Almarsdóttir, Anna Birna3; Sigurðsson, Magnús2; Guðnason, Vilmundur2
1 Clinical Pharmacology and Pharmacy, Department of Public Health, Det Sundhedsvidenskabelige Fakultet, SDU2 unknown3 Clinical Pharmacology and Pharmacy, Department of Public Health, Det Sundhedsvidenskabelige Fakultet, SDU
hönnun og prófun spurningalistadesign and testing of a questionnaire
Background: Epidemiologic studies show that open-ended questionnaire items are unreliable measures of the use of drugs and related substances. It is also important to avoid questions regarding the distant past. These issues were kept in mind when designing a questionnaire on the use of herbal medicines, food supplements, and over-the-counter drugs (OTCs). The objectives of this study were to pre-test the questionnaire and measure the prevalence of use in the last two weeks. Methods: A questionnaire was constructed almost entirely with closed-ended responses. A pre-test was carried out in two phases. Respondents were patrons of the Icelandic Heart Association's clinic. In the utilization study a sample was taken from all 18 079 participants of the MONICA Reykjavík research study and the Descendant Study of the Icelandic Heart Association. The random sample was stratified according to age and sex (N=350). A chi-squared test was used to compare rates. Results: The questionnaire was changed little after pre-testing. The main changes related to wording of questions. In the utilization study, 220 individuals responded (62.9%). The prevalence of herbal use was 46.8%, 75.9% for vitamins, minerals, and micronutrients, and 69.5% for OTCs. Garlic was the most commonly used herbal and food-supplement (14.5%). Codliver oil held a superior position in the group of vitamins, minerals, and micronutrients (59.1%). Vitamin C was the most used single vitamin (20,9%). Calcium was by far the most used mineral (11,4%). Pain relievers were the most common OTCs (38.6%). Conclusions: The format of the questionnaire was satisfactory. The prevalence of use of all groups of substances was high compared to foreign studies. Response bias may make the use seem higher than is true. As the sample was very small, this study should be viewed as a test of a specific method for measuring the use of these substances.