Martijn Sijbom

150 Chapter 6 in the number of prescriptions compared to all previous years (p<0.05) (Figure 1). A statistically significant difference was found for day of the week, with antimicrobial prescriptions on Monday (242,487) and Friday (240,469) dominating compared to other weekdays, which varied between 194,704 and 211,276 prescriptions. Penicillins were the most prescribed antimicrobial group for every year of the study period (Figure 2). Antimicrobial prescriptions according to guideline recommendations Antimicrobial prescriptions with an ICPC code totalled 673,909, of which 585,117 had an ICPC code corresponding to an infection. Table 2 and table 3 show the distribution of determinants for appropriate and inappropriate antimicrobial prescriptions. Prescriptions classified as appropriate amounted to 480,792, compared with 104,325 inappropriate prescriptions. Urinary tract infections (UTI) (37.2%) and RTIs (36.2%) were the most common reason for an antimicrobial prescription. A substantial number of antimicrobial prescriptions were for RTIs and categorized as inappropriate (14.5%). Amongst prescriptions for RTIs alone, 39.6% were therefore classified as inappropriate (Figure 3). For the 480,792 appropriate antimicrobial prescriptions, 72.3% (347,846) corresponded with guidelines for the first or second choice antimicrobial for the diagnosis. With regard to macrolides, 41,363 appropriate prescriptions were for these compounds, of which over three-quarters (77.1%) were not the first or second choice according to Dutch primary care guidelines (Figure 4). Determinants Female gender, age of five years or older, a Turkish-, Surinamese- or Dutch- Caribbean background, a household with one parent, presence of comorbidities, a medium or large primary care practice size and Friday as day of prescription, were positively associated with antimicrobial prescription, meaning over prescription (Table 4). A Moroccan migration background was associated with relatively more appropriate antimicrobial prescriptions compared to a Dutch background. There was no association of household income with appropriateness (Table 4). Determinants associated with antimicrobial overprescribing for RTIs included male gender, age 5 years or older (except age group 15-44 years), Turkish, Surinamese or Dutch Caribbean background, a low household income, presence of a comorbidity, larger primary care practice and weekdays other than Friday as day of prescription (Table 5). A check for bias through a multivariable regression analysis that included patients without data in the SN data did not show different outcomes (Supplement 5).

RkJQdWJsaXNoZXIy MTk4NDMw