%) [1]	>0.999	>0.999
Chest Discomfort	0	0	2 (2.4%) [2]		0.243
Chest Pain	0	0	2 (2.4%) [2]		0.243
Chills	1 (1.2%) [3]	1 (1.2%) [2]	1 (1.2%) [1]	>0.999	>0.999
Fatigue	1 (1.2%) [2]	5 (6.0%) [5]	5 (6.0%) [5]	0.115*	0.115*
Feeling Abnormal	0	0	1 (1.2%) [1]		0.494
Feeling Cold	0	0	1 (1.2%) [1]		0.494
Inflammation	0	1 (1.2%) [1]	0	0.494
Malaise	0	1 (1.2%) [2]	2 (2.4%) [3]	0.494	0.243
Oedema	0	2 (2.4%) [2]	0	0.243
Oedema Peripheral	2 (2.3%) [3]	1 (1.2%) [1]	2 (2.4%) [3]	>0.999	>0.999
Pain	0	1 (1.2%) [2]	1 (1.2%) [1]	0.494	0.494
Pyrexia	2 (2.3%) [2]	0	1 (1.2%) [1]	0.497	>0.999
Secretion Discharge	0	1 (1.2%) [2]	0	0.494
Sudden Death	0	1 (1.2%) [1]	0	0.494
Swelling	0	1 (1.2%) [1]	0	0.494
Ulcer	0	1 (1.2%) [1]	0	0.494
All Hepatobiliary Disorders	1 (1.2%) [1]	0	0	>0.999	>0.999
Hyperbilirubinaemia	1 (1.2%) [1]	0	0	>0.999	>0.999
All Immune System Disorders	0	1 (1.2%) [2]	0	0.494
Hypersensitivity	0	1 (1.2%) [2]	0	0.494
All Infections And Infestations	16 (19%) [35]	9 (11%) [16]	13 (15%) [20]	0.194	0.685
Bronchitis	1 (1.2%) [1]	0	0	>0.999	>0.999
Cellulitis	0	1 (1.2%) [1]	0	0.494
Cervicitis	1 (1.2%) [2]	0	0	>0.999	>0.999
Cystitis	1 (1.2%) [1]	0	1 (1.2%) [1]	>0.999	>0.999
Ear Infection	2 (2.3%) [4]	0	0	0.497	0.497
Gastroenteritis Viral	1 (1.2%) [1]	0	0	>0.999	>0.999
Hordeolum	0	0	1 (1.2%) [1]		0.494
Influenza	1 (1.2%) [2]	1 (1.2%) [1]	1 (1.2%) [1]	>0.999	>0.999
Localised Infection	1 (1.2%) [2]	0	0	>0.999	>0.999
Lower Respiratory Tract Infection	0	0	1 (1.2%) [2]		0.494
Nasopharyngitis	2 (2.3%) [4]	4 (4.8%) [9]	6 (7.1%) [8]	0.441	0.166
Pneumonia	0	1 (1.2%) [2]	0	0.494
Rhinitis	0	0	1 (1.2%) [1]		0.494
Upper Respiratory Tract Infection	6 (7.0%) [12]	1 (1.2%) [2]	3 (3.6%) [5]	0.117*	0.496
Urinary Tract Infection	2 (2.3%) [4]	0	1 (1.2%) [1]	0.497	>0.999
Vaginal Mycosis	1 (1.2%) [2]	0	0	>0.999	>0.999
Viral Infection	0	1 (1.2%) [1]	0	0.494
All Injury, Poisoning And Procedural Complications	4 (4.7%) [9]	5 (6.0%) [12]	5 (6.0%) [8]	0.745	0.745
Contusion	1 (1.2%) [1]	1 (1.2%) [3]	2 (2.4%) [3]	>0.999	0.618
Excoriation	2 (2.3%) [3]	1 (1.2%) [2]	1 (1.2%) [1]	>0.999	>0.999
Facial Bones Fracture	0	0	1 (1.2%) [1]		0.494
Fall	1 (1.2%) [2]	2 (2.4%) [2]	1 (1.2%) [1]	0.618	>0.999
Hip Fracture	1 (1.2%) [2]	0	2 (2.4%) [2]	>0.999	0.618
Joint Dislocation	0	1 (1.2%) [1]	0	0.494
Skin Laceration	1 (1.2%) [1]	2 (2.4%) [2]	0	0.618	>0.999
Wound	0	1 (1.2%) [2]	0	0.494
All Investigations	10 (12%) [19]	6 (7.1%) [7]	6 (7.1%) [8]	0.432	0.432
Biopsy	0	0	1 (1.2%) [1]		0.494
Biopsy Prostate	0	0	1 (1.2%) [1]		0.494
Blood Alkaline Phosphatase Increased	1 (1.2%) [1]	0	0	>0.999	>0.999
Blood Cholesterol Increased	0	0	1 (1.2%) [1]		0.494
Blood Creatine Phosphokinase Increased	1 (1.2%) [2]	0	0	>0.999	>0.999
Blood Glucose Increased	0	1 (1.2%) [1]	1 (1.2%) [2]	0.494	0.494
Blood Urine Present	1 (1.2%) [1]	0	0	>0.999	>0.999
Body Temperature Increased	0	1 (1.2%) [1]	0	0.494
Cystoscopy	1 (1.2%) [1]	0	0	>0.999	>0.999
Electrocardiogram St Segment Depression	4 (4.7%) [4]	1 (1.2%) [2]	0	0.368	0.121*
Electrocardiogram T Wave Amplitude Decreased	1 (1.2%) [1]	1 (1.2%) [1]	0	>0.999	>0.999
Electrocardiogram T Wave Inversion	2 (2.3%) [3]	1 (1.2%) [1]	1 (1.2%) [1]	>0.999	>0.999
Heart Rate Increased	1 (1.2%) [2]	0	0	>0.999	>0.999
Heart Rate Irregular	1 (1.2%) [4]	0	0	>0.999	>0.999
Nasal Mucosa Biopsy	0	1 (1.2%) [1]	0	0.494
Weight Decreased	0	0	1 (1.2%) [2]		0.494
All Metabolism And Nutrition Disorders	6 (7.0%) [8]	1 (1.2%) [1]	2 (2.4%) [4]	0.117*	0.278
Decreased Appetite	1 (1.2%) [2]	0	1 (1.2%) [2]	>0.999	>0.999
Dehydration	1 (1.2%) [1]	0	0	>0.999	>0.999
Diabetes Mellitus	1 (1.2%) [1]	0	0	>0.999	>0.999
Food Craving	1 (1.2%) [1]	1 (1.2%) [1]	0	>0.999	>0.999
Hyponatraemia	1 (1.2%) [1]	0	0	>0.999	>0.999
Increased Appetite	1 (1.2%) [2]	0	1 (1.2%) [2]	>0.999	>0.999
All Musculoskeletal And Connective Tissue Disorders	4 (4.7%) [6]	7 (8.3%) [10]	7 (8.3%) [10]	0.367	0.367
Arthralgia	1 (1.2%) [1]	2 (2.4%) [4]	1 (1.2%) [1]	0.618	>0.999
Arthritis	0	0	1 (1.2%) [1]		0.494
Back Pain	1 (1.2%) [2]	1 (1.2%) [1]	3 (3.6%) [4]	>0.999	0.365
Flank Pain	0	0	1 (1.2%) [1]		0.494
Muscle Spasms	0	1 (1.2%) [1]	1 (1.2%) [2]	0.494	0.494
Muscular Weakness	0	1 (1.2%) [2]	0	0.494
Myalgia	0	0	1 (1.2%) [1]		0.494
Pain In Extremity	1 (1.2%) [1]	0	0	>0.999	>0.999
Shoulder Pain	1 (1.2%) [2]	2 (2.4%) [2]	0	0.618	>0.999
All Neoplasms Benign, Malignant And Unspecified (Incl Cysts And Polyps)	0	2 (2.4%) [3]	1 (1.2%) [1]	0.243	0.494
Colon Cancer	0	1 (1.2%) [1]	0	0.494
Malignant Fibrous Histiocytoma	0	1 (1.2%) [2]	0	0.494
Prostate Cancer	0	0	1 (1.2%) [1]		0.494
All Nervous System Disorders	8 (9.3%) [11]	20 (24%) [40]	25 (30%) [41]	0.013*	<0.001*
Amnesia	0	0	1 (1.2%) [2]		0.494
Balance Disorder	0	1 (1.2%) [3]	0	0.494
Burning Sensation	0	0	2 (2.4%) [2]		0.243
Cognitive Disorder	0	0	1 (1.2%) [1]		0.494
Complex Partial Seizures	0	1 (1.2%) [1]	0	0.494
Coordination Abnormal	0	1 (1.2%) [1]	0	0.494
Dizziness	2 (2.3%) [3]	8 (9.5%) [13]	11 (13%) [15]	0.056*	0.009*
Headache	3 (3.5%) [3]	3 (3.6%) [4]	5 (6.0%) [8]	>0.999	0.493
Hemianopia Homonymous	0	1 (1.2%) [1]	0	0.494
Hypersomnia	0	0	1 (1.2%) [1]		0.494
Lethargy	0	1 (1.2%) [1]	1 (1.2%) [1]	0.494	0.494
Paraesthesia	0	0	1 (1.2%) [1]		0.494
Paraesthesia Oral	0	1 (1.2%) [1]	0	0.494
Parkinson's Disease	1 (1.2%) [1]	0	0	>0.999	>0.999
Parosmia	0	0	1 (1.2%) [2]		0.494
Partial Seizures With Secondary Generalisation	0	0	1 (1.2%) [1]		0.494
Psychomotor Hyperactivity	1 (1.2%) [1]	0	0	>0.999	>0.999
Somnolence	2 (2.3%) [3]	3 (3.6%) [5]	1 (1.2%) [1]	0.680	>0.999
Stupor	0	1 (1.2%) [1]	0	0.494
Syncope	0	4 (4.8%) [6]	3 (3.6%) [4]	0.057*	0.118*
Syncope Vasovagal	0	0	1 (1.2%) [1]		0.494
Transient Ischaemic Attack	0	2 (2.4%) [3]	1 (1.2%) [1]	0.243	0.494
Agitation	2 (2.3%) [2]	2 (2.4%) [2]	1 (1.2%) [1]	>0.999	>0.999
All Psychiatric Disorders	10 (12%) [12]	10 (12%) [14]	8 (9.5%) [11]	>0.999	0.804
Anxiety	0	3 (3.6%) [4]	0	0.118*
Completed Suicide	1 (1.2%) [1]	0	0	>0.999	>0.999
Confusional State	2 (2.3%) [2]	3 (3.6%) [3]	1 (1.2%) [1]	0.680	>0.999
Delirium	0	0	1 (1.2%) [1]		0.494
Delusion	1 (1.2%) [1]	0	1 (1.2%) [1]	>0.999	>0.999
Depressed Mood	0	1 (1.2%) [2]	0	0.494
Disorientation	1 (1.2%) [1]	0	0	>0.999	>0.999
Hallucination	0	0	1 (1.2%) [1]		0.494
Hallucination, Visual	0	0	1 (1.2%) [1]		0.494
Insomnia	2 (2.3%) [3]	0	2 (2.4%) [2]	0.497	>0.999
Irritability	1 (1.2%) [2]	1 (1.2%) [1]	0	>0.999	>0.999
Libido Decreased	0	0	1 (1.2%) [1]		0.494
Listless	0	0	1 (1.2%) [1]		0.494
Nightmare	0	0	1 (1.2%) [1]		0.494
Restlessness	0	1 (1.2%) [2]	0	0.494
All Renal And Urinary Disorders	4 (4.7%) [5]	3 (3.6%) [3]	3 (3.6%) [4]	>0.999	>0.999
Calculus Urethral	0	0	1 (1.2%) [1]		0.494
Dysuria	1 (1.2%) [1]	1 (1.2%) [1]	0	>0.999	>0.999
Incontinence	0	1 (1.2%) [1]	0	0.494
Micturition Urgency	1 (1.2%) [1]	1 (1.2%) [1]	1 (1.2%) [2]	>0.999	>0.999
Nephrolithiasis	1 (1.2%) [1]	0	1 (1.2%) [1]	>0.999	>0.999
Pollakiuria	1 (1.2%) [2]	0	0	>0.999	>0.999
All Reproductive System And Breast Disorders	2 (2.3%) [4]	0	1 (1.2%) [1]	0.497	>0.999
Benign Prostatic Hyperplasia	1 (1.2%) [2]	0	1 (1.2%) [1]	>0.999	>0.999
Pelvic Pain	1 (1.2%) [2]	0	0	>0.999	>0.999
All Respiratory, Thoracic And Mediastinal Disorders	8 (9.3%) [12]	9 (11%) [14]	10 (12%) [22]	0.803	0.626
Allergic Granulomatous Angiitis	0	0	1 (1.2%) [1]		0.494
Cough	1 (1.2%) [1]	5 (6.0%) [7]	5 (6.0%) [7]	0.115*	0.115*
Dysphonia	0	1 (1.2%) [1]	0	0.494
Dyspnoea	1 (1.2%) [1]	1 (1.2%) [1]	1 (1.2%) [1]	>0.999	>0.999
Emphysema	1 (1.2%) [1]	0	0	>0.999	>0.999
Epistaxis	0	1 (1.2%) [1]	2 (2.4%) [2]	0.494	0.243
Haemoptysis	1 (1.2%) [2]	0	0	>0.999	>0.999
Nasal Congestion	3 (3.5%) [3]	1 (1.2%) [1]	3 (3.6%) [4]	0.621	>0.999
Pharyngeal Erythema	0	0	1 (1.2%) [2]		0.494
Pharyngolaryngeal Pain	0	1 (1.2%) [1]	1 (1.2%) [1]	0.494	0.494
Postnasal Drip	1 (1.2%) [2]	0	0	>0.999	>0.999
Productive Cough	0	0	1 (1.2%) [1]		0.494
Rales	1 (1.2%) [2]	0	0	>0.999	>0.999
Respiratory Tract Congestion	0	0	1 (1.2%) [1]		0.494
Rhinorrhoea	0	1 (1.2%) [2]	1 (1.2%) [2]	0.494	0.494
Actinic Keratosis	0	0	1 (1.2%) [1]		0.494
All Skin And Subcutaneous Tissue Disorders	20 (23%) [45]	39 (46%) [111]	40 (48%) [104]	0.002*	0.001*
Alopecia	1 (1.2%) [1]	0	0	>0.999	>0.999
Blister	0	5 (6.0%) [8]	1 (1.2%) [2]	0.028*	0.494
Cold Sweat	1 (1.2%) [3]	0	0	>0.999	>0.999
Dermatitis Contact	0	1 (1.2%) [2]	0	0.494
Drug Eruption	1 (1.2%) [1]	0	0	>0.999	>0.999
Erythema	8 (9.3%) [12]	14 (17%) [22]	14 (17%) [22]	0.175	0.175
Hyperhidrosis	2 (2.3%) [2]	4 (4.8%) [5]	8 (9.5%) [10]	0.441	0.056*
Pruritus	8 (9.3%) [11]	21 (25%) [31]	26 (31%) [38]	0.008*	<0.001*
Pruritus Generalised	0	1 (1.2%) [4]	1 (1.2%) [1]	0.494	0.494
Rash	5 (5.8%) [9]	13 (15%) [18]	9 (11%) [15]	0.048*	0.277
Rash Erythematous	0	1 (1.2%) [1]	0	0.494
Rash Maculo-Papular	0	0	1 (1.2%) [1]		0.494
Rash Pruritic	0	1 (1.2%) [2]	2 (2.4%) [3]	0.494	0.243
Skin Exfoliation	0	1 (1.2%) [2]	0	0.494
Skin Irritation	3 (3.5%) [4]	6 (7.1%) [13]	5 (6.0%) [8]	0.326	0.493
Skin Odour Abnormal	0	0	1 (1.2%) [1]		0.494
Skin Ulcer	1 (1.2%) [2]	0	0	>0.999	>0.999
Urticaria	0	1 (1.2%) [3]	1 (1.2%) [2]	0.494	0.494
Alcohol Use	0	0	1 (1.2%) [1]		0.494
All Social Circumstances	0	0	1 (1.2%) [1]		0.494
Acrochordon Excision	0	0	1 (1.2%) [1]		0.494
All Surgical And Medical Procedures	2 (2.3%) [2]	1 (1.2%) [1]	2 (2.4%) [2]	>0.999	>0.999
Cataract Operation	1 (1.2%) [1]	1 (1.2%) [1]	0	>0.999	>0.999
Eye Laser Surgery	1 (1.2%) [1]	0	0	>0.999	>0.999
Skin Lesion Excision	0	0	1 (1.2%) [1]		0.494
All Vascular Disorders	3 (3.5%) [7]	3 (3.6%) [3]	1 (1.2%) [1]	>0.999	0.621
Hot Flush	0	1 (1.2%) [1]	0	0.494
Hypertension	1 (1.2%) [2]	1 (1.2%) [1]	0	>0.999	>0.999
Hypotension	2 (2.3%) [3]	1 (1.2%) [1]	0	>0.999	0.497
Orthostatic Hypotension	1 (1.2%) [2]	0	0	>0.999	>0.999
Wound Haemorrhage	0	0	1 (1.2%) [1]		0.494
1 Treatment emergent events are defined as events which start on or after the start of treatment. Adverse events are coded using MedDRA. Percentages are based on the number of subjects in the safety population within each treatment group. [AE] represents the total number of times an event was recorded.
2 P-values are based on Fisher’s Exact test for the comparison of placebo versus each active treatment group. An asterisk is appended to p-values that are less than 0.15.

Caution

The order within each AEBODSYS (Body System or Organ Class) is currently alphabetical, but we can apply different sorting methods if desired.

Table 14-5-02

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary, gtreg)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ ADSL
adsl_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt') 

adsl <- adsl_orig %>% 
   filter(SAFFL == 'Y') %>% 
   rename(TRTA = TRT01A) %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose')))

# READ ADAE
adae_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adae.xpt') 

adae <- adae_orig %>% 
   filter(SAFFL == 'Y' & TRTEMFL == 'Y' & AESER == 'Y') %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(across(c(AEBODSYS, AETERM), ~str_to_title(.x))) 

# COMPLETE DATA AE SOC
cmpl_ <- .complete_ae_data(
   data = adae,
   id   = "USUBJID",
   ae   = "AETERM",
   soc  = "AEBODSYS",
   strata = 'TRTA',
   id_df = adsl) %>% 
   mutate(across(c(`..soc..`,`..ae..`), ~ factor(.x)) ) 

# COMPLETE DATA TT
cmpl_tt <- .complete_ae_data(
   data = adae %>% mutate(AETERM   = 'Any Body System',
                          AEBODSYS = 'Any Body System'),
   id   = "USUBJID",
   ae   = "AETERM",
   soc  = "AEBODSYS",
   strata = 'TRTA',
   id_df = adsl) %>% 
   mutate(across(c(`..soc..`,`..ae..`), ~ factor(.x)) ) 

# LONG w/ WITH TOTALS
cmpl_l <- 
   bind_rows(cmpl_,
             cmpl_tt) %>% 
   pivot_longer(cols = c(`..soc..`,`..ae..`)) %>% 
   mutate(ae = ifelse(name == '..soc..', str_glue('All {soc}'), str_glue('{ae}')) ) %>% 
   nest_by(name, soc, ae) %>% 
   filter(! (name == '..ae..' & ae == 'Any Body System')) %>% 
   arrange(soc, ae)

# TABLE
cmpl_tbl <- cmpl_l %>% 
   mutate(
      t_n  = list(
         tbl_summary(
            data = data,
            by = strata,
            include = `by`,
            value = list(`by` ~ 'Overall'),
            label = list(`by` ~ ae),
            statistic = list(everything() ~ '[{n}]')
         ) 
      ),
      t_3g = list( 
         tbl_summary(
            data = data %>% filter(value == TRUE),
            by = strata,
            value = list(`by` ~ 'Overall'),
            include = `by`,
            label = list(`by` ~ ae),
         ) 
      ),
      t_PvsL = list(
         tbl_summary(
            data = data %>% 
               filter(value == TRUE) %>% 
               filter(strata %in% c('Placebo','Low Dose')) %>% 
               droplevels(),
            by = strata,
            value = list(`by` ~ 'Overall'),
            include = `by`,
            label = list(`by` ~ ae),
         ) %>% 
            add_p(
               test = list(all_categorical() ~ 'fisher.test'),
               pvalue_fun = function(x) style_pvalue(x, digits = 3) 
            ) %>% 
            add_significance_stars(
               thresholds = c(0.15),
               hide_p = FALSE,
               pattern = "{p.value}{stars}"
            ) 
      ),
      t_PvsH = list(
         tbl_summary(
            data = data %>% 
               filter(value == TRUE) %>% 
               filter(strata %in% c('Placebo','High Dose')) %>% 
               droplevels(),
            by = strata,
            value = list(`by` ~ 'Overall'),
            include = `by`,
            label = list(`by` ~ ae),
         ) %>% 
            add_p(
               test = list(all_categorical() ~ 'fisher.test'),
               pvalue_fun = function(x) style_pvalue(x, digits = 3) 
            ) %>% 
            add_significance_stars(
               thresholds = c(0.15),
               hide_p = FALSE,
               pattern = "{p.value}{stars}"
            ) 
      ),
      t = list(
         tbl_merge( tbls = list(t_3g,
                                t_n,
                                t_PvsL %>% modify_column_hide(all_stat_cols()) ,
                                t_PvsH %>% modify_column_hide(all_stat_cols()) ),
                    tab_spanner = FALSE)
      )
      
