Symptom onset, diagnosis and management of osteoarthritis

by Karen V. MacDonald, Claudia Sanmartin, Kellie Langlois and Deborah A. Marshall

Osteoarthritis affects more than 10% of Canadians aged 15 or older.Note1,Note2 It can result in disability and the need for joint replacement surgery.Note2-5 The strong association between osteoarthritis and advancing age means that as the population of Canada ages, it is important to understand the prevalence and impact of the condition and strategies that individuals use to manage their symptoms. The time between symptom onset and physician diagnosis is a period when people with osteoarthritis can make lifestyle changes to reduce pain, improve function and delay disability.

The literature on osteoarthritis symptom management focuses on strategies led or suggested by physicians and other health care practitioners after an individual has sought care.Note6-12 However, the length of time that people who experience joint pain, aching and stiffness endure these symptoms before seeking physician diagnosis and care has not been clearly determined. Furthermore, administrative data sources (for example, Discharge Abstract Database, pharmacy claims, practitioner claims) do not provide information about the time between symptom onset and physician diagnosis of osteoarthritis in the population, or joint-specific details, such as side or single versus multiple joints affected.

This article is based on data collected by the 2009 Survey on Living with Chronic Diseases in Canada. The primary objective is to estimate the time between symptom onset and physician diagnosis of osteoarthritis. The secondary objective is to describe management strategies employed by people with the condition.  Osteoarthritis was selected for analysis because it is the most prevalent form of arthritis.  The focus is on hip- and knee-related osteoarthritis, the joints most commonly affected.

Data and methods

Data sources

The Survey on Living with Chronic Diseases in Canada (SLCDC) is a cross-sectional survey conducted by Statistics Canada every two years in collaboration with the Public Health Agency of Canada.Note13 The 2009 SLCDC was developed and funded to gain detailed information specifically about arthritis and hypertension.Note13 The present study is based on the Arthritis component.

Data were collected in February and March 2009 using computer-assisted telephone interviewing.Note13 The sample was drawn from respondents to the 2008 Canadian Community Health Survey (CCHS), which covered the population living in private households in the 10 provinces. The 2008 CCHS excluded residents of the three territories, Indian reserves and institutions, and full-time members of the Canadian Forces. CCHS respondents aged 20 or older who reported having been diagnosed with arthritis by a medical professional, and who provided sufficient information to conduct the interview (contact information) were eligible for the Arthritis component of the SLCDC (n = 7,062). The sample was chosen using systematic sampling after the units were sorted by province, CCHS collection period, and age. The sample size was calculated to provide reliable estimates at the national level by age group and sex. A total of 5,820 CCHS respondents were selected for the SLCDC;  4,565 agreed to participate and have their responses linked to the 2008 CCHS. The overall response rate to the Arthritis component was 78.4%.Note13

The information collected in the Arthritis component included: painful joint and side of body, age at diagnosis, age at symptom onset, arthritis type, medication use (prescription and non-prescription), and use of health care services ( for example, pharmacist, physiotherapist, occupational therapist) (Appendix Table A).Note14

Study sample

Respondents with arthritis diagnosed by a physician were asked what type. They chose from a list of 14 types, including osteoarthritis; respondents could indicate more than one type. This study pertains to people who reported a physician diagnosis of osteoarthritis (n = 1,755), and no other arthritis diagnoses.


Respondents were asked about their experience of joint pain in the past month and which joint was affected. They were given a list from which they could indicate more than one painful joint. Those with hip-related osteoarthritis  (no knee pain), knee-related osteoarthritis (no hip pain), and both hip- and knee-related osteoarthritis were identified. These respondents may have also reported pain in other joints.

The time from symptom onset to diagnosis was calculated as age when diagnosed with osteoarthritis minus age when symptoms (pain, aching or stiffness) were first experienced.

Respondents were asked about their use of prescription and non-prescription drugs for their arthritis in the past month.  They were also asked if they had contacted (seen or talked to) another health care professional (pharmacist, physiotherapist or occupational therapist, class for managing arthritis) about their arthritis in the past year.Note14

Appendix Table A contains a full list of the SLCDC measures used in this analysis.

