Background Sexual partnerships take multiple forms among the Himba and Tjimba people living in Northwestern Namibia, including polygynous and monogamous marriage, and long- and short-term extra-marital partnerships. We hypothesised that HSV-2 prevalence would be higher in regions that are connected to local urban areas, because these should be sites of more frequent anonymous partner exchange.

Methods In 2009, we surveyed 30 villages, obtaining HSV-2 status (using the BioKit Rapid Test for HSV-2) and demographics, home village, recent travel patterns and sex partnerships (n=446). We estimated prevalence (a) for the entire population (b) for all women and men and c) for individual villages. Logistic regression models were fit to test the correlation between risk of infection and travel patterns.

Results Overall HSV-2 prevalence was 36.5%, with higher rates among women (50.5%) than men (23.3%). In contrast to expectation, the highest HSV-2 prevalences were found in some of the most remote regions. To explore this further, we built a network model that superimposes travel patterns onto sex partnership data. Using this model, we observed that older travel corridors (eg, travel from home village to extended-family associated villages) were more associated with higher HSV-2 transmission than newer corridors (eg, travel to local cities to hang out). Ethnographic interviews suggest that differences in the nature of relationships that occur in urban areas (eg, lower contact frequency, higher condom use) may explain the difference between observed and predicted.

Conclusions In this population cultural and economic transitions that lead to more travel into urban areas were not associated with increase in HSV2 prevalence, probably because these relationships are deemed more risky. Shifting attitudes about sex and sex partnerships are likely to affect future STD transmission in this population.