   )

# STACK
with(cmpl_tbl,
     tbl_stack(tbls = t,
               quiet = TRUE) ) %>%
   modify_column_merge( "{stat_1_1} {stat_1_2}") %>% 
   modify_column_merge( "{stat_2_1} {stat_2_2}") %>% 
   modify_column_merge( "{stat_3_1} {stat_3_2}") %>% 
   modify_header(
      label = '',
      c(stat_1_1, 
        stat_2_1,
        stat_3_1) ~ '**{level}<br>(N = {n})**',
      p.value_3 ~ '**Placebo v<br>Low Dose**',
      p.value_4 ~ '**Placebo v<br>High Dose**') %>% 
   modify_spanning_header(
      c(stat_2_1, stat_3_1) ~ "**Xanomeline**",
      c(p.value_3, p.value_4) ~ "**Fisher's Exact p-values**") %>% 
   modify_table_body(
      ~ .x %>%
         mutate(stat_3_2_n =  str_extract(stat_3_2 , "\\d+") %>% as.numeric()) %>% 
         mutate(across(starts_with('stat_'), ~str_replace_all(.x, "\\[0\\]", ""))) %>%
         mutate(across(starts_with('stat_'), ~str_replace(., "0 \\(0%\\)", "0"))) %>% 
         slice(c(n(), 1:(n()-1)) ) ) %>% 
   modify_table_body(
      ~ .x %>%
         mutate(row_type = ifelse( label %in% label[!str_detect(label, "^All")], 'level','label') ) ) %>% 
   modify_column_alignment(columns = all_stat_cols(), align = "left") %>% 
   bold_labels() %>% 
   modify_footnote(
      all_stat_cols() ~ 
         "Treatment emergent events are defined as events which start on or after the start of treatment.<br>
      Adverse events are coded using MedDRA.<br>
      Percentages are based on the number of subjects in the safety population within each treatment group.<br>
      [AE] represents the total number of times an event was recorded.",
      starts_with('p.value') ~ 
         "P-values are based on Fisher's Exact test for the comparison of placebo versus each active treatment
      group. An asterisk is appended to p-values that are less than 0.15.") %>% 
   modify_caption('**Table 14-5.02<br>Incidence of Treatment Emergent Serious Adverse Events by Treatment Group**')

Table 14-5.02
Incidence of Treatment Emergent Serious Adverse Events by Treatment Group

	Placebo (N = 86)1	Xanomeline		Fisher’s Exact p-values
		Low Dose (N = 84)1	High Dose (N = 84)1	Placebo v Low Dose2	Placebo v High Dose2
All Any Body System	0	1 (1.2%) [1]	2 (2.4%) [2]	0.494	0.243
All Nervous System Disorders	0	1 (1.2%) [1]	2 (2.4%) [2]	0.494	0.243
Partial Seizures With Secondary Generalisation	0	0	1 (1.2%) [1]		0.494
Syncope	0	1 (1.2%) [1]	1 (1.2%) [1]	0.494	0.494
1 Treatment emergent events are defined as events which start on or after the start of treatment. Adverse events are coded using MedDRA. Percentages are based on the number of subjects in the safety population within each treatment group. [AE] represents the total number of times an event was recorded.
2 P-values are based on Fisher’s Exact test for the comparison of placebo versus each active treatment group. An asterisk is appended to p-values that are less than 0.15.

Table 14-6

Table 14-6-01

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adlbc <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbc.xpt') %>% 
   filter(SAFFL == 'Y' & (AVISITN != 99 | (AVISITN == 99 & AENTMTFL=='Y'))) 

adlbh <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbh.xpt') %>% 
   filter(SAFFL == 'Y' & (AVISITN != 99 | (AVISITN == 99 & AENTMTFL=='Y')) &
             !(PARAM %in% c('Anisocytes', 'Poikilocytes', 'Microcytes', 'Macrocytes', 'Polychromasia')))

# BIND
adlb <- bind_rows(
   CHEMISTRY = adlbc,
   HEMATOLOGY = adlbh,
   .id = 'ORIG') %>% 
   select(ORIG, PARCAT1, PARAM, PARAMCD, USUBJID , TRTA, AVISIT, AVISITN, AVAL, CHG) %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(AVISIT = str_trim(AVISIT)) %>% 
   mutate(AVISIT = factor(AVISIT) %>% 
             fct_reorder(AVISITN) ) 

# LONG
adlb_l <- adlb %>% 
   pivot_longer(cols = c(AVAL, CHG),
                names_to = 'VARIABLE',
                values_to = 'VALUE')

# NEST
adlb_n <- adlb_l %>% 
   nest_by(ORIG, PARCAT1, PARAM, PARAMCD, AVISIT, AVISITN, VARIABLE) 

# FUNCTION
add_by_n <- function(data, variable, by, tbl, ...) {
   data %>%
      select(all_of(c(variable, by))) %>%
      dplyr::group_by(.data[[by]]) %>%
      dplyr::summarise_all(~sum(!is.na(.))) %>%
      rlang::set_names(c("by", "variable")) %>%
      dplyr::left_join(
         tbl$df_by %>% select(by, by_col),
         by = "by"
      ) %>%
      mutate(
         by_col = paste0("add_n_", by_col),
         variable = style_number(variable)
      ) %>%
      select(-by) %>%
      tidyr::pivot_wider(names_from = by_col, 
                         values_from = variable)
}

# TABLE
adlb_t <- adlb_n %>% 
   filter(!str_starts(PARAMCD, "_")) %>% 
   mutate(t = case_when(
      VARIABLE == 'AVAL' ~ list( 
         tbl_summary(
            data = data,
            by = TRTA,
            include = VALUE,
            type = list(VALUE ~ 'continuous'),
            label = list(VALUE ~ str_glue('{AVISIT}')),
            statistic = list(VALUE ~ '{mean} ({sd})'),
            digits = list(VALUE ~ c(1, 2)),
            missing = 'no') %>% 
            add_stat(
               fns = everything() ~ add_by_n) %>% 
            modify_header(starts_with("add_n_stat") ~ "**N**") %>%
            modify_table_body(
               ~ .x %>%
                  dplyr::relocate(add_n_stat_1, .before = stat_1) %>%
                  dplyr::relocate(add_n_stat_2, .before = stat_2) %>% 
                  dplyr::relocate(add_n_stat_3, .before = stat_3)) ),
      .default =  list(
         tbl_summary(
            data = data,
            by = TRTA,
            include = VALUE,
            type = list(VALUE ~ 'continuous'),
            label = list(VALUE ~ str_glue('{AVISIT}')),
            statistic = list(VALUE ~ '{mean} ({sd})'),
            digits = list(VALUE ~ c(1, 2)),
            missing = 'no')
      ))
   )

# MERGE
adlb_t2 <- adlb_t %>% 
   select(-data) %>% 
   pivot_wider(
      names_from = VARIABLE,
      values_from = t) %>% 
   rowwise() %>% 
   mutate(
      BOTH = list(
         tbl_merge( tbls = list(AVAL, CHG ),
                    tab_spanner = FALSE) )
   )

# STACK & PRINT TABLE
#+ results = 'asis'

for( i in unique(adlb_t2$ORIG) ) {
   
   with(adlb_t2 %>% filter(ORIG == i),
        tbl_stack(BOTH,
                  group_header = str_glue('{PARAM}'),
                  quiet = TRUE) ) %>%
      modify_table_body(
         ~.x %>% 
            select(
               tbl_id1,
               groupname_col, label,
               add_n_stat_1_1, stat_1_1, stat_1_2,
               add_n_stat_2_1, stat_2_1, stat_2_2,
               add_n_stat_3_1, stat_3_1, stat_3_2,
               everything() ) %>% 
            mutate(across(all_stat_cols(), ~str_replace_all(., "NA \\(NA\\)", "")))
      ) %>% 
      modify_header(
         label ~ '',
         c(stat_1_1, stat_2_1, stat_3_1) ~ '**Mean (SD)**',
         c(stat_1_2, stat_2_2, stat_3_2) ~ '**Change<br>from Bsln<br>Mean (SD)**'
      ) %>% 
      modify_spanning_header(
         c(add_n_stat_1_1, stat_1_1, stat_1_2) ~ '**Placebo**',
         c(add_n_stat_2_1, stat_2_1, stat_2_2) ~ '**Xanomeline Low**',
         c(add_n_stat_3_1, stat_3_1, stat_3_2) ~ '**Xanomeline High**'
      ) %>% 
      modify_footnote(all_stat_cols() ~ NA) %>% 
      modify_table_styling(
         columns = label,
         rows =  variable == 'VALUE' & label == 'End of Treatment',
         footnote = "Last observed value while on treatment (prior to or at Week 24)"
      ) %>% 
      modify_caption(
         str_glue('**Table 14-6.01 {i}<br>Summary Statistics for Continuous Laboratory Values**')
      ) %>% 
      bstfun::bold_italicize_group_labels(bold = TRUE) %>% 
      knitr::knit_print() %>% 
      cat()
   
