Nutrition Risk is Associated With Falls Risk in an Observational Study of Community-Dwelling, Rural, Older Adults

Abstract

Objectives:

Poor nutritional status is a risk factor for falls and impedes recovery from falls in older adults. The primary objective of this study was to investigate the relationship between nutrition status and fall risk over time in a cohort of older adults.

Methods:

Using an observational analytic study design, we collected demographic, fall risk, nutrition risk, food insecurity, and incident falls data from community-dwelling older Vermonters.

Results:

Data from 708 participants (70.3 years ± 6.6; 82% female) indicate a significant association between fall risk and nutrition risk (p < 0.001), fall risk and food insecurity (p < 0.001), and food insecurity and nutrition risk (p < 0.001). After adjusting for potential confounders, elevated nutrition risk was significantly associated with an incident fall over the next 6 months (p < 0.05).

Conclusion:

Given the evidence for an association between nutrition status and falls, additional research, in a more diverse population, is needed to understand the nuances of these relationships.

Introduction

The U.S. population is aging and, by 2060, the U.S. Census Bureau projects that one in four Americans will be 65 and older (Vespa et al., 2018). Falls are a leading cause of injury, death, functional impairment, and a driver of health costs and utilization in older adults (Bergen, 2016). Approximately one-third of older adults fall annually, with up to 50% reporting a fall-related injury (Bergen, 2016). Decreasing injury from falls remains a national priority; Healthy People 2030 includes an objective focusing on reducing fall-related deaths in older adults from 94.4 deaths per 100,000 in the population to 93.4 deaths per 100,000 (National Academies of Sciences, 2020).

Malnutrition impedes recovery from falls and is associated with injuries from falls in community-dwelling older adults (Chien & Guo, 2014). Yet, addressing malnutrition is challenging due to multi-factorial determinants of malnutrition in older adults. Physiological changes that occur with age, such as decreased metabolism and reductions in muscle mass and nutrient absorption, can make it difficult for older adults to meet their nutritional needs (Saffel-Shrier et al., 2019). Food insecurity, associated with malnutrition and poorer health status in community-dwelling older adults, is often related to the complex interplay between social determinants of health such as access to housing, employment, education, and services such as home meal delivery (Gundersen & Ziliak, 2015; Lee & Frongillo, 2001; Westergren et al., 2014). During the COVID-19 pandemic, community-dwelling older adults faced increased isolation and decreased access to resources, potentially increasing risk for falls and malnutrition (Fulcher, 2020; National Council on Aging, 2020; Plagg et al., 2020).

Establishing the prevalence of malnutrition in older adults is difficult due to inconsistent definitions, use of varying measurement tools and age groups, and variations in measurement settings and populations (Dwyer et al., 2019). One literature review states that the prevalence of malnutrition in community-dwelling older adults is an estimated 15%, with the greatest risk in women, people of color, people who are poor, and people who live in rural areas (Soenen & Chapman, 2013). Yet, a recent community-based workshop in Maryland for adults 65 years and older, to increase awareness of the relationship between malnutrition and falls, reported that 71% of over 400 participants were at risk for malnutrition as defined by the Seniors in the Community: Risk Evaluation for Eating and Nutrition (SCREEN) Version II (M. L. Smith, Bergeron, Lachenmayr, Eagle, & Simon, 2020).

Unmodifiable risk factors associated with falls and malnutrition are extremely relevant to Vermont’s older adult population, a state in which a majority of older adults are women and falling is the leading cause of injury for adults 65 years and old (A. S. Smith & Trevelyan, 2019; Vermont Department of Health, 2018). Vermont’s older adult population constitutes 19% of the state population, higher than the national average of 16% (United States Census Bureau, 2017). Over 11% of older adults in Vermont are eligible for the Supplemental Nutrition Assistance Program (SNAP), with a majority of those benefits claimed in rural areas (Food Research & Action Center, 2019; E. R. S. United States Department of Agriculture, 2020b). Of particular interest, Vermont has the largest population of rural older adults among all states (65% of all older adults) (A. S. Smith & Trevelyan, 2019). The Healthy People 2020 data highlights a higher rate of falls in women and in people living in rural areas (Office of Disease Prevention and Health Promotion, 2015). Despite these alarming statistics, the extent to which nutrition risk affects falls in Vermont’s community-dwelling older adults is not clear. Therefore, we sought to better understand the relationships between falls, fall risk, nutrition status, and community support over time in community-dwelling, older Vermonters.

