OUP user menu

Oral strategies to supplement older adults’ dietary intakes: comparing the evidence

Heidi J Silver
DOI: http://dx.doi.org/10.1111/j.1753-4887.2008.00131.x 21-31 First published online: 1 January 2009


Despite the current global obesity crisis, undernutrition remains prevalent among older adults worldwide. This review compares the efficacy of the main oral strategies used to increase older adults’ energy and nutrient intakes, i.e., meal enhancement, multivitamin/multimineral supplementation and oral liquid nutrition supplements. Well-designed long-term investigations that are adequately powered to differentiate effects on nutritional, clinical, functional, and cost outcomes are much needed before scientific and clinical consensus can be reached on where and when to implement any strategy as the optimal choice for improving dietary intakes in a specific older adult population.

  • aging
  • dietary intake
  • meal enhancement
  • older adults
  • oral nutrition supplements


The percentage of older adults among the world's population continues to increase. Despite the growing recognition of overweight (25 ≥ BMI ≤ 29.9) and obesity (BMI ≥ 30) among older adults, having a global viewpoint on the state of malnutrition in older adults means also recognizing that undernutrition (BMI ≤ 18.5) continues to be prevalent in approximately 25–40% of older adults worldwide.13 In fact, age ≥ 75 years is an independent risk factor for poor nutritional status.4

As undernutrition remains inconsistently diagnosed and treated, it continues to be associated with poor health outcomes – from inadequate diet quality and having micronutrient deficiencies,5 to loss of lean body and skeletal muscle mass, i.e., sarcopenia,6 as well as inflammatory stress, compromised immune function, impaired wound healing, increased susceptibility to infections, impaired physical performance, functional dependencies, depression, and being homebound. Moreover, undernutrition reduces overall quality of life and is associated with increased mortality risk.7,8 Finally, undernutrition is a healthcare burden that continues to strain the economic resources of developing and developed countries.9

While undernutrition is multifactorial in that it may be a consequence of physiological, psychological, social, and/or environmental factors, changes in age-related physiology predispose older adults to nutrition risk – especially in the presence of other comorbidities. Unfortunately, the most vulnerable older adults, i.e., older women, minorities, rural-living, and the poor are at greatest risk.510 Although age as a risk factor for undernutrition is not modifiable, targeting older adults for dietary supplementation may help them overcome the detrimental effects of undernutrition. In addition to identifying at-risk older adults using valid nutrition screening tools, there is also a need for evidence-based nutrition interventions. However, there remains no consensus on the best method to improve dietary intakes to meet energy, protein, and other nutrient requirements, and thereby improve body weight and body composition, physical and cognitive function, and other health outcomes.

Strategies to complement the habitual dietary intakes of older adults include food-based interventions, such as the following: providing direct feeding assistance; enhancing the dining environment atmosphere; adjusting food form, portion size, and presentation; modifying the texture or consistency of food and beverages; enhancing the flavor of foods served; providing congregate or home-delivered meals; manipulating the energy or nutrient density of recipes; adding snacks or beverages between meals; and interventions using multivitamin/ multimineral, or oral liquid nutrition supplements in addition to meals, as well as providing nutrition education and counseling. Thus, the purpose of this review is to compare and contrast the available evidence on the three main dietary strategies for nutritional supplementation of older adults who are unable to ingest adequate amounts of foods and beverages to meet their daily nutrient requirements: meal enhancement, multivitamin/multimineral (MVM) supplementation and oral liquid nutrition supplementation (ONS).


The feeding environment

Methods designed to improve the dietary intakes of residents in long-term-care facilities (LTC) include enhancing the type of food services provided and modifying the dining environment to be less institution-like and more home-like. For example, providing cafeteria style meals over the course of 3 weeks in a LTC facility in Ontario resulted in significantly higher energy intakes when compared to traditional meal delivery service on trays, P < 0.001.11 Another method that has been tested in LTC is providing direct feeding assistance of food and beverage intake. During a 2-day trial with 134 LTC residents using verbal cuing and individual physical guidance at mealtimes, dietary intakes were 29–56% greater in percentage of food consumed.12

