Early remnants of the Meals on Wheels (MOW) program can be traced back to the United Kingdom in 1943, where the Women's Volunteer Service for Civil Defense organized programs to provide meals to war victims when the blitz rendered many civilians homeless and unable to prepare their own meals. The idea to prepare and deliver meals to these victims evolved into modern-day programs targeted mainly to homebound seniors (Pitocco, 2016). In the United States, the first home-delivered MOW program for older adults was created in Philadelphia in 1954, with other cities such as Columbus, Ohio, and Rochester, New York quickly following. Today, there are over 5,000 programs in all 50 states (Meals on Wheels America [MOWA], 2017), and more throughout Canada, Australia, and most of Europe. It is crucial to review the literature on this topic for the following reasons: (1) There are over 10 million older adults (aged 60+) in the United States who are food insecure (MOWA), (2) This number is projected to increase as the Baby Boomer generation and their children age (Ziliak & Gundersen, 2016), and (3) There have been recent discussions at the federal level proposing possible budget cuts for these programs.
How MOW Works
Although MOW is nationwide, there is a great deal of local autonomy, and thus local variation in how specific programs operate, the services offered, eligibility requirements and restrictions, what meals are delivered, and whether there are provisions for special dietary needs. For example, some MOW programs deliver two meals a day; most only serve lunch and restrict deliveries to Monday through Friday, whereas a few provide weekend coverage. Some allow for special dietary needs, others do not. Some programs sponsor congregate feeding programs for ambulatory seniors in or near senior facilities. Many programs are not even called “Meals on Wheels” but go by other names. Almost all restrict services to persons aged 60 and up (this by virtue of the federal Older Americans Act [OAA] that helps fund these programs). These programs rely heavily on volunteers to get meals delivered to clients and virtually all provide socialization and some degree of safety checking along with home-delivered meals. Thus, the motto, “more than a meal” (Thomas et al., 2015).
The federal OAA does not officially allow means-testing. “However, while there is no means test for participation in the OAA Nutrition Programs, services are targeted to older adults with the greatest social and/or economic need with particular attention to low-income minorities” (OAA, 2002). In practice (in most agencies), the “no means test” provision is interpreted to mean that no client is denied service because of inability to pay. Many local MOW programs do have a sliding scale fee system based on income and assess a per meal fee (up to $7.50 per meal based on local program descriptions we have reviewed). Even programs that do not have a fee system make it known to clients what their meals cost and encourage donations to the extent that clients are able to pay. In an earlier national study of program participants (Ponza et al., 1996), 94% of congregate meal participants and 73% of home-delivered meals participants said they usually make some sort of contribution for their meals. However, these contributions do not fully reimburse the costs of the meals, and typically averages as low as $.20 a meal in some states, such as North Carolina (Teague, 2017).
Meals on Wheels is a benefit funded under the OAA and community block grants, which means it is not a federal entitlement. There was never any intention to feed all seniors who would otherwise be food insecure, nor is the reduction of hunger or food insecurity among seniors the overarching or primary goal. The principal goal, rather, is to provide meals to people who are unable to prepare their own meals so they can “age in place,” that is, remain in their homes as long as possible. For this reason, the program targets the “oldest old,” mainly those age 75 and older who are at a higher risk of social isolation and functional impairments. According to the Administration on Aging, which administers the OAA, “The OAA Nutrition Program targets services to vulnerable older adults who are older, poorer, sicker, at higher nutritional risk, more functionally impaired, more likely to live alone, and more likely to be a minority member than the general U.S. population” (OAA, 2002, p. 5). Although under a federal umbrella, the exact process by which clients enroll in MOW programs, and the precise services clients can expect to receive once enrolled vary from state to state and by local political jurisdictions within states.
