To describe the exercise/physical activity and weight management efforts of Canadians with self-reported arthritis, to examine factors associated with their engagement in these strategies to help manage their arthritis, and to explore reasons for lack of engagement.
Data were from the arthritis component of the 2009 Survey on Living with Chronic Diseases in Canada. The responses (78% response rate; n = 4,565) were weighted to be representative of Canadians (ages ≥20 years) with arthritis. Logistic regression analyses were used to examine factors associated with engaging in exercise/physical activity and weight control/loss (among overweight/obese respondents) for arthritis management purposes.
Individuals with arthritis were mostly women (63%), ages ≥45 years (89%), overweight/obese (67%), married (68%), and white (87%), with postsecondary education (69%). Sixty-three percent were exercising and of those who were overweight or obese, 68% were trying to control/lose weight; only 46% were engaged in both. Having received a clinical recommendation was the factor most strongly associated with engaging in exercise/physical activity and/or controlling/losing weight. The most common reason for not exercising was a coexisting health condition/problem (22%), while the most common reason for not controlling/losing weight among those who were overweight/obese was that it was felt not to be necessary (51%).
The provision of clinical recommendations from a health professional, providing advice on safe and suitable exercises/physical activities, as well as addressing misperceptions of the need to lose weight among the overweight/obese, may facilitate engagement in these health behaviors and ultimately reduce the consequences of arthritis.
In 2007–2008, more than 4.2 million Canadians ages ≥15 years (16% of this population) reported having arthritis, the majority of whom have osteoarthritis (). With the aging of the population, this is expected to increase to approximately 7 million (20%) by 2031 (). Exercise/physical activity and maintaining a healthy weight are key nonpharmacologic management strategies of this chronic condition ().
Regular moderate to vigorous exercise/physical activity is considered safe and has a positive effect on pain and physical functioning in individuals with arthritis ([2-14]). Engaging in exercise/physical activity helps to strengthen bones and muscles, control/lose weight, improve mental health status, reduce the risk of comorbidity, and promote longevity ([15, 16]). Being overweight or obese is an important risk factor for osteoarthritis ([17-19]), and weight loss has been found to reduce pain and physical disability in individuals with osteoarthritis ([20, 21]).
Despite the well-documented benefits of exercise/ physical activity and a healthy weight, individuals with arthritis are less likely to be physically active ([1, 22]) and more likely to be overweight/obese than individuals without arthritis (). Understanding the factors associated with engaging in these self-management strategies is important ().
Using data from the arthritis component of the 2009 Survey on Living with Chronic Diseases in Canada (SLCDC) ([24, 25]), the objectives of our study were to describe the exercise/physical activity and weight management efforts in a nationally representative sample of Canadian adults with self-reported arthritis, to examine factors associated with individuals' engagement in these self-management strategies to help manage their arthritis, and to explore the reasons for lack of engagement.
Box 1. Significance & Innovations
Sixty-three percent of Canadians with arthritis reported they were exercising to help manage their condition. Of those who were overweight or obese, 68% reported they were trying to control/lose weight; however, less than half (46%) reported they were exercising and trying to control/lose weight to help manage their arthritis.
The most common reason for not exercising was a coexisting health condition/problem (22%), while the most common reason for not controlling/losing weight among those who were overweight/obese was that it was felt not to be necessary (51%).
Recipients of clinical recommendations from health professionals were more likely to report exercise/physical activity and/or weight control/loss efforts for arthritis management purposes.
Providing advice on suitable exercises/physical activities that consider an individual's limitations and addressing misperceptions of need to control/lose weight among overweight/obese individuals may help to increase engagement in these health behaviors.
SUBJECTS AND METHODS
The arthritis component of the 2009 SLCDC was developed by the Public Health Agency of Canada and Statistics Canada in consultation with an expert working group of clinicians and researchers. The survey was designed to provide new information on the impact of arthritis on Canadians and their management strategies. The SLCDC was a cross-sectional survey of adults ages ≥20 years living in private dwellings in the 10 Canadian provinces. Participants were drawn from respondents of the 2008 Canadian Community Health Survey (CCHS) who reported being diagnosed with arthritis by a health professional. The SLCDC was administered by trained personnel via a structured telephone interview (in English or French) in February and March 2009 (). To confirm respondents had professional-diagnosed arthritis, the survey included the question, “To begin, do you have arthritis, excluding fibromyalgia, that has been diagnosed by a health professional?” (with the response options “yes” and “no”).
