Objectives: Gini coefficients have been used to describe the distribution of Chlamydia trachomatis (CT) infections among individuals with different levels of sexual activity. The objectives of this study were to investigate Gini coefficients for different sexually transmitted infections (STIs), and to determine how STI control interventions might affect the Gini coefficient over time.
Methods: We used population-based data for sexually experienced women from two British National Surveys of Sexual Attitudes and Lifestyles (Natsal-2: 1999-2001; Natsal-3: 2010-2012) to calculate Gini coefficients for CT, Mycoplasma genitalium (MG), and human papillomavirus (HPV) types 6, 11, 16 and 18. We applied bootstrap methods to assess uncertainty and to compare Gini coefficients for different STIs. We then used a mathematical model of STI transmission to study how control interventions affect Gini coefficients.
Results: Gini coefficients for CT and MG were 0.33 (95% CI [0.18-0.49]) and 0.16 (95% CI [0.02-0.36]), respectively. The relatively small coefficient for MG suggests a longer infectious duration compared with CT. The coefficients for HPV types 6, 11, 16 and 18 ranged from 0.15 to 0.38. During the decade between Natsal-2 and Natsal-3, the Gini coefficient for CT did not change. The transmission model shows that higher STI treatment rates are expected to reduce prevalence and increase the Gini coefficient of STIs. In contrast, increased condom use reduces STI prevalence but does not affect the Gini coefficient.
Conclusions: Gini coefficients for STIs can help us to understand the distribution of STIs in the population, according to level of sexual activity, and could be used to inform STI prevention and treatment strategies.
Keywords: Chlamydia trachomatis; Gini coefficient; HPV; Lorenz curve; Mathematical model; Mycoplasma genitalium; Sexual behavior; Transmission model.
© 2020 Gsteiger et al.