What It Costs
In 2009, total federal funding for the Elderly Nutrition Services Program of the OAA, which includes both home-delivered and congregate meals, was $818 million (Colello, 2011). In 2014, the total was $654 million for 218 million meals served to about 2.4 million people (Mabli et al., 2017). That works out to $3.00 federal dollars per meal served and about $272 federal dollars annually per client served. Only about half (or less) of the total cost of these senior nutrition programs is borne by the federal government; the remainder comes from client fees, churches, grants, foundations, and local donors. The average true dollar cost in 2015 for a congregate meal was $10.96, and for a home-delivered meal, $11.06 (Ziegler et al., 2015). Although the total average per-meal cost is $0.37 more for a home-delivered meal than for a congregate meal, the opposite is true for per-meal paid costs. The average per-meal paid cost of a congregate meal ($9.30) is $0.30 more than that of a home-delivered meal ($9.00). The costs associated with paid labor and the value of volunteer labor are largely responsible for these differences (Ziegler et al.). Furthermore, the cost-per-meal averages vary wildly across the states. The least expensive meals are to be found in Puerto Rico, where the per-meal price of home-delivered meals is $1.93 and the priciest, in Alaska at $13.85 per meal (Colello).
About 60% of the Elderly Nutrition Services funding goes to home-delivered meals, the remainder to congregate meal programs. Because of the increase in the number of homebound older adults, the number of home-delivered meals served has increased more rapidly than the number of congregate meals. Estimates of the percentage of food-insecure, low-income seniors who participate in MOW or associated congregate feeding programs range upward to about 25% to 30% among the total older adult population. In a recent report by the Administration for Community Living, more older adults are using these programs than in the past, and they are more likely to live alone when compared with previous decades. Researchers also found that congregate and home-delivered meals make up a considerable amount of clients' daily nutrient intake (Mabli et al., 2017).
Meals on Wheels is often assumed to be in the business of eliminating food insecurity among seniors and is therefore cited as an excuse not to be too concerned about the food supply to America's elderly. Many assume food stamps and MOW have solved that problem. Although MOW strives to keep older adults living independently in their homes, it does not necessarily succeed and was never intended to succeed as a program that would eliminate food insecurity.
According to the Administration on Aging (n.d.), the Elderly Nutrition Program is intended “...to improve the dietary intakes of participants and to offer participants opportunities to form new friendships and to create informal support networks. The legislative intent is to make community-based services available to older adults who may be at risk of losing their independence.” Other goals and intentions found in reviewing various program statements focus on supporting and improving the lives of older Americans, and more specifically, fostering a community of aging service networks that help maintain quality of life and allow seniors to live independently in their homes as long as possible. Little is known about whether MOW accomplishes any of these goals, as research on the OAA nutrition program is relatively sparse (Payette & Shatenstein, 2005 ; Thomas et al., 2017 ; Wellman, 2010). Before 2017, the last comprehensive evaluation was conducted by Mathematica Policy Research (Ponza et al., 1996) over 20 years ago. Recently, in a collaboration with the Administration for Community Living, Mathematica Policy Research conducted a new evaluation and compared its findings to the earlier report (Mabli et al., 2017). Both reports cover a large number of topics but the ones most relevant to our concerns are program participation and program impacts.
Nutritional support participants are, on average, in their late 70s/early 80s, mainly female, white non-Hispanic, and tend to live alone. Most participants are poor or near-poor, with approximately 31% of congregate meal and 35% of home-delivered meals program participants living below 100% of the federal poverty guidelines, and the rest living at between 100% and 200%. About a quarter are African American or Hispanic. There are some differences between clients served in congregate settings and those who receive home-delivered meals. Home-delivered meals recipients are older, have lower educational attainment, and are more likely to be widowed (Mabli et al., 2017).
The health status of program participants underscores their vulnerability and disability. More than half the participants reported three or more chronic physical health conditions, such as high cholesterol, arthritis or rheumatism, eye conditions, and hypertension. A greater number of home-delivered meals recipients reported they are in fair or poor health, underweight, have difficulty consuming food due to oral health issues, and taking multiple medications. More specifically, 82% of home-delivered meals program recipients take three or more medications daily compared with 68% of congregate meal participants (Mabli et al., 2017). Because many of these older adults are taking multiple medications at once, it increases the likelihood of harmful reactions and interactions of medications, as well as negative effects from malnutrition on drug absorption (Boullata & Armenti, 2004).
