Placing the Emphasis on Real-world Data

For the past 25 years, the wound care community has been working to elevate the level of evidence used to provide care. For much of that time, the holy grail has appeared to be the prospective randomized trial. Now we have come to do these, ad infinitum, especially for relatively easy to heal diabetic foot wounds. However, for many years we have known that those who participate in prospective randomized trials are not the typical patient. In general, they adhere to our plans of care, do not miss appointments, have very good hemoglobin A1c levels, and have adequate perfusion; therefore, they are fundamentally the ideal patient. With this intrinsic bias, the standard of care patient in a clinical trial always does better than the real-world “standard of care” patient.

We are now realizing that we need to collect and pay more attention to real- world data, which are defined as data relating to the patient’s health status and the delivery of care. These data are derived from electronic health records, claims and billing activities, registries, and/or mobile devices.¹ Real-world evidence is the clinical evidence that supports potential benefits and/or potential risks of a medical product or an algorithm of care. The FDA recognizes that these may include large simple trials, pragmatic trials, or observational studies.¹

However, determining how to collect the data remains quite difficult. It would be interesting if all prospective randomized trials had a third arm attached, in which patients with the diagnosis in question were simply followed at the same time as those enrolled in the active arms of the trial. The only inclusion criteria would be to have the same initial diagnosis, regardless of the patient’s adherence to care or meeting other physiologic inclusion criteria. This group would certainly be more “real world” than the current standard of care arms.

As a medical community, we also recognize that in most cases, significant amounts of care need to be aimed at complex patients; this care is often not product-specific, but more algorithmic in nature. Therefore, we are tasked with figuring out where to get our real-world data, especially considering claims and billing activities are not likely to render truly accurate data. However, most other methods of collection are not well incentivized, and even the tools to collect the data only exist in some relatively unique settings. Once better data are collected prospectively, will these patients be “real world” when the mere act of collecting the data may affect the care that is delivered?

Currently, wound center management databases (and a few commercial registries) that are well built include many data points that are needed to provide useful aggregate data. By the nature of their design, retrospective data would inform prospective data collection. There are many nuances to the data that need to be collected. An example is when a well-recognized and effective therapy is not used. Was it because it was not available at that location; did the provider fail to choose that therapy; or did the patient refuse that therapeutic plan? Such data would help us understand the roadblocks to wound closure that occur in so many of our patients.

I applaud the groups within wound care that are working to better define collecting the data, turning it into evidence, and maybe most importantly, using that data for decision-making around the care we deliver to our patients. This is the beginning of a move away from the prospective randomized trial and toward assessing how most of our patients actually fare. Hopefully, we will be able to get this right, and move ahead more quickly than we have been.

John C. Lantis II, MD, FACS

Clinical Editor, Wounds

woundseditor@hmpglobal.com