(I-333) User Adherence to and Design Preferences for a Wearable Device for Cardiac Arrest Detection: A Population Survey

Introduction:: Cardiac arrest is a major population health issue affecting approximately 350,000 individuals annually in the US and Canada. Immediate recognition of arrest is crucial to survival, with survival dropping to 0% for an arrest lasting over 20 minutes. In the out-of-hospital setting, 75% of arrests occur in isolation. This results in approximately 50% of unwitnessed out-of-hospital cardiac arrests (OHCAs) not receiving treatment and being assigned a determination of prolonged death upon arrival of emergency medical services (EMS).

A potential solution to OHCA detection incorporates wearable sensor technologies, capable of measuring OHCA-associated physiological parameters (e.g., heart rate, heart rhythm). Devices that utilize such sensors for general health monitoring have shown commercial success, with estimates suggesting that there are over 440 million connected wearable devices in North America. Integration of such devices into the EMS system allows them to act as artificial bystanders, alerting EMS upon “witnessing” an OHCA.

However, current consumer wearables come in different form factors, with chest patches, wristwatches, and rings being the most popular. From a usability perspective, form factor impacts a user’s willingness to use a device. Other factors, such as awareness of or higher risk for OHCA, also impact user adherence to such a device. As such, the goal of this study is to understand user design preferences and factors impacting adherence to continuous use. It is hypothesized that users will prefer forms that are small and non-intrusive, and that user adherence will increase with increasing perceived risk of OHCA.


Materials and Methods:: In order to assess user design preferences for and willingness to adhere to a wearable device for OHCA detection, an online survey was developed on the Qualtrics survey platform (REB #H21-03589). This survey was advertised via physical and digital posters, as well as an online health research outreach platform, ReachBC. The survey content is divided into three blocks: background, scenario-specific questions, and demographics.

The background block asks participants if they are current/past users or non-users of wearable devices in general, have personal/family cardiac history or personally know someone who’s experienced a cardiac arrest (high risk for or awareness of OHCA), as well as their design preferences. The scenario-specific block guides the participant through hypothetical levels of perceived risk for OHCA and assesses their likelihood to use a wearable device for OHCA detection. Finally, the demographics collects data on stratification variables such as age and gender.

Data analysis was conducted using R (Foundation for Statistical Computing, Vienna, version 3.2.4). Categorical variables were expressed as percentages, and differences between groups of categorical variables were assessed via a chi-squared test. Associations between demographic groups were assessed via an unadjusted odds-ratio. Linear models (LM) were created to analyze the relationship between age and willingness to adhere to continuous wearable use, at different levels of perceived risk. Kurskal-Wallis and Wilcoxon rank-sum tests were performed to assess associations between categorical demographic data and willingness to adhere to continuous wearable use.


Results, Conclusions, and Discussions:: The survey recruited 359 participants (70% women, 26% men, 3% non-binary and genderfluid) with a mean age of 47.4 +/- 18 years. Approximately 82% of participants indicated that they have used wearable devices, and 94% were identified as having higher risk for or awareness of OHCA.

Across participants, wristbands (87%), watches (86%), and rings (62%) were the most popular form factors. Other factors also appeared to influence willingness to use a wearable device for OHCA detection. the LM of willingness to adhere to wearable use as a function of age revealed a significant positive relationship (intercept: 68%, estimate: 0.211%/year, p-value: 0.00709). The Kruskal-Wallis test of gender groups revealed that the non-binary & genderfluid individuals group was significantly different to men and women (p-value: 0.029). Additionally, being a current or past user of wearable technologies impacted willingness to adhere (p-value: 0.027), while having high risk for or awareness of OHCA did not appear to be a significant determinant of adherence. At increasing levels of perceived risk, user demographics tended to have less of an impact on the willingness to adhere.

Form factor preferences fit the hypothesis, with wristbands, watches, and rings being relatively small and non-intrusive. Size (72%) and weight (70%) were amongst the factors participants deemed most relevant to their choice of wearable, further confirming the hypothesis. These preferences are reflected within the commercial sector, with the most popular wearables generally being wrist-worn (e.g., Apple Watch, FitBit). However, these preferences could also be biased due to the commercial saturation of such devices. Generally speaking, a perception of increased risk for OHCA reflected a higher willingness to adhere to continuous use of a wearable, highlighting the importance of education in both risk of OHCA and the role wearables could play in detection. This suggests that an appropriate starting user-base would be individuals inherently at higher risk. This could also suggest that higher awareness of the implications of OHCA could positively impact the uptake of OHCA-centric wearables.


Acknowledgements (Optional): :

References (Optional): :