Abstract
Background
The impact of physical therapy on the outcomes of patients with acute low back pain (LBP) stratified by the STart Back Screening Tool (SBST) is unclear.
Objective
The purpose of this study was to compare the outcomes of patients with acute LBP who were stratified as medium or high risk.
Design
This was a secondary analysis of a randomized trial.
Setting
Patients were recruited between March 2011 and November 2013 from primary care clinics in Salt Lake City, Utah.
Participants
One hundred eighty-one participants with acute LBP who were stratified as medium risk (n = 120) or high risk (n = 61) by the SBST were included. They were aged 18 through 60 years, with duration of symptoms less than 16 days, no symptoms below the knee, no treatment for LBP in the past 6 months, and an Oswestry Disability Index (ODI) score of 20% or greater.
Intervention
After participants received education on how to manage their LBP, they were randomized to receive usual care (n = 97) by their primary care provider or early intervention (n = 84) by a physical therapist.
Measurements
The primary (3-month ODI score) outcome measure was obtained at baseline and at 4 weeks, 3 months, and 1 year.
Results
No differences were detected in the effect of intervention between participants stratified as medium or high risk. For the high-risk subgroup, there was a significant difference between the early intervention and usual care groups for the 3-month ODI (mean difference = –5.87 [95% CI = –11.24, –0.50]) favoring early intervention.
Limitations
The primary study was not designed to examine the SBST.
Conclusions
Patients with acute LBP stratified as high risk seem likely to respond well to one session of education. They may experience additional benefit by 3 months from evidence-based physical therapy treatments. These effects disappear at 1 year.
More than a quarter of patients who consult a primary care provider for acute low back pain (LBP) will continue to have pain and disability at 2 years,1 underscoring the challenge in providing effective interventions to patients with acute LBP. Although a sizable proportion of patients transition to chronic disablement, 1 in 3 patients with acute LBP experiences a degree of recovery sufficient to avoid persistent pain and disability.2 Among the top primary care research priorities are identifying prognostic subgroups of patients with LBP and identifying specific management strategies based on relevant subgroups. More appropriate interventions can then be targeted to patients at increased risk of chronicity while avoiding overtreatment of patients with a good prognosis for recovery.3
The STart Back Screening Tool (SBST) is a decision aid that stratifies patients into prognostic subgroups based on their risk for developing future disabling LBP and makes treatment recommendations based on their risk stratification.4 Compared with patients stratified as low and medium risk, those at high risk have a greater number of psychosocial factors associated with a poor prognosis,5,6 putting them at risk for greater utilization of health care resources.7,8 Recommended interventions common to all SBST risk groups include education to expect a favorable prognosis and advice on optimal levels of activity. Additional treatments for the high-risk group include “psychologically informed” interventions designed to mitigate psychosocial barriers to recovery and evidence-based physical therapy to address impairments in physical function. Patients at medium risk receive only evidence-based physical therapy to address impairments, whereas those at low risk receive guideline-recommended advice on self-management, including education to expect a favorable prognosis and pain reduction strategies.4 Recent research indicates that targeting interventions toward patients with LBP based on their baseline SBST prognostic risk stratification may alter their recovery and reduce future health care utilization and work loss.9,10
Since the initial description and validation of the SBST, there has been considerable attention directed to the high-risk group and defining the parameters of psychologically informed practice for these patients.11 A cognitive-behavioral approach with attention to both psychosocial and biomedical factors is recommended,12 but further research is needed to optimize specific content. In particular, it is unclear if psychologically informed treatment should be further tailored to patients within the SBST high-risk group based on additional characteristics such as duration of current symptoms. The SBST was validated,9 and subsequently implemented,10 in patients with predominately subacute or chronic symptoms; approximately 20% of participants in these studies had symptoms less than 30 days in duration.
For patients with acute, nonspecific LBP, evidence-based physical therapy includes patient education on favorable prognosis and the importance of remaining active, spinal manipulation, and trunk strengthening exercises.13–16 A physical therapy program focused on these interventions, therefore, should be appropriate for patients with acute LBP in the SBST medium-risk category, but it is unclear if these interventions are sufficiently psychologically informed for patients in the SBST high-risk category.
