Malnutrition Risk, Rurality, and Falls among Community-Dwelling Older Adults

Abstract

Objectives

We aimed to evaluate if malnutrition and rurality are associated with fall risk and future falls in community-dwelling older adults.

Design

Prospective Cohort.

Setting

Community, Vermont.

Participants

Older adults receiving home support services who completed a health risk assessment (n=3,300; Mean age 79.6 years ±8.4, 75% female). Additional analysis was completed with a subset of 2,043 participants with two-years of consecutive health assessments.

Measurements

Fall Risk Questionnaire, DETERMINE Nutrition Risk Questionnaire, and fall history.

Results

Independently, high malnutrition risk and rurality were associated with fall risk (p<0.001) and high malnutrition risk was associated with rurality (p<0.001). After adjusting for age, sex, and physical function, individuals with high nutrition risk had a 66% increase in the odds of falling over the next year, but rurality was not significantly associated with a new fall.

Conclusion

These findings suggest that falls are associated with malnutrition risk, but the relationship between falls and rurality is less evident. Further research is needed to identify services that may best alleviate malnutrition risk in older adults and aspects of nutrition that are most protective against fall risk.

Introduction

Falls are a leading cause of injury for older adults in the U.S., resulting in an estimated 2.8 million treated injuries annually and $30.3 billion in costs (1, 2). As consequences of falls are often severe, identifying and addressing factors that increase fall risk is important to promote healthy aging.

Screening is a recommended approach to identify fall risk factors including strength, balance, vision, polypharmacy, environmental risks, and co-morbidities, including sarcopenia and frailty (3, 4, 5). Evidence from prospective studies suggest nutritional status is an independent predictor of falls in community-dwelling older adults (6, 7, 8), and yet screening for malnutrition is not currently recommended as part of fall risk assessments in the U.S.

Living in a rural environment may also impact nutrition status and fall risk (9, 10). Fall risk is mitigated through preventative screenings, exercise programs, and medical care (11) which are more challenging to access in rural communities. Similarly, limited transportation access and increased distance to grocery stores influence nutrition status (10). One study to date examined fall-related rates by urban-rural residence (2), otherwise, to our knowledge, rurality is understudied as a fall risk factor.

Decreasing fall risk and maintaining independence is vital for community-dwelling older adults and their long-term health. Given the limited research assessing the relationship between rurality, nutrition, and falls in community-dwelling, older U.S. adults, we aimed to evaluate the association of nutritional status and rurality with fall risk and incident falls in a sample of community-dwelling older adults.

Methods

We conducted a secondary analysis of health risk assessment data from Medicare beneficiaries in the Support Services at Home (SASH) program in Vermont, collected from 2017–2019. SASH is a social services organization that serves community-dwelling older adults through coordination of housing, community health, and social services. The research on this de-identified dataset was approved by the Institutional Review Board. A total of 3,751 SASH participants, age 65 and older, received services in 2017–2019. Combined data from these three years of health risk assessments resulted in 3,300 (88.0%) participants with data from at least one risk assessment (Year 1) and 2,043 participants (61.9%) with a second, consecutive-year, risk assessment (Year 2).

Measures

Demographics and Rural Residence

Demographic data provided in the limited dataset included age and sex. Using residential zip codes, participants were categorized as living in urban, large rural, or small and isolated rural categories based on Rural-Urban Commuting Area codes (RUCA) (12).

Nutrition Risk

The self-administered, 10-item DETERMINE questionnaire was used to assess risk of malnutrition (13, 14). The previously validated questionnaire has a score range of 0–21 (≥ 6: high malnutrition risk, 3–5: moderate risk, and < 3: low risk).

Fall Risk and History

The SASH health risk assessment uses the Fall Risk Questionnaire (FRQ), as included in the Centers for Disease Control “Stay Independent” brochure for self-assessment of fall risk (15). Fall risk is categorized as ‘Yes or No” based on a score of four or more on a scale range of 0–14. Fall history was assessed with the question, ‘Did you fall in last year” (Yes/No).

Functional Limitation

Participants were asked about independence with nine Activities of Daily Living (ADL) and seven Instrumental Activities of Daily Living (IADL; response options: Independent or Needs Assistance). The ADL items included grooming, toileting, feeding, transfers, ambulation, and dressing. The IADL items were telephone use, travel, shopping, meal preparation, housework, and managing medications and money. Participants who indicated independence on all of the items were categorized as ‘independent with ADLs” and ‘independent with IADLs.”

