Use of a health insurance company database for study of theoretical exposure to hypolipidemic agents

OBJECTIVE: The objective of the investigation was to analyze theoretical exposures to hypolipidaemics in patients treated chronically with these drugs, using the database of the health insurance company. INVESTIGATED GROUP: From the database (with information on age, sex of the insured person, the number of packages and the type of hypolipidemic and year of issue of the prescription) of subjects insured at the Employees Health Insurance Skoda Mlada Boleslav comprising some 100,000 insured subjects in 1994-2000. Patients with long-term (more than one year) hypolipidaemic treatment were selected in years from 1995 to 1999. The group increased every year. In 1995 it comprised 668 cases in 1999, 2396 subjects. METHOD: The consumption of hypolipidaemics was expressed in defined daily doses (DDD). The authors investigated the ratio of chronically treated patients and the proportion of the following groups of patients according to their annual consumption in 1995-1999: group of of drug "vacation" (0 DDD) and the group with a low ( 121.7 DDD) and optimal (> 243.3 DDD) consumption of hypolipidaemics and their relationship to sex and age. For statistical ealuation software SPSS 10.1 was used. RESULTS: In the course of the investigation among the insured subjects the statin consumption increased 76 times and the consumption of fibrates 5 times. The ratio of consumption of resin derivatives and of nicotinic acid was negligible. The size of the group of subjects treated with hypolipidaemics for longer than one year increased from 0.8% in 1995 to 2.2% of the database. The average age increased from 55 to 59 years. The ratio of seniors (> or = 65 years) increased in the course of the investigation and reached 33% in 1999 of all members of the investigated group. The mean annual consumption of hypolipidaemics increased significantly as compared with 1995 and the interannual increase as compared with the previous year was statistically significant in 1997 and 1999. In 1999 it was 237 DDD/per consumer. A lower consumption was recorded in women and in seniors. Drug "vacations" were recorded in 6% of the insured subjects of the group and the frequency did not change significantly in the course of the investigation and no relationship with age and sex was found. A low exposure according to DDD was found in 20%, medium exposure in about 40% and optimal exposure in only one third of the subjects of the investigated group. CONCLUSION: The authors developed a method which makes it possible, when individual data of the health insurance company are available, to investigate the theoretical exposure to hypolipidaemics in insured subjects treated on a long-term basis with these drugs. The authors provided evidence that analysis of the database of the health insurance company can provide certain signals for further pharmacoepidemiological research and for application in a defined medical discipline. Some of the insured subjects are exposed to smaller doses than theoretically assumed. It is necessary to extend the investigation so that the results will better reflect the population of patients and prescribing physicians. Complete evaluation of cases with a low exposure from the aspect of morbidity and drug compliance will be also essential.