The association between the SBST risk stratification and recovery specifically in patients with acute LBP has not been examined. We performed a randomized clinical trial comparing early intervention with evidence-based physical therapy and education with usual care management by a primary care provider in patients with acute LBP,16 providing an opportunity to examine this question in a secondary analysis. The aim of this secondary analysis was to compare the 4-week and 3- and 12-month pain and disability outcomes of patients with acute LBP who were stratified as medium or high risk by the SBST. The 3-month disability outcomes were our main interest. We presumed our evidence-based physical therapy intervention would not be sufficiently psychologically informed for study participants in the high-risk category. Therefore, our primary hypothesis was that the effect of early intervention with evidenced-based physical therapy compared with education with usual care would be greater in the medium-risk group than it would in the high-risk group at 4 weeks, 3 months, and 1 year.
Method
Design Overview
We performed a post hoc secondary analysis of data from a randomized clinical trial (ClinicalTrials.gov Identifier: NCT01726803), comparing the outcomes of patients with a new episode of acute LBP who received either usual care or early intervention.16 The analysis for the present study was not included in the original protocol for the randomized trial. Data from the SBST were available, and that allowed us to study the proposed hypothesis.
Setting and Participants
Participants were recruited from Intermountain Healthcare and University of Utah primary care clinics in Salt Lake City, Utah, between February 2011 and November 2013. Patients who consulted a primary care provider for a new onset of LBP were screened for inclusion. To be eligible, patients were required to have a modified Oswestry Disability Index (ODI)17 score of ≥20%, be aged 18 to 60 years, and have a complaint of LBP or numbness between the 12th rib and buttocks with or without symptoms into one or both legs originating from the lumbar region. Additionally, participants were required to meet 2 criteria previously described to identify patients likely to respond favorably to the protocol used for the early intervention: (1) no LBP extending below the knee and (2) LBP symptom duration <16 days.18 Patients were excluded if they had any of the following: surgery to the lumbosacral spine, current pregnancy, currently receiving treatment for LBP from another health care provider (eg, chiropractor, massage therapist, injections), the presence of neurological symptoms (positive straight leg raise or crossed straight leg raise that reproduced symptoms at <45° or reflex, sensory, or strength deficits indicating lumbar nerve root compression), the presence of “red flags” for a potentially serious condition (cauda equina syndrome, major or rapidly progressing neurological deficit, cancer, fracture, infection, or systemic disease). All participants provided informed consent prior to the study.
Outcomes and Follow-up
After providing informed consent and prior to randomization, all participants provided demographic information and underwent baseline assessment by 1 of 3 physical therapists, each with greater than 15 years of experience. One of these physical therapists performed approximately 90% of the baseline examinations. The SBST was used to categorize patients as high, medium, or low risk for prolonged disability. The SBST is a 9-item multidimensional questionnaire comprising disability and psychosocial subscales. Individual item responses are either positive or negative, and positive responses are summed and used to stratify patients based on published algorithms.4 For this analysis, we included participants categorized as medium or high risk. The SBST has good reliability and validity.4 We eliminated patients stratified as low risk because decision making based on the SBST suggests these patients do not require physical therapist management beyond education,9 and we were interested in evaluating the effects of the early physical therapy treatment.
The primary trial outcome measure was the ODI, a commonly used LBP-related disability measure.19 The ODI is a 10-item questionnaire, with each item scored 0 to 5 points, and the items scores are summed and doubled to produce a final score ranging from 0 to 100. Higher scores indicate greater disability. The ODI has been shown to be a valid and reliable instrument.20 Secondary outcome measures included the Numeric Pain Rating Scale (NPRS). The NPRS is a 0 to 10 scale, with 0 indicating no pain and 10 indicating maximal pain.13–15 The NPRS has good test-retest reliability,21 with a minimal clinically important difference of 2 points among patients with acute LBP.22
We collected several questionnaires at baseline whose constructs also are assessed on the SBST. The Pain Catastrophizing Scale (PCS) is a 13-item scale measuring the extent to which an individual catastrophizes in response to pain. Higher scores indicate greater pain catastrophizing.23 The Fear-Avoidance Beliefs Questionnaire (FABQ) was used to assess fear of pain and beliefs regarding activity avoidance. The FABQ has a 7-item work subscale and a 4-item physical activity subscale. Higher scores indicate greater fear-avoidance beliefs.24
The SBST, FABQ, and PCS scores were collected at baseline, and data for our primary and secondary outcome measures (the ODI and NPRS, respectively) were collected at baseline and at 4-week, 3-month, and 12-month follow-up visits.