Flavor enhancement

As aging reduces sensitivity to the taste and smell of foods, older adults may have an impaired ability to appreciate the palatability of foods.1315 In a randomized trial in which the flavor of protein-rich food items was enhanced with beef, chicken, turkey, or lemon butter flavor, increased caloric intakes resulted in a weight gain of 1.1 kg in LTC residents.16 And, in a residential home setting, flavor enhancement of energy-dense food items for 3 weeks improved grip strength and immune function for 39 older adults.17

Congregate and home-delivered meals

Congregate and home-delivered meals (HDM) are provided to approximately 3 million older Americans under Title IIIC of the Older Americans Act (OAA), which is administered by the U.S. Department of Health and Human Services through the Administration on Aging. In 1993–1995, a national evaluation showed that participation in OAA meal programs resulted in lunch meal dietary intakes that exceeded one-third of the 1989 RDAs.18 Compared to age- and gender-matched non-participants, congregate meal participants had significantly increased intakes for 16 nutrients and HDM participants had significantly increased intakes for 12 nutrients. Similar efficacy of a HDM program was observed in southwestern Ontario, where a telephone-administered follow-up of 263 meal participants showed that 18 months after regular participation there was a significant decrease in nutrition risk for participants who had been categorized as high risk (SCREEN score ≤ 45) when HDM were initiated.19

As home-delivered lunch meals comprise 30–50% of participants’ total daily food and nutrient intakes,18 it is recognized that more comprehensive interventions than 5 meals per week are needed for frail older adults.20 Gollub et al.21 investigated the effect of adding a home-delivered breakfast meal 5 days per week to HDM participants in Texas, Florida, Montana, Virginia, and Maine. Total dietary intakes of the HDM-plus-breakfast group (n = 167) averaged 300 kcal and 14 g protein more per day as compared to the HDM-alone group (n = 214) (P < 0.001), and they consumed significantly greater amounts of folate, calcium, magnesium, potassium, iron, and zinc. Kretser et al.22 assessed outcomes when HDM were provided 7 days per week with an additional two snacks daily to meet 100% of the participants’ Dietary Reference Intakes. After 6 months, recipients of the enhanced HDM program gained significantly more weight (4.3 versus 1.7 lbs, P = 0.004) than recipients of the traditional five-meal program. Despite the positive outcomes that have been demonstrated for various meal programs, only one-third of U.S. adults over the age of 60 years who need these meals services receive them due to inadequate federal, state, and local funding. Thus, about half of the OAA HDM programs have waiting lists.23

Manipulating the energy and nutrient density of foods

Only a few trials have been conducted to assess the effects of increasing the energy and/or nutrient density (kcal or nutrient per gram weight of food) of foods or beverages consumed by older adults. Modifications to increase energy or nutrient density may be achieved by adding, substituting, or enhancing energy- and/or protein-rich ingredients of menu items by using fats such as cream, margarine, or cheese and/or carbohydrates such as glucose polymers. In a study with a case-control design, Odlund-Olin et al.24 observed that residents who received energy-dense meals consumed about 30% more kcal per kg and were able to maintain their ADL functions after 15 weeks. In older adults in a rehabilitation ward, increasing the energy density of food items served in the regular hospital menu for a 2-month period improved average caloric intakes by 25%.25 And in a randomized counterbalanced crossover study, Silver et al.26 compared regularly served HDM menu items to enhanced versions of the same HDM menu items with 45 HDM participant members of a home health agency in Palm Beach, Florida. On the days with the enhanced version, there were significantly improved lunch meal and 24-hour intakes for calories and protein, and several micronutrients including vitamin D, α-tocopherol, riboflavin, niacin, vitamin B6, calcium, magnesium, copper, and selenium.

Between-meal snacks

Since older adults also tend to consume smaller portions as they age, adding between-meal snacks such as milk- or soy-based shakes, puddings, yogurts, juices, snack bars, or cookies is another means of increasing nutrient intakes. Turic et al.27 found that between-meal snacks increased total energy intakes by 30% in LTC residents. Moreover, de Jong et al.28 observed that fruit- and dairy-based snacks given twice daily for 17 weeks to community-residing older adults improved serum 25-hydroxy vitamin D, pyridoxal-5-phosphate, cobalamin, and ascorbic acid levels.

