in: Multiple Criteria Decision Making - Beyond the Information Age , Topcu I. Ozaydın O. Kabak O. Onsel Ekici S, Editor, Springer, London/Berlin , Basel, pp.25-50, 2021
Medicalization of childbirth services and regional differences are the 4
major obstacles in the improvement of women and child health in Turkey. The 5
present study analyzes the geographic distribution of the efficiency of childbirth 6
services in Turkish provinces. Data was collected from the official statistical records 7
of the 2017 Public Hospitals Statistical Yearbook. Charnes, Cooper, and Rhodes’ 8
(Eur J Oper Res 2(6):429–444, 1978) input-oriented data envelopment analysis 9
(DEA) was applied to determine provincial efficiency scores, using childbirth- 10
specific input and output indicators. Jackknife analysis was used for a robustness 11
check of the DEA scores. Four different DEA models were constructed, and the 12
final model’s efficiency scores were recorded. Finally, a decision-tree procedure was 13
integrated into the DEA results, and predictors of efficient and inefficient provinces 14
were examined. A total of 81 provinces in Turkey, representing seven geographic 15
regions, were included in the analysis. The results showed that 18% of the provinces 16
were efficient in terms of childbirth services. Average efficiency scores were high 17
(0.71) for provinces located in the Southeast Anatolia Region. The most important 18
predictor of efficiency for childbirth services is the number of beds in neonatal 19
intensive care units (Neo_int_n_b). A geographic distribution of the provincial 20
efficiency scores of childbirth services shows that eastern Turkey has the highest 21
score. Neo_int_n_b is the most important determinant of efficiency scores. Ensuring 22
public-health managers’ awareness about and continuous monitoring of childbirth 23
services, while focusing more on regional differences, is essential to improve the 24
status of children’s health in Turkey.