Blinding
The nature of the study did not allow for blinding of participants. Group allocation was revealed to the study personnel only after the participant completed all baseline procedures, including the SBST. Research study personnel conducting the 4-week follow-up examination were blinded to treatment group allocation. Participants were reminded to not reveal their treatment group at the follow-up examination.
Randomization and Interventions
Prior to the study, a research assistant prepared a computer-generated random allocation sequence using block sizes of 4. From the randomization allocation, opaque, sequentially numbered envelopes were prepared containing the treatment group assigned for each participant. After the participant completed the baseline examination and education session, the study coordinator opened the envelope, and the participant was assigned to a treatment group.
Prior to randomization and after the baseline assessment, an investigator who was also a licensed physical therapist provided all participants with standard education based on the Back Book.25 Among the key messages was that they did not have a serious medical condition and that acute LBP has a favorable natural history. Participants were advised to avoid bed rest, stay active, and return to normal activities as soon as possible. They were advised that they should take a positive and active approach to their recovery and not wait for someone else to take away their pain. All participants were provided copies of the Back Book, and any questions were answered. The education session lasted approximately 20 minutes.
All participants were encouraged to follow up with their primary care provider on an as-needed basis. Participants randomized to the early intervention group were then scheduled to begin physical therapy treatment within 3 days of the baseline assessment. Participants in the early intervention group received 4 physical therapy sessions over 3 weeks that included spinal manipulation, range-of-motion exercises, and spinal strengthening exercises. The usual care group received no additional intervention. Details of the treatment protocol are available elsewhere.16
Sample Size
The sample size for the original trial was based on detecting a difference of 7 points (SD = 16) on the ODI or a moderate effect size of 0.44, resulting in a sample size of 220. Because we only examined the SBST’s medium-and high-risk groups, our sample size was less than that of the original trial. The secondary sample sizes were n = 97 in the usual care group (medium risk = 61, high risk = 36) and n = 84 in the early intervention group (medium risk = 59, high risk = 25), for a sample size of 181.
Data Analysis
Descriptive statistics were used to characterize the sample. Data were screened for normality and homogeneity of variance, as well as for missing values. Missing data were imputed using the multivariate imputation by chained equations (MICE) method of multiple variable imputation.26 Ten imputed data sets were generated using available primary and secondary outcome scores, treatment group, employment status, sex, age, marital status, education, prior history of LBP, smoking, race, and expectations of benefit from exercise and spinal manipulation. Only participants who completed the SBST and were stratified as either medium or high risk were included in the analyses.
We tested our primary hypothesis using a 2-way analysis of covariance. The hypothesis of interest pertained to the presence of an interaction between treatment group (early intervention or usual care) and SBST category (high or medium risk). Our analysis used the 3-month ODI scores as the dependent variable while controlling for baseline ODI score. Similar analyses were conducted for the 4-week and 12-month time points and for all time points examining the NPRS scores while controlling for baseline NPRS score and were considered secondary. To examine the effect of treatment separately within the medium- and high-risk subgroups, we performed an analysis of covariance comparing the ODI score from baseline to 4-week, 3-month, and 12-month time points while controlling for baseline ODI score. The same approach was used when examining the effect of treatment on NPRS. All analyses used a 2-sided alpha of .05, and no adjustments were made for multiple comparisons. Analyses were conducted using STATA statistical software version 14 (StataCorp LP, College Station, Texas).
Role of the Funding Source
This was a secondary analysis of data from a published clinical trial (NCT01726803) that was funded by the Agency for Healthcare Research and Quality (AHRQ) (grant no. R18HS018672).16 The authors received no funding from AHRQ for this study. This investigation was supported in part by funding from the National Center for Research Resources and the National Center for Advancing Translational Sciences Grant 5UL1TR001067-02.
Results
Of the 220 participants in the primary study, 219 completed the SBST. Thirty-eight participants were stratified as low risk and were not included in this secondary analysis. The final sample (n = 181) had 120 (66%) participants stratified as medium risk and 61 (34%) stratified as high risk. The average age of the sample was 37.5 (SD = 10.2) years, with 52.1% being female. Participants reported moderate pain intensity and disability, and the majority had at least one prior episode of LBP (Tab. 1). The Figure contains the number of patients with complete follow-up data for our primary outcome measure at each time point.
