Spread of HIV continues in rural areas of Tanzania

Samples taken over the past ten years indicate that HIV prevalence in rural northwest Tanzania increased steadily from 6.0% in 1994/1995 to a little over 8% in 2000/2001, levelling out thereafter. Incidence rose sharply from 0.8% to 1.2% per year over the same periods, remaining at 1.1% per year as of 2000/2003. Most recently, HIV incidence has been declining in Tanzanian roadside rural areas, especially among women, but has continued to rise slightly in more remote rural areas. The figures were reported in a paper published in the Journal of Acquired Immune Deficiency Syndromes in December 2007.

An ongoing, longitudinal open cohort study has tracked the epidemiology of HIV infection in rural northwest Tanzania since 1994. The study area is located outside of Mwanza City (Tanzania's second-largest city) and straddles the main road to Kenya. The population of the surrounding area, Kisesa ward, has grown rapidly from 19,530 in 1994 to 26,330 in 2004.

HIV epidemiologic data for this area were collected in four separate serologic surveys, conducted in 1994/95, 1996/97, 1999/2000, and 2003/04. Residents older than 15 years were eligible. The first two surveys had upper age limits of 44 and 46 years respectively; there was no upper age limit for the last two surveys. A total of 28,591 interviews and 28,523 HIV tests were conducted during the four surveys; 6448 adults were tested on two or more occasions.

For this analysis, the area was stratified into remote rural villages (which accounted for 57% of the total population in 1994 and 53% in 2004), and roadside villages plus the central trading centre. Throughout the study, attendance rates were higher in the remote rural villages (70% to 86%) than in the roadside villages (61% to 81%), and higher for women than for men.

In Kisesa ward as a whole, HIV prevalence among adults 15 years of age or older increased from 6.0% to 8.3% between 1994 and 2000, remaining at 8.2% in 2003/2004.

Prevalence has declined slightly for women, from 9/3 in 1999/2000 to 8.8% in 2003/2004. Over the same time intervals, prevalence in men rose from 6.9% to 7.5%.

However, trends in prevalence vary with proximity to the central road. In the trading centre and roadside villages, prevalence levelled off at 8.5% for men and fell sharply for women, from 13.5% in 1999/2000 to 10.6% in 2003/2004. Over the same time in remote rural areas, prevalence remained lower but continued to rise for both genders, from 6.5% to 7.6% for women and from 6.0% to 6.8% in men.

Changes in prevalence are affected by rates of new infection (incidence), mortality, and migration in and out of the region. In Kisesa ward, rates of migration are far larger than incidence and mortality rates; however, in- and out-migration of HIV-positive individuals have roughly balanced each other.

Incidence estimates are based on 33,140 person-years of observation of people between 15 and 45 years of ages, with 362 observed seroconversions. The most recent incidence estimates represent the period between 2000 and 2003. During that time, the crude annual HIV incidence was 1.1% overall. In central and roadside areas, annual incidence fell from 1.9% between 1997 and 2000 to 1.3% between 2000 and 2003; the decrease was sharpest among women (1.9% to 1.0%). However, in remote rural areas, annual incidence continued to rise, from 1.0% between 1997 and 2000, to 1.1% between 2000 and 2003. The rise in remote rural incidence was almost entirely among women (1.0% to 1.2%), while the annual rate among men remained constant at 1.0%.

The researchers conclude that HIV prevalence remains consistently higher in central and roadside areas than in remote rural areas, but that the gap has narrowed over time.

Their data "show that the gap in incidence level … between rural and roadside communities is narrowing, mainly because women in rural areas have recently experienced higher infection rates." The falling incidence in roadside areas "is an encouraging sign, but the continued gradual rise … in rural areas in worrying, especially because most (66%) of the Kisesa population lives in these areas."

They report that there is "an urgent need to promote and expand access to the existing HIV prevention efforts" and to ensure that ART roll-out reaches rural areas. Forthcoming analysis and publication of a fifth serosurvey, data collection for which concluded in July 2007, will extend these results and judge whether ART access has impacted incidence trends.