The completeness of electronic medical record data for patients with type 2 diabetes in primary care and its implications for computer modelling of predicted clinical outcomes.

Abstract

Background: Computer models predicting outcomes among patients with Type 2 Diabetes (T2D) can be used as disease management program evaluation tools. The clinical data required as inputs for these models can include annually updated measurements such as blood pressure and glycated haemoglobin (HbA1c). These data can be extracted from primary care physician office systems but there are concerns about their completeness. Objectives/methods: This study addressed the completeness of routinely collected data extracted from 12 primary care practices in Australia. Data on annual availability of blood pressure, weight, total cholesterol, HDL-cholesterol and HbA1c values for regular patients were extracted in 2103 and analysed for temporal trends over the period 2000 to 2012. An ordinal logistic regression model was used to evaluate associations between patient characteristics and completeness of their records. Primary care practitioners were surveyed to identify barriers to recording data and strategies to improve its completeness. Results: Over the study period completeness of data improved substantially from less than 20% for some parameters up to a level of approximately 80% complete, except for the recording of weight. T2D patients with Ischaemic Heart Disease were more likely to have their blood pressure recorded (OR 1.6, p=0.02). Practitioners’ responses suggest they were not experiencing any major barriers to using their electronic medical record system but did agree with some suggested strategies to improve record completeness. Conclusion: The completeness of routinely collected data suitable for input into computerised predictive models is improving although other dimensions of data quality need to be addressed.