Table 1.
Descriptive Characteristics of the Study Samplea
| Medium-Risk Group (n = 120) | High-Risk Group (n = 61) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Demographicsb | All Participants (N = 181) | Early Intervention Group (n = 59) | Usual Care Group (n = 36) | Early Intervention Group (n = 25) | Usual Care Group (n = 36) | |||||
| Age (y) | 37.5 | (10.2) | 39.9 | (9.9) | 37.1 | (10.4) | 35.1 | (11.3) | 36.1 | (8.9) |
| Sex, male | 92 | (50.8%) | 33 | (56.0%) | 32 | (52.5%) | 14 | (56.0%) | 13 | (36.0%) |
| BMI (kg/m2) | 29.1 | (8.0) | 29.6 | (6.9) | 28.8 | (9.0) | 28.3 | (8.0) | 29.2 | (8.3) |
| Oswestry Disability Index | 42.9 | (13.0) | 44.0 | (13.5) | 38.2 | (11.9) | 45.4 | (14.3) | 47.2 | (11.1) |
| Average LBP | 5.4 | (1.8) | 5.2 | (1.7) | 4.7 | (1.7) | 6.3 | (1.7) | 6.1 | (1.6) |
| Prior history LBP | 124 | (68.5%) | 43 | (72.9%) | 40 | (65.6%) | 18 | (72.0%) | 23 | (63.9%) |
| Currently taking medicine for pain | 156 | (86.1%) | 54 | (91.5%) | 51 | (83.6%) | 22 | (88.0%) | 29 | (80.5%) |
| Pain Catastrophizing Scale | 15.2 | (10.5) | 13.5 | (9.7) | 12.3 | (8.7) | 20.4 | (12.3) | 19.1 | (11.1) |
| Fear-Avoidance Beliefs Questionnaire–Work | 12.5 | (9.6) | 11.1 | (8.4) | 10.7 | (9.1) | 13.3 | (11.1) | 16.1 | (10.8) |
| Fear-Avoidance Beliefs Questionnaire–Physical Activity | 16.0 | (4.4) | 15.3 | (4.3) | 15.3 | (3.9) | 16.5 | (4.8) | 17.5 | (4.7) |
| History anxiety or depression | 50 | (27.6%) | 18 | (30.5%) | 16 | (26.2%) | 4 | (16.0%) | 12 | (33.3%) |
| Upper back or neck pain | 70 | (38.7%) | 32 | (54.2%) | 21 | (34.4%) | 5 | (20.0%) | 12 | (33.3%) |
| Current smoker | 15 | (8.3%) | 7 | (11.9%) | 5 | (8.2%) | 2 | (8.0%) | 1 | (2.8) |
| High school education | 84 | (46.4%) | 31 | (52.5%) | 27 | (44.3%) | 11 | (44.0%) | 15 | (41.7%) |
| Employed outside the home | 147 | (82.3%) | 48 | (84.8%) | 50 | (82.8%) | 29 | (80.0%) | 29 | (80.1%) |
BMI = body mass index, LBP = low back pain.
Continuous variables are reported as mean (SD), and categorical variables are reported as n (%).
Figure.
Flow of participants through the study. ODI = Oswestry Disability Index, SBST = STart Back Screening Tool.
For our primary (ODI) and secondary (NPRS) outcome measures, there were no significant 2-way interactions between SBST risk subgroups and treatment group for the 4-week, 3-month, and 1-year time periods, indicating no difference in treatment effects between medium- and high-risk subgroups over time (Tab. 2).
Table 2.
