Introduction

Older adults with low-income levels are a vulnerable population with poorer health status compared to the general population due to higher rates of chronic disease, advancing age and poverty [1]. Independently, these are all risk factors for poorer health status, but they also act synergistically. For instance, older adults have higher rates of chronic disease relative to their younger counterparts. Canadian statistics indicate that the prevalence of chronic disease in ages 65–79 is 42.7%, and in those aged more than 80, is 51.5% compared to the total population prevalence average of 29.2% [2,3]. Similarly, recently housed individuals who are vulnerable have a high prevalence of chronic disease; based on estimates from three major Canadian urban centers the prevalence was as high as 85% [4].

Seniors comprise a large proportion of the social housing population and by default have a low-income. The Ontario Non-Profit Housing Association reports that one third of individuals waiting for social housing are over age 65 [2,5,6]. These statistics have major implications for future health care provision, as Canada’s population is projected to age rapidly, and the demand for subsidized housing is increasing, with 171,360 households already on the waiting list as of 2016 [2,3,5,6]. Consequently, it is important to study the patterns of disease in vulnerable older adults living in social housing, who may have a higher burden of chronic disease. Though this population has been shown to have health care providers [7] there are some challenges with accessing the healthcare in a timely fashion [8]. Currently, there is limited evidence on the health status and healthcare needs of older adults living in social housing. Due to low literacy rates and research participation levels in low-income social housing residents, traditional population surveys are ineffectual [9]. However, understanding this growing population’s health status and needs are crucial to inform healthcare system resource allocation and efficiency.

We hypothesize that cardiometabolic diseases and their modifiable clinical risk factors (e.g. high blood pressure, high cholesterol, overweight) and modifiable behavioural risk factors (e.g. smoking) exist in greater proportions in older adults residing in social housing. This is consistent with previous research showing that hypertension and diabetes are more common in seniors generally but with higher rates in those of low socioeconomic status [10,11]. Understanding the rates of cardiometabolic risk factors in older adults living in social housing can help optimize future cardiometabolic screening and preventative interventions. Key risk factors included were behavioural (i.e., tobacco/smoking, alcohol, physical activity, diet including fatty food and fruits and vegetables consumption) and cardiometabolic diseases (i.e., overweight/obese, hypertension, diabetes, and elevated serum cholesterol). Key socio-demographic factors and known influencers of cardiometabolic risk, such as age, gender, education level, marital status and ethnicity were also considered.

In order to inform primary care initiatives to reduce health inequities, the objectives of this study were to report, for this vulnerable population of older adults residing in social housing, the:

1. proportions of cardiometabolic risk factors,
2. associations between the modifiable clinical risk factors and social determinants of health,
3. associations between cardiovascular diseases (CVD, which refers to congestive heart failure, hypertension, hypertension, atrial fibrillation, myocardial infarction and cerebrovascular accident) and diabetes, and social determinants of health,
4. associations between CVD and diabetes, and modifiable behavioural risk factors.

Methods

Results

A total of 1065 participants were included in this cross-sectional analysis. The mean age of the sample was 72.4 years (standard deviation (SD) = 8.87); 77.3% were female; 87.2% were White, 6.2% were Indigenous and 6.7% were other ethnicities. Participants typically had low education levels; 48.2% of participants had some high school and 21.0% had completed high school (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Proportion of sociodemographic determinants, clinical risk factors, and cardiometabolic diseases.

https://doi.org/10.1371/journal.pone.0301548.t001

The distributions of three self-reported modifiable clinical risk factors (elevated blood pressure (BP), raised cholesterol, and overweight/obese) were less heterogeneous compared to modifiable behavioural risk factors. A higher proportion of respondents in the over 85-year age group reported hypertension compared to those in the 55-to-64-year age group (60.6% vs 48.6%). However, a smaller proportion of respondents in the over 85-year age group reported high cholesterol compared to the 55-to-64-year age group (30.9% vs 41.9%) and those overweight (48.9% vs 71.3%). In terms of ethnicity, participants who identified as Indigenous had the highest proportion of reported overweight (72.7%) compared to 70.5% for those who were white). No differences in modifiable clinical risk factors were observed based on sex, marital status, and education (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Proportion of self-reported modifiable clinical risk factors stratified by sociodemographic determinants.

