OBJECTIVE: To derive demographic weightings to replace the existing system of prescribing units used in analysing prescribing by general practitioners in England. DESIGN: The prescribing data for one year from a sample of 90 practices in 80 family health service authority areas were used to calculate the relative frequency with which items were prescribed, for each sex, in nine age bands and for temporary residents. Data on the variation in cost per item by age and sex then allowed estimates to be made of the relative costs for these groups. Integer values for both the item based and cost based weightings were obtained by conversion to optimal integer scales. MAIN OUTCOME MEASURES: Item based and cost based weightings for each of the 18 age-sex groups and for temporary residents. The cost based weightings were considered more appropriate to the context in which the new system was to be used. RESULTS: Prescribing costs increased noticeably, for both sexes, in the middle years (ages 35-64). Compared with the existing system, the cost based weightings (ASTRO-PUs) gave greater weight to patients aged 45 and over, especially those in the 55-64 age band, at the expense of younger patients. Children under 5 received twice as many items as those aged 5-14, but the inexpensiveness of their drugs made the cost based weightings of the two groups equal. Similarly, women were generally given more items than men, but at a lower average cost per item, which reduced differences between the sexes in the cost based weightings. Costs for patients aged 75 and over, compared with those aged 65-74, were higher only for women. CONCLUSIONS: The cost based weightings proposed are believed to reflect the present distribution of prescribing costs, in relation to age and sex, in English general practice. They are intended for use in analyses at practice level.
OBJECTIVE: To derive demographic weightings to replace the existing system of prescribing units used in analysing prescribing by general practitioners in England. DESIGN: The prescribing data for one year from a sample of 90 practices in 80 family health service authority areas were used to calculate the relative frequency with which items were prescribed, for each sex, in nine age bands and for temporary residents. Data on the variation in cost per item by age and sex then allowed estimates to be made of the relative costs for these groups. Integer values for both the item based and cost based weightings were obtained by conversion to optimal integer scales. MAIN OUTCOME MEASURES: Item based and cost based weightings for each of the 18 age-sex groups and for temporary residents. The cost based weightings were considered more appropriate to the context in which the new system was to be used. RESULTS: Prescribing costs increased noticeably, for both sexes, in the middle years (ages 35-64). Compared with the existing system, the cost based weightings (ASTRO-PUs) gave greater weight to patients aged 45 and over, especially those in the 55-64 age band, at the expense of younger patients. Children under 5 received twice as many items as those aged 5-14, but the inexpensiveness of their drugs made the cost based weightings of the two groups equal. Similarly, women were generally given more items than men, but at a lower average cost per item, which reduced differences between the sexes in the cost based weightings. Costs for patients aged 75 and over, compared with those aged 65-74, were higher only for women. CONCLUSIONS: The cost based weightings proposed are believed to reflect the present distribution of prescribing costs, in relation to age and sex, in English general practice. They are intended for use in analyses at practice level.