Journal of Contextual Behavioral Science (JCBS)
Volume 14, October 2019, Pages 146-156
Authors
Graciela Rovner, Fredrik Johansson, David Gillanders
Abstract
Despite their strong evidence, group-based ACT-informed pain rehabilitation programs yield medium to low effect sizes pointing to an uneven responsiveness. This has led to the search for ways to identify differential needs and capacity to change. Previous studies found that Pain Willingness and Activity Engagement cluster into four profiles with distinct behavioral and functional patterns of psychological flexibility. Clustering could enable the creation of groups with shared rehabilitation needs and guide the tailoring of process-based rehabilitation.
The statistically created clusters in previous studies are however not easily transferable to clinical practice. This study therefore aimed to create raw score cutoffs to support the clinical implementation of the profiles. The raw scores were developed by visual exploration of the distribution of the CPAQ-8 scores of 1775 patients from a Latent Class Analysis. The cutoffs’ sensitivity and specificity were tested with Operating Characteristic (ROC) and Area Under the Curve (AUC) analysis and capacity to identify the profiles and patterns of behaviors were tested with ANOVA and MANOVA, in comparison to a statistically derived clustering solution. ROC analysis showed that clinical cutoffs could mimic the statistically created clusters with excellent sensitivity and specificity. The AUC ranged between 89.3% and 96.6%. The CPAQ-8 clinical cutoffs identify the same four profiles of Behavioral Flexibility, supporting a systematic assessment, selection and allocation of patients to groups with shared rehabilitation needs. The profiles carry information that will guide the clinician to tailor ACT-informed packages for each group. Implications for triage and treatment-design are discussed.