OBJECTIVES: To determine the prevalence of polypharmacy and the explanation power of Adjusted Clinical Groups (ACG) classification as a measure of morbidity burden in a Spanish population setting. METHODS: Cross-sectional study based on the clinical records from all the attended patients in five primary care centres along the year 2006. Main variables: age, sex, case-mix/episodes, pharmacy costs, therapeutic group, polypharmacy (regular utilization of more than 5 drugs at least 240 days/year), explained variance (predictive analysis and multiple linear regression). The ACG (Johns-Hopkins Case-mix System) classify individuals with similar needs for health care resources based on overall expenditures. For analysis purposes, the ACG are collapsed to one of eight mutually exclusive morbidity classes, known as resource utilization bands (RUB) (collapsed: Low-Comorbidity, Medium-Comorbidity and High-Comorbidity). Logistic regression and ANCOVA (Bonferroni adjustment) analysis were made to adjust the models. The statistical package SPSS was used (p<0.05). RESULTS: Included patients: 80,775. Average number of episodes: 4.8±3.5; mean age: 40.7±22.9 years, males: 46.9%, intensity of use: 72.4%, pharmacy costs: €22.7 millions (55.6% of total costs) and mean cost: €281.05±627.85. Polypharmacy prevalence was 12.2% (CI: 12.0-12.4%), increasing with age (50,1%; in >85 years), female sex and morbidity burden (Low-Comorbidity [16.5%]: 1.1%, Medium-Comorbidity [78.3%]: 11.8% and High-Comorbidity [5.2%]: 40.3%), p<0.0001. A strong association was observed with alcoholism, asthma, depression, and digestive and cardiovascular therapeutic groups (OR>4.3, p<0.0001). Predictive and regression analysis: ROC curve (pharmacy costs): 0.968; sensibility: 96.1%, specificity: 61.7%, R2 (Nagelkerke): 62.2%; explained variance: R2=26.8%; age, sex, comorbidity model: R2=42.2 (p<0.001). Individual cost for patients with High-Comorbidity was €1,436.53 and after adjustment was €1,047.59 (CI: 1,002.87-1,092.31). CONCLUSION: The observed polypharmacy profile was similar to others described in the literature. ACG classification provides a fair good explanation of the pharmacy costs variability after adjustment by age, sex and comorbidity (42.2%), including a significative share of polypharmacy.