Data analysis

All statistical analyses were conducted in Stata (Version 16.0 Stata Corp., College Station, TX). We used descriptive statistics and the responses to the Year 1 health risk assessment to examine participant characteristics. We used chi-square tests to examine the association between malnutrition risk and fall risk, rurality and fall risk, and malnutrition risk and rurality. We conducted multivariate logistic regression among participants with two consecutive years of data and a second model with the subpopulation of participants without a reported fall on the Year 1 health assessment (n=1,189). For the regression models, we used a new fall reported on the Year 2 health risk assessment as the dependent variable and Year 1 malnutrition risk and urban/rural residence as independent variables. The models including the following covariates: age, gender, fall risk, and independence with completing ADLs and IADLs. Participants with incomplete data were retained in the descriptive statistics and chi-square analyses. Among participants with two consecutive years of health risk assessments, 149 did not provide a zip code for RUCA categorization and six did not complete the nutrition screen and therefore these 155 participants were not included in the regression models. Significance level was set at p<0.05 for all analyses a priori.

Results

As shown in Table 1, participants had a mean age of 79.7 (±8.4; range 65–105), were mostly female (74.7%), lived in rural communities (63.3%), and had increased fall risk at Year 1 (59.8%). Over a third of participants were at high malnutrition risk (37.3%).

Table 1.

Participant Characteristics Year 1

Variable	Mean (SD) or N (%)
N=3,300
Age (years)	79.7 (8.4)
Sex (Female)	2,465 (74.7%)
Residence
Urban	1,095 (36.7%)
Large rural	614 (20.6%)
Small and isolated rural	1,277 (42.8%)
Fall risk	1,940 (59.8%)
Fall in the last year	1,237 (39.8%)
Risk of Malnutrition
Low	638 (20.6%)
Moderate	1,305 (42.1%)
High	1,154 (37.3%)
Independent with ADLs	2,327 (70.5%)
Independent with IADLs	1,410 (42.7%)

SD=Standard Deviation, ADL=Activities of Daily Living, LADL=Instrumental Activities of Daily Living

Among participants with elevated fall risk, 46% had high malnutrition risk compared to 13% with low malnutrition risk (p<0.001; See Figure 1) and were more likely to live in isolated rural compared to urban communities (44% vs. 34%, p<0.001). Malnutrition risk was also associated with rural residence, with more participants with high malnutrition risk living in small and isolated rural communities compared to urban communities (43% vs. 30%, p<0.001).

Figure 1.

Malnutrition Risk and Fall Risk by Urban-Rural Residence among Older Adults Receiving Home Services

Among the 2,043 participants who completed two consecutive years of health risk assessments, 1,237 (39.8%) reported a fall during Year 2. In the adjusted logistic regression model, participants with high malnutrition risk had a 66% increase in odds of a new fall compared to those with low malnutrition risk (95% Confidence Interval (CI): 1.2–2.2, p=0.001; Table 2). Rural residence was not associated with an increased odds of a new fall over the next year.

Table 2.

Factors associated with an incident fall after one year

Year 1 Characteristics	Model 1 N=1,777 Odds Ratio (95% CI)	p-value	Model 2* N=1,094 Odds Ratio (95% CI)	p-value
Age	1.00 (0.98–1.01)	0.66	1.01 (1.00–1.03)	0.13
Sex	1.02 (0.81–1.23)	0.87	0.97 (0.71–1.32)	0.83
Dependent with ADL	1.11 (0.86–1.43)	0.43	1.08 (0.76–1.54)	0.66
Dependent with IADL	0.98 (0.78–1.2)	0.44	1.09 (0.81–1.49)	0.55
Fall Risk	2.22 (1.78–2.78)	<0.001	1.63 (1.23–2.17)	0.001
Nutritional Risk
Low (ref.)	1.00		1.00
Moderate	1.01 (0.77–1.34)	0.10	1.11 (0.78–1.58)	0.57
High	1.66 (1.23–2.22)	0.001	1.65 (1.12–2.42)	0.01
Rurality
Urban (ref.)	1.00		1.00
Large Rural	1.19 (0.90–1.58)	0.22	1.17 (0.81–1.70)	0.40
Small Rural	1.20 (0.96–1.50)	0.11	1.06 (0.79–1.43)	0.69

*Sub-population of participants who reported no fall in the past year on the Year 1 health risk assessment

In a sensitivity analysis of the subpopulation of participants without a reported fall at Year 1, 350 (29.4%) reported an incident fall on the second health risk assessment. The results of the multivariable logistic model were similar to Model 1 with a 65% increase in odds of an incident fall among those with high malnutrition risk (95% CI: 1.1–2.4; p=0.01; Table 2) compared to participants at low risk of malnutrition. In the adjusted model, rural residence was not associated with increased odds of falling compared to urban residence.

