To the Editor: Thanks to Dr. Strouse for his thoughtful review of our work. We are in agreement with almost all his themes. It seems important to reiterate that ours was primarily a cross-sectional design, and that we make no claims of a causal relation between sample medication provision and higher blood pressure. This explanation would, of course, represent only one of many plausible reasons, including chance variation. As Dr. Strouse suggests, another possibility is that our study is simply a snapshot of physicians responding appropriately to poorly controlled blood pressure by using every means possible to provide medication for uninsured patients. In fact, much current evidence tells us that the simple condition of lacking health insurance is itself associated with less favorable chronic disease status.1–3
While our study admittedly only begins to explore the relation between insurance status, sample medication use, and hypertension, there is some existing indirect evidence that different physician prescribing habits and access to medication might represent one pathway that links lack of insurance and poor health.2 ,4 ,5 Resolution of these questions awaits a prospective trial, as Dr. Strouse correctly suggests. Meanwhile, the contribution of our study is to remind us all that we cannot simply assume that the availability and the use of free sample medicines improve blood pressure in the uninsured. It also seems salient to note that the modest literature currently available on this subject suggests the availability of free samples is associated with less frequent use of first-line agents for hypertension6 and higher prescribing costs.7
Finally, Dr. Strouse offers his own suggestions for effective use of sample medication, including a description of his own practice of “refilling” sample medicines by telephone. This idea seems to belie the inherent temporary and unpredictable availability of costly new medicines left on a physician’s shelf by pharmaceutical representatives, whose principle goal is sales. Moreover, we do not have any data to suggest that his customs are widespread. The only article we found on this topic8 suggests that attitudes and practices toward pharmaceutical samples vary widely across practices and that the lack of a coherent policy or approach is the norm. We believe that Dr. Strouse is largely correct in assuming our patients had to make an additional visit to collect additional sample medication supplies, but doing so was not a condition of participation. The only additional insight our data provided into this issue was the lack of a significant association between self-reported compliance and the sample medication status.