Data Set and Analysis Code for Out-of-Pocket Health Spending, Governance Capacity, and Antibiotic-Use Composition: Global Evidence from the WHO AWaRe Framework
Description
Research Hypothesis The study investigates whether a country's health financing structure and governance capacity are associated with the quality of its antibiotic consumption. Specifically, the primary hypothesis is that countries with weaker financial protection (characterized by high out-of-pocket health expenditures) and weaker public administration and regulatory capacities exhibit poorer antibiotic use quality. Data and Methodology The study utilizes an unbalanced country-year panel dataset from the 2016–2023 period, consisting of 409 observations across 71 countries, with a heavy representation of the European region. Rather than a fully representative global panel, it represents countries reporting AWaRe data to the WHO. The dependent variables are based on the WHO's AWaRe classification (Access, Watch, Reserve): Access share: Percentage of Access antibiotics in total consumption (higher is positive). Watch share: Percentage of Watch antibiotics (higher indicates a riskier composition). WHO 70% Access target: Binary variable indicating whether a country met the WHO target of at least 70% Access consumption. The main independent variables are: Out-of-pocket (OOP) expenditure share (financial protection proxy). Government effectiveness and Regulatory quality (governance proxies). Controls include the UHC index, log GDP per capita, population aged 65+, urbanization rate, and year/WHO region fixed effects. OLS, Fractional Logit, Binary Logit, and Mundlak Correlated Random Effects (CRE) models were used for estimation and robustness. Key Findings Financial Protection & Governance: Higher OOP share is significantly associated with lower Access share and higher Watch share. Conversely, stronger government effectiveness and regulatory quality correlate with higher Access and lower Watch shares. WHO 70% Target: High OOP reduces the probability of achieving the WHO target, while stronger governance and regulatory quality significantly increase it. Structural vs. Short-Term: The Mundlak CRE model reveals that within-country annual changes are mostly insignificant, meaning results are driven by structural differences between countries rather than short-term fluctuations. Robustness & UHC: Governance effects are highly robust across sample restrictions. OOP results are slightly sensitive to sample size, requiring careful interpretation. The UHC index shows no consistent relationship with antibiotic quality. Interpretation & Use These findings suggest that antibiotic usage quality is deeply rooted in the financial protection of the health system and the state's regulatory and administrative strength. Policymakers can use this data to design targeted interventions, acknowledging that improving rational antibiotic use requires enhancing financial protection and regulatory enforcement, rather than relying solely on expanding general healthcare coverage.