The evaluation of incidentally detected symptomless microhaematuria in schoolchildren is controversial. Some authors advocate varying numbers of immediate tests, while others recommend investigations only in cases who develop systemic symptoms or signs, or a decline in renal function. The objective of this study was to estimate the extent to which this uncertainty affects the declared habits of practising physicians. A sample of 16 family physicians, 42 primary care paediatricians and 26 full-time hospital-based paediatric nephrologists in Israel were asked to complete a survey using a written case of a hypothetical eight-year-old boy with incidentally detected symptomless microhaematuria. Responses were received from 16 (100%), 18 (43%) and 18 (69%), respectively. The mean number of requested tests, other than follow-up examination of the urine, were 1.5 (range 0-5) for family physicians, 2.5 (1-5) for primary care paediatricians and 5.3 (2-12) for paediatric nephrologists, at an average cost of NIS 408 (US$ 136), NIS 454 (US$ 151) and NIS 860 (US$ 286), respectively. There was also a marked variability within subspecialty groups, so that some family physicians recommended more tests at a higher cost than some of the paediatric nephrologists. There was a marked and unexplained variability within and among the three groups of respondents regarding the extent of the evaluation. The main reason for this variability is probably the uncertainty about the scientifically appropriate way to approach this condition in a symptomless child.