Background: . Health economic modeling has paid limited attention to the effects that patients' psychological characteristics have on the effectiveness of treatments. This case study tests 1) the feasibility of incorporating psychological prediction models of treatment response within an economic model of type 1 diabetes, 2) the potential value of providing treatment to a subgroup of patients, and 3) the cost-effectiveness of providing treatment to a subgroup of responders defined using 5 different algorithms.
Methods: . Multiple linear regressions were used to investigate relationships between patients' psychological characteristics and treatment effectiveness. Two psychological prediction models were integrated with a patient-level simulation model of type 1 diabetes. Expected value of individualized care analysis was undertaken. Five different algorithms were used to provide treatment to a subgroup of predicted responders. A cost-effectiveness analysis compared using the algorithms to providing treatment to all patients.
Results: . The psychological prediction models had low predictive power for treatment effectiveness. Expected value of individualized care results suggested that targeting education at responders could be of value. The cost-effectiveness analysis suggested, for all 5 algorithms, that providing structured education to a subgroup of predicted responders would not be cost-effective.
Limitations: . The psychological prediction models tested did not have sufficient predictive power to make targeting treatment cost-effective. The psychological prediction models are simple linear models of psychological behavior. Collection of data on additional covariates could potentially increase statistical power.
Conclusions: . By collecting data on psychological variables before an intervention, we can construct predictive models of treatment response to interventions. These predictive models can be incorporated into health economic models to investigate more complex service delivery and reimbursement strategies.
Keywords: cost utility analysis; decision analysis; logistic regression models; simulation methods.
© The Author(s) 2015.