Cardiometabolic Multimorbidity and Activity Limitation in Canada: A Cross-Sectional Study of Adults Using the Canadian Longitudinal Study of Aging (CLSA) Data

Background: Cardiometabolic multimorbidity (CM) is the diagnosis of at least two of: diabetes, stroke, or heart disease. CM is a common pattern of multimorbidity, however, the association between CM and activity limitation remains unknown. The objectives of this study were to 1) estimate the prevalence of activity limitations among Canadians with CM; and 2) quantify the association between CM and activity limitations. Methods: Using data from the Canadian Longitudinal Study on Aging (CLSA), we estimated the prevalence of CM in Canadians aged 45 to 85 (n=51,022). Multinomial logistic regression was used to quantify the association between CM and activity limitation, evaluated using the Older American Resources and Services (OARS) scale. Results: The prevalence of people living with CM and reporting any activity limitation was 27.4%, with the greatest proportion (47.9%) observed in participants living with all three cardiometabolic conditions. The multinomial odds ratio (or relative risk ratio (RRR)) of activity limitation was greatest amongst participants with all three CM conditions (any limitation: RRR = 11.229, 95% CI = 5.803 to 21.726). Of the two disease combinations, those that included stroke had the greatest odds of activity limitation (stroke and diabetes: RRR = 6.546, 95% CI = 4.436 – 9.661; stroke and myocardial infarction: RRR = 7.029, 95% CI = 4.168 – 11.853). Conclusion: Activity limitation is common amongst Canadians living with CM, and those with CM have an increased odds of reporting activity limitation relative to those with no CM conditions. The odds increase in dose-response relationship as one accumulates more CM conditions. the MOS scale of mellitus;


Introduction
With an aging population, advancing life expectancy, and increasing numbers of people with individual chronic diseases, the prevalence of multimorbidity is increasing. [1][2][3] It is estimated that 40% of Canadians between the ages of 45-85 are living with three or more chronic conditions, 4 with as many as 52% in 60 to 64 years of age range. In the United States, the prevalence of multimorbidity is as high as 70.6% in adults between the ages of 45 and 64, and 91.8% in those over 65 years of age. 5 While the economic burden of multimorbidity is substantial, with costs ranging from 2.7 times 6 to 5.3 times 7 higher than the per person annual cost of those living with only one condition, the burden of multimorbidity on health and well-being is also considerable. Several studies have shown that those living with multimorbidity have more complications of treatment beyond the effects of individual conditions, 8 lower quality of life, [9][10][11] psychological issues, 12,13 increased healthcare use, 11,14 and high levels of disability. 11,15 Cardiometabolic multimorbidity (CM) is a speci c pattern of multimorbidity de ned as the diagnosis of two or more of diabetes (DM), stroke, or myocardial infarction (MI). [16][17][18] Existing studies report a prevalence of CM ranging from 3% in adults over the age of 18, 19 to 6% in older adults. 20 Amongst Canadians, it is estimated that 3.5% (n = 467,749) of those over the age of 50 are living with CM, 21 with another 18% (n = 2.38 million) reporting an individual cardiometabolic condition, and thus at high risk of developing CM. 21 Although CM is a common pattern of multimorbidity, [22][23][24][25][26][27] existing research and clinical guidelines traditionally focus on individual diseases while ignoring the additional complexities resulting from comorbid conditions. 28,29 Thus, treatment recommendations that focus on individual conditions can result in fragmented, redundant, and often expensive care. 8 Consequently, morbidity associated with CM are poorly understood clinically, which in turn has led to unmet health and rehabilitation needs. For example, existing, yet limited research, has shown that patients living with CM typically live with poor management of their conditions. 21 Speci cally, we have recently shown associations between cardiometabolic disease onset with lower levels of physical activity and more issues with stress management. 21 Moreover, research indicates that people with CM are at higher risk of premature mortality and reduced life expectancy, compared to healthy individuals and those diagnosed with only one cardiometabolic condition. 16,[18][19][20]30,31 While research is beginning to develop an understanding of the health, [16][17][18][19][20]30 and lifestyle 21 issues associated with CM, there remains a lack of understanding of the physical and functional complications associated with CM. 32-37 Disability often immediately precedes death, can have a profound impact on an individual's quality of life, and is common sequelae after each of stroke, heart disease, and diabetes. Therefore, addressing disability issues after cardiometabolic disease onset and developing a greater understanding of disability resulting from interacting cardiometabolic diseases is imperative given the increasing numbers of people with cardiometabolic diseases and CM.
Activity limitations are one of the most common forms of disability. 38 De ned as the di culties individuals have in their ability to execute a task or action, 38 activity limitations are in uenced by health condition and personal and environmental contextual factors, and have been shown to be predictive of a person's involvement in desired life situations and roles. 39 Several recent studies have investigated associations between activity limitations and individual cardiometabolic diseases, showing limitations with physical activities 32,37 and performing activities of daily living, 33,35 as compared to people without any cardiometabolic condition.
While results of these studies generally indicate an increased risk of activity limitation with individual stroke, 33,34,36,37,40 diabetes, 35,36,40 or heart disease, 33,37 no study has investigated the association of activity limitations among individuals with multiple cardiometabolic conditions. Thus, the purpose of this population-based study is to examine activity limitation among people with CM. Our objectives are to: 1) estimate the prevalence and severity of activity limitations among Canadian adults with CM; and 2) quantify the association between individual cardiometabolic conditions and CM on activity limitations.

