Associations between Extreme Weather Events, Resource Insecurities, and HIV Vulnerabilities Among Adolescent Girls and Young Women in Kenya

Background and Purpose: There is a growing focus on linkages between climate-related extreme weather events (EWE), such as drought and flooding, and HIV vulnerabilities across Eastern and Southern Africa. Yet the complex pathways accounting for these linkages are understudied, particularly with adolescent girls and young women (AGYW). We applied an ecosocial approach to assess environmental (EWE exposure) and social inequity (resource insecurity)-related factors associated with HIV vulnerabilities. We aimed to examine associations between EWE exposure, resource insecurity (food, water, and sanitation) profiles, and HIV vulnerabilities among AGYW in Kisumu and Nairobi, Kenya.

Methods: We conducted a cross-sectional survey (April-June 2024) with a purposive sample of AGYW (16-24 years old) in Nairobi (informal settlements: Majengo, Mathare) and Kisumu (fishing communities). We first conducted latent profile analysis (LPA) to identify distinct, unobserved subgroups based on food, water, and sanitation insecurity, classifying participants into mutually exclusive profiles according to response patterns. We then conducted multinomial logistic regression to assess associations between EWE exposure and resource insecurity profile membership. Following this, we conducted multivariable linear and logistic regression to examine associations between resource insecurity profile membership and HIV vulnerabilities (sexual relationship power [SRP]; past 12-month transactional sex; past 12-month intimate partner violence [IPV]; adolescent sexual and reproductive health (A-SRH) stigma). All regression models adjusted for age, employment, parental status, and location.

Results: Among participants (n=597; mean age: 20.13 years; standard deviation: [SD]=2.5), nearly one-third (31.8%) reported >5 EWE in the past 12-months, approximately half (51.9%) reported 2-4, and 16.2% reported one. LPA identified three distinct resource insecurity profiles: Class 1 (27.7%) reported water security, mild food insecurity, and low sanitation insecurity; Class 2 (59.1%) reported water insecurity, severe food insecurity, and moderate sanitation insecurity; and Class 3 (13.3%) reported water insecurity, severe food insecurity, and high sanitation insecurity. Increased EWE exposure was associated with higher resource insecurity profile membership. Participants experiencing 2-4 past-year EWE (adjusted relative risk ratio (ARRR): 1.71, 95% Confidence Interval [95%CI]: 1.03-2.83, p=0.038) and >5 EWE (ARRR: 2.24, 95%CI: 1.10-4.53, p=0.025), compared to 1 EWE, had increased odds of Class 2 membership vs. Class 1. Participants reporting >5 past-year EWE (vs. 1) had significantly higher odds of Class 3 (vs. Class 1) membership (ARRR: 6.55, 95%CI: 1.78-24.05, p=0.005). Compared to Class 1, Class 3 participants reported significantly lower SRP (adjusted beta coefficient [aβ]= -3.16; 95%CI: -5.85, -0.47; p=0.021), increased transactional sex (adjusted odds ratio [aOR]=4.72; 95%CI: 2.45, 9.07; p<0.001), increased IPV (aOR=3.02; 95% CI: 1.22, 7.48; p=0.017), and higher A-SRH stigma (aβ=1.88; 95% CI: 0.45, 3.30; p=0.010). Class 2 participants also reported increased transactional sex odds (AOR=1.77, 95%CI: 1.19, 2.64, p=0.005).

Conclusions and Implications: We identified three distinct profiles that reflect lower, moderate, and high resource insecurity across food, water and sanitation access among AGYW in Kenya. Increased EWE exposure was associated with profile membership with higher resource insecurities, and higher resource insecurity profiles were associated with a range of relational-level HIV vulnerabilities. Person-centred approaches such as LPA provide meaningful information to develop climate-informed HIV prevention approaches for AGYW in Kenya.