Validation (in Spanish) of the Mini Nutritional Assessment survey to assess the nutritional status of patients over 65 years of age

Validation (in Spanish) of the Mini Nutritional Assessment survey to assess the nutritional... Abstract Aim To validate the Mini Nutritional Assessment (MNA) in a Spanish population over 65 years of age with varying degrees of independence. Design This cross-sectional validation study used the Chang nutritional assessment method as a reference test. Participants 248 subjects (75.4% female), with a mean age of 83, completed the study. They were classified into three groups: (i) autonomous patients who were able to take part in activities outside their home; (ii) patients who require help with daily self-care; (iii) patients living in a residential health care facility. Setting . Three health centres and three residential care homes situated in Cordoba (Spain). Results The kappa values for intra-observer and inter-observer agreement were 0.870 and 0.784, respectively. The intra-class correlation coefficient intra-observer was 0.874 and the inter-observer was 0.789. The sensitivity and specificity readings for the diagnostic accuracy of MNA were 63.2% and 72.9% in the total sample, respectively. The area under the curve was 0.726. For patients in the Group A, B and C, the sensitivity was 89.3%, 60.7% and 18.8%, and the specificity was 23.3%, 56.8% and 94.1%, respectively. Conclusion The results for the reliability of the survey were excellent, and its internal consistency was acceptable. The diagnostic accuracy, as measured by the sensitivity and specificity readings, was lower than that obtained with the original survey. It can therefore be considered more suitable for a population with limited autonomy, and less appropriate for independent patients. The results may not be relevant to patients outside of the Cordova region in Spain. Aging, geriatrics, nutrition, risk assessment, screening Introduction Nutritional status is a determining factor when assessing health status, especially in the elderly, where it has been shown that it often leads to acute and chronic diseases, which are, in turn, linked to high morbidity and mortality (1,2). It is a high priority, therefore, to diagnose and treat complications associated with poor nutritional status. There are several ways of carrying out nutritional assessment, including dietary history, anthropometric measurements, body composition analysis and biochemical measurements (3,4). This methodology may require prior training. Therefore, nutritional surveys are considered an easier and useful evaluation tool in daily clinical practice. The Mini Nutritional Assessment (MNA) survey, in particular, has a high sensitivity and specificity, which allows the risk of malnutrition to be identified early on (5,6). The survey contained 18 questions, divided into four nutritional areas (anthropometric measurements, a dietary questionnaire, a global assessment and a subjective assessment) carried out during the validation process. The final score classifies patients into three groups: satisfactory nutritional status (score over or equal to 24), risk of malnutrition (between 17 and 23.5) and malnutrition (below 17). The original MNA questionnaire was created and validated in a French population in 1994 (7). Since then, a number of studies have translated and validated the questionnaire into several languages. Subsequently, six questions of this questionnaire were selected to develop the short version of the MNA (MNA-SF), used as a screening method (8,9). Despite its frequent use in Spain among the elderly population, the MNA has not yet been fully validated in our language. In our view, the data available for reliability and validity of the MNA in Spanish are scant. Only one adaptation study (10) and one reliability study (11) have been carried out and validity data (the extent to which the results of the MNA test correlate with the diagnosis of malnutrition) are unknown in different types of outpatient population (autonomous patients, patients being looked after at home and those living in a residential health care facility). Several validations in different languages have been carried out from the development of MNA, highlighting the research from Donini et al. (12) in Italy, Ferreira et al. (13) in Brazil, Kuzuya et al. (14) in Japan or Sarikaya et al. (15) in Turkey. The aim of the present study is, therefore, to validate the MNA in a Spanish population over 65 years old with different degrees of independence, and to establish not only the reliability (internal consistency and concordance), but also the criterion validity. Material and methods Study design and settings This cross-sectional study was carried out in three health centres in Cordoba (Spain), two of which were in the city and another in a village and three residential care homes for the elderly. Participants A random sample of 248 patients over 65 were stratified by place of residence and degree of dependence into three groups: autonomous patients who were able to take part in activities outside their home (Group A), patients who require help with daily self-care (not autonomous; Group B), patients living in a residential health care facility (institutionalized; Group C). Inclusion and exclusion criteria The inclusion criteria were as follows: the age of the patients should be over 65 years, to be able to obtain information from the subject and/or their primary care-giver and to have given their informed consent to participate in the study. The exclusion criteria were: the existence of an underlying illness (dementia, stroke, etc.) that constituted a disability and prevented us from obtaining the informed consent or the information required for the study in cases where a responsible family member/caregiver was