C028: The application of telemedicine to the long term care of a hypertensive population

We sought to determine the efficacy of telemedicine in the care of hypertensive patients. 62 patients attending a hypertension clinic participated in a year long study of telemedicine. The study involved seeing each patient both by telemedicine and in-person on the same day with the order of the visits randomly determined. A telemedicine system utilizing phone lines was employed. The mean study age of participants were 67.1 ± 11.4 years and the study population consisted of 56.6% males. For each type of visit, be it telemedicine (TM) or in-person (IP) clinical data on diagnoses as well as ordering patterns was collected. Intercurrent illnesses were treated. Receptor operator curves (ROC) were used to assess the efficacy of TM as compared to IP in the diagnosis of hypertension. ROC curves for diagnosis of elevated mean pressure provided an area under the curve (auc) of 0.87 (95% CI, .80–0.95). ROC curves for the diagnosis of markedly elevated systolic hypertension provided an area of 0.86 (95% CI 0.78–0.93). This indicates that the telemedicine determined BP differed slightly from IP assessments. There was no difference in diagnostic test ordering or treatment ordering detectable between the two visit types. Patient satisfaction with telemedicine was in general high, although slightly less than that for IP visits. Physician satisfaction with telemedicine in terms of “overall work load”, “mental effort”, “technical skill”, “risk-psychological stress” all ranked statistically in favor of IP. Telemedicine system failures occurred in approximately 16% of visits and diagnostic inadequacies were detected including those associated with inability to adequately auscultate (7/19), poor visualization of certain skin lesions (7/19) and inability to palpate (13/19). In sum, over a year long period telemedicine proved to be an effective means for caring for hypertensive patients. No clinically significant differences in diagnosis or treatment were detected between TM or IP visits, although TM visit failures constituted an incremental cost. Supported by DOE contract AN6271.