Results of Analyses Comparing Outcomes Between Treatment Groups Subgrouped by Risk Stratificationa
| Measure | Medium-Risk Group | High-Risk Group | |||||
|---|---|---|---|---|---|---|---|
| Early Intervention Group | Usual Care Group | Mean Within-Group Difference (95% CI) | Early Intervention Group | Usual Care Group | Mean Within-Group Difference (95% CI) | Mean Interaction Effect (95% CI) | |
| Oswestry Disability Indexb | |||||||
| Baseline | 44.0 | 38.2 | 45.4 | 47.2 | |||
| 4 wk | 10.3 | 12.8 | –2.44 (–7.03 to 2.15) | 13.6 | 17.5 | –3,88 (–10.39 to 2.63) | –1.44 (–9.40 to 6.52) |
| P=.29 | P=.24 | P=.72 | |||||
| 3 mo | 6.6 | 8.5 | –2.00 (–5.85 to 1.85) | 6.1 | 12.0 | –5.87 (–11.24 to –0.50) | –3.87 (–10.50 to 2.75) |
| P=.30 | P=.03c | P=.25 | |||||
| 1 y | 6.9 | 9.5 | –2.63 (–6.95 to 1.68) | 7.1 | 9.5 | –2.36 (–8.33 to 3.61) | –0.26 (–7.11 to 7.70) |
| P=.23 | P=.43 | P=.94 | |||||
| Numeric Pain Rating Scaled | |||||||
| Baseline | 5.2 | 4.7 | 6.3 | 6.1 | |||
| 4 wk | 1.6 | 1.9 | –0.30 (–0.91 to 0.31) | 1.6 | 2.5 | –0.87 (–1.75 to –0.01) | –0.57 (–1.63 to 0.52) |
| P=.34 | P=.05c | P=.89 | |||||
| 3 mo | 1.3 | 1.6 | –0.38 (–0.97 to 0.21) | 1.3 | 2.3 | –0.98 (–1.81 to –0.14) | –0.60 (–1.61 to 0.41) |
| P=.20 | P=.02c | P=.25 | |||||
| 1 y | 1.0 | 1.4 | –0.33 (–0.91 to 0.25) | 1.0 | 1.8 | –0.74 (–1.56 to 0.08) | –0.41 (–1.41 to 0.58) |
| P=.26 | P=.07 | P=.41 | |||||
Results are adjusted for the baseline measure of the outcome variable.
Missing Oswestry Disability Index scores: 0 at baseline, 5 at 4 wk, 5 at 3 mo, and 13 at 1 y; imputed using multiple imputation procedure.
Significant value.
Missing average low back pain scores: 3 at baseline, 3 at 4 wk, 5 at 3 mo, and 13 at 1 y, imputed using multiple imputation procedure.
There was a significant main effect of treatment group at 3 months in favor of early intervention for the ODI (mean difference = –3.31 [95% CI = –6.43, –0.18], P = .038) and at 3 months for the NPRS (mean difference = –0.58 [95% CI = –0.06, 0.09], P = .021), indicating an advantage for receiving early intervention regardless of SBST risk category. There was a significant main effect of SBST at 4 weeks for the ODI (mean difference = 4.01 [95% CI = 0.03, 7.99], P = .048) indicating that participants stratified as high risk had more disability than those stratified as medium risk at the 4-week follow-up.
When examining only the high-risk subgroup, there was a significant difference between the early intervention and usual care groups for the 3-month ODI and NPRS outcomes in favor of early intervention. No other differences were detected for any other time point for either the medium- or high-risk group (Tab. 2).
Discussion
We examined the effects of evidence-based physical therapy in patients with acute LBP based on their SBST stratification as medium or high risk. We found no differences in the effects of treatment over time on disability or pain between the medium- and high-risk groups. Consistent with the outcomes of the original trial,16 we found advantages for patients receiving early intervention involving evidence-based physical therapy when the effect of early intervention was evaluated in both medium- and high-risk groups combined.
Given the evidence for a beneficial effect of early intervention in the full cohort at 3 months, the absence of an interaction between the early intervention and SBST risk category in our secondary analyses can be viewed as consistent with an overall benefit of early intervention, which extends to both the medium- and high-risk SBST categories. However, in contrast to our original hypothesis, the point estimate of the effects of early intervention on the 3-month outcomes suggested an effect that approached a clinically meaningful difference of 6 points within the high-risk group for the ODI,27 but was smaller in the medium-risk group (Tab 2). The point estimate of the 3-month NPRS was larger in the high-risk group than in the medium-risk group but did not approach a clinically meaningful difference22 (Tab. 2). The upper endpoints for the 95% CI values comparing the treatment effects at 3 months between the 2 risk groups indicate that early intervention is unlikely to have more than a 2.75-point more favorable effect for the high-risk group versus the medium-risk group for the ODI or more than a 0.41-point more favorable effect on the NPRS (Tab. 2). Neither of these differences is of clinical significance,22,27 effectively ruling out clinically meaningfully stronger benefits of early intervention on these outcomes in the medium-risk group than in the high-risk group.
