Are depressive symptoms or depression associated with poorer health outcomes in adults with low back pain (LBP) with or without radiculopathy?
We will search MEDLINE, EMBASE, CINAHL, and PsycINFO from database inception to present date. Search terms will include subject headings (e.g., MeSH in MEDLINE) for each database and free text words for LBP and depressive symptoms/depression. We will limit the search to English to increase feasibility. We will also search the reference lists of included studies and reference lists of related systematic reviews.
Eligible studies targeting the population, exposure, and outcomes listed above must meet the following criteria: 1) English language; and 2) cohort or case-control studies. We will exclude the following: 1) guidelines, letters, editorials, commentaries, books and book chapters, conference proceedings, meeting abstracts, lectures and addresses, consensus development statements; 2) case reports, case series, cross-sectional studies, randomized controlled trials, qualitative studies, systematic and non-systematic reviews, biomechanical studies, laboratory studies, studies not reporting on methodology; and 3) cadaveric or animal studies.
Our systematic review will target studies of adults (aged 16 years and older) with LBP with or without radiculopathy. LBP is defined as pain localized below the costal margin and above the inferior gluteal folds, with or without referred leg pain in the absence of any underlying serious or major pathology. Radiculopathy refers to inflammation, injury, or compression of the spinal nerve roots that can present as pain, weakness, or numbness in a myotomal or dermatomal distribution. We will include studies targeting LBP with or without referred leg pain or radiculopathy as described with terms including mechanical LBP, lumbago, lumbar sprain or strain, lumbopelvic pain, lumbar radiculopathy, lumbar disc herniations, sacroiliac syndrome, sciatica, and spinal stenosis.
Adults (aged 16 years and older) with LBP with or without radiculopathy.
We will target studies that assess depressive symptoms or depression as the exposure. We will include depressive symptoms as self-reported symptoms of depression on standardized questionnaires (e.g., Center for Epidemiologic Studies Depression Scale, Beck Depression Index). We will also include diagnosed depression, which has two main categories: 1) major depressive disorder/episode; and 2) dysthymia. Major depressive disorder/episode presents with symptoms such as depressed mood, loss of interest and enjoyment, and decreased energy, and can be categorized as mild, moderate, or severe based on the symptom frequency and severity. Dysthymia is a persistent or chronic form of mild depression with symptoms similar to depressive episodes, but are less intense and persist longer.
We will examine depressive symptoms or depression compared to the absence of depressive symptoms or depression. We will also examine higher severity of depressive symptoms or depression compared to lower severity based on, respectively, scoring of standardized questionnaires (e.g., severe versus mild depressive symptoms based on standardized thresholds on the Beck Depression Index) and standardized diagnoses (e.g., severe depressive episode versus mild depressive episode based on the International Classification of Diseases (ICD) codes).
All health-related contexts will be included.
We will target the following health outcomes: 1) pain (e.g., pain intensity); 2) disability (e.g., impairment, activity limitations, participation restriction); 3) overall health status (e.g., health-related quality of life, recovery); 4) satisfaction with care; and 5) health care utilization (e.g., physician visits, emergency department visits, hospitalizations, spinal imaging). These are informed by core outcome domains that are considered important for LBP research among international panels of experts. We will only include standardized outcome measures (e.g., standardized questionnaires or administrative data) for the aforementioned health outcomes.
Effect measures of interest include odds ratio or risk ratio for dichotomous data, rates or rate ratios for count data, mean differences for continuous data, and survival time or hazard rate ratios for time to event data. If these effect measures are not reported by the studies, we will compute these effect measures, when applicable, based on available data in the studies.
No additional outcomes
We will use a two-phase screening process to select eligible studies. We will first pilot test the eligibility criteria on a random subset of titles and abstracts from the literature search. In phase I screening, pairs of independent reviewers will screen citation titles and abstracts to determine the eligibility of studies (categorizing studies as possibly relevant or irrelevant). We will conduct similar pilot testing prior to screening potentially relevant full-text articles. Pairs of independent reviewers will screen possibly relevant studies in full text during phase II screening to determine eligibility and document reasons for exclusion. Reviewers will meet to discuss disagreements and reach consensus on the eligibility of studies. We will involve a third reviewer if consensus cannot be reached.
Pairs of independent reviewers will critically appraise eligible studies using the Risk of Bias in Non-randomized Studies - of Exposures (ROBINS-E) tool. Before starting critical appraisal, we will conduct a training exercise using the risk of bias appraisal components of ROBINS-E, whereby two reviewers will independently appraise a random subset of included studies and discuss to reach consensus (a third reviewer will be involved if necessary). Paired reviewers will discuss disagreements to reach consensus, and a third reviewer will be involved if consensus cannot be reached. We will use the ROBINS-E tool to evaluate the presence and impact of selection bias, information bias, and confounding on study results. Reviewers will use these criteria to inform their judgment on the internal validity of studies (e.g., low, moderate versus high risk of bias).
We will report the study and patient characteristics, risk of bias results, and outcomes across the included studies. We will assess for clinical, methodological, and statistical heterogeneity among studies. Clinical heterogeneity may result from differences in populations, exposures, comparators, or outcomes across studies. Methodological and statistical heterogeneity may result from differences in risk of bias (i.e., low, moderate, versus high risk of bias) and differences in outcomes across studies beyond what could be expected by chance alone. If studies are deemed homogenous, we will perform a random effects meta-analysis on the association between depressive symptoms and health outcomes. To explore the impact of methodological quality (risk of bias) on study results, we will conduct the following meta-analyses: 1) including all studies (i.e., low, moderate, and high risk of bias studies); and 2) including low risk of bias studies only. If studies are deemed heterogeneous, we will descriptively outline the results of the included studies, stratified by low/moderate risk of bias versus high risk of bias studies.
We will stratify results by type of LBP (with or without radiculopathy), duration of LBP (acute/subacute i.e., less than 3 months’ duration versus chronic i.e., 3 months’ duration or greater), type of exposure (depressive symptoms versus depression) and health outcome (e.g., pain intensity, disability).
Dalla Lana School of Public Health, University of Toronto
Dr Jessica Wong. Dalla Lana School of Public Health, University of Toronto
Dr Laura Rosella. Dalla Lana School of Public Health, University of Toronto
Dr Andrea Tricco. Li Ka Shing Knowledge Institute of St. Michael's Hospital
Dr Pierre Côté. Faculty of Health Sciences, University of Ontario Institute of Technology
Meta-analysis, Systematic review
01 February 2019
01 February 2020
Laura Rosella is funded by a Tier 2 Canada Research Chair in Population Health Analytics. Andrea Tricco is funded by a Tier 2 Canada Research Chair in Knowledge Synthesis. Pierre Côté is funded by a Tier 2 Canada Research Chair in Disability Prevention and Rehabilitation. The funders played no role in the development of this protocol
Subject indexing assigned by CRD
Adult; Depression; Depressive Disorder; Humans; Low Back Pain; Radiculopathy
25 April 2019
25 April 2019
|Piloting of the study selection process||No||No|
|Formal screening of search results against eligibility criteria||No||No|
|Risk of bias (quality) assessment||No||No|