Table 3. The impacts of different mPHR functions on the probability of users abandoning the app.
Likelihood of abandonment | ||
---|---|---|
Coefficient | P-value | |
Average usage of self-monitoring function of patient i per week (AVG_SM) | -0.18 | 0.003 |
Average usage of chart function of patient i per week (AVG_CHART) | 0.02 | 0.374 |
Average usage of medication function of patient i per week (AVG_MED) | 0.10 | 0.044 |
Average usage of outpatient support service of patient i per week (AVG_OSS) | 0.13 | 0.000 |
Standard deviation of usage of self-monitoring function of patient i per week (STD_SM) | 0.10 | 0.001 |
Standard deviation of usage of chart function of patient i per week (STD_CHART) | -0.04 | 0.052 |
Standard deviation of usage of medication function of patient i per week (STD_MED) | -0.09 | 0.065 |
Standard deviation of usage of outpatient support service of patient i per week (STD_OSS) | -0.05 | 0.008 |
Note: The coefficients of the control variables were not reported for brevity. The control variables included patient’s age, gender, number of outpatient visits, number of hospital admissions, number of emergency room visits, disease type, and individual heterogeneity in app usage such as the average and standard deviation of the number of days used per week (See S4 Table for details).
The unused period was one month for this result. See S3 Table for the results of all three periods (one, three, five-month)