Data were obtained as part of a larger longitudinal study of inner-city African-American families with a mother infected with HIV. The current study focused on that portion of the women in the study for whom there was longitudinal data and who reported being in an intimate relationship at the first assessment (the only assessment at which a measure of partner abuse was involved). Data were available for 53 HIV-infected women and 84 noninfected women. No significant differences were found between the two groups on education, employment status, or income; however, the noninfected comparison group was slightly older than the infected group. Each assessment included two interviews, a sociodemographic interview, and a psychosocial interview. Verbal and physical partner abuse were measured with the Conflict Tactics Scale. The Hamilton Rating Scale for Depression was used to measure depressive symptoms. The General Severity Index of the Brief Symptom Inventory was used as a measure of general psychological distress, and the Physical Symptoms Inventory was used to assess the degree to which women were bothered by specific somatic symptoms. Mother-child relationship quality was assessed with the mother-report on the short-form of the Interaction Behavior Questionnaire. As predicted, the women infected with HIV were more likely to acknowledge being both physically and verbally abused during the past year than were demographically similar noninfected women. Also, women, regardless of their HIV status, who reported more physical and/or verbal abuse experienced both acute and chronic effects of abuse on their psychosocial functioning. Further, abuse was more strongly associated with concurrent suicidal ideation for HIV-infected women than for noninfected women. Generally, findings indicate that abuse predicts a poorer prognosis for African-American women regardless of their HIV status. 3 tables and 48 references