Home-delivered meal participants have greater difficulty performing activities of daily living compared with congregate meal participants in every category. For example, 47% of home-delivered meals recipients have difficulty shopping for groceries or personal items compared with just 5.9% of congregate meal program participants. Another example would be the ability to do light housework: 45.4% of home-delivered meals recipients report difficulty in performing this activity compared with 9% of congregate meal program participants. Then, it is no surprise that congregate meal program participants report fewer incidents of falling within the past 3 months than their home-delivered meals recipient counterparts, as well as reporting less fear of falling. Therefore, congregate meal participants are considerably less functionally impaired (Mabli et al., 2017).
Evidence of Success
Reviews of program outcomes have demonstrated program success on a number of outcomes including allowing older adults to age in place, improved nutrition and health, and other benefits. In fact, findings from a systematic review of 80 articles on the topic of home-delivered meals suggest that receiving home-delivered meals has positive outcomes among participants (Campbell et al., 2015).
Aging in Place. One of the goals of home-delivered meals programs is to allow seniors to remain in their homes as opposed to being institutionalized. Here, the programs appear to be succeeding. Using data from the 2009 National Survey of Older Americans Act, Greenlee (2011) determined 93% of home-delivered meals recipients reported the meals allowed them to continue living in their own homes. Similarly, a survey conducted in 2011 by the State of Florida Department of Elder Affairs examined views of participants in a random sample of the state's home-delivered meals program. In this study, 97% said that home-delivered meals helped them continue to live independently.
Improved Nutrition and Health. Diet and nutritional findings suggest that only two thirds of older adults in both congregate meal and home-delivered meals programs eat three meals a day, with approximately 70% of home-delivered meals recipients eating alone. Additionally, many in both programs reported eating special diets due to reasons of health (mainly diabetes, cholesterol, and sodium issues), medication, religion, or cultural reasons.
By calculating the percentage contribution of program meals to participants' daily nutrient intakes from dietary recall, data suggest the meals largely contributed to the daily nutrient and caloric intake of participants. More specifically, congregate meal participants obtained 41% of their daily calories from program meals, and home-delivered meals recipients obtained 38%. Data from the 2009 OAA survey showed that 91% of program participants rate their meal service as “good” or “excellent” and 75% to 85% said they ate healthier meals as a result of their participation in these programs (Greenlee, 2011).
That Elderly Nutrition Services Program participants are better off nutritionally than nonparticipants is confirmed in a recent review of relevant literature on the topic (Zhu & An, 2013). These authors identified eight studies that met their quality, sampling, and design criteria. Of the eight, findings from six suggested MOW programs significantly improve recipients' diets, specifically in terms of quality and nutrient intake. Furthermore, these studies showed that home-delivered meals reduce food insecurity and nutritional risk among participants enrolled in the program and improved dietary adherence.
Other Benefits. As stated earlier, recipients of home-delivered meals are more functionally impaired than congregate meal participants. However, according to Thomas et al. (2018), receiving home-delivered meals actually reduces the risk of falls for older adults. More specifically, older adults who received daily delivered meals reported fewer falls than those who had received meals only once a week or who were on the waiting list for services. Therefore, although the recent report on congregate meal program participants and home-delivered meals recipients suggests that those receiving home-delivered meals report more falls, they are potentially still at less risk of falls compared with older adults receiving services only once a week or not at all. Although additional research is needed to further understand the relationship between receiving meals and falls, this research provides a meritorious foundation. Additional positive outcomes among participants were: increased opportunities for socialization, improved dietary adherence, and an overall higher quality of life (Zhu & An, 2013).
Dissenting Opinions. A dissenting view was expressed in an earlier comprehensive review of the literature by Krassie et al. (2000). Rather than compare diets of participants and nonparticipants, these authors compared home-delivered meals to extant nutritional standards and found that the nutritional adequacy of MOW meals varied according to the following aspects: nutritional standards, menu selection, portion control, level of consumption, and overall customer satisfaction. One interesting variable analyzed in this report was “meal utilization,” which is to say, how much and what of the delivered meal was actually consumed. One study cited in the report showed that on average, only 81% of the energy content of the meal was actually consumed; the other 19% was wasted (Fogler-Levitt et al., 1995). Fogler-Levitt et al. argued that poor taste, unpopular cooking method, disagreeable texture, and unfamiliarity were reasons that older adults did not utilize MOW programs.