Methods

We conducted an observational, prospective study in a cohort of older adults living in Vermont to explore the relationship between nutrition status and falls, fall risk, and food insecurity. Our primary hypotheses were that fall risk in older adults is associated with nutrition risk and that nutrition status is a predictor for future falls. Using a self-report survey, we collected data at two time points: baseline and 6 months later. This study was approved by the Northern Vermont University—Johnson Institutional Review Board.

Participants

All community-dwelling adults, 60 years and older, and residing in Vermont year-round were eligible to participate. Community-dwelling was defined as living in a house, townhouse, apartment, condominium, assisted living facility, or group home.

Recruitment

Participants were recruited through fliers, listservs, news-letters, and community forums during the fall of 2020. The primary recruitment tool was the online community forum Front Porch Forum (FPF). Front Porch Forum is a community-building service in Vermont that allows individuals to post information and view posts from others in their neighborhood (Front Porch Forum, 2021). Recruitment notices through FPF were posted for a total of 4 weeks. Initial recruitment occurred throughout the entire state of Vermont for two consecutive weeks, followed by postings targeted to low responding counties the following 2 weeks. Additionally, fliers were distributed electronically and in paper format. Organizations with outreach to older adults, such as Meals on Wheels (MOW), senior centers, hospitals, and Area Agencies on Aging shared fliers via social media posts and newsletters. Due to the COVID-19 pandemic, the distribution of paper fliers was limited to institutions such as libraries, food banks, and hospitals, which allowed in-person visitors. All outreach provided potential participants with information on the survey and a link to participate. Participants were given the option to complete the survey by phone with research staff. Paper surveys were also made available and mailed by request.

Data Collection

Data were collected during September and October 2020 (baseline) and again 6 months later during March 2021 (follow-up). Participants completed the baseline and follow-up surveys via email, phone, or mail based on preference. Study data were collected and managed using REDCap (Research Electronic Data Capture). REDCap is a secure, web-based software platform designed to support data capture for research studies (Harris et al., 2009).

The baseline survey collected demographic information and responses to three validated instruments: The Centers for Disease Control and Prevention Stopping Elderly Accidents, Deaths, and Injuries (STEADI) fall risk checklist (sensitivity = 0.65 and specificity = 0.65) (Lohman et al., 2017), Seniors in the Community: Risk Evaluation for Eating and Nutrition (SCREEN), Version III (sensitivity = 0.83 (95% CI = 0.82, 0.84); specificity = 0.73 (95% CI = 0.72, 0.74) (Morrison et al., 2019), and the two-item Food Insecurity screen (sensitivity = >97%; specificity = >70%) (Gundersen et al., 2017). Demographic information included age, gender, race and ethnicity, current housing (house, townhouse, condominium, apartment, assisted living, or group home), marital status, education, use of community nutrition programs (MOW, SNAP, senior farmer’s market benefits, food banks, and community congregate meals), chronic health conditions, and medication use. Participants self-reported rural-urban living (“How would you categorize the area of the state in which you live?”) and zip code.

Six months after completion of the baseline survey, each participant received a follow-up survey by mail, phone, or via email with a unique link, based on the mode of administration at baseline. Using the REDCap platform, the follow-up responses were matched to participant baseline data. If an email was found to be invalid and the REDCap link could not be delivered, those individuals were called. If contacted by phone, answers were entered into REDCap by a research assistant. Participants contacted via mail were sent a survey with a self-addressed and pre-paid postage return envelope. The follow-up survey differed from the baseline survey in that demographic information was not asked again. All other questions remained the same.

Data Analysis

Given that our primary hypotheses were that older adults at nutrition risk at baseline, as defined by the SCREEN III, are at a higher fall risk and that nutrition status is a predictor for future falls, our primary outcome was a fall at 6 month follow-up. Secondary outcomes were exploratory and included the relationship between nutrition status and food insecurity, whether food insecurity was a predictor of future falls, and if urban-rural living has a relationship with nutrition status, falls risk, or food insecurity. Participants that did not answer all questions, for the three validated questionnaires pertaining to nutrition risk, food insecurity, and fall risk, at both time points, were excluded.

For baseline and follow-up data, we used descriptive statistics to analyze demographic and use of community program data and chi-square tests to assess the relationships between nutrition risk, food insecurity, fall risk, falls, and rurality. We used linear regression to model the relationship between nutrition status at baseline and an incident fall in the following 6 months and tested the hypothesis by examining the odds ratio (OR) of nutrition risk and the confidence interval (CI), adjusting for demographics, health factors, and fall history. The significance level was set at p < 0.05 for all analyses a priori. Statistical analyses were conducted in STATA (Version 16.1, Stata Corp., College Station, TX).

