Predicting Treatment Response in Schizophrenia
Author Information
Author(s): Ruberg Stephen J, Chen Lei, Stauffer Virginia, Ascher-Svanum Haya, Kollack-Walker Sara, Conley Robert R, Kane John, Kinon Bruce J
Primary Institution: Eli Lilly and Company
Hypothesis
Can early changes in specific symptoms predict longer-term response to atypical antipsychotics in schizophrenia patients?
Conclusion
Early changes in specific PANSS symptoms can accurately predict longer-term treatment response to atypical antipsychotics in schizophrenia patients.
Supporting Evidence
- 92% of patients' treatment responses could be predicted using the decision tree.
- The decision tree identified likely responders with a positive predictive value of 79%.
- The negative predictive value of the decision tree was 75%.
Takeaway
Doctors can tell if a patient will get better on their medication by looking at how their symptoms change in the first two weeks.
Methodology
Data pooled from 6 randomized, double-blind trials; CART analysis used to create a decision tree based on PANSS symptom changes.
Potential Biases
Potential bias due to the retrospective nature of the study and reliance on previously published data.
Limitations
The model may not apply to patients without prominent positive symptoms and requires further validation in diverse populations.
Participant Demographics
Patients with chronic schizophrenia, mean age 35-45, predominantly diagnosed with schizophrenia.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website