Variation in proportion of frequent attenders between Danish general practices

Aim: The authors examined the inter-practice variation in the proportion of adult frequent attenders and whether practice factors may explain some of this variation. Methods: A population-based cross-sectional study was performed on the basis of registers including 262 active general practices and their 419,072 registered adults aged 20 and over in the County of Aarhus, Denmark (630,000 inhabitants) from November 1997 to October 1998. The number of face-to-face daytime contacts with general practitioners was counted for each individual on the basis of data drawn from the files of the National Health Service. Frequent attenders were defined as the 10% most frequent attenders in the county over 12 months for each sex, and four age groups. Both the crude and standardized proportions of frequent attenders in each practice were calculated. Associations between practice factors and the frequent attender proportions were assessed from the squared partial correlation coefficients. Results: The proportion of frequent attenders ranged from 1.9 to 26.2% (factor 13.8) (standardized: 1.6 to 27.8% (factor 17.4)) with 10th to 90th percentiles of 4.4 to 12.9% (factor 2.9) (standardized: 4.3 to 13.2% (factor 3.1)). A low standardized proportion of frequent attenders was statistically significantly associated with a high degree of urbanization, a large number of registered individuals per GP and female GP (multiple R² 0.16 for all practices and 0.22 for solo practices). Conclusions: Variation in the proportion of frequent attenders was considerable between practices and about one-fifth of this variation could be explained by practice factors.