Health Economic Modeling Schizophrenia Outcomes Using Time-to-Event Simulation
Nicolas M. Furiak, MS; Harry J. Smolen, MS; James C. Ghan, BS; and Megha Bansal, MA
From the Medical Decision Modeling, Inc., Indianapolis, Indiana.
This poster presentation was supported by #1R43MH082585.
Background: Schizophrenia is a lifelong debilitating disease affecting roughly 1% of the population worldwide. Stabilization of symptoms in new patients or patients with an exacerbation of symptoms is critical in improving long term patient outcomes and quality of life. Furthermore, patients who relapse have higher rates of subsequent relapse, incur higher costs of care, and are at risk for undesirable lifetime outcomes. Computer models and simulations have been used to predict health and economic outcomes from a general population perspective and have been used to assist decision making for stakeholders in the care of patients with schizophrenia. Recent efforts have focused on patient-specific characteristics such as symptom scores that provide opportunities to modify disease pathways continuously using complex rule sets.
Methods: This study uses a discrete event simulation employing time-to-event methodology to estimate health and economic outcomes for patients with schizophrenia. The model implements Positive and Negative Symptom Scale scores as a central thread that represents a patient’s general disease severity as well as act as a trigger to modify critical parameters. The model was designed such that patients have equal chance of remaining in their initially assigned severity state (mildly ill, moderately ill, severely ill), improve, or worsen over the timeframe of the model. Relapses are defined as a general increase in resource utilization and interaction with healthcare resources and can occur in any setting: in the community or in the hospital. Duration of relapse and time between relapses are scheduled according to disease severity. Costs are calculated assigning units of resources according to the patients health state and setting. The treatment comparators were risperidone standard oral therapy (SOT) and injectable long-acting therapy (LAI) administered monthly. The default time horizon is 5 years.
Results: The model predicted that the total cost of care for SOT was $65,409 with patients experiencing a mean number of 4.35 total relapses. Total mean time spent in relapse was 2.12 years. LAI therapy yielded a $2,255 reduction in total costs, 1.19 fewer total relapses, and a reduction of 0.10 years in relapse. More time was spent in the community for the LAI patients than for SOT patients.
Conclusions: Computer models implementing time-to-event methodology using symptom scores as a central mechanism demonstrate the capability to produce credible health and economic outcomes.
To demonstrate understanding of discrete event simulation in the analysis of care for mental health patients
To validate symptom scores as a primary driver of simulated mental health outcomes
NIMH. Statistics: schizophrenias. Updated 2012. http://www.nimh.nih.gov/statistics/1SCHIZ.shtml. Accessed December 19, 2011.
Ascher-Svanum H, Zhu B, Faries D, et al. A comparison of olanzapine and risperidone on the risk of psychiatric hospitalization in the naturalistic treatment of patients with schizophrenia. Ann Gen Hosp Psychiatry. 2004;3(1):11. PubMed
Edwards NC, Rupnow MF, Pashos CL, et al. Cost-effectiveness model of long-acting risperidone in schizophrenia in the US. Pharmacoeconomics. 2005;23(3):299–314. PubMed
Heeg B, Buskens E, Knapp M, et al. Modelling the treated course of schizophrenia: development of a discrete event simulation model. Pharmacoeconomics. 2005;23(suppl 1):17–33. PubMed