Effect of implementing digital adherence technologies on the use of health care providers’ time and the human resource cost of tuberculosis treatment adherence support in Ethiopia

Abstract

Background. The management of human resources for health is a key priority for resource constrained health systems. Digital adherence technologies (DATs) for treatment support offer an opportunity to reorganise existing healthcare worker workload. Historically, tuberculosis treatment services involved directly observed treatment with frequent patient visits to the clinic for treatment adherence support. More recently, digital technologies have offered an option for the digital monitoring of treatment adherence and remote support. The aim of this study was to determine how health care workers use their time following the implementation of digital adherence support, and the human resource cost of tuberculosis treatment adherence support. Methods. As part of the ASCENT three-arm cluster randomised trial of DATs for tuberculosis treatment support, we conducted time-and-motion studies in 15 of 78 ASCENT-trial health facilities for six-weeks. Healthcare workers recorded time spent on direct patient care and in-person adherence monitoring for 10 patients, as well as time spent on above-service activities. Total time use was calculated by multiplying time spent per visit by number of patient visits at the facility for six weeks and adding above-service time. In a facility-level analysis, we used negative binomial regression models to investigate associations between total time use and facility-level factors. Findings. HCWs spent 4% (126/2892) of time in minutes on in-person adherence monitoring in the Labels, compared to 7% (206/3126) in the Pillbox and 13% (587/4673) in the standard-of-care study arms. More time was spent on above-service activities in the standard-of-care (2360 minutes [SD:1951]) compared to the DAT arms (Pillbox: 2057min [SD:1570]; Labels: 2261min [SD:1360]). After adjusting for facility workload and number of people lost-to-follow-up, we did not find sufficient evidence of a difference in the human resource costs between the study arms (Labels IRR (95%CI): 0.71 (0.33; 1.52); Pillbox (95% CI) 0.71 (0.33; 1.52)). Conclusion. We found a small, although not statistically significant reduction, in human resource costs in the facilities where DATs were implemented. We observed changes in how healthcare workers spent their time, including a shift towards fewer health facility visits but more time spent on in-person adherence support. Further work is ongoing to assess the implications for the cost-effectiveness of the interventions.

Equity, cost and disability adjusted life years of tuberculosis treatment supported by digital adherence technologies and differentiated care in Ethiopia: a trial-based distributional cost-effectiveness analysis

“…Differences in provider costs between the SOC and intervention arms of the study were primarily driven by the number of healthfacility visits per treatment episode. In a separate analysis, we showed that health care workers spent more time with PWTB per visit in the intervention arms than when they were attending more frequently in the SOC arm [23]. This increase in time spent remained lower than the cumulative time of multiple visits in the SOC.…”