The aim of this study was to determine the relationship between patient satisfaction and diabetes- related preventive health care and emergency room (ER) use. We studied 542 urban African-Americans with type 2 diabetes aged > or =25 years who were enrolled in a primary carebased intervention trial to improve diabetes control and reduce adverse health events; 73% female, mean age 58 years, 35% had yearly household incomes of <$7500, and all participants had health insurance. All completed a baseline interview-administered questionnaire. Patient satisfaction was measured using a modified version (nine questions) of the Consumer Assessment of Health Plans Survey (CAHPS) and use of diabetes-related preventive health care and ER were assessed by self-report. We then followed participants for 12 months to determine ER use prospectively. In general, participants gave favorable ratings of their care; over 70% reported that they had no problem getting care, over 60% reported the highest ratings on the communication and courtesy domains, and mean ratings (0-10 scale) for personal doctor and overall health care were high (8.8 and 8.4, respectively). Using poisson regression models adjusted for age, education, and self-reported rating of health, several aspects of patient satisfaction were associated with subsequent ER use. Participants who reported that medical staff were usually helpful or that doctors and nurses usually spent enough time were 0.49 and 0.37 times, respectively, less likely to use the ER (all p < 0.05). However, few aspects of patient satisfaction were associated with better preventive services. These data suggest that greater patient satisfaction was associated with lower ER use in urban African-Americans. Whether measures to improve patient satisfaction would reduce ER use requires further prospective study.