Urine collection devices to reduce contamination in urine samples for diagnosis of uncomplicated UTI: a single-blind randomised controlled trial in primary care

Background Urine collection devices (UCDs) are being marketed and used in clinical settings to reduce urine sample contamination, despite inadequate supporting evidence. Aim To determine whether UCDs, compared with standardised instructions for urine sample collection, reduce the proportion of contaminated samples. Design and setting Single-blind randomised controlled trial in general practices in England and Wales. Method Women aged ≥18 years presenting with symptoms attributable to urinary tract infection (UTI) were randomised (1:1:1) to use either a Peezy UCD or a Whiz Midstream UCD, or were given standardised verbal instructions (SVI) for midstream sample collection. The primary outcome was the proportion of urine samples reported as contaminated by microbiology laboratory analysis. Results A total of 1264 women (Peezy UCD: n = 424; Whiz Midstream UCD: n = 421; SVI: n = 419) were randomised between October 2016 and August 2018. Ninety women were excluded from the primary analysis as a result of ineligibility or lack of primary outcome data, leaving 1174 (Peezy UCD: n = 381; Whiz Midstream UCD: n = 390; SVI: n = 403) for intention-to-treat analysis. The proportion of contaminated samples was 26.5% with the Peezy UCD, 28.2% with the Whiz Midstream UCD, and 29.0% with SVI (relative risk: Peezy UCD versus SVI = 0.91, 95% CI = 0.76 to 1.09, P = 0.32; Whiz Midstream UCD versus SVI = 0.98, 95% CI = 0.97 to 1.20, P = 0.82). There were 100 (25.3%) device failures with the Peezy UCD and 35 (8.8%) with the Whiz Midstream UCD; the proportion of contaminated samples was similar after device failure samples were excluded. Conclusion Neither the Peezy UCD nor the Whiz Midstream UCD reduced urine sample contamination when used by women presenting to primary care with suspected UTI. Their use cannot be recommended for this purpose in this setting.