The aim of the article was to use prospectively collected data on people with type 1 diabetes to examine which routinely collected clinical measures predict the development of peripheral neuropathy in people with type 1 diabetes. Within the Newcastle Diabetes Services, structured data collection at an annual review has been collected since 1985. This includes metabolic measures, cardiovascular risk factors, and markers of complications. From 1990 data collection was standardized and computerized. For this study, all people with type 1 diabetes in the database in both 1992 and 2001 were ascertained. Data were extracted for a diagnosis of peripheral neuropathy (based on neuropathic symptoms, absence of pinprick sensation, and abnormal biothesiometer measurements and/or monofilament sensation) and for the other metabolic and cardiovascular risk measures, as well as markers of other microvascular complications. Associations with the development of neuropathy were sought. Eighteen of 404 people already had peripheral neuropathy in 1992, and 38 others developed neuropathy during follow-up. People who developed neuropathy were older (47 +/- 14 [SD] versus 36 +/- 11 years; P = 0.000), had longer-duration of diabetes (27 +/- 13 versus 18 +/- 10 years; P = 0.001), higher baseline serum cholesterol (5.8 +/- 1.3 versus 5.2 +/- 1.2 mmol/L, P = 0.017), and higher systolic (139 +/- 18 versus 129 +/- 20 mmHg; P = 0.003) and diastolic BP (82 +/- 12 versus 76 +/- 11 mmHg; P = 0.009) than those who remained free of neuropathy. We found no significant difference for BMI and glycated hemoglobin. The multivariate model showed that diastolic BP, duration of diabetes, serum cholesterol, and history of callus/ulcers on the feet predicted the development of peripheral neuropathy. Neuropathy developed in 11.4% of people with type 1 diabetes over a 9-year follow-up, and was predicted by factors normally associated with cardiovascular rather than microvascular disease.