Blog 16 - Self reported outcomes versus sensor generated outcomes
Blog 16 outlines how wearables can play a huge role in improving the measurements of clinical outcomes
The changing nature of healthcare is one of the primary key drivers for the need to change how success in treatments is measured. With the aging population in the western world and the huge success in treating infectious and other acute diseases, chronic disease management has now come to dominate healthcare spending. A 2020 study* at Stanford University School of Medicine found that over 50% of the US population has a chronic disease, and 86% of health care costs are attributable to chronic disease.
Conventional clinical outcomes for measuring drug effectiveness are largely collected at structured time intervals in unnatural settings such as a 6-month clinic follow-up visit. This will typically consist of a doctor asking a patient how they are feeling or how much pain they are in. This practice typically relies on short term memory and generally does not capture the true reality.
The use of continuous self reporting through questionnaires, diaries/logs, surveys, and interviews tends to have a bit more merit, but still leaves scope for over or underestimating the required measures.
Most current outcome measurements in practice are not condition-specific metrics, but reflect abrupt general outcomes such as 5-year survival rate. As a result, current measures are generally unreliable and uncertain.
Wearables sensors are the game changer to improve these outdated measures. With wearables and the data endpoints that can be generated the definition of success for a treatment can be fundamentally altered. Instead of looking at patient reported outcomes or the results of tests done every few months in a doctor’s office, we can put sensors on the patient to actually document his or her progress over a sustained period.
For example, you can put an activity monitor on a Parkinson’s or COPD patient and determine quantitatively if their activity level is increasing or decreasing over time. You can even get detailed gait metrics and analysis of freezing episodes. Joint stiffness related to arthritis in common areas such as ankles, knees and shoulders can be measured to determine level of improvement or otherwise in response to treatments. Everything can be measured up to 24x7 continuously.
These are just a couple of examples, but many of the most common disease states can have measures of success or otherwise that can be determined using wearables. This takes the trial measures from infrequent and often subjective outcomes to continuous objective outcomes.
By moving to continuous quantitative outcomes you dramatically reduce measurement uncertainty. You start to obtain a true picture of what is going on with the patient. Capturing more accurate, quantitative health measures provides the opportunity to improve the accuracy of early-stage trials, leading to fewer later-stage failures. It can also be used to reduce sample sizes or potentially shorten the trial.
Furthermore all of this can be done remotely and continuously, reducing clinic visits, thus lowering the costs. Geographic and logistical constraints to participation can be removed allowing for more rapid and diverse patient enrolment, with higher retention rates. By monitoring patients continuously, you will be able to identify unexpected deterioration in patients overall health, detecting potential adverse events earlier and overall increasing patient safety.
The move to sensor generated outcomes is still somewhat in its infancy, with most trials being exploratory. Much of the current focus is around developing a consensus on accepted endpoint measures for the vast array of data wearables can produce. As convergence in this area emerges, it is reasonable to expect the use of wearables as a leading tool in measuring clinical outcomes to gain major momentum over the coming decade.