There are several potential explanations for our results. We presumed the early intervention management strategy of evidence-based physical therapy was appropriate for the medium-risk subgroup, but potentially suboptimal for the high-risk subgroup because it was not sufficiently psychologically informed. Perhaps the most straightforward explanation for our findings is that our evidence-based physical therapy treatment was sufficiently psychologically informed for patients with very acute, nonspecific LBP. Our intervention was a cognitive-behavioral strategy, although we provided only a single session of education and advice. It may be that this intervention combined with evidence-based physical therapy treatments was sufficient for even patients with acute LBP in the high-risk category. Our findings are consistent with the study by Burton and colleagues,25 who found that merely providing the Back Book to patients consulting primary care with acute or subacute LBP (<3 months’ duration) had a positive effect on their beliefs and outcomes. More research is needed to examine psychologically informed interventions in patients with acute LBP to determine the intensity of intervention required to achieve optimal outcomes.
An explanation for the lack of a detectable interaction effect is that the sample size for the original study was not calculated to detect an interaction between treatment group and the SBST. In addition, we did not stratify randomization to treatment groups by SBST category, creating unequal group sizes. Therefore, this secondary analysis was underpowered to detect an interaction effect. Further research should examine whether patients with acute LBP categorized as high risk on the SBST preferentially benefit from the physical therapy protocol used in this study.
An additional explanation for our findings may be the natural history of acute LBP. Participants in both the medium- and high-risk subgroups demonstrated rapid improvement in disability and reduction in pain from baseline to 4 weeks regardless of intervention, and these improvements were maintained at 1 year. This occurred in the high-risk subgroup despite greater baseline pain and disability in these patients. The rapid improvement among all patients in our study made it difficult to detect differences based on treatment or other factors.
Our inclusion of patients with very acute LBP (<16 days’ duration) certainly influenced our results. Comorbid psychosocial factors such as fear of movement28 and pain catastrophizing29 are known to accompany acute LBP. We found relatively high baseline levels of these constructs in our sample, with 61 participants (28% of our original sample) categorized as high risk on the SBST. The impact of these psychosocial factors on prognosis among patients with acute LBP may differ from their impact on individuals with chronic conditions.6 For example, recent systematic reviews indicate that fear-avoidance beliefs in patients with acute LBP are not prognostic of a poor outcome,30 but higher levels of catastrophizing are associated with a poor outcome in similar patients.29 In our sample of patients with very acute LBP, these psychosocial factors may represent a more transient response to acute pain and, therefore, may be more likely to diminish with time or in response to evidence-based treatments.
The results of this study should be considered in light of important limitations. This was not a preplanned secondary subgroup analysis, which increased the chance of a type 1 error, nor was the SBST used as a stratification factor at randomization, which may have created unbalanced groups and influenced the results.31 We provided the education intervention to all patients; thus, we cannot determine the unique impact of the educational component of our intervention protocol. The primary study was not powered to detect an interaction effect, nor was it designed to examine the SBST. Furthermore, in addition to the analysis of our primary outcome measure, we conducted multiple comparisons of secondary analyses, which increased the chances of committing a type 1 error. The SBST was not intended for use in only patients with very acute LBP. It is notable that the SBST may have limitations and may not be the optimal tool for identifying patients at high risk for delayed recovery due to psychosocial factors. The results of this study may not be generalizable to patients with LBP and symptoms that extend below the knee.
In patients with acute LBP of <16 days’ duration, the effect of the intervention appeared not to be greater in the medium-risk group than in the high-risk group. Patients with acute LBP in the high-risk category seem likely to respond well to one session of education using a cognitive-behavioral strategy. They may experience additional benefit by 3 months from evidence-based physical therapy treatments. These effects seem to disappear at 1 year. This research should be considered hypothesis generating and leads to several recommendations for future research. Future work should examine the protocol used in this study in patients with acute LBP after stratifying intervention groups by baseline status of the SBST. Additionally, we recommend that researchers and clinicians consider comparing different education strategies in patients with very acute LBP who are stratified by the SBST.
