What if decentralized trials also meant better measurement of what matters?
There has been extraordinary growth in the acceptance of at-home, or “decentralized”, clinical trials as a method for capturing high-quality research data while enabling individuals to participate from the comfort of their homes in the past couple of years. It is a big step toward making clinical trials work for patients, rather than the other way around.
We’re just at the beginning, though. There is a world of possibilities out there - the next five years will bring about remarkable strides toward truly patient-driven trial design.
Our two cents: the next step is to update the way that we measure the effectiveness of care, to align with the needs of individuals in their daily lives. Too often, trial endpoints are divorced from the outcomes that matter most to people living with the condition.
How do we use new methods like decentralized trials to measure whether care, or a new therapy, is working for the individual? How do we determine whether a person is feeling better or worse over time? Most people would agree that with positive health outcomes, you know it when you see it - you took Tylenol and your headache went away. But how do we measure “headache went away” in a group of many individuals who are all experiencing different types of headaches, and who all have different thresholds for deciding to start taking Tylenol? How do we measure changes in the headaches that one person experiences over time, or whether Advil works better for them?
There are a few potential answers:
you could develop a device to attempt to objectively measure the strength of a headache
you could ask a clinician to observe the patient and provide an assessment
or, you could ask the individual themselves to describe the headache and how it changes over time and after therapy
The concept behind “patient-reported outcomes,” or PROs, is to take the third approach - measure a health outcome based on how an individual reports it. Arguably, this is the truest form of an outcome. After all, isn’t the aim of medicine to help people feel better, rather than to change a reading on a device? Traditionally, though, our healthcare system has focused more on clinician assessment and device measurement to establish an understanding of outcomes than on patient reporting, particularly in research settings. This has started to change.
The first publication to describe the PRO ran in 1976. Over the next 20 years, the concept of the PRO remained relatively quiet, but starting in the mid-90s, organizations like PHT Corporation began the development of PROs in earnest, and the first studies utilizing PROs were conducted. This work captured the attention of the FDA in the early 2000s, and after several years of development, the first official guidance on the use of PROs was published in 2009.
Last year, we saw more than 4,000 papers mentioning PROs. As this space has evolved, most work on PROs has come to focus on the development of surveys - known either as PROs or PROMs (patient-reported outcome measures). For the sake of clarity, we’ll call them “PROMs” here. PROMs are designed to be used in a defined population, to capture a specific set of information geared toward answering a question like, ‘what are the characteristics and severity levels of these patients’ headaches?’. In that way, they’re similar to device measurements, which provide a lens into a specific component of a patient’s overall outcomes.
Generally, PROMs take the form of multi-question surveys developed for use in research, optimized to fit into the construct of a randomized clinical trial or other traditional study design. For this reason, many PROMs go through psychometric validation to ensure that there’s minimal variation in the way that different patients answer the surveys. They also have very specific rules for use - they’re meant to be completed only at defined intervals, like once per month, can be restricted to be used only for certain diagnoses, and are often not able to be broken down into smaller partial surveys.
This rigidity, which has developed over time as a result of attempting to convince the research community of the validity of PROMs as a measurement tool, has had the unfortunate side-effect of creating cumbersome surveys that are usually time-consuming and frustrating for participants, and artificially constrained in the types of outcome data that they can provide. Their highly specific nature is not designed to enable a complete understanding of the complex individual patient, nor for capturing the types of observations of change in health outcomes that people experience in the course of daily life.
We believe that there’s a better way, and over the past 6 years, we’ve worked with thousands of individuals living with complex conditions, partners in rare disease communities, and research leaders to make our ideas a reality. The concept that we’ve come up with is the HRO, or home-reported outcome.
The basic principle is that the HRO is an outcome that enables greater individual-level flexibility than traditional PRO scales in capturing regularly-reported, longitudinal patient-generated data on all aspects of that individual’s experience of his or her condition. HROs are collected as individual multiple-choice or multi-select questions, delivered electronically to patients and family caregivers on a regular basis. They’re designed to be captured either individually or alongside other HRO questions - to do away with the long-form questionnaire.
Each person answering a set of HROs has the ability to select the symptoms, behaviors, and other types of outcomes that are most relevant to their experience. In this way, the individual has agency in the way that his or her health is measured, and much higher-granularity data is collected by participants over time.
We’ve found that people prefer to collect info this way, leading to very strong frequency of touchpoints with the participant (average 1.2/week per year across 3 years of cohorts), and we see strong coverage of the full patient experience that provides a truly 360° view of outcomes and therapy impact for those individuals.
We’ve also found that HROs can be incredibly helpful to the individual in managing his or her health. After tracking, an individual can then use the record of HRO responses to view their own changes over time and communicate their recent experience to clinicians. In this way, the data captured for research can do double-duty as a tool of substance to improve the standard-of-care.
The possibilities for patient-driven research are beginning to bloom.