https://doi.org/10.1371/journal.pone.0301548.t002

The proportion of reported CVD was higher among the 75–84 age group compared to younger age groups and the over 85 years age group (Table 3). The proportion of individuals with diabetes was higher amongst those who lived alone (25.1% to 33.9%) as opposed to being married (18.7%) (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Proportion of self-reported cardiometabolic diseases (CVD and diabetes) stratified by sociodemographic determinants.

https://doi.org/10.1371/journal.pone.0301548.t003

Among female participants, the proportion that reported having diabetes was highest among non-drinkers (30.3%) compared to low-risk drinkers (19.8%) and high-risk drinkers (12.5%). A higher proportion of overweight individuals had diabetes (34% compared to 15.4% of non-overweight individuals) (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Proportion of cardiometabolic disease (CVD and DM) along with its 95% confidence interval stratified by modifiable behavioural risk factors.

https://doi.org/10.1371/journal.pone.0301548.t004

Over one third of individuals with CVD (33.9%) and with diabetes (39.5%) indicated their self-reported health status as poor. In terms of mobility, 29.4% of individuals with CVD and 35.3% of individuals with diabetes indicated having extreme mobility problems. A significantly higher proportion of individuals with CVD indicated having extreme self care problems (45.2%) compared with those who indicated no problems. Similarly, a significantly higher proportion of individuals with CVD indicated having extreme usual care problems (41.3%) compared with indicating no problems. In terms of pain, individuals with CVD were more likely to indicate having extreme problems (32.4%) compared to no problems (15.9%). Individuals with diabetes were also more likely to indicate extreme problems in terms of usual care (41.3%) and pain (35.8%) (Table 5).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Proportion of self—reported cardiometabolic diseases stratified by self—reported health status and Health-Related Quality of Life.

https://doi.org/10.1371/journal.pone.0301548.t005

Discussion

There are few studies examining modifiable cardiometabolic risk factors among low-income older adults residing in social housing in Ontario. In the current study we explored social determinants of health, modifiable clinical and modifiable behavioural risk factors, self-reported health status, and health-related Quality of Life determinants and their associations with cardiometabolic diseases (CVD and diabetes). Overall, our findings indicate that regardless of their CVD or diabetes status, social housing participants reported poor health behaviours, quality of life, and general health status. A significantly higher proportion of individuals with CVD indicated having extreme problems with self care, usual care, and pain. Similarly, individuals with diabetes were significantly more likely to indicate extreme problems with usual care and with pain.

Overall, these results are consistent with previous studies [9,23–25]. Compared with the average Canadian of a similar age, residents in social housing consumed fewer fruits and vegetables per day. The Canadian Community Health Survey (CCHS) suggests that less than 5% of Canadians over age 65 consume less than one serving of fruits and vegetables per day [26]. Comparatively, one third of participants in our study failed to consume fruits and vegetables even one time per day. The proportion of individuals who had CVD and added salt “sometimes” or “always” to their food was high at 38.7% and 31.9%, respectively. Although there is a lack of Canadian data for this exact measures, national statistics on sodium consumption suggest that of those aged 51–70 years, 70% of males and 31% of females consume excessive sodium, and of those aged 71 years or older, 56% of males and 24% of females consume excessive sodium (which was defined as 2300 mg per day) [15]. The CCHS data only reports that the average Canadian consumes 146 kcal per day from fast food [15]. Similarly, rates of tobacco consumption and alcohol consumption in males in our study, were higher than the general Canadian population of the same age. However, the proportion of participants who exercised for at least 30 minutes per day is similar to 40.6% of Canadians over age 65, who exercised more than 150 minutes per week also reported in the National data [15].

