Introduction

Falls and understanding fall risk are important national health care priorities and public health concerns for Canadians. One third of Canadians fall each year [1], with fall rates increasing with increasing age [2, 3]. Prevention of falls is critical to keeping people independent and active across the lifespan, and particularly in later life. Falls have been found to be associated with an array of risk factors [4–6], and result in a plethora of serious consequences including loss of independence, disability, decline in health and well-being, decreased quality of life, pain, morbidity, institutionalization, and mortality [2, 3, 7]. Almost half of Canadians ages 20 years and older are living with a chronic condition [8], which can put them at increased risk of falls [9]. Falls are the leading cause of injury-related deaths and hospitalizations among all Canadians. Direct and indirect costs of falls are staggering, and societal and personal burden are significant [7, 10, 11].

Canada’s population continues to age [12]. Canada’s rural population is aging faster than its urban and suburban counterparts, with Canadians aged 65 and over comprising 18% of Canada’s rural population [13]. Social and built environment characteristics shape opportunities for and barriers to health promotion [14, 15], and evidence is compelling that social and built environmental characteristics affect health [16, 17]. Research on the connections between community-level factors and health has shown burden of illness to be greater for vulnerable populations, such as older adults [18] and people living in rural and remote areas [18–20].

With the expected accelerating pace of aging in Canada, health declines and challenges, such as falls and their related consequences, carry profound implications and highlight the need to understand when falls present and how they change across the lifespan. Understanding falls and fall risk can inform the timing of programs, services, and public health interventions. The reported high incidence of falls that occur with aging, as well as fall-related mortality and morbidity, underscores the importance of preventive measures and interventions for Canadian adults. No research to date has systematically and comprehensively examined age-group differences in mobility, mobility limitation, and fall profiles in Canadians, nor has any research to date examined differences in mobility, mobility limitations, and fall profiles of adults living in rural versus urban Canada. The objectives of our research were to: 1) comprehensively examine fall profiles by age group e.g., 45 to 54 years, 55 to 64 years, 65 to 74 years, 75 to 85 years; 2) compare falls profiles of adults, by age group, living in rural versus urban areas of Canada; and 3) identify any key issues specific to the rural setting (social, built environment) that affect number of falls.

Materials and methods

This was a cross-sectional study, analyzing population-based data from the Canadian Longitudinal Study on Aging (CLSA). A protocol for the planned analysis was submitted the CLSA Data and Sample Access Committee and the University of Saskatchewan Ethics Committee (Beh #250) for approval prior to accessing the data and analysis. Data were accessed on November 13. 2021 for research purposes. At no time did the authors have access to information that could identify individual participants during or after data collection.

Results

Participant demographics are described in Table 1. Of the 46,762 participants who completed the initial falls related question, “how many times have you fallen in the past 12 months?”, the vast majority reported no falls (94.4%) and a further 4.8% reported a single fall in during the preceding 12 months. Very few participants reported multiple falls (0.8%), with the number of falls ranging from 2 to 30 falls. Table 2 includes information about falls (no fall, single fall, multiple falls) with age group comparisons. Age group was not significantly associated with fall status in the smaller sized rural dwelling sample (N = 6,198) but was associated with fall status for urban dwelling participants, likely a Type I error due to the large N (35,779) for the urban sample. Younger participants in the urban-dwelling group were less likely to be in the multiple faller group and older participants were more likely to be the multiple faller group, but these differences were rather small (phi < 0.1). As there were no substantial differences between the urban and rural dwelling groups, no further results by location are reported.

Download:

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

Table 1. Sample characteristics (N = 51,338).

https://doi.org/10.1371/journal.pone.0300026.t001

Download:

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

Table 2. Number of falls by age group and rural/urban residence (N = 44,455).

https://doi.org/10.1371/journal.pone.0300026.t002

Age group was not associated with whether participants sought medical attention after a fall (p = 0.38). While the chi-square was significant (p = 0.011) for the single fall group, the effect size size (phi = 0.071) was trivial in magnitude (Table 3). In contrast age group was weakly associated with hospitalization after a fall, with participants in the older age groups more likely to report being hospitalized than younger participants. The effect size estimate was small in magnitude for both the single fall (phi = 0.12) and multiple falls groups (phi = 0.14), but only the single fall group had a significant chi-square (p < 0.001). A small magnitude association (single falls phi = 0.11; multiple falls phi = 0.18) between where falls occurred and age group was evident (Table 4), with older participants more likely to fall inside the home than outside the home relative to younger participants.

