Differential effect of China's Zero Markup Drug Policy on provider-induced demand in secondary and tertiary hospitals.

Following the marketization of China's health system in the 1980's, the government allowed public hospitals to markup the price of certain medications by 15% to compensate for reduced revenue from government subsidies. This incentivized clinicians to induce patient demand for drugs which resulted in higher patient out-of-pocket payments, higher overall medical expenditure, and poor health outcomes. In 2009, China introduced the Zero Markup Drug Policy (ZMDP) which eliminated the 15% markup. Using Shanghai as a case study, this paper analyzes emerging and existing evidence about the impact of ZMDP on hospital expenditure and revenue across secondary and tertiary public hospitals. We use data from 150 public hospitals across Shanghai to examine changes in hospital expenditure and revenue for various health services following the implementation of ZMDP. Our analysis suggests that, across both secondary and tertiary hospitals, the implementation of ZMDP reduced expenditure on drugs but increased expenditure on medical services, exams, and tests thereby increasing hospital revenue and keeping inpatient and outpatient costs unchanged. Moreover, our analysis suggests that tertiary facilities increased their revenue at a faster rate than secondary facilities, likely due to their ability to prescribe more advanced and, therefore, more costly procedures. While rigorous experimental designs are needed to confirm these findings, it appears that ZMDP has not reduced instances of medical expenditure provoked by provider-induced demand (PID) but rather shifted the effect of PID from one revenue source to another with differential effects in secondary vs. tertiary hospitals. Supplemental policies are likely needed to address PID and reduce patient costs.

Using publicly available satellite imagery and deep learning to understand economic well-being in Africa.

Accurate and comprehensive measurements of economic well-being are fundamental inputs into both research and policy, but such measures are unavailable at a local level in many parts of the world. Here we train deep learning models to predict survey-based estimates of asset wealth across ~ 20,000 African villages from publicly-available multispectral satellite imagery. Models can explain 70% of the variation in ground-measured village wealth in countries where the model was not trained, outperforming previous benchmarks from high-resolution imagery, and comparison with independent wealth measurements from censuses suggests that errors in satellite estimates are comparable to errors in existing ground data. Satellite-based estimates can also explain up to 50% of the variation in district-aggregated changes in wealth over time, with daytime imagery particularly useful in this task. We demonstrate the utility of satellite-based estimates for research and policy, and demonstrate their scalability by creating a wealth map for Africa's most populous country.