Johnson et al.29 evaluated dietary outcomes of the U.S. Department of Agriculture Seattle Seniors Farmers’ Market Nutrition Program in which a basket of fresh locally grown fruits and vegetables were delivered every other week from June through October 2001 to homebound HDM participants. After 5 months of deliveries, the intervention group (n = 100) consumed 1.3 more servings of fruits and vegetables daily compared to the control group (n = 52), P < 0.001.

In a hospital-based study, Kruizenga et al.3 used the SNAQ screening tool to identify patients who were moderately or severely undernourished and trigger the provision of between-meal snacks that provided an additional 600 kcal and 12 g protein daily. Not only did screening trigger significantly more referrals to the registered dietitian (76% versus 46%) and shorten the interval to dietitian consultation (5.8 days versus 2.6 days, P < 0.001), but between-meal snacks reduced the length of stay in patients with low handgrip strength (11.5 versus 14 days).

Although the evidence reviewed suggests that food-based interventions increase energy and nutrient intakes and improve nutritional status as well as clinical and functional outcomes, very few well-designed RCTs have been conducted to evaluate healthcare outcomes or the cost effectiveness of food-based interventions. Inadequate funding for investigation of these types of dietary interventions and lack of continuity of care at discharge that would promote enrollment of at-risk individuals into food-based programs (such as congregate or home-delivered meals) contributes to the widespread use of other types of dietary supplementation in all care settings.


Daily use of MVM supplements is common among approximately half of older adults in the United States,30,31 and national nutrition monitoring data has shown that MVM improves nutrient intakes for vitamin A, B6, B12, C, E, folate, iron, and zinc. Then again, these data also showed that female older adults exceeded the tolerable upper level intakes for vitamin A and males exceeded the recommended intakes for iron and zinc. Nevertheless, in a RCT with 158 adults who were stratified by age and presence of type II diabetes, Barringer et al.32 reported that daily MVM for 1 year significantly reduced the incidence of self-reported infection based on daily symptom checklists among diabetic participants (93% versus 17%). Conversely, other RCTs of daily MVM supplementation have not shown efficacy for older adults in health-related outcomes. For example, McNeil et al.33 observed no significant effects on cognitive function (defined as immediate memory and executive function) after 12 months of daily MVM in 910 community-residing Scottish older adults. And Liu et al.34 found no significant reductions in infection rates or hospitalization after 18 months of daily MVM in 763 residents of 21 LTC facilities. These negative findings are consistent with a recent meta-analysis of 20 RCTs by Stephen and Avenell,35 which showed no significant effects of MVM on episodes of infection in older adults (relative risk, 0.98; 95% confidence interval, 0.86–1.11). Importantly, a cost-utility analysis performed to determine the benefit of MVM in reducing the use of health service resources in Scotland indicated that once-daily MVM supplementation was not cost effective.36

Since the diets of older adults are often deficient in micronutrients,37 most daily MVM supplements can fill the gap in intakes for vitamins A, B6, C, and E.38 Hence, the lack of efficacy of a single daily MVM on clinical outcomes in older adults may be dose-related. Moreover, MVM supplements can be low in cost, estimated at approximately $0.10 per pill.39 However, MVM do not provide older adults with a source of calories, protein, essential fatty acids, or fiber.40


Since the 1970s, when the first ONS became commercially available,41 the number and variety, the food and nutrient ingredients used, and the health claims for ONS products have increased steadily. ONS include modular components to enhance dietary macronutrient composition, about a dozen specially formulated and processed milk-based formulas or products to be mixed with milk, about two dozen whole-protein and lactose-free formulas, about a dozen whole-protein lactose-free high-calorie formulas, specially designed macronutrient formulations, products with altered carbohydrate ingredients for glucose intolerance, products with altered amino acid concentrations for hepatic disease, pulmonary disease, or renal disease, products for compromised immune function (including those with pre- or probiotic food ingredients), and hypermetabolic states (i.e., burns, trauma, and infection), defined formulas containing increased amounts of conditionally essential nutrients (i.e., amino acids, peptides, or medium-chain triglycerides), as well as specialized pediatric formulas. Typically, ONS are used when an individual is unable to ingest, digest, absorb, or metabolize foods or nutrients; thus, they are used when the habitual diet is not meeting nutrient requirements or for management of specific nutrient needs resulting from a particular injury, illness, or disease condition.


