eRegistries: Electronic registries for maternal and child health

Background: The Global Roadmap for Health Measurement and Accountability sees integrated systems for health information as key to obtaining seamless, sustainable, and secure information exchanges at all levels of health systems. The Global Strategy for Women’s, Children’s and Adolescent’s Health aims to achieve a continuum of quality of care with effective coverage of interventions. The WHO and World Bank recommend that countries focus on intervention coverage to monitor programs and progress for universal health coverage. Electronic health registries – eRegistries – represent integrated systems that secure a triple return on investments: First, effective single data collection for health workers to seamlessly follow individuals along the continuum of care and across disconnected cadres of care providers. Second, real-time public health surveillance and monitoring of intervention coverage, and third, feedback of information to individuals, care providers and the public for transparent accountability. This series on eRegistries presents frameworks and tools to facilitate the development and secure operation of eRegistries for maternal and child health. Methods: In this first paper of the eRegistries Series we have used WHO frameworks and taxonomy to map how eRegistries can support commonly used electronic and mobile applications to alleviate health systems constraints in maternal and child health. A web-based survey of public health officials in 64 low- and middle-income countries, and a systematic search of literature from 2005-2015, aimed to assess country capacities by the current status, quality and use of data in reproductive health registries. Results: eRegistries can offer support for the 12 most commonly used electronic and mobile applications for health. Countries are implementing health registries in various forms, the majority in transition from paper-based data collection to electronic systems, but very few have eRegistries that can act as an integrating backbone for health information. More mature country capacity reflected by published health registry based research is emerging in settings reaching regional or national scale, increasingly with electronic solutions. 66 scientific publications were identified based on 32 registry systems in 23 countries over a period of 10 years; this reflects a challenging experience and capacity gap for delivering sustainable high quality registries. Conclusions: Registries are being developed and used in many high burden countries, but their potential benefits are far from realized as few countries have fully transitioned from paper-based health information to integrated electronic backbone systems. Free tools and frameworks exist to facilitate progress in health information for women and children. The WHO RHR, the Johns Hopkins University Global mHealth Initiative, the United Nations Children’s Fund, and frog Design jointly developed the “mHealth and ICT [Information and Communication Technologies] Framework” to describe commonly used mHealth applications in RMNCH [24], and subsequently used for systematic review of evidence [25]. We used this framework and the taxonomy for primary health systems constraint categories developed by the WHO mHealth Technical and Evidence Review Group (mTERG) for RMNCH [26] to identify applications to alleviate common health systems constraints that can be supported by an eRegistry (Fig. 1). Framework for eRegistries in support of UHC in RMNCH. eRegistries in support of the 12 common electronic and mobile applications to overcome the seven principal constraints for universal health coverage in RMNCH, adapted from [24, 26] Health officials working in any of the 75 LMIC monitored by the Commission on Information and Accountability for Women’s and Children’s Health, and Palestine, were recruited by emails to RMNCH medical and health organizations, ministries of health, public health institutes, and other relevant government offices (e.g., statistics bureaus, RMNCH departments, etc.). The survey was reviewed and provided with a Letter of Exemption by the Regional Committees for Medical and Health Research Ethics in Norway, confirming that the anonymous survey was not medical research on human subjects, and did not need ethical approval (Reference number: IRB 0000 1870). Launched in November 2013, responses were accepted until February 2015. The sample consisted of 298 individuals from 64 countries. Approximately two-thirds of respondents worked at the national or regional level. The survey included questions on national registry infrastructure, reporting and dissemination practices, and data quality. Issues such as privacy legislation, access by individuals and professionals, and data security are presented in the third paper, as well as additional details on the survey methodology (Myhre et al, submitted). Country-level values are presented for all results. Decision rules for combining multiple responses into a single country response were adapted for each question; country averages were calculated for continuous data. Generalized linear models (PROC GLIMMIX) were used for confidence intervals around continuous data, while exact confidence intervals were calculated around binary and categorical data using SAS 9.4. A systematic search of literature was conducted including papers from 2005–March 2015 using Medline, Embase, ISI Web of Science, Cochrane Library and Global Health. The searches used terms indicative of RMNCH registries and were limited to the 76 LMIC as above (Additional file 1). After de-duplication, 4237 articles were identified. We included studies based on longitudinal data collection systems for individual level RMNCH data, and excluded all alternative data collections such as cross sectional surveys and health record document reviews. Two investigators independently scored publications for inclusion and extracted data. We included the following data points (within the categories in parentheses): the country/-ies of operation, the extent of the registry data collection (in facilities only, in community services only, both, or not defined), the scale of the implementation of the registry (national, district, local, or not defined), the specified population captured by the registry data collection (a total population, only subgroups/select population, or not defined), the data collection method used (paper, electronic, both, or not defined), whether the primary data was collected and entered directly into the registry, or if the registry was based on a secondary/duplicate data collection from existing sources (direct, duplicated, or not defined). In cases of conflicting scores/data, consensus was reached after independent scoring by a third investigator. Full text was read for 302 publications with abstracts considered potentially relevant (Additional file 1).