Risk of new HIV diagnosis by intersecting migration, socioeconomic, and mental health vulnerabilities in the Netherlands: a nationwide analysis of the ATHENA cohort and Statistics Netherlands registry data

Abstract

Background: To further reduce new HIV diagnoses in the Netherlands, individual and structural barriers hindering prevention must be addressed. We aimed to estimate the disproportional burden of new HIV diagnoses and explore how intersecting socio-demographic, socio-economic, and health-related factors jointly influence the risk of a new HIV diagnosis.

Methods: We combined data from the ATHENA cohort, an ongoing nationwide HIV cohort, with registry data from Statistics Netherlands. We selected individuals with a new HIV diagnosis between 1 January 2012 and 31 December 2023 and matched them to individuals from the general population. We assessed determinants of a new HIV diagnosis using a multivariable generalized linear model. We used Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) to quantify the joint and individual contribution of intersecting variables.

Findings: 6055 men and 1020 women were newly diagnosed with HIV. Having a migration background and a low to middle income or income below the poverty line was associated with a higher risk of a new HIV diagnosis for both men (low to middle: adjusted odd ratio (aOR) = 1.24, 95% confidence interval (CI) = 1.17-1.31; below the poverty line: aOR = 1.75, 95% CI = 1.62-1.89) and women (low to middle: aOR = 2.49, 95% CI = 2.05-3.01; below the poverty line: aOR = 4.71, 95% CI = 3.80-5.83). Use of mental health care (aOR = 1.14, 95% CI = 1.01-1.27) or antidepressants (aOR = 1.66, 95% CI = 1.50-1.84) also increased the risk among men; while receiving social welfare (aOR = 1.39, 95% CI = 1.15-1.67) and use of antipsychotic medication (aOR = 1.66, 95% CI = 1.21-2.28) increased the risk among women. Of all intersections identified in MAIHDA, men with a first-generation migration background, income below the poverty line, and who used antidepressants had the highest predicted probability of an HIV diagnosis (0.036%, 95% confidence interval (CI) = 0.025-0.052). Women with a first-generation background, income below the poverty line, who received social welfare, and who used antipsychotic medication had the highest predicted risk (0.019%, 95% CI = 0.011-0.035).

Interpretation: A disproportionally higher burden of a new HIV diagnosis was observed for individuals with a migration background and economic and mental health vulnerabilities. HIV prevention and testing need to be reinforced in these groups.

Funding: Dutch Ministry of Health, Welfare and Sport; TKI Health Holland.

Keywords: Demography; HIV; Health inequalities; Healthcare disparities; Socioeconomic factors.

Fig. 1

Predicted probabilities of a new HIV diagnosis for each stratum among men. A) Predicted probabilities, B) Characteristics associated with each stratum. In B) Color coding indicates categories for each variable. Age: dark blue = ≥50 years, medium blue = 25–49 years, light blue <25 years. Migration background: light red = no migration background, medium red = first generation migration background, dark red = second generation migration background. Income: dark green = below poverty line, medium green = low to middle income, light green = high income. For the other variables (mental health care, antidepressant use, antipsychotic use): darker shades = used, lighter shades = not used.

Fig. 2

Predicted probabilities of a new HIV diagnosis for each stratum among women. A) Predicted probabilities, B) Characteristics associated with each stratum. In B) Color coding indicates categories for each variable. Age: dark blue = ≥50 years, medium blue = 25–49 years, light blue <25 years. Migration background: light red = no migration background, medium red = first generation migration background, dark red = second generation migration background. Income: dark green = below poverty line, medium green = low to middle income, light green = high income. For the other variables (received social welfare, antidepressant use, antipsychotic use): darker shades = received/used, lighter shades = not received/used.