ESTIMATING THE SEXUAL MIXING PATTERNS IN THE GENERAL-POPULATION FROM THOSE IN PEOPLE ACQUIRING GONORRHEA INFECTION - THEORETICAL FOUNDATIONAND EMPIRICAL-FINDINGS

Study objectives - To describe mathematically the relationship between patterns of sexual mixing in the general population and those of peop le with gonorrhoea infection, and hence to estimate the sexual mixing matrix for the general population. Design - Integration of data descri bing sexual behaviour in the general population, with data describing sexual behaviour and mixing among individuals infected with gonorrhoea . Use of these data in a simple mathematical model of the transmission dynamics of gonorrhoea infection. Setting - The general population of London and a genitourinary medicine (GUM) clinic in west London. Part icipants - These comprised 1520 men and women living in London who wer e randomly selected for the national survey of sexual attitudes and Li festyles and 2414 heterosexual men and women who presented to the GUM clinic with gonorrhoea. Main results - The relationship between sexual mixing among people with gonorrhoea and sexual mixing in the general population is derived mathematically. An empirical estimate of the sex ual mixing matrix for the general population is presented. The results provide tentative evidence that individuals with high rates of acquis ition of sexual partners preferentially select other individuals with high rates as partners (assortative mixing) . Conclusions - Reliable e stimates of sexual mixing have been shown to be important for understa nding the evolution of the epidemics of HIV infection and other sexual ly transmitted diseases. The possibility of estimating patterns of sex ual mixing in the general population from information routinely collec ted in gonorrhoea contact tracing programmes is demonstrated. Furtherm ore, the approach we describe could, in principle, be used to estimate the same patterns of mixing, using contact tracing data for other sex ually transmitted diseases, thus providing a way of validating our res ults.