Hisamatsu T.M.1, Morelhão P.K.1, Oliveira C.B.2, Damato T.M.M.2, Demarchi S.J.2, Silva F.G.3, Teixeira R.J.2, Franco M.R.1, Pinto R.Z.2
1São Paulo State University (UNESP), Physiotherapy Department, Presidente Prudente, Brazil, 2São Paulo State University (UNESP), Physical Therapy Department, Presidente Prudente, Brazil, 3Sao Paulo State University (UNESP), Physiotherapy Department, Presidente Prudente, Brazil
Background: Studies that clarify the role of physical activity can help in a better understanding of the prognosis of chronic non-specific low back pain (LBP), also helping clinicians and patients for a better understanding the disease process and to predict more accurately the clinical outcomes.
Purpose: To investigate the ability of physical activity level, measured by the Baecke Physical Activity Questionnaire (BPAQ) and accelerometer, in predicting clinical outcomes (i.e. pain and disability) in patients with LBP.
Methods: The study design is a longitudinal cohort with a 6-month follow-up. Participants (N=100) with non-specific low back pain (LBP) completed a set of questionnaires, in which the following data were collected: Anthropometric/ demographic data (i.e., age, Body Mass Index [BMI], work status, education level); duration and intensity of pain; disability measured by the Quebec Back Pain Disability Scale (QUEBEC); fear of movement measured by Tampa Scale of Kinesiophobia (TSK); depression measured by Beck Depression Inventory (BDI) and physical activity levels measured by BPAQ and accelerometer (i.e., time spent on moderate-to-vigorous physical activity per day [MVPA] and light physical activity per day [LPA], steps per day and counts per minutes). A second evaluation was conducted in the 6-month follow-up. Clinical endpoints assessed at 6-month follow-up were: intensity of pain and disability. Correlational analysis was used to verify whether there is any high correlation between variables, univariate and multivariate linear regression analyses were performed separately to verify whether baseline physical activity levels predicts pain and disability at 6-month follow-up. The BPAQ was analysed using the BPAQ total score as well as the three separate domains: occupational, sports and leisure time (excluding sports) activities. The analyses were performed with pain and disability as variable dependent; MVPA, LPA, steps per day, counts per minutes, BPAQ occupational, BPAQ sports, BPAQ leisure time and BPAQ total score as independent variable; and age, BMI, work status, education level, duration of pain, BDI, TSK and baseline, QUEBEC and pain as covariates.
Results: In the multivariate final model, disability had association with age, education level, depression, baseline disability and LPA (β= -.33; 95% IC: -.066 to .000; p .049). This variable explained 37.2% of the variance of disability. Although steps per day showed an association with disability in correlation analysis (r=-.258; p .05) and univariate linear regression (β= -.038; 95%IC: -.067 to -.005; p=.025), these association did not appear in the multivariate final model. None of the physical activity measures remained in the final model when pain was analysed as the dependent variable.
Conclusion(s): This research shows an association between disability and time spent in light activities per day.
Implications: Physical activity level, measured subjectively, apparently is not able to predict pain intensity and disability. A significant association was found between an objective measure of physical activity (i.e. LPA) and disability. However, whether the magnitude of association is clinically relevant is unclear.
Funding acknowledgements: Supported by the São Paulo Research Foundation (grant nos. 2016/03826-5, 2015/17093-7, 2015/07704-9, 2015/02744-2, and 2014/14077-8)
Topic: Musculoskeletal: spine
Ethics approval: Approved by the Ethics Committee FCT/UNESP (CAAE:36285414.4.0000.5402), Brazil.
All authors, affiliations and abstracts have been published as submitted.
