# Longitudinal assessment of changes in hair cortisol levels and associations with violence, poor mental health and harmful substance use among female sex workers in Nairobi, Kenya

## TL;DR

This study explores how hair cortisol levels in female sex workers in Nairobi, Kenya, change over time and how these levels relate to violence, mental health issues, and substance use.

## Contribution

The study provides new evidence linking physical and sexual violence to changes in hair cortisol levels, suggesting a potential pathway to poor health outcomes.

## Key findings

- Hair cortisol concentrations decreased significantly from baseline to endline.
- Physical and/or sexual violence showed significant associations with changes in hair cortisol levels.
- Mental health problems and harmful substance use prevalence decreased over time.

## Abstract

Female sex workers (FSWs) in sub-Saharan Africa commonly experience violence, mental health problems, and harmful substance use. Stressful life events can harm the functioning of the hypothalamic-pituitary-adrenal (HPA) axis, serving as a pathway to increased poor health, including HIV susceptibility through cortisol levels. In this paper, we examine changes in hair cortisol concentration (HCC) levels and associations with experiences of violence, mental health problems and harmful substance use among FSWs in Nairobi, Kenya. We used baseline and endline data from the Maisha Fiti study of FSWs in Nairobi. Participants reported recent violence, poor mental health, and harmful alcohol/substance use at both time points. Hair samples proximal to the scalp were collected to measure HCC levels determined by ELISA technique. We analysed data from 285 HIV-negative respondents who provided a 2 cm hair sample at baseline and endline. Multivariable linear regression models were used to assess the associations between the trajectory of the main exposure variables and the change in HCC levels at endline. Findings showed that HCC levels decreased significantly (p-value = 0.001) from baseline (mean HCC = 316 ng/g) to endline (mean HCC = 238.1 ng/g). Reported prevalence of violence, mental health problems and harmful alcohol/other substances decreased. There was evidence of associations between change in HCC at endline and the trajectories of physical violence (p-value = 0.007) and physical and/or sexual violence (p-value = 0.048). There was weak evidence of an association between the trajectory of exposure to emotional violence but no evidence of other associations. These findings suggest that physical violence and physical and/or sexual violence may lead to HPA axis dysfunction, possibly serving as a pathway linking violence to increased poor health, including HIV acquisition. However, further research with repeated measurements and a larger sample size is needed to examine the associations between violence, HCC levels, and HIV infection.