It is important to recognize that there have been no systematic reviews or meta-analyses comparing isocaloric or isonutrient meal-enhancement methods to ONS. One RCT compared the provision of three between-meal snacks daily to 300 kcal ONS provided three times daily for 6 weeks to residents of four LTC facilities who had a 6–7-pound involuntary weight loss over the prior 6 months.27 In this study, no compensation in habitual caloric intakes was observed with snacks or ONS. Average energy intakes increased by 26–30% from snacks and 46–50% from ONS. It is unclear whether the calorie and nutrient contents of the snacks were comparable to the ONS formulation.

In a meta-analysis of 24 RCTs using the Cochrane method, Baldwin et al.42 compared the dietary advice (i.e., nutrition counseling) provided by a registered dietitian or other healthcare professional in adults with chronic disease or those at nutrition risk to one of the following: no advice, ONS, or advice plus ONS. In the studies comparing dietary advice versus no advice, there was no significant difference in mortality rate. In the studies comparing dietary advice to ONS, ONS resulted in greater energy intakes and weight gain after 3 months. However, no differences were detected in hospital stay or mortality. In the studies comparing dietary advice to advice plus ONS, a small difference was found in handgrip strength with advice plus ONS, but there were no significant differences in energy intakes, weight gain, or mortality.

Although the authors concluded that the evidence base does not support dietary advice, the data do not support this conclusion as they are confounded by the fact that dietary advice was not provided consistently by the most qualified healthcare professional, i.e., the registered dietitian,1 and the comparisons were not equivalent in that a cognitive intervention (dietary advice) was compared to a physical intervention (ONS).


In a series of RCTs in hospitalized older adults in England, Gariballa and Forster4345 found that ONS for 6 months resulted in decreased hospital readmissions, reduced mortality, and improved depression and quality-of-life scores. Several comprehensive meta-analyses have synthesized the evidence on outcomes of ONS in hospitalized patients. Using the literature published from 1970 through 1996, Potter46 found 20 RCTs investigating ONS in 1495 patients. ONS appeared to increase body weight with a pooled weighted mean difference (WMD) of 2.06% (95% confidence interval: 1.63–2.49%) and reduce mortality by 37% (95% confidence interval: 0.45–0.82). These findings are consistent with results from a prospective RCT in 381 patients aged 61–79 years. Patients who received a 180 kcal ONS during daily medication rounds had significantly greater total energy intake (1409 ± 448 vs. 1090 ± 417 kcals), a 1% weight gain compared to a 1% weight loss, and significantly lower mortality (14.7 vs. 35%).47

Likewise, another systematic review of 55 randomized or quasi-randomized trials (n = 9187) showed significantly greater energy and protein intakes with ONS.48 Nineteen of the 55 trials showed a 28% decreased in-hospital incidence of infections, incomplete wound healing, pressure sores, or exacerbation of disease (95% confidence interval, 0.53–0.97). Moreover, 25 trials showed a 14% reduced mortality rate (95% confidence interval, 0.74–1.00), which was greatest in adults aged ≥75 years.

One of the largest meta-analyses using the Cochrane methodology included 49 randomized or quasi-randomized trials with a minimum ONS intervention of 10 days.49 Sixty percent of the 4790 older adults were in acute or chronic care settings and 40% resided in the community. Twenty-nine studies reported increased total daily energy and protein intakes with ONS. However, the variety of dietary assessment methods used did not allow a meta-analysis of dietary data. Thirty-four trials showed a benefit from ONS on percent weight change with a pooled WMD of 2.3% (1.9%–2.7%) and 14 trials showed a benefit for mid-upper-arm circumference with a pooled WMD of 1.2% (0.4%–2.0%). However, 10 studies showed no benefit from ONS on length of hospital stay with a pooled WMD of –1.98 days (5.20–1.24). Yet, 32 studies showed that ONS significantly reduced mortality by 60% (95% confidence interval, 0.59–0.92). Reduced risk appeared to be greater in undernourished subjects (n = 1825) with a relative risk of 0.72 (95% confidence interval, 0.55–0.94), compared to those who were not undernourished (relative risk, 0.78; 95% confidence interval, 0.53–1.15).


