Abstract
Objective
To examine the potential benefit of home-delivered meals for reducing frailty levels among community-dwelling older adults at risk for malnutrition.
Design
A retrospective, single-group observational approach.
Setting
One large home-delivered meal agency in the Midwest United States.
Participants
1090 community-dwelling older adults who received home-delivered meal services, funded through the Older Americans Act, between June 2020 and December 2021.
Measurement
Frailty status was measured by the Home Care Frailty Scale (HCFS) which was routinely administered by agency staff to home-delivered meal clients as part of a quality improvement project. The HCFS was administered at the start of meal services, 3-months after meals began, and 6-months after meals began.
Results
At baseline, 55.4% of clients were found to be at high risk for malnutrition. While there was a significant and consistent decline in HCFS throughout the follow-up period for both high and low nutritional risk groups, the reduction in frailty from baseline to 6-months was greater for the high nutritional risk group (Δ = −1.9; 95% CI: [−2.7, −1.1]; p < 0.001) compared to those with low nutritional risk (Δ = −1.5; 95% CI: [−2.3, −0.7]; p < 0.001). Compared to those who lived alone, clients who lived with other individuals presented with higher levels of frailty at baseline and 3-month follow-up for both low and high malnutrition risk groups.
Conclusion
Home-delivered meal clients are commonly at risk for both frailty and malnutrition. Home-delivered meal programs, which are intended to reduce malnutrition among older adults, may serve as a promising solution for reducing frailty in the vulnerable aging population.
Keywords: Home- and community-based services, Food insecurity, Aging-In-Place, Older Americans Act
1. Introduction
Malnutrition in older adults is a common condition that results from insufficient food intake as well as unmet calorie and nutrient needs [1]. As high as 60% of community-dwelling older adults are malnourished or experience malnutrition [2], increasing their susceptibility to hospitalizations, falls, poor physical function, and reduced quality of life [3,4]. Closely related to malnutrition is frailty – a prevalent condition among older adult populations that is characterized by physiological decline and a heightened vulnerability to adverse health outcomes [5]. For instance, prior evidence has noted a markedly higher risk of frailty in malnourished older adults [6], with up to two-thirds of older adults in the United States who experience both frailty and malnutrition [7]. The combination of both conditions increases susceptibility to functional impairments, such as impaired in-home mobility, decreased independence with meal preparation, and chewing and swallowing difficulties [8], resulting in unintentional weight loss, cognitive decline, and poor quality of life [1].
Home-delivered meal programs may serve as one promising solution for reducing the risk of malnutrition and frailty among community-dwelling older adults. Federally funded meal programs in the United States are mandated to provide meals that meet at least one-third of older adults’ daily dietary needs [9]. In addition to the provision of meals, home-delivered meal programs also provide older adults with the regular opportunity to socially engage with drivers, which is particularly important for older adults who live alone and may not have expansive opportunities for social interaction. Prior work has found that home-delivered meals can reduce feelings of loneliness and improve dietary intake [10,11] – both of which are risk factors for frailty – however the frailty condition has not been extensively studied among home-delivered meal clients [12].
In light of this knowledge gap, the present study was conducted to understand the potential value of home-delivered meals for reducing frailty over time (3-months and 6-months) in older adults at risk of malnutrition. To identify clients in greatest need for services, we also explored changes in frailty levels among home-delivered meal clients who lived alone compared to those who lived with other individuals (e.g., family members; non-relatives). We conclude by presenting our lessons learned for monitoring the frailty levels of home-delivered meal clients using routinely collected data in the social service setting.
2. Methods
2.1. Design and setting
This retrospective, single-group observational study was conducted in partnership with one of the largest non-profit home-delivered meal agencies in the United States. Meals provided by our partner agency were primarily funded through the Older Americans Act, Title III-C Nutrition Services Program and were delivered to the client’s home by either a paid or volunteer driver up to seven days a week. Clients received between 1–2 meals per day and were able to select options that included standard meals, vegetarian, kosher, gluten free, or mechanically altered (e.g., soft texture).
