A New Scale for Measuring Dynamic Patterns of Sexual Partnership and Concurrency

Abstract
Quantitative assessment of multiple sexual partnerships and concurrency may help to elucidate the large observed differences in the prevalence of AIDS among population subgroups and countries. The goals of the study were (1) to develop a global scale of dynamic patterns of sexual partnerships, including concurrency with new partners and stable concurrency; (2) to apply this scale to three Caribbean regions characterized by different cumulative rates of incidence of AIDS; and (3) to compare the concurrency rates given by this scale with those of other published methods. We defined an individual scale based on 6 patterns of sexual behavior over the previous 12-month period, by using a simple algorithm to combine 7 variables. We then applied this scale to cross-sectional data collected from men living in three French Caribbean regions: Guadeloupe, Martinique, and Guyana. We found that all adults of all age classes in the three regions studied frequently had multiple (>2) and concurrent partnerships. The patterns of sexual behavior in the three regions were consistent with the respective cumulative incidence rates of AIDS, and a lower rate of concurrency with new partners and a higher rate of stable partnership concurrency were noted in Martinique, especially among 45- to 59-year-olds. The rate of concurrent partnerships was found to depend on the criteria used to define them and on the observation period (a given moment, or a defined period). Our definition gave a higher rate of concurrency than previously published indicators. The proposed scale can be applied to easy-to-collect data in cross-sectional population surveys and takes into account a wide variety of behaviors, including different types of concurrency.