}

Table 14-6.01 CHEMISTRY
Summary Statistics for Continuous Laboratory Values

	Placebo			Xanomeline Low			Xanomeline High
	N	Mean (SD)	Change from Bsln Mean (SD)	N	Mean (SD)	Change from Bsln Mean (SD)	N	Mean (SD)	Change from Bsln Mean (SD)
Alanine Aminotransferase (U/L)
Baseline	86	17.6 (9.22)		82	18.0 (8.72)		84	19.2 (10.05)
Week 2	83	18.0 (12.53)	0.2 (7.90)	80	20.9 (10.55)	2.8 (8.18)	78	21.0 (8.87)	1.6 (6.83)
Week 4	79	18.7 (12.91)	0.7 (8.66)	72	17.5 (7.66)	-0.7 (5.08)	72	21.3 (9.51)	2.1 (7.08)
Week 6	73	17.0 (9.92)	-0.3 (7.78)	62	17.0 (7.98)	-0.8 (4.87)	66	21.2 (9.49)	1.6 (6.11)
Week 8	72	16.7 (9.34)	-1.1 (5.05)	60	17.6 (7.86)	0.2 (5.45)	56	22.8 (17.49)	3.2 (17.24)
Week 12	67	18.0 (9.16)	0.0 (8.07)	51	18.5 (12.68)	0.1 (9.27)	50	21.0 (10.18)	0.7 (8.74)
Week 16	68	17.1 (7.39)	-0.8 (7.94)	42	17.3 (7.51)	0.7 (5.80)	37	19.6 (7.61)	-0.3 (7.58)
Week 20	65	16.1 (6.56)	-1.9 (7.49)	30	16.7 (6.33)	0.9 (4.77)	31	19.6 (6.82)	-0.3 (8.63)
Week 24	57	17.9 (15.61)	-0.3 (16.62)	26	18.2 (9.17)	1.6 (5.66)	30	21.0 (8.70)	0.2 (8.25)
Week 26	57	16.0 (5.98)	-2.1 (7.70)	25	17.8 (9.51)	1.5 (6.26)	27	18.9 (7.02)	-2.0 (7.01)
End of Treatment1	84	18.1 (16.74)	0.4 (15.40)	82	18.3 (8.26)	0.3 (7.25)	80	19.5 (7.44)	0.1 (8.08)
Albumin (g/L)
Baseline	86	39.8 (2.81)		82	39.8 (2.56)		84	40.3 (2.84)
Week 2	83	38.9 (3.11)	-1.0 (2.49)	80	38.7 (3.17)	-1.1 (2.71)	78	38.9 (2.76)	-1.4 (2.59)
Week 4	79	38.8 (3.29)	-1.0 (2.69)	72	38.6 (2.80)	-1.2 (2.66)	72	39.1 (3.05)	-1.3 (2.70)
Week 6	73	39.1 (2.56)	-1.0 (2.25)	62	38.4 (2.60)	-1.2 (2.49)	66	39.5 (2.76)	-1.0 (2.60)
Week 8	72	39.8 (3.51)	-0.4 (2.70)	60	39.1 (2.93)	-0.5 (2.73)	56	39.8 (2.33)	-0.9 (2.21)
Week 12	67	39.5 (3.49)	-0.5 (2.31)	51	38.9 (2.18)	-0.9 (2.19)	50	39.8 (2.45)	-0.6 (2.77)
Week 16	68	40.4 (3.02)	0.4 (2.45)	42	39.1 (2.98)	-0.4 (2.78)	37	39.9 (1.92)	-0.7 (2.76)
Week 20	65	39.6 (3.47)	-0.5 (2.86)	30	38.6 (2.66)	-1.2 (2.55)	31	39.6 (1.85)	-1.4 (2.86)
Week 24	57	39.7 (3.34)	-0.2 (2.88)	26	40.4 (2.52)	0.4 (2.40)	30	40.5 (2.10)	-0.5 (2.65)
Week 26	57	39.8 (3.02)	0.0 (2.26)	25	39.2 (2.39)	-1.0 (2.85)	27	40.0 (2.26)	-1.2 (2.78)
End of Treatment1	84	39.6 (3.32)	-0.2 (2.69)	82	39.2 (2.97)	-0.5 (2.64)	80	39.8 (2.48)	-0.6 (2.64)
Alkaline Phosphatase (U/L)
Baseline	86	77.7 (58.11)		81	73.3 (20.72)		83	71.0 (38.85)
Week 2	84	77.7 (69.50)	0.1 (15.51)	80	73.0 (21.87)	0.3 (10.25)	78	72.2 (40.22)	-0.1 (8.09)
Week 4	82	78.0 (69.43)	0.4 (14.16)	72	73.2 (23.17)	-0.3 (11.12)	72	71.3 (40.65)	-0.7 (6.88)
Week 6	75	68.4 (21.53)	-0.4 (10.40)	64	72.7 (23.15)	-1.0 (10.83)	67	71.5 (43.40)	-0.2 (12.65)
Week 8	73	70.0 (27.15)	0.0 (10.67)	60	72.9 (23.84)	0.5 (14.09)	56	74.0 (45.60)	0.8 (13.54)
Week 12	67	72.1 (32.73)	1.8 (17.18)	52	69.5 (20.55)	-1.5 (9.41)	50	71.9 (46.65)	-1.7 (8.53)
Week 16	68	70.6 (29.49)	0.3 (16.98)	42	69.5 (19.43)	-1.7 (8.61)	37	74.7 (54.78)	-1.4 (8.29)
Week 20	65	72.6 (37.41)	1.9 (25.77)	31	70.6 (22.20)	-2.2 (9.90)	31	73.5 (55.23)	-2.7 (8.35)
Week 24	56	80.6 (68.06)	10.1 (58.71)	27	72.0 (21.80)	-0.4 (8.93)	30	64.4 (17.63)	-2.4 (7.44)
Week 26	57	81.0 (79.33)	10.1 (72.41)	25	68.6 (21.08)	-4.1 (10.74)	27	61.9 (16.60)	-3.9 (7.65)
End of Treatment1	84	84.6 (84.86)	7.1 (49.37)	82	71.6 (23.80)	-1.1 (13.09)	80	70.3 (37.91)	-1.6 (12.00)
Aspartate Aminotransferase (U/L)
Baseline	86	23.2 (7.50)		82	23.4 (8.24)		84	23.1 (6.61)
Week 2	83	23.6 (12.35)	0.2 (8.96)	80	24.7 (8.06)	1.6 (6.53)	78	23.4 (5.20)	0.4 (5.63)
Week 4	79	23.9 (14.93)	0.5 (11.33)	72	22.3 (6.75)	-1.4 (6.40)	72	23.8 (5.85)	0.6 (5.75)
Week 6	73	22.0 (6.40)	-0.9 (6.38)	62	22.1 (6.11)	-0.4 (4.19)	66	24.1 (7.84)	0.4 (6.10)
Week 8	72	22.3 (7.05)	-1.1 (4.84)	60	22.7 (5.95)	0.3 (4.02)	56	25.7 (13.33)	1.6 (13.98)
Week 12	67	22.8 (7.64)	-0.6 (7.06)	51	24.2 (15.87)	1.5 (12.39)	50	23.3 (6.11)	-1.2 (6.26)
Week 16	68	22.8 (6.42)	-0.6 (6.43)	42	22.4 (10.34)	0.6 (7.62)	37	23.1 (5.78)	-0.7 (4.17)
Week 20	65	21.9 (5.90)	-1.6 (6.07)	30	20.7 (5.74)	0.4 (4.60)	31	24.0 (6.90)	-0.3 (6.18)
Week 24	57	25.2 (21.02)	1.2 (20.43)	26	22.4 (10.78)	2.1 (6.58)	30	24.4 (7.29)	-0.2 (5.48)
Week 26	57	21.5 (6.99)	-2.5 (7.29)	25	22.1 (11.85)	1.4 (7.69)	27	21.6 (5.71)	-3.1 (4.17)
End of Treatment1	84	25.1 (21.33)	1.8 (19.03)	82	23.2 (8.21)	-0.3 (8.06)	80	22.7 (6.13)	-0.4 (6.36)
Bilirubin (umol/L)
Baseline	86	9.7 (3.96)		82	9.4 (4.01)		84	11.0 (5.35)
Week 2	83	10.8 (12.26)	1.1 (10.29)	79	9.4 (4.16)	0.0 (3.08)	78	10.4 (3.94)	-0.6 (3.62)
Week 4	79	11.1 (13.57)	1.4 (11.65)	71	9.2 (3.84)	-0.3 (2.70)	72	10.9 (5.62)	-0.3 (3.07)
Week 6	73	9.6 (3.78)	0.1 (2.88)	62	9.5 (4.03)	0.0 (2.99)	66	10.9 (4.89)	-0.6 (3.18)
Week 8	72	9.4 (3.89)	-0.4 (3.43)	60	9.6 (4.72)	0.5 (3.11)	56	10.8 (5.21)	-0.7 (3.21)
Week 12	67	9.5 (3.56)	-0.1 (2.83)	51	8.8 (4.15)	0.1 (2.64)	50	11.5 (6.16)	-0.3 (4.78)
Week 16	68	10.0 (3.63)	0.2 (2.68)	42	8.8 (3.91)	0.4 (2.14)	37	11.6 (4.89)	-0.7 (4.28)
Week 20	65	10.1 (5.19)	0.3 (3.58)	30	8.8 (4.51)	0.3 (2.82)	31	11.8 (8.69)	-0.2 (6.41)
Week 24	55	9.4 (3.39)	0.1 (2.75)	25	10.1 (4.44)	1.1 (2.84)	30	12.3 (6.52)	-0.6 (2.89)
Week 26	57	10.0 (4.73)	0.4 (3.57)	25	10.2 (6.21)	1.4 (3.44)	27	12.2 (6.82)	-0.6 (5.31)
End of Treatment1	82	11.2 (13.42)	1.4 (11.28)	80	9.8 (4.29)	0.5 (3.08)	80	11.1 (5.36)	0.0 (3.16)
Blood Urea Nitrogen (mmol/L)
Baseline	86	5.5 (1.39)		82	6.4 (1.97)		84	5.8 (1.88)
Week 2	84	5.8 (1.51)	0.3 (1.17)	80	6.6 (1.82)	0.3 (1.53)	78	6.0 (2.06)	0.3 (1.36)
Week 4	82	5.9 (1.47)	0.3 (1.11)	72	6.4 (1.63)	0.0 (1.37)	72	5.9 (1.87)	0.3 (1.30)
Week 6	75	6.1 (1.42)	0.5 (1.16)	64	6.4 (1.76)	-0.1 (1.32)	67	6.0 (2.06)	0.3 (1.43)
Week 8	73	5.6 (1.58)	0.0 (1.22)	60	6.3 (1.92)	0.0 (0.95)	56	5.9 (1.93)	0.1 (1.25)
Week 12	67	5.9 (1.61)	0.2 (1.24)	52	6.2 (1.49)	0.0 (1.45)	50	5.9 (1.76)	0.1 (1.50)
Week 16	68	5.8 (1.60)	0.2 (1.28)	42	6.0 (1.62)	-0.2 (1.25)	37	5.7 (1.92)	0.2 (1.29)
Week 20	66	5.8 (1.53)	0.1 (1.28)	31	5.6 (1.57)	-0.6 (1.39)	31	5.9 (2.25)	0.2 (1.50)
Week 24	57	5.9 (1.32)	0.3 (1.24)	27	6.0 (2.07)	0.0 (1.37)	30	5.3 (1.92)	-0.3 (1.37)
Week 26	57	6.1 (1.62)	0.4 (1.36)	25	5.7 (1.93)	-0.3 (1.14)	27	5.9 (2.47)	0.3 (1.40)
End of Treatment1	84	5.9 (1.38)	0.3 (1.20)	82	6.6 (1.95)	0.2 (1.37)	80	5.7 (1.97)	0.0 (1.57)
Calcium (mmol/L)
Baseline	86	2.3 (0.09)		82	2.3 (0.11)		84	2.3 (0.10)
Week 2	84	2.3 (0.09)	0.0 (0.10)	80	2.3 (0.12)	0.0 (0.10)	78	2.3 (0.11)	0.0 (0.11)
Week 4	82	2.3 (0.09)	0.0 (0.09)	72	2.3 (0.10)	0.0 (0.08)	72	2.3 (0.10)	0.0 (0.10)
Week 6	75	2.3 (0.09)	0.0 (0.10)	64	2.3 (0.10)	0.0 (0.08)	67	2.3 (0.09)	0.0 (0.10)
Week 8	73	2.3 (0.09)	0.0 (0.09)	60	2.3 (0.12)	0.0 (0.09)	56	2.3 (0.12)	0.0 (0.11)
Week 12	67	2.3 (0.09)	0.0 (0.08)	52	2.3 (0.10)	0.0 (0.07)	50	2.3 (0.09)	0.0 (0.09)
Week 16	68	2.3 (0.10)	0.0 (0.11)	42	2.3 (0.11)	0.0 (0.08)	37	2.3 (0.11)	0.0 (0.12)
Week 20	66	2.3 (0.09)	0.0 (0.09)	31	2.3 (0.10)	0.0 (0.09)	31	2.3 (0.08)	0.0 (0.09)
Week 24	57	2.3 (0.09)	-0.1 (0.10)	27	2.3 (0.11)	0.0 (0.12)	30	2.3 (0.10)	-0.1 (0.12)
Week 26	57	2.3 (0.10)	0.0 (0.10)	25	2.3 (0.10)	0.0 (0.08)	27	2.3 (0.09)	0.0 (0.10)
End of Treatment1	84	2.3 (0.09)	0.0 (0.10)	82	2.3 (0.10)	0.0 (0.10)	80	2.3 (0.10)	0.0 (0.11)
Chloride (mmol/L)
Baseline	86	105.7 (3.19)		82	105.8 (3.25)		83	105.4 (3.33)
Week 2	84	106.0 (3.12)	0.3 (3.40)	80	105.6 (3.12)	0.0 (3.44)	77	104.4 (3.49)	-0.8 (3.88)
Week 4	82	105.6 (3.53)	-0.1 (4.24)	72	105.5 (2.88)	-0.3 (3.72)	72	105.0 (3.46)	-0.4 (3.94)
Week 6	75	105.5 (3.14)	-0.1 (3.61)	64	106.3 (3.07)	0.6 (3.78)	67	105.3 (3.09)	-0.3 (3.47)
Week 8	73	106.0 (3.34)	0.3 (3.52)	60	105.6 (3.01)	-0.2 (3.47)	56	105.2 (3.57)	-0.1 (3.71)
Week 12	67	105.3 (2.93)	-0.3 (3.75)	52	105.7 (3.21)	-0.4 (3.63)	49	105.2 (2.40)	-0.1 (2.64)
Week 16	68	105.5 (3.16)	0.0 (3.57)	42	106.2 (2.93)	0.2 (3.17)	37	105.9 (2.66)	0.6 (2.54)
Week 20	65	106.0 (3.68)	0.6 (4.13)	31	106.3 (2.75)	-0.1 (2.41)	31	105.4 (3.03)	-0.2 (3.69)
Week 24	57	105.5 (3.14)	-0.1 (3.82)	27	105.3 (2.52)	-1.2 (2.95)	30	105.3 (3.21)	-0.1 (3.06)
Week 26	57	106.2 (2.58)	0.8 (3.33)	25	105.7 (2.35)	-0.5 (2.72)	27	105.7 (3.44)	0.4 (2.92)
End of Treatment1	84	105.6 (3.42)	-0.1 (3.86)	82	105.5 (3.37)	-0.1 (3.31)	80	105.0 (3.20)	-0.5 (3.17)
Cholesterol (mmol/L)
Baseline	86	5.8 (1.07)		82	5.7 (1.00)		84	5.8 (1.02)
Week 2	84	5.6 (1.02)	-0.1 (0.54)	80	5.6 (0.92)	-0.1 (0.50)	78	5.6 (0.93)	-0.2 (0.53)
Week 4	82	5.5 (0.94)	-0.2 (0.57)	72	5.5 (0.97)	-0.2 (0.49)	72	5.5 (1.00)	-0.3 (0.51)
Week 6	75	5.6 (0.95)	-0.1 (0.67)	64	5.4 (0.95)	-0.2 (0.55)	67	5.5 (0.91)	-0.3 (0.64)
Week 8	73	5.5 (1.02)	-0.2 (0.71)	60	5.5 (0.96)	-0.2 (0.49)	56	5.5 (0.94)	-0.3 (0.54)
Week 12	67	5.5 (0.92)	-0.2 (0.57)	52	5.3 (0.90)	-0.3 (0.47)	50	5.4 (0.91)	-0.3 (0.63)
Week 16	68	5.6 (0.98)	-0.1 (0.58)	42	5.3 (1.00)	-0.3 (0.48)	37	5.4 (0.94)	-0.2 (0.55)
Week 20	66	5.5 (0.94)	-0.2 (0.68)	31	5.2 (0.86)	-0.4 (0.46)	31	5.3 (0.85)	-0.3 (0.49)
Week 24	57	5.5 (1.01)	-0.3 (0.68)	27	5.4 (0.94)	-0.2 (0.65)	30	5.3 (0.89)	-0.3 (0.50)
Week 26	57	5.5 (0.94)	-0.3 (0.62)	25	5.2 (0.76)	-0.4 (0.64)	27	5.4 (0.90)	-0.2 (0.81)
End of Treatment1	84	5.5 (1.02)	-0.3 (0.76)	82	5.4 (0.96)	-0.3 (0.56)	80	5.4 (0.89)	-0.4 (0.60)
Creatine Kinase (U/L)
Baseline	86	86.9 (43.71)		82	100.6 (68.87)		84	104.0 (71.75)
Week 2	83	90.3 (66.11)	2.4 (57.72)	80	106.1 (83.92)	6.0 (84.97)	78	93.8 (53.19)	-12.0 (52.92)
Week 4	79	96.9 (124.45)	8.3 (116.54)	72	93.2 (51.67)	-7.8 (55.54)	72	100.6 (58.39)	-4.8 (53.80)
Week 6	73	88.0 (42.36)	-0.3 (40.55)	62	89.0 (44.29)	-13.7 (59.25)	66	123.5 (227.36)	17.8 (187.17)
Week 8	72	93.8 (45.84)	1.2 (46.26)	60	90.0 (39.16)	-8.3 (61.09)	56	91.8 (52.25)	-9.3 (59.25)
Week 12	67	101.8 (78.66)	8.9 (67.77)	51	94.7 (53.59)	-7.2 (65.59)	50	96.6 (60.55)	-2.8 (66.47)
Week 16	68	104.9 (66.37)	12.7 (52.13)	42	97.5 (56.39)	-6.9 (72.02)	37	86.1 (47.70)	-1.6 (24.32)
Week 20	65	93.8 (46.56)	1.0 (42.07)	30	111.6 (137.04)	21.5 (136.75)	31	97.5 (68.57)	6.0 (28.57)
Week 24	57	127.4 (207.98)	32.6 (200.00)	26	83.6 (38.60)	0.8 (21.99)	30	90.9 (53.97)	-1.1 (23.40)
Week 26	57	94.1 (51.03)	2.0 (46.93)	25	69.9 (23.70)	-15.0 (17.45)	27	93.0 (59.94)	1.4 (43.90)
End of Treatment1	84	112.6 (173.34)	24.6 (165.91)	82	98.2 (61.12)	-1.9 (67.09)	80	95.1 (56.32)	-9.9 (60.94)
Creatinine (umol/L)
Baseline	86	97.7 (17.78)		82	103.5 (20.01)		84	103.7 (19.37)
Week 2	84	99.0 (17.51)	1.4 (8.06)	80	106.2 (21.26)	2.3 (10.51)	78	106.3 (21.18)	2.8 (9.50)
Week 4	82	98.6 (18.56)	1.1 (11.40)	72	105.1 (19.83)	2.0 (8.13)	72	105.8 (20.66)	2.6 (10.43)
Week 6	75	101.7 (18.46)	3.3 (12.14)	64	104.4 (19.85)	1.7 (9.31)	67	105.8 (20.87)	3.2 (9.94)
Week 8	73	99.1 (16.07)	0.0 (9.88)	60	104.3 (21.27)	1.8 (7.54)	56	106.1 (22.24)	3.2 (9.14)
Week 12	67	101.5 (16.53)	2.4 (9.68)	52	101.8 (17.72)	1.8 (7.99)	50	108.2 (20.48)	5.8 (10.95)
Week 16	68	100.4 (16.43)	1.2 (10.57)	42	98.5 (16.63)	-1.3 (9.29)	37	102.5 (17.86)	2.2 (9.18)
Week 20	66	100.3 (17.92)	1.3 (11.31)	31	97.2 (17.23)	-1.8 (8.17)	31	103.5 (19.66)	3.1 (11.76)
Week 24	57	99.3 (15.85)	0.8 (11.37)	27	99.2 (16.14)	1.7 (7.08)	30	100.2 (19.61)	-0.6 (10.36)
Week 26	57	100.0 (18.46)	2.0 (9.30)	25	99.7 (18.84)	2.9 (8.10)	27	101.8 (18.73)	0.3 (8.31)
End of Treatment1	84	98.5 (16.13)	0.8 (10.77)	82	105.6 (19.79)	1.8 (8.60)	80	104.6 (20.50)	1.4 (10.38)
Gamma Glutamyl Transferase (U/L)
Baseline	86	24.9 (49.75)		82	22.2 (15.57)		84	22.8 (17.71)
Week 2	84	24.2 (47.03)	-0.8 (5.52)	80	22.6 (15.52)	0.2 (9.48)	78	23.7 (18.17)	0.6 (8.96)
Week 4	82	24.5 (48.68)	-0.8 (5.69)	72	21.8 (13.82)	-1.1 (7.51)	72	23.3 (16.08)	0.1 (9.29)
Week 6	75	19.9 (15.55)	0.3 (6.11)	64	21.9 (13.72)	-0.6 (6.88)	67	21.6 (13.48)	-0.5 (6.74)
Week 8	73	20.6 (16.38)	0.1 (6.76)	60	23.6 (15.55)	1.1 (11.36)	56	26.8 (29.96)	4.0 (24.63)
Week 12	67	21.4 (15.04)	0.5 (5.70)	52	21.9 (13.97)	-0.7 (8.80)	50	24.7 (21.26)	1.2 (8.19)
Week 16	68	20.8 (13.00)	-0.1 (5.65)	42	22.8 (14.86)	-0.1 (6.38)	37	23.9 (17.79)	-1.3 (6.16)
Week 20	66	20.4 (13.90)	-0.5 (5.42)	31	24.1 (17.52)	-0.4 (5.90)	31	22.2 (11.28)	-0.5 (5.98)
Week 24	57	22.0 (16.68)	0.5 (10.90)	27	23.7 (14.75)	0.5 (7.28)	30	25.5 (22.64)	-0.9 (6.48)
Week 26	57	21.4 (13.91)	-0.4 (7.08)	25	22.0 (15.31)	-1.0 (8.44)	27	23.4 (18.73)	-3.4 (7.97)
End of Treatment1	84	25.7 (48.78)	0.7 (9.60)	82	22.4 (14.03)	0.3 (10.36)	80	22.3 (15.66)	-0.6 (9.81)
Glucose (mmol/L)
Baseline	86	5.6 (2.14)		82	5.4 (0.94)		84	5.4 (1.34)
Week 2	83	5.6 (1.87)	0.0 (1.37)	80	5.6 (1.75)	0.1 (1.50)	78	6.1 (2.92)	0.7 (2.06)
Week 4	79	5.6 (1.87)	0.0 (1.55)	70	5.4 (1.41)	-0.1 (1.09)	72	5.9 (1.84)	0.4 (1.25)
Week 6	73	5.7 (2.22)	0.1 (1.37)	62	5.3 (1.34)	-0.1 (1.12)	66	6.0 (2.80)	0.5 (2.20)
Week 8	72	5.5 (1.35)	-0.1 (2.09)	59	5.5 (1.76)	0.1 (1.34)	56	5.8 (2.15)	0.2 (1.68)
Week 12	67	6.1 (1.97)	0.4 (1.97)	51	5.9 (3.18)	0.4 (2.78)	49	6.0 (2.28)	0.4 (1.64)
Week 16	68	5.5 (1.42)	-0.2 (1.69)	42	5.3 (0.83)	-0.2 (0.89)	37	5.9 (2.30)	0.2 (1.74)
Week 20	65	5.8 (1.50)	0.1 (2.08)	30	5.7 (1.73)	0.1 (1.39)	31	5.8 (1.61)	0.2 (1.47)
Week 24	57	5.7 (1.83)	-0.1 (2.68)	26	5.7 (1.26)	0.2 (0.82)	30	6.0 (1.92)	0.5 (1.94)
Week 26	57	5.8 (1.85)	0.0 (1.60)	25	5.5 (1.72)	0.1 (1.35)	27	5.6 (1.01)	0.1 (1.66)
End of Treatment1	84	5.6 (1.61)	0.0 (2.26)	82	5.4 (1.07)	-0.1 (1.03)	80	5.9 (2.15)	0.5 (1.62)
Phosphate (mmol/L)
Baseline	86	1.2 (0.15)		81	1.2 (0.11)		83	1.2 (0.15)
Week 2	84	1.2 (0.15)	0.0 (0.16)	80	1.2 (0.15)	0.0 (0.16)	78	1.2 (0.18)	0.0 (0.19)
Week 4	82	1.1 (0.16)	0.0 (0.17)	71	1.1 (0.14)	0.0 (0.14)	72	1.2 (0.18)	0.0 (0.16)
Week 6	75	1.2 (0.15)	0.0 (0.17)	64	1.2 (0.17)	0.0 (0.17)	67	1.2 (0.15)	0.0 (0.16)
Week 8	73	1.1 (0.14)	0.0 (0.13)	60	1.1 (0.16)	0.0 (0.16)	56	1.2 (0.16)	0.0 (0.18)
Week 12	67	1.2 (0.16)	0.0 (0.18)	52	1.1 (0.14)	0.0 (0.18)	50	1.1 (0.15)	0.0 (0.17)
Week 16	68	1.1 (0.14)	0.0 (0.15)	42	1.1 (0.17)	0.0 (0.19)	37	1.2 (0.17)	0.0 (0.17)
Week 20	65	1.2 (0.17)	0.0 (0.15)	31	1.1 (0.13)	0.0 (0.14)	31	1.2 (0.15)	0.0 (0.16)
Week 24	56	1.2 (0.15)	0.0 (0.16)	27	1.2 (0.19)	0.1 (0.19)	30	1.1 (0.17)	0.0 (0.17)
Week 26	57	1.2 (0.15)	0.0 (0.18)	25	1.1 (0.17)	0.0 (0.17)	27	1.2 (0.19)	0.0 (0.20)
End of Treatment1	84	1.2 (0.17)	0.0 (0.18)	82	1.2 (0.17)	0.0 (0.18)	80	1.2 (0.16)	0.0 (0.17)
Potassium (mmol/L)
Baseline	86	4.3 (0.43)		81	4.3 (0.34)		82	4.3 (0.41)
Week 2	84	4.2 (0.41)	0.0 (0.37)	80	4.3 (0.41)	0.0 (0.38)	77	4.3 (0.40)	0.0 (0.46)
Week 4	82	4.2 (0.37)	-0.1 (0.38)	71	4.3 (0.40)	-0.1 (0.36)	72	4.2 (0.37)	-0.1 (0.46)
Week 6	75	4.3 (0.33)	0.0 (0.41)	64	4.3 (0.39)	-0.1 (0.41)	67	4.2 (0.32)	-0.1 (0.44)
Week 8	73	4.2 (0.37)	-0.1 (0.43)	60	4.3 (0.39)	0.0 (0.38)	56	4.2 (0.39)	0.0 (0.50)
Week 12	67	4.2 (0.41)	0.0 (0.46)	52	4.2 (0.42)	-0.2 (0.40)	49	4.2 (0.26)	-0.1 (0.39)
Week 16	68	4.2 (0.41)	0.0 (0.46)	42	4.3 (0.33)	-0.1 (0.35)	37	4.3 (0.27)	0.0 (0.35)
Week 20	64	4.3 (0.46)	0.0 (0.44)	31	4.3 (0.36)	0.0 (0.44)	31	4.3 (0.37)	0.0 (0.40)
Week 24	56	4.3 (0.44)	0.1 (0.44)	27	4.3 (0.41)	0.0 (0.35)	30	4.2 (0.42)	0.0 (0.50)
Week 26	57	4.2 (0.38)	0.0 (0.34)	25	4.3 (0.40)	0.0 (0.36)	27	4.3 (0.39)	0.0 (0.54)
End of Treatment1	84	4.3 (0.43)	0.0 (0.41)	82	4.3 (0.43)	-0.1 (0.39)	80	4.2 (0.36)	-0.1 (0.47)
Protein (g/L)
Baseline	86	70.5 (4.72)		82	70.4 (4.29)		84	71.0 (4.24)
Week 2	84	69.2 (4.93)	-1.3 (4.53)	80	69.5 (4.48)	-0.9 (3.63)	78	69.5 (4.31)	-1.6 (3.91)
Week 4	82	69.0 (4.18)	-1.6 (3.93)	72	68.8 (4.30)	-1.5 (3.97)	72	69.8 (4.54)	-1.4 (3.90)
Week 6	75	69.6 (3.95)	-1.2 (4.01)	64	68.8 (4.33)	-1.3 (4.17)	67	70.2 (4.28)	-1.1 (4.10)
Week 8	73	70.5 (5.20)	-0.4 (4.35)	60	69.2 (3.95)	-0.9 (3.71)	56	70.6 (3.78)	-1.0 (3.57)
Week 12	67	70.0 (4.28)	-0.6 (3.86)	52	69.3 (4.98)	-0.8 (3.62)	50	70.4 (3.83)	-1.0 (4.11)
Week 16	68	71.2 (4.57)	0.4 (4.39)	42	69.3 (4.79)	-0.5 (4.27)	37	70.5 (4.93)	-1.0 (4.57)
Week 20	66	70.0 (4.92)	-1.1 (4.53)	31	68.7 (4.82)	-1.3 (3.88)	31	69.5 (4.15)	-1.8 (4.13)
Week 24	57	70.1 (4.33)	-0.8 (4.81)	27	70.9 (6.11)	0.7 (4.03)	30	70.7 (3.47)	-0.7 (3.83)
Week 26	57	70.4 (4.48)	-0.5 (4.56)	25	69.4 (5.29)	-1.1 (4.26)	27	69.4 (4.89)	-2.5 (4.01)
End of Treatment1	84	70.2 (4.26)	-0.3 (4.35)	82	70.1 (4.95)	-0.4 (3.97)	80	70.4 (3.91)	-0.7 (3.63)
Sodium (mmol/L)
Baseline	86	140.3 (2.74)		82	140.0 (2.61)		83	140.0 (3.11)
Week 2	84	140.4 (2.62)	0.0 (3.35)	80	139.6 (2.47)	-0.3 (2.86)	77	139.1 (2.74)	-0.6 (2.87)
Week 4	82	139.9 (2.74)	-0.4 (3.58)	72	140.1 (2.40)	0.2 (3.16)	72	139.7 (2.75)	-0.3 (3.39)
Week 6	75	140.3 (2.58)	0.0 (3.08)	64	140.6 (2.69)	0.8 (3.25)	67	140.1 (2.61)	-0.2 (3.23)
Week 8	73	140.7 (2.48)	0.4 (3.33)	60	140.5 (2.58)	0.6 (3.33)	56	140.5 (3.15)	0.3 (3.79)
Week 12	67	140.4 (2.39)	0.2 (2.98)	52	141.1 (2.77)	1.1 (3.06)	49	140.2 (2.39)	-0.1 (2.79)
Week 16	68	141.1 (2.37)	0.9 (3.10)	42	141.0 (2.62)	1.2 (2.93)	37	141.6 (2.96)	1.4 (3.18)
Week 20	65	141.2 (2.53)	1.1 (3.01)	31	141.4 (2.53)	1.6 (3.42)	31	141.8 (3.87)	1.2 (3.61)
Week 24	57	141.7 (2.23)	1.6 (3.27)	27	141.5 (2.12)	1.2 (3.12)	30	141.6 (2.99)	1.0 (3.10)
Week 26	57	142.6 (2.25)	2.5 (2.90)	25	142.1 (2.08)	1.9 (3.13)	27	142.4 (3.07)	1.9 (3.43)
End of Treatment1	84	141.5 (2.74)	1.1 (3.51)	82	141.1 (2.65)	1.3 (3.09)	80	140.5 (3.22)	0.5 (3.50)
Urate (umol/L)
Baseline	86	285.0 (74.45)		82	300.7 (77.78)		84	302.2 (78.01)
Week 2	84	284.0 (68.72)	-1.8 (50.16)	80	305.8 (74.25)	1.0 (40.56)	78	301.7 (81.96)	-2.7 (33.85)
Week 4	82	285.6 (69.22)	-0.6 (35.59)	72	299.5 (79.44)	-3.7 (38.80)	72	291.4 (79.28)	-9.5 (42.17)
Week 6	75	288.0 (68.73)	-1.1 (38.24)	64	298.1 (74.46)	-6.2 (30.89)	67	291.6 (74.82)	-9.8 (34.48)
Week 8	73	285.6 (65.63)	-5.1 (36.28)	60	290.9 (71.15)	-15.6 (30.49)	56	291.6 (76.76)	-11.2 (35.13)
Week 12	67	291.1 (70.07)	3.2 (40.20)	52	290.3 (63.20)	-11.5 (31.83)	50	294.2 (78.84)	-9.0 (36.29)
Week 16	68	285.8 (68.72)	-3.6 (41.33)	42	288.8 (58.78)	-12.9 (39.98)	37	284.1 (80.47)	-19.9 (36.13)
Week 20	66	301.9 (73.43)	12.1 (40.18)	31	279.7 (58.71)	-16.5 (44.34)	31	295.1 (85.89)	-9.6 (43.81)
Week 24	57	293.5 (73.47)	4.6 (42.98)	27	274.9 (57.72)	-19.2 (31.26)	30	288.5 (88.37)	-19.4 (50.03)
Week 26	57	291.9 (74.43)	5.1 (44.39)	25	279.3 (60.01)	-13.6 (39.12)	27	301.1 (86.31)	-12.6 (47.68)
End of Treatment1	84	290.5 (73.86)	4.7 (46.43)	82	298.3 (80.54)	-3.1 (37.11)	80	292.4 (80.83)	-10.5 (42.35)
1 Last observed value while on treatment (prior to or at Week 24)

