A study of drug utilisation in scotland

Singh M1, Hind C2, Williams D1. 1Department of Clinical Pharmacology, Aberdeen Royal Infirmary. 2Medicines Unit, NHS Grampian.

Introduction: Prescribing is not uniform with, variations existing between prescribers, practices, health authorities, and regions (Morton-Jones et al 1993). Previous studies have shown wide variation in the prescribing rate of a number of drugs frequently used in GP practices including antidepressants, non-steroidal anti-inflammatory drugs (NSAIDs), corticosteroids and antibiotics (McGavock 1992). Controlling the cost and improving the quality of prescribing is an issue of concern to many health authorities. Appropriate, drug therapy is often more cost effective than other kinds of treatment whilst inappropriate prescribing can cause adverse outcomes, deplete health care resources and compromise the quality of care. Aims: To determine whether significant variation exists in the prescribing and cost of key drug groups (i.e. specific, symptomatic and often presumptive drugs) within Scottish health boards. Methods: Data on prescribing for all 15-health boards in Scotland were extracted from the Prescribing Information System for Scotland (PRISMS) prescribing database for April 2004 – March 2005. Therapeutic groups were analysed according to judgement of the degree of specificity of their use into specific, symptomatic and often presumptive (Mcgavock 1992). Prescribing data together with Scotland average were sorted for each health board for Specific drug groups. Descriptive statistics for these data were determined using SPSS version 12. The variability of prescribing for each therapeutic group (i.e. specific, symptomatic and often presumptive drugs) was determined by estimating, differences from the Scottish national average. Health boards were stratified into, high and low prescribing health boards, according to their percentage difference of prescribing rate and cost from Scotland average for each therapeutic group. Results: An almost two fold variation in prescribing rate and cost of drugs in most therapeutic drug groups (Table1), between lowest and highest prescribing health boards existed. Overall

Western Isles, Argyll and Clyde, Dumfries, and Lanarkshire health boards prescribed a high percentage of drugs in all therapeutic drugs groups, with Lothian, Tayside and Orkney being one of the lowest prescribing health boards. Prescribing cost of drugs was higher in Lanarkshire, Fife and Forth valley health boards in all therapeutic drug groups, whereas Lothian and Orkney health boards tended to spend less on prescribing drugs in all therapeutic drug groups. Conclusions:A considerable inter – health board variation was found in the number of different drugs prescribed. A low prescription rate of often – presumptive and symptomatic drug groups are recommended and can act as a measure of prescribing quality. Variability in the prescription of specific drug group may be as a result of differing levels of morbidity in each health board. Further research is needed to find out causation. Acknowledgements: We would like to thank Information Statistics Division (ISD) Scotland for the data on which these results were based.

McGavock, H. 1992, Pharmacoepidemiology & Drug Safety, Vol.1, pp. 73-83.
Morton-Jones, T. & Pringle, M. 1993, BMJ, vol. 306, no. 6894, pp. 1731-1734.