Can we convert between outcome measures of disability for chronic low back pain?

Developing a shortened spine functional index (SFI-10) for patients with sub-acute/chronic spinal disorders: a cross-sectional study

BACKGROUND

Brief whole-spine patient-reported outcome measures (PROMs) provide regional solutions and future directions for quantifying functional status, evidence, and effective interventions. The whole-spine regional Spine Functional Index (SFI-25) is used internationally in clinical and scientific contexts to assess general sub-acute/chronic spine populations. However, to improve structural validity and practicality a shortened version is recommended. This study developed a shortened-SFI from the determined optimal number of item questions that: correlated with criteria PROMs being highly with whole-spine, moderately with regional-spine, condition-specific and patient-specific, and moderately-low with general-health and pain; retained one-dimensional structural validity and high internal consistency; and improved practicality to reduce administrative burden.

METHODS

A cross-sectional study (n = 505, age = 18-87 yrs., average = 40.3 ± 10.1 yrs) of sub-acute/chronic spine physiotherapy outpatients from an international sample of convenience. Three shortened versions of the original SFI-25 were developed using 1) qualitative 'content-retention' methodology, 2) quantitative 'factorial' methodology, and 3) quantitative 'Rasch' methodology, with a fourth 'random' version produced as a comparative control. The clinimetric properties were established for structural validity with exploratory (EFA) and confirmatory (CFA) factorial analysis, and Rasch analysis. Criterion validity used the: whole-spine SFI-25 and Functional Rating Index (FRI); regional-spine Neck Disability Index (NDI), Oswestry Disability Index (ODI), and Roland Morris Questionnaire (RMQ), condition-specific Whiplash Disability Questionnaire (WDQ); and patient-specific functional scale (PSFS); and determined floor/ceiling effect. A post-hoc pooled international sub-acute/chronic spine sample (n = 1433, age = 18-91 yrs., average = 42.0 ± 15.7 yrs) clarified the findings and employed the general-health EuroQuol-Index (EQ-5D), and 11-point Pain Numerical Rating Scale (P-NRS) criteria.

RESULTS

A 10-item SFI retained structural validity with optimal practicality requiring no computational aid. The SFI-10 concept-retention-version demonstrated preferred criterion validity with whole-spine criteria (SFI-25 = 0.967, FRI = 0.810) and exceeded cut-off minimums with regional-spine, condition-specific, and patient-specific measures. An unequivocal one-dimensional structure was determined. Internal consistency was satisfactory (α = 0.80) with no floor/ceiling effect. Post-hoc analysis of the international sample confirmed these findings.

CONCLUSION

The SFI-10 qualitative concept-retention version was preferred to quantitative factorial and Rasch versions, demonstrated structural and criterion validity, and preferred correlation with criteria measures. Further longitudinal research is required for reliability, error, and responsiveness, plus an examination of the practical characteristics of readability and administrative burden.

Systematic Review Procedures for the World Health Organization (WHO) Evidence Syntheses on Benefits and Harms of Structured and Standardized Education/Advice, Structured Exercise Programs, Transcutaneous Electrical Nerve Stimulation (TENS), and Needling Therapies for the Management of Chronic Low Back Pain in Adults

As commissioned by the WHO, we updated and expanded the scope of four systematic reviews to inform its (in development) clinical practice guideline for the management of CPLBP in adults, including older adults. Methodological details and results of each review are described in the respective articles in this series. In the last article of this series, we discuss methodological considerations, clinical implications and recommendations for future research.