Table 14-6.01 HEMATOLOGY
Summary Statistics for Continuous Laboratory Values

	Placebo			Xanomeline Low			Xanomeline High
	N	Mean (SD)	Change from Bsln Mean (SD)	N	Mean (SD)	Change from Bsln Mean (SD)	N	Mean (SD)	Change from Bsln Mean (SD)
Basophils (GI/L)
Baseline	85	0.1 (0.04)		81	0.1 (0.03)		81	0.1 (0.02)
Week 2	83	0.0 (0.02)	0.0 (0.03)	80	0.1 (0.03)	0.0 (0.03)	80	0.0 (0.03)	0.0 (0.02)
Week 4	79	0.0 (0.02)	0.0 (0.03)	71	0.0 (0.03)	0.0 (0.03)	71	0.0 (0.02)	0.0 (0.02)
Week 6	73	0.0 (0.03)	0.0 (0.03)	62	0.0 (0.03)	0.0 (0.03)	64	0.1 (0.03)	0.0 (0.03)
Week 8	72	0.0 (0.02)	0.0 (0.03)	59	0.1 (0.02)	0.0 (0.04)	56	0.1 (0.03)	0.0 (0.02)
Week 12	66	0.0 (0.02)	0.0 (0.04)	50	0.0 (0.02)	0.0 (0.03)	50	0.1 (0.03)	0.0 (0.02)
Week 16	68	0.0 (0.03)	0.0 (0.04)	42	0.0 (0.02)	0.0 (0.03)	37	0.1 (0.03)	0.0 (0.02)
Week 20	65	0.0 (0.03)	0.0 (0.03)	30	0.0 (0.02)	0.0 (0.03)	31	0.0 (0.02)	0.0 (0.03)
Week 24	58	0.0 (0.03)	0.0 (0.03)	25	0.0 (0.03)	0.0 (0.02)	30	0.1 (0.02)	0.0 (0.02)
Week 26	57	0.0 (0.02)	0.0 (0.03)	25	0.0 (0.03)	0.0 (0.02)	27	0.1 (0.03)	0.0 (0.03)
End of Treatment1	84	0.0 (0.03)	0.0 (0.03)	82	0.0 (0.03)	0.0 (0.03)	81	0.1 (0.02)	0.0 (0.02)
Eosinophils (GI/L)
Baseline	85	0.1 (0.12)		81	0.1 (0.12)		81	0.1 (0.10)
Week 2	83	0.1 (0.10)	0.0 (0.09)	80	0.2 (0.17)	0.0 (0.14)	80	0.1 (0.11)	0.0 (0.07)
Week 4	79	0.1 (0.13)	0.0 (0.09)	71	0.2 (0.14)	0.0 (0.11)	71	0.2 (0.15)	0.0 (0.13)
Week 6	73	0.1 (0.13)	0.0 (0.11)	62	0.2 (0.21)	0.1 (0.16)	64	0.2 (0.28)	0.1 (0.25)
Week 8	72	0.1 (0.10)	0.0 (0.10)	59	0.3 (0.26)	0.2 (0.23)	56	0.2 (0.25)	0.1 (0.19)
Week 12	66	0.1 (0.09)	0.0 (0.11)	50	0.3 (0.21)	0.1 (0.19)	50	0.2 (0.17)	0.1 (0.14)
Week 16	68	0.1 (0.08)	0.0 (0.10)	42	0.2 (0.16)	0.1 (0.13)	37	0.2 (0.21)	0.1 (0.17)
Week 20	65	0.1 (0.08)	0.0 (0.10)	30	0.2 (0.19)	0.1 (0.16)	31	0.1 (0.11)	0.0 (0.12)
Week 24	58	0.1 (0.08)	0.0 (0.09)	25	0.2 (0.14)	0.0 (0.12)	30	0.2 (0.15)	0.0 (0.13)
Week 26	57	0.1 (0.10)	0.0 (0.11)	25	0.2 (0.15)	0.0 (0.13)	27	0.1 (0.10)	0.0 (0.09)
End of Treatment1	84	0.1 (0.08)	0.0 (0.10)	82	0.2 (0.17)	0.0 (0.14)	81	0.2 (0.20)	0.1 (0.18)
Ery. Mean Corpuscular HGB Concentration (mmol/L)
Baseline	85	20.6 (0.87)		80	20.2 (0.98)		80	20.5 (0.87)
Week 2	83	20.3 (0.77)	-0.3 (1.00)	80	20.3 (0.85)	0.0 (1.04)	80	20.4 (0.74)	0.0 (0.98)
Week 4	77	20.2 (0.78)	-0.4 (1.01)	69	20.2 (0.72)	0.0 (0.94)	70	20.5 (0.71)	-0.1 (0.90)
Week 6	72	20.5 (0.77)	-0.2 (0.98)	60	20.1 (0.84)	-0.2 (0.99)	65	20.4 (0.89)	-0.2 (1.00)
Week 8	71	20.3 (0.68)	-0.2 (0.85)	56	20.2 (0.83)	0.0 (0.92)	56	20.5 (0.76)	-0.1 (0.93)
Week 12	65	20.3 (0.74)	-0.3 (0.96)	49	20.3 (0.68)	0.0 (0.99)	50	20.5 (0.76)	-0.1 (1.03)
Week 16	68	20.2 (0.75)	-0.4 (1.02)	42	20.0 (0.82)	-0.2 (1.02)	37	20.4 (0.83)	-0.2 (1.11)
Week 20	64	20.3 (0.80)	-0.3 (0.94)	30	20.0 (0.85)	-0.4 (0.91)	31	20.3 (0.88)	-0.3 (1.22)
Week 24	58	20.1 (0.89)	-0.5 (1.01)	25	20.3 (0.91)	-0.3 (0.95)	30	20.4 (0.85)	-0.3 (1.29)
Week 26	56	20.3 (0.99)	-0.4 (1.16)	25	19.9 (0.94)	-0.6 (1.09)	27	20.2 (0.92)	-0.6 (1.12)
End of Treatment1	83	20.2 (0.85)	-0.4 (0.99)	82	20.2 (0.81)	0.0 (0.98)	81	20.4 (0.78)	-0.1 (1.08)
Ery. Mean Corpuscular Hemoglobin (fmol(Fe))
Baseline	85	1.9 (0.12)		81	1.9 (0.09)		81	1.9 (0.13)
Week 2	83	1.9 (0.11)	0.0 (0.07)	80	1.9 (0.09)	0.0 (0.07)	80	1.9 (0.13)	0.0 (0.07)
Week 4	79	1.9 (0.12)	0.0 (0.07)	71	1.9 (0.09)	0.0 (0.07)	71	1.9 (0.13)	0.0 (0.06)
Week 6	73	1.9 (0.13)	0.0 (0.08)	62	1.9 (0.08)	0.0 (0.05)	65	1.9 (0.13)	0.0 (0.08)
Week 8	72	1.9 (0.12)	0.0 (0.06)	59	1.9 (0.09)	0.0 (0.06)	56	1.9 (0.10)	0.0 (0.07)
Week 12	66	1.9 (0.12)	0.0 (0.06)	50	1.9 (0.09)	0.0 (0.05)	50	1.9 (0.09)	0.0 (0.07)
Week 16	68	1.9 (0.12)	0.0 (0.05)	42	1.9 (0.10)	0.0 (0.07)	37	1.9 (0.11)	0.0 (0.08)
Week 20	65	1.9 (0.11)	0.0 (0.06)	30	1.9 (0.08)	0.0 (0.06)	31	1.9 (0.12)	0.0 (0.08)
Week 24	58	1.9 (0.12)	0.0 (0.06)	25	1.9 (0.09)	0.0 (0.06)	30	1.9 (0.13)	0.0 (0.09)
Week 26	57	1.9 (0.12)	0.0 (0.07)	25	1.9 (0.08)	0.0 (0.08)	27	1.9 (0.12)	0.0 (0.08)
End of Treatment1	84	1.9 (0.12)	0.0 (0.07)	82	1.9 (0.09)	0.0 (0.06)	81	1.9 (0.13)	0.0 (0.07)
Ery. Mean Corpuscular Volume (fL)
Baseline	85	92.9 (5.69)		80	94.6 (4.90)		80	93.3 (5.91)
Week 2	83	94.1 (5.31)	1.2 (3.93)	80	94.6 (4.12)	0.2 (3.57)	80	93.4 (5.93)	0.2 (3.91)
Week 4	77	94.6 (5.87)	1.7 (3.47)	69	95.1 (4.22)	0.6 (3.69)	70	93.0 (5.52)	0.0 (3.69)
Week 6	72	94.2 (6.21)	1.3 (3.79)	60	95.2 (4.51)	0.7 (3.68)	65	94.0 (6.61)	0.6 (3.99)
Week 8	71	94.3 (5.51)	1.5 (3.27)	56	94.9 (3.97)	0.6 (3.35)	56	94.4 (4.77)	0.4 (4.03)
Week 12	65	94.2 (5.66)	1.6 (3.81)	49	95.1 (4.17)	0.9 (3.91)	50	94.9 (4.90)	0.8 (4.20)
Week 16	68	94.1 (6.01)	1.7 (4.07)	42	95.8 (5.17)	1.2 (4.18)	37	94.7 (6.27)	0.6 (5.05)
Week 20	64	93.3 (5.58)	1.3 (4.14)	30	95.4 (5.43)	2.0 (3.29)	31	95.1 (5.91)	0.9 (5.12)
Week 24	58	93.9 (5.90)	1.6 (3.85)	25	94.3 (3.85)	0.8 (3.94)	30	95.0 (5.99)	0.7 (4.97)
Week 26	56	93.9 (6.40)	1.7 (4.48)	25	95.8 (4.33)	2.6 (3.70)	27	95.4 (6.19)	1.2 (3.93)
End of Treatment1	83	94.4 (5.93)	1.4 (4.07)	82	95.0 (3.98)	0.6 (3.83)	81	93.8 (6.14)	0.7 (4.40)
Erythrocytes (TI/L)
Baseline	85	4.5 (0.45)		81	4.5 (0.42)		81	4.7 (0.47)
Week 2	83	4.4 (0.40)	-0.1 (0.22)	80	4.4 (0.43)	-0.1 (0.27)	80	4.6 (0.47)	-0.1 (0.22)
Week 4	79	4.4 (0.45)	-0.1 (0.25)	71	4.4 (0.38)	-0.2 (0.30)	71	4.5 (0.49)	-0.1 (0.25)
Week 6	73	4.4 (0.37)	-0.1 (0.24)	62	4.3 (0.38)	-0.2 (0.27)	65	4.5 (0.50)	-0.1 (0.24)
Week 8	72	4.4 (0.41)	-0.1 (0.26)	59	4.4 (0.34)	-0.2 (0.24)	56	4.5 (0.45)	-0.1 (0.19)
Week 12	66	4.4 (0.44)	-0.1 (0.26)	50	4.3 (0.32)	-0.2 (0.20)	50	4.6 (0.47)	-0.1 (0.27)
Week 16	68	4.5 (0.41)	0.0 (0.27)	42	4.3 (0.36)	-0.1 (0.24)	37	4.6 (0.44)	-0.2 (0.26)
Week 20	65	4.5 (0.41)	-0.1 (0.30)	30	4.3 (0.32)	-0.1 (0.21)	31	4.5 (0.50)	-0.2 (0.30)
Week 24	58	4.4 (0.45)	-0.1 (0.23)	25	4.4 (0.38)	-0.1 (0.19)	30	4.6 (0.58)	-0.1 (0.24)
Week 26	57	4.4 (0.42)	0.0 (0.25)	25	4.4 (0.35)	-0.1 (0.20)	27	4.5 (0.47)	-0.2 (0.27)
End of Treatment1	84	4.4 (0.43)	-0.1 (0.27)	82	4.4 (0.38)	-0.2 (0.25)	81	4.6 (0.49)	-0.1 (0.22)
Hematocrit
Baseline	85	0.4 (0.04)		80	0.4 (0.04)		80	0.4 (0.04)
Week 2	83	0.4 (0.04)	0.0 (0.03)	80	0.4 (0.04)	0.0 (0.03)	80	0.4 (0.04)	0.0 (0.03)
Week 4	77	0.4 (0.04)	0.0 (0.03)	69	0.4 (0.04)	0.0 (0.03)	70	0.4 (0.04)	0.0 (0.03)
Week 6	72	0.4 (0.04)	0.0 (0.03)	60	0.4 (0.04)	0.0 (0.03)	65	0.4 (0.04)	0.0 (0.03)
Week 8	71	0.4 (0.04)	0.0 (0.03)	56	0.4 (0.03)	0.0 (0.02)	56	0.4 (0.03)	0.0 (0.03)
Week 12	65	0.4 (0.04)	0.0 (0.03)	49	0.4 (0.03)	0.0 (0.02)	50	0.4 (0.04)	0.0 (0.03)
Week 16	68	0.4 (0.04)	0.0 (0.03)	42	0.4 (0.03)	0.0 (0.03)	37	0.4 (0.03)	0.0 (0.04)
Week 20	64	0.4 (0.04)	0.0 (0.03)	30	0.4 (0.03)	0.0 (0.02)	31	0.4 (0.04)	0.0 (0.04)
Week 24	58	0.4 (0.04)	0.0 (0.03)	25	0.4 (0.04)	0.0 (0.02)	30	0.4 (0.04)	0.0 (0.04)
Week 26	56	0.4 (0.04)	0.0 (0.03)	25	0.4 (0.04)	0.0 (0.03)	27	0.4 (0.04)	0.0 (0.03)
End of Treatment1	83	0.4 (0.04)	0.0 (0.03)	82	0.4 (0.04)	0.0 (0.03)	81	0.4 (0.04)	0.0 (0.03)
Hemoglobin (mmol/L)
Baseline	85	8.6 (0.83)		81	8.6 (0.77)		81	8.9 (0.78)
Week 2	83	8.4 (0.76)	-0.2 (0.42)	80	8.4 (0.76)	-0.2 (0.46)	80	8.7 (0.83)	-0.2 (0.42)
Week 4	79	8.4 (0.82)	-0.2 (0.44)	71	8.4 (0.78)	-0.3 (0.50)	71	8.6 (0.78)	-0.2 (0.45)
Week 6	73	8.4 (0.77)	-0.2 (0.44)	62	8.3 (0.76)	-0.3 (0.46)	65	8.6 (0.75)	-0.3 (0.44)
Week 8	72	8.5 (0.76)	-0.2 (0.46)	59	8.3 (0.71)	-0.3 (0.39)	56	8.7 (0.73)	-0.3 (0.39)
Week 12	66	8.4 (0.83)	-0.2 (0.42)	50	8.3 (0.69)	-0.3 (0.34)	50	8.8 (0.79)	-0.3 (0.42)
Week 16	68	8.5 (0.73)	-0.1 (0.46)	42	8.3 (0.69)	-0.2 (0.42)	37	8.7 (0.70)	-0.3 (0.47)
Week 20	65	8.4 (0.80)	-0.2 (0.51)	30	8.2 (0.63)	-0.2 (0.46)	31	8.7 (0.79)	-0.4 (0.46)
Week 24	58	8.3 (0.82)	-0.2 (0.40)	25	8.4 (0.69)	-0.2 (0.36)	30	8.9 (0.85)	-0.3 (0.36)
Week 26	57	8.4 (0.80)	-0.2 (0.47)	25	8.3 (0.67)	-0.3 (0.39)	27	8.7 (0.74)	-0.5 (0.39)
End of Treatment1	84	8.4 (0.78)	-0.2 (0.44)	82	8.4 (0.69)	-0.2 (0.44)	81	8.7 (0.79)	-0.2 (0.36)
Leukocytes (GI/L)
Baseline	85	6.9 (1.76)		81	6.6 (1.95)		81	6.5 (1.54)
Week 2	83	6.5 (1.55)	-0.4 (1.40)	80	6.9 (1.99)	0.2 (1.37)	80	6.5 (1.59)	-0.1 (1.18)
Week 4	79	6.4 (1.63)	-0.4 (1.17)	71	6.8 (2.12)	0.1 (1.53)	71	6.6 (1.42)	0.1 (1.21)
Week 6	73	6.5 (1.57)	-0.4 (1.36)	62	6.9 (1.99)	0.1 (1.14)	65	7.0 (1.84)	0.4 (1.65)
Week 8	72	6.5 (1.70)	-0.3 (1.28)	59	7.1 (1.88)	0.3 (1.26)	56	6.9 (1.95)	0.4 (1.64)
Week 12	66	6.3 (1.28)	-0.5 (1.09)	50	6.8 (2.03)	0.2 (0.97)	50	6.6 (1.51)	0.0 (1.21)
Week 16	68	6.4 (1.50)	-0.5 (1.29)	42	6.6 (2.11)	0.2 (1.44)	37	6.8 (1.72)	-0.1 (1.37)
Week 20	65	6.5 (1.77)	-0.5 (1.40)	30	6.5 (1.94)	0.1 (1.24)	31	6.6 (1.67)	-0.2 (1.21)
Week 24	58	6.7 (1.77)	-0.2 (1.37)	25	6.3 (1.84)	0.0 (1.16)	30	6.7 (1.80)	-0.1 (1.16)
Week 26	57	6.4 (1.47)	-0.4 (1.24)	25	6.1 (1.93)	-0.3 (1.15)	27	6.7 (1.78)	-0.2 (1.42)
End of Treatment1	84	6.6 (1.80)	-0.2 (1.32)	82	6.8 (2.17)	0.1 (1.35)	81	6.7 (1.76)	0.2 (1.37)
Lymphocytes (GI/L)
Baseline	85	1.8 (0.57)		81	1.8 (0.57)		81	1.7 (0.52)
Week 2	83	1.7 (0.50)	-0.1 (0.37)	80	1.8 (0.66)	0.0 (0.45)	80	1.7 (0.50)	0.0 (0.38)
Week 4	79	1.7 (0.56)	0.0 (0.38)	71	1.8 (0.67)	0.0 (0.49)	71	1.7 (0.50)	0.0 (0.44)
Week 6	73	1.8 (0.58)	0.0 (0.41)	62	1.9 (0.64)	0.1 (0.47)	64	1.7 (0.48)	0.0 (0.44)
Week 8	72	1.8 (0.67)	0.0 (0.44)	59	1.9 (0.65)	0.1 (0.45)	56	1.7 (0.50)	0.0 (0.36)
Week 12	66	1.7 (0.58)	-0.1 (0.42)	50	1.8 (0.58)	-0.1 (0.45)	50	1.6 (0.47)	0.0 (0.37)
Week 16	68	1.7 (0.58)	-0.1 (0.42)	42	1.8 (0.63)	-0.1 (0.34)	37	1.7 (0.62)	0.0 (0.41)
Week 20	65	1.7 (0.50)	-0.1 (0.45)	30	1.7 (0.59)	-0.1 (0.50)	31	1.7 (0.59)	0.0 (0.36)
Week 24	58	1.8 (0.65)	0.0 (0.48)	25	1.8 (0.56)	-0.1 (0.37)	30	1.7 (0.56)	0.0 (0.37)
Week 26	57	1.8 (0.59)	0.0 (0.43)	25	1.7 (0.62)	-0.1 (0.39)	27	1.8 (0.61)	0.1 (0.50)
End of Treatment1	84	1.8 (0.59)	0.0 (0.46)	82	1.8 (0.62)	0.0 (0.46)	81	1.6 (0.54)	0.0 (0.38)
Monocytes (GI/L)
Baseline	85	0.4 (0.15)		81	0.5 (0.16)		81	0.4 (0.12)
Week 2	83	0.4 (0.15)	0.0 (0.14)	80	0.5 (0.16)	0.0 (0.13)	80	0.4 (0.15)	0.0 (0.12)
Week 4	79	0.4 (0.12)	0.0 (0.11)	71	0.5 (0.17)	0.0 (0.15)	71	0.4 (0.13)	0.0 (0.13)
Week 6	73	0.4 (0.15)	0.0 (0.13)	62	0.5 (0.19)	0.0 (0.16)	64	0.5 (0.15)	0.0 (0.13)
Week 8	72	0.4 (0.14)	0.0 (0.14)	59	0.5 (0.14)	0.0 (0.14)	56	0.5 (0.15)	0.0 (0.14)
Week 12	66	0.4 (0.13)	0.0 (0.13)	50	0.5 (0.15)	0.0 (0.12)	50	0.4 (0.15)	0.0 (0.11)
Week 16	68	0.5 (0.15)	0.0 (0.17)	42	0.5 (0.16)	0.0 (0.12)	37	0.5 (0.16)	0.0 (0.14)
Week 20	65	0.5 (0.16)	0.0 (0.15)	30	0.5 (0.13)	0.0 (0.13)	31	0.5 (0.17)	0.0 (0.18)
Week 24	58	0.5 (0.17)	0.0 (0.17)	25	0.5 (0.19)	0.0 (0.11)	30	0.5 (0.18)	0.0 (0.12)
Week 26	57	0.4 (0.14)	0.0 (0.14)	25	0.4 (0.17)	0.0 (0.12)	27	0.5 (0.14)	0.0 (0.13)
End of Treatment1	84	0.5 (0.17)	0.0 (0.17)	82	0.5 (0.17)	0.0 (0.13)	81	0.5 (0.15)	0.0 (0.12)
Platelet (GI/L)
Baseline	84	250.3 (65.52)		79	233.8 (58.58)		81	227.7 (54.74)
Week 2	83	246.2 (57.55)	-4.4 (36.21)	78	247.4 (59.37)	9.1 (46.36)	80	238.2 (59.77)	8.4 (31.55)
Week 4	78	243.8 (54.30)	-7.4 (40.04)	70	239.2 (57.80)	5.7 (39.75)	70	238.3 (49.60)	9.8 (29.21)
Week 6	73	250.8 (58.97)	-3.3 (36.82)	62	238.3 (55.74)	3.5 (43.39)	63	239.1 (56.79)	9.2 (37.82)
Week 8	72	246.0 (66.61)	-6.2 (42.35)	59	245.4 (60.21)	5.1 (46.18)	55	236.9 (70.88)	11.0 (48.13)
Week 12	65	241.9 (53.71)	-12.8 (42.62)	50	238.8 (49.49)	1.9 (24.69)	49	236.1 (53.03)	9.1 (34.47)
Week 16	68	241.8 (55.65)	-13.7 (43.45)	41	244.5 (57.60)	2.8 (32.93)	37	230.9 (58.00)	5.3 (25.87)
Week 20	65	248.4 (60.70)	-8.4 (30.42)	30	240.7 (64.07)	7.3 (42.65)	30	235.1 (65.46)	2.3 (22.80)
Week 24	57	238.8 (51.89)	-11.3 (35.06)	24	249.7 (63.44)	1.8 (33.48)	29	238.3 (67.53)	4.5 (26.74)
Week 26	56	247.6 (60.11)	-2.7 (41.73)	25	241.3 (57.16)	-1.9 (38.84)	27	237.4 (67.14)	0.3 (26.99)
End of Treatment1	84	241.5 (59.49)	-8.5 (35.55)	82	236.7 (63.56)	1.1 (36.31)	81	233.8 (60.79)	4.1 (35.89)
1 Last observed value while on treatment (prior to or at Week 24)

