Abstract
Introduction
Malnutrition is a frequent condition in the elderly especially in hospitals and in nursing homes, and even among the free-living elders the prevalence is not negligible (5-10%). Awareness towards malnutrition is still limited. The lack of time for nutritional assessment by the overcommitted healthcare personnel, including the general practitioners (GPs), may represent one possible explanation for limited recognition of malnutrition. Therefore, a self-administered instrument could be useful in raising alert on the GPs and allow early detection of malnutrition and early care provision. The aim of the present study was to analyze the validity of the Self-MNA that takes cue from the Mini Nutritional Assessment-Short Form (MNA-SF) and has been adapted to be self-administered by free-living elderly subjects.
Methods
Participants were recruited from patients referring to the GP offices in Italy. Nutritional evaluation was performed through the administration of Full-MNA, MNA-SF and Self-MNA. The comorbidity level was assessed through the Cumulative Illness Rating Scale (CIRS). The level of difficulty in filling out the test was reported by the participants, and the time spent to complete the Self-MNA was also registered.
Results
A total of 226 subjects, 125 women and 101 men (75.1 ± 8 and 75.3 ± 8 years old, respectively; p=0.89) were enrolled, and 214 (94.7%) of them completed the Self-MNA. According with the Full-MNA test score, 8.4% of women and 3.5% of men were classified as malnourished, whereas 32.7% of women and 31.4% of men were at risk of malnutrition. Agreement between Self-MNA and Full-MNA, and Self-MNA vs. MNA-SF was classified as “moderate (k = 0.476 and 0.496 respectively; p < 0.001). Self-MNA showed a fair predictive value compared to the Full-MNA and MNA-SF tests (76.6 and 79.9%, respectively) with a barely adequate sensitivity (70.9 and 75.4%, respectively). The analysis of the characteristics of FN (false negative: subjects who were considered at risk of malnutrition or malnourished at Full-MNA but not at Self-MNA) showed that the clinical and functional aspects of these subjects (age, comorbidity and severity, time necessary to complete the Self-MNA, decrease in food intake, severe illness in the past 3 months, dementia and depression, fluid intake, need for feeding assistance, arm and calf circumferences) were very similar to the characteristics of true positive subjects. Patients required 6.7 ± 4.5 minutes to complete the test and 25 subjects (11.7%) needed more than 10 minutes, up to a maximum of 30 minutes. Patients who stated a greater difficulty were older (79.8 ± 7 vs. 73.5 ± 7 years; p< 0.001), they were more «malnourished» at Full-MNA (10.7 vs. 1,7%; p= 0.006) and clinical status was characterized by a higher severity index (1.72 ± 0.6 vs. 1.41 ± 0.4; p= 0.008).
Conclusion
In the present study we investigated the validity of the Self-MNA in a sample of free-living elderly subjects. The results obtained confirm the validity of the test that may represent a useful tool for the GPs, although some important limitations need to be considered, limiting its use in clinical practice.
Key words: Nutritional risk, Mini nutritional assessment, Self-MNA, general practitioner
Introduction
Malnutrition is a frequent condition in older adults especially in hospitalized elder and in nursing home residents; indeed, the prevalence of malnutritiondiffers significantly across the healthcare settings (1, 2), and even among the free-living elderly, it is not negligible (5-10%) (3, 4). Moreover, 38% of community- dwelling older adults are at risk of malnutrition (5).
The pathophysiology of age-related malnutrition is complex and multifactorial, mainly due to senile anorexia, increased energy and nutrient requirements, comorbidity, polypharmacy, psychological and socio-economic status (6, 7, 8).
Malnutrition can worsen age-related sarcopenia (in turn increasing the risk of disability), decrease immune response, precipitate the onset of pressure sores (9, 10, 11), contribute to cognitive decline and depression (12, 13), leading,finally,to frailty and increased rate of hospital admission and readmissions (14, 15, 16).
Despite the availability of a number of validated screening tools for malnutrition (17, 18, 19, 20, 21), awareness towards malnutrition is still limited (22, 23, 24, 25). The lack of time for nutritional assessment by the overcommitted healthcare personnel (26), including the general practitioners (GPs), may represent one possible explanation for the limited recognition of malnutrition (27, 28). Therefore, a self-administered instrument may be very useful in the general practice setting, allowing early detection of malnutrition and an early care provision.
The Mini Nutritional Assessment (MNA) (24) is, currently, the most used and validated tool for the nutritional status evaluation in older adults; it is divided into a short-form version (MNA-SF) for the screening of malnutrition and a full version (Full-MNA) for a thorough assessment of nutritional risk. In addition, already widely validated in the literature, it has recently been proposed a self-administered version, namely the Self-MNA, that takes its origin from the MNA-SF (29).
The aim of the present study was to analyze the validity of the Self-MNA in elderly free-living subjects.
Materials and methods
Subjects and Setting
Participants were recruited among patients referring to 24 GPs offices in Italy throughout the national territory. The GPs were selected on a voluntary basis among those affiliated with the SIMPeSV (Italian Society of Preventive Medicine and Lifestyles). Eligible participants were 65 years and older, without severe cognitive impairment. Recruitment took place from January to April 2016 during routine visits performed by the GPs in their offices.
Oral and written informed consent was obtained from all participants.
The nutritional evaluation was performed by the GPs, who administered the Full-MNA and MNA-SF, whereas Self-MNA was self-administered: p ]- Full-MNA (24). It consists of 18 questions grouped into 4 parts: anthropometry [body mass index (BMI), weight loss, mid upper arm and calf circumferences], clinical status (medications, mobility, pressure sores and skin ulcers, lifestyle, psychological stress or neuropsychological problems), dietary assessment (autonomy on feeding, quality and number of meals, fluid intake) and self- perception about health and nutrition. A weighted score is assigned to each question/item. The total score is the sum of screening and assessment scores and ranges from 0 to 30 points. Results exceeding 23.5 points correspond to a normal, adequate state of nutrition in healthy individuals. Scores ranging from 17 to 23.5 points are indicative of increased nutritional risk, and values below 17 points strongly suggest a state of malnutrition;
- MNA-SF (30, 31). The MNA-SF has 6 questions instead of 18, it eliminates time-consuming and subjective items. A screening score ≥ 12 corresponds to a normal nutritional status; between 8 and 11 points subjects are considered at risk of malnutrition; below 8 points subjects are considered malnourished;
- The Self-MNA (29) represents a new version of MNA-SF with the same 6 questions adapted to be self-administered in adults aged 65 years and older (or with the assistance of their caregivers). The Self-MNA was translated from English to Italian using a standard forward-backward translation procedure. To fill out the Self-MNA, participants were asked to follow the instructions in the form. No additional instructions or assistance to complete the questionnaire were provided in order to allow the tool's intended use as a self-administered assessment.
