Diagnostic clinical prediction rules for categorising low back pain: A systematic review

Low back pain remains a challenging condition due to the difficulty in identifying specific diagnoses. While clinical prediction rules are recognized for supporting decision-making, their development and validation status for low back pain is unclear. Prior research has shown that some diagnostic tools are in early stages, but recent progress is unknown. This gap motivated a new systematic review to assess the current state of diagnostic clinical prediction rules for low back pain. No prior work had resolved the full scope of these tools' readiness for clinical use. The review aimed to clarify which tools are externally validated and which remain in derivation phases. This uncertainty drove the need for an updated evaluation of the literature. The lack of impact analysis for some tools raises questions about their practical utility. This review provides a synthesis of the available evidence to guide future development and implementation.

The study aimed to evaluate the current status of diagnostic clinical prediction rules for low back pain. It focused on identifying new and previously known tools and assessing their development stages. The motivation stemmed from the need to understand which tools are ready for implementation. The specific problem addressed was the lack of recent data on these diagnostic tools. The review sought to determine whether any tools have progressed beyond derivation phases. The researchers aimed to clarify the readiness of these tools for clinical use. This included evaluating their methodological quality and risk of bias. The study's goal was to inform future research and funding priorities in this area.

A systematic review was conducted across five databases, including Medline and CINAHL. The search was supplemented with hand-searching and citation tracking. Eligible studies were selected based on predefined criteria. Quality assessment tools were used to evaluate risk of bias and methodological standards. The Quality Assessment of Diagnostic Accuracy Studies-2 was applied to assess study quality. A checklist was used to evaluate the methodological quality of the CPRs. The GRade and ASsess Predictive tool was used for CPR appraisal. The review included 11 diagnostic CPRs identified from 5021 screened studies.

Eleven diagnostic CPRs were identified, including six previously known and five new tools. Three of the previously known CPRs have been externally validated. Five new CPRs remain in the derivation phase. The most validated CPRs include the Lumbar-Spinal-Stenosis-Self-Administered-Self-Reported-History-Questionnaire. Another validated tool is the Diagnosis-Support-Tool-to-Identify-Lumbar-Spinal-Stenosis. The StEP-tool differentiates radicular from axial low back pain. These tools have not yet undergone impact analysis. Their use should be approached with caution until further evaluation is completed.

The updated review found five new diagnostic CPRs for low back pain, all in early development stages. Three previously known CPRs have been externally validated but require impact analysis. The authors suggest that future funding should prioritize external validation and impact assessment. These findings highlight the need for further research on existing CPRs. The current evidence does not support widespread implementation of these tools. The review provides a foundation for guiding future development efforts. The authors propose that impact studies are necessary before clinical adoption. This synthesis informs the next steps for improving diagnostic accuracy in low back pain.