Point-of-care testing and antibiotics prescribing in out-of-hours general practice: a register-based study in Denmark

• Published in: BMC family practice Vol. 25; no. 1; pp. 31 - 8
• Main Authors: Christensen, Line Due, Vestergaard, Claus Høstrup, Keizer, Ellen, Bech, Bodil Hammer, Bro, Flemming, Christensen, Morten Bondo, Huibers, Linda
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 23.01.2024; BioMed Central; BMC
• Subjects: After-Hours Care; Anti-Bacterial Agents; Antibiotics; C-Reactive Protein; Decision making; Denmark; Dosage and administration; Family medicine; General Practice; Humans; Medical tests; Methods; Out-of-hours medical care; Patients; Point-of-Care Testing; Practice; Prescriptions; Primary care; Registration; Urine; Workloads; Denmark; Denmark; Point-of-care testing; General practice; Out-of-hours medical care; Anti-bacterial agents
• ISSN: 2731-4553; 2731-4553; 1471-2296
• DOI: 10.1186/s12875-024-02264-0

Point-of-care testing may reduce diagnostic uncertainty in case of suspicion of bacterial infection, thereby contributing to prudent antibiotic prescribing. We aimed to study variations in the use of point-of-care tests (C-reactive protein test, rapid streptococcal antigen detection test, and urine dipstick) among general practitioners (GPs) and the potential association between point-of-care testing and antibiotic prescribing in out-of-hours general practice. We conducted a population-based observational register-based study, based on patient contacts with out-of-hours general practice in the Central Denmark Region in 2014-2017. The tendency of GPs to use point-of-care testing was calculated, and the association between the use of point-of-care testing and antibiotic prescribing was evaluated with the use of binomial regression. Out-of-hours general practice conducted 794,220 clinic consultations from 2014 to 2017, of which 16.1% resulted in an antibiotic prescription. The GP variation in the use of point-of-care testing was largest for C-reactive protein tests, with an observed variation (p90/p10 ratio) of 3.0; this means that the GPs in the 90th percentile used C-reactive protein tests three times as often as the GPs in the 10th percentile. The observed variation was 2.1 for rapid streptococcal antigen detection tests and 1.9 for urine dipsticks. The GPs who tended to use more point-of-care tests prescribed significantly more antibiotics than the GPs who tended to use fewer point-of-care tests. The GPs in the upper quintile of the tendency to use C-reactive protein test prescribed 22% more antibiotics than the GPs in the lowest quintile (21% for rapid streptococcal antigen detection tests and 8% for urine dipsticks). Up through the quintiles, this effect exhibited a positive linear dose-response correlation. The GPs varied in use of point-of-care testing. The GPs who tended to perform more point-of-care testing prescribed more antibiotics compared with the GPs who tended to perform fewer of these tests.