Caution

The order of items within Chemistry and Hematology is currently alphabetical, based on laboratory names. However, we can adopt alternative sorting methods if preferred

Table 14-6-02

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adlbc_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbc.xpt')  

adlbh_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbh.xpt')

# COMBINE & SUBSEST
adlb <- bind_rows(
   CHEMISTRY = adlbc_orig,
   HEMATOLOGY = adlbh_orig,
   .id = 'ORIG') %>% 
   filter(SAFFL == 'Y', ANL01FL == 'Y', AVISITN != 99) %>% 
   select(ORIG, PARCAT1, PARAM, PARAMCD, USUBJID , TRTA, AVISIT, AVISITN, AVAL, CHG, LBNRIND) %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(LBNRIND = factor(LBNRIND,
                           levels = c('LOW','NORMAL','HIGH'),
                           labels = c('Low','Normal','High')) ) %>% 
   mutate(PARAM = str_remove_all(PARAM, "\\(.*?\\)") %>% str_trim() ) %>% 
   filter(!is.na(PARAM), !is.na(TRTA), !is.na(LBNRIND)) %>%
   filter(LBNRIND %in% c('Low','Normal','High')) %>% 
   filter(!str_starts(PARAMCD, '_') )

# NEST
adlb_n <- adlb %>% 
   nest_by(ORIG) 

# TABLES
adlb_t <- adlb_n %>% 
   mutate(t_cross   = list(
      tbl_strata(
         data = data %>% set_variable_labels(PARAM = str_glue('{ORIG}')),
         strata = PARAM,
         .tbl_fun = 
            ~.x %>% 
            tbl_cross(row = TRTA,
                      col = LBNRIND,
                      margin = NULL,
                      percent = 'row',
                      digits = c(0, 0)) %>% 
            add_p(),
         .quiet = TRUE,
         .combine_with = 'tbl_stack'
      )
   ),
   t_summary = list(
      tbl_strata(
         data = data %>% set_variable_labels(PARAM = str_glue('{ORIG}')),
         strata = TRTA,
         .tbl_fun = 
            ~.x %>% 
            tbl_summary(
               by = LBNRIND,
               include = PARAM,
               percent = 'row',
               digits = everything() ~ c(0, 0) ) %>%
            add_p() %>% 
            modify_header(all_stat_cols() ~ '**{level}**'),
         .header = '**{strata}**',
         .quiet = TRUE
      )
   ),
   t_combine = list(
      t_summary %>% 
         modify_column_hide(columns = c(test_name_1, test_result_1, p.value_1,
                                        test_name_2, test_result_2, p.value_2)) %>% 
         modify_table_body(
            ~.x %>% 
               mutate(across(starts_with('stat_'), ~str_replace(., "0 \\(0%\\)", "0"))) %>% 
               select(-test_name_3, -test_result_3, -p.value_3) %>% 
               left_join(
                  t_cross$table_body %>% 
                     filter(!is.na(alternative)) %>% 
                     mutate(variable = 'PARAM',
                            var_label = str_glue('{ORIG}'),
                            row_type = 'level',
                            label = groupname_col,
                            test_name_3 = test_name,
                            test_results_3 = test_result,
                            p.value_3 = p.value) %>% 
                     select(variable, var_label, row_type, label, test_name_3, test_results_3,  p.value_3)
               ) 
         ) 
   )
   )

# STACK & PRINT
tbl_stack(tbls = adlb_t$t_combine) %>%
   bold_labels() %>%
   modify_column_alignment(
      all_stat_cols(), 'left') %>% 
   modify_spanning_header(
      c(p.value_3) ~ '**&nbsp;**') %>% 
   modify_fmt_fun(
      update = p.value_3 ~ function(x) style_pvalue(x, digits = 3)) %>% 
   modify_header(
      label  ~ '' ) %>% 
   modify_footnote(
      all_stat_cols() ~ 
         'Percentages are based on the number of subjects with non-missing assessments (i.e., the total of the
subjects in the low, normal, and high categories) within each treatment group') %>% 
   modify_caption(
      '**Table 14-6.02<br>Frequency of Normal and Abnormal (Beyond Normal Range) Laboratory Values During Treatment**' )

Table 14-6.02
Frequency of Normal and Abnormal (Beyond Normal Range) Laboratory Values During Treatment

	Placebo			Low Dose			High Dose
	Low1	Normal1	High1	Low1	Normal1	High1	Low1	Normal1	High1	p-value
CHEMISTRY
Alanine Aminotransferase	1 (1%)	74 (88%)	9 (11%)	5 (6%)	68 (83%)	9 (11%)	0	70 (88%)	10 (13%)	0.185
Albumin	16 (19%)	65 (77%)	3 (4%)	15 (18%)	66 (80%)	1 (1%)	5 (6%)	74 (93%)	1 (1%)	0.042
Alkaline Phosphatase	4 (5%)	72 (86%)	8 (10%)	2 (2%)	74 (90%)	6 (7%)	3 (4%)	73 (91%)	4 (5%)	0.776
Aspartate Aminotransferase	0	74 (88%)	10 (12%)	0	72 (88%)	10 (12%)	0	72 (90%)	8 (10%)	0.907
Bilirubin	0	78 (93%)	6 (7%)	0	79 (98%)	2 (2%)	0	75 (94%)	5 (6%)	0.402
Blood Urea Nitrogen	0	75 (89%)	9 (11%)	0	60 (73%)	22 (27%)	0	68 (85%)	12 (15%)	0.023
Calcium	5 (6%)	78 (93%)	1 (1%)	7 (9%)	72 (88%)	3 (4%)	7 (9%)	71 (89%)	2 (3%)	0.789
Chloride	0	74 (88%)	10 (12%)	0	74 (90%)	8 (10%)	0	78 (98%)	2 (3%)	0.058
Cholesterol	5 (6%)	73 (87%)	6 (7%)	6 (7%)	72 (88%)	4 (5%)	6 (8%)	70 (88%)	4 (5%)	0.962
Creatine Kinase	1 (1%)	66 (79%)	17 (20%)	1 (1%)	68 (83%)	13 (16%)	0	67 (84%)	13 (16%)	0.816
Creatinine	0	80 (95%)	4 (5%)	0	76 (93%)	6 (7%)	0	73 (91%)	7 (9%)	0.572
Gamma Glutamyl Transferase	4 (5%)	73 (87%)	7 (8%)	4 (5%)	69 (84%)	9 (11%)	1 (1%)	71 (89%)	8 (10%)	0.689
Glucose	0	82 (98%)	2 (2%)	0	81 (99%)	1 (1%)	0	77 (96%)	3 (4%)	0.534
Phosphate	1 (1%)	83 (99%)	0	0	80 (98%)	2 (2%)	1 (1%)	78 (98%)	1 (1%)	0.518
Potassium	3 (4%)	80 (95%)	1 (1%)	3 (4%)	79 (96%)	0	2 (3%)	78 (98%)	0	>0.999
Protein	1 (1%)	76 (90%)	7 (8%)	2 (2%)	75 (91%)	5 (6%)	1 (1%)	77 (96%)	2 (3%)	0.536
Sodium	4 (5%)	75 (89%)	5 (6%)	3 (4%)	74 (90%)	5 (6%)	7 (9%)	62 (78%)	11 (14%)	0.177
Urate	3 (4%)	77 (92%)	4 (5%)	1 (1%)	75 (91%)	6 (7%)	4 (5%)	69 (86%)	7 (9%)	0.564
HEMATOLOGY
Basophils	0	84 (100%)	0	0	82 (100%)	0	0	81 (100%)	0	>0.999
Eosinophils	0	84 (100%)	0	0	71 (87%)	11 (13%)	0	74 (91%)	7 (9%)	<0.001
Ery. Mean Corpuscular Hemoglobin )	0	81 (96%)	3 (4%)	0	81 (99%)	1 (1%)	1 (1%)	75 (93%)	5 (6%)	0.186
Ery. Mean Corpuscular HGB Concentration	15 (18%)	68 (82%)	0	17 (21%)	65 (79%)	0	9 (11%)	72 (89%)	0	0.231
Ery. Mean Corpuscular Volume	1 (1%)	53 (64%)	29 (35%)	0	64 (78%)	18 (22%)	1 (1%)	64 (79%)	16 (20%)	0.077
Erythrocytes	13 (15%)	71 (85%)	0	18 (22%)	64 (78%)	0	11 (14%)	68 (84%)	2 (2%)	0.238
Hematocrit	5 (6%)	74 (89%)	4 (5%)	9 (11%)	72 (88%)	1 (1%)	1 (1%)	78 (96%)	2 (2%)	0.052
Hemoglobin	14 (17%)	68 (81%)	2 (2%)	10 (12%)	72 (88%)	0	5 (6%)	73 (90%)	3 (4%)	0.093
Leukocytes	7 (8%)	73 (87%)	4 (5%)	6 (7%)	68 (83%)	8 (10%)	4 (5%)	74 (91%)	3 (4%)	0.462
Lymphocytes	6 (7%)	73 (87%)	5 (6%)	4 (5%)	69 (84%)	9 (11%)	4 (5%)	76 (94%)	1 (1%)	0.103
Monocytes	2 (2%)	80 (95%)	2 (2%)	1 (1%)	77 (94%)	4 (5%)	0	79 (98%)	2 (2%)	0.626
Platelet	0	81 (96%)	3 (4%)	2 (2%)	77 (94%)	3 (4%)	2 (2%)	77 (95%)	2 (2%)	0.681
1 Percentages are based on the number of subjects with non-missing assessments (i.e., the total of the subjects in the low, normal, and high categories) within each treatment group

Table 14-6-03

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adlbc_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbc.xpt')  

adlbh_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbh.xpt')


# COMBINE & SUBSEST
adlb <- bind_rows(
   CHEMISTRY = adlbc_orig,
   HEMATOLOGY = adlbh_orig,
   .id = 'ORIG') %>% 
   filter(SAFFL == 'Y', ANL01FL == 'Y', AVISITN != 99) %>% 
   select(ORIG, PARCAT1, PARAM, PARAMCD, USUBJID , TRTA, AVISIT, AVISITN, AVAL, CHG, LBNRIND) %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(LBNRIND = factor(LBNRIND,
                           levels = c('LOW','NORMAL','HIGH'),
                           labels = c('Low','Normal','High')) ) %>% 
   mutate(PARAM = str_remove_all(PARAM, '\\(.*?\\)') %>% str_trim() ) %>% 
   mutate(PARAM = str_remove_all(PARAM, 'change from previous visit, relative to normal range') %>% str_trim() ) %>% 
   mutate(PARAM = str_remove_all(PARAM, 'change from previous visit, relative to normal rang') %>% str_trim() ) %>% 
   filter(!is.na(PARAM), !is.na(TRTA), !is.na(LBNRIND)) %>%
   filter(LBNRIND %in% c('Low','Normal','High')) %>% 
   filter(str_starts(PARAMCD, '_') )

# NEST
adlb_n <- adlb %>% 
   nest_by(ORIG) 

# TABLES
adlb_t <- adlb_n %>% 
   mutate(t_cross   = list(
      tbl_strata(
         data = data %>% set_variable_labels(PARAM = str_glue('{ORIG}')),
         strata = PARAM,
         .tbl_fun = 
            ~.x %>% 
            tbl_cross(row = TRTA,
                      col = LBNRIND,
                      margin = NULL,
                      percent = 'row',
                      digits = c(0, 0)) %>% 
            add_p(),
         .quiet = TRUE,
         .combine_with = 'tbl_stack'
      )
   ),
   t_summary = list(
      tbl_strata(
         data = data %>% set_variable_labels(PARAM = str_glue('{ORIG}')),
         strata = TRTA,
         .tbl_fun = 
            ~.x %>% 
            tbl_summary(
               by = LBNRIND,
               include = PARAM,
               percent = 'row',
               digits = everything() ~ c(0, 0) ) %>%
            add_p() %>% 
            modify_header(all_stat_cols() ~ '**{level}**'),
         .header = '**{strata}**',
         .quiet = TRUE
      )
   ),
   t_combine = list(
      t_summary %>% 
         modify_column_hide(columns = c(test_name_1, test_result_1, p.value_1,
                                        test_name_2, test_result_2, p.value_2)) %>% 
         modify_table_body(
            ~.x %>% 
               mutate(across(starts_with('stat_'), ~str_replace(., '0 \\(0%\\)', '0'))) %>% 
               select(-test_name_3, -test_result_3, -p.value_3) %>% 
               left_join(
                  t_cross$table_body %>% 
                     filter(!is.na(alternative)) %>% 
                     mutate(variable = 'PARAM',
                            var_label = str_glue('{ORIG}'),
                            row_type = 'level',
                            label = groupname_col,
                            test_name_3 = test_name,
                            test_results_3 = test_result,
                            p.value_3 = p.value) %>% 
                     select(variable, var_label, row_type, label, test_name_3, test_results_3,  p.value_3)
               ) 
         ) 
   )
   )

# STACK & PRINT
tbl_stack(tbls = adlb_t$t_combine) %>%
   bold_labels() %>%
   modify_column_alignment(
      all_stat_cols(), 'left') %>% 
   modify_spanning_header(
      c(p.value_3) ~ '**&nbsp;**') %>% 
   modify_fmt_fun(
      update = p.value_3 ~ function(x) style_pvalue(x, digits = 3)) %>% 
   modify_header(
      label  ~ '') %>% 
   modify_footnote(
      all_stat_cols() ~ 
         'Percentages are based on the number of subjects with non-missing assessments (i.e., the total of the
      subjects in the low, normal, and high categories) within each treatment group') %>% 
   modify_caption(
      '**Table 14-6.03<br>Frequency of Normal and Abnormal (Clinically Significant Change from Previous Visit) Laboratory Values During Treatment**')

Table 14-6.03
Frequency of Normal and Abnormal (Clinically Significant Change from Previous Visit) Laboratory Values During Treatment

	Placebo			Low Dose			High Dose
	Low1	Normal1	High1	Low1	Normal1	High1	Low1	Normal1	High1	p-value
CHEMISTRY
Alanine Aminotransferase	1 (1%)	78 (93%)	5 (6%)	2 (2%)	73 (89%)	7 (9%)	0	74 (93%)	6 (8%)	0.743
Albumin	9 (11%)	74 (88%)	1 (1%)	10 (12%)	71 (87%)	1 (1%)	3 (4%)	77 (96%)	0	0.152
Alkaline Phosphatase	1 (1%)	75 (89%)	8 (10%)	1 (1%)	76 (94%)	4 (5%)	1 (1%)	74 (93%)	5 (6%)	0.891
Aspartate Aminotransferase	0	76 (90%)	8 (10%)	0	76 (93%)	6 (7%)	0	73 (91%)	7 (9%)	0.884
Bilirubin	0	80 (95%)	4 (5%)	0	81 (100%)	0	0	77 (96%)	3 (4%)	0.150
Blood Urea Nitrogen	0	77 (92%)	7 (8%)	0	67 (82%)	15 (18%)	0	73 (91%)	7 (9%)	0.094
Calcium	2 (2%)	82 (98%)	0	3 (4%)	79 (96%)	0	4 (5%)	75 (94%)	1 (1%)	0.538
Chloride	0	77 (92%)	7 (8%)	0	75 (91%)	7 (9%)	0	78 (98%)	2 (3%)	0.223
Cholesterol	3 (4%)	78 (93%)	3 (4%)	4 (5%)	75 (91%)	3 (4%)	4 (5%)	75 (94%)	1 (1%)	0.883
Creatine Kinase	0	73 (87%)	11 (13%)	1 (1%)	72 (88%)	9 (11%)	0	72 (90%)	8 (10%)	0.850
Creatinine	0	80 (95%)	4 (5%)	0	77 (94%)	5 (6%)	0	78 (98%)	2 (3%)	0.593
Gamma Glutamyl Transferase	2 (2%)	76 (90%)	6 (7%)	2 (2%)	74 (90%)	6 (7%)	1 (1%)	72 (90%)	7 (9%)	0.983
Glucose	0	83 (99%)	1 (1%)	0	82 (100%)	0	0	77 (96%)	3 (4%)	0.127
Phosphate	1 (1%)	82 (98%)	1 (1%)	0	79 (98%)	2 (2%)	0	80 (100%)	0	0.697
Potassium	1 (1%)	83 (99%)	0	2 (2%)	79 (98%)	0	1 (1%)	79 (99%)	0	0.846
Protein	1 (1%)	79 (94%)	4 (5%)	1 (1%)	79 (96%)	2 (2%)	0	79 (99%)	1 (1%)	0.673
Sodium	3 (4%)	76 (90%)	5 (6%)	1 (1%)	78 (95%)	3 (4%)	2 (3%)	74 (93%)	4 (5%)	0.859
Urate	1 (1%)	80 (95%)	3 (4%)	1 (1%)	76 (93%)	5 (6%)	1 (1%)	76 (95%)	3 (4%)	0.942
HEMATOLOGY
Basophils	0	84 (100%)	0	0	82 (100%)	0	0	80 (100%)	0	>0.999
Eosinophils	0	84 (100%)	0	0	73 (89%)	9 (11%)	0	76 (95%)	4 (5%)	0.003
Ery. Mean Corpuscular Hemoglobin )	0	82 (98%)	2 (2%)	0	81 (99%)	1 (1%)	1 (1%)	75 (94%)	4 (5%)	0.251
Ery. Mean Corpuscular HGB Concentration	11 (13%)	72 (87%)	0	5 (6%)	76 (94%)	0	1 (1%)	79 (99%)	0	0.009
Ery. Mean Corpuscular Volume	1 (1%)	61 (73%)	21 (25%)	0	70 (86%)	11 (14%)	1 (1%)	67 (84%)	12 (15%)	0.145
Erythrocytes	7 (8%)	77 (92%)	0	16 (20%)	66 (80%)	0	8 (10%)	71 (89%)	1 (1%)	0.081
Hematocrit	3 (4%)	79 (95%)	1 (1%)	6 (7%)	75 (93%)	0	1 (1%)	77 (96%)	2 (3%)	0.172
Hemoglobin	6 (7%)	77 (92%)	1 (1%)	9 (11%)	73 (89%)	0	5 (6%)	74 (93%)	1 (1%)	0.697
Leukocytes	4 (5%)	77 (92%)	3 (4%)	5 (6%)	72 (88%)	5 (6%)	2 (3%)	75 (94%)	3 (4%)	0.734
Lymphocytes	4 (5%)	77 (92%)	3 (4%)	3 (4%)	75 (91%)	4 (5%)	3 (4%)	76 (95%)	1 (1%)	0.765
Monocytes	2 (2%)	78 (93%)	4 (5%)	0	78 (95%)	4 (5%)	0	79 (99%)	1 (1%)	0.241
Platelet	0	84 (100%)	0	1 (1%)	79 (96%)	2 (2%)	2 (3%)	76 (95%)	2 (3%)	0.342
1 Percentages are based on the number of subjects with non-missing assessments (i.e., the total of the subjects in the low, normal, and high categories) within each treatment group

