Friday, May 01, 2015

Study Day in Clinical Study Protocol

Any clinical trial should always have a study protocol. The study protocol will contain a schedule of events table to specify the study visits and the study procedures. For some protocols, study visits will just be the study days (Day 1, Day 7, Day 21,…).

A while ago, I wrote an article “counting the study day” to describe two ways in counting the study day: starting with zero or starting 1. At that time, I said that both ways were acceptable.

It turns out that starting with study day 0 can cause a lot of trouble for programming, especially in SDTM (study data tabulation model).  According to SDTM Implementation Guidelines, there should not be a study day 0 in SDTM data set. If a study protocol uses a study day 0 as the first day of study drug administration, there will be an inconsistency when comparing the SDTM data sets (therefore, the data listings) with the protocol. Here is what it is said about the study day in SDTM implementation guidelines (SDTMIG).

To avoid the inconsistency issue, it is advisable that in study protocols, study day 0 should not be used. In clinical study protocols, the first day of the study drug administration (or randomization day) should always be counted as day 1 (instead of day 0).  With study day 1 being the first day of the study drug administration, the 7 days after treatment will be study day 8, and so on.

The schedule of events table below is advisable. It correctly uses Day 1 as the first day of dosing.

The schedule of events table below is not advisable since it uses Day 0. To avoid the trouble for downstream SDTM programming, it is better to have study visits listed as Day 1, Day 2, Day 8, Day 15,  Day 29.

Additional References:

1. Studyday calculation ( --DY Variable in SDTM)
2. SAS Programming in the Pharmaceutical Industry