How Direct-to-Patient Registries inform Clinical Trial Research
By: Nancy Dreyer, MPH, PhD | June 25, 2015
Patient registries in particular are a useful approach to getting systematic information about the patient experience in dealing with diseases and the health care system. Getting perspective on that patient experience, including how they respond to various treatments, is critically important to helping to shape meaningful clinical trials, and to put the results of those clinical trials in context. For example, how many patients in the “real world” are like those tested in the clinical trial, and for those who are different, in what ways are they different and are those differences likely to affect how they might respond to treatment? We need both types of information (clinical trials and real-world evidence) to help us deliver truly personalized medical care.
At Quintiles, we are especially interested in ‘direct-to-patient registries’, where patients record health information outside of the clinical trial experience. One of the benefits of such registries is that patients record data that are relevant to their own experience, including details about the impact of their illness or condition on day-to-day life (sometimes called the “Burden of illness”); what treatments they use, including those that were prescribed as well as any other treatments they may be using, including like non-prescriptions drugs, meditation, etc.; whether they adhere to treatments prescribed by their doctors, and why or why not. The use of direct-to-patient registry data is a new approach, and one that is quite different form the highly controlled world of clinical trials. Some worry about whether patients will be accurate and honest reporters and will they provide clinically interpretable information, since most don’t have medical training. However, we see tremendous value when these tools are used appropriately, and that data derived from patients can provide vital insights that might not otherwise come to attention.
Pregnant women report medicines used
We recently demonstrated the benefits of this approach through a public-private partnership through the Innovative Medicines Initiative in Europe , in a project known as PROTECT (the Pharmacoepidemiological Research on Outcomes of Therapeutics by a European ConsorTium), which was coordinated by the European Medicines Agency (EMA).
The focus of our work was to learn whether data that was reported directly by patients, without the intervention of a health care provider, could be a useful approach for understanding the safety of medications used in pregnancy. This is a particularly important area for real-world research because not many clinical trials are conducted on pregnant women, unless the product is being developed specifically for use in pregnancy. Nonetheless, women who use medications for chronic disease or who want to use a medication for an acute illness all want to know about whether these treatments could harm the fetus, and we have little information which can be used to help clinicians respond. After all, would you want to base your decision on whether to use a drug during pregnancy on the experience of a mouse or a hamster? Or would you be satisfied to base your decision on whether to use a medication in pregnancy using evidence from voluntary reports from health care providers who called in to report events that they suspected might be caused by drug use? Good clinical intuition is tremendously valuable, but I wouldn’t want to make my choices based on reviewing lots of anecdotal evidence. After all, clinicians would never call in to report that they had a good experience and that all of their pregnant patients who stayed on their antidepressants, for example, had normal birth outcomes. So we need systematic data collection and evaluation in order to evaluate the safety of medications used in pregnancy.
Currently one of the better alternatives to spontaneous reports that we currently use to understand the safety of medications used during pregnancy is to look at electronic medical record data to characterize treatments and outcomes. However, these types of data generally do not contain information about lifestyle and other factors, which patients may or may not share with their doctor and which may affect the outcome of pregnancy, whether pregnancy women actually took the drugs prescribed for them, whether they used drugs that were prescribed for others (borrowed medications), etc.. These issues demonstrate the need to develop alternative methods of data collection for pharmacovigilance.
Our part of this PROTECT program was a direct-to-patient non-interventional study in which we invited pregnant women in Denmark, the UK, the Netherlands and Poland to share information in a registry about their medical history, medications taken during their pregnancy, lifestyle factors that may have an impact on birth outcome (like use of alcohol, recreational drugs, etc.), and family history. We also collected information about the outcomes of their pregnancies. The goal of the study wasn’t to answer every possible question about the impact of medication and behavior on pregnancy. Rather, it was to demonstrate how such data can be collected and evaluated and then to figure out if it was reasonable accurate and complete.
Data collection closed in the first quarter of 2015, and analyses of the results are currently in progress, including comparisons of data reported by pregnant women with that available from electronic medical records (in one country, with patient permission) as well with as the Danish National Prescription Registry, which tracks all medicines prescribed and filled (also, with full patient knowledge).
Such studies are still novel, yet we believe they will be used more often as a way to supplement clinical trial information and provide a full picture of the risks and benefits of medications used in pregnancy as well as other applications. These direct-to-patient registries will allow researchers to address myriad research questions relating to everyday clinical practice, including treatment heterogeneity in population subgroups not traditionally included in clinical trials, like pregnant women as well as children, the elderly and people with multiple and complex diseases. Used in conjunction with clinical trial research, they could be tremendously valuable to inform drug safety by obtaining information that is hard to come by otherwise. Internet-based studies such as this may also be useful for studying illicit drug use and other behaviors that may affect the fetus, since there is some evidence suggesting that patients will tell computers things that they might not tell health care professionals. But these tools must be used wisely, and with strict attention to maintaining patient privacy while also using the data to promote public health.
One of the big obstacles going forward will be to address the variety of administrative barriers relating to data protection and informed consent, independently and also within the structure of a public-private partnership. I hope we will not let our fears about privacy outweigh our need to for information about what factors really can harm the fetus. There are strong research practices that can promote both.