Roger Chou, MD
Is risk-stratified primary care management of low back pain more effective than nonstratified current best practice? Do effects differ in risk subgroups?
Randomized controlled trial (Sub-grouping for Targeted Treatment [STarT Back] trial). Current Controlled Trials ISRCTN37113406.
Blinded (data collectors).*
4 and 12 months.
10 general practices in the Keele General Practice Research Partnership, England, UK.
851 patients ≥ 18 years of age (mean age 50 y, 59% women) who consulted their physician about back pain of any duration. Exclusion criteria included spinal surgery in the past 6 months, receipt of back treatments outside of primary care, pregnancy, and serious or potentially serious illness or comorbid conditions.
Stratified management with an initial 30-minute protocol-directed physiotherapy assessment and treatment session followed by referral for further physiotherapy according to risk for poor prognosis (low, medium, or high) assessed using the 9-item Keele STarT Back Screening Tool (stratified group, n = 568), or current best practice with an initial 30-minute nonprotocolized physiotherapy assessment and treatment session followed by referral according to physiotherapist clinical judgment (control group, n = 283). In the stratified group, low-risk patients received only the initial physiotherapy session, medium-risk patients were referred for standardized physiotherapy, and high-risk patients were referred for psychologically informed physiotherapy.
Roland and Morris Disability Questionnaire (RMDQ) score and adverse events.
81% at 4 months (intention-to-treat analysis). 12-month follow-up was < 80% and is not reported here.
At 4 months, more patients in the stratified group than in the control group had ≥ 30% improvement on RMDQ score overall and in medium-risk and high-risk subgroups; groups did not differ in the low-risk subgroup of patients (Table). No adverse events were reported in either group.
Risk-stratified primary care management reduced disability more than nonstratified current best practice in patients with low back pain.
†RMDQ = Roland and Morris Disability Questionnaire; other abbreviations defined in Glossary. RBI and CI calculated from data in article.
‡Based on a mean of 5 imputed data sets.
§Adjusted for age, sex, baseline RMDQ score, and back pain duration.
||Good outcome defined as ≥ 30% change from baseline on RMDQ score.
Chou R. Risk-stratified primary care management of low back pain reduced disability. Ann Intern Med. ;156:JC2–7. doi: 10.7326/0003-4819-156-4-201202210-02007
Download citation file:
Published: Ann Intern Med. 2012;156(4):JC2-7.
Back Pain, Rheumatology.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use