학술논문

A meta-analysis approach for estimating average unit costs for ART using pooled facility-level primary data from African countries.
Document Type
Article
Source
African Journal of AIDS Research (AJAR). Dec2019, Vol. 18 Issue 4, p297-305. 9p.
Subject
*MEDICAL economics
*HEALTH facilities
*HEALTH services administration
*HIV infections
*HOSPITAL costs
*META-analysis
*PATIENTS
*POLICY sciences
*POPULATION geography
*QUALITY assurance
*SYSTEMATIC reviews
*HEALTH insurance reimbursement
*ANTIRETROVIRAL agents
*ELECTRONIC health records
*DESCRIPTIVE statistics
Language
ISSN
1608-5906
Abstract
Objective: To estimate facility-level average cost for ART services and explore unit cost variations using pooled facility-level cost estimates from four HIV empirical cost studies conducted in five African countries. Methods: Through a literature search we identified studies reporting facility-level costs for ART programmes. We requested the underlying data and standardised the disparate data sources to make them comparable. Subsequently, we estimated the annual cost per patient served and assessed the cost variation among facilities and other service delivery characteristics using descriptive statistics and meta-analysis. All costs were converted to 2017 US dollars ($). Results: We obtained and standardised data from four studies across five African countries and 139 facilities. The weighted average cost per patient on ART was $251 (95% CI: 193–308). On average, 46% of the mean unit cost correspond to antiretroviral (ARVs) costs, 31% to personnel costs, 20% other recurrent costs, and 2% to capital costs. We observed a lot of variation in unit cost and scale levels between countries. We also observed a negative relationship between ART unit cost and the number of patients served in a year. Conclusion: Our approach allowed us to explore unit cost variation across contexts by pooling ART costs from multiple sources. Our research provides an example of how to estimate costs based on heterogeneous sources reconciling methodological differences across studies and contributes by giving an example on how to estimate costs based on heterogeneous sources of data. Also, our study provides additional information on costs for funders, policy-makers, and decision-makers in the process of designing or scaling-up HIV interventions. [ABSTRACT FROM AUTHOR]