Skip to content

The tables and their numbers

In this chapter

  • Read the primary table of the pilot and its p-value.
  • Read the disposition table, and connect it to the missing-data question of chapter 7.
  • Find a number in the ARD: the analysis, the operation, the raw value and the formatted value.
  • Apply the co-primary rule to the two primary results.
Time
25 minutes
At the end you have
the result of the primary analysis (p = 0.245), the disposition of the 254 patients, and the ARD row behind a number.

The programs of chapter 12 wrote one listing per table. Here is Table 14.3.1, the shell of chapter 14 with its numbers:

Table 14.3.1 ADAS-Cog (11): Change from Baseline to Week 24 - LOCFPopulation: Efficacy population
StatisticPlacebo (N=79)Xanomeline Low Dose (N=81)Xanomeline High Dose (N=74)p-value
Baseline
n798174
Mean (SD)24.1 (12.19)24.4 (12.92)21.3 (11.74)
Median21.021.018.0
Min, Max5, 615, 573, 57
Week 24
n798174
Mean (SD)26.7 (13.79)26.4 (13.18)22.8 (12.48)
Median24.025.020.0
Min, Max5, 626, 623, 62
Change from baseline to Week 24
n7981740.245
Mean (SD)2.5 (5.80)2.0 (5.55)1.5 (4.26)
Median2.02.01.0
Min, Max-11, 16-11, 17-7, 13

Program: t14_3_1.R · Output: T14_3_1 · Results: output/ard/t14_3_1.csvSource: out/tlf/listings/t14_3_1.lst

What to look at:

  • The N of each column: 79, 81 and 74 patients of the efficacy population.
  • The change from baseline to week 24: the mean ADAS-Cog (11) worsened in every group: by 2.5 points with placebo, 2.0 with the low dose, 1.5 with the high dose. A higher score is worse.
  • The p-value, 0.245: the test of a linear dose response in the ANCOVA (chapter 7). At the level of the plan, it is not significant. The difference in the means is in the expected direction, but the trial does not show that it is more than chance.

This is the value of the study’s own report (Table 14-3.01 of the CSR).

Table 14.1.2 Subject DispositionPopulation: Intent-to-treat population
StatisticPlacebo (N=86)Xanomeline Low Dose (N=84)Xanomeline High Dose (N=84)Total (N=254)
End of study status
COMPLETED58 (67.4%)25 (29.8%)27 (32.1%)110 (43.3%)
DISCONTINUED28 (32.6%)59 (70.2%)57 (67.9%)144 (56.7%)
Reason for discontinuation
ADVERSE EVENT8 ( 9.3%)44 (52.4%)40 (47.6%)92 (36.2%)
COMPLETED58 (67.4%)25 (29.8%)27 (32.1%)110 (43.3%)
DEATH2 ( 2.3%)1 ( 1.2%)0 ( 0.0%)3 ( 1.2%)
LACK OF EFFICACY3 ( 3.5%)0 ( 0.0%)1 ( 1.2%)4 ( 1.6%)
LOST TO FOLLOW-UP1 ( 1.2%)1 ( 1.2%)0 ( 0.0%)2 ( 0.8%)
PHYSICIAN DECISION1 ( 1.2%)0 ( 0.0%)2 ( 2.4%)3 ( 1.2%)
PROTOCOL VIOLATION2 ( 2.3%)1 ( 1.2%)3 ( 3.6%)6 ( 2.4%)
STUDY TERMINATED BY SPONSOR2 ( 2.3%)2 ( 2.4%)3 ( 3.6%)7 ( 2.8%)
WITHDRAWAL BY SUBJECT9 (10.5%)10 (11.9%)8 ( 9.5%)27 (10.6%)

Program: t14_1_2.R · Output: T14_1_2 · Results: output/ard/t14_1_2.csvSource: out/tlf/listings/t14_1_2.lst

Of the 254 patients, 110 completed the study. The two xanomeline groups lost about 70% of their patients, most of them for an adverse event; placebo lost 32.6%. Go back to chapter 7: this is why the missing-data handling matters here. With LOCF, a patient who left at week 8 keeps the week 8 score as the week 24 value, and most of those patients are in the active groups.

Every number of every table is also a row of the ARD, the analysis results data (CDISC ARS, the Analysis Results Standard). Open out/ard/t14_3_1.csv: here is its header and the row of the p-value.

out/ard/t14_3_1.csv
"analysis_id","op_id","g1_grouping","g1_id","g1_value","g2_grouping","g2_id","g2_value","g3_grouping","g3_id","g3_value","g4_grouping","g4_id","g4_value","g5_grouping","g5_id","g5_value","raw","formatted","sec","seclab","rowlab","kind","colname","lab2","lab3"
"T14_3_1_CHG_CMP","T14_3_1_CHG_DR_1_pval","PLANNED","","","","","","","","","","","","","","",0.244705673868529,"0.245",3,"Change from baseline to Week 24","Comparison by Treatment","P",,"",""

The row holds the analysis (T14_3_1_CHG_CMP), the operation (T14_3_1_CHG_DR_1_pval: the dose response p-value), the grouping, the raw value 0.244705673868529, the formatted value 0.245, and the labels. The dashboard, the QC and the clinical study report read the numbers from the ARD, never from the text of a table. The pilot has 18,392 such rows from 53 analyses (chapter 12).

Your turn 15.1 The second co-primary endpoint

Find the p-value of the dose response of the CIBIC+ (Table 14.3.2) in the ARD. Under the co-primary rule of chapter 7, did the trial show a dose response?

Show the answer
out/ard/t14_3_2.csv
"analysis_id","op_id","g1_grouping","g1_id","g1_value","g2_grouping","g2_id","g2_value","g3_grouping","g3_id","g3_value","g4_grouping","g4_id","g4_value","g5_grouping","g5_id","g5_value","raw","formatted","sec","seclab","rowlab","kind","colname","lab2","lab3"
"T14_3_2_V_CMP","T14_3_2_V_DR_1_pval","PLANNED","","","","","","","","","","","","","","",0.959670868583973,"0.960",1,"Week 24","Comparison by Treatment","P",,"",""

The p-value of the CIBIC+ is 0.960. Under the co-primary rule, the trial wins only if both tests win. Neither of the two is significant (0.245 and 0.960), so the trial did not show a dose response on its primary endpoints.

What you learned

  • The primary dose response of the ADAS-Cog (11) has p = 0.245: not significant, the value of the study’s report.
  • About 70% of each xanomeline group discontinued, against 32.6% of placebo: the reason why the missing-data handling matters.
  • Every number is a row of the ARD with its analysis, its raw value and its formatted value.
  • Co-primary endpoints must both win; here neither does.

Next: 16 · Readable R programs →

Found an error or have a question? Write to datastep.studio@gmail.com· The plan is written in Sapling; every R program runs in DataStep Studio for R.

Data: the CDISC pilot study (cdiscpilot01). The protocol as CDISC USDM from the CDISC 360i project (MIT licence); the SDTM data from the PhUSE scripts repository (MIT licence). Public example data: no real patient.

R is a project of the R Foundation. SAS is a registered trademark of SAS Institute Inc. CDISC, SDTM, ADaM and USDM are trademarks of the Clinical Data Interchange Standards Consortium. Claude is a trademark of Anthropic. This book is not affiliated with or endorsed by any of them.

© 2026 DataStep Labs. All rights reserved.