RATIONALE: Although posture is a known contributor to musculoskeletal pain, efforts to improve posture and reduce pain have achieved mixed results. BBF has potential for retraining posture, but its use has not been tested in relation to occupational performance. This project aims to evaluate the effectiveness of real-time BBF to retrain habitual, in-task posture and reduce pain in computer users.

DESIGN: A two-way independent-groups matched-pairs, block design with the pairs randomly allocated to one of two treatment groups and 12-wk follow-up.

PARTICIPANTS: We recruited 42 computer-based workers with recent and ongoing neck, shoulder, or upper back pain from staff at the university and the local municipal council.

DATA TOOLS: Pain in 10 body regions (defined by a modified body chart from the Nordic questionnaire) was assessed using Numerical Rating Scales (0 = no pain, 10 = severe pain). Posture was assessed at participants’ usual work station while they performed their usual work tasks using in-task lateral marker imaging. Images were drawn at 2-min intervals from 16-min videos. Posture angles were measured from lines drawn between key anatomical reference points on the images and averaged for each data-gathering episode.

METHOD: After participants were initially assessed (T0), they were matched to pairs and randomly allocated to one of two 3-wk equivalent-intensity interventions. One group received a workplace-based postural education program. The other group received a real-time, biomechanically based, postural biofeedback program utilizing the BackTone® biofeedback device. Participants were reevaluated at end of intervention (T1) and at 12-wk follow-up (T2).

ANALYSIS: Within- and between-group improvements in posture and pain scores between T0 and T1 and between T0 and T2 were examined using the Wilcoxon signed-rank test. Chi-square analyses were applied to comparisons of the number of participants in each group who completely resolved their combined neck, shoulder, and upper back pain at T1 and T2.

RESULTS: We recruited 42 computer users with neck, shoulder, or upper back pain who were primarily middle aged and female. Participants in the BBF group achieved significant improvements in upper thoracic and cervical spine angles at T1 (p = .034 and p = .018, respectively), and the improvements persisted at T2 (p = .041 and p = .020, respectively). Participants in the education group achieved improvement in upper thoracic angle at T1 only (p = .008); the improvement did not persist at T2. There was no improvement in education group cervical angles.

Average biofeedback group pain improvement (63%, standard deviation [SD] = 55.95%) was greater than average education group pain improvement (35.6%, SD = 38.6%) at T1 (p = .022) and at T2 (BBF 78.1%, SD = 57.6%; ED 40.3%, SD = 50.5%; p = .048). Thus, BBF was more effective and sustained than education.

DISCUSSION: The results support our hypotheses that BBF is more effective than postural education in improving posture and musculoskeletal pain in computer users. Inclusion of real-time BBF will enhance outcomes for treatment of computer users with musculoskeletal pain.

IMPACT STATEMENT: These findings provide support for occupational therapists to incorporate BBF as an effective tool for achieving sustained improvement of habitual posture in the context of reducing symptoms and risk of upper body musculoskeletal problems.