In adults with type I or II diabetes, a systematic review of ONS resulted in 16 studies for meta-analysis.50 In short-term RCTs, ONS reduced postprandial glucose, postprandial insulin, and glucose area under the curve compared to standard care. Moreover, diabetes-specific formulas (high fat, fructose, and fiber) resulted in lower postprandial glucose compared to isocaloric food-based afternoon snacks. In addition to improving glycemic control by lowering postprandial glucose, diabetes-specific formulas significantly lowered peak blood glucose concentration and glucose area under the curve.

A systematic review of ONS in subjects with chronic renal disease resulted in 18 studies (n = 541), but there was insufficient data to perform a meta-analysis on clinical or functional outcomes.51 From the 14 studies that compared disease-specific renal ONS to standard care, it appeared that disease-specific formulas increased energy and protein intakes, and serum albumin concentration.

Several RCTs in radiation oncology patients, especially those with head and neck cancer, show benefits from ONS in weight, nutritional status, and quality-of-life outcomes. In a RCT with 60 radiation oncology patients in Australia, significantly greater mean energy and protein intakes and a trend toward higher fiber intakes were observed after 3 months of ONS.52 In a meta-analysis of three RCTs, Elia et al.53 observed that ONS increased energy intakes by 381 kcal per day, as compared to standard hospital meal service. On the other hand, a RCT with 75 head and neck patients showed that nutrition counseling was more beneficial than ONS in improving oral toxicity symptoms and quality-of-life ratings.54

Weight loss and wasting of lean body mass are also common features of chronic obstructive pulmonary disease (COPD). In an 8-week inpatient rehabilitation program in the Netherlands, the provision of two to three ONS daily improved body weight (2.1 ± 2.1 kg) and fat-free mass significantly, while also improving respiratory measures, handgrip strength, and peak workload.55 Likewise, a 12-week ONS intervention in Spain showed improved weight in stable COPD patients.56 Notably, the authors determined that patients who were randomized to ONS using a prescription of REE × 1.3 showed greater improvements in respiratory and functional outcomes than those receiving ONS at REE × 1.7.

In contrast, a 15-country RCT of acute stroke patients who were not dysphagic showed no difference in the frequency of in-hospital complications (defined as pneumonia and urinary tract infections), hospital stay, or mortality between the usual hospital diet (n = 2007) and the usual diet plus ONS (n = 2016).57 However, only 8% of patients in both groups were categorized as undernourished (by observation) and dietary data were not collected to evaluate actual consumption.

Avenell and Handoll58 identified eight RCTs for meta-analysis of ONS after hip fracture (n = 448 patients). No significant effects on hospital stay or mortality were detected. However, when the data were pooled to combine the outcomes of complications and mortality into one variable, ONS appeared to reduce the number of patients with the combined outcome by 48% (95% confidence interval, 0.32–0.84). Similarly, a systematic review by Milne et al.59 of four RCTs showed a 52% reduced risk for complications of hip fracture (95% confidence interval, 0.24–0.96). Importantly, Boudville and Bruce60 observed that consuming ONS 30 or 90 min before an ad libitum lunch meal did not decrease meal energy intakes in women who were recovering from hip fracture.

A pilot study with nine adults aged 55–84 years with surgically treated chronic skin ulcers who received daily ONS providing 35% of total energy intakes, as well as daily MVM and 500 mg vitamin C, suggested improvement in wound healing as patients experienced a 90% skin graft acceptance rate at 10 days postoperatively.61 A larger study in 672 older adults showed that ONS reduced the risk for developing pressure ulcers by 43% (95% confidence interval, 1.03–2.38) after 15 days of twice-daily ONS totaling 400 kcal/d.62 A systematic review of ONS in pressure ulcers by Stratton et al.63 showed that high-protein ONS reduced the incidence of pressure ulcers by 25% (95% confidence interval, 0.62–0.89) in four studies (n = 1224). Interestingly, the authors estimated that 19.25 patients would have to receive ONS to prevent one pressure ulcer.