Methods
The reporting of the methods and results in this population-based cross-sectional study follows the Statistical Analyses and Methods in the Published Literature Guidelines.

Data source and study population
We used data from the Canadian Longitudinal Study on Aging (CLSA) 41 to address our study objectives. The CLSA data set is a nationally representative sample of Canadians recruited using the Canadian Community Health Survey on Healthy Aging, provincial health registries, and random-digit dialing. Ethics approval for CLSA data collection was obtained by thirteen research ethics boards across Canada.
Baseline data from volunteer participants was collected between 2010 to 2015 and included two cohorts: 1) data collected using self-report questionnaires administered through 60-minute computer-assisted telephone interviews (n = 21,241), and 2) data collected at designated data collection sites or at the participant's home (n = 30,097). For the purposes of this study, data from the cohorts were grouped as there was no difference in the variables of interest other than how the data was collected (Fig. 1). We included respondents between the ages of 45 and 85 who had data available for our dependent and independent variables.
Excluded from the CLSA and thus our research, were residents in the three territories, persons living on federal First Nations reserves, full-time members of the Canadian Armed Forces, individuals living in institutions, and people who were not able to respond in English or French or who had cognitive impairment. Furthermore, individuals were excluded from our analysis if any one of the CM health conditions being assessed was unknown to the participant, if the participant refused to answer the question, or if the participant did not complete any variables included in this study (n = 316).

