Related Experiment Videos
The reliability of Marlatt's taxonomy for classifying relapses
R Longabaugh1, A Rubin, R L Stout
1Center for Alcohol and Addiction Studies, Brown School of Medicine, Providence, Rhode Island 02912, USA.
Addiction (Abingdon, England)
|December 1, 1996
Summary
Marlatt's relapse taxonomy, used in addiction research, shows inconsistent reliability across studies. This challenges the comparability of findings and the aggregation of addiction treatment knowledge.
Area of Science:
- Addiction research
- Clinical psychology
- Behavioral science
Background:
- Marlatt's relapse taxonomy is widely used in addiction research and clinical practice.
- It classifies precipitants of relapse, aiding in understanding addiction recovery.
- The taxonomy's reliability across independent studies has not been previously assessed.
Purpose of the Study:
- To evaluate the inter-rater and inter-laboratory reliability of Marlatt's relapse taxonomy.
- To determine if the taxonomy can be consistently applied across different research settings.
- To assess the impact of inconsistent reliability on addiction research comparability.
Main Methods:
- Independent classification of 149 relapse episodes by trained raters from three research laboratories.
- Utilized Marlatt's taxonomy for categorizing relapse precipitants.
- Statistical analysis to compare reliability across raters and laboratories.
Main Results:
- Inconsistent reliability was found in classifying relapse episodes across the three research laboratories.
- Despite extensive cross-laboratory training, significant variability in classification persisted.
- The findings indicate challenges in applying the taxonomy uniformly in research.
Conclusions:
- Comparability of addiction research findings based on Marlatt's relapse taxonomy is questionable.
- Assumptions for aggregating knowledge in addiction treatment using this taxonomy are not supported.
- Recommendations are proposed to enhance taxonomy reliability and data collection methods.