ONS is the most frequently prescribed strategy to prevent or attenuate unintentional weight loss in LTC residents, and consumption is high when residents are offered choices that meet taste preferences and receive adequate assistance to promote consumption.64,65 However, LTC residents receive less than one-third of prescribed ONS and LTC staff spend as little as 1–6 min encouraging oral consumption, despite evidence that encouragement is required for 30 min per meal to improve intakes.66 In LTC, inadequate staffing ratios and staff training, lack of specificity in the written nutrition order for the ONS regimen (i.e., time of delivery), and lack of documentation of delivery and consumption interfere with the ability to assess the efficacy of ONS.

In contrast, LTC residents consume more calories when provided with between-meal snacks like juices, puddings, and yogurts. Additionally, family members rate providing multiple between-meal snacks as a more preferable nutrition intervention than ONS.67 Nevertheless, for institutionalized adults with Alzheimer's disease who are able to eat independently, providing ONS for only 3 weeks improved energy intakes and mid-arm muscle circumferences.68,69 Young et al.68 also showed that regularly served meals with enhanced carbohydrate content improved meal and 24-hour energy intakes in patients with late-stage Alzheimer's.


An important concern regarding ONS consumption is whether older adults compensate for the energy and nutrient content of ONS by reducing subsequent meal intakes. A few small controlled experiments have investigated the effects of ONS on short-term energy intakes. In a laboratory-based experiment with 15 young (20–40 years) and 15 older (>70 years) adults who consumed ONS preloads either 5 or 60 min before a test meal, no difference was detected in ratings of palatability, hunger, or fullness between young and older adults.70 Notably, energy intake was higher in young and older subjects when the ONS was consumed 60 min before the test meal. Thus, the authors concluded that ONS should be administered between meals rather than with meals. In a 3-day trial involving 16 older adults with low BMI, 24-hour energy intakes were greatest with high or low carbohydrate ONS compared to no ONS.71 In a 4-month RCT with 83 older adults receiving home healthcare services, it was determined that ONS increased energy intakes on average by 332 kcal per day and weight gain was higher in the ONS group (1.62 vs. 0.04 kg, P < 0.001).72 In a 6-month RCT in which 68 adults who were ≥ 65 years of age and had a BMI ≤ 25 kg/m2 consumed ONS twice-daily between meals, ONS increased total energy intakes significantly and resulted in a 2.5% greater weight gain (P = 0.031); however, no differences in body composition or physical function were detected. Importantly, no compensation was observed in the amount of caloric intakes from regular meals with between-meal ONS.73Tables 1 and 2 summarize the significant outcomes from the evidence on ONS.

View this table:
Table 1

Significant outcomes of clinical trials investigating oral liquid nutrition supplements in older adults.

