Reducing low-value imaging for low back pain : Systematic review with meta-analysis

Objective
To examine the effectiveness of implementing interventions to improve guideline-recommended imaging referrals in low back pain.

Design
Systematic review with meta-analysis.

Literature Search
We searched MEDLINE, EMBASE, the Cumulative Index to Nursing and Allied Health Literature, Web of Science Core Collection, and the Cochrane Central Register of Controlled Trials from inception to June 14, 2021, as well as Google Scholar and reference lists of relevant systematic reviews published in the last 10 years. We conducted forward and backward citation tracking.

Study Selection Criteria
Randomized controlled or clinical trials in adults with low back pain to improve imaging referrals.

Data Synthesis
Bias was assessed using the Cochrane Risk of Bias 2 tool. Data were synthesized using narrative synthesis and random-effects meta-analysis (Hartung-Knapp-Sidik-Jonkman method). We assessed the certainty of evidence using the Grading of Recommendations Assessment, Development and Evaluation approach.

Results
Of the 2719 identified records, 8 trials were included, with 6 studies eligible for meta-analysis (participants: N = 170 460). All trials incorporated clinician education; 4 included audit and/or feedback components. Comparators were no-intervention control and passive dissemination of guidelines. Five trials were rated as low risk of bias, and 2 trials were rated as having some concerns. There was low-certainty evidence that implementing interventions to improve guideline-recommended imaging referrals had no effect (odds ratio [95% confidence interval]: 0.87 [0.72, 1.05]; I2 = 0%; studies: n = 6). The main finding was robust to sensitivity analyses.

Conclusion
We found low-certainty evidence that interventions to reduce imaging referrals or use in low back pain had no effect. Education interventions are unlikely to be effective. Organizational- and policy-level interventions are more likely to be effective.