IRSST - Institut de recherche Robert-Sauvé en santé et en sécurité du travail

Review of knowledge on risk factors for knee osteoarthritis and evaluation tools and care and service interventions


Osteoarthritis is a chronic disease characterized by progressive degeneration of the cartilage and subchondral bone, resulting in pain, functional limitations and long-term disability. The weight-bearing joints are especially vulnerable to developing the disease, but knee osteoarthritis (OA) is more likely to cause functional disabilities. Although the disease is mainly found in people aged 65 and up, the Public Health Agency of Canada estimates that the prevalence of knee OA in people aged 55 to 64 was 66% in 2026. This anticipated growth in prevalence is a serious concern given that osteoarthritis is one of the most important chronic diseases in terms of using health care services. Knee OA has many adverse social and economic impacts. These data highlight the importance of proposing more efficient knee OA management throughout the care and services continuum (prevention-treatment-rehabilitation).

In current evidence-based practice trends, clinicians and managers are increasingly required to consult the scientific literature in order to propose better interventions to clients. This disease has given rise to a considerable number of publications in a variety of research areas. It has become difficult for people who provide care and services to knee OA patients to make sense of such an abundance of information and use it effectively.

The general objective of this project was to develop a synthesis of knowledge on risk factors for knee OA, evaluation tools and care and services for people with the disease. The first specific objective was to produce a synthesis of scientific evidence on all risk factors associated with the development and progress of knee OA. The second specific objective was to identify evaluation tools used in the rehabilitation of knee OA patients and analyze their relevance and metrological characteristics. The third specific objective was to produce a synthesis of scientific evidence on intervention options for knee OA patients.

The approach used was a systematic or critical review of the scientific literature. Research strategies were adopted for each specific objective in order to identify the relevant literature in various electronic databases (MEDLINE, SCOPUS, AMED, etc.). Manual searches were done as well. Lastly, a synthesis of best evidence was done based on high-quality studies (specific objectives 1 and 3) and all studies considered of interest (objective 2) and is presented in this report. Summary tables of detailed results are found in a separate document (available at the site of the REPAR/FRQS hyperlink).

For specific objective 1, it was determined that advanced age, being female, obesity and high body mass index (BMI), working in a kneeling or squatting position and handling heavy loads, high-intensity physical activities performed over a long period and high bone mineral density are the most significant risk factors for knee OA. The evidence for these factors was moderate to strong. Significant inter-study heterogeneity was found in the characterization of exposure.

For objective 2, the criterion validity of a clinical measurement of knee alignment using an inclinometer was demonstrated. The pain subscales of most of the algofonctional questionnaires have good validity and reliability. Some tools, such as the Intermittent and Constant Osteoarthritis Pain (ICOAP), are useful for assessing more specific aspects such as intermittent and constant pain.  For joint and muscle function, the concept of Cyriax’s capsular pattern has  not been validated, whereas methods for measuring the isometric and isokinetic muscle strength of knee OA patients are reliable. The results of our study confirm the robustness of the psychometric properties (which are good to excellent) of several tools for assessing the Activity and/or Participation components, specifically the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), the Oxford Knee Score (OKS), the Patient Function Numerical Rating Scale (NRS) and the Lequesne Algofunctional Index (LAI). Lastly, the Work Limitations Questionnaire (WLQ) scored well for internal consistency and content and construct validity in measuring the effects of OAK on work performance. However, the WLQ is less responsive to change than the Work Instability Scale for Rheumatoid Arthritis (RA-WIS).

As for specific objective 3, physical exercise has been shown to be up to moderately effective; it must be practiced regularly in order to maintain its positive effect on pain and function. Hyaluronic acid injections are effective in relieving pain: their action is neither immediate nor lasting. Non-steroidal anti-inflammatories (NSAIDs) are good painkillers, but have significant side effects. As for supplements, electric therapy, acupuncture, heat, cold, ortheses and laser therapy, evidence of their respective effectiveness is often contradictory or based on low-quality or non-homogeneous studies.

In conclusion, for objective 1, our findings are consistent overall with those of the authors of previously published systematic reviews or meta-analyses. However, the scientific evidence probably would have been stronger for certain risk factors if the methods of characterizing exposure had been more consistent. Likewise, the role of several risk factors could have been clarified if more high-quality observational studies had been found on the subject. Although overall we were able to develop quite a comprehensive profile of factors associated with the development of knee OA, the same cannot be said for progression-related factors. This would have required more cohort studies.

For objective 2, we recognize that there is great interest in cross-cultural validation of algofonctional questionnaires and studies to validate measurement tools that provide an overall score based on the sum of scores in several dimensions or categories of the international classification of functioning, disability and health (ICF). Although most such tools are valid and reliable, several are generic. This indicates a need to develop or validate tools that will be more effective for measuring each ICF category. This applies especially to the Activities and Participation components.

For objective 3, one finding emerges: at present, there is no miracle treatment or therapy for people suffering from knee OA. Exercise is definitely recommended and must be practiced regularly to maintain its positive effects. Hyaluronic acid injections are effective, but their action is neither immediate nor lasting. NSAIDs are effective painkillers but are viable only in the short term and, furthermore, have serious side effects. Accordingly, different treatment options should be considered for effective management of knee OA. This is consistent with the points raised in many practice guides based on literature reviews.