Predictive validity of the STarT Back screening tool among older adults with back pain

Abstract Background The predictive validity of the STarT Back screening tool among older adults is uncertain. This study aimed to assess the predictive validity of the SBT among older adults in primary care. Methods This prospective cohort study included 452 patients aged ≥55 years seeking Norwegian primary care with a new episode of back pain. A poor outcome (persistent disabling back pain) was defined as a score of ≥7/24 on the Roland‐Morris Disability Questionnaire (RMDQ) at 3, 6 or 12 months of follow‐up. The ability of the SBT risk groups to identify persistent disabling back pain was assessed with multivariable logistic regression, area under receiver operating characteristics curve (AUC), and with the accuracy measures sensitivity, specificity, predictive values and likelihood ratios. Results The adjusted odds ratios (95% CI) for persistent disabling back pain were 2.40 (1.34–4.30) at 3 months, 3.42 (1.76–6.67) at 6 months and 2.81 (1.47–5.38) at 12 months for the medium‐risk group ( n = 118), and 8.90 (1.83–43.24), 2.66 (0.81–8.67) and 4.53 (1.24–16.46) for the high‐risk group ( n = 27), compared to the low‐risk group ( n = 282). There were no statistically significant differences in odds between the medium‐ and high‐risk groups at any time point. AUC values (95% CI) were 0.65 (0.59–0.71), 0.67 (0.60–0.73) and 0.65 (0.58–0.71) at 3, 6 and 12 months. Accuracy measures were poor at all time points, with particularly poor sensitivity and negative likelihood ratio values. Conclusion The predictive validity of the SBT risk groups in predicting persistent disabling back pain in older adults was poor. Significance Statement This study found that the STarT Back screening tool had poor predictive validity among older adults and that it may need recalibration or extension before widespread implementation among older adults. Having valid tools for this population may aid clinicians with allocating scarce healthcare resources, which is especially important considering the rapidly ageing population and its expected challenge to the healthcare systems.