M-Health Intervention Has the Potential to Alleviate Constraints in Using Administrative Data Systems for Improving Service Delivery in India

Abstract

Abstract Objectives Administrative Data (AD) are a key component of nutrition data systems, but little is known about how AD are used or can be reshaped for accountability and action. We studied factors influencing AD use in India's Integrated Child Development Services (ICDS), in two Indian states; and assessed how an m-Health intervention (Common Application Software (CAS)) changes it. ICDS-AD system consolidates data from 11-recording registers updated every month by ∼1.3 million frontline workers (FLW). CAS digitizes these records and creates dashboards for real-time monitoring. Methods We conducted semi-structured interviews (n = 82) with supervisory staff at sub-district [Lady Supervisors (LS); Child Development Project Officer (CDPO)] and district level [District Program Officer (DPO)] to understand AD use patterns, facilitators and barriers. Data were coded, summarized and analysed across levels of supervisory staff. Results LS supervise FLWs, visit villages monthly to monitor data, and collate it to the sub-district level. Due to limited time and pressures to update AD on time, LS focus on data completeness and not on quality or data use. At the next level, CDPOs review and discuss program targets for indicators with LS, and check data completeness. They create monthly reports on targets for the districts, based on LS reports. At the district, DPOs use compiled reports to review subdistrict performance and discuss any gaps with CDPOs. CAS makes the data entry, compilation and report generation easier for all users. LS can monitor data entry daily in CAS and data collation is automated. CDPOs can review the CAS dashboard and use it to discuss issues with LS and DPOs. However, program target indicators dominate use of data. In CAS areas, these include growth monitoring completion, home visits, identification of malnourished children, and their referral. Overall, data use is target-driven, with indicators for monitoring prioritized by program management. Conclusions Focus of data use has been on reporting up on target indicators rather than on local program improvement, due to staff and time constraints. CAS can address some AD use constraints by easing data entry and data compilation but a data culture that focuses on predetermined target indicators can limit use for local program improvement. Funding Sources Bill & Melinda Gates Foundation via grants to IFPRI and UCSF.