Download:

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

Table 3. Frequency of falls requiring medical attention by age group and single/multiple falls (N = 2,566).

https://doi.org/10.1371/journal.pone.0300026.t003

Download:

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

Table 4. Frequency of falls in different locations by age group and single/multiple falls (N = 2,592).

https://doi.org/10.1371/journal.pone.0300026.t004

Age group was associated (moderate effect size) with how falls occurred for both the single fall group (phi = 0.24) and the multiple fall group (phi = 0.30) (Table 5). The most pronounced age group differences were for falls while exercising. Not surprising, older participants were less likely to fall during exercise and younger participants were more likely to fall during exercise. Older participants were more likely to report falling while standing or walking than younger participants in both the single and multiple falls group. Older participants who reported a single fall were more likely to report falling on snow or ice than were younger participants, but the converse was apparent for participants in the multiple falls group, with younger participants more likely to report falling on snow or ice. Older participants who were in the single fall group were more likely to report falling from furniture, however reported falls getting in and out of the shower and bath were infrequent regardless of age group Table 6 includes the type of injury experienced during the most serious fall experienced by participants, and these were not significantly associated with age group.

Download:

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

Table 5. Frequency of falls in different manners by age group and single/multiple falls (N = 2551).

https://doi.org/10.1371/journal.pone.0300026.t005

Download:

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

Table 6. Frequency of different types of the most serious injury by age group and single/multiple falls (N = 419).

https://doi.org/10.1371/journal.pone.0300026.t006

Odds ratios for single and multiple falls for participants with chronic conditions are reported in Table 7. The likelihood of having multiple falls was substantially higher (i.e., large magnitude of effect) for participants who reported being diagnosed with memory problems or with multiple sclerosis. Moderately higher likelihood of having multiple falls was also apparent for participants diagnosed with osteoarthritis in the hands, asthma, bowel disorder, anxiety disorder, angina, rheumatoid arthritis, transient ischemic attack, kidney disease, bowel incontinence, and for the few participants who reported diagnoses of Parkinson’s disease or Alzheimer’s disease. For the following other chronic conditions there was an increased likelihood of reporting a single fall or multiple falls, with the magnitude of effect stronger for the multiple fall vs single fall groups: mood disorders, osteoarthritis of the knee, osteoarthritis of the hip, osteoporosis, urinary incontinence, lung related disorders, stroke, and epilepsy. Other significant associations (i.e., where the OR CI did not include 1) were small or trivial.

Download:

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

Table 7. Prevalence of chronic conditions and odds of single or multiple falls relative to participants who did not fall.

https://doi.org/10.1371/journal.pone.0300026.t007

Discussion

Our study has contributed important information regarding the incidence of falls and fall-related injuries in community-dwelling Canadians ages 45 to 75+ years. We sought to explore rural–urban location as a factor to advance the understanding of falls and fall rates in high-income countries. We hypothesized there may be urban-rural differences, as disparities in health status [11, 28, 29], risk factors associated with poorer health including physical inactivity [30–32], and barriers to accessing and utilizing health services and facilities and health information [33], all of which can affect falls, have been reported. While we did not find rural-urban differences in fall rates across any age group in the CLSA baseline data, it is not known whether this finding will continue to be observed as data are examined over time.

Our findings indicate that several health conditions typically not thought of to be associated with falls, such as mood disorder, anxiety disorder, and hyperthyroidism are associated with higher likelihood of multiple falls. Chronic health conditions and diseases may directly, indirectly, and synergistically increase fall risk and fall rates such as through the direct effects of the condition or disease; indirectly with the health condition affecting other physiological systems resulting in decreased physical activity and/or causing muscle weakness, sensory or other impairments and causing poor balance; or a combination of direct and indirect effects. Previous research has suggested chronic diseases and multiple pathologies to be most important predictors of falls and more important than other risk factors such as polypharmacy [34], which was not explored in the current study.

