Decomposing rural-urban preterm birth variations in Tanzania

Article type: Research Article

Authors: Ndifwa, Nelson^{a; *} | Mukyanuzi, Elevatus^b | Lameck, Tolo^a

Affiliations: [a] Eastern Africa Statistical Training Centre, Tanzania | [b] University of Dar es Salaam, Dar es Salaam, Tanzania

Correspondence: [*] Corresponding author: Nelson Ndifwa, Eastern Africa Statistical Training Centre (EASTC), Tanzania. E-mail: [email protected].

Abstract: This study aimed at decomposing sources of rural-urban preterm birth variations in Tanzania. It used cross-sectional secondary data from the 2015/16 Tanzania Demographic and Health Survey. The study determined the extent to which rural-urban preterm birth variations are attributed to socio-economic characteristics (endowments) variations. The study employed binary logistic regression and multivariate Oaxaca blinder decomposition method, and findings reveal that, in urban areas, low parity, women anemia, high birth interval, women being employed and women BMI were statistically significant associated with low preterm births. Moreover, findings also show that domestic severe violence was statistically significantly associated with a high prevalence of preterm births. Furthermore, the findings showed that if the distribution in factor endowments including parity, anemia and birth interval were the same in both rural and urban areas preterm birth caused by the mentioned factors would drop in rural by 491.7%, 13.8% and 9% respectively. Therefore the study recommends tailoring the policies that empower women accessibility of resources and opportunities and creates awareness to women’s health issues which directly affect preterm births so as to reduce the problem.

Keywords: Preterm birth, Oaxaca Blinder decomposition method

DOI: 10.3233/SJI-220944

Journal: Statistical Journal of the IAOS, vol. 38, no. 4, pp. 1251-1259, 2022