2.2. Data collection
Data were drawn retrospectively from our agency’s main electronic health record (EHR) system for clients who received meal services between June 1, 2020 and December 31, 2021. All newly enrolled clients underwent a standard, telephone-based initial eligibility assessment completed by trained staff members. During this 20−30 min assessment, staff collected self-reported information from clients about their demographic characteristics, health conditions, and nutritional risk – all of which constituted standard data often collected by home-delivered meal agencies across the U.S. Nutritional status was categorized as either “high” nutritional risk (score of 6 or greater on the Older Americans Act nutrition risk assessment) [13] or “low” nutritional risk (score of 0–5). Additionally, our partner agency also collected data on clients’ frailty status, as described below.
2.3. Measurement of frailty status
Newly enrolled home-delivered meal clients who underwent an initial eligibility assessment also underwent a frailty assessment. The Home Care Frailty Scale (HCFS) [14] was implemented by the agency as part of a larger quality improvement initiative to help agency staff identify clients in need of supplemental services (e.g., additional meals; homemaking assistance) [15]. Agency staff administered the HCFS, via telephone, at the start of meal services (baseline) and again 3- months and 6-months after meal services began. The HCFS, which was originally developed from internationally representative data drawn from the interRAI Home Care scale, assessed the following domains: function, movement, cognition and communication, social interaction, and nutritional status, Of note, nutritional status of the HCFS was scored separately from the Older Americans Act nutrition risk assessment. HCFS scores ranged from 0 to 24 with higher scores indicative of greater levels of frailty.
2.4. Analysis
Descriptive statistics of client demographic and baseline health conditions were stratified by level of nutritional risk (“low” or “high”) and included median and interquartile range (IQR) for continuous variables and frequencies and percentages for categorical variables. A repeated measures mixed effects model was used to assess the trend in total HCFS scores over time, where the correlation between time points was assumed to follow a first-order autoregressive (i.e., AR(1)) correlation structure. Mixed effects regression models allow all data to be included regardless of missingness across time. This approach was ideal for these data, as it provided the least biased estimates by inclusion of all participants [16]. Using this model, we assessed the differences in total HCFS scores between home-delivered meal clients with low and high nutritional risk and explored the predictive effect of living alone on total HCFS score. Average HCFS scores and corresponding 95% confidence intervals were provided at baseline, 3-month, and 6-month follow-up, adjusting for gender, race, and baseline age. Since there was a high prevalence of loss to follow-up in this cohort, a secondary analysis exploring risk factors for study dropout was also conducted to understand their characteristics. All analyses were performed using SAS 9.4. Research activities were approved by the Institutional Review Board at The Ohio State University (#2021E1355).
3. Results
A total of 1090 home-delivered meal clients were included in our sample (Table 1). At baseline, the median client age was 75 years with the majority being female and the predominant race being white. The most prevalent baseline health conditions in our sample were orthopedic issues, hypertension, CHF/heart failure/cardiac issues, diabetes/pre-diabetes, and arthritis.
Table 1.
Demographic and baseline health conditions of home-delivered meal clients by level of nutritional risk.