Krassie et al. (2000) and De Graaf et al. (1990) took up the issue of consumer satisfaction and raised the important methodological caveat that surveys may overestimate satisfaction because respondents may fear loss of benefits if they express overly critical or unappreciative opinions. In an examination of the predictors of satisfaction with home-delivered meals among 209 home-delivered meals recipients in an unspecified U.S. location, food quality was found to be the most important predictor of meal satisfaction (Joung et al., 2011).
In summary, it is evident from several lines of research that MOW and the associated congregate feeding programs have been properly targeted (i.e., reach the populations they were intended to reach), are popular with recipients, make a measurable difference in the nutritional status of recipients, promote socialization, keep mobility-limited persons in their homes and out of institutions, and increase physical, mental, and social well-being among vulnerable seniors. What more could one ask of any social program, most of all a program that only costs the federal government a billion or so dollars a year?
Current Issues Impacting MOW
There are quite a number of issues and controversies surrounding the OAA nutrition programs. Some of the more important are noted and discussed below:
Waiting Lists. Critics often respond to the 5-day-per-week home-delivered lunch programs with an impatient query, What about breakfast and dinner? What about the weekends? And there is no doubt these are serious questions in light of consistent findings that the daily lunch often represents half or more of a recipient's total nutritional intake for the day. But currently, more than 70% of MOW programs across the nation report they have a waiting list (Thomas et al., 2015) with the wait time somewhere between several months and a year (Gualtieri & Donley, 2016). How do seniors on the waiting lists feed themselves while they are waiting?
Thomas et al. (2015) studied a national sample of persons on MOW waiting lists and reported that, in comparison to the national population of community-dwelling seniors, the wait-listed are more likely to be widowed, less educated, older, Black, Hispanic, and receive Medicaid. Furthermore, 31% of seniors on MOW waiting lists reported being depressed, compared with 12% of seniors in the national population, and 28% of those examined displayed signs and symptoms of anxiety, compared with 10% of the national population of older adults. With regard to falls, the older adults on waiting lists were significantly more likely to have fallen in the last month and/or have a fear of falling compared with the national population of seniors. But these are also, in general, the characteristics of clients who receive MOW services (e.g., Greenlee, 2011) so the proper conclusion is that the entire MOW population, both those receiving and those waiting to receive services are a frail, vulnerable, high-risk group.
Gualtieri and Donley (2016) interviewed 21 people on the waiting list of the Orange County, Florida MOW program to determine how they feed themselves while they are waiting for services. A consistent theme throughout the interviews was that although finances do contribute to senior food insecurity, a lack of transportation or health and mobility issues also play an important role. Gualtieri and Donley also learned that, although some seniors use local food pantries to supplement purchased food, most do not because they are unaware of the location of these pantries. For the latter group, local grocery stores are their only source of food. Complaints about the prices of food in the local groceries were common, and although most do what they can to save money (use coupons, buy items on sale, etc.), all the respondents said they do not get enough food when going to the grocery store. Some stretch the food they do have by reducing portions or skipping meals. Almost all relied on others to either bring them food or take them to the grocery store or food pantries. They worried more about the willingness and availability of persons in their social network than about running out of money. This is assuredly not to argue that their low, fixed incomes were not a problem. The point, rather, is that even when they had money or food stamps to use, they still had to worry about getting to the store or to food pantries.
The study results also make it clear that seniors on the MOW waiting lists, particularly those whose assessment scores mark them as critical or high priority, are seriously in need of food assistance. The most galling thing is that the people on these waiting lists have already been identified as food insecure; their locations and contact information are known; their need for assistance is obvious. Yet still they wait. Putting adequate resources into MOW programs around the nation to at least clear existing waiting lists would be a good start on a sensible national policy to feed the nation's older adults.
The existence of waiting lists confirms that the demand for home-delivered meals surpasses the supply, and this will only worsen as the Baby Boom generation ages into retirement. We can confidently predict the demand for MOW programs will continue to increase for another two decades, as we know the prevalence of food insecurity among older adults is projected to increase over time (Ziliak & Gundersen, 2016). Moreover, life expectancy continues to increase (Abramson, 2015) so the length of time seniors will need food assistance is also increasing. With vast increases in funds available for MOW programs unlikely, some analysts have begun to wonder what MOW needs to do differently to keep pace with the demand.
