This month’s intriguing question…
Should the statistician sign the clinical study report?
During report audits we often see that a clinical study report is finalized and signed by the sponsor and the principal- or coordinating investigators, no-one else signs the report. Is that correct or should a statistician sign the report too?
Let’s first look at some results from our audits: When we review clinical study reports, the major errors are not so much encountered in the tables, figures and listings. Nevertheless, these tables, figures and listings are very often the highest time-consuming factor in the quality control of a clinical study report. Where do we find most of the errors? Well, we find often inconsistencies within the report when the data in the text do not match with the data in the tables. Where else? Another major source of errors is found in some simple statistical calculations within the text; typically, an error is made in the calculation of a percentage. Often this happens because the author is not asking the statistician to calculate simple percentages but uses his/her own hand-held calculator. It is not always clear if the author has hit the wrong button or selected the wrong data to calculate from. Worst are the fundamental statistical issues when the Statistical Analysis Plan is not respected and another approach or population is selected in the clinical study report.
What do the regulations say? In ICH E9 “Statistical Principles for Clinical Trials” it is stated in chapter 7.1. “Finally statistical judgement should be brought to bear on the analysis, interpretation and presentation of the results of a clinical trial. To this end the trial statistician should be a member of the team responsible for the clinical study report, and should approve the clinical report.” This Guideline thus calls for the statistician to “approve” the clinical report. In ICH E3 “Structure and Content of Clinical Study Reports” it is mentioned in chapter 6 that the listing of persons whose participation materially affected the conduct of the study, should include the responsible biostatistician.
Conclusion: In all clinical study reports, the responsible statistician should be named and there should be somewhere a proof that the statistician approved the study report, either as per applicable SOP or by the statistician’s signature on the study report.