Estimating the costs of developing and implementing a digital quality improvement tool designed to improve newborn care and survival in Malawi and Zimbabwe: Neotree (Preprint)

Abstract

Two thirds of the 2.4 million newborn deaths that occur within the first 28 days of life could potentially be avoided through ensuring existing low-cost evidence-based interventions are implemented for all sick and small newborns. An open-source digital quality improvement tool (Neotree) that combines data capture with education and clinical decision support is a promising solution to this implementation gap.

We present results from cost analysis of pilot implementation of Neotree in three hospitals in Malawi and Zimbabwe.

We used a combination of activity-based costing and expenditure approaches to estimate the cost of developing and pilot implementation of Neotree in one hospital in Malawi (Kamuzu Central Hospital (KCH)) and two hospitals in Zimbabwe (Sally Mugabe Central Hospital (SMCH) and Chinhoyi Provincial Hospital (CPH)). We estimated the costs from a provider perspective for the time horizon of 12 months. Data were collected through expenditure reports, monthly staff time-use surveys and interviews with the project staff. Sensitivity and scenario analyses were conducted to assess the impact of uncertainties on the results or estimate potential costs at scale. A pilot time-motion survey was conducted at KCH and a comparable hospital where Neotree was not implemented.

: Total cost of pilot implementation of Neotree at KCH, SMCH and CPH was US$37,748, US$52,331 and US$41,764, respectively. Average monthly cost per admitted child was US$15, US$15 and US$58, respectively. Staff costs were the main cost component with on average 73% of total costs, ranging from 63% to 79%. The results from sensitivity analysis showed that uncertainty around the number of admissions had a significant impact on the costs in all hospitals. In Malawi, replacing monthly web hosting with a server also had a significant impact on the costs. Under routine (non-research) conditions and at scale, total costs are estimated to reduce substantially, up to 76%, reducing cost per admitted child to as low as $5 in KCH, $4 in SMCH and $14 in CPH. Median time to admit a baby was 27 minutes (IQR: 20-40; n=250) using Neotree, compared to 26 minutes (IQR: 21-30; n=34) using paper-based systems and to discharge a baby was 9 minutes (IQR: 7-13; n=246) versus 3 minutes (IQR: 2-4, n=50), respectively.

Neotree is a time- and cost-efficient tool, comparable with the results from limited similar mHealth decision support tools in low- and middle-income countries. Implementation costs of Neotree varied substantially between the hospitals, mainly due to hospital size. The implementation costs could be substantially reduced at scale due to economies of scale because of integration to the health systems and reductions in cost items such as staff and overhead. More studies on assessment of impact and cost-effectiveness of large scale mHealth decision support tools are needed.