Effectiveness of robot-assisted arm exercise on arm and hand function in stroke survivors - A systematic review and meta-analysis

Objective: To examine the treatment effect of commercially available robotic-assisted devices, compared to traditional occupational- and physiotherapy on arm and hand function in persons with stroke. Methods: A systematic literature search was conducted in Medline, EMBASE, CINAHL and Cochrane Central Register of Controlled Trials up to January 2022. Randomized controlled trials (RCT's) involving persons with stroke of all ages and robot-assisted exercise as method for arm and hand function, compared to traditional therapy methods were included. Three authors performed the selection independently. The quality of evidence across studies was assessed using GRADE. Results: Eighteen RCT's were included in the study. A random effects meta-analysis showed a statistically significantly higher treatment effect in the robotic-assisted exercise group (p=<0.0001) compared to the traditional treatment group, with a total effect size of 0.44 (CI = 0.22-0.65). Heterogeneity was high, measured with I2 of 65%). Subgroup analyses showed no significant effects of the type of robotic device, treatment frequency or duration of intervention. Discussion and conclusion: Even though the analysis showed significant improvement in arm and hand function in favor of the robotic-assisted exercise group, the results in this systematic review should be interpreted with caution. This is due to high heterogeneity among the studies included and the presence of possible publication bias. Results of this study highlight the need for larger and more methodological robust RCT's, with a focus on reporting training intensity during robotic exercise.

Measurement properties of patient-reported outcome measures used in rehabilitation of adults with chronic musculoskeletal pain: A mapping review

BACKGROUND

Choosing measurement tools for diagnostic, prognostic, or evaluative purposes in a chronic musculoskeletal pain (CMP) population is challenging for rehabilitation practice. Implementation of measurement tools for clinical practice is impaired by gaps in knowledge about measurement properties.

OBJECTIVE

Identifying evidence about the measurement properties of tools frequently used in Dutch pain rehabilitation practice.

METHODS

A mapping review was conducted of eligible studies that investigated reliability, validity, or responsiveness, and interpretability, as defined by the COSMIN taxonomy, of original versions or Dutch translations of predefined Patient-Reported Outcome Measures (PROMs) in a CMP population. MEDLINE, PsycINFO, EMBASE, and CINAHL were searched in March 2021. Results were visually mapped.

RESULTS

Thirty-five studies were included. The results show many knowledge gaps in both original and translated versions. In general, aspects of validity were most frequently reported. The Pain Disability Index, Pain Catastrophizing Scale, and the 12-Item Short Form Health Survey were the most studied measurement tools. No results were found for the Checklist Individual Strength, Illness Perception Questionnaire, and Utrecht Coping List.

CONCLUSION

Little evidence of the measurement properties of PROMs used in rehabilitation of patients with CMP in the Netherlands was found. PROMs need to be used and interpreted with caution in daily practice.

Development of a mapping function ("crosswalk") for the conversion of scores between the Oswestry Disability Index (ODI) and the Core Outcome Measures Index (COMI)

INTRODUCTION

The Oswestry Disability Index (ODI) and the Core Outcome Measures Index (COMI) are two commonly used self-rating outcome instruments in patients with lumbar spinal disorders. No formal crosswalk between them exists that would otherwise allow the scores of one to be interpreted in terms of the other. We aimed to create such a mapping function.

METHODS

We performed a secondary analysis of ODI and COMI data previously collected from 3324 patients (57 ± 17y; 60.3% female) at baseline and 1y after surgical or conservative treatment. Correlations between scores and Cohen's kappa for agreement (κ) regarding achievement of the minimal clinically important change (MCIC) score on each instrument (ODI, 12.8 points; COMI, 2.2 points) were calculated, and regression models were built. The latter were tested for accuracy in an independent set of registry data from 634 patients (60 ± 15y; 56.8% female).

RESULTS

All pairs of measures were significantly positively correlated (baseline, 0.73; 1y follow-up (FU), 0.84; change-scores, 0.73). MCIC for COMI was achieved in 53.9% patients and for ODI, in 52.4%, with 78% agreement on an individual basis (κ = 0.56). Standard errors for the regression slopes and intercepts were low, indicating excellent prediction at the group level, but root mean square residuals (reflecting individual error) were relatively high. ODI was predicted as COMI × 7.13-4.20 (at baseline), COMI × 6.34 + 2.67 (at FU) and COMI × 5.18 + 1.92 (for change-score); COMI was predicted as ODI × 0.075 + 3.64 (baseline), ODI × 0.113 + 0.96 (FU), and ODI × 0.102 + 1.10 (change-score). ICCs were 0.63-0.87 for derived versus actual scores.