Table 14-6-04

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adlbc_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbc.xpt')  

adlbh_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbh.xpt')

# COMBINE & SUBSEST
adlb <- bind_rows(
   CHEMISTRY = adlbc_orig,
   HEMATOLOGY = adlbh_orig,
   .id = 'ORIG') %>% 
   filter(VISIT != 'SCREENING 1') %>% 
   filter(SAFFL == 'Y', AVISITN != 99) %>% 
   filter(!is.na(VISIT), !is.na(TRTA), !is.na(BNRIND), !is.na(ANRIND), !is.na(PARAM)) %>%
   mutate(TRTP = factor(TRTP, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Low Dose','High Dose'))) %>% 
   mutate(ANRIND = factor(ANRIND,
                          levels = c('N','H'),
                          labels = c('Normal','High')) ) %>% 
   mutate(BNRIND = factor(BNRIND,
                          levels = c('N','H'),
                          labels = c('Normal','High')) ) %>%
   mutate(VISIT = factor(VISIT) %>% fct_reorder(AVISITN)) %>% 
   mutate(PARAM = str_remove_all(PARAM, '\\(.*?\\)') %>% str_trim() ) %>% 
   select(ORIG, PARAM, PARAMCD, USUBJID , TRTP, VISIT, ANRIND, BNRIND) %>% 
   filter(!str_starts(PARAMCD, '_') ) %>% 
   filter(PARAM %in% c('Alanine Aminotransferase', 'Albumin', 'Basophils', 'Eosinophils'))

# NEST
adlb_n <- adlb %>% 
   nest_by(ORIG, PARAM, PARAMCD, TRTP, VISIT)

# TABLE
#+ message = FALSE
adlb_t <- adlb_n %>% 
   mutate(t_cross = list(
      tbl_cross(
         data = data,
         col = ANRIND,
         row = BNRIND,
         label = list(BNRIND ~ str_glue('{VISIT}') ),
         percent = 'row',
         margin = 'row',
         margin_text = 'n',
         missing = 'no') %>% 
         modify_table_body(
            ~.x %>% 
               arrange(row_type, var_type) ) )
   )

# SUMMARISE: STACK
adlb_w <- adlb_t %>% 
   select(-data) %>% 
   pivot_wider(names_from = TRTP,
               values_from = t_cross) %>%
   rowwise() %>% 
   mutate(t_cross = 
             tbl_merge(tbls = list(Placebo, `Low Dose`, `High Dose`),
                       tab_spanner  = c('**Placebo**','**Low Dose**','**High Dose**')) %>% list()
   ) %>% 
   group_by(ORIG, PARAM, PARAMCD) %>% 
   summarise(t_cross = tbl_stack(t_cross, group_header = NULL, quiet = TRUE) %>% list() ) %>% 
   group_by(ORIG) %>% 
   summarise(t_cross = tbl_stack(t_cross, group_header = PARAM, quiet = TRUE) %>% list() ) 

# PRINT
#+ results = 'asis'
for(i in 1:2){
   adlb_w$t_cross[[i]] %>% 
      modify_table_body(
         ~.x %>% 
            mutate(row_type = ifelse(label == 'n', 'level', row_type)) %>% 
            mutate(across(starts_with('stat_'), ~str_replace_all(.x, '0 \\(0%\\)|0 \\(NA%\\)', '0'))) %>% 
            mutate(across(starts_with('stat_'), 
                          ~case_when(
                             label == 'n' ~ str_replace_all(.x, ' \\(.*?\\)', ''),
                             .default = .x) ) ) ) %>% 
      modify_header(
         starts_with('stat_1_') ~ '**Normal at<br>Baseline**',
         starts_with('stat_2_') ~ '**High at<br>Baseline**') %>% 
      modify_footnote(
         all_stat_cols() ~ 'Only subjects with baseline results are included in the summary.<br>
                            There were no subjects with abnormal low values at baseline.') %>% 
      modify_column_alignment(
         all_stat_cols(), 'left') %>% 
      modify_caption(
         str_glue('**Table 14-6.04 {adlb_w$ORIG[[i]]}<br>Shifts of Laboratory Values During Treatment, Categorized Based on Threshold Ranges, by Visit**')) %>% 
      bstfun::bold_italicize_group_labels(bold = TRUE) %>% 
      knitr::knit_print() %>% 
      cat()
}

Table 14-6.04 CHEMISTRY
Shifts of Laboratory Values During Treatment, Categorized Based on Threshold Ranges, by Visit

	Placebo		Low Dose		High Dose
	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1
Alanine Aminotransferase
WEEK 2
Normal	81 (100%)	0	77 (100%)	0	78 (100%)	0
High	0	2 (100%)	0	1 (100%)	0	0
n	81	2	77	1	78	0
WEEK 4
Normal	77 (100%)	0	69 (100%)	0	72 (100%)	0
High	1 (50%)	1 (50%)	0	1 (100%)	0	0
n	78	1	69	1	72	0
WEEK 6
Normal	71 (99%)	1 (1.4%)	59 (100%)	0	66 (100%)	0
High	1 (100%)	0	0	1 (100%)	0	0
n	72	1	59	1	66	0
WEEK 8
Normal	71 (100%)	0	57 (100%)	0	54 (96%)	2 (3.6%)
High	0	1 (100%)	0	1 (100%)	0	0
n	71	1	57	1	54	2
WEEK 12
Normal	65 (98%)	1 (1.5%)	48 (98%)	1 (2.0%)	49 (98%)	1 (2.0%)
High	1 (100%)	0	1 (100%)	0	0	0
n	66	1	49	1	49	1
WEEK 16
Normal	67 (100%)	0	40 (100%)	0	37 (100%)	0
High	1 (100%)	0	0	0	0	0
n	68	0	40	0	37	0
WEEK 20
Normal	64 (100%)	0	29 (100%)	0	31 (100%)	0
High	1 (100%)	0	0	0	0	0
n	65	0	29	0	31	0
WEEK 24
Normal	55 (98%)	1 (1.8%)	25 (100%)	0	30 (100%)	0
High	1 (100%)	0	0	0	0	0
n	56	1	25	0	30	0
WEEK 26
Normal	56 (100%)	0	23 (96%)	1 (4.2%)	27 (100%)	0
High	1 (100%)	0	0	0	0	0
n	57	0	23	1	27	0
Albumin
WEEK 2
Normal	83 (100%)	0	78 (100%)	0	78 (100%)	0
High	0	0	0	0	0	0
n	83	0	78	0	78	0
WEEK 4
Normal	79 (100%)	0	70 (100%)	0	72 (100%)	0
High	0	0	0	0	0	0
n	79	0	70	0	72	0
WEEK 6
Normal	73 (100%)	0	60 (100%)	0	66 (100%)	0
High	0	0	0	0	0	0
n	73	0	60	0	66	0
WEEK 8
Normal	72 (100%)	0	58 (100%)	0	56 (100%)	0
High	0	0	0	0	0	0
n	72	0	58	0	56	0
WEEK 12
Normal	67 (100%)	0	50 (100%)	0	50 (100%)	0
High	0	0	0	0	0	0
n	67	0	50	0	50	0
WEEK 16
Normal	68 (100%)	0	40 (100%)	0	37 (100%)	0
High	0	0	0	0	0	0
n	68	0	40	0	37	0
WEEK 20
Normal	65 (100%)	0	29 (100%)	0	31 (100%)	0
High	0	0	0	0	0	0
n	65	0	29	0	31	0
WEEK 24
Normal	57 (100%)	0	25 (100%)	0	30 (100%)	0
High	0	0	0	0	0	0
n	57	0	25	0	30	0
WEEK 26
Normal	57 (100%)	0	24 (100%)	0	27 (100%)	0
High	0	0	0	0	0	0
n	57	0	24	0	27	0
1 Only subjects with baseline results are included in the summary. There were no subjects with abnormal low values at baseline.

Table 14-6.04 HEMATOLOGY
Shifts of Laboratory Values During Treatment, Categorized Based on Threshold Ranges, by Visit

	Placebo		Low Dose		High Dose
	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1
Basophils
WEEK 2
Normal	82 (100%)	0	77 (100%)	0	77 (100%)	0
High	0	0	0	0	0	0
n	82	0	77	0	77	0
WEEK 4
Normal	78 (100%)	0	69 (100%)	0	69 (100%)	0
High	0	0	0	0	0	0
n	78	0	69	0	69	0
WEEK 6
Normal	72 (100%)	0	59 (100%)	0	62 (100%)	0
High	0	0	0	0	0	0
n	72	0	59	0	62	0
WEEK 8
Normal	71 (100%)	0	56 (100%)	0	55 (100%)	0
High	0	0	0	0	0	0
n	71	0	56	0	55	0
WEEK 12
Normal	65 (100%)	0	48 (100%)	0	49 (100%)	0
High	0	0	0	0	0	0
n	65	0	48	0	49	0
WEEK 16
Normal	67 (100%)	0	39 (100%)	0	36 (100%)	0
High	0	0	0	0	0	0
n	67	0	39	0	36	0
WEEK 20
Normal	64 (100%)	0	29 (100%)	0	31 (100%)	0
High	0	0	0	0	0	0
n	64	0	29	0	31	0
WEEK 24
Normal	57 (100%)	0	24 (100%)	0	29 (100%)	0
High	0	0	0	0	0	0
n	57	0	24	0	29	0
WEEK 26
Normal	56 (100%)	0	24 (100%)	0	26 (100%)	0
High	0	0	0	0	0	0
n	56	0	24	0	26	0
Eosinophils
WEEK 2
Normal	82 (100%)	0	74 (97%)	2 (2.6%)	77 (100%)	0
High	0	0	1 (100%)	0	0	0
n	82	0	75	2	77	0
WEEK 4
Normal	78 (100%)	0	69 (100%)	0	68 (99%)	1 (1.4%)
High	0	0	0	0	0	0
n	78	0	69	0	68	1
WEEK 6
Normal	72 (100%)	0	59 (100%)	0	58 (94%)	4 (6.5%)
High	0	0	0	0	0	0
n	72	0	59	0	58	4
WEEK 8
Normal	71 (100%)	0	53 (95%)	3 (5.4%)	52 (95%)	3 (5.5%)
High	0	0	0	0	0	0
n	71	0	53	3	52	3
WEEK 12
Normal	65 (100%)	0	46 (96%)	2 (4.2%)	49 (100%)	0
High	0	0	0	0	0	0
n	65	0	46	2	49	0
WEEK 16
Normal	67 (100%)	0	39 (100%)	0	35 (97%)	1 (2.8%)
High	0	0	0	0	0	0
n	67	0	39	0	35	1
WEEK 20
Normal	64 (100%)	0	29 (100%)	0	31 (100%)	0
High	0	0	0	0	0	0
n	64	0	29	0	31	0
WEEK 24
Normal	57 (100%)	0	24 (100%)	0	29 (100%)	0
High	0	0	0	0	0	0
n	57	0	24	0	29	0
WEEK 26
Normal	56 (100%)	0	24 (100%)	0	26 (100%)	0
High	0	0	0	0	0	0
n	56	0	24	0	26	0
1 Only subjects with baseline results are included in the summary. There were no subjects with abnormal low values at baseline.

Caution

For the upcoming tables, we will only display 2 chemistry and 2 hematology variables to highlight the code without making the table excessively long.

Table 14-6-05

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adlbc_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbc.xpt')  

adlbh_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbh.xpt')

# COMBINE & SUBSEST
adlb <- bind_rows(
   CHEMISTRY  = adlbc_orig,
   HEMATOLOGY = adlbh_orig,
   .id = 'ORIG') %>% 
   filter(SAFFL == 'Y', ANL01FL == 'Y') %>% 
   filter(!is.na(PARAM), !is.na(TRTP), !is.na(BNRIND), !is.na(ANRIND), AVISITN != 99) 

adlb_n <- adlb %>% 
   mutate(ANRIND = factor(ANRIND, levels = c('N','H'), labels = c('Normal','High') ) ,
          BNRIND = factor(BNRIND, levels = c('N','H'), labels = c('Normal','High')),
          TRTP = factor(TRTP, levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose')) ) %>% 
   select(ORIG, PARAM, PARAMCD, USUBJID, TRTP, ANRIND, BNRIND) %>% 
   drop_na(ANRIND, BNRIND) %>% 
   mutate(ALL = 1) %>% 
   filter(!str_starts(PARAMCD, '_') ) %>% 
   mutate(PARAM = str_remove_all(PARAM, '\\(.*?\\)') %>% str_trim() ) %>% 
   filter(PARAM %in% c('Alanine Aminotransferase', 
                       'Albumin', 
                       'Alkaline Phosphatase',
                       'Aspartate Aminotransferase',
                       'Bilirubin',
                       'Basophils', 
                       'Eosinophils',
                       'Hemoglobin',
                       'Lymphocytes')) %>% 
   nest_by(ORIG, PARAM, PARAMCD)

# CMH FUNCTION
my_cmh <- function(data, variable, by, ...) {
   table( data[['ANRIND']],
          data[['TRTP']],
          data[['BNRIND']] ) %>% 
      mantelhaen.test() %>% 
      broom::tidy() %>% 
      select(method, p.value)
}

# SUMMARISE: STACK
adlb_t <- adlb_n %>% 
   mutate(
      t = list(
         tbl_summary(
            data = data %>% mutate(inter = interaction(BNRIND, TRTP) ),
            by = inter,
            label = list(ALL ~ 'n'),
            include = c(ALL, ANRIND),
            statistic = list(
               ALL ~ '{n}',
               ANRIND ~ '{n} ({p}%)'),
            digits = list(
               ANRIND ~ c(0, 0)) ) %>% 
            add_stat(fns = list( ALL ~ my_cmh) ) )
   ) %>% 
   group_by(ORIG) %>% 
   summarise(t_orig = tbl_stack(tbls = t, group_header = PARAM, quiet = TRUE) %>% list() )

# PRINT
#+ results = 'asis'
for(i in 1){
   adlb_t$t_orig[[1]]  %>% 
      modify_table_body(
         ~.x %>% 
            filter(!(row_type == 'label' & label != 'n')) %>% 
            mutate(across(starts_with('stat_'), ~str_replace_all(.x, '0 \\(0%\\)|0 \\(NA%\\)', '0'))) %>% 
            janitor::remove_empty(which = 'cols')
      ) %>% 
      modify_header(
         label ~ ' ',
         c(stat_1, stat_3, stat_5) ~ '**Normal at<br>Baseline**',
         c(stat_2, stat_4, stat_6) ~ '**High at<br>Baseline**',
         p.value ~ '**p-value**'
      ) %>% 
      modify_spanning_header(
         c(stat_1, stat_2) ~ '**Placebo (N=86)**',
         c(stat_3, stat_4) ~ '**Xan. Low (N=84)**',
         c(stat_5, stat_6) ~ '**Xan. High (N=84)**'
      ) %>% 
      modify_column_hide(method) %>% 
      modify_column_alignment(
         all_stat_cols(), 'left'
      ) %>% 
      modify_fmt_fun( c(p.value) ~ function(x) style_pvalue(x, digits = 3) ) %>% 
      modify_footnote(
         all_stat_cols() ~ 
            'NOTES: Only subjects with baseline results are included in the summary.<br>
      There were no subjects with abnormal low values at baseline.<br>
      A subject is counted only once for each analyte. A change will be considered <br>
      shifting from normal at baseline to abnormal or from abnormal at baseline to normal at any visit during the treatment.<br>
      The treatment period is defined as any planned visit after Week 0 (Visit 3), up to and including Week 24 (Visit 12).',
      p.value ~ 
         'CMH test for general association, controlling for status at baseline.') %>% 
      modify_column_indent(columns = label, undo = TRUE) %>% 
      modify_caption(
         '**Table 14-6.05 {adlb_t$ORIG[[i]]} <br>Shifts of Laboratory Values During Treatment, Categorized Based on Threshold Ranges**') %>% 
      bstfun::bold_italicize_group_labels(bold = TRUE) %>% 
      knitr::knit_print() %>% 
      cat()
}

Table 14-6.05 CHEMISTRY
Shifts of Laboratory Values During Treatment, Categorized Based on Threshold Ranges

	Placebo (N=86)		Xan. Low (N=84)		Xan. High (N=84)		p-value2
	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1
Alanine Aminotransferase
n	82	2	79	1	80	0	0.828
Normal	80 (98%)	0	78 (99%)	0	78 (98%)	0
High	2 (2%)	2 (100%)	1 (1%)	1 (100%)	2 (3%)	0
Albumin
n	84	0	80	0	80	0
Normal	84 (100%)	0	80 (100%)	0	80 (100%)	0
High	0	0	0	0	0	0
Alkaline Phosphatase
n	82	2	79	0	79	1	0.611
Normal	81 (99%)	0	78 (99%)	0	79 (100%)	0
High	1 (1%)	2 (100%)	1 (1%)	0	0	1 (100%)
Aspartate Aminotransferase
n	82	2	79	1	80	0	0.770
Normal	79 (96%)	1 (50%)	77 (97%)	1 (100%)	78 (98%)	0
High	3 (4%)	1 (50%)	2 (3%)	0	2 (3%)	0
Bilirubin
n	84	0	79	0	79	1
Normal	83 (99%)	0	79 (100%)	0	77 (97%)	0
High	1 (1%)	0	0	0	2 (3%)	1 (100%)
1 NOTES: Only subjects with baseline results are included in the summary. There were no subjects with abnormal low values at baseline. A subject is counted only once for each analyte. A change will be considered shifting from normal at baseline to abnormal or from abnormal at baseline to normal at any visit during the treatment. The treatment period is defined as any planned visit after Week 0 (Visit 3), up to and including Week 24 (Visit 12).
2 CMH test for general association, controlling for status at baseline.

Table 14-6-06

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adlbhy_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adlbhy.xpt')  

# COMBINE & SUBSEST
adlbhy <-adlbhy_orig  %>% 
   filter(SAFFL == 'Y', PARAMCD %in% c('TRANSHY', 'HYLAW'), !is.na(BASE), AVISITN > 0) %>% 
   group_by(USUBJID) %>%
   filter(AVAL    == max(AVAL)) %>%
   filter(AVISITN == max(AVISITN)) %>%
   ungroup()

adlbhy2 <- adlbhy %>%
   select(USUBJID, PARAMCD, PARAM, TRTP, BASE, AVAL) %>% 
   mutate(AVAL = factor(AVAL, labels = c('Normal','High') ) ,
          BASE = factor(BASE, labels = c('Normal','High')),
          TRTP = factor(TRTP, levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose')) ) %>% 
   mutate(ALL = 1)

adlbhy_n <- adlbhy2 %>% 
   nest_by(PARAM, PARAMCD) %>% 
   arrange(desc(PARAM))

# CMH FUNCTION
my_cmh <- function(data, variable, by, ...) {
   table( data[['BASE']],
          data[['TRTP']],
          data[['AVAL']] ) %>% 
      mantelhaen.test() %>% 
      broom::tidy() %>% 
      select(method, p.value)
}

# SUMMARISE: STACK
adlbhy_t <- adlbhy_n %>% 
   mutate(
      t = list(
         tbl_summary(
            data = data %>% mutate(inter = interaction(BASE, TRTP) ),
            by = inter,
            label = list(ALL ~ 'n'),
            include = c(ALL, AVAL),
            statistic = list(
               ALL ~ '{n}',
               AVAL ~ '{n} ({p}%)'
            ),
            digits = list(
               AVAL ~ c(0, 0))) %>% 
            add_stat(fns = list( ALL ~ my_cmh) ) 
      )
   ) 

# PRINT
with(adlbhy_t,
     tbl_stack(tbls = t, group_header = PARAM, quiet = TRUE) )  %>% 
   modify_table_body(
      ~.x %>% 
         filter(!(row_type == 'label' & label != 'n')) %>% 
         mutate(across(starts_with('stat_'), ~str_replace_all(.x, '0 \\(0%\\)|0 \\(NA%\\)', '0'))) %>% 
         janitor::remove_empty(which = 'cols')
   ) %>% 
   modify_header(
      label ~ ' ',
      c(stat_1, stat_3, stat_5) ~ '**Normal at<br>Baseline**',
      c(stat_2, stat_4, stat_6) ~ '**High at<br>Baseline**',
      p.value ~ '**p-value**'
   ) %>% 
   modify_spanning_header(
      c(stat_1, stat_2) ~ '**Placebo (N=86)**',
      c(stat_3, stat_4) ~ '**Xan. Low (N=84)**',
      c(stat_5, stat_6) ~ '**Xan. High (N=84)**'
   ) %>% 
   modify_column_hide(method) %>% 
   modify_column_alignment(
      all_stat_cols(), 'left'
   ) %>%
   modify_fmt_fun( c(p.value) ~ function(x) style_pvalue(x, digits = 3) ) %>% 
   modify_footnote(
      all_stat_cols() ~ 
         'NOTES: Only subjects with baseline results are included in the summary.<br>
      The single subject with elevated transaminase and elevated bilirubin also had elevated alk phos (>3xULN).<br>
      There were no subjects with abnormal low values at baseline.<br>
      A subject is counted only once for each analyte. A change will be considered shifting from normal at<br>
      baseline to abnormal or from abnormal at baseline to normal at any visit during the treatment.<br> 
      The treatment period is defined as any planned visit after Week 0 (Visit 3), up to and including Week 24 (Visit 12).',
      p.value ~ 'CMH test for general association, controlling for status at baseline.') %>% 
   modify_column_indent(columns = label, undo = TRUE) %>% 
   modify_caption(
      "**Table 14-6.06<br>Shifts of Hy's Law Values During Treatment**") %>% 
   bstfun::bold_italicize_group_labels(bold = TRUE)

Table 14-6.06
Shifts of Hy’s Law Values During Treatment

	Placebo (N=86)		Xan. Low (N=84)		Xan. High (N=84)		p-value2
	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1	Normal at Baseline1	High at Baseline1
Transaminase 1.5 x ULN
n	82	2	78	2	80	0	0.232
Normal	79 (96%)	0	77 (99%)	1 (50%)	77 (96%)	0
High	3 (4%)	2 (100%)	1 (1%)	1 (50%)	3 (4%)	0
Total Bili 1.5 x ULN and Transaminase 1.5 x ULN
n	80	0	78	0	77	0
Normal	79 (99%)	0	78 (100%)	0	77 (100%)	0
High	1 (1%)	0	0	0	0	0
1 NOTES: Only subjects with baseline results are included in the summary. The single subject with elevated transaminase and elevated bilirubin also had elevated alk phos (>3xULN). There were no subjects with abnormal low values at baseline. A subject is counted only once for each analyte. A change will be considered shifting from normal at baseline to abnormal or from abnormal at baseline to normal at any visit during the treatment. The treatment period is defined as any planned visit after Week 0 (Visit 3), up to and including Week 24 (Visit 12).
2 CMH test for general association, controlling for status at baseline.

