Effect of Prior Health Knowledge on the Usability of Two Home Medical Devices: Usability Study (Preprint)

Abstract

BACKGROUND Studies on the usability of health care devices are quietly becoming more common even though usability standards are not necessarily specified and followed. Yet there is little knowledge about the impact of the context of use on the usability outcome. While it is specified in the usability standard (ISO 9241-11, 2018) of a medical device that it may be affected by its context of use and especially by the characteristics of its users. Among these, prior health knowledge is crucial. However, no study has so far shown that prior health knowledge influences the usability of a medical device. OBJECTIVE Our study was designed to fill this gap, by analyzing the relationship between the usability of home medical devices and prior health knowledge through an experimental approach METHODS In order to assessed the usability of two devices (blood pressure monitor and pulse oximeter) user tests were conducted with 149 students. Mixed-methods (subjective vs. objective) approach using a variety of standard instruments (direct observation, video analysis, questionnaires) were used. They completed a questionnaire to show the extent of their previous health knowledge and then operated both devices randomly. Efficiency and effectiveness measures were collected by video analysis. Satisfaction measures were collected by questionnaire (System Usability Scale). The qualitative observational data was coded using inductive analysis by two independent researchers. Correlational analyses and clusters were performed to test how usability relates to sociodemographic and prior health knowledge. RESULTS The results indicate a lack of usability for both devices. Users made approximately 0.77 errors (SD = 1.49) and the mean of the System Usability Scale (SUS) score was 72.4 (SD = 21.07), which is considered "satisfactory" for the blood pressure monitor (N = 137 participants). The pulse oximeter (N = 147) therefore appears to be easier to use but participants made more errors (M = 0.99, SD = 0.92). The mean SUS score was 71.52 (SD = 17.29), which indicates a "satisfactory" score. Results also showed a low negative and significant correlation only between the effectiveness of the two devices (blood pressure monitor: r = -0.191, P = 0.026; and pulse oximeter: r = -0.263, P = 0.001) and previous knowledge. More subtly, the authors experimentally identified the existence of a threshold level (² = 10.89, P =.004) on health knowledge to correctly use the pulse oximeter, but which is missing for the blood pressure monitor. CONCLUSIONS Thus, this study has two contributions: (1) a theoretical interest highlighting the importance of user characteristics including prior health knowledge of usability outcome and (2) an applied interest to provide recommendations to designers and medical staff.