CONCLUSION

Predictions at the group level were very good and met standards justifying the pooling of data. However, we caution against using individual values for treatment decisions, e.g. attempting to monitor patients over time, first with one instrument and then with the other, due to the lower statistical precision at the individual level. The ability to convert scores via the developed mapping function should open up more centres/registries for collaboration and facilitate the combining of data in meta-analyses.

Measurement Properties of the Patient-Specific Functional Scale and Its Current Uses: An Updated Systematic Review of 57 Studies Using COSMIN Guidelines

OBJECTIVE

To systematically review measurement properties, including acceptability, feasibility, and interpretability, and current uses of the Patient-Specific Functional Scale (PSFS).

DESIGN

Systematic review of a patient-reported outcome measure using the COnsensus-based Standards for the selection of health status Measurement INstruments (COSMIN) guidelines.

LITERATURE SEARCH

We searched 11 databases from January 2010 to July 2020 for articles on measurement properties or use of PSFS.

STUDY SELECTION CRITERIA

Published primary articles without language restrictions.

DATA SYNTHESIS

Two independent reviewers screened all records, extracted data, and performed risk of bias assessments using COSMIN guidelines. We qualitatively synthesized findings for each measurement property in musculoskeletal and nonmusculoskeletal conditions, and 2 reviewers independently performed Grading of Recommendations Assessment, Development and Evaluation assessments. This study was preregistered with the Open Science Framework (https://doi.org/10.17605/OSF.IO/42UZT).

RESULTS

Of the 985 articles screened, we included 57 articles on measurement properties and 255 articles on the use of PSFS. The PSFS had sufficient test-retest reliability in musculoskeletal (22 studies, 845 participants, low-to-moderate certainty) and nonmusculoskeletal conditions (6 studies, 197 participants, very low certainty), insufficient construct validity as a measure of physical function (21 studies, 2 945 participants, low-to-moderate certainty), and sufficient responsiveness (32 studies, 13 770 participants, moderate-to-high certainty). The standard error of measurement ranged from 0.35 to 1.5. The PSFS was used in 87 unique health conditions, some without prior evidence of validity.

CONCLUSION

The PSFS is an easy-to-use, reliable, and responsive scale in numerous musculoskeletal conditions, but the construct validity of PSFS remains uncertain. Further study of the measurement properties of the PSFS in nonmusculoskeletal conditions is necessary before clinical use. J Orthop Sports Phys Ther 2022;52(5):262-275. Epub: 05 Feb 2022. doi:10.2519/jospt.2022.10727.

Formulation of Japanese Orthopaedic Association (JOA) clinical practice guideline for the management of low back pain- the revised 2019 edition

BACKGROUND

The latest clinical guidelines are mandatory for physicians to follow when practicing evidence-based medicine in the treatment of low back pain. Those guidelines should target not only Japanese board-certified orthopaedic surgeons, but also primary physicians, and they should be prepared based entirely on evidence-based medicine. The Japanese Orthopaedic Association Low Back Pain guideline committee decided to update the guideline and launched the formulation committee. The purpose of this study was to describe the formulation we implemented for the revision of the guideline with the latest data of evidence-based medicine.

METHODS

The Japanese Orthopaedic Association Low Back Pain guideline formulation committee revised the previous guideline based on a method for preparing clinical guidelines in Japan proposed by Medical Information Network Distribution Service Handbook for Clinical Practice Guideline Development 2014. Two key phrases, "body of evidence" and "benefit and harm balance" were focused on in the revised version. Background and clinical questions were determined, followed by literature search related to each question. Appropriate articles were selected from all the searched literature. Structured abstracts were prepared, and then meta-analyses were performed. The strength of both the body of evidence and the recommendation was decided by the committee members.

RESULTS

Nine background and nine clinical qvuestions were determined. For each clinical question, outcomes from the literature were collected and meta-analysis was performed. Answers and explanations were described for each clinical question, and the strength of the recommendation was decided. For background questions, the recommendations were described based on previous literature.

