By David Kiefer, MD, Editor

Clinical Assistant Professor, Department of Family Medicine, University of Wisconsin; Clinical Assistant Professor of Medicine, Arizona Center for Integrative Medicine, University of Arizona, Tucson

Dr. Kiefer reports no financial relationships relevant to this field of study.

SYNOPSIS: A nutritional supplement helped malnourished nursing home residents gain weight and improve quality of life, more so than dietary advice and counseling.

SOURCE: Parsons EL, Stratton RJ, Cawood AL, et al. Oral nutritional supplements in a randomised trial are more effective than dietary advice at improving quality of life in malnourished care home residents. Clin Nutr 2016; Jan 11. pii: S0261-5614(16)00003-0. [Epub ahead of print].


  • The authors randomized 104 people (average age 88 years) in nursing homes to receive either nutritional supplementation (target 600 calories and 16 g of protein) or dietary advice for 12 weeks.
  • The nutritional supplementation group had 6-8% greater quality of life by standardized measures than the dietary advice group, and gained 1.2 kg (vs. 0.5 kg) during the study period.

This study compared oral nutritional supplements (ONS) with dietary advice in a malnourished institutionalized population with respect to quality of life (QL; the primary outcome) and nutritional intake (the secondary outcome). The U.K. study was undertaken to address the problem of malnutrition in nursing homes. A total of 104 residents of said facilities were recruited and randomized to ONS or dietary advice provided by a dietitian for 12 weeks.

The researchers started by searching for all nursing/care homes in Hampshire, England (n = 633), and excluded facilities owned locally or by the National Health Service, those with less than 10 beds or for people younger than 50 years of age, and those just for advanced dementia, drug dependency, and learning disabilities; the authors did not note the reason for these exclusions. All told, 53 homes participated, and the focus was on adults 50 years of age who were at medium or high risk for malnutrition based on the MUST scale. (See Table 1.) A total of 1,455 residents were screened, 598 of whom met adequate MUST criteria (score 1), although only 132 passed a variety of exclusion criteria (outlined in detail in the article). Of the 132, only 104 gave informed consent for participation in the procedure.

The 104 residents selected for the trial were then randomized to ONS or dietary advice. For the ONS group, residents were able to select a variety of ONS depending on personal preference, but all provided 600 kilocalories (kcal) and 16 grams of protein. Examples of ONS chosen by residents include drinks, puddings, and soups, and they also varied by flavor, volume, and energy density; most residents chose a fortified drink providing 1.5-2.4 kcal/mL. Those residents in the dietary advice group received written information requesting that they focus on eating foods and drinks of “high energy.” This written information was reviewed, in person, with a dietitian at baseline and week 6.

Measurements were taken at baseline, six weeks, and 12 weeks, and included MUST criteria, body weight, height, BMI, and QL (using EuroQol that measures mobility, self-care, usual activities, pain, anxiety; and EuroQol visual analog scale [VAS], ranging 0-100, a self-assessment). (See Table 1.) Both the EuroQol and the VAS were adjusted to “incorporate judgments made by representative samples of the general population,” essentially a referenced calculation allowing ranges of -0.059 to 1.0 and -0.073 to 1.0, respectively. In addition, a 24-hour dietary recall was used to estimate energy and macronutrient intake, and residents were asked to verbally respond to a series of questions pertaining to hunger and satiation.

Table 1: Malnutrition Universal Screening Tool (MUST) Criteria

  • Body mass index (BMI): > 20 kg/m2 = 0 score, 18.5-20 kg/m2 = 1, < 18.5 kg/m2 = 2
  • Unplanned weight loss in past 3-6 months: < 5% = 0 score, 5-10% = 1, > 10% = 2
  • Acute illness and no nutritional intake for more than five days: score 2
  • Low risk if score 0, medium risk if score 1, high risk if score 2 or more

An intention-to-treat analysis of the 104 study subjects was undertaken on the ONS (n = 53) and dietary advice (n = 51) groups. The average age was 88.5 years, and 86% were female. Per MUST, 46% were medium risk and 54% were high risk. Baseline characteristics were similar, except for an eight-point higher VAS score in the ONS group. Thirty-four residents did not complete the study (most commonly due to “confusion” secondary to “ill health”), 14 in the ONS group and 20 in the dietary advice group. There was no significant demographic differences between these groups, nor with those residents who completed the study.

