File
Dhondt E.1, Van Oosterwijck J.1,2, Cagnie B.1, Adnan R.1,3, Schouppe S.1, Van Akeleyen J.4, Logghe T.4, Danneels L.1, Goubert D.1
1Ghent University, Rehabilitation Sciences and Physiotherapy, Ghent, Belgium, 2Research Foundation - Flanders (FWO), Brussels, Belgium, 3Universiti Teknologi MARA, Faculty of Sports Science and Recreation, Shah Alam, Malaysia, 4General Hospital St-Dimpna, Physical and Rehabilitation Medicine, Geel, Belgium
Background: Since evidence indicates that treatment content should be adapted according to patients their individual needs, a first step aimed at increasing the effectiveness of various therapies for patients with low back pain (LBP) is to identify clinical relevant subgroups of patients and match them with targeted treatment approaches. Consequently, there is a growing need to determine patient pre-treatment characteristics that could predict adherence and clinical outcome following specific interventions.
Purpose: To identify predictors of adherence to and clinical outcome of an outpatient multimodal rehabilitation program for patients with chronic LBP.
Methods: The data of 273 patients with chronic LBP who participated in a multimodal rehabilitation program of an outpatient rehabilitation centre in Belgium were analysed. The rehabilitation program consisted out of 36 therapeutic sessions and the content was mainly exercise-based. Possible predictors were obtained from pre- and post-treatment assessments existing out of a medical examination and standardized questionnaires. Univariate and multivariate logistic regression analyses were performed to identify predictors of treatment adherence and outcome. Patients who completed ≥70% of the treatment course were classified as adherent. Patients showing a post-treatment reduction of ≥30% in Oswestry Disability Index (ODI) and Visual Analogue Scale (VAS) back pain intensity scores were assigned to the favourable outcome group.
Results: While 83% of the total patient sample was adherent to the therapy, only 33% out of the 227 patients who completed the program had a favourable clinical outcome. Multivariate logistic regression revealed that older age (OR=1.039) and a higher Tampa Scale of Kinesiophobia (TSK) score (OR=1.150) increased the odds to complete the treatment program, whereas higher VAS back pain intensity scores (OR=0.802) reduced the odds for adherence. A higher ODI score (OR=0.961) decreased the odds for a favourable clinical outcome.
Conclusion(s): Older patients and patients with higher levels of kinesiophobia at intake are more likely to adhere to a multimodal rehabilitation program for chronic LBP, whereas patients with higher levels of back pain intensity are more likely to withdraw early. Patients with lower levels of LBP related disability at intake are more likely to have a favourable outcome. While this study identified treatment predictors, moderators of the obtained treatment effect could not be investigated since a non-randomized trial was performed. Consequently, there is a high need for randomized trials that further develop and test a prediction rule. The findings of this study may be helpful to guide the selection of variables.
Implications: Assessment of chronic LBP patient pre-treatment characteristics such as age, degree of kinesiophobia, back pain intensity and disability levels may be of great interest to clinical practice. Directing a treatment according to those characteristics could function as a tool to optimize treatment benefits. At the same time, patients at risk for showing a non-favourable outcome could be identified and referred to a different treatment approach. The results of this study has provided some valuable insights which may assist therapists in everyday clinical decision making in the treatment of individual patients with chronic LBP.
Funding acknowledgements: This work was supported by the Research Foundation - Flanders (FWO) and the Special Research Fund of Ghent University [BOF14/IOP/067].
Topic: Musculoskeletal: spine
Ethics approval: Ethical approval was received from the local ethics committee of the St.-Dimpna General Hospital.
All authors, affiliations and abstracts have been published as submitted.
