Among occupational risk factors of recurrence, chronicity and no return to work in low back pain, poor job satisfaction is the only high evidence-based factor.
To find out any validated questionnaire usable to assess job satisfaction in low back pain patients, both in clinical practice and research setting.
A systematic literature search on Pubmed and Cochrane library databases and un-indexed literature was made. “Job satisfaction” and “low back pain” keywords were used. Only English and French relevant articles were retained. A double assessment was made of listed questionnaires according to psychometric properties and daily practice use.
Among the 40 articles retained only four used a validated questionnaire. Among the 12 different questionnaires, only two are validated in their English version (Job Descriptive Index [JDI] and the Work Environment Scale [WES]) and one in its French Version (JDI). Because they are time consuming, use these questionnaires in daily practice seems difficult.
Based on literature review and questionnaire heterogeneity, at this time, there is no reference job satisfaction questionnaire. For daily practice, global job satisfaction visual analog scale could be useful. For research and intervention, JDI is more suitable despite its validity is still questionable.
Un faible niveau de satisfaction professionnelle est le seul facteur professionnel retrouvé avec un fort niveau de preuve dans la lombalgie comme facteur de risque de récidive, de chronicité et de non retour au travail.
Rechercher les outils référencés pour l’évaluation de la satisfaction professionnelle des patients lombalgiques utilisables en pratique courante ou en recherche clinique.
Une revue bibliographique systématique a été effectuée. Les mots clés utilisés étaient : job satisfaction et low back pain . Seuls les articles pertinents rédigés en français ou en anglais étaient retenus. Les questionnaires utilisés étaient analysés selon des critères psychométriques et sur les qualités qu’un praticien recherche en pratique courante.
Quarante articles sont retenus dont quatre employant un outil validé. Douze outils référencés d’évaluation de la satisfaction professionnelle sont recensés dont deux validés en anglais (Job Descriptive Index [JDI] et Work Environment Scale [WES]) et un en français (JDI). Leur temps de passation est long.
La disparité des outils retrouvés et le recours fréquent à des questionnaires non validés confirment l’absence d’outil de référence. En pratique courante, une échelle globale d’évaluation de la satisfaction professionnelle peut être utilisée. En recherche et en intervention une échelle multidimensionnelle tel que le JDI est plus appropriée, cependant leur validité reste à développer.
Non-specific low back pain is a significant health problem and a frequent reason for consulting a physician. It generates high rates of morbidity, disability and sick leave. Low back pain has a high economic cost . In fact, 15 to 20% of adults experience an episode of low back pain in a given year, and 60 to 90% of people experience low back pain at least once in their lifetime . Low back pain is one of the main causes of occupational disease and absenteeism .
The main risk factors for recurrence, chronicity and failure to return to work are the clinical criteria of the low back pain episode (pain intensity, disability severity, development duration), the existence of a prior low back pain episode, a low level of job satisfaction, the general state of health, age and psychological factors . Numerous occupational factors are found, with varying levels of proof, as risk factors for chronicity and recurrence: workload, decision-making freedom, social support at the workplace, compensation, sick leave of longer than 8 days, poor posture at work, daily time spent bearing loads, the absence of an ergonomic workstation, low levels of professional qualification, salary inadequacy, no development prospects and task monotony . Occupational stress only has limited scientific proof and among occupational factors, only a low level of job satisfaction as evaluated by the patient has a high level of scientific proof .
In occupational and public health, stress in the workplace, job satisfaction, and more generally, psychosocial factors at the workplace have been extensively studied. The concept of job satisfaction is used in various disciplines (e.g., sociology, psychology, epidemiology, human resources management and medicine…). In psychology, this is traditionally defined as “the positive or pleasing emotional state resulting from the evaluation by a person of their workplace or work experiences” (from Locke, 1976 ) or more recently, job satisfaction has been conceptualized as a “general attitude towards work” (from Fisher and Locke, 1992 ).
Many tools for evaluating the effects of low back pain and risk factors for chronic low back pain are used in everyday practice or clinical research. The Quebec scale evaluates disability, the Dallas questionnaire – quality of life, the Visual Analog Scale (VAS) – pain intensity, the WHO score – general state, the Hospital Anxiety and Depression Scale – psychological state, the Fear Avoidance and Beliefs Questionnaire (FABQ) – fears avoidance and beliefs . Some occupational psychosocial factors can be evaluated by Karasek’s Job Content Questionnaire . However, tools to evaluate job satisfaction, if they exist, are poorly known, and subsequently are less frequently used.
The objective of this study was to catalogue job satisfaction evaluation tools used in low back pain patients to determine a reference tool that can be used in everyday practice and/or clinical research.