The GPs were asked also to evaluate the comorbidity level of the participants according with the Cumulative Illness Rating Scale (CIRS) (32). Both GPs and patients were finally asked, separately, to define the level of difficulty (from 0-no difficulty to 10-maximum difficulty) that the patients showed in completing the; also the time necessary to fill out the Self- MNA was recorded.
Data analysis and statistics
The principal goal of the study was to test the difference in the estimated prevalence of correctly identified malnourished subjects at the Full-MNA vs. the Self-MNA. Sample size calculation was based assuming significance (alpha) set at 0.05 and power (beta) set at 0.90. Therefore, approximately 200 participants were required.
The predictive value of Self-MNA versus MNA-SF and MNA nutritional categorization was estimated through the evaluation of overall predictive value, sensitivity, specificity, predictive value of positive or negative test (33). For this analysis we considered only two levels of risk for MNA, MNASF and Self-MNA, aggregating subjects with high or moderate risk of malnutrition in a single level of risk.
Agreement between the assessment methods was also analyzed by kappa (k) statistic. The value of k varies from 0 to 1, with values < 0.2 indicating poor, 0.21–0.4 fair, 0.41– 0.6 moderate, 0.61–0.8 substantial and > 0.8 almost perfect concordance (34).
Distributional assumptions were evaluated. Data are presented as mean ± SD or percentage, as appropriate. Student's t-test and ANOVA were used. ANCOVA was used for adjustment for confounding variables, as specified in the text and tables. Pearson χ2 was used for the comparison of the distribution of categorical variables. Differences were considered to be statistically significant for p< 0.05. Statistical analysis was performed using SPSS statistical software (SPSS Inc. Wacker Drive, Chicago, IL, USA).
Results
Participants' Characteristics
A total of 226 subjects, 125 women and 101 men (75.1 ± 8 and 75.3 ± 8 years old, respectively; p=0.89) were enrolled, and 214 (94.7%) of them completed the Self-MNA (Figure 1). Characteristics of study participants are summarized in Table 1. According with the Full-MNA test score, 8.4% of women and 3.5% of men were classified as malnourished (Full-MNA test score < 17), whereas 32.7% of women and 31.4% of men were at risk of malnutrition (Full-MNA test score: 17- 23.5). BMI ≤ 18.5 kg/m2 was observed in one participant only while 16.8% of patients were obese (BMI ≥ 30 kg/m2).
Figure 1.
Flowchart
Table 1.
Clinical/ functional characteristics and nutritional status in the study population
| Males | Females | p | ||
|---|---|---|---|---|
| N | 101 | 125 | ||
| Age | Years | 75.3±8 | 75.1±8 | 0.89 |
| > 75 years (%) | 47 | 43.5 | 0.35 | |
| BMI | kg/m2 | 26.6±4 | 26.3±5 | 0.65 |
| < 18.5 kg/m2 (%) | 0 | 0.9 | 0.47 | |
| 18.5-24.9 kg/m2 (%) | 37.6 | 45.2 | ||
| 25-29.9 kg/m2 (%) | 45.5 | 37.1 | ||
| ≥ 30 kg/m2 (%) | 16.8 | 16.9 | ||
| CIRS scale | Comorbity index | 1.9±1.8 | 1.68±1.8 | 0.67 |
| High Comorbidity index (>4) (%) | 8.8 | 8.9 | 0.58 | |
| Severity index | 1.54±0.4 | 1.42±0.4 | 0.03 | |
| High Severity index (> 1.5) (%) | 49 | 35.8 | 0.03 | |
| Full-MNA | Score | 24.7±4 | 23.9±4 | 0.18 |
| Malnourished (score < 17) (%) | 3.5 | 8.4 | 0.34 | |
| Patients at risk of malnutrition (score 17-23.5) (%) | 31.4 | 32.7 | ||
| Normal nutritional status (score ≥ 24) (%) | 65.1 | 58.9 | ||
| MNA-SF | Score | 12.2±2 | 11.6±3 | 0.09 |
| Malnourished (score 0-7) (%) | 4.9 | 11.3 | 0.22 | |
| Patients at risk of malnutrition (score 8-11) (%) | 25.5 | 22.6 | ||
| Normal nutritional status (score ≥ 12) (%) | 69.6 | 66.1 | ||
| Self-MNA | Score | 11.9±2 | 11.4±3 | 0.16 |
| Malnourished (score 0-7) (%) | 8.2 | 9.2 | 0.12 | |
| Patients at risk of malnutrition (score 8-11) (%) | 21.4 | 33 | ||
| Normal nutritional status (score ≥ 12) (%) | 70,4 | 57.8 |
Data are presented as mean ± standard deviation, or as percentage; Legend: BMI: Body Mass Index; CIRS: Cumulative Illness Rating Scale; MNA: Mini Nutritional Assessment; MNA-SF: Mini Nutritional Assessment – Short Form; Self-MNA: self Mini Nutritional Assessment.
8.9% of men and 8.8% of women had a high comorbidity index (> 4), and a high severity index (>1.5) was observed in 49% of men and 35.8% of women.
No gender differences emerged in MNA classification, in BMI classes and in the level of comorbidity defined by the CIRS.
Nutritional Risk: Agreement between Self-MNA and MNASF or Full-MNA Agreement between Self-MNA and Full- MNA, and between Self-MNA and MNA-SF was classified as “moderate (k = 0.476 and 0.496 respectively; p < 0.001, Table 2.
Table 2.
Agreement between Self-MNA test with Full-MNA, and between Self-MNA and MNA-SF
| Self- MNA (2 levels of risk) | |||
|---|---|---|---|
| Increased or moderate risk of malnutrition (n) | Normal nutritional status (n) | ||
| Full-MNA (2 levels of risk) Malnutrition and at risk of malnutrition (n) | 56 | 23 | |
| Normal nutritional status (n) | 27 | 108 | |
| Overall predictive value (%) | 76.6 | ||
| Sensitivity (%) | 70.9 | ||
| Specificity (%) | 80 | ||
| Positive predictive value (%) | 67.5 | ||
| Negative predictive value (%) | 82.4 | ||
| MNA-SF (2 levels of risk) Malnutrition and at risk of malnutrition (n) | 49 | 16 | |
| Normal nutritional status (n) | 27 | 122 | |
| Overall predictive value (%) | 79.9 | ||
| Sensitivity (%) | 75.4 | ||
| Specificity (%) | 81.9 | ||
| Positive predictive value (%) | 64.5 | ||
| Negative predictive value (%) | 88.4 | ||
For the evaluation of predictive value of Self-MNA versus Full-MNA and MNA-SF, nutritional categorization was estimated considering only two levels of risk. Subjects with high or moderate risk of malnutrition were aggregated in a single level of risk; 12 subjects did not complete the Self-MNA; Legend. MNA: Mini Nutritional Assessment; MNA-SF: Mini Nutritional Assessment – Short Form; Self-MNA: Mini Nutritional Assessment- self-administered.
The Self-MNA showed a fair predictive value compared to the Full-MNA and MNA-SF tests (76.6 and 79.9%, respectively) with a barely adequate sensitivity (70.9 and 75.4%, respectively) (Figure 2 and 3).
Figure 2.
Agreement between Self-MNA and Full-MNA
Figure 3.
Agreement between Self-MNA and MNA-SF
The analysis of the single items showed that the evaluations of GPs and patients were not always consistent. For example, no decline in appetite was reported by 8.2% subjects in whom the GPs found a severe or moderate decrease in food intake; no weight loss was declared by 15.9% of subjects in whom, however, it was reported by the GPs; no dementia or sadness were reported by 13.5% of subjects inwhom cognitive impairment or depression had been detected by the GPs. Finally, 10.1% of subjects classified their BMI class at a higher level compared to GPs (Table 3).
Table 3.
Agreement between the evaluation of patients and GPs considering the single items of Self-MNA and MNA-SF
| Item of MNA-SF or Self-MNA | Agreement(%) | Patients reporting a better condition than GPs(%) | Patients reporting a worst condition than GPs (%) |
|---|---|---|---|
| A. Has your food intake declined over the past 3 months? 0 = severe decrease in food intake 1 = moderate decrease in food intake 2 = no decrease in food intake | 88.4 | 8.2 | 3.4 |
| B. How much weight have you lost in the past 3 months? 0 = weight loss greater than 3 kg 1 = do not know 2 = weight loss between 1 and 3 kg 3 = no weight loss or weight loss < 1 kg | 78.7 | 15.9 | 5.4 |
| C. How would you describe your current mobility? 0 = unable to get out of a bed, a chair, or a wheelchair without the assistance of another person 1 = able to get out of bed or a chair, but unable to go out of my home 2 = able to leave my home | 94.2 | 2.4 | 3.4 |
| D. Have you been stressed or severely ill in the past 3 months? 0 = yes 2 = no | 86.5 | 5.3 | 8.2 |
| E. Are you currently experiencing dementia and/or prolonged severe sadness? 0 = yes, severe dementia and/or prolonged severe sadness | 80.7 | 13.5 | 5.8 |
| 1 = yes, mild dementia, but no prolonged severe sadness 2 = neither dementia nor prolonged severe sadness F. BMI 0 = BMI less than 19 kg/m2 1 = BMI 19 to less than 21 2 = BMI 21 to less than 23 3 = BMI 23 or greater | 77.9 | 10.1 | 12 |
Data are presented as percentage; Legend: GP: General Practitioner; BMI: Body Mass Index; MNA-SF: Mini Nutritional Assessment – Short Form; Self-MNA: Mini Nutritional Assessment- self-administered.
The analysis of the characteristics of false negative (FN) participants (FN subjects: considered at risk of malnutrition or malnourished at the Full-MNA but not at the Self-MNA) showed that the clinical and functional aspects of these subjects (age, comorbidity and severity of illness, time necessary to complete the Self-MNA, decrease in food intake, severe illness in the past 3 months, dementia and depression, fluid intake, need for feeding assistance, arm and calf circumference) were very similar to the characteristics of TP subjects (Table 4).
Table 4.
Clinical/ functional characteristics and nutritional status comparing FP and FN to TP and TN at Self-MNA
| TP | TN | FP | FN | p between groups | p between TP and FN | ||
|---|---|---|---|---|---|---|---|
| N | 49 | 122 | 27 | 16 | |||
| Age | Years | 78.4±8 | 73.4±7 | 75.2±8 | 76.2±9 | 0.002 | 0.32 |
| > 75 years (%) | 66 | 36.8 | 37 | 46.7 | 0.007 | 0.11 | |
| BMI | kg/m2 | 26.2±5 | 26.6±4 | 27.1±5 | 25.4±3 | 0.72 | 0.87 |
| CIRS scale | Comorbity index | 2.38±2 | 1.3±1.4 | 1.62±1.6 | 2.8±2.1 | < 0.001 | 0.39 |
| High Comorbidity index (>4) (%) | 14.6 | 4.2 | 7.4 | 20 | 0.05 | 0.53 | |
| Severity index | 1.74±0.5 | 1.36±0.3 | 1.46±0.4 | 1.55±0.5 | < 0.001 | 0.31 | |
| High Severity index (> 1.5) (%) | 58.3 | 30.5 | 44.4 | 73.3 | < 0.001 | 0.38 | |
| Difficulty in completing the Self-MNA > 6 | Evaluated by the GP (%) | 27.7 | 6.4 | 22.2 | 21.4 | < 0.001 | 0.6 |
| Self assessed by the patient (%) | 25.5 | 5.4 | 22.2 | 21.4 | < 0.001 | 0.17 | |
| Time necessary to complete the Self-MNA (mn) | 8.22±5.5 | 5.88±4.1 | 6.23±3.4 | 7.6±4.9 | 0.406* | 0.84* | |
| Full-MNA items | |||||||
| Severe or moderate decrease in food intake (%) | 67.6 | 4.2 | 0 | 40 | < 0.001 | 0.19 | |
| Weight loss > 1 kg in the past 3 months (%) | 75 | 16.9 | 11.1 | 20 | < 0.001 | 0.003 | |
| Disability (unable to get out of a bed, a chair, or a wheel-chair without the assistance of another person or able to get out of bed or a chair, but unable to go out of my home) (%) | 34.3 | 4.2 | 11.1 | 13.3 | < 0.001 | 0.12 | |
| Severe illness in the past 3 months (%) | 75 | 2.5 | 14.8 | 86.7 | < 0.001 | 0.63 | |
| Presence of dementia or prolonged severe sadness (%) | 64.6 | 9.3 | 14.8 | 46.7 | < 0.001 | 0.78 | |
| BMI less than 21 kg/m2 (%) | 85.3 | 97.5 | 92.6 | 86.7 | 0.02 | 0.6 | |
| Lives independently (not in nursing home or hospital) (%) | 8.3 | 2.2 | 19 | 13.3 | 0.03 | 0.6 | |
| Takes more than 3 prescription drugs per day (%) | 66.7 | 51.6 | 66.7 | 80 | 0.09 | 0.6 | |
| Pressure sores or skin ulcers (%) | 6.3 | 4.4 | 4.8 | 6.7 | 0.96 | 0.99 | |
| ≤ 2 full meals eaten by the patient daily (%) | 22.9 | 27.5 | 14.3 | 26.7 | 0.85 | 0.94 | |
| Reduced consumption of markers for protein intake (dairy products, eggs, legumes, meat, fish or poultry) (%) | 42.1 | 46.1 | 57.2 | 46.7 | 0.89 | 0.71 | |
| ≤ 2 servings of fruit or vegetables per day (%) | 23 | 8.8 | 14.3 | 6.7 | 0.31 | 0.19 | |
| ≤ 5 cups of fluid (water, juice, coffee, tea, milk …) consumed per day (%) | 64.6 | 53.8 | 42.9 | 66.6 | 0.38 | 0.81 | |
| Unable to eat without assistance or self-fed with some difficulty (%) | 22.9 | 2.2 | 4.8 | 20 | 0.005 | 0.83 | |
| Self view of nutritional status as being malnourished or uncertain of nutritional state (%) | 41.7 | 6.6 | 4.8 | 20 | < 0.001 | 0.36 | |
| Self view of health status in comparison with other people of the same age as “not as good (or do not know) (%) | 64.6 | 15.4 | 9.6 | 33.3 | < 0.001 | 0.08 | |
| Mid-arm circumference < 22 cm | 27.1 | 7.7 | 14.3 | 40 | 0.001 | 0,47 | |
| Calf circumference < 31 cm | 39.6 | 8.8 | 9.5 | 26.7 | < 0.001 | 0.32 | |
Legend: BMI: Body Mass Index; CIRS: Cumulative Illness Rating Scale; MNA: Mini Nutritional Assessment; Self-MNA: Mini Nutritional Assessment- self- administered; Data are presented as mean ± standard deviation, or as percentage; TP: true positive; TN: true negative; FP: false positive; FN: false negative were defined trough the comparison between Full-MNA (considered as the reference tool) and Self-MNA; * After adjustment for age, comorbidity and severity.
Difficulties in the self-administration of the Self-MNA
As reported by the GPs, the study participants required 6.68±4.50 minutes to complete the test [25 subjects (11.7%) needed more than 10 minutes, up to a maximum of 30 minutes, to complete the test]. Participants reported a difficulty of 2.64 ± 3.10 points on a scale from 0 (no difficulty) to 10 (maximum difficulty). This value was similar to that recorded by the GPs (2.85 ±3.10; p= 0.16). Patients who stated a greater difficulty (>6 points on the scale from 0 to 10) compared to those who claimed to have found a minimum difficulty in completing the test (0-3 points on the scale from 0 to 10 points) were older (79.8 ± 7 vs. 73.5 ± 7 years; p< 0.001) and more frequently were aged over-75 years (77.8 vs 33.8 %; p< 0.001). These subjects were more «malnourished» at the Full-MNA (10.7 vs. 1.7 %; p= 0.006) and their clinical status was characterized by higher illness severity index (1.72 ± 0.6 vs. 1.41 ± 0.4; p= 0.008) (Table 5).
Table 5.
Clinical/ functional characteristics and nutritional status according to the degree of difficulty experienced by the patients in completing the Self-MNA
| Difficulty in | completing the Self-MNA | 0-3 | 4-6 | >6 | p between groups | P between groups A and C |
|---|---|---|---|---|---|---|
| N | 151 | 35 | 28 | |||
| Gender | Male (%) | 48.3 | 44.4 | 37 | 0.55 | 0.28 |
| Age | Years | 73.5±7 | 77.6±8 | 79.8±7 | <0.001 | <0.001 |
| > 75 years (%) | 33.8 | 61.1 | 77.8 | <0.001 | <0.001 | |
| BMI | kg/m2 | 26.3±4 | 27.1±5 | 27.3±4 | 0.25 | 0.1 |
| CIRS scale | Comorbity index | 1.54±1.6 | 2.2±2 | 2.21±2 | 0.06 | 0.08 |
| High Comorbidity index (>4) (%) | 4.2 | 17.1 | 14.3 | 0.01 | 0.04 | |
| Severity index | 1.41±0.4 | 1.6±0.4 | 1.72±0.6 | 0.006 | 0.008 | |
| High Severity index (> 1.5) (%) | 35 | 54.3 | 60.7 | 0.01 | 0.01 | |
| Full-MNA | Score | 25.3±3 | 22.4±5 | 22±5 | 0.02* | 0.03* |
| Malnourished (score < 17) (%) | 1.7 | 16.1 | 10.7 | 0.002 | 0.006 | |
| At risk of malnutrition (score 17-23.5) (%) | 27.8 | 38.7 | 46.4 |
Data are presented as mean ± standard deviation, or as percentage; 12 subjects did not complete the Self-MNA;
After adjustment for age, comorbidity and severity; Legend: BMI: Body Mass Index; CIRS: Cumulative Illness Rating Scale; MNA: Mini Nutritional Assessment; Self-MNA: Mini Nutritional Assessment- self-administered.
Twelve subjects (5.3%) were not able to complete the Self- MNA. These subjects were more frequently women (88.9 vs. 52.7 %; p=0.03), older (80.6 ± 8 vs. 74.9 ± 8 years; p=0.04) and had a higher level of comorbidity (3.44 ± 2.6 vs. 1.71 ± 1.7; p=0.01) (Table 6). The assessment made by the GPs through the Full-MNA indicated that these subjects had a decline in food intake ( 66.7% vs. 20.2% of those who completed the test; p= 0.02) and/ or a significant weight loss in the last 3 months (55.6% vs. 29.8%; p= 0.02).
Table 6.
Clinical and functional characteristics and nutritional status of non-completers of the Self-MNA compared to completers
| Non- completers | Completers | p | ||
|---|---|---|---|---|
| N | 12 | 214 | ||
| Gender | Male (%) | 11.1 | 47.3 | 0.03 |
| Age | Years | 80.6±8 | 74.9±8 | 0.04 |
| > 75 years (%) | 66.7 | 44.2 | 0.18 | |
| BMI | kg/m2 | 24.1±4 | 26.5±4 | 0.2 |
| CIRS scale | Comorbity index | 3.44±2.6 | 1.71±1.7 | 0.01 |
| High Comorbidity index (>4) (%) | 44.4 | 7.7 | < 0.001 | |
| Severity index | 1.5±0.2 | 1.48±0.4 | 0.18 | |
| High Severity index (> 1.5) (%) | 66.7 | 41.3 | 0.13 | |
| Full- MNA | Score | 22.1±5 | 24.3±4 | 0.94* |
| Malnourished (score < 17) (%) | 22.2 | 5.7 | 0.13 | |
| Patients at risk of malnutrition (score 17-23.5) (%) | 33.3 | 32 | ||
| Normal nutritional status (score ≥ 24) (%) | 44.4 | 62.3 |
Data are presented as mean ± standard deviation, or as percentage; * After adjustment for age and comorbidity; Legend: BMI: Body Mass Index; CIRS: Cumulative Illness Rating Scale; Self-MNA: Mini Nutritional Assessment- self-administered.
Discussion
In the present study we examined the validity of the Self- MNA in free-living elderly subjects. The results obtained confirm the validity of the test that may represent a useful tool for the GPs although some important limitations to the test need to be considered, limiting its use in clinical practice.
To the best of our knowledge, only two papers concerning the Self-MNA were published with discordant results.
In the study by Huhmann et al. (29) the Authors proposed for the first time the Self-MNA and administered it in 463 subject/ caregiver pairs (298 community- dwelling older adults and 165 caregivers). In this study population, age was 76.8 ± 6.8 years (60% women). Additionally, 102 healthcare professionals (HCP) participated in the study. HCP-administered MNA-SF was considered the gold standard. The agreement between the Self-MNA and the MNA-SF was 99% (Self-MNA sensitivity: 99%, specificity: 98%, false negative rate: 1%, false positive rate: 2%) for identifying malnourished subjects and 83% (sensitivity: 89%, specificity: 77%, false negative rate: 11%, false positive rate: 23%) for identifying patients at risk of malnutrition compared to the MNA-SF administered by a HCP.
The aim of the second study conducted by Moore et al. (35) was to compare results of the Self-MNA to those of the MNASF among a sample of homebound older adults who were newly enrolled in a Meals on Wheels program. Results suggested that the Self-MNA screening tool was not practical for this sample, as evidenced by low rates of completion. Fewer than 80% completed the front of the form on the first attempt, and only 40% recorded a final screening score on the back of the form. The Authors' conclusion was that to increase usability of the Self-MNA among older adults, it may be necessary to modify instructions and formatting to improve clarity.
In our study we verified the applicability of the test in a sample of free-living elderly subjects. The test was completed by more than 95% of the enrolled subjects. Nevertheless a substantial proportion of patients (approximately 30%), even among those who completed the test, declared to have had difficulty, taking up to 30 minutes to complete the test. It has to be considered that these subjects, as well as those who were not able to complete the test, were older, with more comorbidities and a higher risk of malnutrition. The simplicity of administration for a screening test is a prerequisite, and this is true especially for self-administered measures. In our study, we observed that patients who had difficulty in filling out the Self-MNA were also those who had a greater need for a nutritional support according to the scores obtained at the Full- MNA administered by clinicians which represents an important limitation to the tool.
In our study the analysis of the single items showed that in some cases the evaluations of GPs and patients were not consistent. In these cases patients usually declared a more favourable clinical condition (no decline in appetite, no weight loss, no dementia or sadness) while these symptoms were detected by the GPs. Under- or mis-reporting especiallyof food intake and anthropometric data is very frequent in nutritional surveys not only in elderly subjects (36, 37) and may represent a reason for caution in the analysis of the data collected through the Self-MNA.
Both in the study by Huhmann et al. and in that one by Moore et al. (29, 35), the Authors used the MNA-SF as a gold standard tool. In our study, we used the MNA-SF as a test to validate the Self-MNA but we adopted the Full-MNA for the global definition of nutritional risk. We believe that this approach allows a better assessment of the validity of the test, and represents a point of strength of our study. In fact, the Full-MNA takes into account some aspects that the MNA-SF does not consider (ability to eat without assistance, protein intake, fluid intake, mid-arm and calf circumferences) and prevents a certain amount of misclassification that is «implicit» in the MNA-SF as in any other screening test (31). Misclassification into a lower nutritional risk category (false negative) may be potentially harmful to the patient because of a missing nutritional intervention when it would be necessary and appropriate.
Execution time by healthcare professionals represents one of the criteria for choosing a nutritional evaluation tool (38- 40). In particular, the Full-MNA takes 10 to 15 minutes while 3-5 minutes are necessary to administer the MNA-SF (30, 41, 42, 43). In our study the time needed to complete the Self-MNA was relatively high (around 7 minutes) although some people required a much longer time ([24 subjects (11.5%) needed more than 10 minutes, up to a maximum 30 minutes to complete the test]. This may make the test less acceptable in poorly motivated subjects and in patients with a deteriorated clinical and functional condition.
It should be noted that the Self-MNA is designed to estimate the risk of undernutrition. Actually both over- and undernutrition are highly prevalent in hospitalized patients and are both well-known risk factors for increased morbidity and mortality. In the present study 16.8% of men and 16.9% of women were obese (BMI≥ 30 kg/m2). The presence of obesity is underestimated in the elderly, and especially the coexistence of over- and under-nutrition is not usually recognized. In a previous study we developed and validated a screening tool for the easy detection and reporting of both under- and over-nutrition overnutrition, specifically identifying the clinical conditions where the two types of malnutrition coexist (44, 45, 46). A short form of this tool will probably be useful.
Our study has some limitations. We did not have information regarding the specific diagnoses. Weanalyzed the clinical status by a comorbidity index considering 13 relatively independent areas grouped under body systems. This approach seems to be more suitable to the geriatric population instead of the analysis and the staging of individual diseases. Moreover, we enrolled patients who spontaneously referred to the office of the GPs during the observation period. This may represent a selection bias because those who do not refer to the GPs' office may have different clinical and nutritional characteristics. The validity of Self-MNA should therefore be checked by enlarging the survey to all free-living subjects including those who for different reasons do not frequently walk in the GP's office.
In conclusion, the Self-MNA may be considered a useful tool for assessing the nutritional risk in the elderly although some important limitations to the test need to be considered, limiting its use in clinical practice. The results need to be shared with the GPs who will eventually perform a Full-MNA and/ or a more in-depth evaluation of nutritional status. The fact that the Self-MNA can be self-administered is definitely an advantage. It should however be considered that an important proportion of subjects may have difficulties in completing it. In these subjects, often more at risk for the presence of malnutrition, other instruments (MNA-SF and Full-MNA) directly administered by the GPs become essential. Finally it has to be considered that the Self-MNA is aimed at the recognition of the presence of a risk of under-nutrition while, as noted in our sample, the prevalence of obesity in the elderly is growing significantly.
Acknowledgements
Special thanks to the members of the Research Group of the SIMPeSV, Albano Vito, Auriemma Lucia, Bariletto Giuseppe, Bellani Francesco, Bruschelli Carla, Cadeddu Maria Grazia Rita, Cogorno Barbara, Di Modica Rosario Salvatore, Gori Fernando, Livadiotti Daniela. Marini Giulia, Marolla Matteo, Mastroianni Serafina, Mocerino Maria Felicia, Monaco Antonella, Morbiato Francesco Filippo, Omenetti Sauro, Panzera Tiziana, Pelini Elisabetta, Petruzzi Sara, Pirro Maurizio, Ricagni Italo Guido, Roscilli Franco, Siviero Guido Alberto, Vaccaro Francesco, Pietro Scalera, Giuseppe Nicodemo Bombardiere for test administration and data collection.
Conflict of interest
The authors declare they have no conflict of interest
Ethical standard
Oral and written informed consent was obtained from all participants.
References
- 1.Cereda E, Pedrolli C, Klersy C, Bonardi C, Quarleri L, Cappello S, Turri A, Rondanelli M, Caccialanza R, Nutritional status in older persons according to healthcare setting: A systematic review and meta-analysis of prevalence data using MNA(®), Clin Nutr., 2016 [DOI] [PubMed]
- 2.Schrader E, Grosch E, Bertsch T, Sieber CC, Volkert D. Nutritional and Functional Status in Geriatric Day Hospital Patients-MNA Short Form Versus Full MNA. J Nutr Health Aging. 2016;20:918–926. doi: 10.1007/s12603-016-0691-4. 10.1007/s12603-016-0691-4 PubMed PMID: 27791222. [DOI] [PubMed] [Google Scholar]
- 3.Vellas B, Lauque S, Andrieu S, Nourhashemi F, Rolland Y, Baumgartner R, Garry P. Nutrition assessment in the elderly. Curr Opin Clin Nutr Metab Care. 2001;4(1):5–8. doi: 10.1097/00075197-200101000-00002. 10.1097/00075197-200101000-00002 PubMed PMID: 11122552. [DOI] [PubMed] [Google Scholar]
- 4.de van der Schueren MA, Wijnhoven HA, Kruizenga HM, Visser M. A critical appraisal of nutritional intervention studies in malnourished, community dwelling older persons. Clin Nutr. 2015;pii:S0261–5614. doi: 10.1016/j.clnu.2015.12.013. [DOI] [PubMed] [Google Scholar]
- 5.Kaiser MJ, Bauer JM, Ramsch C, et al. Frequency of malnutrition in older adults: a multinational perspective using the mini nutritional assessment. J Am Geriatr Soc. 2010;58:1734–1738. doi: 10.1111/j.1532-5415.2010.03016.x. 10.1111/j.1532-5415.2010.03016.x PubMed PMID: 20863332. [DOI] [PubMed] [Google Scholar]
- 6.Donini LM, Poggiogalle E, Piredda M, Pinto A, Barbagallo M, Cucinotta D, Sergi G. Anorexia and eating patterns in the elderly. PLoS One. 2013; 8: e63539. [DOI] [PMC free article] [PubMed]
- 7.Lattanzio F, Laino I, Pedone C, Corica F, Maltese G, Salerno G, Garasto S, Corsonello A, Incalzi RA, on behalf of the Pharmaco surVeillance in the elderly Care PVC Study Group Geriatric Conditions and Adverse Drug Reactions in Elderly Hospitalized Patients. J Am Med Dir Assoc. 2012;13:96–99. doi: 10.1016/j.jamda.2011.04.006. 10.1016/j.jamda.2011.04.006 PubMed PMID: 21621472. [DOI] [PubMed] [Google Scholar]
- 8.Torheim LE, Ferguson EL, Penrose K, Arimond M. Women in resource-poor settings are at risk of inadequate intakes of multiple micronutrients. J Nutr. 2010;140:2051S–2058S. doi: 10.3945/jn.110.123463. 10.3945/jn.110.123463 PubMed PMID: 20881075. [DOI] [PubMed] [Google Scholar]
- 9.Hassen TA, Pearson S, Cowled PA, Fitridge RA. Preoperative nutritional status predicts the severity of the Systemic Inflammatory Response Syndrome (SIRS) following major vascular surgery. Eur J Vasc Endovasc Surg. 2007;33:696–702. doi: 10.1016/j.ejvs.2006.12.006. 10.1016/j.ejvs.2006.12.006 PubMed PMID: 17276097. [DOI] [PubMed] [Google Scholar]
- 10.Schneider SM, Veyres P, Pivot X, Soummer AM, Jambou P. Malnutrition is an independent factor associated with nosocomial infections. Br J Nutr. 2004;92:105–111. doi: 10.1079/BJN20041152. 10.1079/BJN20041152 PubMed PMID: 15230993. [DOI] [PubMed] [Google Scholar]
- 11.Donini LM, De Felice MR, Tagliaccica A, De Bernardini L C Cannella. Nutritional status and evolution of pressure sores in geriatric patients. J Nutr Health Aging. 2005;9(6):446–454. PubMed PMID: 16395517. [PubMed] [Google Scholar]
- 12.Donini LM, De Felice MR, Cannella C. Nutritional status determinants and cognition in the elderly. Arch Gerontol Geriatr. 2007;1:143–153. doi: 10.1016/j.archger.2007.01.022. 10.1016/j.archger.2007.01.022 [DOI] [PubMed] [Google Scholar]
- 13.Grieger JA, Nowson CA, Ackland LM. Nutritional and functional status indicators in residents of a long-term care facility. J Nutr Elder. 2009;28:47–60. doi: 10.1080/01639360802633979. 10.1080/01639360802633979 PubMed PMID: 19234994. [DOI] [PubMed] [Google Scholar]
- 14.Friedmann JM, Jensen GM, Smicklas-Wright J, McCamish MA. Initial malnutrition predicting early non-elective hospital readmission in nutritionally compromised older adults. Am J Clin Nutr. 1997;65:1714–1720. doi: 10.1093/ajcn/65.6.1714. 10.1093/ajcn/65.6.1714 PubMed PMID: 9174465. [DOI] [PubMed] [Google Scholar]
- 15.Muscaritoli M, Molfino A. Malnutrition: the hidden killer in healthcare systems. BMJ. 2013;346:f1547. doi: 10.1136/bmj.f1547. 10.1136/bmj.f1547 PubMed PMID: 23485639. [DOI] [PubMed] [Google Scholar]
- 16.van Wissen J, van Stijn MF, Doodeman HJ, Houdijk AP. Mini Nutritional Assessment and Mortality after Hip Fracture Surgery in the Elderly. J Nutr Health Aging. 2016;20(9):964–968. doi: 10.1007/s12603-015-0630-9. 10.1007/s12603-015-0630-9 PubMed PMID: 27791228. [DOI] [PubMed] [Google Scholar]
- 17.Elia M, Stratton RJ. Considerations for screening tool selection and role of predictive and concurrent validity. Curr Opin Clin Nutr Metab Care. 2011;14:425–433. doi: 10.1097/MCO.0b013e328348ef51. 10.1097/MCO.0b013e328348ef51 PubMed PMID: 21832898. [DOI] [PubMed] [Google Scholar]
- 18.Donini LM, De Bernardini L, De Felice MR, Savina C, Coletti C. Effect of nutritional status on clinical outcome in a population of geriatric rehabilitation patients. Aging Clin Exper Res. 2004;16:132–138. doi: 10.1007/BF03324542. 10.1007/BF03324542 [DOI] [PubMed] [Google Scholar]
- 19.Pirlich M, Schutz T, Norman K, Gastell S, Lubke HJ. The German hospital malnutrition study. Clin Nutr. 2006;25:563–572. doi: 10.1016/j.clnu.2006.03.005. 10.1016/j.clnu.2006.03.005 PubMed PMID: 16698132. [DOI] [PubMed] [Google Scholar]
- 20.Stratton RJ, Elia M. Deprivation linked to malnutrition risk and mortality in hospital. Br J Nutr. 2006;96(5):870–876. doi: 10.1017/bjn20061852. 10.1017/BJN20061852 PubMed PMID: 17092375. [DOI] [PubMed] [Google Scholar]
- 21.Kubrak C, Jensen L. Malnutrition in acute care patients: a narrative review. Int J Nurs Studies. 2007;44:1036–1054. doi: 10.1016/j.ijnurstu.2006.07.015. 10.1016/j.ijnurstu.2006.07.015 [DOI] [PubMed] [Google Scholar]
- 22.Hassan A, Ramadan M, Abdel-Rahman MO, Kerkadi A. Prevalence of undernutrition in a long-term care facility in Qatar. J Nutr Elder. 2009;28(3):313–322. doi: 10.1080/01639360903140353. 10.1080/01639360903140353 PubMed PMID: 21184373. [DOI] [PubMed] [Google Scholar]
- 23.Cereda E, Pedrolli C, Zagami A, Vanotti A, Piffer S. Body mass index and mortality in institutionalized elderly. J Am Med Dir Assoc. 2001;12(3):174–178. doi: 10.1016/j.jamda.2010.11.013. 10.1016/j.jamda.2010.11.013 [DOI] [PubMed] [Google Scholar]
- 24.Guigoz Y, Vellas B, Garry PJ. Mini nutritional assessment: a practical assessment tool for grading the nutritional status of elderly patients. Facts Res Gerontol. 1994;2:15–59. [Google Scholar]
- 25.Cederholm T, Bosaeus I, Barazzoni R, Bauer J, van Gossum A, Klek S, Muscaritoli M, Nyulasi I, Ockenga J, Schneider SM, de van der Schueren MA, Singer P. Diagnostic criteria for malnutrition -An ESPEN Consensus Statement. Clin Nutr. 2015;34:335–340. doi: 10.1016/j.clnu.2015.03.001. 10.1016/j.clnu.2015.03.001 PubMed PMID: 25799486. [DOI] [PubMed] [Google Scholar]
- 26.Schönherr S, Halfens RJG, Meijers JMM, Schols JMGA, Lohrmann C. Structural and process indicators of nutritional care: A comparison between Austrian hospitals and nursing homes. Nutrition. 2012;28:868–873. doi: 10.1016/j.nut.2011.11.007. 10.1016/j.nut.2011.11.007 PubMed PMID: 22305532. [DOI] [PubMed] [Google Scholar]
- 27.Söderhamn U, Dale B, Sundsli K, Söderhamn O. (2012) Nutritional screening of older home-dwelling Norwegians: a comparison between two instruments. Clin Interv Aging. 2012;7:383–391. doi: 10.2147/CIA.S35986. PubMed PMID: 23049250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hamirudin AH, Charlton K, Walton K, Bonney A, Potter J, Milosavljevic M, Hodgkins A, Albert G, Ghosh A, Dalley A. Feasibility of implementing routine nutritional screening for older adults in Australian general practices: a mixed-methods study. BMC Fam Pract. 2014;15:186. doi: 10.1186/s12875-014-0186-5. 10.1186/s12875-014-0186-5 PubMed PMID: 25421546, PMCID 4247715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Huhmann MB, Perez V, Alexander DD, Thomas DR. A self-completed nutrition screening tool for community-dwelling older adults with high reliability: a comparison study. J Nutr Health Aging. 2013;17:339–344. doi: 10.1007/s12603-013-0015-x. 10.1007/s12603-013-0015-x PubMed PMID: 23538656. [DOI] [PubMed] [Google Scholar]
- 30.Rubenstein LZ, Harker JO, Salva A, et al. Screening for undernutrition in geriatric practice: Developing the short-form Mini-Nutritional Assessment (MNA-SF) J Gerontol A Biol Sci Med Sci. 2001;56A:M366–372. doi: 10.1093/gerona/56.6.m366. 10.1093/gerona/56.6.M366 [DOI] [PubMed] [Google Scholar]
- 31.Kaiser MJ, Bauer JM, Uter W, Donini LM, Stange I, Volkert D, Diekmann R, Drey M, Bollwein J, Tempera S, Guerra A, Ricciardi LM, Sieber CC. Prospective validation of the modified mini nutritional assessment short-forms in the community, nursing home, and rehabilitation setting. J Am Geriatr Soc. 2011;59:2124–2128. doi: 10.1111/j.1532-5415.2011.03659.x. 10.1111/j.1532-5415.2011.03659.x PubMed PMID: 22092071. [DOI] [PubMed] [Google Scholar]
- 32.Parmelee PA, Thuras PD, Katz IR, Lawton MP. Validation of the Cumulative Illness Rating Scale in a geriatric residential population. J Am Geriatr Soc. 1995;43:130–137. doi: 10.1111/j.1532-5415.1995.tb06377.x. 10.1111/j.1532-5415.1995.tb06377.x PubMed PMID: 7836636. [DOI] [PubMed] [Google Scholar]
- 33.Galen R, Gambino S. The Predictive Value and Efficiency of Medical Diagnosis. John Wiley: New York (USA), 1975.
- 34.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174. 10.2307/2529310 PubMed PMID: 843571. [PubMed] [Google Scholar]
- 35.Moore BE, Friedman BJ, Crixell SH. Usability of the Self-Mini Nutritional Assessment among Older Adults Receiving Home-delivered Meals in Texas. Family & Consumer Sci Res J. 2015;44:159–171. 10.1111/fcsr.12133 [Google Scholar]
- 36.Pfrimer K, Vilela M, Resende CM, Scagliusi FB, Marchini JS, Lima NK, Moriguti JC, Ferriolli E. Under-reporting of food intake and body fatness in independent older people: a doubly labelled water study. Age Ageing. 2015;44:103–108. doi: 10.1093/ageing/afu142. 10.1093/ageing/afu142 PubMed PMID: 25341675. [DOI] [PubMed] [Google Scholar]
- 37.Gemming L, Jiang Y, Swinburn B, Utter J, Mhurchu CN. Under-reporting remains a key limitation of self-reported dietary intake: an analysis of the 2008/09 New Zealand Adult Nutrition Survey. Eur J Clin Nutr. 2014;68:259–264. doi: 10.1038/ejcn.2013.242. 10.1038/ejcn.2013.242 PubMed PMID: 24300904. [DOI] [PubMed] [Google Scholar]
- 41.Stratton RJ, Hackston A, Longmore D, Dixon R, Price S, Stroud M, King C, Elia M. Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the ‘Malnutrition Universal Screening Tool (‘MUST’) for adults. Br J Nutr. 2004;92:799–808. doi: 10.1079/bjn20041258. 10.1079/BJN20041258 PubMed PMID: 15533269. [DOI] [PubMed] [Google Scholar]
- 42.Raslan M, Gonzalez MC, Gonçalves Dias MC, Nascimento M, Castro M, Marques P, Segatto S, Torrinhas RS, Cecconello I, Linetzky Waitzberg D: Comparison of nutritional risk screening tools for predicting clinical outcomes in hospitalized patients. Nutrition 2010; 26: 721–726. [DOI] [PubMed]
- 43.Donini LM, Poggiogalle E, Molfino A, Rosano A, Lenzi A, Rossi Fanelli F, Muscaritoli M, Mini-Nutritional Assessment, Malnutrition Universal Screening Tool, and Nutrition Risk Screening Tool for the Nutritional Evaluation of Older Nursing Home Residents, J Am Med Dir Assoc., 2016 [DOI] [PubMed]
- 44.Donini LM, Savina C, Gennaro E, De Felice MR, Rosano A, Pandolfo MM, Del Balzo V, Cannella C, Ritz P, Chumlea WC.: A Systematic Review of the Literature Concerning the Relationship between Obesity and Mortality in the Elderly. J Nutr Health Aging. 2012;16: 89–98. [DOI] [PMC free article] [PubMed]
- 45.Donini LM, Ricciardi LM, Neri B, Lenzi A, Marchesini G. Risk of malnutrition (over and under-nutrition): Validation of the JaNuS screening tool. Clin Nutr. 2014;33:1087–1094. doi: 10.1016/j.clnu.2013.12.001. 10.1016/j.clnu.2013.12.001 PubMed PMID: 24373664. [DOI] [PubMed] [Google Scholar]
- 46.Donini LM, Poggiogalle E, Migliaccio S, Pinto A, Lubrano C, Lenzi A. Sarcopenic Obesity: Correlation with Clinical, Functional, and Psychological Status in a Rehabilitation Setting. Food & Nutr Sci. 2014;5:2020–2031. 10.4236/fns.2014.520213 [Google Scholar]
Uncited references
- 38.Törmä J, Winblad U, Cederholm T, Saletti A. Does undernutrition still prevail among nursing home residents? Clin Nutr. 2013;32:562–568. doi: 10.1016/j.clnu.2012.10.007. 10.1016/j.clnu.2012.10.007 PubMed PMID: 23137706. [DOI] [PubMed] [Google Scholar]
- 39.Donini LM, Savina C, Rosano A, Cannella C. Systematic review of nutritional status evaluation and screening tools in the elderly. J Nutr Health Aging. 2007;11:421–432. PubMed PMID: 17657364. [PubMed] [Google Scholar]
- 40.Beck AM, Gøgsig Christensen A, Stenbæk Hansen B, Damsbo-Svendsen S, Kreinfeldt Skovgaard Møller T, Boll Hansen E, Keiding H. Study protocol: costeffectiveness of multidisciplinary nutritional support for undernutrition in older adults in nursing home and home-care: cluster randomized controlled trial. Nutr J. 2014;13:86. doi: 10.1186/1475-2891-13-86. 10.1186/1475-2891-13-86 PubMed PMID: 25163483, PMCID 4153908. [DOI] [PMC free article] [PubMed] [Google Scholar]