Kristefer Stojanovski

Associations between racial discrimination, smoking during pregnancy and low birthweight among Roma.

Background Racial discrimination may increase the risk of low birthweight (LBW), but has not been studied among Roma, the largest minority population in Europe. Moreover, few studies test both institutional and interpersonal forms of racial discrimination on health. Our objective was to examine associations between institutional and interpersonal racial discrimination with LBW, and to test potential mediation by smoking during pregnancy. In 2012-2013, Romani women interviewers surveyed 410 Romani women in Serbia and Macedonia. We measured institutional discrimination (neighborhood segregation, legal status of housing and neighborhood socioeconomic status), interpersonal discrimination [Everyday Discrimination Scale (EDS)], birthweight and smoking by self-report or interviewer report. We estimated relative risks for discrimination on LBW and separately on smoking during pregnancy using log-binomial regression, adjusting for age, parity, years at residence and wealth. The indirect effect of high EDS via smoking on LBW was estimated using inverse odds weighting mediation. Living in a low SES neighborhood showed a 2-fold risk of LBW [adjusted risk ratio (aRR) = 2.4, 95% CI = 1.2, 5.0]; aRRs for segregation and illegal housing were weaker (aRR = 1.8, 95% CI = 0.7, 4.3; aRR = 1.3, 95% CI = 0.6, 2.6, respectively). Institutional measures were not associated with smoking. High EDS was associated with LBW (aRR = 2.4, 95% CI = 1.1, 5.2) and smoking during pregnancy (aRR = 1.4, 95% CI = 1.1, 1.8); the indirect effect of EDS on LBW via smoking was not significant. Interpersonal discrimination and living in a low SES neighborhood were associated with LBW among Roma. Interventions to improve Romani health may benefit from a human rights approach.

Health insurance and accessibility to health services among Roma in settlements in Belgrade, Serbia—the journey from data to policy making

The Serbian constitution and health-related laws assert that citizens and residents are universally entitled to health care, provided that they hold health insurance. However, until 2010, persons who did not hold a national identification number (ID) were required to present a plethora of documents to be granted one. We assessed the relationship between citizenship, residency and possession of health insurance cards, together with utilization of health services, among Roma residing in disadvantaged settlements in Belgrade. The Roma Health and Nutrition Survey was conducted in 2009 to assess the social determinants of health among Roma. Data were analysed, using logistic regression, to examine health insurance status and utilization of services by citizenship and residency. Eighty-nine per cent of respondents said they were Serbian citizens. Approximately 11% were refugees, 7% internally displaced persons (IDPs) and remainder domicile. Multivariate analysis revealed that non-citizens were more likely to lack health insurance [odds ratio (OR) = 9.2, confidence interval (CI) (3.5, 24.1)], as were refugees and IDPs [OR = 3.1, CI (1.4, 6.9), OR = 4.0, CI (1.4, 11.5), respectively]. Having health insurance was a positive predictor for being seen by a physician [OR = 2.3, CI (1.3, 4.2), OR = 2.3, CI (1.3, 3.9)]. Data from this survey indicated that non-citizen Roma had limited access to health services. These findings led the Serbian Ministry of Health and National Health Insurance Fund to reduce the administrative and legislative hurdles in obtaining health insurance, to ensure the Roma rights to health care. This demonstration of data-driven policies on Roma health could serve as a model for other countries.

Low levels of modern contraceptive use and associated factors in the Western Balkans

Abstract Objectives: The objective of this research was to determine modern contraceptive use (MCU) and explore factors associated with MCU in the Western Balkans. Methods: Multiple Indicator Cluster Survey (MICS) data from Bosnia and Herzegovina, Kosovo, Macedonia, Montenegro and Serbia, including the datasets from Roma settlements, were merged. Bivariate and multivariate analyses were used to assess which factors were associated with MCU among married women aged 15–49 in the region. Results: Among the 11,381 women surveyed, 13.75% reported MCU and 23.74% justified domestic violence. While bivariate analysis indicated that women who expressed justification of domestic violence were less likely to report MUC, this was not the case after adjusting for sociodemographic factors. Multivariate analysis showed that Roma were less likely than non-Roma to engage in MCU. Additionally, increased education, increased wealth, and urban residency were significantly associated with increased likelihood of MCU. Conclusion: The results highlight the need for more attention from research, programming and policy communities in order to better understand and address the overall problematically low level of MCU and factors associated with low MCU. More efforts are needed to address the disparities in MCU among Roma women and women of lower socioeconomic status.

Predictors of prenatal smoking and attempted smoking cessation during pregnancy: a community-based study of Romani women in Southeastern Europe

Background The objective of this study was to examine predictors of prenatal smoking, and attempted smoking cessation during pregnancy among Romani women. Methods A community‐based, cross‐sectional study (November 2012 to February 2013) of 410 Romani women in Roma settlements in Serbia and Macedonia was conducted. Logistic regression was used to identify predictors of prenatal smoking and attempted smoking cessation during pregnancy. Results Romani women older than 30 years and those who were living with a man were over twice as likely (adjusted odds ratio (aOR) 2.48, 95% confidence interval (CI) 1.12‐5.46; aOR 2.09, 95% CI 1.27‐3.43) to smoke during pregnancy, compared to women <20 and married women, respectively. An inverse relationship between education and prenatal smoking was observed (for primary education versus no education, aOR 0.56, 95% CI 0.32‐0.98; for secondary or higher education versus no education, aOR 0.38, 95% CI 0.16‐0.90). Having a husband/partner who smokes was associated with significantly increased likelihood of prenatal smoking (aOR 3.71, 95% CI 2.20‐6.25) and decreased likelihood of attempting to quit (aOR 0.51, 95% CI 0.24‐1.06). Conclusions Culturally sensitive and comprehensive prevention strategies and intervention programs are needed to reduce smoking during pregnancy among Romani women, including interventions targeting male partners.