Using early childhood infections to predict late childhood antibiotic consumption: a prospective cohort study

• Main Authors: Kristian Gjessing, Johnny Ludvigsson, Åshild Olsen Faresjö, Tomas Faresjö
• Format: Article
• Language: English
• Published: Royal College of General Practitioners 2020-10-01
• Series: BJGP Open
• Subjects: anti-bacterial agents; prescriptions; child; prospective studies; socioeconomic factors; primary health care
• Online Access: https://bjgpopen.org/content/4/5/bjgpopen20X101085

Background: In the Swedish welfare system, the prescription and price of antibiotics is regulated. Even so, socioeconomic circumstances might affect the consumption of antibiotics for children. Aim: This study aimed to investigate if socioeconomic differences in antibiotic prescriptions could be found for children aged 2–14 years, and to find predictors of antibiotic consumption in children, especially if morbidity or socioeconomic status in childhood may function as predictors. Design & setting: Participants were from All Babies In Southeast Sweden (ABIS), a prospectively followed birth cohort (N = 17 055), born 1997-1999. Pharmaceutical data for a 10-year period, from 2005–2014 were used (the cohort were aged from 5–7, up to 14–16 years). Participation at the 5-year follow-up was 7443 children. All prescriptions from inpatient, outpatient, and primary care were included. National registries and parent reports were used to define socioeconomic data for all participants. Most children’s infections were treated in primary healthcare centres. Method: Parents of included children completed questionnaires about child morbidity at birth and at intervals up to 12 years. Their answers, combined with public records and national registries, were entered into the ABIS database and analysed. The primary outcome measure was the number of antibiotic prescriptions for each participant during a follow-up period between 2005–2014. Results: The most important predictor for antibiotic prescription in later childhood was parent-reported number of antibiotic-treated infections at age 2–5 years (odds ratio (OR) range 1.21 to 2.23, depending on income quintile; P<0.001). In the multivariate analysis, lower income and lower paternal education level were also significantly related to higher antibiotic prescription. Conclusion: Parent-reported antibiotic-treated infection at age 2–5 years predicted antibiotic consumption in later childhood. Swedish doctors are supposed to treat all patients individually and to follow official guidelines regarding antibiotics, to avoid antibiotics resistance. As socioeconomic factors are found to play a role, awareness is important to get unbiased treatment of all children.