While the results related to where falls occur based on age group and the higher likelihood of falls in individuals with some of the chronic conditions were not surprising, the low incidence of falls overall and lack of difference in fall rate between age groups were unexpected. Similar to Verma and colleagues’ 2016 study of falls and fall-related injuries of adults in the United States of America [35], the number of fallers in the older age groups was lower than previously reported. Several studies have reported that about a third of community-dwelling adults 65 years of age and older report falling in one year. Much of the literature to date has focused on falls in the older adult population, while data on fall incidence among younger, working aged adults is rare [35]. Our findings and the work of Verma et al suggest falls screening and identification of individuals at high risk of falls and fall-related injuries should be comprised of a more multi-faceted pragmatic approach across the adult lifespan; for example, considering falling and falls as a universal vital sign of potential functional or other health issues, regardless of age e.g., versus only people who are 60 or 65 years and older; condition or disease e.g., versus only individuals who have a neurological condition that affects balance or muscle strength; or health care context e.g., expanding beyond settings such as acute or long-term care to public health, ambulatory and mental health care settings.

Routine inquiring about falls and fall-related injuries, for younger aged adults or those without known risk factors or health conditions, is limited, as the vast majority of falls and falls screening evidence to date is focused on older adults. For example, the American Geriatrics Society and British Geriatrics Society published a clinical practice guideline on fall risk screening, assessment, and management, with recommendations to screen all adults aged 65 years and older for fall risk annually [36]; recently published world guidelines on falls prevention and management are focused on older adults [37]; and government initiatives also suggest people aged 65 years and older be asked about their fall risk annually [38, 39], or with a significant change in clinical status [39].

Incorporating questions about falls and fall-related injuries on an annual basis or with significant change in health status, as per the guidelines for older adults, into all clinical and health visits regardless of age of the individual and care context would serve as a basis to direct more robust screening and assessment as needed. Interventions and education to increase awareness implemented early on and earlier may be useful in preventing falls as the individual ages or the health condition progresses. While recognizing that fall prevention in older adults is a priority, targeting increased awareness about falls and fall risk identification and prevention across all age groups, as well as addressing and controlling chronic health conditions and diseases earlier may also be important and beneficial public health strategies.

Identifying individuals at high-risk of fall is complex, as evidenced by the ever-increasing number of falls screening and risk tools that have been developed for older adults specifically, people with various health conditions, as well as for specific care continuum contexts e.g., acute care, inpatient care, long-term care and rehabilitation facilities. Many individuals are referred to and engage in a fall prevention program (screening, assessment, intervention) after sustaining a fall-related injury, which may be too late to be consequential in the longer-term. Due to the significant morbidity, mortality, social and economic burden of falls and fall-related injuries in the Canadian population [7, 39–43], a primary versus secondary prevention approach to fall risk and falls across the lifespan is warranted.

Conclusion

Approximately one in twenty adults in the CLSA baseline sample reported experiencing a fall in the preceding year, and about one in one hundred reported falling multiple times. Likelihood of falls was similar regardless of age or urban/rural residence, but age was associated with fall location or situation. Risk factors for single and multiple falls were consistent with previous literature, although some unexpected risk factors for multiple falls were identified (self-reported mood and anxiety disorders, hyperthyroidism). A more multifaceted approach, not based solely on older age, to screening, primary prevention, assessment and management of falls and fall-related risk factors, including chronic disease management across the lifespan is warranted to reduce the personal, social, and economic burden of falls and fall-related injuries.

Acknowledgments

This research was made possible using the data collected by the Canadian Longitudinal Study on Aging (CLSA). This research has been conducted using the CLSA datasets Baseline Tracking Dataset version 3.4, Baseline Comprehensive Dataset version 4.0, under Application Number 171011. The CLSA is led by Drs. Parminder Raina, Christina Wolfson and Susan Kirkland.

We thank Shoshanna Green, Research Coordinator, University of Saskatchewan; and Katelyn Madigan, Research Assistant, McMaster University for their editorial assistance with the manuscript.