oa Applying a Novel Smart Insole System that Reduces Re-Ulceration Risk among Diabetics with Peripheral Neuropathy: Do Users Adhere and Comply?

Background & Aim: People with diabetes carry a 25% lifetime risk of foot ulceration. It is well-established that high plantar pressures increase the risk of developing foot ulcers, and that managing peak pressure is an important strategy in reducing such risk. This study tested a novel smart insole system designed to reduce ulceration risk by alerting patients using a smartwatch when their plantar pressure was too high. The device was tested for degree of adherence, compliance, and successful offloading responses among users. Outcomes of users triggering many alerts were compared to those triggering few alerts to see if alert frequency affected adherence and compliance to a novel mobile health device. Method: Participants with diabetes, peripheral neuropathy, and a history of foot ulcers were instructed to wear a smart insole system. Pressure sensors inside the insole were placed in strategic areas where foot ulceration risk has been shown to be high. The sensors were wirelessly connected to a smartwatch through a transmitter. The smartwatch alerted participants when plantar pressure exceeded 50 mmHg over 95% of a moving 15 minute window. Adherence, defined as the number of hours the device was worn, was determined with sensor data and via questionnaires. A successful response to an alert was recorded when patient-initiated offloading occurred within 20 minutes. The length of time an alert lasted (measured as the median time between alert onset and successful offloading) served as a measure of compliance. Results: Participants who increased adherence over time tended to have more alerts (0.82 ±  0.31 alerts/hr) than those who did not improve (0.36 ±  0.46 alerts/hr, p =  0.09). Users receiving a high number of alerts (HA) began with similar levels of successful response to those receiving a low number of alerts (LA), but the HA group successfully offloaded significantly more often than the LA group by the last segment of the study (55.0 ±  6.6% vs. 16.6 ±  11.9%, p <  0.01). Median alert durations increased for LA relative to HA (p =  0.10). Participants tended to overestimate their adherence compared to objective sensor measurements (7.60 ±  2.50 hours/day vs. 5.38 ±  3.43 hours/day, p =  0.10). Conclusion: The results of this study suggests that there appears be to a minimum number of alerts that a user must experience (1 alert every 2 hours of wear time) to maintain adherence and successful response to alerts over time. Above this level, median alert durations decrease, user adherence improves, and successful response rates increase. This suggests that within the range of alerts typically received by someone wearing a smart insole system, relatively more alerts may be preferred, and increasing the number of alerts a user receives by lowering the pressure threshold may be a viable path to maintaining adherence. In addition, self-reported adherence measures may exaggerate usage of novel mobile health devices.