Intended for healthcare professionals


Home collection of urine for culture from infants by three methods: survey of parents' preferences and bacterial contamination rates

BMJ 2000; 320 doi: (Published 13 May 2000) Cite this as: BMJ 2000;320:1312
  1. Lynster C T Liaw, specialist registrar in paediatric nephrology,
  2. Deepa M Nayar, specialist registrar in microbiology,
  3. Stephen J Pedler, consultant microbiologist,
  4. Malcolm G Coulthard, consultant paediatric nephrologist (malcolm.coulthard{at}
  1. Royal Victoria Infirmary, Queen Victoria Road, Newcastle upon Tyne NE1 4LP
  1. Correspondence to: M G Coulthard
  • Accepted 22 February 2000

Urinary tract infection is common in childhood. Infants are most likely to scar and often have non-specific symptoms. Because of practical difficulties with collecting urine, samples are often not obtained.1 Most samples are collected by parents,2 yet nobody has sought parents' views on available methods. We assessed contamination rates and parents' opinions of three common methods used at home.

Subject, methods, and results

Parents of children aged 1 to 18 months volunteered to collect urine at home by pads,bags, and clean catch in a randomised order, on one day. The study had ethics committee approval. Demonstrations and instruction sheets were given. Parents washed their hands before each procedure and the child's perineum before each collection. Pads (Newcastle sterile urine collection packs, Ontex UK, Corby) were placed inside the nappy and checked every 10 minutes until wet (but not soiled), then urine aspirated with a syringe. Bags (Hollister U-Bag, Hollister, Libertyville, IL) were applied and inspected every 10 minutes and removed to decant the urine. For clean catch samples, infants were nursed with a sterile bottle ready. Samples were immediately instilled on to dipslides (Till-U-Test, Dimanco, Bedfordshire) with sterile swabsticks and returned with forms recording parents' collection times, comments, and rankings. Equipment costs were: pads 40 pence for 10 (or 59p for a pack containing syringe, bottle, and two pads); bag 89p; sterile bottle 7p; dipslide and swabstick 59p.

Forty four parents attempted collections (29 boys, median age 4 months, range 1 to 18months). No samples were obtained from one baby with diarrhoea, and no other child had a urine infection. Bacterial counts were <104/ml (typically reported as “insignificant” or “no growth”) from 31 (70%) pads, 29 (66%) bags, and 33 (75%) clean catch collections. Seven samples from pads, eight from bags, and one from clean catch collection had contamination (>104/ml of one or more organism). Nine samples (8%)from five children (four boys) grew>105 coliforms/ml, suggesting infection. However, this was excluded by sterile samples collected on the same day or on immediate repeat in hospital.

Assessments by 44 parents of three methods of home urine collection.

View this table:

Parents found the pads and bags easy to use and preferred them to clean catch collections (table) for both sexes. The pad was considered comfortable, whereas the bag was distressing, particularly on removal, often leaking and leaving red marks. Some found extracting the urine from the pad or emptying it from the bag to be awkward. Most parents complained that clean catch collections were time consuming and often messy; nine gave up after prolonged attempts. Five parents whose infants voided immediately ranked it best. The median collection time was 25 minutes for each method, but parents resented constraining their children this long for clean catch collections.


This is the first study of parents' views of infant urine collection methods. Pad, bag, and clean catch samples were equally effective at excluding an infection; variations in contamination rates balanced collection failures. Most parents disliked clean catch collections; their views should be heeded. Most preferred pads to bags, and they are cheaper. They also found inoculating dipslides with swabsticks easy; this technique may have contributed to their relatively low contamination rates.3 4

Since Kass suggested a diagnostic cut off of a single bacterial species cultured at >105/ml, it has been widely taken as proof of a urine infection and assumed not to occur from skin contamination, even though his study 5 and others 3 recorded similar false positive rates to ours. False positive results potentially lead to inappropriate treatment and imaging. Suprapubic puncture is an unrealistic alternative in primary care. Although collecting multiple samples would reduce the false positive rate, it might delay antibiotic treatment.


We thank the parents for volunteering and their thoughtful comments and Dr Mohammad Raza for microbiological help.

Contributors: MGC had the original idea for the study and is the guarantor. The studywas designed, the data analysed, and the paper written jointly by all the authors. LDTL carried out the clinical aspects and DMN and SJP the laboratory aspects of the study.


  • Funding None.

  • Competing interests The Children's Kidney Fund of the Newcastle University Hospitals special trustees receives royalties from Ontex UK from sales of Newcastle sterile urine collection packs.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract