The effect of hosting 3.4 million refugees on native population mortality

Aysun Aygün, Murat Güray Kırdar, and Berna Tuncay

Journal of Health Economics, Volume 80 (2021), Article Number 102534
https://doi.org/10.1016/j.jhealeco.2021.102534

Review

This paper examines the effects of Syrian refugees in Turkey on the health care resources and the mortality outcomes of natives, focusing on neonatal, infant, child, and elderly mortality.

At the end of 2017, when this analysis was undertaken, Turkey hosted 3.4 million Syrian refugees who can access public health services free of charge. Syrian refugees arriving in Turkey had substantial health care needs and high utilization of health services due to injuries, poor living conditions and the disruption of health services in Syria, and higher prevalence of several diseases.

The authors compare provinces with high concentrations of refugees to those with low concentrations of refugees, before and after the arrival of refugees. The analysis is based on several data sources, including (1) annual, province-level data from the Turkish Statistical Institute (TurkStat) covering population, births, deaths, and causes of deaths; (2) data from the Turkish Ministry of Health on numbers of health workers and hospital beds in the health sector; and (3) data from multiple sources on the number of Syrian refugees across provinces from 2009 to 2017.

Main results:

  • The Turkish government responded to the Syrian refugee influx by increasing the number of nurses, pediatricians, and hospital beds in refugee hosting provinces. A 10-percentage point increase in the ration of refugees to natives led to a 7-8 percent increase in the number of nurses and an 8 percent increase in the number of hospital beds. There isn’t any evidence of an increase in the overall number of doctors, midwives, or intensive care beds in response to the arrival of Syrian refugees.
  • The investment of the Turkish government in extra nurses and hospital beds prevented an adverse effect of the refugee influx on these inputs in per capita terms. There isn’t any effect of refugees on the number of nurses or hospital beds per capita.
  • The numbers of doctors, midwives, hospitals, and adult intensive care beds did not keep up with the increase in population due to refugee arrivals, and the numbers of these health inputs per capita all declined. A 10-percentage point increase in the ratio of refugees to natives decreased the number of doctors per person by around 6-9 percent.
  • Refugees are more likely to settle in provinces where infant and child mortality is declining more slowly than elsewhere in the country. Refugee arrivals have an adverse effect on infant and child mortality because refugees are more likely to settle in provinces close to the Syrian border, where infant and child mortality is declining at a slower rate relative to other provinces.
  • After accounting for the settlement patterns of refugees, there isn’t any evidence of an effect of refugee arrivals on native mortality for any age group.

The authors conclude that the arrival of large numbers of Syrian refugees has strained the physical and human resources of the Turkish health care system. However, this shock does not translate into increases in native mortality for any age group. The authors suggest several possible reasons for this result: (1) the increase in the number of pediatricians in refugee-hosting provinces may have prevented or reduced any impact of refugee arrivals on infant and child mortality in the native population; (2) a Family Medicine Program implemented between 2005 and 2009 increased resources for maternal and infant health and substantially increased immunization levels; (3) Turkey has a universal health coverage system; (4) there may have been excess capacity in the health system; and (5) the native population may have switched to the private health system (although this is not evident in the data). The author cautions, however, that even though pressure on the public health care system did not adversely affect mortality rates, it could worsen the satisfaction of the native population with health services or have eventual effects on health outcomes of the native population.