Statistical methods

Descriptive statistics (means for continuous variables, percentages for categorical variables) were used to report the prevalence of osteoarthritis (any joint), and hip-, knee-, or hip- and knee-related osteoarthritis; the mean age of symptom onset and diagnosis; medication use; and contacts with health professionals during the previous year (Appendix Tables B and C). Estimates were weighted to represent the household population with arthritis.Note13 To account for the sampling design of the SLCDC, the bootstrap method was used to estimate 95% confidence intervals (CI). All analyses were performed using SUDAAN version 10 software (Research Triangle Institute). To be included in the study sample, respondents were required to report type of arthritis and joint (non-missing data). There were no missing data; however, respondents could answer  “don’t know” to some questions.


Prevalence of osteoarthritis 

An estimated 37% of Canadians aged 20 or older who had been diagnosed with arthritis reported osteoarthritis as their only form of the condition. Of these, 12% experienced pain in their hip(s); 29%, in their knee(s), and 29%, in both (Table 1).

The prevalence of osteoarthritis (any joint) and hip- and knee-related osteoarthritis rose with advancing age (Table 1). For example, among people with arthritis, 16% of those aged 20 to 49 reported osteoarthritis (any joint), compared with 45% of those aged 65 or older. The pattern was similar for hip- and knee-related osteoarthritis.

Symptom onset and diagnosis

Nearly half (48%) of people with osteoarthritis (any joint) experienced symptoms and were diagnosed in the same year (Table 2). Another 42% had symptoms at least a year before diagnosis, and for  about 10%, symptoms emerged after they had been diagnosed. This pattern held for hip- and knee-related osteoarthritis.

Seniors were more likely than 20- to 49-year-olds to experience symptoms and be diagnosed with osteoarthritis (any joint) in the same year (55% versus 27%).

For people who experienced symptoms of osteoarthritis (any joint) before they were diagnosed, the average time to diagnosis was 7.7 years (Table 2). The length of this period ranged from 6.8 years to 9.0 years across joint pain sites, and varied by age group from 5.6 years among those aged 20 to 49 (95% CI: 4.1-7.1) to 10 years (95% CI: 7.0-13.0) among seniors.

Age at onset and age at diagnosis

The average age at which people with osteoarthritis (any joint) had been diagnosed was  50.4 years (95% CI: 49.4-51.4); the average age at which symptoms were first experienced was 47.6 years (95% CI: 46.5-48.7) (Table 3). These estimates include all individuals with osteoarthritis, regardless of the timing of events—symptoms preceded, occurred same year, or followed diagnosis. As well, the averages pertaining to osteoarthritis in the knee or hip do not account for the possibility that these people had been diagnosed with osteoarthritis in another joint before they experienced symptoms in the hip and/or knee.

Prescription and non-prescription medications

An estimated 39% of individuals with osteoarthritis (any joint) managed their symptoms with prescription medications (Table 4). Among those with both hip- and knee-related osteoarthritis, the figure was 52%. The use of prescription medications for osteoarthritis (any joint) was fairly stable across age groups, ranging narrowly from 37% to 42% (Table 5).

Two-thirds (66%) of individuals with osteoarthritis (any joint) used non-prescription medications. Among those with hip- and knee-related osteoarthritis  and hip-related osteoarthritis, the figure was 74% (Table 4).

The prevalence of non-prescription medication use for osteoarthritis (any joint) ranged from 60% at ages 20 to 49 to 70% at ages 50 to 64 (Table 5).

Seeking  professional advice

In the previous 12 months, 20% of people with osteoarthritis (any joint) had consulted a pharmacist; 22% had consulted a physiotherapist or occupational therapist; and 12% had attended an educational class to help them manage problems related to their arthritis (Table 6). The percentage who sought professional advice varied by the site of the joint pain. For example, 31% of those with both hip- and knee-related osteoarthritis consulted a pharmacist, compared with 18% of those with knee-related osteoarthritis.


The literature on osteoarthritis symptom management tends to focus on strategies led by or suggested by physicians and other health care practitioners after an individual has sought their care.Note6-12 As well, although the literature has identified many reasons why individuals delay seeking treatment, information about how long they do so is lacking.

