Dataset Information

Improving imperfect data from health management information systems in Africa using space-time geostatistics.

ABSTRACT:

Background

Reliable and timely information on disease-specific treatment burdens within a health system is critical for the planning and monitoring of service provision. Health management information systems (HMIS) exist to address this need at national scales across Africa but are failing to deliver adequate data because of widespread underreporting by health facilities. Faced with this inadequacy, vital public health decisions often rely on crudely adjusted regional and national estimates of treatment burdens.

Methods and findings

This study has taken the example of presumed malaria in outpatients within the largely incomplete Kenyan HMIS database and has defined a geostatistical modelling framework that can predict values for all data that are missing through space and time. The resulting complete set can then be used to define treatment burdens for presumed malaria at any level of spatial and temporal aggregation. Validation of the model has shown that these burdens are quantified to an acceptable level of accuracy at the district, provincial, and national scale.

Conclusions

The modelling framework presented here provides, to our knowledge for the first time, reliable information from imperfect HMIS data to support evidence-based decision-making at national and sub-national levels.

SUBMITTER: Gething PW

PROVIDER: S-EPMC1470663 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundCOVID-19 rapidly spread around the world, putting health systems under unprecedented pressure and continuous adaptations. Well-established health information systems (HIS) are crucial in providing data to allow evidence-based policymaking and public health interventions in the pandemic response. This study aimed to compare morbidity information between two databases for COVID-19 management in Portugal and identify potential complementarities.MethodsThis is an observational study using records from both COVID-19 cases surveillance (National Epidemiological Surveillance System; SINAVE) and related deaths (National e-Death Certificates Information System; SICO) systems, which were matched on sex, age, municipality of residence and date of death. After the linkage, morbidity reported in SINAVE and identified in SICO, through the application of Charlson and Elixhauser comorbidity indexes algorithms, were compared to evaluate agreement level.ResultsOverall, 2285 matched cases were analyzed, including 53.9% males with a median age of 84 years. According to the method of data reporting assessment, the presence of any morbidity ranged between 26.3% and 62.5%. The reporting of ten morbidities could be compared between the information reported in SINAVE and SICO databases. The proportion of simultaneous reporting in both databases ranged between 5.7% for diabetes and 0.0% for human immunodeficiency virus infection or coagulopathy. Minimal or no agreement was found when assessing the similarity of the morbidity reporting in both databases, with neoplasms showing the highest level of agreement (0.352, 95% IC: 0.277-0.428; p < 0.001).ConclusionDifferent information about reported morbidity could be found in two HIS used to monitor COVID-19 cases and related deaths, as data are independently collected. These results show that the interoperability of SICO and SINAVE databases would potentially improve available HIS and improve available information to decision-making and address COVID-19 pandemic management.

Project description:BACKGROUND:As part of a partnership between the Institute for Healthcare Improvement and the Ethiopian Federal Ministry of Health, woreda-based quality improvement collaboratives took place between November 2016 and December 2017 aiming to accelerate reduction of maternal and neonatal mortality in Lemu Bilbilu, Tanqua Abergele and Duguna Fango woredas. Before starting the collaboratives, assessments found inaccuracies in core measures obtained from Health Management Information System reports. METHODS AND RESULTS:Building on the quality improvement collaborative design, data quality improvement activities were added and we used the World Health Organization review methodology to drive a verification factor for the core measures of number of pregnant women that received their first antenatal care visit, number of pregnant women that received antenatal care on at least four visits, number of pregnant women tested for syphilis and number of births attended by skilled health personnel. Impact of the data quality improvement was assessed using interrupted time series analysis. We found accurate data across all time periods for Tanqua Abergele. In Lemu Bilbilu and Duguna Fango, data quality improved for all core metrics over time. In Duguna Fango, the verification factor for number of pregnant women that received their first antenatal care visit improved from 0.794 (95%CI 0.753, 0.836; p<0.001) pre-intervention by 0.173 (95%CI 0.128, 0.219; p<0.001) during the collaborative; and the verification factor for number of pregnant women tested for syphilis improved from 0.472 (95%CI 0.390, 0.554; p<0.001) pre-intervention by 0.460 (95%CI 0.369, 0.552; p<0.001) during the collaborative. In Lemu Bilbilu, the verification factor for number of pregnant women receiving a fourth antenatal visit rose from 0.589 (95%CI 0.513, 0.664; p<0.001) at baseline by 0.358 (95%CI 0.258, 0.458; p<0.001) post-intervention; and skilled birth attendance rose from 0.917 (95%CI 0.869, 0.965) at baseline by 0.083 (95%CI 0.030, 0.136; p<0.001) during the collaborative. CONCLUSIONS:A Data quality improvement initiative embedded within woreda clinical improvement collaborative improved accuracy of data used to monitor maternal and newborn health services in Ethiopia.