Results

Of the 746 older adults that initiated the survey at baseline, 705 participants (95%) provided complete responses. Five-hundred and eighty-nine participants (83%) completed the required questions at follow-up. One participant was excluded from the analyses as they moved out of state during the study period. Therefore, 588 older Vermonters completed the follow-up survey. Participants that did not complete the follow-up questionnaire were not significantly different from the completers at baseline.

The average age of the cohort at baseline (N = 705) was 70 years (±6.6) and 82% were female. A majority of participants identified as non-Hispanic (99%), white (99%), married (60%), living in a house, condominium, or town-house (89%), college educated (78%), and living in a rural area (72%). Fifty-three percent of respondents were identified as being at risk for a fall, 42% reported a fall in the past 6 months, 55% were identified as being at elevated nutrition risk, and 11% were experiencing food insecurity. Thirty-seven percent were classified as having both falls and nutrition risk (Figure 1). At baseline, there was a significant relationship between fall risk and nutrition risk (p < 0.001), fall risk and food insecurity (p < 0.001), and food insecurity and nutrition risk (p < 0.001). There was no relationship between fall risk, nutrition status, or food insecurity and self-reported urban-rural living. The most commonly used resources at baseline were food banks (8%), closely followed by SNAP (7%), and community congregate meals (6%). Other less commonly used resources included Meals on Wheels (2%) and Senior Farmer’s Markets (1%). We compared self-identified rurality with Vermont Rural-Urban Commuting Area (RUCA) Codes defined by the U.S. Department of Agriculture Economic Research Service (United States Department of Agriculture & Economic Research Service 2020a). We did not find a significant difference in self-categorization and RUCA categories for urban-rural designation.

Figure 1.

Baseline fall risk and nutrition risk.

When responding to questions around what would encourage or inhibit participation in community programs for older adults at baseline, 51% of participants stated they were likely to participate in community events involving healthy activities such as balance, strength, or yoga classes and almost half (44%) indicated that weather influenced participation.

The characteristics of the sample at baseline based on nutrition risk appear in Table 1. Older adults at nutrition risk were more likely to live alone, have at least some college education, utilize community nutrition resources, have multiple chronic conditions, take multiple medications, be at risk for a fall, and to be experiencing food insecurity.

Table 1.

Baseline Demographics by Nutrition Risk (N = 705).

	Nutrition Risk
	Not at Nutrition Risk	At Nutrition Risk
Characteristic	317 (45.0%)	388 (55.0%)	p-value
Age, mean (SD)	70.03 (6.5)	70.42 (6.6)	0.18
Gender
Female	260 (82.0%)	319 (82.2%)	0.66
Male	57 (18.0%)	68 (17.5%)
Other	0	1 (0.3%)
Race
White	316 (99.7%)	380 (97.9%)	0.12
American Indian/Alaska Native	1 (0.3%)	7 (1.8%)
Asian	0	1 (0.3%)
Ethnicity
Hispanic	2 (0.6%)	3 (0.8%)	0.82
Non-Hispanic	315 (99.4%)	385 (99.2%)
Housing
House	305 (96.2%)	323 (83.2%)	<0.001
Apartment	7 (2.2%)	54 (13.9%)
Marital status
Married or living as married	237 (74.8%)	186 (47.9%)	<0.001
Separated	0 (0%)	6 (1.5%)
Divorced	49 (15.5%)	107 (27.6%)
Widowed	22 (6.9%)	48 (12.4%)
Never married	9 (2.8%)	41 (10.6%)
Self-reported rurality
Rural	227 (71.6%)	280 (72.2%)	0.87
Urban	90 (28.4%)	108 (27.8%)
Education
Some high school	1 (0.3%)	0	<0.001
High school	9 (2.8%)	24 (6.2%)
Vocational training	1 (0.3%)	11 (2.8%)
Some college	34 (10.7%)	77 (19.8%)
College	106 (33.4%)	104 (26.8%)
Graduate/Professional	166 (52.4%)	172 (44.3%)
Community resource usage
MOW	1 (0.3%)	16 (4.1%)	<0.001
SNAP	5 (1.6%)	47 (12.1%)
Senior Farmer’s market	0	9 (2.3%)
Food bank	6 (1.9%)	51 (13.1%)
Congregate meals	9 (2.8%)	30 (7.7%)
Chronic condition diagnosis
Myocardial infraction	9 (2.8%)	17 (4.4%)	<0.001
Hypertension	103 (32.5%)	166 (42.8%)
Osteo/rheumatoid arthritis	141 (44.5%)	226 (58.2%)
Osteoporosis	63 (19.9%)	92 (23.7%)
Diabetes	16 (5.0%)	65 (16.8%)
Lung disease	18 (5.7%)	57 (14.7%)
Stroke	3 (0.9%)	15 (3.9%)
Dementia/Alzheimer’s	0	1 (0.3%)
Cancer	52 (16.4%)	71 (l8.3%)
None	88 (27.8%)	56 (14.4%)
≥ 5 medications per day	47 (14.8%)	l39 (35.8%)	<0.001
At fall risk	102 (32.2%)	261 (67.3%)	<0.001
Experiencing food insecurity	6 (1.9%)	71 (18.3%)	<0.001

