Family formation among South African youth: the role of HIV and socio-demographic factors

Abstract

BACKGROUND: In South Africa, the number of civil unions dropped from 186,522 in 2008 to 135,458 in 2017, while cohabitation and premarital childbearing have been on the rise. Research has found that female employment and education, cultural beliefs, place of residence and contraceptive use are some of the factors that are associated with family formation among youth in South Africa. However, given that young women aged 15-24 years are at heightened risk of HIV transmission in South Africa, little is known about how the pandemic affects the ability of youth to form families. Furthermore, literature in South Africa, has compared the fertility intentions of HIV Positive and HIV Negative women, however, it has not looked at family formation in its entirety. Therefore, this study examined the association between socio-demographic factors, HIV status and family formation among youth in South Africa. DATA AND METHODS: This study utilised secondary data which was acquired from the 2016 South African Demographic and Health Survey. The 2016 Demographic and Health Survey is a survey that included a sample of 8,514 females aged 15-49 years. The study population, however, consists of youth aged 15-34 years in 2016. Therefore, the study had a weighted sample size of 2 357 women aged 15-34 years, of which 1913 were HIV negative and 444 were HIV positive. The dependent variable in this study was family formation and it had four categories, namely, single and without children, single and with children, married and without children and married and with children. On the other hand, the independent variables were, HIV status, respondent’s age, place of residence, population group, highest level of education, employment status, household wealth index, told about family planning, pregnancy intention and condom use. The data was analysed in STATA version 15, and it was done in two phases. The first phase included a descriptive of the variables in the study, presented through a series of graphs and frequency tables. In the second phase, multinomial regression models were used to assess the association between socio-demographic factors, HIV status and family formation among youth in South Africa. The Multinomial Regression Models consisted of five models. Model 1 assessed the association between HIV status and Family formation. Model 2 assessed the effects of Model 1 and demographic factors. Model 3assessed the effects of Model 2 and socio-economic factors. Lastly, Model4 assessed the effects of model 3,and family planning and proximate factors. RESULTS: Among the HIV Negative youth, 17.88% were single and without children, 44.84% were single and with children, 4.12% were married and without children, while 33.15% were married and with children. Among the HIV Positive youth, 19.26% were single and without children, 46.78% were single and with children, 3.93% were married and without children, and 30.86% were married and with children. This therefore indicates that there were marginal differences in the family formation patterns of HIV Negative and HIV Positive youth in South Africa. After controlling for the socio-demographic characteristics, the regression analysis showed that HIV Positive youth had a 1.29 [CI: 0.673-2.473] times higher risk of being single and with children, compared to HIV Negative youth. The relative risk of being married and without children was 0.57 [CI: 0.197-1.667] times lower for HIV Positive youth as compared to HIV Negative youth. Lastly, the relative risk of being married and with children was 1.09 [CI: 0.483-2.444] times higher for HIV positive youth. However, HIV status was not significantly associated with family formation. The relative risk of being single and with children was significantly associated with respondent’s age [RRR=5.26, CI: 3.920-7.057] population group [RRR=3.14, CI: 1.413-6.984] and an unintended pregnancy [RRR=14.43, CI: 2.245-92.688]. The relative of being married and without children was significantly associated with the respondent’s age [RRR=2.89, CI: 1.952-4.280], rural residence [RRR=0.31, CI: 0.155-0.631], rich household wealth index [RRR=0.35, CI: 0.156-0.767] and condom use [RRR=0.20, CI: 0.096-0.434]. Lastly, the relative risk of being married and with children was significantly associated with the respondent’s age [RRR=10.03, CI: 7.127-14.135], rural residence [RRR=0.50, CI: 0.309-0.800], and condom use[RRR=0.44, CI: 0.270-0.728]. CONCLUSION AND RECCOMENDATION: The overall inference that was drawn from this study is that there are marginal differences in the family formation patterns of HIV Positive and HIV Negative youth in South Africa. This therefore suggests that HIV Positive youth are forming families that are similar to that of their HIV Negative peers. Furthermore, HIV status was not a significant predictor of family formation. This can be attributed to the increased access to ART, which has been found to decrease the risk of sexual and mother-to-children-transmission (MTCT) of HIV. However, unintended pregnancies were found to be significantly associated with family formation. This raises concerns of possible HIV transmission. Therefore, the National Contraception and Family Planning Policy needs to ensure that there is a proper integration of contraception and family planning services, with HIV and MTCT prevention services. Lastly, the Health Sector HIV Prevention Strategy of 2016, needs to integrate HIV services into their framework