Results are displayed in Table 2. After 12 weeks, the ONS group showed statistically significant improvements in QL, both in the EuroQol calculation and VAS (P values 0.005 and 0.006, respectively). For both of these parameters, when the measurements were averaged over the 12-week period to estimate an average QL for that period, the ONS group was again statistically higher: for EuroQol, 0.489 for ONS vs. 0.449 for dietary advice (P = 0.013), and for VAS, 0.531 for ONS vs. 0.507 for dietary advice (P = 0.015). The researchers reported some of these findings as a percentage improvement to make the numbers more clinically relevant. (See Table 2.)

Table 2: Study Results for the Oral Nutritional Supplements and Dietary Advice Groups


Oral Nutritional Supplements

Dietary Advice

Quality of life, as per EuroQol

0.496 (8.3%)*


Quality of life, as per VAS

0.535 (7.3%)*


Body weight change (kilograms)

+ 1.22


* Percentage that the ONS scores were above the dietary advice scores; this percentage represents “…a proportion of the entire range of possible scores.”

Other results included a significant increase in body weight in the ONS group over the 12 weeks (1.22 kg; P = 0.010) vs. an insignificant increase in the dietary advice group (0.48 kg; P > 0.05). An extremely detailed breakdown of micro- and macro-nutrient intake revealed that the ONS group consumed 351 kcal of calories and 12.2 grams of protein more than the dietary advice group per day. Obviously, most of this amount was due to the ONS ingestion, although higher voluntary food intake in the ONS group also accounted for some of this. The ONS also tipped the scales in favor of higher intake of “… most other nutrients …” as detailed in article.


In the midst of the discussion of food over supplements, à la Michael Pollan, it was difficult to stomach (pun intended) the results of this study. Isn’t it always better to consume whole foods, organically grown if possible, to provide nutrition and prevent illness? The researchers of this study looked at just this question in a very select group of people: those who were malnourished, elderly, and institutionalized. They found that dietary advice and two meetings with a dietitian couldn’t compete with ONS of a variety of formulations with respect to QL, weight gain, and nutrient intake. Even though the target calorie supplementation wasn’t reached, the residents displayed significant benefit over the study period.

Obviously, this demographic isn’t relevant to all clinical practices. But the results for this obscure subgroup of people nonetheless possibly could be extrapolated to other populations who may have marginal nutrition. And ONS is not that uncommon of a nutritional approach when we think of how many of our patients (and selves, friends, families) turn to protein shakes, smoothies, and juices to take the place of skipped meals or fortify one’s nutrition during extremes of activity, stress, work, etc.

What should we tell patients based on this study? Certainly, an individualized decision, although one approach might be to follow the methodology of this study and use the MUST criteria. For example, the MUST criteria would whittle down significantly who might truly benefit from ONS; a 0 on the MUST scale may dissuade the use of ONS to achieve nutritional goals, whereas those patients at medium and high risk (scores 1+) should seriously consider ONS to boost QL and weight. There are few reasons not to jump on the ONS bandwagon other than cost (not mentioned in the trial) and logistics (extra staff involvement to prepare and distribute the ONS). There was some inkling from this study that a boost of nutrition from ONS might even nudge voluntary food intake — a nice side benefit, and one that the food > supplements advocates certainly could support.

Before we pass out smoothies to all our malnourished patients, a few comments on the study itself may prompt some reflection. There were some unknowns about the true ONS intake in the intervention group; the ONS was just a target amount, but residents remained in the study regardless of true intake. It is difficult to say whether true intake of ONS would have been more or less than the estimates, clearly affecting study results. Also, the control group, dietary advice, may not have had a fighting chance. Other dietary intervention studies often include weekly dietitian visits, so with only two visits over 12 weeks, it could be argued that the residents would be unlikely to incorporate such dietary suggestions into their lives. Furthermore, it would be interesting to note what might happen with a true control group, either one with no intervention or one with a nutritional supplement that provided little to no supplemental nutrition. Ideally, these control interventions, in addition to being randomized, would be blinded — a difficulty with dietary intervention studies, but nonetheless an approach that would enhance the believability of the results. The age of the study population (average 88.5 years) also at least gives one pause about the relevance of a dietary advice arm based on written information, supplemented only twice with a dietitian consult. Perhaps the significant percentage of dropouts (in each arm) speaks to the onerous aspect of the intervention attempted in this population, if not simply because of the complicated health histories in this demographic.

Despite these criticisms, this was a gallant effort to reach a deserving population with one of life’s basic needs, healthy nutrients, and demonstrate what we all know, that adequate nutrition certainly correlates with improved quality of life. And the demonstrated improvements (7-8%) with ONS compared to dietary advice may even be on par with other medical and non-medical interventions in this population.