Author Contributions
Concept/idea/research design: J. Magel, J.M. Fritz, P. Kjaer, R.L. Marcus, G.P. Brennan
Writing: J. Magel, J.M. Fritz, R.L. Marcus, T Greene
Data collection: J.M. Fritz, G. P. Brennan, J Magel
Data analysis: J. Magel, T. Greene, J.M. Fritz
Project management: J.M. Fritz
Fund procurement: J.M. Fritz
Providing participants: G.P. Brennan, J.M. Fritz, J. Magel
Providing institutional liaisons: G.P. Brennan
Consultation (including review of manuscript before submitting): J.M. Fritz, T Greene, P. Kjaer, R.L. Marcus, G.P. Brennan
Dr Magel holds an Orthopaedics Specialist Certification (OCS) and is a Fellow of the American Academy of Orthopaedic and Manual Physical Therapists. Dr Marcus holds an Orthopaedics Specialist Certification (OCS). Dr Fritz and Dr Brennan are Catherine Worthingham Fellow of the American Physical Therapy Association.
Ethics Approval
The institutional review boards at the University of Utah and Intermountain Healthcare, Salt Lake City, Utah, approved the trial.
Funding
This was a secondary analysis of data from a published clinical trial (NCT01726803) that was funded by the Agency for Healthcare Research and Quality (AHRQ) (grant no. R18HS018672).16 The authors received no funding from AHRQ for this study. This investigation was supported in part by funding from the National Center for Research Resources and the National Center for Advancing Translational Sciences Grant 5UL1TR001067-02.
Disclosures/Presentations
This article was adapted, in part, from a platform presentation given at the 2016 Combined Sections Meeting of the American Physical Therapy Association, February 4–7, 2016, Anaheim, California (abstract published as: Magel J, Fritz JM, Green T, et al. Associations between the STarT Back Screening Tool and the outcomes of patients with acute low back pain who meet a clinical prediction rule. J Orthop Sports Phys Ther. 2016;41:A16).
References
- 1. Mehling WE, Gopisetty V, Bartmess E et al. The prognosis of acute low back pain in primary care in the United States: a 2-year prospective cohort study. Spine (Phila Oa 1976). 2012, 37:678–684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Itz CJ, Geurts JW, van Kleef M, Nelemans P. Clinical course of non-specific low back pain: a systematic review of prospective cohort studies set in primary care. Eur J Pain. 2013, 17:5–15. [DOI] [PubMed] [Google Scholar]
- 3. Costa Lda C, Koes BW, Pransky G et al. Primary care research priorities in low back pain: an update. Spine (Phila Pa 1976). 2013, 38:148–156. [DOI] [PubMed] [Google Scholar]
- 4. Hill JC, Dunn KM, Lewis M et al. A primary care back pain screening tool: identifying patient subgroups for initial treatment. Arthritis Rheum. 2008, 59:632–641. [DOI] [PubMed] [Google Scholar]
- 5. Chou R, Shekelle P. Will this patient develop persistent disabling low back pain? JAMA. 2010, 303:1295–1302. [DOI] [PubMed] [Google Scholar]
- 6. Grotle M, Foster NE, Dunn KM, Croft P. Are prognostic indicators for poor outcome different for acute and chronic low back pain consulters in primary care? Pain. 2010, 151:790–797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Gore M, Sadosky A, Stacey BR et al. The burden of chronic low back pain: clinical comorbidities, treatment patterns, and health care costs in usual care settings. Spine (Phila Pa 1976). 2012, 37:E668–E677. [DOI] [PubMed] [Google Scholar]
- 8. Becker A, Held H, Redaelli M et al. Low back pain in primary care: costs of care and prediction of future health care utilization. Spine (Phila Pa 1976). 2010, 35:1714–1720. [DOI] [PubMed] [Google Scholar]
- 9. Hill JC, Whitehurst DG, Lewis M et al. Comparison of stratified primary care management for low back pain with current best practice (STarT Back): a randomised controlled trial. Lancet. 2011, 378:1560–1571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Foster NE, Mullis R, Hill JC et al. Effect of stratified care for low back pain in family practice (IMPaCT Back): a prospective population-based sequential comparison. Ann Fam Med. 2014, 12:102–111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Main CJ, George SZ. Psychologically informed practice for management of low back pain: future directions in practice and research. Phys Ther. 2011, 91:820–824. [DOI] [PubMed] [Google Scholar]
