What are the comparative benefits and harms of different oral or topical pharmacological therapies or combinations thereof (combinations may include both pharmacological and nonpharmacological components) for acute, subacute, or chronic nonradicular low back pain, radicular low back pain, or spinal stenosis?
Key Question 2 (KQ2)
What are the comparative benefits and harms of different nonpharmacological, noninvasive therapies, or combinations thereof (combinations may include both pharmacological and nonpharmacological components) for acute, subacute, or chronic nonradicular low back pain, radicular low back pain, or spinal stenosis, including but not limited to exercise and related interventions, complementary and alternative therapies, psychological therapies , physical modalities, and interdisciplinary rehabilitation?
Key Question 3 (KQ3)
How do the benefits of pharmacological or nonpharmacological therapies for low back pain vary according to patient characteristics (e.g., demographic, clinical, and psychosocial risk factors)?
Searches will begin in January 2006, as systematic evidence reviews conducted for an ACP/APS clinical practice guideline addresses the interventions covered in the current review and conducted searches through November 2006.
Electronic database searches will be updated while the draft report is posted for public comment and peer review to capture any new publications. Literature identified during the updated search will be assessed by following the same process of dual review as all other studies considered for inclusion in the report. If any new pertinent literature is identified for inclusion in the update search process, it will be incorporated prior to final submission of the report.
Literature Databases. Ovid MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Database of Systematic Reviews, and the National Guideline Clearinghouse will be searched to capture both published and grey literature.
Scientific Information Packets. Companies that manufacture non-generic medications for low back pain will be invited to provide Scientific Information Packets.
Hand Searching. Reference lists of included articles will be reviewed for includable literature.
Contacting Authors. In the event that information regarding methods or results appears to be omitted from the published results of a study, or if we are aware of unpublished data, we will query the authors to obtain this information.
Process for Selecting Studies. Pre-established criteria based on the PICOTS will be used to determine eligibility for inclusion and exclusion of abstracts in accordance with the AHRQ Methods Guide.19 All excluded abstracts will be reviewed by at least two reviewers. All citations deemed appropriate for inclusion by at least one of the reviewers will be retrieved. Each full-text article will be independently reviewed for eligibility by two team members, including any articles suggested by peer reviewers or that arise from the public posting process. Any disagreements will be resolved by consensus. A record of studies excluded at the full-text level with reasons for exclusion will be maintained.
Types of study to be included
Given the large number of interventions and comparisons addressed in this review, systematic reviews of randomized trials will be used if they address a key question, include studies that meet the PICOTS as defined above, and are assessed as good-quality using the AMSTAR quality assessment tool. We will exclude outdated systematic reviews, defined as systematic reviews for which searches ended prior to 2009. If systematic reviews are included, we will update findings with any new primary trials identified in our searches, update meta-analyses if appropriate, and re-assess strength of evidence based on the totality of evidence. If multiple systematic reviews are relevant and good-quality, we will focus on the findings from the most recent reviews, evaluate areas of consistency and inconsistency across the reviews, and assess strength of evidence based on the totality of evidence. For harms, we will include cohort studies or interventions and comparisons when randomized trials are sparse or unavailable. We will exclude case-control studies, case reports, and case series.
Condition or domain being studied
Low back pain is one of the most frequently encountered conditions in clinical practice. Up to 84 percent of adults have low back pain at some time in their lives, and over one quarter of U.S. adults report recent (in the last three months) low back pain. Low back pain can have major adverse impacts on quality of life and function; it is frequently associated with depression or anxiety. Low back pain is also costly - in 1998, total U.S. health care expenditures for low back pain were estimated at $90 billion. Since that time, costs of low back pain care have risen at a rate higher than observed for overall health expenditures. Low back pain is one of the most common reasons for missed work or reduced productivity while at work, resulting in high indirect costs.
• Adults with acute (<4 weeks), subacute (4-12 weeks), or chronic (>12 weeks) nonradicular low back pain, radicular low back pain, or symptomatic spinal stenosis.