Staffing Innovations and Consequences
Among the more serious and pressing issues are these:
- All MOW programs rely on volunteers to deliver services, but “ironically, as the need for service grows, there has been a drop in volunteers” (Winterton et al., 2013, p. 142). Historically, MOWs and most other nonprofits have been very passive and reactive about volunteers—generally, they take what walks through the door. Sensing the growing inadequacy of this passive approach, many MOWs “have moved from a volunteer-run service towards new funding and operational models. Others have developed innovative approaches to funding or recruiting less traditional sources of volunteers” (Winterton et al., p. 143).
- More and more MOWs are looking to commercial entities to prepare and deliver meals; in most cases, a hybrid mix of volunteers and commercial providers ensues. It is probably true that the commercial sector can prepare and deliver meals more efficiently than an all-volunteer operation can, but this conflicts with the next bullet point.
- The social dimensions of MOW. It is clear that MOW clients derive as much benefit from the increased social interaction as they do from the meals themselves, and it would be tragic to lose sight of this fact (Mabli et al., 2017).
Perhaps there are better ways to encourage broader community approaches to create social capital. Among the innovations that are being tested in MOWs around the world are: Privatization of meal preparation, particularly frozen meals; cost-sharing between program and clients; delivery of enough food to last more than 1 day (a full week in some cases, and two or three in others); use of volunteers to transport homebound seniors to congregate meal programs rather than to deliver food; computerized food preference profiles on each client; personalized food preparation; alternate modes of food delivery (e.g., by bicycle); involving volunteers in governance and administration; developing new approaches to volunteer recruitment and retention; better understanding of volunteer motivations and needs; and greater corporate involvement.
Coverage and Outreach. As the preceding review of client characteristics makes clear, most MOWs do target resources to the priority clients specified in the federal enabling legislation. Clients served are mostly poor, food insecure, aged, and mobility restricted. At the same time, available guesses (“estimates” would say too much) about the extent to which MOW cover the target population (% of those technically eligible who in fact receive services) fall somewhere in the 5% to 20% range (depending on how strictly one defines “technically eligible”). So at least four fifths and maybe nineteen twentieths of eligible seniors do not receive MOW services. What accounts for who makes it into an MOW program and who does not?
Many MOW assign potential clients a priority score. In jurisdictions where demand greatly exceeds supply (which seems to be the case about 70% of the time), the local MOW may never clear all the high-priority cases, much less those of lower priority. Low-priority clients can, therefore, languish on the waiting list for years. A second factor is whether potential clients live within the delivery area of the MOW. Potential clients who do not live in the designated delivery areas are usually just out of luck.
Do MOW do any sort of systematic outreach to identify clients in need? According to the most recent study we have found of this question (Lee et al., 2008), the answer is yes. A national phone survey of 29 Area Agency on Aging directors and 64 local MOW providers revealed only about 10% reported “little or no outreach.” However, one in four listed “word of mouth” as a local outreach strategy, which seems a bit of a stretch. (Is “word of mouth” outreach?) Then there were the predictable agency newsletters, speaking engagements, health fairs and awareness events, media spots, hospital, church, and senior center postings. Not one mention was made of direct, proactive efforts to identify dense clusters of seniors and do door-to-door canvassing to identify those in need. The authors comment that current outreach strategies, need assessments, and service delivery systems did not directly target populations that are considered “in need.” Instead, the reported outreach strategies were directed toward the general public (i.e., maximum outreach). These strategies have been in the aging service network for several decades, and have been found to be particularly helpful in increasing program awareness in the community. However, these strategies do not proactively try to identify and assist older adults who are most “in need” (Campbell et al., 2015; Lee et al., p. 411). This is consistent with findings from national surveys of OAA program participants that the authors cite, most of whom had either heard about the program through their friends (26.8%) or family (17.9%), but not much through media (4.8%), or any of the other outreach strategies just mentioned (Lee et al., p. 411).
Participants in the survey were also asked to identify barriers to reaching those most in need. Area Aging Agency directors cited geographical location, such as living in rural areas, as the most difficult barrier, mainly due to the fact that rural clients are more distant and costly to reach (p. 410). Other barriers cited were that older adults were too proud to participate in services, and limited funding kept agencies from being able to serve everyone. Findings from local service providers were similar; they mentioned that older adults' negative attitude toward the program and stigma factored into their unwillingness to participate, and even presented themselves as bigger barriers followed by funding.