Reasons that populations in social housing have high rates of behavioural risk factors are many. Systematic barriers faced by individuals living in poverty may be important factors. For instance, social housing participants may lack access to healthy food choices due to geography, income, poor mobility, and social isolation [27]. Neighborhoods with lower socioeconomic status are less likely to have grocery stores or healthier fast food options [28,29]. Secondly, the few fruits and vegetables offered in areas with poor grocery store access often have higher prices [30]. Thirdly, our participants had a high proportion of pain and mobility issues, which can directly limit their ability to physically access healthy food on a regular basis. Lastly, poverty and poor health literacy may also represent potential multidirectional mechanisms for higher rates of smoking and alcohol consumption [31]. Likewise, participants had poorer health status than the average Canadian within a similar age range. Diabetes and hypertension rates were higher in our population than the national average for people aged 65 and older (28.6% vs. 17.9%; 59.1% vs. 43.7%) [15]. Similarly, a larger proportion of social housing participants were either overweight or obese (69.3% of all participants and especially people with diabetes at 79.4%). Compared to the age—matched national average (68.3%), people with diabetes in our study were more likely to be overweight or obese [15]. In terms of quality of life, social housing participants reported higher proportion of pain and mobility issues (69.7–74.1% vs. 51%; 57.9–67.9% vs. 22%).%) [17]. These findings are consistent with higher rates of chronic medical diseases, and comorbidities in social housing residents and those with lower socioeconomic status [32]. Underlying reasons may include decreased access to primary and ambulatory care in low—income populations [32–37]. Another possible reason underlying our findings is that this population may be at risk because they have markedly lower education levels. According to the 2021 Census, Ontario seniors (65 years and older) generally have high educational attainment with 47.9% having completed a post-secondary education, 28.3% a high school diploma, and only 23.8% have not graduated from high school [38]; however, 48.2% of seniors in this study had not completed high school. This is consistent with literature which has found both poorer knowledge of cardiometabolic disease risk factors and disparity in cardiac outcomes in lower-level socioeconomic groups [39,40].

Our results suggest that seniors living in social housing are a vulnerable group with poor cardiometabolic health. As such, there may be a role for cardiometabolic preventative public health programs, tailored toward this population that is distinct from those programs currently available. Previous studies have shown that interventions that target the entire community have lower rates of effectiveness for traditionally marginalized populations, and that if not targeted carefully, can exacerbate health disparities [41,42]. Thus, public health programs for this population must take participants’ lack of financial resources, poor health literacy and poor mobility into account [9]. Lastly, specific interventions that aim to diagnose and better control hypertension are needed. These public health screening programs could target older age groups and those lacking in physical activity. In terms of program development for those already diagnosed with hypertension, interventions may present a more significant challenge. A multifaceted interventional program will be needed as other factors not evaluated in our study, such as lack of access to medication may also have a role in the observed associations.

Our study has several strengths. First, our study is one of the first to examine cardiometabolic risk factors in social housing residents using validated tools. This population is “hard to reach” with low response rates [9]. The total sample size of 1065 participants was relatively large. The study was conducted in diverse Ontario cities of varying sizes accounting for different environmental factors, all of which improve the generalizability of the findings to the larger Canadian population of adults residing in social housing. The limitations of the study include consecutive sampling, which may be affected by response bias. The study utilized self—selection, meaning that participants were possibly more health literate than the average senior living in social housing. Nonetheless our sample had remarkably lower education levels than the average population. Additionally, we collected self—reported health data, and thus, diagnoses of hypertension and diabetes, and self-reported health behaviours and weight/height used to calculate BMI, may be prone to errors. In addition, the majority of our study sample was White (87%), with a very small proportion of each non-white ethnicity represented in the remaining 13%. As such, generalizability of results to non-White populations is limited. The study utilized consecutive sampling to access the intended population who are underserved and understudied [9,23] and achieve the highest coverage possible. Initially, a random sampling design was used with a predefined list of unit numbers to be reached by door-knocking. However, this approach was not successful due to the general unwillingness of residents to open their doors to researchers, even with advanced notice. Consequently, the list of random unit numbers was continuously expanded until enough responders were reached and the required sample size was achieved, effectively becoming a consecutive sample. Therefore, consecutive sampling was ultimately used successfully from the outset for the remaining buildings. This may have increased the risk of response bias in our sample whereby the characteristics of responders may be expected to differ from non-responders.

Conclusion

Overall, this study has described poor health behaviours, health status, and quality of life in social housing residents from thirty social housing buildings throughout Ontario. Poverty, poor health literacy, and social isolation may be potential mediators for these associations, but requires future research. Cardiometabolic preventative public health programs targeted toward health behaviours must take into account such factors as social and environmental circumstances to improve health outcomes in older adults living in social housing.

Supporting information