CONCLUSIONS

The 2019 clinical practice guideline for the management of low back pain was completed according to the latest evidence-based medicine. We strongly hope that this guideline serves as a benchmark for all physicians, as well as patients, in the management of low back pain.

Comparison of the Effectiveness of Pharmacological Treatments for Patients with Chronic Low Back Pain: A Nationwide, Multicenter Study in Japan

Introduction

Chronic low back pain (CLBP) is a leading cause of disability, yet there is limited high-quality evidence to identify the most suitable pharmacological therapy. The purpose of this Japanese nationwide, multicenter, prospective study was to compare the effectiveness of four representative drug therapies-acetaminophen, celecoxib, loxoprofen, and a tramadol and acetaminophen (T+A) combination drug-to establish evidence for a drug of choice for CLBP.

Methods

Patients with CLBP (N=471) received one of the four treatments and were evaluated, prospectively and comprehensively, once every month for six months using a visual analog scale (VAS) for LBP, the Japanese Orthopedic Association (JOA) score, the JOA Back Pain Evaluation Questionnaire (JOABPEQ), the Roland-Morris Disability Questionnaire (RDQ), the EuroQol five-dimensions three-levels (EQ-5D-3L), and the Short Form-8 item health survey (SF-8). We conducted multivariable linear regression analyses of the four drugs at 1 and 6 months after drug allocation. Differences with P<0.05 were considered statistically significant.

Results

Patients who received acetaminophen showed a significant improvement from baseline in the mental health subscale of the JOABPEQ at one month (P=0.02) and the JOA score at six months (P<0.01). None of the other outcome measures among the four drugs differed significantly. Across groups, all outcome measures, except the mental component summary (MCS) score of the SF-8, improved equivalently, although most measurements showed no obvious cumulative effect over six months. The MCS score of the SF-8 decreased gradually over six months in all groups.

Conclusions

Most of the outcome measures among the treated groups were not significantly different, indicating similar treatment effects of the four drugs for CLBP. Our study indicated the limit of each outcome measure for evaluating the patient status, suggesting that a single outcome measure is insufficient to reflect treatment effectiveness.

Identification of subgroup effect with an individual participant data meta-analysis of randomised controlled trials of three different types of therapist-delivered care in low back pain

Background

Proven treatments for low back pain, at best, only provide modest overall benefits. Matching people to treatments that are likely to be most effective for them may improve clinical outcomes and makes better use of health care resources.

Methods

We conducted an individual participant data meta-analysis of randomised controlled trials of three types of therapist delivered interventions for low back pain (active physical, passive physical and psychological treatments). We applied two statistical methods (recursive partitioning and adaptive risk group refinement) to identify potential subgroups who might gain greater benefits from different treatments from our individual participant data meta-analysis.

Results

We pooled data from 19 randomised controlled trials, totalling 9328 participants. There were 5349 (57%) females with similar ratios of females in control and intervention arms. The average age was 49 years (standard deviation, SD, 14).

Participants with greater psychological distress and physical disability gained most benefit in improving on the mental component scale (MCS) of SF-12/36 from passive physical treatment than non-active usual care (treatment effects, 4.3; 95% confidence interval, CI, 3.39 to 5.15). Recursive partitioning method found that participants with worse disability at baseline gained most benefit in improving the disability (Roland Morris Disability Questionnaire) outcome from psychological treatment than non-active usual care (treatment effects, 1.7; 95% CI, 1.1 to 2.31). Adaptive risk group refinement did not find any subgroup that would gain much treatment effect between psychological and non-active usual care. Neither statistical method identified any subgroups who would gain an additional benefit from active physical treatment compared to non-active usual care.

Conclusions

Our methodological approaches worked well and may have applicability in other clinical areas. Passive physical treatments were most likely to help people who were younger with higher levels of disability and low levels of psychological distress. Psychological treatments were more likely to help those with severe disability. Despite this, the clinical importance of identifying these subgroups is limited. The sizes of sub-groups more likely to benefit and the additional effect sizes observed are small. Our analyses provide no evidence to support the use of sub-grouping for people with low back pain.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12891-021-04028-8.

