Background: We have shown that a fourth-generation telehealth program that analyzes and responds synchronously to data transferred from patients is associated with fewer hospitalizations and lower medical costs. Whether a fourth-generation telehealth program can reduce all-cause mortality has not yet been reported for patients with chronic cardiovascular disease.
Objective: We conducted a clinical epidemiology study retrospectively to determine whether a fourth-generation telehealth program can reduce all-cause mortality for patients with chronic cardiovascular disease.
Methods: We enrolled 576 patients who had joined a telehealth program and compared them with 1178 control patients. A Cox proportional hazards model was fitted to analyze the impact of risk predictors on all-cause mortality. The model adjusted for age, sex, and chronic comorbidities.
Results: There were 53 (9.3%) deaths in the telehealth group and 136 (11.54%) deaths in the control group. We found that the telehealth program violated the proportional hazards assumption by the Schoenfeld residual test. Thus, we fitted a Cox regression model with time-varying covariates. The results showed an estimated hazard ratio (HR) of 0.866 (95% CI 0.837-0.896, P<.001; number needed to treat at 1 year=55.6, 95% CI 43.2-75.7 based on HR of telehealth program) for the telehealth program on all-cause mortality after adjusting for age, sex, and comorbidities. The time-varying interaction term in this analysis showed that the beneficial effect of telehealth would increase over time.
Conclusions: The results suggest that our fourth-generation telehealth program is associated with less all-cause mortality compared with usual care after adjusting for chronic comorbidities.
Keywords: all-cause mortality; cardiovascular diseases; outcome assessment (health care); telemedicine.