Special Dietary Needs and Preferences. It would be convenient if everyone enrolled in an MOW program could eat the same lunch, but a variety of health, religious, and cultural factors make this unworkable. Many seniors suffer chronic physical disorders, the proper management of which requires special diets. Diabetes is an example of a condition that requires nutritional management, but often it is the same meal everyone else is served except that dessert is a fruit cup rather than pudding. For people with diabetes, the key nutritional issue is the glycemic index of the food being eaten, not the sugar content, which in turn depends on the total carbohydrate content of a meal. In the contemporary vernacular, “carbs is carbs” whether they come from a pile of mashed potatoes or a piece of cake.
Other chronic diseases that require special diets include hypertension (low sodium) and cardiovascular disease (low fat and low cholesterol). There is also an assortment of food allergies that should be considered (lactose-intolerance, nut allergies, gluten allergies). Probably the most common issue of all: poor dentition and the consequent need for soft foods. Add to those issues religious restrictions, a range of important self-identifications (vegan, vegetarian, ovo-lacto-vegetarian, etc.), and finally various cultural preferences in food that vary among African Americans, Hispanics, and so on and one quickly senses that the “one lunch for everyone” model just won't do. And yet in the study by Lee et al. (2008), only about half the programs studied offered special meals for those with health-related dietary needs. Even fewer accommodate special religious or cultural needs. Therefore, the data suggest that doing so would not be impossible; it would, for example, be feasible to maintain a dietary need/preference profile on every MOW client. But that might entail the preparation and delivery of 15 or 20 different lunches, a complication when trying to get a thousand lunches prepared and out the door.
Adequacy of the Food Served. Most MOWs serve only lunches and the standards require that these lunches provide at least 30% of all dietary intake necessary for good health. And yet we know empirically that, on average, these lunches comprise about 50% of participants' total daily dietary intake, at least for most nutrients and most clients. As we are probably not going to change the latter truth anytime soon, the implication is that the MOW lunch should be amped up to contain about half the minimum daily nutrient intake. This, of course, would have significant cost implications.
Considerations of Expanding Services
With respect to expanding services, funding is always a significant limitation. Some have suggested that sliding scales be tilted upward so that participants' share comes progressively closer to the true cost of meal provision. In light of the research reviewed earlier, one must also ask whether one meal a day is adequate. As one wag editorialized, “At least more than one meal, please.” A two-meal (breakfast and lunch) option, delivered between, say, 8 and 10, would certainly increase the usefulness of MOW in reducing food insecurity among homebound seniors and get them closer to the recommended allowances for all nutrients. That said, a survey in Toronto found that clients only consumed, on average, 81% of the energy provided in the MOW lunch and discarded the rest (Fogler-Levitt et al., 1995). Another study showed that older adults who were assigned to a treatment group that received 21 full meals and 14 snacks per week felt overwhelmed by all the food, were unable to eat it, and then dropped out of the study because they felt guilty for wasting food. So although we would very much like to see more food being delivered to frail low-income seniors via MOW, “efforts should be made to provide elderly participants with meals that are stimulating to their diminished appetites and not so large that they induce feelings of guilt when food goes unconsumed” (Moran, 2004, p. 1,220).
In the end, MOW is to senior food insecurity what Habitat for Humanity is to homelessness. Within the confines of programmatic intent, these are wonderful programs fully deserving of public and private support. But in both cases, the numbers are such that making a measurable dent in the larger problem is just not in the cards. In theory, the world can be saved one person, one family, or one neighborhood at a time. In practice, the urgency of our situation does not afford us that luxury. Demographically, the United States is witnessing an influx of people aging, and because of medical advancements, people are living decades longer than in previous years. Although the data suggest these programs yield positive outcomes for clients in terms of nutrition, functionality, and socialization, we still do not have enough supply to meet the demand. The OAA is set to be rewritten in the next few years. Although there have been threats at the federal government level to cut funding to MOW programs, findings from research covered in this review suggest the rewrite should include an increase in funding so these agencies can provide services to everyone in need, and even expand what they are able to offer to older adults in terms of choice, as well as the amount and frequency of meals received.