App-based multidisciplinary back pain treatment versus combined physiotherapy plus online education: a randomized controlled trial

Non-specific low back pain (LBP) is one of the leading causes of global disability. Multidisciplinary pain treatment (MPT) programs comprising educational, physical, and psychological interventions have shown positive treatment effects on LBP. Nonetheless, such programs are costly and treatment opportunities are often limited to specialized medical centers. mHealth and other digital interventions may be a promising method to successfully support patient self-management in LBP. To address these issues, we investigated the clinical effects of a multidisciplinary mHealth back pain App (Kaia App) in a randomized controlled trial (registered at German Clinical Trials Register under DRKS00016329). One-hundred one adult patients with non-specific LBP from 6 weeks to 1 year were randomly assigned to an intervention group or a control group. In the intervention group, the Kaia App was provided for 3 months. Control treatment consisted of six individual physiotherapy sessions over 6 weeks and high-quality online education. The primary outcome, pain intensity, was assessed at 12-week follow-up on an 11-point numeric rating scale (NRS). Our per-protocol analysis showed no significant differences between the groups at baseline (Kaia App group: M = 5.10 (SD = 1.07) vs. control group: M = 5.41 (SD = 1.15). At 12-week follow-up the Kaia App group reported significantly lower pain intensity (M = 2.70 (SD = 1.51)) compared to the control group (M = 3.40 (SD = 1.63)). Our results indicate that the Kaia App as a multidisciplinary back pain app is an effective treatment in LBP patients and is superior to physiotherapy in combination with online education.

Outcome measurement in patients with low back pain undergoing epidural steroid injection

Objectives

This study aims to evaluate the outcomes of epidural steroid injection (ESI) in patients with low back pain.

Patients and methods

This prospective study included a total of 82 patients (51 females; 31 males; mean age 50.8±14.2 years; range, 17 to 86 years) who underwent ESI due to lumbar disc hernia-induced radiculopathy between September 2014 and May 2015. The Numeric Rating Scale (NRS), Oswestry Disability Index (ODI), Istanbul Low Back Pain Disability Index (ILBPDI), and the Short Form-36 (SF-36) were administered to all patients before and three weeks and three months after ESI.

Results

The mean scores of all scales were significantly lower at three weeks and three months following ESI compared to the baseline scores. There were no significant differences between the mean scores at three weeks and three months. The NRS yielded the highest post-ESI change from baseline.

Conclusion

Our study results showed that all scales used in this study were effective tools for the evaluation of outcomes of EPI in patients with low back pain. Although the NRS yielded the highest sensitivity for detecting change, evaluating functional state and quality of life is essential for multivariate analyses.

Patient-Reported Outcome Measures in Neurosurgery: A Review of the Current Literature

BACKGROUND

Patient-reported outcome measures (PROMs) play an important role in the evaluation of health outcomes, quality of life, and satisfaction, and have been successfully utilized in many areas of clinical medicine and surgical practice. The prevalence of PROMs in neurosurgery is not known.

OBJECTIVE

To review the PROMs that have been utilized in the published neurosurgery literature to date.

METHODS

Articles were searched in MEDLINE, EMBASE, HMIC Health Management Information Consortium, PsycARTICLES, and PsycINFO using search terms related to neurosurgery and PROMs, published from 1806 to August 2016. A total of 268 articles were identified that were stratified by the inclusion and exclusion criteria leading to a total of 137 articles. Twenty-six PROMs, involving both adult and pediatric populations, were identified.

RESULTS

A large number of generic and disease-specific PROMs are used in the neurosurgical literature. Generic PROMs are usually nonspecific measures of health status. Disease-specific PROMs may not address issues relevant to neurosurgical procedures. There are very few neurosurgery-specific PROMs that take into account the impact of a neurosurgical procedure on a specific condition.

