OBJECTIVE: To find an equation that can calculate quality reference standards on the sum expected per doctor, adjusted for the size of his/her list, his/her characteristics and patients included in chronically ill services. This objective is posed within a strategy of individualizing information and evaluation as a tool to promote improvement. DESIGN: Crossover, descriptive study. SETTING: Autonomous community of the Canary Islands. Tenerife Health Area, primary care management. PARTICIPANTS: The study was performed with 51 doctors belonging to 10 primary care teams, for whom there was quality information available due to their having used the CRONOS computer programme for a sufficient period of time. The period analyzed ran from January to November 1997. MEASUREMENTS AND MAIN RESULTS: Multiple regression analysis through automatic algorithm of step-by-step selection of variables. There was a close relationship between the variability of the pharmacy sum per doctor (adjusted r2: 0.81; F, 66.05; p < 0.000) and the size of the list; also between the former and the characteristics of the users on the list. The most relevant variables were the percentage of women over 55 and the mean drug consumption per consultation with patients during the study period. It was seen, in relation to this last variable, that the percentage of pensioners and the number of hyper-intense patients included were important. CONCLUSION: Given the high variability explained, we propose broadening the study to all the doctors, so as to give the results greater consistency and offer quality reference information, in line with the above-mentioned improvement plan.
OBJECTIVE: To find an equation that can calculate quality reference standards on the sum expected per doctor, adjusted for the size of his/her list, his/her characteristics and patients included in chronically ill services. This objective is posed within a strategy of individualizing information and evaluation as a tool to promote improvement. DESIGN: Crossover, descriptive study. SETTING: Autonomous community of the Canary Islands. Tenerife Health Area, primary care management. PARTICIPANTS: The study was performed with 51 doctors belonging to 10 primary care teams, for whom there was quality information available due to their having used the CRONOS computer programme for a sufficient period of time. The period analyzed ran from January to November 1997. MEASUREMENTS AND MAIN RESULTS: Multiple regression analysis through automatic algorithm of step-by-step selection of variables. There was a close relationship between the variability of the pharmacy sum per doctor (adjusted r2: 0.81; F, 66.05; p < 0.000) and the size of the list; also between the former and the characteristics of the users on the list. The most relevant variables were the percentage of women over 55 and the mean drug consumption per consultation with patients during the study period. It was seen, in relation to this last variable, that the percentage of pensioners and the number of hyper-intense patients included were important. CONCLUSION: Given the high variability explained, we propose broadening the study to all the doctors, so as to give the results greater consistency and offer quality reference information, in line with the above-mentioned improvement plan.