OBJECTIVES: To compare open heart surgery services provided by public and private hospitals in Catalonia (Spain) according to case mix, procedures undergone and surgical mortality. METHODS: Data on all adult patients undergoing open heart surgery procedures were collected prospectively in a sample of public and privately owned centres for a 6.5-month period in 1994. Sociodemographic, clinical and procedural variables were collected. A predictive model stratifying patients according to their surgical mortality risk was used to adjust for differences in case mix between providers. RESULTS: Included were 1287 open heart surgery procedures. Public and private patients differed significantly in terms of gender, clinical history (e.g. hypertension, pulmonary disease, recent infarction) and procedural variables (e.g. reoperation, type of intervention). There were also statistically significant differences related to educational level, with better educated patients more likely to be treated in private centres. Crude surgical mortality rates differed between providers, although public centres operated on higher-risk patients. After adjusting for differences in case mix, the association between the type of provider and surgical mortality was not statistically significant (odds ratio 1.68; 95% CI from 0.94 to 3.0). CONCLUSIONS: Although crude mortality rates differ between public and private providers, there is a significant trend towards higher surgical risk in public centres. After adjusting for surgical risk, differences between types of provider decreased and were no longer statistically significant. The importance of other social and health-related factors, such as educational level, may explain differences between providers in their patients' surgical risk and in their performance in open heart surgery.
OBJECTIVES: To compare open heart surgery services provided by public and private hospitals in Catalonia (Spain) according to case mix, procedures undergone and surgical mortality. METHODS: Data on all adult patients undergoing open heart surgery procedures were collected prospectively in a sample of public and privately owned centres for a 6.5-month period in 1994. Sociodemographic, clinical and procedural variables were collected. A predictive model stratifying patients according to their surgical mortality risk was used to adjust for differences in case mix between providers. RESULTS: Included were 1287 open heart surgery procedures. Public and private patients differed significantly in terms of gender, clinical history (e.g. hypertension, pulmonary disease, recent infarction) and procedural variables (e.g. reoperation, type of intervention). There were also statistically significant differences related to educational level, with better educated patients more likely to be treated in private centres. Crude surgical mortality rates differed between providers, although public centres operated on higher-risk patients. After adjusting for differences in case mix, the association between the type of provider and surgical mortality was not statistically significant (odds ratio 1.68; 95% CI from 0.94 to 3.0). CONCLUSIONS: Although crude mortality rates differ between public and private providers, there is a significant trend towards higher surgical risk in public centres. After adjusting for surgical risk, differences between types of provider decreased and were no longer statistically significant. The importance of other social and health-related factors, such as educational level, may explain differences between providers in their patients' surgical risk and in their performance in open heart surgery.