A total of 7,062 respondents ages ≥20 years who reported having diagnosed arthritis were selected from the 2008 CCHS for the arthritis component of the 2009 SLCDC. Upon determining the out-of-scope respondents (because of death, moving outside of Canada, not having the chronic condition, etc.), 5,820 respondents were deemed eligible and contacted to take part in the 2009 SLCDC. Of these, 4,565 (78.4%) responded to the survey and consented to share their data with the Public Health Agency of Canada, Health Canada, and provincial governments and to link their 2009 SLCDC information to their 2008 CCHS responses ().
Respondents were asked the following 2 questions: “Do you currently exercise or participate in physical activities to help manage problems related to your arthritis?” (with the response options “yes” and “no”) and “Are you currently trying to control your weight or lose weight to help manage your arthritis?” (with the response options “yes” and “no”). Those that responded “no” to either of these questions were then asked the reasons why they were not exercising/participating in physical activities and/or trying to control/lose weight to help manage their arthritis. Among the many response options was that they were already doing so for other reasons.
Sociodemographic characteristics assessed included sex, age, marital status, race, education level, and body mass index (BMI). Clinical characteristics evaluated included years living with arthritis, number of painful joints in the past month, joint pain intensity (range 1–10, where 10 = “pain as bad as it can be”) in the past month, joint pain frequency (“always,” “often,” “sometimes,” “rarely,” or “never”) in the past month, fatigue (range 1–10, where 10 = “fatigue as bad as it can be”) in the past month, frequency of fatigue (“always,” “often,” “sometimes,” “rarely,” or “never”) in the past month, limitations (“a lot,” “a little,” or “not at all”) in instrumental activities of daily living (IADL) in the past month, and use of any medications (prescription or nonprescription) for their arthritis (“yes”/“no”) in the past month. In addition, knowing their arthritis type, having made contact with a health professional about their arthritis in the past 12 months, and having ever received a clinical recommendation (exercise/physical activity, weight control/loss, course/class, and/or use of an assistive device) from a health professional for arthritis management (“yes”/“no”) were assessed. These characteristics/factors were selected a priori, since most have been shown to be associated with whether or not adults with arthritis engage in exercise ([23, 26]) and weight management efforts ().
Respondents were categorized as overweight or obese based on the World Health Organization standards for BMI, calculated from self-reported weight and height (BMI 25–29.9 kg/m2 or ≥30 kg/m2, respectively) (). Polyarticular joint pain was defined as reporting ≥4 painful joints in the past month (). Severe joint pain and fatigue were defined by self-rating ≥7 of 10 and were considered frequent if experienced “always” or “often” ([30, 31]). Severe limitation in IADL was defined as “being limited a lot” in ≥3 of 5 IADL (bathing/dressing, getting around the house, household chores, errands/shopping, and recreation).
Reasons for lack of engagement
Respondents not exercising and/or managing their weight (overweight/obese individuals only) were asked the reasons why. Among the response categories included lack of willpower/self-discipline, time constraints, does not think it is important, and other reasons. Respondents were asked to indicate all relevant response categories.
To account for sample allocation and survey design, all estimates were weighted using survey weights generated by Statistics Canada in order to reflect the number of people in the Canadian population (ages ≥20 years) with arthritis (). The 95% confidence intervals were calculated using exact SEs generated through bootstrap resampling ().
Descriptive analyses were conducted comparing sociodemographic and clinical characteristics of all those engaged in exercise/physical activity (n = 2,911), and among the overweight or obese respondents, those engaged in weight control/loss (n = 1,853) and those engaged in exercise/physical activity and weight control/loss (n = 1,323) to help manage their arthritis. Those engaged in these strategies for other reasons were excluded (trying to exercise for other reasons: n = 503 [9.7%] of all respondents; controlling/losing weight for other reasons: n = 185 [6.5%] of overweight/obese respondents). Multivariate logistic regression analyses were used to estimate adjusted odds ratios (ORs) ([33, 34]) for associations between sociodemographic and clinical characteristics and engagement in these strategies for arthritis management purposes. Fewer than 11% of the data were missing. Our primary analyses were performed using respondents with complete data; we then repeated the analyses using all respondents, with multiple imputation for missing data (). Because the results were similar, only the former are reported.
Descriptive analyses were performed to examine reasons for lack of engagement in these self-management strategies. Multivariate logistic regression analysis was used to explore the association of characteristics with the perception of not needing to control/lose weight because they are already a healthy weight among overweight/obese individuals. All analyses were performed with SAS Enterprise Guide, version 5.1.
