This study used a panel design to examine the relative impact of characteristics of people with psychiatric disabilities and their service use on hospitalization. Secondary data, including 4 years of community mental health service records and 5 years of state hospitalization records maintained by the county authority of a major midwestern metropolitan area, were employed in the analyses. The inclusion of multiple waves of data allowed for both cross-sectional and longitudinal analyses. Results indicated that previous hospitalization was the most powerful predictor of future hospitalization, followed closely by community service utilization. Persons who were low utilizers of community services in the previous year and high utilizers in the current year were significantly more likely to be hospitalized and to have a greater number of hospitalizations in that year. These variables were significant predictors of hospitalization in each of the pairs of years that were studied. Implications of these findings for estimating risk for hospitalization and for future research are discussed.