Variables
Activity Limitation (dependent variable) -Activity limitation was evaluated using the Older American Resources and Services (OARS) scale. This 14-item questionnaire assesses an individual's functioning with Activities of Daily Living (ADLs -dressing, feeding, appearance management, walking, getting out of bed, bathing, toileting) and Instrumental Activities of Daily Living (iADLs -telephone use, travel, shopping, meal preparation, housework, taking medication, nances). Item responses produce an unadjusted OARS score. For each ADL or iADL task included in the questionnaire, participants had three possible response options (able to complete the task, able to complete the task with assistance, unable to complete the task). One question regarding toileting asked if the participant had di culty with incontinence and what was the frequency of the incontinence (never or less than once a week, once or twice a week, three times a week or more, don't know/no answer, refused to answer).
To determine the level of activity limitation, participants were rst assigned an ADL problem score based on the number of activities a participant identi ed as being able to complete with assistance (from the unadjusted OARS questionnaire) and the number of activities missing (no response provided by the participant). A participant's ability to prepare their own meals was then considered in conjunction with the ADL problem score to determine their level of activity limitation, which was categorized as: no functional impairment, mild impairment, moderate impairment, severe impairment, total impairment. For the purposes of our analysis, the severe and total impairment categories were grouped. Table 1 outlines the process by which these variables were considered to determine activity limitation. Health Condition (independent variable of interest) -To determine a diagnosis of diabetes, heart attack, or stroke, participants were asked: 1) Has a doctor ever told you that you have diabetes, borderline diabetes or that your blood sugar is high? 2) Has a doctor ever told you that you have had a heart attack or myocardial infarction? 3) Has a doctor ever told you that you have experienced a stroke or a cerebrovascular accident? Participant responses were classi ed as: yes or no.
Respondents were divided into eight mutually exclusive groups. The control group included individuals reporting the absence of all of diabetes, myocardial infarction, and stroke. Three experimental groups were developed among those reporting single cardiometabolic diseases: 1) diabetes, 2) myocardial infarction, 3) stroke. Four additional experimental groups were also created for all possible cardiometabolic disease combinations: 1) Diabetes and Stroke, 2) Diabetes and Myocardial Infarction, 3) Stroke and Myocardial Infarction, 4) Diabetes and Stroke and Myocardial Infarction.
Contextual factor covariates -Environment: (i) Social support was measured using the 19-item Medical Outcomes Study (MOS): Social Support Survey. 42 This valid and reliable questionnaire consists of four separate social support domains and scores (emotional/informational, tangible, affectionate, and positive social integration), as well as a composite social support index score. Each item is scored on a 5-point response scale (i.e., none of the time, a little of the time, some of the time, most of the time, all of the time). A mean total score is derived for each of the domains and for the composite score. We used the composite score in our analyses, in which higher scores indicate more social support. (ii) Rural or urban location of residence was identi ed via postal code data. 43 Personal: Demographic and socioeconomic variables included for study were age, biological sex (male/female), ethnic background (white, non-white), marital status (single, never married or never lived with a partner; married/living with a partner in a common-law relationship; widowed; divorced / separated), employment status (working, retired), household income (less than $20,000; $20,000 or more, but less than $50,000; $50,000 or more, but less than $100,000; $100,000 or more, but less than $150,000; $150,000 or more), and education (less than secondary school graduation; secondary school graduation, no post-secondary education; some post-secondary education; post-secondary degree/diploma).

Statistical Analyses
Weighted analysis was conducted for all analyses to ensure generalizability to the Canadian population and to account for the complex sampling method. In ation weights were used in the descriptive analysis and analytic weights in the regression analysis, as per guidelines and weights provided by the CLSA. 44 Sample characteristics were described with the number and proportion for categorical variables and mean (SD) or median (Interquartile range) for continuous variables, strati ed by health condition(s) and presented as population estimates. The prevalence of activity limitations was estimated from the number of respondents reporting none, mild, moderate, severe/total activity limitation by diagnostic group, relative to the number of respondents in that group.
A series of logistic and multinomial logistic regression models were constructed to quantify the association between activity limitations (dependent variable) and cardiometabolic conditions (independent variable). In the logistic regression models, all levels of activity limitations were combined to report 'any' limitations vs 'no' limitations (i.e., 'none' base category). In the multinomial models, there were four categories of activity limitations, including: none (base category), mild, moderate, and severe/total combined. In both logistic and multinomial regression models, cardiometabolic condition was examined in eight mutually exclusive groups, as well as combination groups: any one CM health condition, any two CM health condition combinations, all three CM health conditions. For all models, no CM health condition was the base category. All models were adjusted for personal and environmental confounding variable, known to in uence disability (i.e., age, sex, level of education, total household income, marital status, retirement status, rural/urban location, ethnicity, and the MOS scale of functional social support). Both unadjusted and adjusted multinomial odds ratios (reported as relative risk ratios (RRR)), along with 95% con dence intervals (CI), are reported.
All statistical analyses were completed using Stata S/E Version 15 software 45 using the SVY commands for survey data analysis. All statistical tests were two-sided with alpha set at 0.05.

