Antibiotic management of urinary tract infection in elderly patients in primary care and its association with bloodstream infections and all cause mortality: population based cohort study

Abstract

Objective: To evaluate the association between antibiotic treatment for urinary tract infection (UTI) and severe adverse outcomes in elderly patients in primary care.

Design: Retrospective population based cohort study.

Setting: Clinical Practice Research Datalink (2007-15) primary care records linked to hospital episode statistics and death records in England.

Participants: 157 264 adults aged 65 years or older presenting to a general practitioner with at least one diagnosis of suspected or confirmed lower UTI from November 2007 to May 2015.

Main outcome measures: Bloodstream infection, hospital admission, and all cause mortality within 60 days after the index UTI diagnosis.

Results: Among 312 896 UTI episodes (157 264 unique patients), 7.2% (n=22 534) did not have a record of antibiotics being prescribed and 6.2% (n=19 292) showed a delay in antibiotic prescribing. 1539 episodes of bloodstream infection (0.5%) were recorded within 60 days after the initial UTI. The rate of bloodstream infection was significantly higher among those patients not prescribed an antibiotic (2.9%; n=647) and those recorded as revisiting the general practitioner within seven days of the initial consultation for an antibiotic prescription compared with those given a prescription for an antibiotic at the initial consultation (2.2% v 0.2%; P=0.001). After adjustment for covariates, patients were significantly more likely to experience a bloodstream infection in the deferred antibiotics group (adjusted odds ratio 7.12, 95% confidence interval 6.22 to 8.14) and no antibiotics group (8.08, 7.12 to 9.16) compared with the immediate antibiotics group. The number needed to harm (NNH) for occurrence of bloodstream infection was lower (greater risk) for the no antibiotics group (NNH=37) than for the deferred antibiotics group (NNH=51) compared with the immediate antibiotics group. The rate of hospital admissions was about double among cases with no antibiotics (27.0%) and deferred antibiotics (26.8%) compared with those prescribed immediate antibiotics (14.8%; P=0.001). The risk of all cause mortality was significantly higher with deferred antibiotics and no antibiotics than with immediate antibiotics at any time during the 60 days follow-up (adjusted hazard ratio 1.16, 95% confidence interval 1.06 to 1.27 and 2.18, 2.04 to 2.33, respectively). Men older than 85 years were particularly at risk for both bloodstream infection and 60 day all cause mortality.

Conclusions: In elderly patients with a diagnosis of UTI in primary care, no antibiotics and deferred antibiotics were associated with a significant increase in bloodstream infection and all cause mortality compared with immediate antibiotics. In the context of an increase of Escherichia coli bloodstream infections in England, early initiation of recommended first line antibiotics for UTI in the older population is advocated.

Fig 1

Timeline of study and criteria for differentiating independent episodes of urinary tract infection (UTI)

Fig 2

Flowchart of study cohort. *Observations define all general practitioner (GP) contacts (rows in database) in Clinical Research Datalink database (CPRD). †UTI episode contains all GP contacts that define a single event of UTI for a patient. UTI episode includes 30 day period before diagnosis and 60 day follow-up period after diagnosis. HES=hospital episodes statistics

Fig 3

Kaplan-Meier survival curves by antibiotic management over 60 days