Optimizing chronic disease management mega-analysis: economic evaluation

Abstract

Background: As Ontario's population ages, chronic diseases are becoming increasingly common. There is growing interest in services and care models designed to optimize the management of chronic disease.

Objective: To evaluate the cost-effectiveness and expected budget impact of interventions in chronic disease cohorts evaluated as part of the Optimizing Chronic Disease Management mega-analysis.

Data sources: Sector-specific costs, disease incidence, and mortality were calculated for each condition using administrative databases from the Institute for Clinical Evaluative Sciences. Intervention outcomes were based on literature identified in the evidence-based analyses. Quality-of-life and disease prevalence data were obtained from the literature.

Methods: Analyses were restricted to interventions that showed significant benefit for resource use or mortality from the evidence-based analyses. An Ontario cohort of patients with each chronic disease was constructed and followed over 5 years (2006-2011). A phase-based approach was used to estimate costs across all sectors of the health care system. Utility values identified in the literature and effect estimates for resource use and mortality obtained from the evidence-based analyses were applied to calculate incremental costs and quality-adjusted life-years (QALYs). Given uncertainty about how many patients would benefit from each intervention, a system-wide budget impact was not determined. Instead, the difference in lifetime cost between an individual-administered intervention and no intervention was presented.

Results: Of 70 potential cost-effectiveness analyses, 8 met our inclusion criteria. All were found to result in QALY gains and cost savings compared with usual care. The models were robust to the majority of sensitivity analyses undertaken, but due to structural limitations and time constraints, few sensitivity analyses were conducted. Incremental cost savings per patient who received intervention ranged between $15 per diabetic patient with specialized nursing to $10,665 per patient wth congestive heart failure receiving in-home care.

Limitations: Evidence used to inform estimates of effect was often limited to a single trial with limited generalizability across populations, interventions, and health care systems. Because of the low clinical fidelity of health administrative data sets, intermediate clinical outcomes could not be included. Cohort costs included an average of all health care costs and were not restricted to costs associated with the disease. Intervention costs were based on resource use specified in clinical trials.

Conclusions: Applying estimates of effect from the evidence-based analyses to real-world resource use resulted in cost savings for all interventions. On the basis of quality-of-life data identified in the literature, all interventions were found to result in a greater QALY gain than usual care would. Implementation of all interventions could offer significant cost reductions. However, this analysis was subject to important limitations.

Plain language summary: Chronic diseases are the leading cause of death and disability in Ontario. They account for a third of direct health care costs across the province. This study aims to evaluate the cost-effectiveness of health care interventions that might improve the management of chronic diseases. The evaluated interventions led to lower costs and better quality of life than usual care. Offering these options could reduce costs per patient. However, the studies used in this analysis were of medium to very low quality, and the methods had many limitations.

Figure 1:. Diabetes Cost Curves for 5 Patient Subgroups (FY 2006–2010)

Figure 2:. Coronary Artery Disease Cost Curves for 5 Patient Subgroups (FY 2006–2010)

Figure 3:. Congestive Heart Failure Cost Curves for 5 Patient Subgroups (FY 2006–2010)

Figure 4:. Chronic Obstructive Pulmonary Disease Cost Curves for 5 Patient Subgroups (FY 2006–2010)