Individuals with arthritis were mostly women (63.2%), ages ≥45 years (88.6%), overweight/obese (67.3%), married (68%), and white (87%), with postsecondary education (68.7%) (Table 1).
Table 1. Sociodemographic characteristics in individuals with arthritis from the household population age ≥20 years of the 2009 Survey on Living with Chronic Diseases in Canada (n = 4,565)
Study population according to exercise/physical activity and/or weight management efforts for their arthritis
Fewer than two-thirds (63.0%) reported they were exercising to manage their arthritis (Table 2). Among those who were overweight/obese, more than two-thirds (68.0%) reported they were trying to control/lose weight, and only 45.7% were engaged in both to help manage their arthritis.
Table 2. Sociodemographic and clinical characteristics in individuals with arthritis according to their exercise and weight management status from the household population age ≥20 years of the 2009 Survey on Living with Chronic Diseases in Canada*
Exercising for arthritis
Controlling/losing weight for arthritis (overweight/obese)
Exercising and controlling/losing weight for arthritis (overweight/obese)
Engaged (n = 2,911 [63.0%])
Not engaged (n = 1,651 [37.0%])
Engaged (n = 1,853 [68.0%])
Not engaged (n = 1,014 [32.0%])
Engaged (n = 1,323 [45.7%])
Not engaged (n = 1,545 [54.3%])
Values are the proportion (95% confidence interval [95% CI]) unless indicated otherwise. All values are based on weighted data. BMI = body mass index; IADL = instrumental activities of daily living.
bHigh sampling variability (coefficient of variation between 16.6% and 33.3%).
A significantly higher proportion of those engaged in each of the 3 self-management strategy groups reported having made contact with any health professional and having received a clinical recommendation to exercise and control/lose weight compared to those not engaged. Other significant differences in respondent characteristics (sociodemographic and clinical) were found among those engaged versus not engaged; however, these differences were not common to all 3 groups (Table 2).
Factors associated with engagement in exercise/physical activity and/or weight control/loss for arthritis management purposes
Having contacted any health professional in the past 12 months about their arthritis and having received any clinical recommendation(s) from a health professional to help manage their arthritis (i.e., exercise/physical activity, weight control/loss, course/class, and/or use of an assistive device) were associated with engaging in exercise/physical activity (Table 3). Being obese, having lived with a diagnosis of arthritis for ≥10 years, and having received any clinical recommendation(s) were associated with engaging in weight control/loss. Factors associated with engaging in both self-management strategies included having made contact with any health professional and having received any clinical recommendation(s). Other factors (i.e., multiple painful joints and knowing type of arthritis) that demonstrated marginally significant associations with engagement in these self-management strategies are also shown in Table 3.
Table 3. Association between characteristics and engaging in exercise and/or weight control/loss for arthritis from the household population age ≥20 years of the 2009 Survey on Living with Chronic Diseases in Canada*
Exercising for arthritis (n = 4,562)
Controlling/losing weight for arthritis (overweight/obese; n = 2,867)
Exercising and controlling/losing weight for arthritis (overweight/obese; n = 2,868)
Values are the odds ratio (95% confidence interval), adjusted for all variables in the model. All values are based on weighted data. BMI = body mass index; N/A = not applicable; IADL = instrumental activities of daily living.
aP < 0.001.
bP < 0.01.
cP < 0.00001.
dP < 0.05.
eAny of the following: exercise/physical activity, control/lose weight, take a course/class, or use of an assistive device.
Being obese was associated with not engaging in exercise/physical activity, whereas being white (although the sample was predominately white) and having severe limitations in ≥3 IADL was associated with not attempting to control/lose weight. Finally, severe limitations in ≥3 IADL was associated with not attempting to engage in both self-management strategies.
Reasons for lack of engagement
The most common reason for not exercising to manage arthritis was a coexisting health condition/problem (21.7%), while the most common reason for not controlling/losing weight among those who were overweight/obese was that they felt they did not need to, i.e., they were already a healthy weight (51.2%) (Figure 1). Being male was associated with the perception of not needing to control/lose weight among overweight/obese individuals not engaging in weight control/loss (Table 4).