Table 14-7

Table 14-7-01

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
advs_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/advs.xpt')

advs <- advs_orig %>% 
   filter(SAFFL == 'Y', ANL01FL == 'Y') %>% 
   mutate(EOTFL = ifelse(AVISIT == 'End of Treatment', 'Y', ''),
          W24FL = ifelse(AVISIT == 'Week 24', 'Y', '') ) %>% 
   filter(EOTFL == 'Y' | W24FL == 'Y' | ABLFL == 'Y') %>%
   filter(PARAM %in% c('Diastolic Blood Pressure (mmHg)',
                       'Pulse Rate (BEATS/MIN)',
                       'Systolic Blood Pressure (mmHg)')) %>% 
   mutate(TRTP = factor(TRTP,
                        levels = c('Placebo', 'Xanomeline Low Dose', 'Xanomeline High Dose'),
                        labels = c('Placebo', 'Xan. Low', 'Xan. High')),
          AVISIT = factor(AVISIT,
                          levels = c('Baseline', 'Week 24', 'End of Treatment'),
                          labels = c('Baseline','Week 24', 'End of Trt.')),
          PARAM = factor(PARAM,
                         levels = c('Systolic Blood Pressure (mmHg)',
                                    'Diastolic Blood Pressure (mmHg)',
                                    'Pulse Rate (BEATS/MIN)'),
                         labels = c('Systolic Blood Pressure (mmHg)',
                                    'Diastolic Blood Pressure (mmHg)',
                                    'Pulse (bpm)'))
   ) %>% 
   select( PARAM, PARAMCD, USUBJID, ATPT, TRTP, AVISIT, AVAL)


# NEST
advs_n <- advs %>% 
   nest_by(PARAM, PARAMCD, ATPT, TRTP, AVISIT)

# STATS
my_stats <- function(data, variable, ...) {
   data %>% 
      select({{variable}}) %>% 
      drop_na() %>% 
      summarise( # n = length(AVAL),
         Mean = mean(AVAL),
         SD = sd(AVAL),
         Median = median(AVAL),
         `Min.` = min(AVAL),
         `Max.` = max(AVAL))
}

# TABLES
advs_t <- advs_n %>% 
   mutate( t = list( 
      tbl_summary(
         data = data,
         include = AVAL,
         label = list(AVAL ~ str_glue('{AVISIT}')),
         statistic = list(everything() ~ '{N_nonmiss}'),
         missing = 'no'
      ) %>% 
         add_stat(
            fns = everything() ~ my_stats
         )
   )
   )

# STACK
t <- with(advs_t,       
          tbl_stack(t, group_header = str_glue('{PARAM}_{ATPT}_{TRTP}'), quiet = TRUE) )

# MODIFY OBJECT TABLE
t$table_body <- t$table_body %>% 
   separate(groupname_col, 
            into = c('groupname_col_1',
                     'groupname_col_2',
                     'groupname_col_3' ),
            sep = '_') %>% 
   mutate(across(starts_with('groupname_col_'), 
                 ~case_when(
                    is.na(lag(.x))             ~ .x,
                    .x == lag(.x) ~ NA_character_, TRUE ~ .x) ) 
   )

# MODIFY OBJECT STYLING HEADER
t$table_styling$header <- 
   t$table_styling$header %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_1', label = 'Measure') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_2', label = 'Position') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_3', label = 'Treatment') ) %>% 
   slice(-1)

# PRINT
t %>% 
   modify_fmt_fun(
      list(
         Mean ~ \(x) style_number(x, digits = 1),
         SD ~ \(x) style_number(x, digits = 2),
         c('Median','Min.','Max.') ~ \(x) style_number(x, digits = 1) ) ) %>% 
   modify_header(
      label ~ '**Planned<br>Relative<br>Time**',
      stat_0 ~ '**n**',
      Mean ~ '**Mean**',              
      SD ~ '**SD**',                 
      Median~ '**Median**',             
      Min. ~ '**Min.**',               
      Max. ~ '**Max.**',  
      groupname_col_1  ~ '**Measure**',
      groupname_col_2 ~ '**Position**',
      groupname_col_3 ~ '**Treatment**') %>% 
   modify_table_styling(
      columns = label,
      rows = label == 'End of Trt.',
      footnote = 'End of treatment is the last on-treatment assessment of the specified vital sign (on or before the Week 24 visit).' ) %>% 
   modify_footnote(
      all_stat_cols() ~ NA) %>% 
   modify_column_alignment(where(is.numeric), 'right') %>% 
   modify_caption(
      '**Table 14-7.01<br>Summary of Vital Signs at Baseline and End of Treatment**')

Table 14-7.01
Summary of Vital Signs at Baseline and End of Treatment

Measure	Position	Treatment	Planned Relative Time	n	Mean	SD	Median	Min.	Max.
Systolic Blood Pressure (mmHg)	AFTER LYING DOWN FOR 5 MINUTES	Placebo	Baseline	85	138.6	16.75	140.0	90.0	180.0
			Week 24	59	135.8	17.30	131.0	100.0	180.0
			End of Trt.1	82	134.9	19.44	132.5	88.0	180.0
		Xan. Low	Baseline	84	138.8	16.55	138.0	100.0	178.0
			Week 24	27	134.1	16.74	136.0	100.0	173.0
			End of Trt.1	72	135.5	17.22	133.0	102.0	190.0
		Xan. High	Baseline	84	140.1	17.82	141.0	100.0	188.0
			Week 24	30	132.2	18.18	130.0	101.0	178.0
			End of Trt.1	72	130.9	17.48	128.0	104.0	176.0
	AFTER STANDING FOR 1 MINUTE	Placebo	Baseline	85	135.3	17.89	134.0	90.0	180.0
			Week 24	59	133.5	19.23	130.0	90.0	199.0
			End of Trt.1	82	130.9	17.50	130.0	84.0	172.0
		Xan. Low	Baseline	84	135.6	18.04	136.0	100.0	186.0
			Week 24	27	131.0	17.82	130.0	92.0	168.0
			End of Trt.1	72	131.1	18.54	130.0	99.0	180.0
		Xan. High	Baseline	84	137.3	19.71	138.0	100.0	194.0
			Week 24	30	130.4	20.83	128.0	96.0	198.0
			End of Trt.1	73	127.3	16.51	124.0	99.0	170.0
	AFTER STANDING FOR 3 MINUTES	Placebo	Baseline	85	136.5	18.77	136.0	80.0	184.0
			Week 24	59	134.8	17.35	131.0	100.0	190.0
			End of Trt.1	82	131.3	19.30	130.0	80.0	176.0
		Xan. Low	Baseline	84	136.4	18.11	134.5	104.0	182.0
			Week 24	27	131.0	17.92	130.0	100.0	168.0
			End of Trt.1	72	132.2	18.56	130.0	98.0	200.0
		Xan. High	Baseline	84	138.8	18.75	138.0	100.0	186.0
			Week 24	30	129.2	16.95	126.0	90.0	172.0
			End of Trt.1	73	128.1	16.02	128.0	95.0	179.0
Diastolic Blood Pressure (mmHg)	AFTER LYING DOWN FOR 5 MINUTES	Placebo	Baseline	85	75.7	11.09	76.0	40.0	99.0
			Week 24	59	72.9	11.32	74.0	44.0	109.0
			End of Trt.1	82	74.0	10.39	74.0	46.0	100.0
		Xan. Low	Baseline	84	76.3	9.77	76.0	57.0	100.0
			Week 24	27	76.1	9.14	76.0	60.0	90.0
			End of Trt.1	72	73.6	8.33	72.0	56.0	90.0
		Xan. High	Baseline	84	77.2	9.80	78.0	51.0	98.0
			Week 24	30	73.9	9.23	74.0	60.0	92.0
			End of Trt.1	73	73.5	9.58	74.0	50.0	94.0
	AFTER STANDING FOR 1 MINUTE	Placebo	Baseline	85	77.9	10.63	78.0	51.0	104.0
			Week 24	59	74.2	12.89	74.0	45.0	117.0
			End of Trt.1	82	74.3	11.67	75.0	39.0	100.0
		Xan. Low	Baseline	84	76.2	10.14	78.0	54.0	98.0
			Week 24	27	76.3	10.28	78.0	60.0	98.0
			End of Trt.1	72	73.9	9.55	73.0	51.0	92.0
		Xan. High	Baseline	84	78.1	10.77	78.0	56.0	108.0
			Week 24	30	74.9	11.00	76.0	50.0	97.0
			End of Trt.1	73	76.2	10.49	77.0	50.0	98.0
	AFTER STANDING FOR 3 MINUTES	Placebo	Baseline	85	77.7	11.00	78.0	46.0	110.0
			Week 24	59	74.3	11.38	74.0	51.0	110.0
			End of Trt.1	82	73.8	11.17	74.0	49.0	100.0
		Xan. Low	Baseline	84	76.6	10.93	76.0	48.0	108.0
			Week 24	27	76.2	10.18	76.0	57.0	98.0
			End of Trt.1	72	74.0	9.57	73.0	57.0	102.0
		Xan. High	Baseline	84	79.6	10.19	80.0	51.0	104.0
			Week 24	30	76.0	10.63	78.5	50.0	98.0
			End of Trt.1	73	75.8	10.96	78.0	50.0	99.0
Pulse (bpm)	AFTER LYING DOWN FOR 5 MINUTES	Placebo	Baseline	85	70.4	10.46	70.0	51.0	100.0
			Week 24	59	69.1	9.46	68.0	50.0	92.0
			End of Trt.1	82	70.2	11.14	68.0	50.0	104.0
		Xan. Low	Baseline	84	68.8	9.52	68.0	50.0	88.0
			Week 24	27	68.1	9.28	68.0	52.0	90.0
			End of Trt.1	72	67.1	10.47	66.0	48.0	100.0
		Xan. High	Baseline	84	70.1	9.27	68.0	52.0	98.0
			Week 24	30	69.3	11.88	68.0	47.0	96.0
			End of Trt.1	72	67.4	10.30	67.0	47.0	92.0
	AFTER STANDING FOR 1 MINUTE	Placebo	Baseline	85	75.5	12.68	76.0	56.0	133.0
			Week 24	59	72.8	8.98	74.0	52.0	88.0
			End of Trt.1	82	74.5	10.18	75.0	52.0	108.0
		Xan. Low	Baseline	84	73.5	10.59	72.0	53.0	100.0
			Week 24	27	72.1	9.53	74.0	52.0	88.0
			End of Trt.1	72	73.1	11.06	72.0	51.0	104.0
		Xan. High	Baseline	84	75.0	10.89	72.0	56.0	104.0
			Week 24	30	73.4	11.93	72.0	54.0	98.0
			End of Trt.1	73	72.7	10.61	72.0	52.0	100.0
	AFTER STANDING FOR 3 MINUTES	Placebo	Baseline	85	74.6	11.94	74.0	54.0	134.0
			Week 24	59	72.8	8.73	74.0	56.0	88.0
			End of Trt.1	82	74.6	10.00	75.5	52.0	100.0
		Xan. Low	Baseline	84	72.3	10.99	70.0	51.0	104.0
			Week 24	27	70.7	10.78	72.0	52.0	96.0
			End of Trt.1	72	71.6	9.94	72.0	53.0	97.0
		Xan. High	Baseline	84	74.0	10.76	72.0	52.0	100.0
			Week 24	30	72.4	11.92	71.5	54.0	96.0
			End of Trt.1	73	72.5	10.06	72.0	54.0	100.0
1 End of treatment is the last on-treatment assessment of the specified vital sign (on or before the Week 24 visit).

Warning

Table 14-7-01. Difference in values for End of Treatment for all groups. The End of Treatment flag is used as is from the ADaM which indicates there are likely discrepancies for end of treatment between the original CDISC Pilot analysis data and the PHUSE CDISC Pilot replication data.

Table 14-7-02

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
advs_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/advs.xpt')

advs <- advs_orig %>% 
   filter(SAFFL == 'Y', ANL01FL == 'Y', !is.na(BASE)) %>% 
   mutate(EOTFL = ifelse(AVISIT == 'End of Treatment', 'Y', ''),
          W24FL = ifelse(AVISIT == 'Week 24', 'Y', '') ) %>% 
   filter(EOTFL == 'Y' | W24FL == 'Y' | ABLFL == 'Y') %>%
   filter(PARAM %in% c('Diastolic Blood Pressure (mmHg)',
                       'Pulse Rate (BEATS/MIN)',
                       'Systolic Blood Pressure (mmHg)')) %>% 
   mutate(TRTP = factor(TRTP,
                        levels = c('Placebo', 'Xanomeline Low Dose', 'Xanomeline High Dose'),
                        labels = c('Placebo', 'Xan. Low', 'Xan. High')),
          AVISIT = factor(AVISIT,
                          levels = c('Baseline', 'Week 24', 'End of Treatment'),
                          labels = c('Baseline','Week 24', 'End of Trt.')),
          PARAM = factor(PARAM,
                         levels = c('Systolic Blood Pressure (mmHg)',
                                    'Diastolic Blood Pressure (mmHg)',
                                    'Pulse Rate (BEATS/MIN)'),
                         labels = c('Systolic Blood Pressure (mmHg)',
                                    'Diastolic Blood Pressure (mmHg)',
                                    'Pulse (bpm)'))) %>% 
   filter(AVISIT != 'Baseline') %>% 
   select( PARAM, PARAMCD, USUBJID, ATPT, TRTP, AVISIT, CHG)

# NEST
advs_n <- advs %>% 
   nest_by(PARAM, PARAMCD, ATPT, TRTP, AVISIT)

# STATS
my_stats <- function(data, variable, ...) {
   data %>% 
      select({{variable}}) %>% 
      drop_na() %>% 
      summarise( # n = length(AVAL),
         Mean = mean(CHG),
         SD = sd(CHG),
         Median = median(CHG),
         `Min.` = min(CHG),
         `Max.` = max(CHG))}

# TABLES
advs_t <- advs_n %>% 
   mutate( t = list( 
      tbl_summary(
         data = data,
         include = CHG,
         label = list(CHG ~ str_glue('{AVISIT}')),
         statistic = list(everything() ~ '{N_nonmiss}'),
         missing = 'no') %>% 
         add_stat(
            fns = everything() ~ my_stats))
   )

# STACK
t <- with(advs_t,       
          tbl_stack(t, group_header = str_glue('{PARAM}_{ATPT}_{TRTP}'), quiet = TRUE))

# MODIFY OBJECT TABLE
t$table_body <- t$table_body %>% 
   separate(groupname_col, 
            into = c('groupname_col_1',
                     'groupname_col_2',
                     'groupname_col_3' ),
            sep = '_') %>% 
   mutate(groupname_col_4 = case_match(
      groupname_col_3,
      'Placebo' ~ '86',
      'Xan. Low' ~ '84',
      'Xan. High' ~ '84 '), .after = groupname_col_3) %>% 
   mutate(across(starts_with('groupname_col_'), 
                 ~case_when(
                    is.na(lag(.x))             ~ .x,
                    .x == lag(.x) ~ NA_character_, TRUE ~ .x)) ) 

# MODIFY OBJECT STYLING HEADER
t$table_styling$header <- 
   t$table_styling$header %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_1', label = 'Measure') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_2', label = 'Position') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_3', label = 'Treatment') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_4', label = 'N') ) %>% 
   slice(-1)

# PRINT
t %>% 
   modify_fmt_fun(
      list(
         Mean ~ \(x) style_number(x, digits = 1),
         SD ~ \(x) style_number(x, digits = 2),
         c('Median','Min.','Max.') ~ \(x) style_number(x, digits = 1))) %>% 
   modify_header(
      label ~ '**Planned<br>Relative<br>Time**',
      stat_0 ~ '**n**',
      Mean ~ '**Mean**',              
      SD ~ '**SD**',                 
      Median~ '**Median**',             
      Min. ~ '**Min.**',               
      Max. ~ '**Max.**',  
      groupname_col_1  ~ '**Measure**',
      groupname_col_2 ~ '**Position**',
      groupname_col_3 ~ '**Treatment**',
      groupname_col_4 ~ '**N**') %>% 
   modify_table_styling(
      columns = label,
      rows = label == 'End of Trt.',
      footnote = 'End of treatment is the last on-treatment assessment of the specified vital sign (on or before the Week 24 visit).') %>% 
   modify_footnote(all_stat_cols() ~ NA) %>% 
   modify_column_alignment(where(is.numeric), 'right') %>% 
   modify_caption(
      '**Table 14-7.02<br>Summary of Vital Signs Change from Baseline at End of Treatment**')

Table 14-7.02
Summary of Vital Signs Change from Baseline at End of Treatment

Measure	Position	Treatment	N	Planned Relative Time	n	Mean	SD	Median	Min.	Max.
Systolic Blood Pressure (mmHg)	AFTER LYING DOWN FOR 5 MINUTES	Placebo	86	Week 24	58	-2.1	14.73	-4.0	-28.0	50.0
				End of Trt.1	81	-3.7	18.98	-5.0	-46.0	48.0
		Xan. Low	84	Week 24	27	-0.3	17.19	2.0	-48.0	30.0
				End of Trt.1	72	-3.5	16.69	-4.0	-42.0	34.0
		Xan. High	84	Week 24	30	-5.6	17.18	-7.0	-36.0	26.0
				End of Trt.1	72	-8.9	16.54	-10.0	-54.0	30.0
	AFTER STANDING FOR 1 MINUTE	Placebo	86	Week 24	58	-1.7	16.87	0.0	-32.0	40.0
				End of Trt.1	81	-4.9	18.42	-2.0	-48.0	48.0
		Xan. Low	84	Week 24	27	-0.1	17.73	-1.0	-30.0	48.0
				End of Trt.1	72	-4.3	16.61	-3.0	-52.0	40.0
		Xan. High	84	Week 24	30	-6.3	19.49	-9.0	-36.0	42.0
				End of Trt.1	73	-10.0	18.41	-14.0	-62.0	35.0
	AFTER STANDING FOR 3 MINUTES	Placebo	86	Week 24	58	-1.0	15.80	-3.5	-36.0	38.0
				End of Trt.1	81	-5.5	17.57	-6.0	-50.0	48.0
		Xan. Low	84	Week 24	27	-0.1	16.20	0.0	-30.0	30.0
				End of Trt.1	72	-3.9	16.83	-4.0	-52.0	60.0
		Xan. High	84	Week 24	30	-9.0	16.88	-8.0	-40.0	30.0
				End of Trt.1	73	-10.3	16.79	-10.0	-48.0	30.0
Diastolic Blood Pressure (mmHg)	AFTER LYING DOWN FOR 5 MINUTES	Placebo	86	Week 24	58	-0.8	10.82	-0.5	-18.0	41.0
				End of Trt.1	81	-1.5	9.41	-1.0	-24.0	20.0
		Xan. Low	84	Week 24	27	-0.9	7.71	-2.0	-20.0	16.0
				End of Trt.1	72	-3.5	8.56	-2.0	-30.0	16.0
		Xan. High	84	Week 24	30	-2.2	9.20	-1.0	-20.0	21.0
				End of Trt.1	73	-3.5	9.20	-4.0	-24.0	20.0
	AFTER STANDING FOR 1 MINUTE	Placebo	86	Week 24	58	-2.3	10.08	-4.0	-23.0	24.0
				End of Trt.1	81	-3.3	9.96	-2.0	-27.0	24.0
		Xan. Low	84	Week 24	27	1.0	7.30	2.0	-20.0	18.0
				End of Trt.1	72	-2.2	9.20	-2.0	-30.0	20.0
		Xan. High	84	Week 24	30	-2.3	10.85	-7.0	-18.0	22.0
				End of Trt.1	73	-2.2	11.77	-1.0	-34.0	22.0
	AFTER STANDING FOR 3 MINUTES	Placebo	86	Week 24	58	-2.3	9.56	-3.5	-22.0	20.0
				End of Trt.1	81	-3.9	8.77	-4.0	-26.0	16.0
		Xan. Low	84	Week 24	27	-1.6	8.29	0.0	-20.0	10.0
				End of Trt.1	72	-3.0	10.19	-2.0	-24.0	38.0
		Xan. High	84	Week 24	30	-2.1	9.77	-3.5	-20.0	16.0
				End of Trt.1	73	-3.7	10.47	-2.0	-40.0	20.0
Pulse (bpm)	AFTER LYING DOWN FOR 5 MINUTES	Placebo	86	Week 24	58	-0.3	8.77	-1.0	-24.0	24.0
				End of Trt.1	81	0.1	10.48	0.0	-24.0	40.0
		Xan. Low	84	Week 24	27	-1.6	10.53	0.0	-24.0	25.0
				End of Trt.1	72	-1.4	11.40	-2.0	-24.0	32.0
		Xan. High	84	Week 24	30	-2.0	11.16	-2.0	-34.0	20.0
				End of Trt.1	72	-2.1	9.13	-2.0	-26.0	24.0
	AFTER STANDING FOR 1 MINUTE	Placebo	86	Week 24	58	-1.7	11.72	0.5	-53.0	18.0
				End of Trt.1	81	-0.9	11.72	-2.0	-45.0	44.0
		Xan. Low	84	Week 24	27	-1.3	9.05	0.0	-20.0	12.0
				End of Trt.1	72	0.0	13.09	-1.5	-24.0	34.0
		Xan. High	84	Week 24	30	-1.8	13.41	-3.0	-36.0	20.0
				End of Trt.1	73	-1.6	9.57	-2.0	-28.0	22.0
	AFTER STANDING FOR 3 MINUTES	Placebo	86	Week 24	58	-1.5	10.47	0.0	-46.0	14.0
				End of Trt.1	81	0.1	11.09	0.0	-46.0	32.0
		Xan. Low	84	Week 24	27	-2.1	8.77	-2.0	-20.0	16.0
				End of Trt.1	72	-0.2	11.71	-1.0	-26.0	29.0
		Xan. High	84	Week 24	30	-2.7	11.12	-2.0	-40.0	14.0
				End of Trt.1	73	-1.2	9.36	0.0	-32.0	22.0
1 End of treatment is the last on-treatment assessment of the specified vital sign (on or before the Week 24 visit).