CONCLUSION

PROMs that currently feature in the neurosurgical literature may not address the specific outcomes relevant to neurosurgical practice. There is an emergent need for generic and disease-specific PROMs to be validated in neurosurgical patients and neurosurgery-specific PROMs developed to address unmet needs of patients undergoing neurosurgical procedures.

A Comparison of Back Pain Functional Scale With Roland Morris Disability Questionnaire, Oswestry Disability Index, and Short Form 36-Health Survey

STUDY DESIGN

A comparison study of Back Pain Functional Scale (BPFS) with Roland Morris Questionnaire (RMQ), Oswestry Disability Index (ODI), and Short Form 36-Health Survey (SF-36).

OBJECTIVE

The aim of this study is to investigate the correlation of BPFS with RMQ, ODI, and SF-36.

SUMMARY OF BACKGROUND DATA

The primary goal in the treatment of patients with low back pain is to improve the patients' levels of activities and participation. Many questionnaires focusing on function have been developed in patients with low back pain. BPFS is one of these questionnaires. No studies have investigated the correlation of BPFS with ODI and SF-36.

METHODS

This study was conducted with 120 patients receiving outpatient and inpatient treatment in physiotherapy and rehabilitation units of a state hospital. BPFS, RMQ, ODI, and SF-36 questionnaires were used to assess the disability in low back pain. Spearman and Pearson Correlation were used to compare the data obtained in the study.

RESULTS

There was a good correlation among the 5 functional outcome measures (correlation r = -0.693 for BPFS/RMQ, r = -0.794 for BPFS/ODI, r = 0.697 for BPFS/SF-36 Physical function and r = 0.540 for BPFS/SF-36 Pain).

CONCLUSION

BPFS demonstrated good correlation with RMQ, ODI, SF-36 physical function, and SF-36 pain.

LEVEL OF EVIDENCE

2.

The Chronic Pain Grade Questionnaire: validity, reliability and responsiveness in Greek chronic hip pain sufferers

BACKGROUND

The Chronic Pain Grade Questionnaire (CPGQ) was developed to assess the global severity of chronic pain based on pain intensity and pain-related disability. This study aimed to translate, culturally adapt, and validate the Greek version of the CPGQ (CPGQ-Gr).  Methods: Adaptation into Greek followed established guidelines. We invited orthopedic outpatients suffering from chronic hip pain to participate in the study. The validity, reliability, and responsiveness of the CPGQ-Gr were assessed.

RESULTS

Factor analysis yielded two factors (subscales), disability score (DS) and characteristic pain intensity (CPI). CPGQ-Gr items, total and subscale scores were highly correlated with the 12-Item Short Form Health Survey (SF-12) physical component summary score, and slightly correlated or not correlated with the SF-12 mental component summary score. Cronbach's alpha correlation coefficients for the CPGQ-Gr total scale, DS, and CPI subscales were 0.90, 0.95, and 0.83 respectively. All measures showed excellent temporal stability (intraclass correlation coefficients of 0.84, 0.92, and 0.91, respectively). Cliff's delta effect sizes ranged from 0.47 to 0.82. The values of the area under the receiver operating characteristic curve were consistent with good to excellent discriminatory ability (range: 0.747-0.902).

CONCLUSION

Our findings suggest that the Greek version of the CPGQ is a valid, reliable, and sensitive to changes, instrument for grading the severity of chronic hip pain. HIPPOKRATIA 2018, 22(1): 37-42.

The Power of Low Back Pain Trials: A Systematic Review of Power, Sample Size, and Reporting of Sample Size Calculations Over Time, in Trials Published Between 1980 and 2012

STUDY DESIGN

A systematic review of nonspecific low back pain trials published between 1980 and 2012.

OBJECTIVE

To explore what proportion of trials have been powered to detect different bands of effect size; whether there is evidence that sample size in low back pain trials has been increasing; what proportion of trial reports include a sample size calculation; and whether likelihood of reporting sample size calculations has increased.

