Management of non-specific low back pain in primary care - A systematic overview of recommendations from international evidence-based guidelines.

Aim: Systematic identification, characterization and analysis of recommendations concerning the diagnosis and treatment of non-specific low back pain (LBP) in primary care provided in international evidence-based guidelines from high-income countries.
Background: LBP is one of the most common reasons for consulting a primary care physician and its prevalence is higher in high-income than in middle- or low-income countries. The majority of LBP is non-specific and treatment recommendations are not often based on high-quality and patient-oriented evidence.
Methods: We systematically searched PubMed and major guideline databases from 2013 to 2020. Two independent reviewers performed literature selection and the quality assessment of included guidelines using the AGREE II tool. We extracted all relevant recommendations including the corresponding Grade of Recommendation. We grouped all included recommendations by topic and compared them to each other.
Findings: This overview includes 10 current guidelines and overall 549 relevant recommendations. Recommendations covered aspects of assessment and diagnosis (15%), non-pharmacological interventions (46%), pharmacological interventions (26%), invasive treatments (8%) and multimodal pain management (5%). In total, 30% of all recommendations were strong and 57% weak or very weak. The proportion of recommendations for and against an intervention was 45% and 38%, respectively. The recommendations from the different guidelines were largely in good agreement. We identify only a small number of contradictory recommendations, mostly dealing with very specific interventions.
Conclusion: In conclusion, current evidence-based guidelines published in high-income countries provide recommendations for all major aspects of the management of people with LBP in primary care. Recommendations from different guidelines were largely consistent. More than 50% of these recommendations were weak or very weak and a high proportion of recommendation advised against an intervention.