Warning

Table 14-7-02. Difference in values throughout table. The End of Treatment flag is used as is from the ADaM which indicates there are likely discrepancies for end of treatment between the original CDISC Pilot analysis data and the PHUSE CDISC Pilot replication data.

Table 14-7-03

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
advs_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/advs.xpt')

advs <- advs_orig %>% 
   filter(SAFFL == 'Y', ANL01FL == 'Y', !is.na(BASE)) %>% 
   mutate(EOTFL = ifelse(AVISIT == 'End of Treatment', 'Y', ''),
          W24FL = ifelse(AVISIT == 'Week 24', 'Y', '') ) %>% 
   filter(EOTFL == 'Y' | W24FL == 'Y' | ABLFL == 'Y') %>%
   filter(PARAM %in% c('Weight (kg)')) %>% 
   mutate(TRTP = factor(TRTP,
                        levels = c('Placebo', 'Xanomeline Low Dose', 'Xanomeline High Dose'),
                        labels = c('Placebo', 'Xan. Low', 'Xan. High')),
          AVISIT = factor(AVISIT,
                          levels = c('Baseline', 'Week 24', 'End of Treatment'),
                          labels = c('Baseline','Week 24', 'End of Trt.'))) %>% 
   select( PARAM, PARAMCD, USUBJID, TRTP, AVISIT, AVAL, CHG)

advs_l <- advs %>% 
   pivot_longer( cols = c('AVAL','CHG'),
                 names_to = 'VAR',
                 values_to = 'AVAL') %>% 
   filter(!(VAR == 'CHG' & AVISIT == 'Baseline'))

# NEST
advs_n <- advs_l %>% 
   nest_by(PARAM, PARAMCD, VAR, TRTP, AVISIT)

# STATS
my_stats <- function(data, variable, ...) {
   data %>% 
      select({{variable}}) %>% 
      drop_na() %>% 
      summarise( # n = length(AVAL),
         Mean = mean(AVAL),
         SD = sd(AVAL),
         Median = median(AVAL),
         `Min.` = min(AVAL),
         `Max.` = max(AVAL))
}

# TABLES
advs_t <- advs_n %>% 
   mutate( t = list( 
      tbl_summary(
         data = data,
         include = AVAL,
         label = list(AVAL ~ str_glue('{AVISIT}')),
         statistic = list(everything() ~ '{N_nonmiss}'),
         missing = 'no') %>% 
         add_stat(
            fns = everything() ~ my_stats))
   )

# STACK
t <- with(advs_t,       
          tbl_stack(t, group_header = str_glue('{PARAM}_{VAR}_{TRTP}'), quiet = TRUE))

# MODIFY OBJECT TABLE
t$table_body <- t$table_body %>% 
   separate(groupname_col, 
            into = c('groupname_col_1',
                     'groupname_col_2',
                     'groupname_col_3' ),
            sep = '_') %>% 
   mutate(groupname_col_4 = case_match(
      groupname_col_3,
      'Placebo' ~ '86',
      'Xan. Low' ~ '84',
      'Xan. High' ~ '84 '), .after = groupname_col_3) %>% 
   mutate(across(starts_with('groupname_col_'), 
                 ~case_when(
                    is.na(lag(.x))             ~ .x,
                    .x == lag(.x) ~ NA_character_, TRUE ~ .x)) ) 

# MODIFY OBJECT STYLING HEADER
t$table_styling$header <- 
   t$table_styling$header %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_1', label = 'Measure') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_2', label = 'Position') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_3', label = 'Treatment') ) %>% 
   add_row( t$table_styling$header[1,] %>% mutate(column = 'groupname_col_4', label = 'N') ) %>% 
   slice(-1)

# PRINT
t %>% 
   modify_fmt_fun(
      list(
         Mean ~ \(x) style_number(x, digits = 1),
         SD ~ \(x) style_number(x, digits = 2),
         c('Median','Min.','Max.') ~ \(x) style_number(x, digits = 1))) %>% 
   modify_header(
      label ~ '**Planned<br>Relative<br>Time**',
      stat_0 ~ '**n**',
      Mean ~ '**Mean**',              
      SD ~ '**SD**',                 
      Median~ '**Median**',             
      Min. ~ '**Min.**',               
      Max. ~ '**Max.**',  
      groupname_col_1  ~ '**Measure**',
      groupname_col_2 ~ '**Position**',
      groupname_col_3 ~ '**Treatment**',
      groupname_col_4 ~ '**N**') %>% 
   modify_table_styling(
      columns = label,
      rows = label == 'End of Trt.',
      footnote = 'End of treatment is the last on-treatment assessment of the specified vital sign (on or before the Week 24 visit).') %>% 
   modify_footnote(all_stat_cols() ~ NA) %>% 
   modify_column_alignment(where(is.numeric), 'right') %>% 
   modify_caption(
      '**Table 14-7.03<br>Summary of Weight Change from Baseline at End of Treatment**')

Table 14-7.03
Summary of Weight Change from Baseline at End of Treatment

Measure	Position	Treatment	N	Planned Relative Time	n	Mean	SD	Median	Min.	Max.
Weight (kg)	AVAL	Placebo	86	Baseline	86	62.8	12.77	60.6	34.0	86.2
				Week 24	59	63.2	12.58	63.5	34.0	86.6
				End of Trt.1	82	63.0	12.71	63.1	34.5	87.1
		Xan. Low	84	Baseline	83	67.3	14.13	64.9	45.4	106.1
				Week 24	27	67.4	14.07	62.6	45.5	106.1
				End of Trt.1	71	67.0	14.30	66.2	41.7	104.8
		Xan. High	84	Baseline	84	70.0	14.65	69.2	41.7	108.0
				Week 24	30	71.1	15.82	68.7	49.9	105.7
				End of Trt.1	72	69.3	14.48	68.0	42.2	105.2
	CHG	Placebo	86	Week 24	59	0.1	2.30	0.0	-4.5	8.2
				End of Trt.1	82	0.0	1.89	0.0	-5.0	5.0
		Xan. Low	84	Week 24	27	-0.3	2.04	0.0	-5.4	3.2
				End of Trt.1	71	-0.3	2.73	0.0	-14.5	5.9
		Xan. High	84	Week 24	30	1.0	6.47	-0.2	-4.5	33.3
				End of Trt.1	72	0.1	4.82	-0.3	-5.9	37.0
1 End of treatment is the last on-treatment assessment of the specified vital sign (on or before the Week 24 visit).

Warning

Table 14-7-03. Difference in values for End of Treatment for all groups. The End of Treatment flag is used as is from the ADaM which indicates there are likely discrepancies for end of treatment between the original CDISC Pilot analysis data and the PHUSE CDISC Pilot replication data.

Table 14-7-04

Code

##+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven, labelled)
pacman::p_load(gtreg, gtsummary)

# THEME
theme_gtsummary_language('en', big.mark = '')
theme_gtsummary_compact()

# READ
adsl_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adsl.xpt') %>% 
   mutate(ARM = factor(ARM,
                       levels = c('Placebo', 'Xanomeline Low Dose', 'Xanomeline High Dose'),
                       labels = c('Placebo<br>', 'Xanomeline<br>Low Dose', 'Xanomeline<br>High Dose')))

cm_orig <- haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/tabulations/sdtm/cm.xpt')

cm <- left_join(
   cm_orig,
   adsl_orig %>% 
      select(USUBJID, ARM) ) %>% 
   mutate(CMDECOD_ALL = 'Patients receiving at least one concomitant medication') 

t1 <- tbl_ae(
   data = cm,
   id = USUBJID,
   ae = CMDECOD_ALL,
   strata = ARM,
   id_df = adsl_orig,
   statistic = '{n} ({p}%)',
   digits = c(0, 0),
   zero_symbol = '0') %>% 
   modify_header(
      label ~ '**Therapeutic class, n (%)**'
   )

t2 <- tbl_ae(
   data = cm,
   id = USUBJID,
   ae = CMDECOD,
   soc = CMCLAS,
   strata = ARM,
   id_df = adsl_orig,
   statistic = '{n} ({p}%)',
   digits = c(0, 0),
   zero_symbol = '0') 

tbl_stack(tbls = list(t1, t2), quiet = TRUE) %>% 
   modify_header(
      label ~ '**Therapeutic class, n (%)**') %>% 
   modify_caption(
      '**Table 14-7.04<br>Summary of Concomitant Medications (Number of Subjects)**')

Table 14-7.04
Summary of Concomitant Medications (Number of Subjects)

Therapeutic class, n (%)	Placebo , N = 86	Xanomeline Low Dose, N = 84	Xanomeline High Dose, N = 84
	Overall	Overall	Overall
Patients receiving at least one concomitant medication	77 (90%)	74 (88%)	78 (93%)
ALIMENTARY TRACT AND METABOLISM	12 (14%)	11 (13%)	9 (11%)
ALGELDRATE	2 (2%)	0	2 (2%)
CALCIUM	7 (8%)	6 (7%)	3 (4%)
CALCIUM CARBONATE	0	0	1 (1%)
CIMETIDINE	0	1 (1%)	0
LOPERAMIDE HYDROCHLORIDE	1 (1%)	1 (1%)	1 (1%)
METFORMIN HYDROCHLORIDE	1 (1%)	1 (1%)	0
NIZATIDINE	1 (1%)	1 (1%)	4 (5%)
SIMETICONE	0	2 (2%)	0
ANTINEOPLASTIC AND IMMUNOMODULATING AGENTS	1 (1%)	0	1 (1%)
LEUPRORELIN ACETATE	1 (1%)	0	1 (1%)
BLOOD AND BLOOD FORMING ORGANS	0	1 (1%)	0
FERROUS SULFATE	0	1 (1%)	0
CARDIOVASCULAR SYSTEM	12 (14%)	12 (14%)	7 (8%)
AMLODIPINE	8 (9%)	1 (1%)	2 (2%)
DIGOXIN	0	3 (4%)	2 (2%)
DILTIAZEM HYDROCHLORIDE	0	0	1 (1%)
DOXAZOSIN MESILATE	1 (1%)	2 (2%)	1 (1%)
FELODIPINE	0	1 (1%)	0
FLUVASTATIN	0	2 (2%)	0
FUROSEMIDE	2 (2%)	2 (2%)	1 (1%)
LOSARTAN POTASSIUM	0	2 (2%)	0
NIFEDIPINE	2 (2%)	0	0
DERMATOLOGICALS	0	0	1 (1%)
CLOBETASOL PROPIONATE	0	0	1 (1%)
GENITO URINARY SYSTEM AND SEX HORMONES	6 (7%)	10 (12%)	5 (6%)
ESTROGENS CONJUGATED	6 (7%)	10 (12%)	5 (6%)
NERVOUS SYSTEM	23 (27%)	14 (17%)	8 (10%)
ACETYLSALICYLIC ACID	21 (24%)	11 (13%)	6 (7%)
ALPRAZOLAM	1 (1%)	0	0
DONEPEZIL HYDROCHLORIDE	1 (1%)	2 (2%)	2 (2%)
HALOPERIDOL	0	1 (1%)	0
PAROXETINE HYDROCHLORIDE	0	1 (1%)	0
SUMATRIPTAN	1 (1%)	0	0
RESPIRATORY SYSTEM	4 (5%)	1 (1%)	4 (5%)
BUDESONIDE	0	0	1 (1%)
GUAIFENESIN	1 (1%)	0	0
IPRATROPIUM BROMIDE	1 (1%)	0	0
NAPROXEN SODIUM	1 (1%)	0	3 (4%)
SALBUTAMOL SULFATE	2 (2%)	1 (1%)	0
SYSTEMIC HORMONAL PREPARATIONS, EXCL.	2 (2%)	13 (15%)	8 (10%)
HYDROCORTISONE	2 (2%)	13 (15%)	8 (10%)
UNCODED	74 (86%)	70 (83%)	77 (92%)
UNCODED	74 (86%)	70 (83%)	77 (92%)

Figures

Figure 14-1

Code

#+ message = FALSE
# PACKAGES
pacman::p_load(tidyverse, haven)
pacman::p_load(gtsummary)
pacman::p_load(ggsurvfit)

# THEME
theme_gtsummary_compact()

# READ ADSL
adtte_orig <-  haven::read_xpt(
   'https://raw.githubusercontent.com/cdisc-org/sdtm-adam-pilot-project/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/analysis/adam/datasets/adtte.xpt')

# UPDATE
adtte <- adtte_orig %>% 
   filter(SAFFL == 'Y') %>% 
   filter(PARAMCD == 'TTDE') %>% 
   mutate(TRTA = factor(TRTA, 
                        levels = c('Placebo','Xanomeline Low Dose','Xanomeline High Dose'),
                        labels = c('Placebo', 'Xanomeline Low Dose','Xanomeline High Dose')))

# FIGURE
survfit2(Surv(AVAL, 1-CNSR) ~ TRTA, data = adtte) |>
   ggsurvfit(linewidth = 1) +
   add_quantile(y_value = 0.5, linetype = 'dotted', color = 'grey30', linewidth = 0.8) +
   add_censor_mark(size = 2,
                   show.legend = FALSE) +
   add_confidence_interval(show.legend = FALSE) +
   add_pvalue(caption = 'Logrank {p.value}',
              location = 'annotation',
              x = 15, y = 0.05,
              size = 3) +
   add_risktable(
      risktable_stats = c('n.risk')
   ) +
   add_risktable_strata_symbol(symbol = '\U25AC', size = 16) +
   scale_ggsurvfit() +
   labs(title = 'KM plot for Time to First Dermatologic Event',
        subtitle = 'Safety population',
        y = 'Survival Probability (%)',
        x = 'Time (Days)') +
   guides(colour = ggh4x::guide_stringlegend(ncol = 1)) +
   theme(legend.position = c(0.99, 0.99),
         legend.justification = c('right','top'),
         legend.background = element_rect(fill = 'transparent'))

Code

# TABLE
tbl_merge(
   tbls = list(
      tbl_summary(
         data = adtte,
         include = TRTA,
         statistic = list(all_categorical() ~ '{n}')
      ) %>% 
         modify_header(all_stat_cols() ~ '**No Subjects**<br>{n}'),
      tbl_summary(
         data = adtte %>% mutate(CNSR = factor(CNSR, label = c('Event','Censored'))),
         by = CNSR,
         include = TRTA,
         statistic = list(all_categorical() ~ '{p}% ({n})'),
         percent = 'row'
      ) %>% 
         modify_header(all_stat_cols() ~ '**{level}**<br>{n} ({style_percent(p)}%)'),
      tbl_survfit(
         survfit2(Surv(AVAL, 1-CNSR) ~ TRTA, data = adtte, conf.type = 'log-log'),
         probs = 0.5,
         label_header = '**Median Survival**',
         estimate_fun  = function(x) style_number(x, digits = 1),
      ) %>% 
         modify_header(all_stat_cols() ~ '**Median Survival (95% CL)**')
   ),
   tab_spanner = FALSE) %>%
   modify_header(
      label ~ '') %>% 
   modify_footnote(
      all_stat_cols() ~ NA,
      stat_1_2 ~ 'Dermatologic events were identified as adverse events associated with skin conditions such as rash, pruritus, dermatitis. A full list of adverse event terms is presented in the final study report') %>% 
   modify_table_body(
      ~.x %>% 
         slice(-1) %>% 
         mutate(row_type = 'label'))

	No Subjects 254	Event 152 (60%)1	Censored 102 (40%)	Median Survival (95% CL)
Placebo	86	34% (29)	66% (57)	— (—; —)
Xanomeline Low Dose	84	74% (62)	26% (22)	33.0 (27.0; 48.0)
Xanomeline High Dose	84	73% (61)	27% (23)	36.0 (23.0; 46.0)
1 Dermatologic events were identified as adverse events associated with skin conditions such as rash, pruritus, dermatitis. A full list of adverse event terms is presented in the final study report

PACKAGES

Package Links

gtsummary

gtreg

bstfun

tidyverse

haven

ggsurvfit

mmrm

broom.helpers

SESSION INFO

Code

sessionInfo()

R version 4.3.3 (2024-02-29 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 17763)

Matrix products: default


locale:
[1] LC_COLLATE=English_United States.1252 
[2] LC_CTYPE=English_United States.1252   
[3] LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    

time zone: America/Los_Angeles
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] ggsurvfit_1.0.0 labelled_2.12.0 mmrm_0.3.11     gtreg_0.3.0    
 [5] tictoc_1.2      emmeans_1.10.0  gt_0.10.1       gtsummary_1.7.2
 [9] haven_2.5.4     lubridate_1.9.3 forcats_1.0.0   stringr_1.5.1  
[13] dplyr_1.1.4     purrr_1.0.2     readr_2.1.5     tidyr_1.3.1    
[17] tibble_3.2.1    ggplot2_3.5.0   tidyverse_2.0.0

loaded via a namespace (and not attached):
 [1] tidyselect_1.2.0     farver_2.1.1         fastmap_1.1.1       
 [4] TH.data_1.1-2        janitor_2.2.0        pacman_0.5.1        
 [7] broom.helpers_1.14.0 digest_0.6.34        estimability_1.5    
[10] timechange_0.3.0     lifecycle_1.0.4      ellipsis_0.3.2      
[13] survival_3.5-8       magrittr_2.0.3       compiler_4.3.3      
[16] rlang_1.1.3          sass_0.4.8           tools_4.3.3         
[19] utf8_1.2.4           yaml_2.3.8           knitr_1.45          
[22] labeling_0.4.3       htmlwidgets_1.6.4    curl_5.2.1          
[25] here_1.0.1           xml2_1.3.6           multcomp_1.4-25     
[28] bstfun_0.5.1.9002    withr_3.0.0          ggh4x_0.2.8         
[31] grid_4.3.3           fansi_1.0.6          xtable_1.8-4        
[34] colorspace_2.1-0     scales_1.3.0         MASS_7.3-60.0.1     
[37] cli_3.6.2            mvtnorm_1.2-4        rmarkdown_2.26      
[40] generics_0.1.3       rstudioapi_0.15.0    tzdb_0.4.0          
[43] commonmark_1.9.1     splines_4.3.3        parallel_4.3.3      
[46] vctrs_0.6.5          Matrix_1.6-5         sandwich_3.1-0      
[49] jsonlite_1.8.8       patchwork_1.2.0      hms_1.1.3           
[52] glue_1.7.0           codetools_0.2-19     stringi_1.8.3       
[55] gtable_0.3.4         munsell_0.5.0        pillar_1.9.0        
[58] htmltools_0.5.7      R6_2.5.1             TMB_1.9.10          
[61] Rdpack_2.6           rprojroot_2.0.4      evaluate_0.23       
[64] lattice_0.22-5       markdown_1.12        rbibutils_2.2.16    
[67] backports_1.4.1      broom_1.0.5          snakecase_0.11.1    
[70] Rcpp_1.0.12          nlme_3.1-164         checkmate_2.3.1     
[73] xfun_0.42            zoo_1.8-12           pkgconfig_2.0.3