• Exclude: Children, pregnant women
• Exclude: Patients with low back pain related to cancer, infection, inflammatory arthropathy, high velocity trauma, fracture; or low back pain associated with severe or progressive neurological deficits
KQ1: Oral or topical pharmacologic therapies (or combinations thereof)
• Nonsteroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen, naproxen, celecoxib, acetylsalicylic acid (aspirin)
• Nonopioid analgesics, such as acetaminophen
• Opioid analgesics, such as oxycodone, hydrocodone, hydromorphone, morphine, fentanyl
• Tramadol and tapentadol
• Antidepressants, such as tricyclic antidepressants, serotonin-norepinephrine reuptake inhibitors (SNRIs), and selective serotonin-reuptake inhibitors (SSRIs), or serotonin antagonist and reuptake inhibitors (SARIs)
• Skeletal muscle relaxants
• Corticosteroids, such as prednisone or prednisolone
• Anti-epileptic drugs, such as gabapentin or pregabalin
• Exclude: Invasive, nonsurgical therapies (e.g., injections) and surgical therapies
Any included pharmacological or non-pharmacological intervention or combination of interventions (combinations may include both pharmacological and nonpharmacological components) versus any other included intervention or combination of interventions, placebo (drug trials), sham (functionally-inert) treatments, or no treatment.
Any nonhospital setting or in self-directed care
• Final health outcomes
o Reduction or elimination of low back pain, including related leg symptoms
o Improvement in back-specific and overall function
o Improvement in health-related quality of life (HRQOL)
o Reduction in work disability/return to work
o Global improvement
o Number of back pain episodes or time between episodes
o Patient satisfaction
• Adverse effects of intervention(s)
o Pharmaceutical: serious (anaphylaxis, death) and nonserious (mild allergic or untoward) drug reactions or effects; opioid addiction or overdose
o Nonpharmaceutical: serious (death, neurological including cauda equine syndrome, fracture, local skin burns, etc.) and nonserious (mild transient local or general soreness, stiffness, aching; local skin irritation, etc.)
Data extraction, (selection and coding)
After studies are selected for inclusion, we will abstract the following data for each randomized trial or cohort study: study design, year, setting, country, sample size, eligibility criteria, population and clinical characteristics, intervention characteristics, and results. Information relevant for assessing applicability will also be abstracted, including the characteristics of the population, interventions, and care settings; the use of run-in or washout periods, and the number of patients enrolled relative to the number assessed for eligibility.
For systematic reviews we will abstract the following data: Inclusion criteria, search strategy, databases searched, search dates, the number of included studies, study characteristics of included studies (e.g., sample sizes, interventions, comparison, and results), methods of quality assessment, quality ratings for included studies, methods for synthesis, and results.
All study data will be verified for accuracy and completeness by a second team member.
Risk of bias (quality) assessment
Predefined criteria will be used to assess the quality of individual controlled trials, systematic reviews, and observational studies by using clearly defined templates and criteria as appropriate. Randomized trials will be evaluated with appropriate criteria and methods developed by the Cochrane Back Review Group and cohort studies will be evaluated using criteria developed by the U.S. Preventive Services Task Force. Systematic reviews will be assessed using the AMSTAR quality rating instrument.20 These criteria and methods will be used in conjunction with the approach recommended in AHRQ Methods Guide. Studies will be rated as “good,” “fair,” or “poor.” We will re-review the quality ratings of studies included in the prior American Pain Society review to insure consistency in quality assessment.
Primary studies rated “good” will be considered to have the least risk of bias, and their results will be generally considered valid. Good-quality studies use valid methods to select patients for inclusion and allocate patients to treatment; report similar baseline characteristics in different treatment groups; clearly report attrition and have low attrition; use appropriate methods to reduce performance bias, for example, blinding of patients, care providers, and outcome assessors; and use appropriate analytic methods, for example, intention-to-treat analysis and, for cohort studies, adjustment for potential confounders (which may include demographics or social and behavioral factors).
Studies rated “fair” will be susceptible to some bias, though not enough to invalidate the results. These studies may not meet all the criteria for a rating of good quality, but no flaw is likely to cause major bias. The study may also be missing information, making it difficult to assess limitations and potential problems. The fair-quality category is broad, and studies with this rating will vary in their strengths and weaknesses. The results of some fair-quality studies are likely to be valid, while others may be only possibly valid.
