Examining changes in sexual risk in Britain between 1990-2010: A Latent Class Analysis approach

Abstract

Abstract Background: Understanding how sexual risk changes over time is important for determining the likelihood of adverse sexual health outcomes. There are different risk profiles in the population and standard descriptive or regression methods are limited in their ability to capture this multidimensional concept. Latent Class Analysis (LCA) is a mixture modelling method that generates a categorical latent variable classifying a heterogeneous population into homogenous groups. We therefore use LCA to investigate sexual risk in Britain, and if and for whom, this has changed between 1990 and 2010. Methods: Data from three rounds of Britain’s decennial National Survey of Sexual Attitudes and Lifestyle (Natsal) were used, restricted to sexually-active participants (those reporting any sexual partners in the past year) aged 16-44-years (N1990=11,738; N2000=9,690; N2010=8,397). LCA was used with four variables: partner numbers (past year), number of partners without a condom (past year), age at first sex and self-perceived HIV risk, as variables common to all three Natsals. Covariates included age, ethnicity, educational attainment, same-sex attraction, and marital status. Multinomial regression analyses and Chi-Squared tests assessed change over time. Results: Partner number was the key driver of class membership, the resulting three classes we labelled: Long-Term Monogamy (41.3% of men; 60.9% of women), Risk-Conscious Monogamy (32.5%; 23.4%) and Sexual Risk Taking (26.2%; 15.7%) There was a statistically significant increase between 1990 and 2010 in the proportion of men (c2=739.49, p<0.01) and women (c2=1270.43, p<0.01) in the Sexual Risk-Taking class, increasing from 19.5% to 31.1% (men) and 9.9% to 22.1% (women). Marital status, specifically being single/not cohabiting, was strongly associated with Sexual Risk-Taking class membership at all time points. Conclusion: Our results indicate the viability of LCA models to assess change over time for complex behavioural phenomena, whilst supporting previous findings, namely an increase in sexual risk-taking in recent decades, partnership number driving class assignment, and significant sex differences in risk-taking. This methodological approach to assessing change over time can be used to extend previous LCA models (such as investigating the impact of COVID-19 on latent class composition) and to support empirical evidence of change over time to facilitate more nuanced public health policy.