| Low risk (n = 486) | High risk (n = 604) | Total (N = 1090) | |
|---|---|---|---|
| Demographics | |||
| Male, n (%) | 204 (42.0%) | 245 (40.6%) | 449 (41.2%) |
| Marital Status, n (%) | |||
| Not Married or Separated | 314 (64.6%) | 463 (76.7%) | 777 (71.3%) |
| Married | 172 (35.4%) | 141 (23.3%) | 313 (28.7%) |
| Household Composition, n (%) | |||
| Missing | 0 (0.0%) | 2 (0.3%) | 2 (0.2%) |
| Lives alone | 220 (45.3%) | 327 (54.1%) | 547 (50.2%) |
| Lives with others | 266 (54.7%) | 275 (45.5%) | 541 (49.6%) |
| Hispanic or Latino, n (%) | 4 (0.8%) | 4 (0.7%) | 8 (0.7%) |
| Race, n (%) | |||
| Black | 82 (16.9%) | 144 (23.8%) | 226 (20.7%) |
| White | 398 (81.9%) | 443 (73.3%) | 841 (77.2%) |
| Othera | 6 (1.2%) | 17 (2.8%) | 23 (2.1%) |
| Age at baseline, median (IQR) | 79 (70, 86) | 72 (65, 81) | 75 (67, 84) |
| Baseline health conditions | |||
| Orthopedic issues, n (%) | 117 (24.1%) | 204 (33.8%) | 321 (29.4%) |
| Hypertension, n (%) | 116 (23.9%) | 132 (21.9%) | 248 (22.8%) |
| CHF/heart failure/cardiac issues, n (%) | 104 (21.4%) | 143 (23.7%) | 247 (22.7%) |
| Diabetes/Pre-diabetes, n (%) | 87 (17.9%) | 153 (25.3%) | 240 (22.0%) |
| Arthritis, n (%) | 88 (18.1%) | 106 (17.5%) | 194 (17.8%) |
| COPD/asthma/bronchitis, n (%) | 53 (10.9%) | 115 (19.0%) | 168 (15.4%) |
| Cancer, n (%) | 32 (6.6%) | 51 (8.4%) | 83 (7.6%) |
| Renal issues or dialysis, n (%) | 25 (5.1%) | 53 (8.8%) | 78 (7.2%) |
| Neurological issues or seizures, n (%) | 29 (6.0%) | 35 (5.8%) | 64 (5.9%) |
| MI/heart attack, n (%) | 7 (1.4%) | 13 (2.2%) | 20 (1.8%) |
Other race included Native American Indian/Alaskan Native, Asian, and those who identified as “other”.
High nutritional risk clients accounted for about 55% of our sample with a median age of 72. Their most prevalent baseline health conditions included orthopedic issues, diabetes/pre-diabetes, CHF/heart failure/cardiac issues, and hypertension. Similar to high nutritional risk clients, the most prevalent baseline health conditions among low nutritional risk clients included orthopedic issues, diabetes, hypertension, and CHF/heart failure/cardiac issues.
Home-delivered meal clients with high nutritional risk had higher HCFS (i.e., more severe frailty) scores at each time point compared to clients at low nutritional risk (Table 2). There was a significant and consistent decline in HCFS scores throughout the follow-up period for both nutritional risk groups, and the reduction in frailty from baseline to 6-months was greater for the high nutritional risk group (Δ = −1.9; 95% CI: [−2.7, −1.1]; p < 0.001) compared to those with low nutritional risk (Δ = −1.5; 95% CI: [−2.3, −0.7]; p < 0.001). However, the change in HCFS from baseline to 6-months was not significantly different between the two nutritional risk groups (p = 0.50).
Table 2.
Adjusted mean HCFS and 95% Confidence Interval over time stratified by level of nutritional risk, and by household composition.
| Low risk | High risk | ||||
|---|---|---|---|---|---|
| Overall | |||||
| Baseline (n = 486) |
3-months (n = 168) | 6-months (n = 79) | Baseline (n = 604) | 3-months (n = 190) | 6-months (n = 82) |
| 7.2 (5.5, 8.8) | 7.2 (5.4, 8.9) | 5.6 (3.8, 7.4) | 8.3 (6.7, 10.0) | 7.9 (6.2, 9.6) | 6.4 (4.7, 8.2) |
| Lives Alone | |||||
| Baseline (n = 220) |
3-months (n = 76) | 6-months (n = 39) | Baseline (n = 327) | 3-months (n = 111) | 6-months (n = 49) |
| 6.7 (5.0, 8.5) | 6.2 (4.4, 8.0) | 6.0 (4.0, 7.9) | 7.9 (6.3, 9.5) | 7.6 (5.9, 9.3) | 5.9 (4.0, 7.7) |
| Lives with others | |||||
| Baseline (n = 266) |
3-months (n = 92) | 6-months (n = 40) | Baseline (n = 275) | 3-months (n = 78) | 6-months (n = 32) |
| 7.7 (6.0, 9.4) | 8.1 (6.4, 9.9) | 5.4 (3.5, 7.4) | 9.1 (7.4, 10.8) | 8.4 (6.7, 10.2) | 7.5 (5.5, 9.5) |
Among high nutritional risk clients, those who lived with someone had higher HCFS at each time point compared to those who lived alone. For low nutritional risk clients, those who lived with someone had higher HCFS scores at baseline and 3-month follow-up compared to those who lived alone, while the HCFS score at 6-months was marginally lower for clients who lived with someone (Adj. mean: 5.4; 95% CI: [3.5, 7.4]) than for those who do not (Adj. mean: 6.0; 95% CI: [4.0, 7.9]).