SUMMARY OF BACKGROUND DATA

Clinical trials should have a sample size sufficient to detect a minimally important difference for a given power and type I error rate. An underpowered trial is one within which probability of type II error is too high. Meta-analyses do not mitigate underpowered trials.

METHODS

Reviewers independently abstracted data on sample size at point of analysis, whether a sample size calculation was reported, and year of publication. Descriptive analyses were used to explore ability to detect effect sizes, and regression analyses to explore the relationship between sample size, or reporting sample size calculations, and time.

RESULTS

We included 383 trials. One-third were powered to detect a standardized mean difference of less than 0.5, and 5% were powered to detect less than 0.3. The average sample size was 153 people, which increased only slightly (∼4 people/yr) from 1980 to 2000, and declined slightly (∼4.5 people/yr) from 2005 to 2011 (P < 0.00005). Sample size calculations were reported in 41% of trials. The odds of reporting a sample size calculation (compared to not reporting one) increased until 2005 and then declined (Equation is included in full-text article.).

CONCLUSION

Sample sizes in back pain trials and the reporting of sample size calculations may need to be increased. It may be justifiable to power a trial to detect only large effects in the case of novel interventions.

LEVEL OF EVIDENCE

3.

Identifying back pain subgroups: developing and applying approaches using individual patient data collected within clinical trials

Background

There is good evidence that therapist-delivered interventions have modest beneficial effects for people with low back pain (LBP). Identification of subgroups of people with LBP who may benefit from these different treatment approaches is an important research priority.

Aim and objectives

To improve the clinical effectiveness and cost-effectiveness of LBP treatment by providing patients, their clinical advisors and health-service purchasers with better information about which participants are most likely to benefit from which treatment choices. Our objectives were to synthesise what is already known about the validity, reliability and predictive value of possible treatment moderators (patient factors that predict response to treatment) for therapist-delivered interventions; develop a repository of individual participant data from randomised controlled trials (RCTs) testing therapist-delivered interventions for LBP; determine which participant characteristics, if any, predict clinical response to different treatments for LBP; and determine which participant characteristics, if any, predict the most cost-effective treatments for LBP. Achieving these objectives required substantial methodological work, including the development and evaluation of some novel statistical approaches. This programme of work was not designed to analyse the main effect of interventions and no such interpretations should be made.

Methods

First, we reviewed the literature on treatment moderators and subgroups. We initially invited investigators of trials of therapist-delivered interventions for LBP with > 179 participants to share their data with us; some further smaller trials that were offered to us were also included. Using these trials we developed a repository of individual participant data of therapist-delivered interventions for LBP. Using this data set we sought to identify which participant characteristics, if any, predict response to different treatments (moderators) for clinical effectiveness and cost-effectiveness outcomes. We undertook an analysis of covariance to identify potential moderators to apply in our main analyses. Subsequently, we developed and applied three methods of subgroup identification: recursive partitioning (interaction trees and subgroup identification based on a differential effect search); adaptive risk group refinement; and an individual participant data indirect network meta-analysis (NWMA) to identify subgroups defined by multiple parameters.

Results

We included data from 19 RCTs with 9328 participants (mean age 49 years, 57% females). Our prespecified analyses using recursive partitioning and adaptive risk group refinement performed well and allowed us to identify some subgroups. The differences in the effect size in the different subgroups were typically small and unlikely to be clinically meaningful. Increasing baseline severity on the outcome of interest was the strongest driver of subgroup identification that we identified. Additionally, we explored the application of Bayesian indirect NWMA. This method produced varying probabilities that a particular treatment choice would be most likely to be effective for a specific patient profile.

Conclusions

These data lack clinical effectiveness or cost-effectiveness justification for the use of baseline characteristics in the development of subgroups for back pain. The methodological developments from this work have the potential to be applied in other clinical areas. The pooled repository database will serve as a valuable resource to the LBP research community.

Funding

The National Institute for Health Research Programme Grants for Applied Research programme. This project benefited from facilities funded through Birmingham Science City Translational Medicine Clinical Research and Infrastructure Trials Platform, with support from Advantage West Midlands (AWM) and the Wolfson Foundation.