Efficiency in COVID-19 inpatient care: findings from public hospitals in Iran

Improving efficiency is one of the high-potential options for expanding fiscal space for health. During the pandemic, as health systems' financial challenges intensify, the importance of utilizing resources efficiently also increases. Therefore, this study was conducted to estimate the efficiency of Iran's public hospitals in treating COVID-19 inpatient cases.

This descriptive-analytical study was based on administrative claims data from the Iran Health Insurance Organization and included 439,327 COVID-19 inpatient cases across 493 public hospitals in 2021. Epidemic waves were first identified using time-series data on daily admissions, and each patient was assigned to the corresponding wave. A logistic regression model was then fitted to estimate the probability of death based on age, sex, ICU admission, and epidemic wave. From this, a risk-adjusted survival variable (RA_surv) was calculated at the individual level and subsequently aggregated at the hospital level. Hospital efficiency was assessed using input-oriented data envelopment analysis (DEA) under variable returns to scale, with total inpatient billed charges for COVID-19 as the input and RA_surv as the output. Finally, to account for structural and contextual differences, DEA efficiency scores were adjusted using a fractional logit regression model that incorporated teaching status, specialty type, hospital size, and province fixed effects.

The mean of per-patient charge for COVID-19 treatment was estimated at USD 236.46 (SD = 234.48; median = 185.73), and the mean daily hospital charge was USD 46.34 (SD = 26.41; median = 41.62). These figures varied considerably across provinces, with the highest per-patient charge observed in Tehran (USD 364.98) and the lowest in South Khorasan (USD 171.37). Overall hospital efficiency scores before contextual factors adjustment ranged from 0.083 to 1.00. After adjustment, the national mean remained 0.49, although the distribution and ranking of hospitals shifted. A strong positive correlation was found between non-adjusted and adjusted efficiency scores (Spearman's rho = 0.707, p < 0.001).

This study indicated significant variation in hospital charges in COVID-19 inpatient bills in Iran's public hospitals, and there was a relatively significant potential to save resources during the financial difficulties of Iran's health system during the pandemic. Adopting appropriate strategies to reduce variation in clinical practice, for example, promoting the use of clinical guidelines, can significantly help reduce variation in hospital charges and subsequently improve the system's efficiency.