| To Register SMDM Homepage | |
|
|
|
||||
Purpose: To develop a formulation of schizophrenia progression by modeling the 1- year time profile of severities of positive and negative symptoms associated with the disease under the influence of one of the three standard pharmacological treatments.
Methods: We develop a representative model for the natural course of schizophrenia in 18-65 year old patients. The model not only accounts for the positive and the negative domains of the disease but also accounts for the severities within each domain. Item-specific scores from the Positive and Negative Syndrome Scale (PANSS) are used to define four severity states within each domain. We employ Bayesian synthesis of published clinical trials and observational studies to estimate the transition probabilities between symptom severity states over time, while properly reflecting the overall uncertainty in the parameters that guide these transitions. Lower transition probabilities to moderate and severe symptom states indicates better control of symptoms
Results: Based on the posterior mean estimates, we find several detailed transition patterns between severities of positive, negative and joint symptoms over time and by treatment, that were masked in published transition rates. Namely, the transition probabilities indicate that risperidone is best in controlling severe positive symptoms while olanzapine is the worst during the first quarter of drug treatment. However, haloperidol turns out to be the best in controlling severe positive symptoms for the next two quarters of treatment. Olanzapine appears to be the best drug to control severe negative symptoms across all four quarters of treatment while haloperidol is the worst in this regard. Finally, all three drugs are able to prevent deterioration of negative symptoms as long as they can effectively control the positive symptoms.
Conclusions: Various fine levels of detail on the transition probabilities may serve to estimate quality of life of schizophrenic patients and resource utilizations in this field more accurately. Bayesian synthesis of evidence also reveals considerable uncertainty in the time-profile parameters and thereby the transition rates in the published findings. Estimates of parameter uncertainty by itself have important implications for the practice of cost-effectiveness analysis and future resource allocation policies in schizophrenia treatments.
See more of Poster Session - CEA: Methods and Applications; Health Services Research
See more of The 26th Annual Meeting of the Society for Medical Decision Making (October 17-20, 2004)