Anderegg Nanina, Slabbert Mariette, Buthelezi Kholi, Johnson Leigh F
Centre for Infectious Disease Epidemiology and Research, School of Public Health, University of Cape Town, South Africa.
Institute of Social and Preventive Medicine, University of Bern, Switzerland.
Infect Dis Model. 2024 Jan 23;9(1):263-277. doi: 10.1016/j.idm.2024.01.006. eCollection 2024 Mar.
In sub-Saharan Africa, accurate estimates of the HIV epidemic in female sex workers are crucial for effective prevention and care strategies. These estimates are typically derived from mathematical models that assume certain demographic and behavioural characteristics like age and duration of sex work to remain constant over time. We reviewed this assumption for female sex workers in South Africa.
We reviewed studies that reported estimates on either the age or the duration of sex work among female sex workers in South Africa. We used Bayesian hierarchical models to synthesize reported estimates and to study time trends. In a simulation exercise, we also investigated the potential impact of the "constant age and sex work duration"-assumption on estimates of HIV incidence.
We included 24 different studies, conducted between 1996 and 2019, contributing 42 estimates on female sex worker age and 27 estimates on sex work duration. There was evidence suggesting an increase in both the duration of sex work and the age of female sex workers over time. According to the fitted models, over each decade the expected duration of sex work increased by 55.6% (95%-credible interval [CrI]: 23.5%-93.9%) and the expected age of female sex workers increased by 14.3% (95%-CrI: 9.1%-19.1%). Over the 23-year period, the predicted mean duration of sex work increased from 2.7 years in 1996 to 7.4 years in 2019, while the predicted mean age increased from 26.4 years to 32.3 years. Allowing for these time trends in the simulation exercise resulted in a notable decline in estimated HIV incidence rate among sex workers over time. This decline was significantly more pronounced than when assuming a constant age and duration of sex work.
In South Africa, age and duration of sex work in female sex workers increased over time. While this trend might be influenced by factors like expanding community mobilization and improved rights advocacy, the ongoing criminalisation, stigmatisation of sex work and lack of alternative employment opportunities could also be contributing. It is important to account for these changes when estimating HIV indicators in female sex workers.
在撒哈拉以南非洲地区，准确估计女性性工作者中的艾滋病流行情况对于有效的预防和护理策略至关重要。这些估计通常来自数学模型，这些模型假设某些人口统计学和行为特征，如年龄和性工作时长，随时间保持不变。我们对南非女性性工作者的这一假设进行了审查。
我们审查了报告南非女性性工作者年龄或性工作时长估计值的研究。我们使用贝叶斯分层模型来综合报告的估计值并研究时间趋势。在一次模拟练习中，我们还调查了“年龄和性工作时长恒定”这一假设对艾滋病发病率估计值的潜在影响。
我们纳入了1996年至2019年间开展的24项不同研究，这些研究提供了42项关于女性性工作者年龄的估计值和27项关于性工作时长的估计值。有证据表明，随着时间推移，女性性工作者的性工作时长和年龄都有所增加。根据拟合模型，每过十年，性工作的预期时长增加55.6%（95%可信区间[CrI]：23.5% - 93.9%），女性性工作者的预期年龄增加14.3%（95% CrI：9.1% - 19.1%）。在这23年期间，性工作的预测平均时长从1996年的2.7年增加到2019年的7.4年，而预测平均年龄从26.4岁增加到32.3岁。在模拟练习中考虑这些时间趋势导致性工作者中估计的艾滋病发病率随时间显著下降。这种下降比假设性工作年龄和时长恒定的情况明显更显著。
在南非，女性性工作者的年龄和性工作时长随时间增加。虽然这一趋势可能受到社区动员扩大和权利倡导改善等因素的影响，但性工作持续被定罪、污名化以及缺乏替代就业机会也可能起到了作用。在估计女性性工作者的艾滋病指标时考虑这些变化很重要。
