Time trend analysis of sociodemographic characteristics and HIV risk factors of clients presenting for VCT in Moshi, Tanzania from 2003-2006

Mayhood M, Ostermann J., Crump J, Tribble A., Itemba D., Mgonja A., Mtalo A., Mtweve S., Bartlett J., Shao J., Thielman N. Conference abstract: IAS2007 • Abstract: A-042-0079-04627http://www.ias2007-abstracts.org
Publication year: 

Objectives: Continuing assessment of clients presenting to HIV Voluntary Counseling and Testing (VCT) is essential in order to target education and outreach. We investigated time-dependent trends in sociodemographic characteristics and HIV risk factors of clients presenting for VCT in Moshi,Tanzania. Methods: 8357 clients presenting consecutively for VCT between November2003 and December 2006 were administered a questionnaire on sociodemographic characteristics and HIV risk factors. Rapid HIV antibody testing was performed. Associations between individual demographic and risk characteristics and test date were assessed using logistic and linear regression models. Results for proportions are reported in average change in percentage points per year (pp/y; sample mean; p value). Results: The overall median (range) age was 30 (18-87) years and did not change over time. Regression models showed significant increase in females (2.2pp/y; sample mean 55.4%; p=0.001) and those citing multiple sexual partners (1pp/y; 16.9%; p=0.039), a sexual partner being ill or dead (0.8pp/y; 10.0%; p=0.031), and suspecting an unfaithful partner (9.8pp/y; 36.2%; p<0.001) as reasons for testing. There was a decrease in clients reporting symptoms of HIV (0.9pp/y; 11.5%; p=0.027) and the number of HIV symptoms (p=0.001). For risk behaviors, there was a decrease in clients with concurrent partners (7.8pp/y; 12.6%; p<0.001) and partners who tested positive (2.1pp/y; 8.0%; p<0.001); there was a small increase in average number of lifetime sexual partners (p=0.021). Overall, HIV seroprevalence declined (1.7pp/y; 14.8%; p<0.001) from 20.5% in the first quarter 2004 to 10.6% in the last quarter 2006.Conclusions: Among clients accessing VCT services between 2003 and 2006, we observed increases in females and clients citing partner-dependent exposure risks as reasons for testing, but noted fewer HIV-related symptoms at presentation and an overall decrease in seroprevalence. These trends suggest that educational programs are appropriately targeting at-risk populations, and that accessibility to VCT programs for these populations is improving.