Gender-Based Violence, Physiological Stress, and Inflammation: A Cross-Sectional Study

Abstract

Background: Female sex workers (FSWs) are at high risk for gender-based violence (GBV) and HIV infection. This study aimed to identify associations between GBV exposure in the past 12 months and biomarkers of physiologic stress and inflammation that may play a role in increased HIV risk among Kenyan FSWs.

Materials and methods: Participating women responded to a detailed questionnaire on GBV and mental health. Plasma was collected for assessment of systemic C-reactive protein (CRP) and interleukin-6 (IL-6) levels. Hair proximal to the scalp was collected to measure cortisol concentration. CRP and IL-6 were measured by enzyme-linked immunosorbent assay, and hair cortisol concentration was determined by enzyme immunoassay. Log-transformed biomarker values were compared across GBV exposure categories using Kruskal-Wallis or Wilcoxon rank sum tests. Multivariable linear regression was used to explore associations between recent GBV and hair cortisol concentration.

Results: Two hundred eighty-three women enrolled, of whom 112 (39.6%) reported physical, sexual, or emotional violence in the past 12 months, 134 (47.3%) reported more remote exposure, and 37 (13.1%) reported no exposure. CRP and IL-6 levels did not differ across groups (p = 0.57 and p = 0.62, respectively). Among 141 women who provided hair, cortisol concentrations were higher among recently exposed women compared to the other two groups combined (p = 0.02). In multivariable regression, recently exposed women had higher hair cortisol levels than remotely exposed or unexposed women (adjusted beta = 0.52, 95% confidence interval 0.02-1.02, p = 0.04).

Conclusions: While CRP and IL-6 levels did not differ by GBV category, recent GBV was associated with increased hair cortisol concentration. GBV-related increases in cortisol could affect health outcomes and merit study in relation to HIV acquisition risk.

Keywords: abuse; cortisol; inflammation; physiological stress; sex work; violence.

FIG. 1.

Log-transformed C-reactive protein (A), interleukin 6 (B), and hair cortisol concentration (C) by exposure group. Individual results are shown by each dot, and median values for each group (i.e., women exposed to GBV in the past 12 months, women exposed to GBV over 12 month ago, and women not reporting GBV exposure) are indicated with a black line. Kruskal–Wallis p-values are presented for across-group comparisons (i.e., overall p-values), and Wilcoxon rank-sum p-values are presented for pairwise comparisons. GBV, gender-based violence.

FIG. 2.

Scatter plots of days since any physical, sexual, or emotional violence and log-transformed hair cortisol concentration. (A) Presents data for women exposed within 3.8 months of the study visit. (B) Presents data for women exposed more than 3.8 months but within 12 months of the study visit. (C) Presents data for women exposed more than 12 months before the study visit. Measured values, a fitted line, and 95% confidence intervals are presented for each panel, as are Spearman correlation coefficients and p-values.