At the 6 month follow-up (N = 588), 48% of participants were at risk for a fall, 32% experienced a fall within the past 6 months, 61% were at nutrition risk, and 8% were experiencing food insecurity. Living in a rural part of the state was not associated with nutrition or fall risk and use of community nutrition programs did not change significantly between baseline and follow-up.

The univariate regression of baseline nutrition risk and the occurrence of a fall between baseline and 6 month follow-up demonstrated a statistically significant association, with an OR of 2.0 (95% CI 1.4, 2.9; p < 0.01) (Table 2). Potential confounders included in the multivariate model were age, gender, education (high school education or less vs. college education), marital status (married or living as married vs. single), number of chronic conditions, living in a rural part of the state, taking five or more medications per day, baseline fall history, and baseline food insecurity. After adjusting for the potential confounders, the strength of the association between nutrition risk and a new fall was attenuated, but remained statistically significant (OR 1.7; 95% CI 1.1, 2.6; p < 0.01).

Table 2.

Unadjusted and Adjusted Associations between Nutrition Risk, Food Insecurity, and Rural Residence and Incident Falls.

	Unadjusted OR (95% CI)	p Value	Adjusteda OR (95% CI)	p Value
At nutrition risk (baseline)	2.0 (1.4, 2.9)	<0.01	1.7 (1.1, 2.6)	<0.01
Experiencing food insecurity (baseline)	1.5 (0.8, 2.6)	0.18	1.1 (0.6, 2.0)	0.86
Rural residence	1.6 (1.1, 2.3)	0.03	1.3 (0.8, 2.0)	0.26

^aAdjusted for age, gender, education, marital status, urban-rural, multi-morbidity, 5 or more medications, baseline fall history, baseline food insecurity. OR, odds ratio; CI, confidence interval.

Also shown in Table 2, food insecurity and rural residence did not increase the odds of a future fall, after adjusting for covariates (OR 1.2; 95% CI 0.6–2.2; p = 0.63 and OR 1.3; 95% CI 0.8, 1.9; p = 0.29, respectively). Multivariate models adjusted for urban-rural living based on RUCA instead of self-reported urban-rural living did not change the results.

Discussion

In our cohort of community-dwelling, mostly rural, older adults, baseline nutrition risk was significantly associated with a new fall at 6 month follow-up, even after adjusting for potential confounders. At both baseline and follow-up, fall risk was associated with nutrition risk and food insecurity. To our knowledge, this is the first study to investigate the nuances of the relationship between nutrition risk, food insecurity, and fall risk in this population. These results support the notion that nutrition is an important component, and possible predictor, of fall risk and should be monitored in community-dwelling older adults.

Our findings support previous work in which poor nutrition status was associated with falls risk (Andre et al., 2013; Chien & Guo, 2014; Eckert et al., 2021; Meijers et al., 2012; Torres et al., 2015; Westergren et al., 2014). Our recent secondary analysis of a cohort of 2986 community-dwelling, older Vermonters found those at high risk of malnutrition were 66% more likely to experience a fall within a year (p = 0.01) (Eckert et al., 2021). Other research has found that adults over age 50 with poor appetite and challenges eating are more likely to fall (Lin & Chang, 2021). Additional cross-sectional studies from the Democratic Republic of the Congo (Andre et al., 2013), Sweden (Westergren et al., 2014), and the Netherlands (Meijers et al., 2012) as well as longitudinal studies from Taiwan (Chien & Guo, 2014), and France (Torres et al., 2015) found that risk for malnutrition was associated with a risk for falling. A longitudinal study conducted in Spain did not find a relationship between falls and nutrition risk; however, they did report that community-dwelling older adults who expressed a fear of falling were more likely to be malnourished (p < 0.001) (Lavedán et al., 2018). Previous studies have also identified an association between malnutrition and impaired activity, with an increased fall risk in older adults living in long term care facilities (Misu et al., 2017; Neyens et al., 2013). One observational study conducted in Canada involving community-dwelling older adults receiving home care did not find a relationship between nutrition status and risk for a fall over time (Leclerc et al., 2009). However, participants at higher nutrition risk were more likely to be lost to follow-up. Varying tools and definitions for malnutrition and fall risk complicate comparisons across research studies and community level recommendations for decreasing fall and nutrition risk in older adults are still needed.