According to the present study, the average age at symptom onset of osteoarthritis was 47.6 years, slightly higher than results of other studies, which reported symptom onset in the 40 to 45 age range.Note15-17 Based on SLCDC data, the mean time to diagnosis of osteoarthritis was almost three years later. For people who experienced symptoms before they were diagnosed, the average time between symptom onset and diagnosis was 7.7 years.

Osteoarthritis-related pain and disability are commonly perceived as part of normal aging, an assumption that can lead to patient acceptance of symptoms and no treatment.Note18,Note19 For example, Gignac  et al.Note18 found that focus group participants (with and without osteoarthritis) not only described osteoarthritis as a normal part of aging, but in some cases, reported that their physicians also viewed it this way.

Gender differences may also play a role in self-reporting functional limitations related to arthritis, as they are associated with duration of arthritis symptoms, disparities in health behaviours, household income and region of residence.Note20

However, osteoarthritis is not a normal part of agingNote1; timely and proper management of symptoms is important in delaying disability, and thereby, reducing long-term resource use.Note21,Note22

According to the Public Health Agency of Canada, one in five individuals with arthritis reported that they did not have enough information about their condition.Note23 In the present analysis, the percentage of people with osteoarthritis (any joint) who had ever taken a class to manage their arthritis was 12%. This finding suggests that an opportunity may exist to increase the use of educational programs for the management of hip and knee pain. For example, results of the 2007 to 2009 Canadian Health Measures Survey show that 24% of Canadians are obese.Note24 Obesity is a primary modifiable risk factor for osteoarthritis.  Patient education programs related to exercise, healthy diets, and strategies to avoid joint stresses have been shown to be effective for managing symptoms and improving function, and are recommended in many arthritis care guidelines.Note25-32

Treatments provided by physiotherapists can help decrease pain and disability, while improving function in individuals with osteoarthritis.Note26,Note33,Note34 As well, many arthritis care guidelines recommend treatments by physical or occupational therapists.Note26-28,Note30,Note35 Even so, the results of the SLCDC for people with arthritis of any type indicate that comparatively few sought advice from a physical or occupational therapist.Note23 The findings of the present analysis were similar—22% of individuals with osteoarthritis had consulted a physical or occupational therapist in the previous year. Furthermore, a notable 60% of individuals younger than 50 took non-prescription medications.  The relative levels of use of physical or occupational therapist services and of non-prescription medications could be influenced by the perception that symptoms are a normal part of aging, the cost of services, or a lack of extended health insurance coverage.

The main limitations of this study are the cross-sectional design of the SLCDC and the use of self-reported data, which have not been clinically validated and which are subject to recall bias. Also, it was not possible to determine if symptom onset and diagnosis involved the same joint.  For some variables, small sample sizes resulted in high variability of estimates. In addition, the generalizability of the results is limited. The data pertain only to osteoarthritis and exclude several population groups—members of the Canadian Forces, residents of the three territories and of Indian reserves, and residents of institutions. Finally, the analysis did not examine the availability and use of community programs (for example, fitness facilities, services or programs) for the management of symptoms.

Although costly and logistically difficult to implement at the population level, a longitudinal study would be helpful in understanding the factors (voluntary and involuntary) that influence the length of time between symptom onset and diagnosis of osteoarthritis.


For many people with osteoarthritis, several years elapse between symptom onset and diagnosis. This is a key period during which individuals have an opportunity to make dietary and lifestyle choices that might reduce pain, improve function and delay disability, and ultimately, reduce resource use. The present study fills a data gap by providing estimates about the length of this period, as well as post-diagnosis management strategies—information that is not available from administrative sources.


This study was funded by grants from Alberta Innovates—Health Solutions and the Canadian Institutes of Health Research/The Arthritis Society of Canada. Study funders had no involvement in the study design or collection, analysis and interpretation of the data. Deborah A. Marshall is a Canada Research Chair in Health Systems and Services Research and Arthur J.E. Child Chair in Rheumatology.

Date modified: