Effect of oral nutrition supplements on anthropometric and functional parameters among community-dwelling older adults: a systematic review and meta-analysis of randomized controlled trials

Abstract

Background

Oral nutrition supplement (ONS) prescription in older adults is affordable for preventing malnutrition through the use of multiple nutrients and energy products. However, there is a gap in knowledge regarding the benefits of ONS supplementation on anthropometric and functional parameters in malnourished older adults without complicated conditions, known as community-dwelling older people.

Methods

Electronic databases, including PubMed/Medline, SCOPUS, Web of Science, and the Cochrane Central Register of Controlled Trials, were searched to answer this question through August 10, 2023. The study adhered to the PRISMA guidelines and was registered on PROSPERO. The Cochrane risk of bias tool assessed the quality of randomized controlled trials (RCTs). The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) tool was used to evaluate the certainty of evidence. In total, 12 RCTs (n = 1320) were initially included.

Results

ONS had a positive effect on body weight (weighted mean difference (WMD): 1.33, 95% CI: 0.94, 1.72, P = 0.00) with moderate certainty, BMI (WMD: 0.36, 95% CI: 0.06, 0.68, P = 0.02), calf circumference (WMD: 0.27, 95% CI: 0.08, 0.46, P < 0.001) with low certainty and midarm circumference (WMD: 0.38, 95% CI: -0.14, 0.89, P = < 0.15, I2: 87.2 P < 0.00) with very low certainty. Significant improvements in grip strength (WMD: 1.012, 95% CI: 0.37, 1.65, P < 0.00) and a slightly positive change in gait speed (WMD: 0.04, 95% CI: 0.004.0.083, P = 0.03, I2:0.0, P:0.72) were also observed.

Conclusion

ONS in community-dwelling older persons without complicated conditions can also have beneficial effects in terms of anthropometric and functional parameters.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40795-025-01010-8.

Highlights

• Malnutrition and unintentional weight loss affect a large proportion of older adults.

• Oral nutrition supplements (ONS) can significantly improve energy and body weight in community-dwelling older individuals

• Calf circumference as an important indicator of disability improved by ONS consumption

• Community-dwelling older persons without complicated conditions can also benefit from ONS

Supplementary Information

The online version contains supplementary material available at 10.1186/s40795-025-01010-8.

Introduction

One of the most significant changes in the twenty-first century is aging, which has had a profound effect on all sectors of society. The proportion of adults aged 65 and older is expected to increase from 9.1% in 2019 to 16.7% in 2050, reaching an estimated 1.5 billion older adults worldwide [1]. This aging trend prompts the need to promote healthy aging not only for disease prevention but also for preserving independence in daily routines [2]. The World Health Organization (WHO, 2012), in their global brief on healthy aging, reported that good health can “add life to years” [3].

There are significant challenges associated with this aging trend, such as protein-energy malnutrition, which is often poorly managed or unrecognized in free-living older adults [4]. Malnutrition prevalence worldwide varies from 10 to 85% across different age groups. A multinational study from all five continents reported a mean prevalence of 22.8%, with considerable differences among settings. On the basis of this combined dataset, 12.5% of community-dwelling older people were malnourished, and 57.5% were at risk of malnutrition [5, 6].Approximately 27% of community-living older people also experience unintentional weight loss, which is defined as more than 5% weight loss over 6–12 months, and 15–30% of community-dwelling older individuals experience appetite loss, which is associated with poor health outcomes [7, 8].

Aging and malnutrition have a bidirectional link. While aging vulnerability leads to nutrition-related disorders, malnutrition results in poor health problems, including lower quality of life and more frequent and longer hospital stays, representing a vicious cycle [9–11]. Older persons often experience natural age-related changes, such as changes in cognition and memory, declining activity, impaired sensory perception, malabsorption, and poor dentition, which may affect dietary intake due to their aging process [12, 13] Dietary intake in terms of quality and quantity are important modifiable risk factors that can be used to both prevent and treat malnutrition. A common approach for treating malnutrition is the use of nutritional supplements. ONS is defined as a commercially available, multinutrient, ready-to-consume product designed to provide a nutritious solution in terms of calories and all other nutritional contents that can be taken directly or added to food. These supplements can contain nonnutritive dense chemicals claimed to have biologically beneficial effects or provide nutrients to increase intake [14]. For many people, ONS is an affordable way to increase their caloric and protein intake. This form of energy supplementation helps maintain or improve a patient's nutritional status, addresses the problem of uniform energy in a natural diet, and provides balanced nutrients to meet the patient's needs. For hospitalized or community-dwelling older adults, ONS can also be used as a regular meal or as a supplement with 200–600 kcal per day between meals [15, 16]. Using oral nutrition supplements alone or in combination with other nutrition interventions among the older population living in the community may result in significant health benefits and cost savings [17, 18]. A growing portion of the aging population that is at risk of malnutrition may respond favorably to this straightforward preventive nutritional intervention. Research has demonstrated the effects on the clinical, nutritional, and functional properties [19] of ONS in older persons, mostly malnourished hospitalized patients receiving enteral nutrition along with other interventions (i.e., dietary advice), or in more complicated patients, mostly malnourished hospitalized patients [20–22] through enteral nutrition along with other interventions (i.e., dietary advice) [23–25] or in more complicated patients [26]. Owing to different supplementation strategies, study settings, and types of participants, the existing evidence is still not univocal in the free-living older population who do not have significant impairments in daily functioning or advanced stages of chronic illnesses that require special medical care.

Given this background, we aimed to summarize the effects of nutritional supplementation on key health outcomes, including anthropometric indices, physical function (i.e., hand grip and gait speed), and caloric and macronutrient intake outcomes in community-dwelling older people, stratifying our results by calorie, duration, placebo, and quality assessment.

Materials and methods

Study protocol

The current systematic review and meta-analysis adhered to PRISMA guidelines for the identification, screening, eligibility, and inclusion of studies [27]. We followed a structured protocol that is registered in the International Prospective Register of Systematic Reviews (PROSPERO), registration number: CRD42023467793.

Data sources and literature search strategy

Two investigators (AHN and AP) developed a search strategy using keywords and Medical Subject Headings (i.e., MeSH) in Medline; then, other database search strategies were modified. The detailed search strategy is shown in supplementary Table 1. Two researchers independently searched and screened all available trials in PubMed/Medline, SCOPUS, Web of Science, and the Cochrane Central Register of Controlled Trials up to August 10th, to investigate the effects of ONS on the anthropometric and functional status of older adults. English-language publications were considered. Any inconsistencies during the extraction process were discussed and resolved by consensus with a third author (AL). The included studies were chained backward and forward. Duplicate entries were eliminated by importing the search results into EndNote.

Table 1.

The PICO criteria for inclusion criteria are summarized in Table 1

Population	Older people (mean age > 60) living in the community
intervention	Treatment that used ready-to-use multi-nutrient product
comparator	Standard diet with or without placebo, or any other intervention that is identical in the intervention and control groups except for ONS
outcome	With at least one assessment among (weight, BMI, MAC,CC, hand grip, gait speed)

The inclusion criteria for the evaluation of the population, intervention, comparison, results, and study design were determined a priori. The included articles had to meet the following characteristics: 1) original research paper with a randomized control trial study design; 2) studies that involved older individuals (aged > 60 years) living in the community (i.e., free-living or receiving community home care services or community sheltered); and 3) studies that used ONS alone or in combination with other interventions as long as the comparator was the same set of interventions without the mentioned nutritional intervention (i.e., dietary counseling plus oral nutrition supplement as dietary counseling). ONS is defined as a commercially available, ready-to-consume product to provides a nutritious solution in terms of calories and all other nutritional contents that can be taken directly or added to food and drinks [28]. Table 1 summarizes the PICO criteria for study inclusion.. Articles were excluded if they met any of the following criteria: 1) meta-analysis, systematic review, case report, letter, or review, 2) animal studies on lactation or pregnancy, 3) patient older people and those living in nursing homes, 4) parenteral and enteral tube feeding, 5) home-prepared or disease-specific ONS, 6) intervention that uses specific vitamins, minerals or any particular nutrient alone or in combination with ONS, 7) intervention that uses a single specific type of macronutrient, and 8) studies that use ONS with the same amount of calories as the intervention.

Data analysis

Data extraction

All important features and results, including author, publication year, nation, design of the studies, dose and duration, intervention protocol and control, age, sex, and sample size, were recorded via a prepiloted data extraction table. Data extraction and risk of bias evaluation were also carried out by two independent researchers, and any inconsistencies were resolved by a third experienced researcher. In articles that reported outcome assessment at a different time point, the end of the intervention was considered. Furthermore, only individual ONS outcomes were collected from those studies that combined ONS with other interventions.

Quality assessment

The Cochrane risk of bias tool was used to assess the quality of the RCTs. Selection bias, performance bias, detection bias, attribution bias, reporting bias, and other forms of bias were among the six domains that were evaluated. The answers “yes” and “NO” were representative of low and high risk of bias, respectively. In some cases, bias may have been categorized as 'unclear' due to inadequate and incomplete reporting of methodological details in the original studies. The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) tool was used to assess the certainty of evidence.

Statistical analysis

Random effect models were used to perform a meta-analysis if at least two trials assessed an outcome and if the desired data (i.e., mean ± SD for continuous variables) were published or calculable. In studies that used standard errors, the data were converted to SDs. The 95% confidence interval (95% CI), mean difference, and weighted mean difference were used as numerical indicator values. As there were differences in placebo/control feeding, ONS dose, and duration that may have had an impact on effect size between studies, a random model effect was used. Forest plot visualizing the meta-analysis results. Meta-analyses were conducted with Stata software, version 17 (Stata Corp, College Station, TX).

The degree of heterogeneity was calculated via the Q test, the I² statistic, and the p value to measure the strength of evidence. A p-value < 0.05 and I² over 50% indicate a high level of heterogeneity [29]. Publication bias was assessed via Egger’s and Begg's tests, and visual inspection of the funnel plots was used to assess publication bias. Sensitivity analysis was estimated via the Hartung‒Keuls estimator for the variance of the pooled effect estimate, and the between-study variance for continuous outcomes was estimated via the limited maximum likelihood estimator. In the case of publication bias, this was corrected via trim-and-fill analysis.

Results

The flowchart of the meta-analysis and systematic review is presented in Fig. 1. The database search identified 12,950 records. An additional 32 other records were identified through other sources. After 4468 duplicates and 8521 prevalent studies were removed, 4037 records remained for title and abstract screening. The full texts of 33 records were screened. In total, 12 articles (n = 1317) met the inclusion criteria, and 21 articles were excluded for not reporting the outcome of interest ( [30–32]n = 3); combining with other interventions ( [33–35]n = 3); starting at the hospital ( [36–39], n = 4); (study protocol, i.e., type of oral supplement with specific nutrients [40, 41], type of placebo that increased energy content [25, 41]; prescribing multivitamins or minerals along with oral supplements [42–44], n = 7 and non-RCT trial [45–48], n = 4). PRISMA 2020 Flow diagram of the selection study process is presented in Fig. 1.

Fig. 1.

PRISMA 2020 flow diagram of the selection study process

Study characteristics

The study characteristics and demographic information of the included articles are summarized in Table 2. The 12 studies included were published between 1995 and 2023. The largest study recruited 308 [49] participants and the smallest study recruited 38 [50] participants. Two studies took place in China [51, 52], two studies in the UK [49, 53], two studies in the Netherlands [54, 55], three studies in Canada [56–58], one study in Korea [59], one study in Thailand [50] and one study in Hong Kong [60]. Two other studies in Thailand and Netherland reported that recruitment took place in an outpatient clinic [50, 54]. The intervention duration ranged between 8 and 24 weeks. All of the studies included male and female participants Table 2.

Table 2.

Characteristics of included studies

						Sample size			Age (mean ± SD)		Intervention		Study funding source/ conflict of interest
Author	Year	Location	Study design	Patient source	Female Gender (Int/Con)	INT	Con	Duration (weeks)	Int	Con	Int	Con	Calorie provided daily
Assantachai, P [50]	2020	Thailand	RCT	Outpatient Geriatric clinic	57.9 overall	20	15	24	81.5 ± 5.3	81.5 ± 6.11	chicken based formula	similarly flavored drinking water	150	Educational Fund
Kim [59]	2013	Korea	RCT	Health care service	79.1/79.6	41	43	12	78.9 ± 5.5	78.4 ± 6.0	Commercial Greenbia HP	none	400	Health Promotion Fund, Ministry of Health & Welfare, Republic of Korea
van de Bool [54]	2017	Netherlands	RCT	Outpatient clinic	41/57.1	38	35	16	62.8 ± 1.3	62.2 ± 1.3	Commercial, Nutricia NV,Zoetermeer, the Netherlands	Similarly flavored drink	250–375	grants from Netherlands Lung Foundation and Nutricia Research,
Xie [51]	2023	China	RCT	Community Health Service Centers	36/15	101	100	12	74.65 ± 7.13	75.76 ± 7.73	Commercial, Abbott Trading	Starch capsule	400	Non declared
Yeung [52]	2022	China	RCT	Community living	51/49	42	40	12	66.3 ± 9.5	66.7 ± 10.0	Commercial (Fresubin® Powder	none	600	funding and study products from Fresenius-Kabi Deutschland
Edington [53]	2004	Uk	RCT	Post discharge	56/53	51	49	8	76.8 ± 5.3	79.3 ± 8	Esure Plus tetrapak, Enlives tetrapak, Formances Pudding or Ensure Bars, Abbott Laboratories	none	600–1000	Abbott Laboratories
Payette [56]	2002	Canada	RCT	Community living	71/71	42	39	16	81.6 ± 7.5	78.6 ± 6.1	Ensure or Ensure Plus	none	400–700	Abbott Laboratories
Smith [49, 62]	2020	Uk	RCT	Primary care	66/68	154	154	12	71.3 ± 11.1	71.63 ± 10.31	Fortisip Compact range, Nutricia	none	300–600	National Institute of Health Research and nutricia company
Wouters-Wesseling [55]	2003	Netherland	RCT	Free living	62/53	38	38	24	83.97 ± 6.3	81.07 ± 6.9	Liquid calorie supplement	Similarly flavored drink	500	Numico Research B.V
Gray Donald [57]	1995	Canada	RCT	Free living	74/67	24	24	12	76 ± 7	79 ± 8	Ensure/Ensure plus/Enrich	none	500–700	National health Research and development program
Zhu [60]	2019	Hong Kong	RCT	Community dwelling	80.6/72.5	36	40	12	74.8 ± 6.9	74.5 ± 7.1	Ensure NutriVigor	none	460	none
Krondal [58]	1999	Canada	RCT	Community dwelling	75/77	36	35	16	70 ± 7a		boost, mead Johnson nutritional’s	none	235	INNOVUS research Inc

Abbreviations: F Female, (female percentage in each group is presented). IN Intervention group, CON Control group, RCT Randomized controlled trial, A Overall reported

Intervention protocols

All of the included studies used liquid commercial formulas in the form of ready-made or powder that should be mixed with water. One of the studies used dietary advice [49], the other used nutrition education [51], and two other studies conducted exercise sessions [54, 60] plus oral nutrition supplements, all of which were controlled, as they were identical in both groups. Three studies used similarly flavored noncaloric drinks [50, 54, 55], one of the studies used starch-made capsules [51], and the other eight studies did not use a placebo. The recommended doses across the studies were between one and three servings a day with no daily limit [58]. The daily energy provided from ONS ranged between 150 [50] and 1000 kcal [53]. To distinguish the effects of different supplementation volumes, a calorie subgroup analysis was conducted. The inclusion criteria used in different studies differ based on body mass index (BMI), Mini Nutrition Assessment (MNA), or Malnutrition Universal Screening Tool (MUST) score BMIs lower than 20, 23, 24,25, or 27 (alone or in a combination with considerable weight loss), or loss of muscle mass (i.e., fat mass index (FMI) under the sex‐ and age‐specific 25th criteria) is considered to predict the risk of malnutrition or not.

Among the 12 included studies, six (50%) reported full or partial funding from the industry, while four (33.3%) reported alternate sources of funding. Two studies did not declare a conflict of interest or funding details.

Among the 12 studies identified, 9 studies were judged to be at low risk of random sequence generation [50, 52, 53, 55, 59–62], six studies of allocation concealment [49, 50, 54, 55, 59, 60], and six studies were considered to be at high risk of performance bias due to blinding [51, 53, 56–58, 62]. Risk of bias assessments are summarized in Fig. 2, supplementary table.

Fig. 2.

Risk of bias judgment in the included studies

Effects of ONS on anthropometric indices

Effects of ONS on body weight

An analysis of eight studies that reported the effect of ONS on body weight revealed that ONS had a positive effect on body weight (WMD: 1.33, 95% CI: 0.94, 1.72; P = 0.00), with low heterogeneity (I²: 12.8, p = 28.6). Subgroup analysis was performed based on duration, age (less than or greater than 12 weeks), placebo, and calorie intake (less than or greater than 400 kcal), and quality assessment revealed no significant differences between the groups. The findings are summarized in Fig. 3, Table 3.

Fig. 3.

Forest plot comparing the effect of ONS on body weight

Table 3.

Results of subgroup analysis for the effects of oral nutrition supplement on anthropometric indices, functional parameters and dietary intake in community dwelling elderly

	Number of studies	WMD (95% CI)	P-value1	I2 (%)2	P for heterogeneity3	P between Subgroups
Anthropometric indices
Body weight
Overall	8	1.33 (0.94,1.72)	<0.000	18.2	0.286
Duration						0.38
≤ 12 weeks	6	1.20 (0.65, 1.74)	<0.00	35.7	0.16
> 12 weeks	2	1.58 (0.91, 2.25)	<0.001	0.0	0.59
Placebo						0.5
with	2	1.55 (0.74, 2.37)	<0.001	37.7	0.59
without	6	1.23 (0.71, 1.74)	<0.001	0	0.15
Calorie						0.38
< 400 kcal	4	1.06 (0.19, 1.94)	<0.001	44.8	0.14
≥ 400 kcal	4	1.33 (0.94, 1.74)	<0.001	0.0	0.63
Quality assessment						0.5
Good	3	0.81(-0.35,1.98)	0.17	69.2	0.039
Fair	3	1.47(1.03,1.98)	<0.00	0	0.95
Poor	2	1.63(0.85,2.40)	<0.00	0	0.75
Body mass index
Overall	5	0.36 (0.06, 0.68)	0.02	68.5	0.013
Duration						0.25
≤12 weeks	3	0.55 (0.37, 0.73)	0.01	0	0.47
>12 weeks	2	0.22 (-0.31, 0.76)	0.2	83	0.07
Placebo						0.43
with	2	0.53 (0.12, 0.94)	0.381	86.9	0.88
without	3	0.28 (-0.16, 0.73)	0.232	99.7	0.00
calorie						0.38
< 400 kcal	3	0.28 (-0.11, 0.68)	0.16	55.3	0.10
≥ 400 kcal	2	0.51 (0.2, 0.82)	<0.00	0.32	0.22
Quality assessment
						0.15
Good	1	0.22 (-0.36, 0.80)	0.45	.	.
Fair	2	0.59 (0.41,0.78)	<0.00	0.0	0.89
poor	2	0.17 (-0.29,0.64)	0.46	53.4	0.14
Calf circumference
Overall	5	0.27 (0.08, 0.46)	<0.0	27.6	0.23
Duration						0.44
≤12 weeks	3	0.41 (0.21,0.58)	0.00	0.0	0.45
>12 weeks	2	0.30 (0.16, 0.44)	0.11	0.0	0.32
Placebo						0.91
with	2	0.36 ( -0.79, 1.53)	0.53	66.1	0.08
Without	3	1.23 (0.71, 1.74)	0.0	0	0.37
Calorie						0.91
≤400 kcal	2	0.36 (-0.79, 1.53)	0.53	66.1	0.08
>400 kcal	3	0.30 ( 0.16, 0.44)	0.0	0	0.37
Quality assessment						0.81
Good	1	0.20(-0.01,0.41)	0.06	0.0	0.0
Fair	2	0.21(-0.25, 0.69)	0.36	64.4	0.09
Poor	2	0.49(-0.37,1.36)	0.26	36.8	0.20
Mid Arm Circumference
Overall	3	0.37 (-0.13, 0.88)	0.15	0.0	0.72
Functional parameters
Hand grip
Overall	6	1.01 (0.37,1.65)	0.002	20	0.28
Duration						0.73
< 12 weeks	4	1.11 (0.24, 1.98)	0.01	44.2	0.14
≥ 12 weeks	2	0.85 (-0.41, 2.12)	0.18	0	0.35
calorie						0.54
< 400 kcal	2	1.43 (0.71, 2.16)	0.00	0.0	0.51
≥ 400 kcal	3	0.36 (-0.44,1.17)	0.37	0.0	0.65
placebo						0.45
with	4	1.36 (0.51,2.20)	0.002	0	0.00
without	2	-0.81 (-0.35,1.97)	0.17	59.0	0.59
Bias						0.20
Good	3	1.31 (0.43, 2.19)	0.003	0.00	0.91
Fair	1	-0.30 ( -3.07,2.47)	0.83	.	.
Poor	2	1.05 (-0.54, 2.65)	0.19	77	0.03
Gait Speed
overall	4	0.043 (0.004.0.083)	0.03	0.0	0.72
Macronutrient and energy intake
Energy
Overall	9	130.09 (2.09, 258.09)	0.04	96.7	0
Duration						0.18
≤ 12 weeks	5	225.53 (-1.48, 452.54)	0.05	96.9	≤0.000
> 12 weeks	4	16.07 (-197.41,229.57)	0.88	96.9	≤0.000
Placebo						0.61
with	3	46.38(-403.39,496.17)	0.84	97.6	<0.00
without	6	166.92(25.49,308.347)	0.02	96.8	<0.00
Calorie						0.75
< 400 kcal	5	110.09 (-37.01,257.21)	0.14	95.8	<0.00
≥ 400 kcal	4	157.31(-102.01,416.64)	0.23	97	<0.00
Bias						0.72
Good	4	133.57(113.92,381.079)	0.29	96.9	0.00
Fair	3	51.39(-195.63,298.426)	0.68	89.4	0.00
Poor	2	236.47(-151.63,624.58)	0.23	97.5	0.00
Protein Intake
Overal	9	8.41 (1.78, 15.04)	0.013	97.6	0.00
Duration						0.38
≤ 12 weeks	5	11.38 (0.01, 22.768)	0.05	97.7	≤0.000
> 12 weeks	4	5.21 ( -2.70, 13.137)	0.19	95.7	≤0.000
Placebo						0.95
with	3	8.03 (-11.38, 27.460)	0.41	97.5	<0.00
without	6	8.62 (0.38, 16.861)	0.04	98	<0.00
calorie						0.31
< 400 kcal	5	11.56 (-1.28, 24.41)	0.07	98.5	<0.00
≥ 400 kcal	4	8.41 (1.78, 15.3)	0.22	93.1	<0.00
Bias						0.08
Good	4	6.98 (-1.078, 15.041)	0.09	95.6	<0.00
Fair	3	1.51(-7.49, 10.52)	0.74	89.4	0.005
Poor	2	19.56 (6.24, 32.88)	0.004	94.3	0.00
carbohydrate Intake
Overall	4	19.16(-9.95, 48.29)	0.19	88.9	<0.00
Fat Intake
Overall	4	30.54 (-30.40, 91.48)	0.32	99.5	<0.00

WMD Weighted Mean Difference

¹P for heterogeneity, within subgroup

²An I² value >50% shows significant between-study heterogeneity

³P for heterogeneity, between subgroups

Effect of ONS on BMI

Five studies with a total of 554 participants evaluated the effect of ONS on BMI. An RM model was used for the meta-analysis. ONS consumption led to a significant increase in BMI (WMD: 0.36 95% CI; 0.05, 0.67 P = 0.02), which was not significant in the subgroup analysis. The results are summarized in Fig. 4, Table 3. Moderate heterogeneity was also observed (I²:68.5 P = 0.013).

Fig. 4.

Forest plot comparing the effect of ONS on BMI

Effects of ONS on calf and midarm circumferences

Five studies with 522 participants reported a significant positive effect of ONS on calf circumference (CC) with low heterogeneity (WMD: 0.27, 95% CI: 0.08, 0.46, P ≤ 0.0; I²: 27.6, P = 0.23), which did not lead to between-group significant results. Three studies reported an inconsistent effect of ONS on mid-arm circumference (WMD: 0.37, 95% CI: −0.14, 0.89 P = 0.15). The findings are detailed in Tables 3, Figs. 5 and 6.

Fig. 5.

Forest plot comparing the effect of ONS on calf circumference

Fig. 6.

Forest plot comparing the effect of ONS on mid arm circumference

Effects of ONS on functional parameters

ONS supplementation in six included studies that reported grip strength was associated with a significant increase in this parameter (WMD: 1.012, 95% CI: 0.37, 1.65, P < 0.00), with a low level of heterogeneity (I²:20. P = 0.28), which did not rely on dose, duration of placebo or quality assessment. This estimate was also not significant in terms of sensitivity. A slightly positive effect of ONS consumption on gait speed was observed (WMD: 0.04, 95% CI: 0.004.0.083, P = 0.03; I²: 0.0, P = 0.72). The results are presented in Table 3, Figs 7 and 8.

Fig. 7.

Forest plot comparing the effect of ONS on hand grip

Fig. 8.

Forest plot comparing the effect of ONS on gait speed

Effects of ONS on dietary intake

Effect of ONS on energy intake

Nine studies reported the effect of ONS on energy intake, and the results revealed that ONS consumption led to a significant increase in energy intake (WMD: 130.09, 95% CI: 2.08, 258.09 P = 0.04), CIs were very wide, and substantial heterogeneity was reported (I²:96.7 P < 0.00). One of the studies [54] reported the previous month's habitual dietary intake over a total of 3 months. After removing this study from the analysis, the results changed to (WMD: 174.21 95% CI: 38.81, 309.62, p = 0.012), again with high heterogeneity (I2:96.7 P > 0.000) and between-group analysis revealed no difference.

Effects of ONS on protein intake

For energy intake, nine studies reported positive changes in protein intake following supplementation with ONS, with evidence of substantial heterogeneity (WMD: 8.41, 95% CI: 1.78, 15.04; P = 0.013; I²:96.7, P < 0.00). The differences between the groups were not significant. As the Egger test result of 0.028 indicated possible evidence of publication bias, the trim and fill adjustment approach was carried out and demonstrated that 14 studies had to be trimmed and filled, and the resulting combined effect size became insignificant (WMD: −3.16 95% CI: −9.90– 3.59). Four studies that reported carbohydrate and fat intake changes following supplementation reported inconsistent effects of ONS on the abovementioned parameters (WMD, 19.16; 95% CI: −9.95, 48.29; P = 0.19; I²:88.9; and WMD: 30.54; 95% CI:−30.40, 91.48, p = 0.32; I2: 99.5). The results are summarized in Fig. 9. Supplementary material 2.

Fig. 9.

Comparing the effect of ONS on dietary intake

Publication bias and sensitivity analysis

Egger’s regression test indicated no publication bias for weight (p = 0.365), BMI (p = 0.74), CC (p = 0.81), MAC (p = 0.93), hand grip (p = 0.82), gait speed (p = 0.21), energy intake (p = 0.11), carbohydrate intake (p = 0.26), or fat intake (p = 0.96). However, there was significant publication bias for protein intake (P = 0.029). Owing to high publication bias, we performed a trim-and-fill sensitivity analysis and found that the addition of two unpublished studies changed the protein intake value to insignificant results. The results of the sensitivity analysis revealed that the overall results did not change even after each study was excluded, and the estimates of all the results of the sensitivity analysis were not significant. Supplementary material 2.

Quality assessment

The certainty of evidence was judged via the GRADE approach. The certainty of the evidence was rated moderate for body weight and gait speed; low for BMI, CC, and grip strength; and very low for mid-arm circumference. Each outcome of the GRADE assessment is available in the supplementary table.

Discussion

Since older adults tend to have poor appetite and lower energy and protein intake either in the fasting or postprandial state, calls for effective treatment strategies to prevent nutritional-associated problems at the early stage of aging are crucial [63, 64]. Among therapeutic approaches, ONS might achieve sufficient health outcomes in terms of the nutritional and functional status of older people [16, 65]. Several systematic reviews and meta-analyses have investigated the effects of ONS in older adults with various health problems in different clinical settings [22, 28, 66]. However, there is conflicting evidence and a gap in knowledge regarding the benefits of ONS supplementation in community-dwelling older people [51, 53, 56]. To the best of our knowledge, the present meta-analysis is the first to evaluate the effectiveness of ONS in community-dwelling older persons, i.e., the free-living older population. Based on these findings, ONS could significantly increase energy and body weight, BMI, and CC in community-dwelling older people. Additionally, a significant improvement in functional measures, including grip strength, and slight changes in gait speed were detected. However, there was no significant alteration in mid-arm circumference or fat or carbohydrate intake. The results remained unchanged after subgroup analysis in terms of duration of intervention, amount of calorie intake from supplements, type of intervention in the control group, and quality of the studies.

With respect to anthropometric indices, in line with our findings, a recent systematic review and meta-analysis investigating the effects of ONS in older people with anorexia proposed remarkable improvements in body weight and BMI. Additionally, ONS supplementation significantly increased energy and protein intake [67]. Similarly, another meta-analysis by Cawood et al. [68] evaluating the clinical benefits of high-protein ONS in a study population with a mean age of 74 years proposed a significant improvement in body weight and energy and protein intake. Allen et al. [69] performed a meta-analysis on the use of ONS drinks in older adults with dementia. In agreement with our conclusion, they reported that ONS significantly improved body weight and BMI [69]. Moreover, on the basis of the conclusion of a meta-analysis by Reinders et al. [70], the use of ONS combined with dietary counseling, as the most effective nutritional intervention, could significantly increase body weight and energy intake in older adults at risk of malnutrition. In contrast, on the basis of the findings of a meta-analysis byCorrea-Pérez, A., no beneficial effects of ONS on anthropometric parameters such as body weight and BMI were detected in older malnourished individuals [71]. However, it should be noted that the overall quality of evidence in the aforementioned review was very low, particularly because of the small sample size and the risk of bias [71]. Notably, alterations in BMI and body weight might not always lead to improved overall clinical results. However, improved functional status by adequate nutritional intake in older adults living in the community and not in the acute phase can sustain their freedom and quality of life.

Among the clinical community, an association between functional and nutritional status is of growing interest and one of the six recommended characteristics used by the Academy of.

Nutrition and Dietetics/American Society for Parenteral and Enteral Nutrition is a hand grip.

strength [72]. Mendes et al. reported that a lower CC is associated with lower grip strength in community-dwelling older persons [72] whereas a lower CC is associated with a greater risk of mortality in hospitalized and community-living older individuals [73].

One of the key functional parameters for older people is gait speed and muscle strength, which predict 10-year cognitive decline. [74]. Consistent with the data obtained from this meta-analysis demonstrating the beneficial effects of ONS on functional parameters, a systematic review and meta-analysis revealed that high protein ONS resulted in a meaningful increase in grip strength in a study population with a mean age of 74 years [68]. A systematic review focused onhow nutritional interventions can affect the functional outcomes of geriatric rehabilitation patients consistently revealed that ONS could improve mobility and functional outcomes [75]. Thomson et al. [28] conducted a meta-analysis to analyze the effectiveness and cost-effectiveness of ONS in frail, older subjects who are at risk of malnutrition or malnourished. The conclusions of their review were in line with those of our study. The present study results did not reveal statistically significant changes in gait speed. Veronese et al. reported that while nutritional supplementation improves grip strength, it can increase gait speed. It is hypothesized that gait speed is more dependent on pulmonary and neurological function than muscle mass alone [76]. They concluded that ONS might have some positive effect on mobility [28]. In contrast, grip strength and activities of daily living, which are functional indices of aging, did not significantly change following ONS treatment [71]. However, the quality of evidence for this conclusion was reported to be low because of the high risk of bias and small sample size [71].

It is estimated that approximately 20% of community-dwelling older people experience appetite loss. Numerous mechanisms have been proposed, including alterations in gut function, the neuroendocrine system and sensory perception. Even living environment alterations and the routines of older people can physiologically affect appetite via unknown mechanisms [77, 78]. Similarly, some pathophysiological factors in older adults, including increased activity of cholecystokinin and elevated anal sphincter muscle tension can lead to satiety and poor appetite [79–82]. Free-living older people who experience a poor appetite have a significantly lower BMI, fat-free mass, and grip strength. [83] Mechanisms of the beneficial effects of ONS on anthropometric and consequently functional parameters in older people might be explained by increased appetite. The beneficial effects of ONS on anthropometric and consequently functional parameters in older people with poor appetites may be achieved by adding ONS to normal foods such as porridge or rice soup to improve compliance [81, 84–87]. Therefore, ONS, which is convenient, simple, affordable, and provides balanced nutrients, can be a good choice for older persons, but previous studies have focused mainly on late-stage malnutrition or complicated patients when there is no other option than ONS [24, 26, 65, 75].The present meta-analysis suggests a beneficial effect of ONS in older adults in preventing unwanted weight loss and improving handgrip strength associated with better health outcomes. However, while ONS may offer benefits in preventing weight loss in older adults which does not necessarily translate to malnutrition, its prescription should be considered as part of a more comprehensive and multifaceted approach. Furthermore, more comprehensive research is needed to clarify, the role of ONS in improving the overall quality of life.

Study strengths and limitations

The present quantitative review possesses several strong points, including a relatively large sample size providing an adequate level of statistical power, conducting subgroup analysis for numerous probable confounding variables, and not sensitiveness of overall effects to any single study for anthropometric and functional parameters. However, some weaknesses should be considered when interpreting the obtained results. The use of various types of ONS for a time series that varies and is not able to run dose–response meta-analysis are among the most common important limitations. Not taking into account lifestyle changes such as exercise during intervention measurement errors, and misclassification of the participants in terms of assessing energy and dietary intake are also the other limitations. Additionally, including English language articles might overlook some valuable data. It should be noted that statistically but not clinically significant findings of the present meta-analysis are not generalizable for application in clinical settings.

Conclusion

Overall, ONS could significantly improve anthropometric and functional parameters, including body weight, BMI, CC, grip strength, and gait speed, in community-dwelling older people. A significant increase in energy intake following ONS consumption might explain such improvements in anthropometric and consequently functional parameters. Notably, changes in the abovementioned parameters may not necessarily translate to better final clinical outcomes, butcommunity-dwelling older people who are not in the acute phase can delay institutionalization, manage chronic disease, reduce the risk of falls, maintain independence, and enhance overall quality of life. Further well-designed research assessing more functional parameters with a large sample size representing the community is suggested to investigate the effectiveness of ONS in community-dwelling older people. In further studies, consideration should also be given to mediating factors such as exercise, healthcare education, fun activities, and psychotherapeutic support to assess the impact of ONS on nutritional status and function in community-dwelling older individuals.

Abbreviations

ONS

Oral nutrition supplements

GRADE

The Grading of Recommendations, Assessment, Development, and Evaluation

WMD

weighted mean difference,

BMI

body mass index

MNA

Mini Nutrition Assessment

MUST

Malnutrition Universal Screening Tool

CC

calf circumference

Authors’ contributions

Conceptualization and methodology: Azadbakht, Asghari Hanjani, Arzhang. Data interpretation and analysis: Asghari Hanjani, Arzhang. Manuscript drafting: Asghari Hanjani, Arzhang. Statistical analysis: Arzhang. Supervision and critical revision: Azadbakht.

Funding

This study was funded by the Tehran University of Medical Sciences (grant number: 1402–4-212–697790.

Data availability

Data sets generated during the current study are available from the corresponding author on reasonable requests.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.