Table 4. Association between characteristics with perception of not needing to control/lose weight among overweight/obese individuals not engaging in weight control/loss from the household population age ≥20 years of the 2009 Survey on Living with Chronic Diseases in Canada (n = 1,009)
Received any clinical recommendations from a health professional
Exercise/physical activity and maintaining a healthy weight are recommended by public health and medical authorities to help manage arthritis-related symptoms and disability ([1, 36, 37]). However, many individuals with arthritis are not engaging in these self-management strategies and are less likely to do so compared to the general population. For instance, according to results from a national survey, only 41% of Canadians with arthritis were physically active during their leisure time compared to 50% of the general Canadian population (). Similarly, results from a US survey showed a lower prevalence of recommended physical activity among those with arthritis (24%) compared to the general US adult population (26.2%) ().
The factor most strongly associated with individuals' engagement in exercise/physical activity and/or weight control/loss (among those who were overweight or obese) was having received a clinical recommendation(s) from a health professional (adjusted ORs ranged from 2.7–3.0). Having coexisting health conditions/problems was identified as the most common reason for not engaging in exercise/physical activity, likely due to additional condition-specific symptoms that interfere with being physically active. The perception of not needing to control/lose weight because they were already a healthy weight among those who were overweight/obese was found to be the most common barrier to weight management efforts.
Our study population is similar to Canadians with arthritis from other population-based surveys with respect to sex and age; however, the proportion of overweight/obese patients and the proportion with postsecondary education were higher (). Given the high proportion of respondents with postsecondary education, our results may reflect the best-case scenario with respect to engagement in self-management strategies given the wealth of evidence demonstrating that those who are less educated are less healthy and are not as likely to engage in good health behaviors ([26, 38]).
Results from the Behavioral Risk Factor Surveillance System of the Centers for Disease Control and Prevention also showed clinical recommendations from a health professional to be strongly associated with exercise () and weight management efforts () among adults with arthritis. In addition to making the necessary clinical recommendations, health professionals should identify and address any reasons for their patient's lack of engagement in these health behaviors.
Given the high prevalence of coexisting conditions such as high blood pressure (34.7%), diabetes mellitus (14.4%), heart disease (14.7%), and mood and anxiety disorders (13.3%) among individuals with arthritis (), it is not surprising that having coexisting health conditions/problems that may interfere with an individual's ability to be physically active was reported to be the most common barrier to engaging in exercise/physical activity.
The impact(s) of other chronic conditions can affect the appropriate management of arthritis. Conversely, given that exercise is often the first line of treatment for other chronic conditions, arthritis may negatively impact the management of these other conditions (). For example, issues such as pain, fear, joint or muscle stiffness, fatigue, and/or impaired balance may prevent an individual from exercising or being physically active (). In such instances, health professionals should make a particular effort to recommend suitable and safe activities that consider their patient's limitations in an effort to reduce barriers (). Furthermore, health professionals could provide encouragement by educating their patients about the many health benefits of physical activity (even of minimal intensity such as standing and walking) in managing chronic conditions ([39, 41]).
Addressing misperceptions of the need to control/lose weight among those that are overweight/obese is another important consideration in the treatment of arthritis. There is a growing body of evidence regarding body weight misperception ([42-45]). To our knowledge, this is the first study to have explored this issue in a nationally representative sample of individuals with arthritis. An accurate perception of one's weight status is critical for individuals to be receptive to recommendations and counseling regarding weight maintenance/loss. Therefore, it is imperative that health professionals assess their patient's weight, educate patients about their weight status, and know when to initiate the appropriate management (). Recommended guidelines for body weight classification in adults include measuring a patient's BMI and if their BMI is >25 and <35 kg/m2, measuring their waist circumference. Waist circumference is an indicator of abdominal fat, which is associated with greater health risk than fat located in the hip/thigh area (). Professional advice about weight management strategies is especially important for individuals with arthritis, since these individuals often have physical limitations and a decreased ability to engage in some forms of physical activity ().
In addition to the routine evaluation of a patient's BMI and waist circumference, there are initiatives that advocate for the routine assessment of a patient's physical activity. For instance, Exercise is Medicine is a new initiative jointly sponsored by the American College of Sports Medicine and the American Medical Association that calls on all health care professionals to assess every patient's physical activity program at every visit (online at exerciseismedicine.com). Furthermore, Kaiser Permanente in Southern California has made a major commitment to getting their patients more active by recording physical activity as a vital sign and advocating that there is no greater indicator of an individual's health than how many minutes per week he/she exercises (). The routine assessment of patients' BMI/waist circumference and physical activity would not only increase their awareness of these aspects of their physical health, but also form the basis from which to provide recommendations, set targets/goals, track progress, and address any potential barriers that may affect their development of such health behaviors.
