Quality of routine health data at the onset of the COVID-19 pandemic in Ethiopia, Haiti, Laos, Nepal, and South Africa

Population Health Metrics

Research

Anna Gage¹  Catherine Arsenault¹  Neena R. Kapoor¹  Jean Paul Joseph²  Gebeyaw Molla³  Suresh Mehata⁴  Solomon Kassahun Gelaw⁵  Daniella Myriam Pierre⁶  Nompumelelo Gloria Mfeka-Nkabinde⁷  Londiwe Mthethwa⁷  Anagaw Derseh Mebratie⁸  Wondimu Ayele⁸  Adiam Nega⁸  Roody Thermidor⁹  Daisuke Asai^1,10  Dilipkumar Hensman^1,10 

[1] Department of Global Health and Population, Harvard T.H. Chan School of Public Health, Boston, USA;Division d’Épidémiologie et de Laboratoire, Zanmi Lasante, Mirebalais, Plateau Central, Haiti;Ethiopian Public Health Institute, Addis Ababa, Ethiopia;Ministry of Health and Population, Government of Nepal, Kathmandu, Nepal;Ministry of Health of Ethiopia, Addis Ababa, Ethiopia;Programme National de Lutte contre les IST/VIH/SIDA (PNLS) Unite de Coordination des Maladies Transmissibles (UCMIT), Ministère de la Sante Publique et de la Population (MSPP), Port-au-Prince, Haiti;School of Nursing and Public Health, University of KwaZulu-Natal, Durban, South Africa;School of Public Health, College of Health Sciences, Addis Ababa University, Addis Ababa, Ethiopia;Studies and Planning Unit, Ministry of Public Health and Population, Port-au-Prince, Haiti;World Health Organization, Vientiane, Lao People’s Democratic Republic;

关键词: Health management information systems;    DHIS2;    Data quality;    COVID-19;   

DOI  :  10.1186/s12963-023-00306-w

 received in 2022-08-05, accepted in 2023-05-14,  发布年份 2023

来源: Springer

【 摘 要 】

BackgroundDuring the COVID-19 pandemic, governments and researchers have used routine health data to estimate potential declines in the delivery and uptake of essential health services. This research relies on the data being high quality and, crucially, on the data quality not changing because of the pandemic. In this paper, we investigated those assumptions and assessed data quality before and during COVID-19.MethodsWe obtained routine health data from the DHIS2 platforms in Ethiopia, Haiti, Lao People’s Democratic Republic, Nepal, and South Africa (KwaZulu-Natal province) for a range of 40 indicators on essential health services and institutional deaths. We extracted data over 24 months (January 2019–December 2020) including pre-pandemic data and the first 9 months of the pandemic. We assessed four dimensions of data quality: reporting completeness, presence of outliers, internal consistency, and external consistency.ResultsWe found high reporting completeness across countries and services and few declines in reporting at the onset of the pandemic. Positive outliers represented fewer than 1% of facility-month observations across services. Assessment of internal consistency across vaccine indicators found similar reporting of vaccines in all countries. Comparing cesarean section rates in the HMIS to those from population-representative surveys, we found high external consistency in all countries analyzed.ConclusionsWhile efforts remain to improve the quality of these data, our results show that several indicators in the HMIS can be reliably used to monitor service provision over time in these five countries.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

40517_2023_256_Article_IEq62.gif

Fig. 3

40517_2023_256_Article_IEq60.gif