Discussion

To our knowledge, this is the first study in the U.S. to examine nutrition in relation to falls in community-dwelling, older adults. We identified a significant association between malnutrition risk and fall risk using data from screening assessments among older adults receiving supportive home services. After adjusting for demographic and other fall risk factors, we found that high, compared to low, malnutrition risk was associated with a significant increase in the odds of a prospective fall. We posit that nutrition contributes to fall risk through co-morbidities such as sarcopenia and frailty (16, 17), and its role in muscle and bone health (18). Food insecurity, common among Medicare beneficiaries and associated with increased fall risk, is also a likely contributor to the current findings (19).

Although both fall and malnutrition risk were independently associated with rural residence, after adjusting for function and demographic factors, residence in a small and isolated town was not associated with an incident fall in the next year. We expected to find higher risk of falls among participants living in small and isolated areas, given health disparities in rural older adults (20, 21, 22). Based on the current results, the relationship between rural residence and falls may be more complex and require additional study. To date, there is a paucity of research examining the role of urban-rural residence on falls and fall risk factors. One study found a significantly higher rate of falls among rural dwelling older adults in China compared to urban dwellers (23). In the U.S., deaths from falls among older adults varies between urban and rural locations, with older adults in small metropolitan areas reporting higher rates of fall-related deaths compared to large metropolitan and small isolated/rural areas (24). A challenge in examining urban-rural variation in falls is defining rurality, particularly in low-population states. Research by Onega et al. found that individuals’ self-report of rural classification can be discordant from RUCA-defined categories (25). Collecting self-report of rural classification in future studies may clarify the relationship between nutrition, falls, and rural residence.

Among our participants, 37% were at high malnutrition risk and 42% were classified as having moderate malnutrition risk. Currently, little data is available to understand how many older adults in the U.S. are at a high risk for malnutrition. One study documented that 60% of cognitively intact, non-critically ill older adults receiving care in an urban emergency department were malnourished or at risk for malnutrition (26). Among older adult veterans receiving home-based primary care, 15% were found to be malnourished with an additional 40% at malnutrition risk (27). Given the relatively high percentages of malnutrition risk across studies, concerted efforts to assess malnutrition risk in community-based settings will help to better quantify the prevalence of malnutrition in U.S. older adults. Understanding the prevalence can lead to improved clarity about the relationship with falls, and how nutrition-related community resources can assist with addressing both nutrition and fall risk.

Our research has several limitations. As provided, the data set did not include a comprehensive set of health indicators and fall risk factors. Therefore, other variables that may interact with or confound the association between nutrition, rurality, and falls could not be assessed or adjusted for in our analyses. Additionally, screenings provide valuable but limited data. The health risk assessment included nutrition and fall risk screenings, but these indicators do not substitute for the diagnosis of malnutrition, sarcopenia, or other conditions associated with malnutrition or falls. The sample had a mean age of 79.7 years and over half of participants reported needing assistance with IADLs. By comparison, an estimated 19% of U.S. adults aged 75 years and older need help with tasks related to IADLs (28). Therefore, the findings are not generalizable to all community-dwelling older adults; however, they do contribute to a better understanding of older adults who are at higher fall risk based on older age, decreased functional ability, and receiving services to maintain independent living.

Even with these limitations, our findings indicate the need to consider nutrition as an important variable in fall prevention research and practice, particularly in a country with an aging population and increasing fall related injuries and deaths from falls (1, 24). Community-based programs for Medicare beneficiaries should consider malnutrition as a possible predictor of falls and provide screenings related to this social determinant of health. Further research could identify services that may alleviate high malnutrition in older adults and aspects of nutrition that are protective against a new fall.

Funding

Research reported in this publication was supported by an Institutional Development Award (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under grant number P20GM103449. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NIGMS or NIH.

Ethics Declarations

This study was considered “non-human subjects” research by the University of Vermont Internal Review Board. Informed consent was deemed as not applicable for secondary analysis of the de-identified dataset.

Conflicts of interest

The authors have no potential conflicts of interest to declare with respect to the research, authorship, or publication of this article. No copyrighted materials were used in the conduct of this study or the preparation of this article.