Results
In our study population, 41.0% (N=203,281) women were married in childhood and 38.8% (N=195,707) gave birth during adolescence. Adolescent childbearing was more likely among women wo were married as children. Approximately 79.4% of the women married in childhood gave birth by age 19 years, while only 10.7% of the women married as adults gave birth during adolescence. Table 1 presents characteristics of the study population by child marriage and adolescent childbearing groups. Larger percentages of women married in childhood and who gave birth in adolescence were from poor households, with lower educational attainment, and living in rural areas. While the distribution of BMI categories was comparable across the groups, the percentage of women giving birth to 5 or more children was significantly higher among women who were married before age 18 years and gave birth by age 19 years.
Prevalence of all-cause hysterectomy in our study population was 4.3% (N=20,659). Among women married as adults, prevalence of all-cause hysterectomy was 2.7% (95% CI: 2.6 – 2.8) for those who did not give birth in adolescence, and 4.4% (95% CI: 4.1 – 4.8) for those who gave birth in adolescence. Among women who were married in childhood, the prevalence of hysterectomy was 5.5% (95% CI: 5.2 – 5.8) and 6.7% (95% CI: 6.5 – 6.8) for those who did not and did give birth in adolescence, respectively. Figure 1 presents the prevalence rates of hysterectomy by the child marriage and adolescent childbearing groups across respondents’ demographic and socioeconomic (e.g., religion, household wealth) characteristics and risk-factors (e.g., BMI). Results of the adjusted Wald tests suggest that across all attributes, women who were married in childhood and who gave birth in adolescence had a significantly higher prevalence of hysterectomy than that of the reference group (i.e., women who were married as adults and did not give birth in adolescence). In general, the prevalence rates were also higher among women who were married in childhood and did not give birth in adolescence and women who were married as adults but gave birth in adolescence, compared to their counterparts in the reference group.
Unadjusted and adjusted odds ratios of hysterectomy for the child marriage and adolescent childbearing groups are presented in Table 2. After accounting sociodemographic attributes and risk factors (such as parity), women married in childhood who gave birth in adolescence were 1.87 times more likely to have a hysterectomy compared to women who were married as adults and did not give birth in adolescence. The adjusted odds were 1.53 and 1.40 times higher for women who were married as adults but gave birth in adolescence, and women who were married in childhood and did not give birth in adolescence, respectively.
The results for gynecologic problem specific hysterectomy outcomes were very similar to those of all-cause hysterectomy outcome (Table 2). For example, adjusted odds of hysterectomy due to excessive menstrual bleeding or due to fibroids/cysts were 1.82 and 1.71 times higher, respectively, for women who were married in childhood and gave birth in adolescence compared to their counterparts who were married as adults and did not give birth in adolescence.
Table S1 (in supplementary material) presents the results by sub-samples of household wealth index quintiles. Of note, compared to women from wealthier household, women from poorer households were more likely to get married before age 18 years and to give birth by age 19 years. We found that across all different wealth levels, women who were married in childhood and/or gave birth in adolescence had a significantly higher likelihood of having hysterectomy compared to women who were married as adults and did not give birth in adolescence.
Along with wealth, educational attainment and urban/rural residence were other factors considered to influence child marriage and adolescent childbearing. To mitigate the influence of such heterogeneities, models were also estimated by sub-samples of educational attainment and urban/rural residence, results of which are presented in Table S2. The greater likelihood of hysterectomy among women who were married in childhood and/or gave birth during adolescent age was evident across all educational levels and in both urban and rural areas.
Previous studies also indicated regional variations in hysterectomy prevalence in India.8, 9 Table S3 presents the results by samples of geographic regions in India. The relationship between child marriage, adolescent childbearing, and hysterectomy persisted across all regions despite differences in hysterectomy prevalence rates across the regions. For example, in the Southern region where the prevalence of hysterectomy was the highest, women who were married in childhood and gave birth in adolescence were 1.99 times more likely to undergo a hysterectomy compared to women who were married as adults and did not give birth in adolescence. The odds of hysterectomy were also higher (AOR: 1.56) for women who were married in childhood and gave birth in adolescence in the Northeastern region, where the prevalence of hysterectomy was the lowest.
Lastly, the K-M survivor functions for the event of having a hysterectomy by child marriage and adolescent childbearing groups are presented in Figure 2. Survival rates among women married as adults and not giving birth in adolescence were 98.1%, 96.3%, 94.4% and 92.5% at ages 35, 40, 45, and 44 years, respectively. Survival rates among women married in childhood who gave birth in adolescence were 95.8%, 93.6%, 91.3%, and 89.0%, respectively. Results of the log-rank test suggested that the survivor functions for different groups were not the same (i.e., the null hypothesis of equality of survivor functions across the groups was rejected). At any age (between 20 and 49), the survival probability was the lowest for women who were married before the age of 18 years and gave birth by the age of 19 years.