Using computer modelled life expectancy to evaluate the impact of Australian Primary Care Incentive programs for patients with type 2 diabetes

- We propose a novel methodology for evaluating a primary care initiative for patients with T2D.
- We modelled life expectancy using routinely collected general practice data from Australia.
- Diabetes cycle of care completion did not significantly increase life expectancy.
- The potential for substantial gains in life expectancy were identified.
- Better recording of clinical data is necessary to fully realise the value of the approach.

AimsTo evaluate the impact of enhanced primary care and practice incentive programs on the care of patients with type 2 diabetes in the Australian primary care setting using routinely collected data and computer modelling software.MethodsPrimary care patient data were electronically extracted from practices and inputted into the United Kingdom Prospective Diabetes Study (UKPDS) Outcomes model. A retrospective cohort study design was employed with predicted life expectancies compared between patients who had a recorded diabetes cycle of care (DCoC) and those who did not. Changes in glycated haemoglobin (HbA1c) were also analysed using a mixed-effects regression model. Potential life expectancy gains were estimated by inputting theoretical risk factors data consistent with current guidelines.ResultsTwelve primary care practices were recruited and suitable data were available for 559 people with type 2 diabetes. Two hundred and twenty five patients (40%) were identified as having completed at least one DCoC and as a group had a predicted additional life expectancy of 0.65 years (95% CI, −0.22 to 1.5). However, once this was adjusted for comorbidities the difference reduced to 0.03 years. There was no significant difference in HbA1c levels attributable to the intervention. An estimated 0.5 year of additional life expectancy was predicted should all patients have complied with current risk factor guideline recommendations.ConclusionsComputer modelling using routinely collected primary care data can be used to evaluate the effectiveness of primary care programs. However, there are some data availability and linkage limitations in the Australian setting.