A prospective population-based cohort study conducted across three hospitals in North Denmark from September 20 through October 23, 2021, evaluated the diagnostic accuracy of urinary dipstick tests for urinary tract infections in acutely hospitalized adults. Among 2,495 patients admitted to emergency departments, 42% (1,052) underwent dipstick testing. The tested population had a median age of 73 years (IQR 57-82), was 51% female, and 71% were admitted to internal medicine departments.

Of those tested, 8% (89/1,052) met urinary tract infection (UTI) criteria, defined by symptoms such as dysuria, frequency, or suprapubic tenderness combined with a positive urine culture. Classic UTI symptoms were present in 17% of the dipstick-tested patients. The type of urine sample was not reported in 81% of cases, and the indication for dipstick analysis was not documented in 60% of cases.

Urine cultures were performed in 58% (607/1,052) of patients who underwent dipstick testing. The dipstick's sensitivity and specificity for UTI detection were 87% (95% confidence interval [CI], 78%-93%) and 45% (95% CI, 41%-50%), respectively, with a positive predictive value of 21% and a negative predictive value of 95%. Pretest probabilities for UTI ranged from 29% to 60% depending on the presence of specific symptoms. Following dipstick testing, positive results raised the probability of UTI to between 35% and 68%, whereas negative results lowered it to between 12% and 27%.

The findings, published in BMJ Evidence-Based Medicine, showed that a positive dipstick in patients without specific UTI symptoms was associated with an increased risk for empirical antibiotic therapy (relative risk 4.41; 95% CI, 2.40-8.11). The study, conducted with clinicians unaware of the research to avoid influencing management decisions, concluded that dipsticks yielded limited clinical decision support compared to a symptom-driven approach.

Full disclosures can be found in the published study.