Intersectional inequalities in healthcare utilisation in informal settlements in Freetown, Sierra Leone a multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

Introduction: Residents of informal settlements face significant intersectional inequalities, due to the overlapping and compounding effects of multiple social factors. This study aims to explore how these intersecting social factors, identified by community members, combine to shape household-level inequalities healthcare utilisation (HU) among residents of informal settlements in Freetown, Sierra Leone. 

Methods: This study employed participatory action research to collaboratively identify key social determinants affecting healthcare utilisation in Freetown’s informal settlements. A cross-sectional health and wellbeing survey was implemented in April-May 2023 and collected data from 4,871 households in Cockle Bay, Dwazark, and Moyiba informal settlements. The survey questions were codesigned by researchers and community fieldworkers, informed by prior qualitative research. Two outcomes were analysed: HU within the settlement (n = 4,821), and outside the settlement (n = 4,616). A multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) was conducted, nesting households within 122 intersectional strata. These strata were defined by six social factors: head of household gender, marital composition, engagement in income-generating activity, food security, disability and the household’s settlement. Intersectional measures included variance partition coefficient (VPC), the proportional change in variance (PCV), and residual intersectional effects. 

Results: VPCs of 0.9% (PCV, of 92.8%) for HU within the settlements and of 3.9% (PCV, 81.7%) for HU outside the informal settlements suggest moderate but meaningful intersectional effects in shaping HU inequalities. The lowest levels of HU within informal settlements were observed among single, male, disabled individuals in Moyiba who lacked income-generating activities and experienced food insecurity. For HU outside the settlement, the lowest levels were found among female-headed households in Moyiba who were married, cohabiting, or engaged with a disabled household member, experienced food insecurity, and were engaged in income-generating activities. 

Conclusion: This study identifies and quantifies inequalities in HU at the household level across three informal settlements in Freetown, driven by intersecting social factors. Addressing these inequalities requires policies that are universally accessible but implemented with an intensity proportionate to the level of vulnerability, ensuring that support is targeted to those most in need.