A cross-sectional study to assess home glucose monitoring practices was conducted in 200 non-insulin-treated diabetic patients consecutively attending our hospital clinic. Of the 200, 97 (48%) patients (Group 1) regularly monitored urine (n = 74), blood (n = 19) or both (n = 4); 103 (52%) patients (Group 2) performed no home monitoring. The two groups were similar in terms of age, sex, duration of diabetes and type of treatment. The prevalence of diabetic complications was also closely comparable and only peripheral neuropathy differed between the groups, being more common in Group 1 (n = 12) than Group 2 (n = 4); p < 0.05. There was also no significant difference between the HbA1 concentration (mean +/- SD) in Group 1 (9.7 +/- 2.2%) and Group 2 (9.4 +/- 2.0%). The mean frequency of home monitoring was four tests weekly, but only 21 (22%) kept a written record and 60 (62%) would never alter their treatment on the basis of their results. Almost a third of patients could not interpret the results of monitoring or give the normal range of values. Home glucose monitoring, particularly of urine, is widely practised in Type 2 diabetes, at considerable overall expense. However, convincing evidence of its value in helping patients improve their blood glucose control or preventing the complications of the disease is lacking.