Does movement change when low back pain changes? A systematic review

Key points

1. 01A relationship between spinal movement change and pain or disability change was found in only 31% of the 65 associations tested across 27 studies.
2. 02When a relationship was found, increased range of motion, velocity, or flexion relaxation of the back extensors was consistently associated with improvement (93% of cases).
3. 03Movement change alone does not appear necessary for pain or activity limitation to improve in most people with non-specific low back pain.
4. 0493% of the included studies were rated high risk of bias, mainly due to lack of reported assessor blinding.
5. 05The overall GRADE quality of evidence was rated low, limiting confidence in the findings.

How it was conducted

Design

Aetiology systematic review (PROSPERO CRD42017064436)

Databases

Medline, EMBASE, CINAHL, AMED searched from inception to January 2020

Participants

2739 participants across 27 included studies

Included studies

27 studies (7 used data from an RCT); both cohort and RCT designs analysed as cohort studies

Movement measures

Lumbopelvic kinematics, lumbar flexion/extension ROM, velocity, flexion relaxation, and muscle activity measured with instrumented tools

Data synthesis

Vote counting used due to large heterogeneity precluding meta-analysis

What they found

• A relationship between change in movement and change in pain was found in 22% (5.6/26) of the 26 associations investigated in 16 studies.
• A relationship between change in movement and change in activity limitation was found in 37% (14.4/39) of the 39 associations investigated in 23 studies.
• Combining pain and activity limitation, a relationship was observed in 31% (20.0/65) of all 65 associations tested.
• When a relationship was observed, increased movement range, velocity, or flexion relaxation was associated with improved clinical outcome 93% (18.5/20.0) of the time.
• 25 of 27 studies (93%) were categorised as high risk of bias, primarily due to uncertainty about assessor blinding.
• Inter-rater agreement on quality assessment items ranged from 92.3% to 100% (median 97.2%, IQR 95.9%).

Limitations

• 93% of studies were high risk of bias, mainly due to absent or unreported assessor blinding, preventing a low-risk-only sensitivity analysis.
• Substantial heterogeneity in movement parameters, measurement tools, interventions, and analysis techniques precluded meta-analysis.
• Movement parameters measured were often not individually tailored to participants, potentially underestimating true relationships.
• Only English-language, peer-reviewed studies were included, introducing possible language and publication bias.

Why it matters

For patients

Most people with non-specific low back pain do not need to change the way they move in order to experience less pain or improved function.

For clinicians

Targeting movement change as the primary mechanism of treatment may not be warranted for most patients, and a multidimensional, individualised, biopsychosocial approach is better supported by the evidence.

For readers

This systematic review challenges movement-focused treatment models for low back pain by showing movement change and pain improvement rarely co-occur at the individual level.