Other factors such as age, sex, and level of education previously have been found to be significantly associated with exercise/physical activity () and weight loss efforts (). We did not establish any definite relationships between these demographic factors (as measured) and engagement in exercise/physical activity and/or weight control/loss in our primary analysis. However, lower socioeconomic status has been shown to be related to obesity in women in developed countries such as Canada ([48, 49]), which may mediate the effect of these factors on activity and/or weight control/loss. Furthermore, the lack of an association between exercise/physical activity efforts and severity of symptoms (i.e., joint pain and fatigue), severe limitations in IADL, as well as multiple painful joints was unanticipated given the evidence that symptoms such as pain are known to limit exercise participation ([23, 50]). Potential explanations may include the fact that we did not have very sophisticated measures of severity of symptoms and limitations in IADL and/or the sample reflected a heterogeneous group in terms of type of arthritis, and the scales used to capture these clinical characteristics might be capturing very different constructs in an individual with rheumatoid arthritis versus osteoarthritis.
Our study has a number of strengths, including the large, randomly selected population-based sample and the administration of the survey by trained personnel using a structured format. However, the findings within should be considered in light of several limitations. First, the cross-sectional study design precluded an examination of how respondents' engagement in these self-management activities might have varied over time.
Second, all variables relied on self-report data; therefore, recall and social desirability biases may have resulted in misclassification of the outcome and/or explanatory variables. However, self-reported arthritis diagnoses have been shown to have high reliability, fairly high sensitivity (approaching 84%), and moderate specificity (71%) (). In addition, respondents were asked if their arthritis was diagnosed by a health professional, which has been shown to optimize specificity (). Furthermore, respondents were asked the case finding question twice, first in the 2008 CCHS interview and then in the 2009 SLCDC interview, providing further confirmation of their arthritis diagnosis ().
With respect to self-reported BMI, studies have shown that mean BMI and the prevalence of obesity are lower when calculated using self-reported (versus measured) height and weight (). As a result, the associations we found between obesity and engagement in physical activity/exercise and/or weight management strategies may be underestimated.
Regarding self-reported exercise/physical activity and weight control/loss, it remains unclear to what extent respondents' self-reported engagement in these strategies reflects true engagement. In addition, the dichotomous response option used to classify respondents' engagement does not quantify their efforts in any way; consequently, we do not know if an effective exercise dose and/or weight management strategies were used.
Self-report measures of physical activity have been reported to be both higher and lower than directly measured levels of physical activity (e.g., accelerometry), which makes it very difficult to correct for inherent error in self-report measures (). However, due to social desirability, it is plausible that a response bias exists as a result; the questions regarding engagement in this health behavior (as well as weight control/loss efforts) will include some positive responses that reflect less than optimal exercise intensity. Therefore, engagement in effective exercise in this population with arthritis may actually be lower than we report.
In surveys requiring a measure of physical activity, where physical activity is not the primary focus and more detailed measures are not feasible, the use of a single-item question has been found to have utility ([55, 56]). In order to further develop the physical activity component of the survey used for this study, future iterations could consider capturing levels of physical activity by way of including response options such as “very active,” “moderately active,” “somewhat active,” “rarely active,” or “not active at all.”
Third, the 3–14-month lag time between the 2008 CCHS and 2009 SLCDC interviews may have affected how current some variables were, such as an individual's body weight (). Misclassification of a respondent's weight is possible; however, this is unlikely to be a significant source of error because the majority of people do not lose a considerable amount of weight over such a short timeframe ().
Finally, one may question whether our results present the worse-case scenario by not including those who are engaged in exercise/physical activity and/or weight control/loss for reasons other than to manage their arthritis. However, we explored factors associated with engagement in these self-management strategies in those doing so for their arthritis and those doing so for other reasons in order to determine the impact on our results. Only a slight reduction in the strength of the main associations was found, while remaining statistically significant (data not shown).
In summary, exercise/physical activity and weight control/loss are the cornerstones of arthritis management. Recipients of clinical recommendations from health professionals are more likely to report exercise/physical activity and weight control/loss efforts to help manage their arthritis. Providing advice that considers an individual's limitations from an exercise/physical activity perspective and addressing misperceptions of the need to control/lose weight among overweight/obese individuals may help to increase engagement in these health behaviors.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Ms O'Donnell had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. O'Donnell, Rusu, Bernatsky, Hawker, MacKay, Badley.
Acquisition of data. O'Donnell.
Analysis and interpretation of data. O'Donnell, Rusu, Bernatsky, Hawker, Canizares, MacKay, Badley.