Prognostic factors in the assessment of low back pain: a systematic review and evidence gap map


This review synthesises prospective prognostic studies of short -, mid- and long-term disability in people with LBP to inform treatment delivery.


PubMed, Medline, Web of Science and Embase were searched for studies in English, from inception to December 2019. Risk of bias was assessed using the QUIPs (Quality In Prognosis Studies) tool as recommended by the PROGRESS series on prognosis studies. Data from the included studies was extracted: first author and year of publication; study type and follow-up duration; number and mean age of participants; setting; LBP diagnosis; outcome measures; and results concerning prognostic factors. Pooling and meta-analysis was performed where possible. The results were collated into an Evidence Gap Map (EGM).


Results: Included studies were prospective cohort studies with at least one follow-up. Timeframes ranged from seven days to five years. Studies varied in types of LBP, from simple through to chronic or nerve-related LBP. The heterogeneity of LBP presentations was accepted to reflect clinical practice. Participants’ ages ranged from 21 to 64. The main outcome of interest was disability. High scores on work-specific Fear Avoidance Beliefs Questionnaire (OR 1.69; CI, 1.10-2.59) and the presence of neurological signs (OR 2.87; CI, 1.37-6.03) were negatively associated with long-term recovery after meta-analysis. Having completed higher education was positively associated with long-term recovery (OR 0.73; CI, 0.53-1.00), but had moderate evidence of not being prognostic in the mid-term (OR 0.95; CI, 0.49-1.84), as did depression (OR 1.04; CI, 0.58-1.87) and episodic LBP (OR 0.67; CI, 0.22, 2.04). Single studies identified that employment status, physical characteristics and psychological factors are associated with long-term disability. In the short-term, there was only weak evidence that any of the factors considered were prognostic.

Conclusion(s): This systematic review synthesised high and moderate quality evidence to determine prognostic factors for recovery from disabling LBP. The EGM grouped factors by domain and time period, highlighting available evidence and gaps in knowledge. Factors from six domains that predict long-term disability have been identified; clinical characteristics, psychological factors and employment related factors can predict long-term non-recovery from disabling LBP and therefore should be considered in clinical practice. There are limitations of the identified factors, from quality of the reported studies and appearance in single studies. The EGM highlighted that further research is required for prognostic factors related to short-term outcomes, since most studies focussed on mid to long-term follow up.

Cost and savings

No further data 


The EGM gives clinicians a steer on which factors have good evidence for consideration when assessing prognosis for LBP. The EGM highlights areas where high quality research is absent, notably prognostic factors for short term outcomes. Non-modifiable factors, such as demographics or employment status may be considered by clinicians when assessing prognosis.

Top three learning points

1) There is very little research focussing on short term predictors of outcome in low back pain therefore predictors should be used with caution in an acute presentation

2) Non-modifiable factors such as sex or employment status may be considered as predictors when assessing patients

3) Further high quality research is required to determine further predictors; a significant number of studies were excluded from this analysis due to risk of bias

Funding acknowledgements

This report is independent research arising from a Pre-Doctoral Research Fellowship, Adele Hill, ICA-PCAF-2018-01-137 supported by the National Institute for Health Research. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health.

Additional notes

For further information on this and related projects, please see:

MSK AI Research:

Back pain decision support app: