Abstract. To determine the prevalence and risk factors of anemia among human immunodeficiency virus (HIV)-infected women in Rwanda and the influence of highly active antiretroviral therapy (HAART) on anemia, we analyzed 200 HIV-positive women and 50 HIV-negative women in a cross-sectional study. Clinical examinations and iron and vitamin B 12 assays were performed, and complete blood counts, serum folic acid levels, and CD4 cell count determined. The prevalence of anemia was significantly higher among HIV-positive women (29%) than among HIV-negative women (8%) ( P < 0.001). Risk factors for anemia were lower body mass index (odds ratio [OR] = 3.4, 95% confidence interval [CI] = 2.4-4.1), zidovudine use (OR = 1.14, 95% CI = 1.01-1.29), lack of HAART (OR = 1.44, 95% CI = 1.21-1.67), oral candidiasis (OR = 1.4, 95% CI = 1.2-1.6), pulmonary tuberculosis (OR = 1.8, 95% CI = 1.7-2.2), cryptococcal meningitis (OR = 1.6, 95% CI = 1.21-1.8), Pneumocystis jiroveci pneumonia (OR = 1.41, 95% CI = 1.20-1.65) and CD4 lymphocyte count < 200 cells/µL (OR = 2.41, 95% CI = 2.01-3.07). The mean ± SD hemoglobin level of 10.9 ± 1.6 g/dL at HAART initiation significantly increased to 12.3 ± 1.5 g/dL in 8 months ( P < 0.001). Anemia increases with HIV stage, and HAART is associated with a significant improvement in hemoglobin levels.*Address correspondence to Jean Bosco Gahutu, Faculty of Medicine, National University of Rwanda, PO Box 30, Huye, Rwanda. E-mail: jgahutu@nur.ac.rw 457 ANEMIA IN HIV-INFECTED AND UNINFECTED WOMEN, RWANDA level < 11.5 mg/dL by using reference ranges based upon local control measurements in our laboratory, which is located at a moderate altitude of 1,768 meters. This limit was also shown by Beutler and Waalen. 20 Serum iron, vitamin B 12 , and folic acid levels were measured in the clinical laboratory of Ghent University Hospital. However, erythropoietin assays and the screening for malaria and chronic helminthes were not systematically conducted.The first-line HAART regimen at the time of the study consisted of stavudine or zidovudine, plus lamivudine, plus nevirapine or efavirenz; more patients used stavudine and nevirapine than zidovudine and efavirenz. For all the patients who initiated HAART at the time of entry into the study, they were all given a stavudine-based regimen. No patients were taking other antiretroviral medications such as tenofovir, abacavir, or protease inhibitors. Women receiving zidovudine for prevention of mother-to-child transmission of HIV were not included in the study population.Most patients were given cotrimoxazole prophylaxis with a dose of 1 tablet of 960 mg daily when CD4 cell counts were less than 350 cells/μL when possible. However, because many patients initially only sought care when they have reached advanced stages of immunosuppression, some patients were given prophylaxis at lower CD4 cell counts. No persons were offered ferrous sulfate and/or multivitamins for the treatment of anemia.Statistical analysis. All statistical analyses were performed with SPSS software ...