Evaluation of the National Health Services (NHS) Direct Pilot Telehealth programme: cost-effectiveness analysis

Abstract

Objective to evaluate the cost-effectiveness of a pilot telehealth programme applied to a wide population of patients with chronic obstructive pulmonary disease (COPD). Design: Vital signs data was transmitted from the home of the patient on a daily basis using a patient monitoring system for review by community nurse to assist decisions on management. Setting: Community services for patients diagnosed with COPD Participants: Two Primary Care Trusts (PCTs) enrolled 321 patients diagnosed with COPD into the telehealth programme. 227 patients having a complete baseline record of at least 88 days of continuous remote monitoring and meeting all inclusion criteria were included in the statistical analysis. Intervention: Remote monitoring Methods: Resource and cost data associated with patient events (in-patient hospitalisation, accident and emergency (A&E) and home visits) 12 months before, immediately before and during monitoring, equipment, start-up and administration were collected and compared to determine cost-effectiveness of the programme. Main outcome measures: Cost-effectiveness of programme, impact on resource usage, and patterns of change in resource usage. Results: Cost-effectiveness was determined for the two PCTs and the two periods before monitoring to provide four separate estimates. Cost-effectiveness had high variance both between the PCTs and between the comparison periods ranging from a saving of £140800 ($176,000) to an increase of £9600 ($12,000). The average saving was £1023 ($1280) per patient per year. The largest impact was on length of stay with a fall in the average length of in-patient care in PCT1 from 11.5 days in the period 12 months before monitoring to 6.5 days during monitoring, and similarly in PCT2 from 7.5 days to 5.2 days. Conclusion: There was a wide discrepancy in the results from the two PCTs. This places doubt on outcomes and may indicate also why the literature on cost-effectiveness remains inconclusive. The wide variance on savings and the uncertainty of monitoring cost does not allow a definitive conclusion on the cost effectiveness as an outcome of this study. It might well be that the average saving was £1023 ($1280) per patient per year but the variance is too great to allow this to be statistically significant. Each locality based clinical service provides a service to achieve the same clinical goal, but they do so in significantly different ways. The introduction of remote monitoring has a profound effect on team learning and clinical practice and thus distorts the cost effectiveness evaluation of the use of the technology. Cost-effectiveness studies will continue to struggle to provide a definitive answer because outcome measurements are too dependent on factors other than the technology.