StudyPopulationSample sizeDesignInterventionDurationOutcome
Turic et al. (1998)76LTC91Case control300 kcal ONS 3/day vs. 3 snacks/day6 weeksEnergy intake; P < 0.01
Protein intake; P < 0.01
Vitamin and mineral intakes; P < 0.01
Gariballa et al. (2006)43Acutely ill445RCT400 mL ONS6 weeksReduced readmission rate (adjusted hazard ratio: 0.68)
Reduced 6-month mortality (9% vs. 14%)
Gariballa et al. (2007)44,45Acutely ill225RCT400 mL ONS6 weeksReduced depression; P = 0.021
Greater serum folate and B12
Higher quality-of-life scores; P = 0.04
Potter et al. (2001)47New admissions381RCT120 mL ONS 3/dayEnergy intake: P = 0.001
Weight gain: P = 0.003
Reduced mortality (OR: 0.62)
Isenring et al. (2007)52Cancer60RCTONS3 monthsEnergy intake; P = 0.02
Protein intake: P < 0.001
Ravasco et al. (2005)54Cancer75RCT400 mL ONS3 monthsEnergy intake; P = 0.005
Creutzberg et al. (2003)55COPD64Prospective cohortONS 2–3/day8 weeksWeight gain; P < 0.001
Fat-free-mass gain; P < 0.001
Handgrip strength; P = 0.004
Planas et al. (2005)56COPDRandomizedONS @ REE × 1.312 weeks
ONS @ REE × 1.7
FOOD Trial Collab. (2005)57Stroke2149RCT360 ml ONS6 monthsNo significant outcomes
Boudville et al. (2005)60Females with hip fracture7Randomized crossoverWater2 hoursNo compensation in energy intakes
ONS 30 mins pre-meal
ONS 90 mins pre-meal
Raffoul et al. (2006)61Pressure ulcer9Prospective cohort300 kcal ONS16 daysSkin graft rate >90%
Bourdel-Marchasson et al. (2000)62Critically ill672RCTONS 2/day15 daysEnergy intake; P = 0.006
Protein intake; P < 0.001
Incidence of pressure ulcers 40.6% vs. 47% (RR: 1.28 vs. 1.57; P = 0.04)
Lauque et al. (2000)64LTC88RCT400 kcal ONS60 daysEnergy intake; P < 0.001
Weight gain; P < 0.001
Young et al. (2004)65Alzheimer's34Case controlONS mid-morning21 days68% had greater energy and protein intakes
Parrott et al. (2006)69Alzheimer's30Case controlONS mid-morning3 weeks59% maintained increased energy intakes
Wilson et al. (2002)70Age >70 years15Randomized crossoverWater vs. high-fat, high-carb, high-protein ONS5 vs. 60 min pre- mealEnergy intake higher with ONS 60 min pre-meal; P < 0.05
Age 20–40 years15
Ryan et al. (2004)71BMI ≤ 2416RCT250 mL ONS at breakfast3 daysEnergy intake; P = 0.0035
Payette et al. (2002)72Frail with home health83RCTONS16 weeksEnergy intake; P < 0.001
Weight gain; P < 0.001
Wouters-Wesseling et al. (2003)73BMI ≤ 2568RCT250 mL ONS6 monthsWeight gain; P = 0.03
Better sleep score; P = 0.03
  • Abbreviations: AUC, area under the curve; BG, blood glucose; BMI, body mass index (kg/m2); CI, 95% confidence interval; COPD, chronic obstructive pulmonary disease; LTC, long-term care; MAMC, mid-arm muscle circumference; ONS, oral liquid nutrition supplement; OR, odds ratio; RCT, randomized controlled trial; RR, relative risk; WMD, weighted mean difference.

View this table:
Table 2

Significant outcomes of meta-analyses of studies investigating oral liquid nutrition supplements in older adults.

StudyPopulationSample sizeNo. of studiesInterventionDurationOutcomes
Baldwin et al. (2001)42Chronic illness213525Dietary advice vs. ONS16 days–24 monthsWeight gain WMD 1.09 kg (95%CI: 0.29–1.90 kg)
Potter (2001)46Hospitalized206230ONSWeight gain (%) WMD 2.06% (95%CI: 1.63–2.49%)
MAMC WMD 3.16% (95%CI: 2.43–3.9%)
Milne et al. (2006)48Age ≥65 years918755175–1000 kcal ONS/day10 days–18 monthsMortality benefit: ≥75 years (OR: 0.64); ≥400 kcal/d (OR: 0.85)
Reduced complications with hip fracture (OR: 0.48)
Weight gain (%) WMD 1.75% (95%CI: 1.12–2.3%)
MAMC WMD 1.41% (95%CI: 0.46–2.35%)
Milne et al. (2005)49Age ≥65 years479049175–1000 kcal ONS/day10 days – 18 monthsMortality benefit: undernutrition (RR: 0.72); ≥400 kcal/d (RR: 0.71); ≥75 years (RR: 0.69); ONS ≥ 35 days (RR: 0.75)
Reduced complications with hip fracture (RR: 0.60)
Weight gain (%) WMD 2.3% (95%CI: 1.9–2.7%)
MAMC WMD 1.2% (95%CI: 0.4–2.0%)
Elia et al. (2005)50Diabetes78423Disease-specific ONS1 day–3 monthsLower postprandial BG 1.03 mmol/l (95%CI: 0.58–1.47)
Lower peak BG 1.59 mmol/l (95%CI: 0.6–2.32)
Lower glucose AUC 7.96 mmol/l/min (95%CI: 2.25–13.66)
Stratton et al. (2005)51Chronic kidney disease54118ONS14 days–11 monthsSerum albumin increase 0.23 g/dL
Energy and protein intakes increase by 20–50%
Elia et al. (2006)53Cancer62ONS6–70 daysEnergy intake increased 381 kcal/day
Avenell and Handoll (2006)58Hip fracture172721ONSReduced unfavorable outcome (death or complications) RR: 0.52 (95%CI: 0.32–0.84)
Stratton et al. (2005)63Postoperative pressure ulcer risk122415250–500 kcal ONS2–26 weeksReduced incidence (OR: 0.75)
  • Abbreviations: AUC, area under curve; BG, blood glucose; BMI, body mass index (kg/m2); CI, confidence interval; COPD, chronic obstructive pulmonary disease; LTC, long-term care; MAMC, mid-arm muscle circumference; ONS, oral liquid nutrition supplement; OR, odds ratio; RCT, randomized controlled trial; RR, relative risk; WMD, weighted mean difference.