Systematic review of the literature
The keywords used to search the Medline and CISMeF databases were: “job satisfaction” [Mesh] and “low back pain” [Mesh] for English language sites and satisfaction professionnelle and lombalgie for French language sites. They include respectively, “job satisfaction”, “work satisfaction”, “lumbago”, “lower back pain”, “low back ache”, “low back pain, recurrent”, “low back pain, postural”, “low back pain, mechanical”, and “low back pain, posterior compartment”.
The search was performed in three different databases: the Medline database, the Cochrane library and non-indexed literature (Abes-Sudoc, Google Scholar, the website of the French INRS institute of occupational risk prevention). The search was for any published articles, regardless of how old the publication date. Additional articles and works were extracted from bibliographic references mentioned in the relevant articles.
To be selected, an article needed to report the results of a prospective study of a population at least partially composed of low back pain patients. The study was to have analyzed job satisfaction as a primary or secondary endpoint. For literature reviews, the referenced articles discussing studies corresponding to the accepted criteria were analyzed to make sure that no relevant studies were missed. The studies needed to be in English or French and published in a scientific journal or on a reference site in the areas of public health, occupational health, rheumatology or physical medicine and rehabilitation (PMR). Several steps eliminated articles that did not correspond to our criteria. An initial selection was performed by reading titles or abstracts only, when available, to eliminate unrelated articles. A second selection was performed by reading the “Methods” section or the entire article, depending on what was necessary.
Inventory of tools used
In each selected study, the method for evaluating job satisfaction – questionnaire referenced in the bibliography or questionnaire devised specifically for the needs of the study – was recorded. For each of these methods, a search was performed on the references cited in article bibliographies or on database research studies for the first article to ever validate the research tool.
Critical analysis of the questionnaires
To objectively study the tools found, a reading table ( Table 1 ) based on both the metrological properties of the questionnaires and the qualities that a practitioner would look for in standard practice was provided. To diminish the risk of interpretation error, two independent examiners from different specialties (PMR [EC] and occupational medicine [MR]) evaluated each tool. Psychometric characteristics, such as validity (construct validity, criterion validity), reliability (interclass correlation coefficient, k coefficient, α internal consistency) and sensitivity to change were examined. The validity of the content was studied by referencing the various explored areas. Finally, the overall validity was characterized as satisfactory, moderate or insufficient by the observers depending on the number of criteria found (insufficient < 1 criterion, moderate 2 criteria, satisfactory > 2 criteria). The visible characteristics of the scales (appearance validity) were also analyzed: execution time, item number, item description, response modalities, rating difficulties, the possibility for the patient to answer the questionnaire alone, the cost and the availability of an English and French version. The execution time was evaluated according to a three-level, semi-quantitative scale based on the number of items in the questionnaire (short < 3 items, moderate 3 to 30 items, long > 30 items). To evaluate the frequency of use, we relied on the frequency of use reported in studies and the number of references found for the tool in the literature.
|Sensitivity to change
|Overall validity (satisfactory/insufficient/unknown)
|Number of items
|Execution time (short/moderate/long)
|Response method (Likert/VAS/binary/open-ended)
|Rating difficulties (yes/no)
|Respondent can answer alone (yes/no)
|Available in English (yes/no)
|Available in French (yes/no)
|Frequency of use
|Social support at the workplace (colleagues, managers)
|Overall job satisfaction
At the end of this search, we found 63 articles corresponding to our criteria, including 56 in Medline, zero in the Cochrane Library and seven in the Gray literature. After reading the summary, 36 articles were selected. None of the Gray literature articles were selected, since they only contained literature reviews and not original studies. After fully reading the 36 articles, 23 were selected and provided 22 additional references, of which 17 fulfilled the selection criteria. In the end, a total of 40 studies were examined ( Fig. 1 ).
Non-referenced evaluation methods
Eighteen studies used an evaluation method that was not referenced in the bibliography . Thirteen of these studies used a single question with variable formulations ( Table 2 ) . Four of these studies used a questionnaire comprised of several items , one included four items, one included eight items and in one study, the number of items was not provided . For eight of the studies, the responses to the question(s) were in the form of a Likert scale. The Likert scales had four to seven points. The expected response was in the form of a numeric scale of 0 to 10 for one questionnaire and in the form of multiple-choice (categorizing the responses into several categories, such as “high satisfaction” and “low satisfaction”) for two questionnaires . The response form was not provided for seven of the questionnaires . Thirteen of these questionnaires were in English , and two of these were also available in French . One author had tested the validity of the scale used by submitting the same question to the patients at a 1-week time interval with a fairly decent result , and another author found satisfactory internal consistency . The other studies did not mention any questionnaire validity analyses. Two studies did not specify their method for evaluating professional satisfaction . One study used one single question on job task satisfaction .