- 12. Main CJ, Sowden G, Hill JC et al. Integrating physical and psychological approaches to treatment in low back pain: the development and content of the STarT Back trial's “high-risk” intervention (STarT Back; ISRCTN 37113406). Physiotherapy. 2012, 98:110–116. [DOI] [PubMed] [Google Scholar]
- 13. Flynn T, Fritz JM, Whitman J et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short term improvement with spinal manipulation. Spine (Phula Pa 1976). 2002, 27:2835–2843. [DOI] [PubMed] [Google Scholar]
- 14. Brennan GP, Fritz JM, Hunter SJ et al. Identifying sub-groups of patients with “non-specific” low back pain: results of a randomized clinical trial. Spine (Phila Pa 1976). 2006, 31:623–631. [DOI] [PubMed] [Google Scholar]
- 15. Cleland JA, Fritz JM, Kulig K et al. Comparison of the effectiveness of three manual physical therapy techniques in a subgroup of patients with low back pain who satisfy a clinical prediction rule: a randomized clinical trial. Spine (Phila Pa 1976). 2009, 232, 34:2720–2729. [DOI] [PubMed] [Google Scholar]
- 16. Fritz JM, Magel JS, McFadden M et al. Early physical therapy vs usual care in patients with recent-onset low back pain: a randomized clinical trial. JAMA. 2015, 314:1459–1467. [DOI] [PubMed] [Google Scholar]
- 17. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976). 2000, 25:2940–2952. [DOI] [PubMed] [Google Scholar]
- 18. Fritz JM, Childs JD, Flynn TW. Pragmatic application of a clinical prediction rule in primary care to identify patients with low back pain with a good prognosis following a brief spinal manipulation intervention. BMC Fam Pract. 2005, 6:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy. 1980, 66:271–273. [PubMed] [Google Scholar]
- 20. Vianin M. Psychometric properties and clinical usefulness of the Oswestry Disability Index. J Chiropr Med. 2008, 232, 7:161–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Jensen MP, Turner JA, Romano JM. What is the maximum number of levels needed in pain intensity measurement? Pain. 1994, 58:387–392. [DOI] [PubMed] [Google Scholar]
- 22. Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine (Phila Pa 1976). 2005, 30:1331–1335. [DOI] [PubMed] [Google Scholar]
- 23. Osman A, Barrios FX, Gutierrez PM et al. The Pain Catastrophizing Scale: further psychometric evaluation with adult samples. J Behav Med. 2000, 232, 23:351–365. [DOI] [PubMed] [Google Scholar]
- 24. Waddell G, Newton M, Henderson I et al. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain. 1993, 52:157–168. [DOI] [PubMed] [Google Scholar]
- 25. Burton AK, Waddell G, Tillotson KM, Summerton N. Information and advice to patients with back pain can have a positive effect: a randomized controlled trial of a novel educational booklet in primary care. Spine (Phila Pa 1976). 1999, 24:2484–2491. [DOI] [PubMed] [Google Scholar]
- 26. Royston P. Multiple imputation of missing values. Stata J. 2004, 4:227–241. [Google Scholar]
- 27. Fritz JM, Irrgang JJ. A comparison of a modified Oswestry Low Back Pain Disability Questionnaire and the Quebec Back Pain Disability Scale. Phys Ther. 2001, 81:776–788. [DOI] [PubMed] [Google Scholar]
- 28. Fritz JM, George SZ. Identifying psychosocial variables in patients with acute work-related low back pain: the importance of fear-avoidance beliefs. Phys Ther. 2002, 82:973–983. [PubMed] [Google Scholar]
- 29. Wertli MM, Eugster R, Held U et al. Catastrophizing-a prognostic factor for outcome in patients with low back pain: a systematic review. Spine J. 2014, 14:2639–2657. [DOI] [PubMed] [Google Scholar]
- 30. Wertli MM, Rasmussen-Barr E, Weiser S et al. The role of fear avoidance beliefs as a prognostic factor for outcome in patients with nonspecific low back pain: a systematic review. Spine J. 2014, 14:816–836e4. [DOI] [PubMed] [Google Scholar]
- 31. Saragiotto BT, Maher CG, Moseley AM et al. A systematic review reveals that the credibility of subgroup claims in low back pain trials was low. J Clin Epidemiol. 2016, 79:3–9. [DOI] [PubMed] [Google Scholar]