Introduction

Identifying subgroups that respond to targeted treatments is a key objective for low back pain (LBP) research.  The Keele STarT Back Screening Tool (SBT) assigns patients to one of three subgroups, according to the risk of persistent disability (low risk, medium risk, high risk).  Several items in the SBT were chosen because they measured psychosocial domains that were thought to be modifiable in primary care.  In primary care settings the SBT  successfully predicts long term disability, and targeting treatment to the risk subgroup has been shown to reduce disability.  There is much less information on it’s performance in specialty clinics. The objectives of our study were to a) determine whether SBT can predict change in disability after referral to a low back pain specialty clinic and b) determine if incorporating features not included in the SBT may improve ability to predict disability.

Methods 

This is a retrospective observational cohort study involving LBP patients referred by primary care providers and seen in a US chronic LBP (cLBP) specialty clinic.  Patients referred to the clinic are seen for back-to-back appointments with a physical therapist and a physician, who formulate a joint treatment plan.  Patients receive consistent messaging about spinal pain from their providers, organized around the principles of pain neuroscience education.  There are no standardized treatment pathways, but patients who fall into the SBT high-risk subgroup are discussed at monthly multi-disciplinary case conferences so that progress is closely monitored and treatment adjusted as indicated.  SBT and covariates were measured at baseline.  Outcome measures were pain intensity (VAS), PROMIS-10 global physical health (PH) and PROMIS-10 global mental health (PH).  VAS, PH, and MH were measured at baseline and at pragmatic timepoints following.  Multivariable linear regression was used to identify correlations between SBT and covariates with changes in PH and MH.

Results: 

241 patients were followed for a mean of 17.0±7.5 months. Baseline pain was 6.6 (SD 2.1), PROMIS-global MH score was 44.4 (SD 9.6), and PH score was 38.5 (SD 8.6). 29.7% were low-risk on the SBT, 41.8% were medium-risk, and 28.5% were high-risk. Mean change in MH and PH scores were 0.8 (SD 8.11) and 2.39 (SD 7.52) respectively. High-risk SBT positively predicted change in MH and both medium- and high-risk status predicted change in PH (Table 1). Additional variables that predicted change in MH include chronic overlapping pain conditions (COPC) and in PH included every-day smoker status, pain down buttock/thigh, Charlson Comorbidity Index (CCI) score, COPCs, and recent injections.

Discussion

The SBT predicts change in dysfunction following treatment in a cLBP specialty clinic, but including other variables, most notably COPC’s, improves performance.  Effort is needed to identify other factors, even if nonmodifiable, that can help stratify risk in cLBP patients.