Eighty-seven percent (n = 949) of clients missed one one or both missing follow-up phone calls to collect HCFS scores. Loss to follow up was significantly associated with living alone (p = 0.048) and race (p = 0.041). That is, those who lived alone had 30% lower odds of dropout compared to clients who lived with someone (OR: 0.7; 95% CI: 0.5–1.0) and non-white clients had 30% lower odds of dropout compared to white clients (OR: 0.7; 95% CI: 0.4–1.0). Age at baseline, nutritional risk, and gender were not found to be significant predictors of follow-up.
4. Discussion
The present study examined changes in frailty levels over time among home-delivered meal clients at risk of malnutrition. Given that we noted a significant and consistent decline in HCFS scores (i.e., an improvement in frailty status) throughout the follow-up period across our sample, home-delivered meals may serve as a promising solution for improving the frailty levels among community-dwelling older adults. However, our low retention rate limits our ability to make strong claims about the value of home-delivered meals on frailty levels. Conversely, our data do help us understand the potential needs of clients who are categorized as “high” nutrition risk. For instance, all clients in our sample had to complete the Older Americans Act nutrition risk assessment at the start of their meal services – a federal requirement for all clients who receive meals funded through the Older Americans Act, Title III-C Nutrition Services Program. Our findings indicated that clients with high nutritional risk at baseline had consistently higher HCFS scores across all time points compared to those with low nutrition risk scores. Given the literature that has confirmed the association between malnutrition and frailty [1,6], clients nationwide with high nutritional risk at the start of meal services may be more susceptible to health decline and more warrant more comprehensive services (e.g., wellness calls; referrals to other community-based programs) from their home-delivered meal agencies. The need for more comprehensive services may be particularly relevant for meal clients who live with other individuals given that their HCFS scores were consistently higher compared to clients who lived alone, and their increased frailty status may have contributed to their greater likelihood of being lost to follow-up at 3- and 6-months. Both of these characteristics – high nutritional risk and living with other individuals – may help home-delivered meal agencies “flag” clients most susceptible to adverse health outcomes and coordinate additional supportive services as able.
Notably, and consistent with national home-delivered meal data sets [17], our sample most commonly reported experiencing orthopedic issues and arthritis as well as diet-related health conditions, such as cardiovascular disease (e.g., hypertension, congestive heart failure) and diabetes, all of which have been correlated with frailty [18,19]. These health characteristics further indicate that meal clients may be highly susceptible to frailty and disease complications that threaten their ability to remain living safely at home. To proactively address these complications, there is an opportunity to enhance home-delivered meal programming through the provision of specialized services and interventions. For instance, skilled guidance from a dietitian can assist clients with blood glucose measurement and management, proper food consumption, and nutrient (e.g., sodium, carbohydrate) monitoring [20]. Additionally, with the high prevalence of orthopedic disorders present in our study sample and the well-established association between orthopedic issues (e.g., arthritis) and falls [21], consultation referrals to occupational and/or physical therapy may help reduce rates of fall-related morbidity and mortality [22].
4.1. Lessons learned
We conducted this study by leveraging data that were routinely collected by staff at our partner agency rather than using data collected specifically for research purposes. While this increased the pragmatism of our data collection procedures, it also resulted in low rates of follow-up data being obtained from clients at 3- and 6-month time points. For teams attempting to collect frailty data from home-delivered meal clients in future projects, we recommend that teams (a) select frailty instruments that can easily be administered in-person by paid or volunteer meal drivers and (b) administer frailty instruments in real-time when meals are being delivered. We attribute our low rates of follow-up to our HCFS administration procedures in that the HCFS was administered only via telephone, and the vast majority of clients were unreachable by telephone at 3- and 6-month time points.
