Three Reasons Why Patient Preference Studies are NOT Market Research
Patient preference studies in healthcare have been around since the 1990’s, with early studies using various survey and conjoint analysis techniques to determine patients’ preferences for treatment or procedures of a variety of disease.[1] Since then, dozes of quantitative patient preference studies have been published using increasingly sophisticated modeling and other observational research techniques. Yet, the purpose of these studies, how they are done, and what value they bring is still poorly understood by most of the medical products industry. One comment that I often hear when I talk about patient preference is, “Oh! So, it’s like market research.” My answer is always an unequivocal “No, not really.” But, it’s sometimes difficult to describe precisely why. So, here are three reasons that I hope will explain why patient preference studies are not market research. I look forward to seeing your additions in the comments.
1. We aren’t actually asking patients’ preference. Despite the fact that these studies are called “patient preference” studies, we aren’t actually interested in preference, per se. In fact, what we want to know is such more complicated than that. Patient preference studies are designed to elicit complex behavioral characteristics from patients including patients’ tolerance for risk, patients’ priorities for treatment outcomes and/or treatment characteristics, the amount of time a patient is willing to trade-off of their lifespan to receive treatment benefits, or a patients’ willingness to pay.
As you can readily see, these concepts go beyond an individuals’ preference for a male or female doctor, a pill or an injection, or how they receive disease awareness materials, for example. “Patient preference” is an overly simplified way to describe incredibly sophisticated observational research to support medical products development and commercialization.
2. The purpose of patient preference research is different. There are a variety of reasons to conduct a patient preference study, none of which are to determine the market for a new therapy. In recent years, industry might feel compelled to conduct patient preference research to support regulatory submissions given FDA’s priority on patient-centered medical product development.[2] Patient preference studies could be used to argue for a single-arm clinical trial, or approval of a device with a higher risk profile, where patients have a higher tolerance for risk given the benefits of the treatment. There is also hope that patient preference studies can influence reimbursement decisions and health technology assessments globally. Because of the rigorous methods used to assess patient preference, these studies carry significant weight with regulators and decision-makers.
3. The stakes are very high. When patients participate in a patient preference study, they are making hard choices about their own health. Even if the treatment scenarios presented in a discrete choice survey are hypothetical, patients take them seriously. They are making trade-offs between mortality and recovery time, durability and symptom relief, survival and time in the hospital. For patients suffering with any number of chronic, debilitating diseases, these choices are very real for them, and their answers are sincere. Patients are not given enough credit for their sophisticated understanding of their bodies and how their condition impacts their day to day life and the lives of those around them. When presented with treatment choices, they make difficult trade-offs that we can never fully understand.
I hope I’ve convinced you that patient preference research is not market research. I mean no disrespect to my market research colleagues.You all do amazing work with great purpose. My intent here was to draw the distinction between two very different areas of observational study. At the very least, I have my talking points next time someone asks me what patient preference research is and what makes is so different. What would you add to this list?
[1] See Wright et al. (1994) in Radiotherapy and Oncology; Haymen et al. (1997) in Journal of Clinical Oncology; and Hornberger (1999) in Surgery to name just a few.
[2] https://www.fda.gov/about-fda/cdrh-patient-engagement/patient-preference-information-ppi-medical-device-decision-making