For all the methods reviewed for improving the dietary intakes of older adults in habitual and institutional settings, it can be asserted that further large, long-term, placebo-controlled randomized trials are much needed that have adequate power to detect differences in clinical and functional outcomes. Such trials are especially indicated for older adults who have the greatest nutrition risk, that is, those who are at least moderately undernourished (BMI ≤ 20) and those aged ≥75 years old. For nutrition counseling and meal enhancement methods, the evidence base is especially limited with regard to well-designed RCTs that investigate clinical and functional outcomes as well as cost effectiveness. The lack of reimbursement for dietitian services and consequent reduced availability of registered dietitians, along with lack of coordination in the delivery of nutrition care from acute to post-acute care settings contributes to the gap in evidence.74 Further, dietitians continue to rely on referrals for nutrition service provision, which contributes to inconsistent, delayed, and unmonitored nutrition intervention.75,76

While the intakes of some nutrients, such as vitamin D, will always require supplementation for most older adults, the evidence base for daily, single MVM supplementation is even weaker than for food-based and meal-enhancement methods. Barriers to investigating the efficacy of MVM in older adults not only include the lack of programs available for large-scale distribution of MVM, but also the limited information regarding dosage and the need for valid biomarkers to evaluate micronutrient intakes. As with food-based interventions, it is also necessary to show effectiveness on clinical, functional, and cost outcomes beyond the evaluation of micronutrient intakes and blood concentrations.

The evidence base for ONS is far more comprehensive than for other dietary strategies. ONS appear to improve energy intakes as well as have a positive impact on some clinical and functional outcomes and reduced mortality risk in certain disease, illness, or injury conditions such as cancer, COPD, hip fracture, pressure ulcers, and independently functioning Alzheimer's patients. Disease-specific formulations appear to be beneficial in diabetes and renal disease. Although the evidence does not appear to support efficacy of ONS in stroke nor for some other LTC residents, consensus has not been reached to delineate for which specific disease states the widespread use of ONS is and is not reliably indicated. Moreover, little evidence exists on the cost effectiveness of ONS, especially in well-defined, well-matched, adequately powered RCTs in which ONS is equitably compared to meal-enhancement strategies, i.e., those that are designed to be isocaloric and isonutrient as well as having the same delivery regimen.

For all oral strategies, many other barriers challenge healthcare professionals worldwide in making decisions for the best method to prevent or remedy undernutrition in older adult patients. Unfortunately, there remain inadequate resources for clinical nutrition research, and comprehensive nutrition interventions are hampered by inadequate ratios of professional and non-professional staff to implement and evaluate outcomes. Moreover, synthesis of the available evidence is limited by the range of methods used to define and measure outcomes. Ultimately, an efficacious dietary strategy is one that is safe, clinically relevant, and cost effective. Moreover, it is one that not only improves the nutrient intakes, nutritional status, and body weight of older adults, it also enhances body composition and functionality; prevents, delays, or shortens institutionalization; and enriches overall quality of life.


Portions of this review will appear in the Nestlé Nutrition Workshop Series: Clinical and Performance Program, Volume 12, 2009, S. Karger AG, Basel, Switzerland.


View Abstract