4.2. Limitations
We recognize that our single-group, observational study design does not allow us to make conclusions about the causal effect of home-delivered meal programs on the frailty levels of clients over time. Further, a substantial proportion of our sample could not be reached for follow-up at 3-month and 6-month time points, limiting the strength and generalizability of our findings; however, observed attrition rates in this study may inform future trials to improve precision in effect estimates.
5. Conclusion
Home-delivered meal clients are commonly at risk for malnutrition and the related condition of frailty. Both conditions are multifaceted in their causes and outcomes, leading to limited understanding of effective interventions and programs to concurrently reduce malnutrition and frailty risk. Our findings suggest that home-delivered meal programs may serve as one promising solution for mitigating both malnutrition and frailty and confirms that meal clients at high nutritional risk and who live with other individuals may be particularly susceptible to health decline, jeopardizing their ability to remain living in the community.
Author contributions
LAJ led study conceptualization activities and manuscript development; MN and JMH led data analysis and interpretation; JNK assisted with interpretation of study findings and their relevance to community-based practice; LEB and MLR led data collection and processing activities; GH assisted with study conceptualization and interpretation of study findings. All authors contributed to manuscript development and refinement.
Financial disclosure
This project was supported, in part by grant number 90INNU0016, from the Administration for Community Living, US Department of Health and Human Services, Washington, D.C. 20201. Grantees undertaking projects with government sponsorship are encouraged to express freely their findings and conclusions. Points of view or opinions do not, therefore, necessarily represent official ACL policy.
Ethical approval and informed consent statement
This study was conducted in accordance with regulations set forth by the Institutional Review Board at The Ohio State University (#2021E1355; approved October 10, 2022). This was determined to be an exempt study and no informed consent was required from participants.
Conflict of interest
The authors have no conflicts of interest to disclose.
Declaration of interests
The authors have no financial interests/personal relationships to disclose.
References
- 1.Wei K., Nyunt M.S.Z., Gao Q., Wee S.L., Yap K.B., Ng T.P. Association of frailty and malnutrition with long-term functional and mortality outcomes among community-dwelling older adults. JAMA Netw Open. 2018;1(3) doi: 10.1001/jamanetworkopen.2018.0650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pereira G.F., Bulik C.M., Weaver M.A., Holland W.C., Platts-Mills T.F. Malnutrition among cognitively intact, noncritically ill older adults in the emergency department. Ann Emerg Med. 2015;65(1):85–91. doi: 10.1016/j.annemergmed.2014.07.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chan H.C.N., Fei X., Leung E.L.Y., Langston K., Marshall S., van der Meij B.S. Post-discharge consequences of protein-energy malnutrition, sarcopenia, and frailty in older adults admitted to rehabilitation: a systematic review. Clin Nutr ESPEN. 2023;54:382–397. doi: 10.1016/j.clnesp.2023.01.023. [DOI] [PubMed] [Google Scholar]
- 4.Trevisan C., Crippa A., Ek S., Welmer A.-K., Sergi G., Maggi S., et al. Nutritional status, body mass index, and the risk of falls in community-dwelling older adults: a systematic review and meta-analysis. J Am Med Dir Assoc. 2019;20(5):569–582.e7. doi: 10.1016/j.jamda.2018.10.027. [DOI] [PubMed] [Google Scholar]
- 5.Fried L.P., Ferrucci L., Darer J., Williamson J.D., Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci. 2004;59(3):M255–M263. doi: 10.1093/gerona/59.3.M255. [DOI] [PubMed] [Google Scholar]
- 6.Liang H., Li X., Lin X., Ju Y., Leng J. The correlation between nutrition and frailty and the receiver operating characteristic curve of different nutritional indexes for frailty. BMC Geriatrics. 2021;21(1):619. doi: 10.1186/s12877-021-02580-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Verlaan S., Ligthart-Melis G.C., Wijers S.L.J., Cederholm T., Maier A.B., de van der Schueren M.A.E. High prevalence of physical frailty among community-dwelling malnourished older adults-a systematic review and meta-analysis. J Am Med Dir Assoc. 2017;18(5):374–382. doi: 10.1016/j.jamda.2016.12.074. [DOI] [PubMed] [Google Scholar]
- 8.Woo J. Nutrition and frailty. J Nutr Health Aging. 2018;22(9):1025–1027. doi: 10.1007/s12603-018-1120-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Older Americans Act Of 1965 as amended by Public Law 116-131 on 3-25-2020.pdf. Accessed August 3, 2023. https://acl.gov/sites/default/files/about-acl/2020-04/Older%20Americans%20Act%20Of%201965%20as%20amended%20by%20Public%20Law%20116-131%20on%203-25-2020.pdf.