Results
Sociodemographic characteristics are presented in Table 2   Participants with two cardiometabolic conditions had a greater odds of reporting mild, moderate, and severe/total activity limitation compared to those with no cardiometabolic condition (Fig. 2). The multinomial odds ratio (i.e., relative risk ratio) increased in all activity limitation categories with increasing number of conditions, ranging from 1.9 to 9.9 in the mild activity limitation category to 3.5 to 14.0 in the severe/total activity limitation category. Of the two disease CM combinations, those that included stroke (DM + Stroke and MI + Stroke) showed an increased odds of developing any (6.546 and 7.029, respectively), mild (5.655 and 5.802, respectively), moderate (14.550 and 14.098, respectively), and severe/total (8.115 and 13.085, respectively) activity limitations.
Participants with all three cardiometabolic conditions had the highest odds of reporting activity limitations relative to other diagnostic groups.

Discussion
Existing research on multimorbidity has de ned the condition quite broadly and focussed primarily on mortality rates. However, patients living with multimorbidity can still live with various severities of activity limitation prior to death and different patterns of multimorbidity will result in different severity of activity limitations. Our analysis was conducted to determine the prevalence and severity of activity limitation among Canadians with cardiometabolic conditions, and to examine the association between activity limitation with increasing numbers of cardiometabolic conditions. Our ndings of increasing activity limitation with increasing onset of cardiometabolic conditions, highlights the importance of both primary and secondary prevention of cardiometabolic diseases in the Canadian adult population.
Our  49,50 Issues with mood after stroke are also common with as many as 36% reporting depressive symptoms, 51 23% reporting anxiety, 52 and 25% experiencing psychological stress. 53 These motor and cognitive impairments may impact a patient's ability to engage in prevention activities (e.g., daily exercise, healthy meal preparation), resulting in the acceleration of other underlying disease process, such as diabetes and myocardial infarction. Conversely, myocardial infarction and diabetes typically spare neurological function and result in less direct activity impairment and a more gradual progression of morbidity.
The multiplicative effect of activity limitation with onset of additional cardiometabolic conditions seen in our analysis is also observed in similar studies examining the associations between CM and other health outcomes. 16,18,21 For two disease combinations, the observed odds of activity limitation were greatest at all severities when one of the two diseases included stroke. Similar results are reported in other studies. For example, Hoeymans et al. reported that each of stroke and diabetes were found to have the greatest odds of activity limitation relative to other chronic conditions, including heart disease. 40 Similarly, Hung et al.
reported that stroke, congestive heart failure, and diabetes were all associated with mobility and complex task activity limitation, 37 and Liang et al. observed that all CM health conditions were associated with an increased likelihood of ADL activity limitation. 33 Our ndings combined with previous evidence suggest an increasing importance of secondary prevention efforts after an initial diagnosis of a cardiometabolic condition. Prevention of CM will lead to fewer activity limitations, and improved independence and quality of life.

Study Limitations
Limitations of our study include the self-reported nature of disease status. For example, the question relating to diabetes asked the participants if a doctor had ever told them that they had high blood sugar. This may have resulted in an overestimation of the prevalence of diabetes as it could have included some participants with pre-diabetes. Additionally, the CLSA survey is limited in that it excluded residents from the three territories, persons living on federal First Nations reserves, full-time members of the Canadian Armed Forces, and individuals living in institutions. This may have contributed to an underestimation of CM in the Canadian population. Furthermore, the cross-sectional nature of our data limits conclusions that may be made between activity limitations and cardiometabolic conditions.

Conclusion
Activity limitations increase with increasing onset of cardiometabolic conditions and this is most obvious in those who are living with the effects of stroke. Our ndings indicate the importance of prevention in patients living with one or more cardiometabolic conditions. Preventative and/or treatment regimens that target stroke are of particular importance as the occurrence of stroke greatly increases an individual's risk of activity limitation. This is of particular importance in patients that have diabetes, have suffered a myocardial infarction, or both, as the occurrence of a stroke substantially increases the individual's risk of activity limitation. Although our analyses demonstrate a clear association between CM and activity limitation, further investigation is required to better understand the causal nature between CM and activity limitation. Availability of data and materials: The data that support the ndings of this study are available from the Canadian Longitudinal Study on Aging, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission from the Canadian Longitudinal Study on Aging.