A model for reconstructing trends and distribution in age at first sex from multiple household surveys with reporting biases

Abstract

AbstractAge at first sex (AFS) is a key indicator for monitoring sexual behaviour risk for HIV and sexually transmitted diseases. Reporting of AFS data, however, suffer social-desirability and recall biases which obscure AFS trends and inferences constructed from it. We illustrated examples of the biases using data from multiple nationally-representative Demographic and Health Surveys household surveys conducted between 1992 and 2019 in Ethiopia (4 surveys), Guinea (4 surveys), Senegal (8 surveys), and Zambia (8 surveys). Based on this, we proposed a time-to-event, interval censored model for the AFS that uses overlapping reports by the same birth cohort in successive surveys to adjust reporting biases. The three-parameter log-skew-logistic distribution described the asymmetric and nonmonotonic hazard exhibited by empirical AFS data. In cross-validation analysis, incorporating a term for AFS reporting bias as a function of age improved model predictions for the trend of AFS over birth cohorts. The interquartile range for the AFS was 16 years to 23 years for Ethiopian and Senegalese women and 15 years to 20 years for Guinean and Zambian men. Median AFS increased by around one to 1.5 years between the 1960 and 1989 birth cohorts for all four datasets. Younger male respondents tended to report a younger AFS while female respondents tended to report an older AFS than when asked in later surveys. Above age 30, both male and female respondents tended to report older AFS compared to when surveyed in their late twenties. Simulations validated that the model recovers the trend in AFS over birth cohorts in the presence of reporting biases. At least three surveys are needed to obtain reliable trend estimate for a 20-years trend. Mis-specified reference age at which reporting is assumed unbiased did not affect the trend estimate but resulted in biased estimates for the median AFS in the most recent birth cohorts.