- 10.Walton K., Rosario V.A., Pettingill H., Cassimatis E., Charlton K. The impact of home‐delivered meal services on the nutritional intake of community living older adults: a systematic literature review. J Hum Nutr Diet. 2020;33(1):38–47. doi: 10.1111/jhn.12690. [DOI] [PubMed] [Google Scholar]
- 11.Thomas K.S., Akobundu U., Dosa D. More than a meal? A randomized control tria l comparing the effects of home-delivered meals programs on participants’ feelings of loneliness. J Gerontol B Psychol Sci Soc Sci. 2016;71(6):1049–1058. doi: 10.1093/geronb/gbv111. [DOI] [PubMed] [Google Scholar]
- 12.Hutchins-Wiese H.L., Walsh S.E. Frailty and nutrition risk screening in home-delivered meal clients. J Nutr Gerontol Geriatr. 2020;39(2):114–130. doi: 10.1080/21551197.2020.1719258. [DOI] [PubMed] [Google Scholar]
- 13.NSI_checklist_508 with citation.pdf. Accessed April 26, 2024. https://acl.gov/sites/default/files/nutrition/NSI_checklist_508%20with%20citation.pdf.
- 14.Morris J.N., Howard E.P., Steel K.R. Development of the interRAI home care frailty scale. BMC Geriatr. 2016;16(1):188. doi: 10.1186/s12877-016-0364-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Juckett L.A., Oliver H.V., Hariharan G., Bunck L.E., Devier A.L. Strategies for implementing the interRAI home care frailty scale with home-delivered meal clients. Front Public Health. 2023:11. doi: 10.3389/fpubh.2023.1022735. https://www.frontiersin.org/articles/10.3389/fpubh.2023.1022735 Accessed January 23, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kaplan D. SAGE; 2004. The SAGE handbook of quantitative methodology for the social sciences. [Google Scholar]
- 17.Shan M., Gutman R., Dosa D., Gozalo P.L., Ogarek J.A., Kler S., et al. A New data Resource to examine meals on wheels clients’ health care utilization and costs. Med Care. 2019;57(3):e15–e21. doi: 10.1097/MLR.0000000000000951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cook M.J., Verstappen S.M.M., Lunt M., O’Neill T.W. Increased frailty in individuals with osteoarthritis and rheumatoid arthritis and the influence of comorbidity: an analysis of the UK biobank cohort. Arthritis Care Res. 2022;74(12):1989–1996. doi: 10.1002/acr.24747. [DOI] [PubMed] [Google Scholar]
- 19.Shakya S., Bajracharya R., Ledbetter L., Cary M.P. The association between cardiometabolic risk factors and frailty in older adults: a systematic review. Innov Aging. 2022;6(5):igac032. doi: 10.1093/geroni/igac032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Juckett L.A., Hariharan G., Dodonova D.C., Klaus J., Rowe M., Burak E., et al. Implementing a community-based initiative to improve nutritional intake among home-delivered meal recipients. Nutrients. 2022;14(5):944. doi: 10.3390/nu14050944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Iijima H., Aoyama T. Increased recurrent falls experience in older adults with coexisting of sarcopenia and knee osteoarthritis: a cross-sectional study. BMC Geriatr. 2021;21(1):698. doi: 10.1186/s12877-021-02654-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Stevens J.A., Lee R. The potential to reduce falls and avert costs by clinically managing fall risk. Am J Prev Med. 2018;55(3):290–297. doi: 10.1016/j.amepre.2018.04.035. [DOI] [PMC free article] [PubMed] [Google Scholar]