OBJECTIVES: To propose a case-mix methodology for primary care, based on chronicity, type and age. To describe the explanatory value of these variables in the variability of the medical case-load. DESIGN: Observation, descriptive and retrospective study. SETTING: Primary care. Rochapea Health Centre, Pamplona. MATERIAL AND METHODS: Computer records of all the consultations between January 1996 and June 1997. Dependent variable: case-load. INDEPENDENT VARIABLES: age, type, chronic pathologies (diabetes, lipaemia, chronic neurological diseases, COPD-asthma, chronic psychiatric illnesses, cardiopathy, hypertension, alcohol and other drug abuse). The Kruskal-Wallis test was used to compare work-loads by age groups; and multiple linear regression analysis to calculate the predictive power of the independent variables. RESULTS: Significant differences were observed for age groups. In the multivariate model used for general practitioners, all the variables could be included. They explained 24.2% of the variability in work load (R2). For paediatricians, age and asthma, explaining 23.48%, could also be included. CONCLUSIONS: Age, type and chronicity are useful variables for predicting case load from administrative data bases. They can be used in adjustments for case load applicable to capitation payment systems.
OBJECTIVES: To propose a case-mix methodology for primary care, based on chronicity, type and age. To describe the explanatory value of these variables in the variability of the medical case-load. DESIGN: Observation, descriptive and retrospective study. SETTING: Primary care. Rochapea Health Centre, Pamplona. MATERIAL AND METHODS: Computer records of all the consultations between January 1996 and June 1997. Dependent variable: case-load. INDEPENDENT VARIABLES: age, type, chronic pathologies (diabetes, lipaemia, chronic neurological diseases, COPD-asthma, chronic psychiatric illnesses, cardiopathy, hypertension, alcohol and other drug abuse). The Kruskal-Wallis test was used to compare work-loads by age groups; and multiple linear regression analysis to calculate the predictive power of the independent variables. RESULTS: Significant differences were observed for age groups. In the multivariate model used for general practitioners, all the variables could be included. They explained 24.2% of the variability in work load (R2). For paediatricians, age and asthma, explaining 23.48%, could also be included. CONCLUSIONS: Age, type and chronicity are useful variables for predicting case load from administrative data bases. They can be used in adjustments for case load applicable to capitation payment systems.