Macroeconomic Indicators and Health Service Efficiency in Iran across Stress and Normal Regimes: A Markov Switching Approach

Abstract

Background: The efficiency of health services is a fundamental indicator of health system performance, and macroeconomic conditions are linked to it through public budgets, the cost of imported medicines and equipment, and households' ability to pay. This study examined the association of Gross Domestic Product (GDP), exchange rate, and inflation with the efficiency of health services in Iran across distinct regimes over 2000–2023.

Methods: This applied, analytical time-series study used annual data for 2000–2023 collected from the World Bank and the Central Bank of Iran. Technical efficiency was estimated with an output-oriented Banker–Charnes–Cooper (BCC) data envelopment analysis model with five inputs and the inverse crude mortality rate as the output. The association of the cyclical components of GDP, exchange rate, and inflation with efficiency was estimated with a two-regime Markov switching regression controlling for COVID-19, in Python.

Results: Mean technical efficiency was 0.969; 11 of 24 years lay on the frontier, and the lowest score occurred in 2021 (0.739). The two-regime model outperformed the linear model (bootstrap p = 0.005). In the stress regime (2003, 2010, 2011, and 2021–2023), the deviation of the exchange rate from its trend was negatively and significantly associated with efficiency (p < 0.001), The relationship remained statistically significant across all sensitivity specifications, but its direction was sensitive to how exchange-rate fluctuations were measured; specifically, the exchange-rate coefficient in the stress regime became positive in the growth-rate specification. (0.017); whereas the associations of GDP and inflation were not robust. No indicator was significant in the normal regime. Expected durations were 2.50 years for the stress regime and 6.00 years for the normal regime.

Conclusion: The association between macroeconomic indicators and health service efficiency in Iran is not uniform. During periods of economic stress, the significant association between exchange rate fluctuations and efficiency was robust across all specifications, but its direction depended on how the fluctuation was measured (negative for deviation from trend and positive for growth rate); thus, only the significance of the association is robust, not its direction. Accordingly, continuous monitoring of efficiency and protective arrangements for financing imported health inputs during currency stress are recommended. The findings indicate association, not causation.