Feature story

Consultation held on definition and measurement of concurrent sexual partnerships

24 April 2009

It has long been suggested that concurrent sexual partnerships are one component responsible for creating sexual contact networks conducive to the rapid spread of HIV. Recently the idea has gained prominence and momentum, with national HIV programmes beginning to plan prevention campaigns targeted at reducing concurrent partnerships. However, in both the scientific and programmatic communities, research and discussion about concurrent partnerships and their role in the spread of HIV have been impeded by lack of clarity about the definition and appropriate indicators of concurrency.

On 20-21 April 2009, the UNAIDS Reference Group on Estimates, Modelling, and Projections convened a meeting to discuss the definition and measurement of concurrent sexual partnerships in Nairobi, Kenya. The meeting brought together a group of 35 experts, representing those working on large-scale cross-sectional surveys, longitudinal cohort studies, researchers using sexual behaviour and concurrent partnership data, and monitoring and evaluation specialists.

The main aim of the meeting was to reach consensus on what should be measured in large-scale surveys to best capture the amount of concurrency in a population and to monitor the impact of programmes. Topics covered included a history and overview of concurrent partnerships and HIV, comparisons of the definitions, methods, and questions used to measure concurrency, the reliability of sexual behaviour data, the measurement of concurrency in large cross sectional surveys and in longitudinal cohort studies.

Overlapping sexual partnerships where sexual intercourse with one partner occurs between two acts of intercourse with another partner.

Agreed definition of "concurrent partnerships"

At the close of the meeting, meeting participants reached consensus that the definition of concurrent partnerships should be “overlapping sexual partnerships where sexual intercourse with one partner occurs between two acts of intercourse with another partner”, and recommended that the point prevalence of concurrent partnerships in the adult population, that is the proportion of adults aged 15-49 reporting more than one ongoing sexual partnership at an instant in time, be used as the main indicator of concurrent partnerships in a population. This measure best distinguishes between concurrency and rapid serial monogamy.

Additionally, the Reference Group discussed other useful indicators of concurrent partnerships in a population (including the cumulative prevalence of concurrent partnerships in the adult population over a 1 year period), and made recommendations for further research into the methods of measuring concurrent partnerships and sexual behaviour, the relationship between concurrency and HIV transmission, and social norms around concurrent partnerships.

A set of specific recommendations on how to capture the proposed indicator will be provided to the UNAIDS Monitoring and Evaluation Reference Group and to the implementing organizations of large household surveys by end of April, 2009. A full meeting report will be available by end of May, 2009.