OBJECTIVES: To develop and validate a population-level Discrete Event Simulation model of OA (DESOA) in Alberta, Canada, to forecast the prevalence and per patient average cost of OA from 2010 to 2035.

METHODS: The DESOA uses incidence, stage duration and resource use estimates across Primary care (PC), Specialist Care (SP), Acute Care (AC) and Sub-acute Care (SAC) stages according to patients’ demographics, obesity and comorbidity status. Parameters were estimated using Alberta Administrative Health Data from 1994/95 to 2012/13 in addition to Canadian Institute of Health Information inpatient cost data and population projection from Statistics Canada. Validated algorithms were used to identify patients with OA and eight major chronic diseases. All costs are reported in current year CAD$ using resource-specific inflation rates.

RESULTS: Projected by DESOA, the total number of patients with OA in Alberta will increase from 376,000 in 2010 to 439,000 in 2035. In 2035, 51% of patients will be in PC and will not be referred to SP, 27% will be non-surgical who will be referred back from SP to PC and 22% will be surgical. The highest proportion of patients in 2035 will be comprised of those with no comorbidity (42%), CVD/Diabetes (24%), multi-morbidity (18%), depression (10%) and COPD (6%). The highest average cost in PC will be associated with those with multi-morbidity ($2900), and those with CVD/diabetes in AC ($15,680). Furthermore, in 2035, the non-surgical cohort will have the highest average per patient cost in PC ($3820 in 2035) compared to surgical ($2430) and non-referred cohort ($480).

CONCLUSIONS: Our estimates suggest that OA’s average cost is significantly higher among those with co-occurring diseases and that the average cost at the PC stage was highest for the non-surgical cohort. An improvement in reducing average cost for such sub-population would substantially reduce the OA’s direct cost burden over the future years.