Lucia Fiestas Navarrete (Bocconi University)

Background

The consequences of women’s social status on health are widely discussed in the literature, yet partly owing to the difficulty in measuring the complex phenomenon of disempowerment, few have offered insight into how the conditions of social life affect women’s health utilization. Our main objective is to investigate the effect of female empowerment, in the form of marital age, on women’s use of cervical and breast screening in India, while exploring the channels that mediate and moderate the relationship.

Methods

We linked data from a sample of 15,903 ever-married women (15-49 years) who participated in the National Family Health Survey in 2015-16 with district-level data from the Indian Ministry of Health on per capita health facility supply. Drawing on Lauver’s theory of care-seeking behaviour, we incorporate women’s health insurance status into the analysis as a critical factor facilitating the external conditions that enable women to seek care. Our empirical strategy relies on a twofold instrumental variable approach seeking to generate quasi-random sources of variation in marital age and insurance participation. Specifically, we instrument age of first union by age of first menstruation and insurance status by cluster insurance rate in order to isolate their causal influence on women’s health utilization. Using structural equation modelling, we test (i) whether financial self-efficacy mediates the relationship between marital age and opportunistic screening and (ii) to what extent the district-level per capita supply of community health centres (CHCs) moderates the strength of the mediating channel.

Findings

The prevalence of ever having cervical and breast screening is 23 and 9 percent, respectively. We show that with every additional year that marriage is postponed, the probability of a woman ever having a cervical exam significantly increases by 7.9 percentage points and that of having a breast exam by 5.1. We find that higher marital age is associated with improvements in financial self-efficacy, which in turn, positively influence women’s use of cervical and breast screening. This is an important finding as it reveals financial self-efficacy to be consistent with a ‘second-stage’ form of empowerment resulting from marriage postponement. We also show that the indirect effect of marital age on cervical screening mediated through financial self-efficacy is positive and significant for any value of district supply greater than 3.6 CHCs per million inhabitants. This points to the critical burden that a poor supply context may have on women’s health utilization despite the relative improvements in autonomy associated with marital age.

Interpretation

Our findings suggest that losses in female empowerment attributed to early marriage partly explain Indian women’s low cervical and breast screening participation. Interventions aiming to address the uniquely high prevalence of women’s cancers in India would be wise to reflect on the health utilization consequences that can be attributed to socio-cultural practices exacerbating female disempowerment. Enforcing legal protections against underaged marriage is an actionable arena where both social and health agendas may converge in the interest of women – and good public policy.