I'm co-investigator on a clinical trial where the we are studying the effect of an intervention. The study is powered to detect a clinically meaningful change for some primary endpoint. My questions are around the planned secondary endpoints.

There are two secondary endpoints: 1) whether or not individuals randomized in the trial were prescribed medications (in this case statins) and 2) among those who were prescribed medications, whether or not individuals adhered to those medications. The first secondary endpoint is the main one and we have decided to use a regression model to estimate the intervention effect. The latter secondary endpoint hinges on the results of the first one. The PI wants to look at whether the intervention had an effect on statin adherence among those who were prescribed. Specifically, he wants to use a regression model to do this.

My concerns are 1) by subsetting to this group, we no longer have the randomization with respect to the intervention, so that the "effect of intervention on adherence" is meaningless, and 2) in line with the first point, the underlying mechanisms may vary between the main secondary outcome and the adherence outcome. In other words, we may have many unmeasured factors. My suggestion was to keep the analysis descriptive: tabulate the binary adherence outcome by the binary intervention, and restricting the analysis to those who were prescribed medications. However, a reviewer (non-statistician) insists that we can still use regression.

Are there things that I can add in order to make a convincing case for keeping it descriptive?

  • $\begingroup$ I find this not so easy to follow. Could you maybe draw a diagram of the situation? $\endgroup$ Commented Nov 11, 2020 at 19:01
  • 1
    $\begingroup$ @SextusEmpiricus - I've simplified the language to clarify the main points that I would like addressed. Please let me know if it's still unclear. $\endgroup$ Commented Nov 11, 2020 at 19:22
  • $\begingroup$ Can you give a sense of what the intervention is? It sounds like these are mediation analyses, which would be perfectly reasonable. The main analysis for the primary outcome would be intervention -> improved biomarker. The 1st secondary analysis would be intervention -> increased Rx rate. The 2nd secondary analysis would be intervention -> adherence rate (among patients w/ Rx). $\endgroup$ Commented Nov 11, 2020 at 19:36
  • 1
    $\begingroup$ It sounds to me like the implied full causal theory is that the intervention increases the rate of prescriptions (to patients who would not otherwise have received them), and adherence does not decrease amongst patients who normally don't get the Rx, so the intervention ultimately leads to improvements in the biomarker. For example, is the intervention something like helping patients w/o insurance afford the Rx? If so, it all seems perfectly sensible to me. $\endgroup$ Commented Nov 11, 2020 at 19:37
  • $\begingroup$ The main target is primary care patients who are at high risk for CVD and eligible for statin therapy based on current guidelines, but have not received one. The intervention is essentially additional pharmacogenetic testing, the results of which the care-provider and patient can discuss together to deliver appropriate statin prescribing. The control is the usual care based on currently established guidelines. $\endgroup$ Commented Nov 11, 2020 at 19:46

1 Answer 1

  1. You are right that the specified analysis is among a non-randomized set. You are wrong however that the intervention is meaningless because we restrict analyses to this non-randomized set. One does, however, need to adjust for possible confounders. In fact, this analysis does not "depend" on the significance of the first secondary hypothesis as one might suppose. Given statins are the most widely prescribed medication (period), I'm assuming the issue doesn't concern whether any people start statins, but it rather has to do with the failure of the intervention to delay the time at which one starts statin medications. The failure to declare statistical significance of the first secondary endpoint simply means you did not have enough to power to detect a change if there was one. So the second secondary hypothesis needs to consider that intervention might affect compliance regardless. Identifying the right confounds to adjust for in an analysis is a serious statistical hurdle that needs careful deliberation in the protocol (via an amendment) or the SAP (if database lock has not been achieved yet).

  2. You're absolutely right that intervention may have opposite effects in the risk of starting a medication versus the risk of adhering to that medication. I can think of numerous examples. Consider fluoridation of water: it offsets time to dental carries, but it does nothing to improve dental hygiene so gum recess, gingivitis, etc. so the major surgical dental adverse events may not be affected at all. In other words: this is why we need a complete set of secondary endpoints, otherwise we are doing salami science.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.