Detection of impaired glucose regulation and/or type 2 diabetes mellitus, using primary care electronic data, in a multiethnic UK community setting

Aims/hypothesis The aim of this study was to develop and validate a score for detecting the glycaemic categories of impaired glucose regulation (IGR) and type 2 diabetes using the WHO 2011 diagnostic criteria. Methods We used data from 6,390 individuals aged 40–75 years from a multiethnic population based screening study. We developed a logistic regression model for predicting IGR and type 2 diabetes (diagnosed using OGTT or HbA 1c ≥6.5% [48 mmol/mol]) from data which are routinely stored in primary care. We developed the score by summing the β coefficients. We externally validated the score using data from 3,225 participants aged 40–75 years screened as part of another study. Results The score includes age, ethnicity, sex, family history of diabetes, antihypertensive therapy and BMI. Fifty per cent of a population would need to be invited for testing to detect type 2 diabetes mellitus on OGTT with 80% sensitivity; this is slightly raised to 54% that need to be invited if using HbA 1c . Inviting the top 10% for testing, 9% of these would have type 2 diabetes mellitus using an OGTT (positive predictive value [PPV] 8.9% [95% CI 5.8%,12.8%]), 26% would have IGR (PPV 25.9% [95% CI 20.9%, 31.4%]). Using HbA 1c increases the PPV to 19% for type 2 diabetes mellitus (PPV 18.6% [95% CI 14.2%, 23.7%]) and 28% for an HbA 1c between 6.0% and 6.4% (PPV 28.3% [95% CI 23.1%, 34.0%]). Conclusions The score can be used to reliably identify those with undiagnosed IGR and type 2 diabetes in multiethnic populations. This is the first score developed taking into account HbA 1c in the diagnosis of type 2 diabetes.