A possible causal pathway for the relationship between nutrition status and falls could be the relationship between frailty and sarcopenia and malnutrition (Lorenzo-López et al., 2017; Robinson et al., 2017). Frailty is associated with poor health outcomes and increased mortality (Koller & Rockwood, 2013). Inadequate nutrition in the frail is very common secondary to disease, swallowing disorders, bad oral health, lack of taste and smell, social isolation, depression, chewing problems and anorexia of aging (Meijers et al., 2012). Sarcopenia, the slow, progressive loss of lean muscle mass over time (Paddon-Jones & Rasmussen, 2009), is associated with frailty, falls, fractures, functional decline, increased morbidity and mortality, and low quality of life in the elderly (Arango-Lopera et al., 2013; H. K. Kim et al., 2012; J. S. Kim, Wilson, & Lee, 2010; Volpi et al., 2012). Sarcopenia leads to a cycle in which immobilization leads to malnutrition and frailty, which leads to further risk for falls and fractures (Muhlberg & Sieber, 2004).

Although 11% of our participants at baseline were at risk for food insecurity, we did not find a relationship between food insecurity and a future fall. In Vermont, the rate of food insecurity in all households during the pandemic increased (Niles et al., 2020), but our data indicates there may have been a decrease in food insecurity in community-dwelling older adults during this time frame. Food insecurity in our cohort decreased to 8% at follow-up. This finding may be explained by the fact that during the height of the pandemic, Vermont MOW lifted their usual criteria for enrollment and all Vermonters aged 60 and older, regardless of income, were eligible to participate (Fulcher, 2020). In fact, across the country, the number of older adults utilizing MOW services during July 2020 increased by 47% (Meals on Wheels America, 2021). Home-delivered meals are associated with lower food insecurity (Lloyd & Wellman, 2015) and daily delivered meals provided by such programs as MOW have been shown to decrease fall risk (Thomas et al., 2018). Regular deliveries from MOW also decrease isolation and the resultant lack of assistance with meal preparation and feeding (Tilley, 2017).

While we did see a decrease in food insecurity during the time period, interestingly we also found an increase in nutrition risk. In our study group, nutrition risk increased by 6% between baseline and 6 months follow-up. This may reflect environmental factors related to the COVID-19 pandemic. Research indicates that depression symptoms increased across demographic groups in US adults during COVID-19 (Visser et al., 2020). Some research has shown a relationship between depression and nutritional status, particularly in rural older adults (Cabrera et al., 2007; Gougeon et al., 2015; Jung et al., 2017). In addition, depression and loneliness have been linked to poor nutrition and less physical activity (Ong et al., 2016), both of which have increased in this age group during the pandemic (Steinman et al., 2020). The pandemic and resulting life style changes had a negative effect on physical activity and nutrition in independently living, Dutch older adults (Visser et al., 2020).

The relationship between urban-rural living, falls risk, and nutrition risk is complex. The majority (72%) of our participants at baseline reported living in a rural area, but we did not identify a significant association between rurality and fall risk, nutrition status, or food insecurity. Previous Vermont-based research suggests that the relationship between rural or urban setting and physical activity could be non-linear (Bonnell et al., 2021; Troy et al., 2018). This could relate to community walkability, reliance on vehicles for transportation, the built environment, and access to recreation, among other factors (Bonnell et al., 2021; Troy et al., 2018). Participants dwelling in rural areas may be more physically active and maintain a lower body mass index (Troy et al., 2018). Additionally, access to healthy food, which could act as an intermediary between location and nutrition risk, may be influenced by the characteristics of the specific geographic region (i.e. food deserts, socioeconomic status or poverty, physical and cognitive functioning, and social support) (Hilmers et al., 2012).

