Using computerised general practice patient records to evaluate primary care programs: Modelling the impact of an Australian initiative to improve type 2 diabetes care

Abstract

This research proposes and assesses the utility of an approach for evaluating the impact of primary health care initiatives within the Australian health sector. It uses the Australian Government Diabetes Practice Inventive Program as a case study and analyses historical data recorded by general practitioners over the period 2000 to 2012. Data on regular patients with type 2 diabetes mellitus were electronically extracted from the routinely used medical record software of 12 general practices located within the Northern Sydney Local Health District in NSW. After checking these data for completeness, parameters on 559 (37%) regular patients with type 2 diabetes were inputted into the UKPDS Outcome Model V1 to predict individual patient life expectancy. The recording of at least one annual Diabetes Cycle of Care was used to define an intervention group and the impact of the program was assessed in terms of life expectancy differences. Two hundred and twenty five patients (40%) were included in the intervention group and as a group they had a predicted additional life expectancy of 0.65 years (95% CI, -0.22 to 1.5). However, once this was adjusted for baseline comorbidities using a regression analysis the difference reduced to 0.03 years. An estimated 0.5 of a year of additional life expectancy was predicted should all patients have complied with current risk factor guideline recommendations. An analysis of data completeness of medical records demonstrated substantial improvement over the period 2000 to 2012. The exception to this finding was seen in the recording of patient weight that remained at only 55% completeness over the period. A survey of general practitioners participating in the study identified “pop up” reminders, routine auditing of records and encouraging practice nurses to record clinical measurements as strategies to assist in improving the completeness of records. The method described was able to quantitatively evaluate the impact of a primary care program in terms of patient-important outcomes utilising existing data routinely stored within Australian general practice records. Further development and standardisation of primary care electronic medical record systems would enhance the utility of the methodology.