Studies rated “poor” will have significant flaws that imply biases of various types that may invalidate the results. They will have a serious or “fatal” flaw in design, analysis, or reporting, such as inadequate methods for allocating patients to treatment; large amounts of missing information; discrepancies in reporting; or serious problems in the delivery of the intervention. The results of these studies will be least as likely to reflect flaws in the study design as the true difference between the compared interventions. We will not exclude studies rated as being poor in quality a priori, but poor-quality studies will be considered to be less reliable than higher quality studies when synthesizing the evidence, particularly when discrepancies between studies are present.
For systematic reviews, we will only include studies rated “good,” based on use of multiple sources in the literature search, application of pre-defined inclusion and exclusion criteria, assessment of risk of bias using an appropriate tool, use of methods to reduce errors in data abstraction and quality rating (e.g., multiple independent reviewers), appropriate methods for evidence synthesis (qualitative or quantitative), and an explicit system for considering the body of evidence that includes the major domains of strength of evidence (risk of bias, consistency, precision, and directness).
Each study evaluated will be dual-reviewed for quality by two team members. Any disagreements will be resolved by consensus.
Strategy for data synthesis
We will construct evidence tables identifying the study characteristics (as discussed above), results of interest, and quality ratings for all included studies, and summary tables to highlight the main findings. We will review and highlight studies by using a hierarchy-of-evidence approach, where the best evidence is the focus of our synthesis for each key question. In the evidence tables, we will include relevant studies from the prior ACP/APS review as well as new studies identified in current searches. We will summarize findings from prior good-quality systematic reviews, including the number and types of studies included and overall findings, separately from newly identified studies.
We will synthesize data qualitatively (based on ranges and descriptive analysis, with interpretation of results) and quantitatively (meta-analysis) when appropriate. Meta-analyses will be conducted to summarize data and obtain more precise estimates on outcomes for which studies are homogeneous enough to provide a meaningful combined estimate. The feasibility of a quantitative synthesis will depend on the number and completeness of reported outcomes and a judgment of adequate homogeneity among the reported results. In general, we would pool if three or more trials were available for a specific comparison and outcome; pooling would only be considered for two trials with high clinical homogeneity and low statistical heterogeneity. To determine whether meta-analysis could be meaningfully performed, we will consider the quality of the studies and the heterogeneity among studies in design, patient population, interventions, and outcomes, and may conduct sensitivity analyses. We will consider using meta-analysis results from good-quality systematic reviews that used appropriate methods for pooling. If new studies not included in the meta-analysis are identified, decisions regarding whether to perform an updated meta-analysis will be based on the precision of the pooled estimate, the consistency of results from new studies compared to the pooled estimate, and the likelihood that results from new studies would impact conclusions and estimates. Meta-regression may be conducted to explore statistical heterogeneity using patient demographics, comorbidities, pain types, treatment features (including specific techniques and number and intensity of treatments) and dosing strategies and additional variables on methodological or other characteristics (e.g., quality, randomization or blinding, outcome definition and ascertainment) given the availability of at least six to ten studies for continuous variables and four studies for categorical variables.
Results will be presented as structured by the key questions and organized by the duration of symptoms (acute, subacute, or chronic) type of low back pain (non-radicular low back pain, radicular low back pain, spinal stenosis), with prioritized outcomes (pain, function) presented
Analysis of subgroups or subsets
Contact details for further information
Oregon Health & Science University, Mail code: BICC
3181 S.W. Sam Jackson Parkway Rd.
Portland, OR 97239-3098
Organisational affiliation of the review
Pacific Northwest Evidence-based Practice Center
Ms Elaine Graham, OHSU Dr Roger Chou, OHSU Dr Janna Friedly, University of Washington Dr Robin Hashimoto, Spectrum Dr Andrea Skelly, Spectrum Dr Sean Sullivan, University of Washington Ms Melissa Weimer, University of Washington Ms Rochelle Fu, OHSU Mr Paul Kraegel, University of Washington Ms Erika Brodt, Spectrum Ms Tracy Dana, OHSU Ms Leah Williams, OHSU Ms Sara Grusing, OHSU
Anticipated or actual start date
18 September 2014
Anticipated completion date
14 May 2015
Agency for Healthcare Research and Quality, U.S. Department of Health and Human Services
Formal screening of search results against eligibility criteria
Risk of bias (quality) assessment
PROSPERO This information has been provided by the named contact for this review. CRD has accepted this information in good faith and registered the review in PROSPERO. CRD bears no responsibility or liability for the content of this registration record, any associated files or external websites.