Community-based events are one approach for increasing awareness of fall and nutrition risk status and learning about options to reduce these risks (Centers for Disease Control and Prevention, 2019; Karlsson et al., 2020). But strategies for implementing effective programs to decrease food insecurity and malnutrition must be tailored to the culture and individuality of communities (Saffel-Shrier et al., 2019). In this study, participants indicated interest in community events related to healthy activities, such as balance, strength, or yoga classes. However, almost half indicated that weather is a consideration when deciding to participate in community events, highlighting a major barrier specific to rural older Vermonters. Understanding disparities related to nutrition risk based on rural versus urban living is imperative. Additional barriers to adequate nutrition related to rural living include access and education (Saffel-Shrier et al., 2019). Ensuring Internet access to rural areas is an important step. Offering healthcare visits and education programs online can lessen transportation and weather barriers (M. L. Smith et al., 2020; Steinman et al., 2020). Internet access also allows rural older adults to take advantage of online grocery ordering and delivery services.

An increasing number of older adults are remaining in their homes instead of moving to retirement or assisted living communities (Roberts et al., 2018). Decreasing risk for falls and maintaining independence is vital for community-dwelling older adults. Adequate nutrition may lead to an increased ability to “age in place” since early identification and management of nutrition status is essential for identifying factors leading to impaired mobility (Meijers et al., 2012). Most importantly, malnutrition is modifiable and is reversible. Nutrition screening, as part of a falls prevention program, could lead to early identification of malnutrition or risk for malnutrition. According to national recommendations and clinical practice guidelines, older adults should be screened for fall risk and provided with recommendations and referrals based on fall risk status (Panel on Prevention of Falls in Older Persons & Society, 2011). While previous work indicates that nutrition screenings in the setting of fall risk assessment are feasible and can produce valuable information on community-dwelling older adults (Karlsson et al., 2019; Wingood et al., 2019) inclusion of nutrition assessment and interventions as part of a falls prevention program are not common (Hopewell et al., 2018).

Strengths and Limitations

While our study contributes to the growing literature exploring the relationship between falls and nutrition, it has some limitations. As our study was mostly white (99%), non-Hispanic (99%), and female (82%), the potential relationships between food insecurity, malnutrition, and falls risk may not have been adequately represented. However, our data is more closely representative of Vermont’s population, which is 94% white and 98% non-Hispanic (United States Census Bureau, 2019). While a majority of our participants were female, our results are an important addition to the literature since previous research has indicated that there may be a higher rate of falls in women, a higher rate of injury, as well as different perceptions of falls among women (Office of Disease Prevention and Health Promotion, 2015). Women living in rural areas are of particular concern for fall risk (Office of Disease Prevention and Health Promotion, 2015; Soenen & Chapman, 2013; Sohng et al., 2004; Tay et al., 2020). Further exploration of potential gender and ethnicity differences to better understand the relationship between fall risk and nutrition is needed.

Our research strategy and outcomes were undoubtedly impacted by the COVID-19 pandemic. Our reliance on listservs and social media for recruitment due to COVID-19 restrictions likely impacted our findings and may have led to lower recruitment of older adults of lower socioeconomic status. Older adults living in regions with limited broadband are also most likely underrepresented. In addition to changes to our recruitment strategy, changes to usual physical activity, food intake, and social support during the pandemic may have impacted our results. All in-person programs targeting older adults were canceled due to the pandemic, while other programs, such as MOW, saw an increase in utilization. As a result, the relationships identified between nutrition, falls, and food security, may only be reflective of this particular time period.

This study has several strengths. We examined an understudied area, contributing to the knowledge base on the relationship between nutrition status and fall risk, and providing additional evidence for the need for nutrition screening in community-dwelling older adults. We also studied a population that is unique in its high number of rural living older adults. By collecting data at two time points, we were able to determine changes in risk over time and identified an important connection between nutrition risk and risk for a future fall. While the COVID-19 pandemic impacted our recruitment strategy and required us to pivot from in-person to primarily online recruitment we enrolled older adults across the entire state.

Conclusion

In older adults, falls are preventable and nutrition status is modifiable; therefore, both are high value targets for interventions. Early identification and management of nutrition status is essential for identifying reasons for impaired mobility and implementing fall prevention strategies, ultimately allowing older adults to maintain independence as they age. As the number of older adults in our population rises, monitoring nutrition status can improve quality of life, may decrease fall risk, decrease health care spending, and reduce morbidity and mortality in older adults.