not available to supply these details. Sample size calculation The epidemiological-statistical package Epidat (version 4.2) (16) was used to calculate the sample size. For an accuracy of 3% and an alpha error of 5%, the necessary sample size was 248 subjects. Based on the resulting losses in a pilot study, it was decided to increase the sample size by 3% to get a final sample of 255 subjects. Study variables and measurements Patients were characterized on sociodemographics, anthropometrics, clinimetric scales, haematological, biochemical variables and therapeutic variables. The variables for the validation and the description of the sample are described below. Sociodemographic: age (years), gender (male/female), education level (illiterate, able to read, primary, secondary and superior), residence (autonomous in health centres, not autonomous in health centres, institutionalized patients). Anthropometric: size (cm), weight (kg), body mass index (BMI; kg/m2), triceps fold (mm), calf circumference (cm), arm circumference (cm) and knee-heel length (cm). Clinimetric scales: cognitive impairment (Pfeiffer) (17), instrumental dependence (Lawton y Brody) (18) and dependence in basic activities of daily life (Barthel) (19), nutritional evaluation (Chang) (20,21). Haematological and biochemical variables: Haemoglobin (g/dl), Nº lymphocytes (103/µl), total proteins (g/dl), albumin (g/dl). Therapeutic variables: Nº illnesses, Nº drugs taken, degree of mobility (bedridden, bed-chair, able to walk with help, independent). The resulting variable produced after completing the MNA survey (5) was considered as ordinal (the total score obtained for each patient) and categorical (being normal/malnourished or the risk thereof). All measurements were taken by specially trained medical and nursing staff to minimize the coefficients of variation, with three measurements for each anthropometric variable and choosing the mean as the final value. The set recommendations were followed when carrying out the anthropometric measurements (22). To carry out the nutritional assessment, the Chang method was used as the reference test. This method is based on anthropometric (percentage of weight loss compared with the ideal, brachial perimeter and triceps fold), biochemical (serum albumin level) and immunological measurements (peripheral blood lymphocyte count), and it produces a nutritional score that allows us to quantify the type and degree of malnutrition suffered by the patient. To perform the statistical analysis, the variables were dichotomized, labelling those with a normal score (equal to or above 24 in the MNA and a score of 1 in the Chang method) as ‘normally nourished’ patients, and those who presented some degree of malnutrition or risk of malnutrition (a score below 24 in the MNA and above 1 in the Chang method) as ‘malnourished’ patients. Statistical analysis Student’s parametric t-test was used to compare the two means, according to the normality of the data. The analysis of variance (ANOVA) was used as a parametric test to compare three or more means, using the Bonferroni method (23) for post hoc contrasts and for non-parametric tests, Kruskal–Wallis (24). A Kolmogorov–Smirnov test (n > 50) was carried out to find the goodness of fit for a normal distribution of the quantitative variables, with the Lilliefors test correction (25), as well as graphic representation tests such as a histogram or QQ (quantil–quantil) and PP (percent–percent) plots. Inter-observer reliability was evaluated through data collected by two different observers who used the MNA independently, and in the same time period, to evaluate 32 subjects. To establish intra-observer reliability, 31 patients were re-evaluated by the same observer conducting the survey again, 15–20 days after the first interview. Both inter-observer and intra-observer reliability were finally established by calculating the Cohen kappa index (26), the intra-class correlation coefficient and the Bland Altman score (27) (represents the pairs of results of the measurements made with the same method or instrument, to check the reproducibility of the test results). Internal consistency was assessed using Cronbach’s alpha coefficient (28). The sensitivity, specificity and likelihood ratios for the positive and negative results, as well as the Youden index (29) and the validity index to validate the criterion or the practice were calculated. A receiver operating characteristic (ROC) curve (30) was used to correlate the score obtained from the MNA survey and the diagnosis of malnutrition measured by the Chang method. These statistical tests were carried out first on the total population and later on each type of sample according to the place of residence and degree of autonomy. The best cut-off point for each type of population was obtained through ROC curves. For all these contrast tests, the statistical significance level was set for a two-tailed alpha error equal to or less than 5% and confidence intervals (CI) were set at 95%. SPSS Version 22 statistical software (IBM Corp, New York, USA) was used to carry out the statistical analysis. Bioethical concerns The research protocol was approved by the Cordoba Research Ethics Committee, Public Andalusian Health Service (Spain). All the patients and/or the main caregiver were informed verbally and in writing of the aims of the study and, after they had agreed to take part, they gave their informed consent in writing, as set down by Law 41/2002 on patient autonomy and their right to decide. The research was carried out in agreement with the Declaration of Helsinki, backed by the World Medical Association for conducting research studies with human beings. Results Of the 255 patients selected, 248 completed the study, with 7 losses due to lack of analytical and/or anthropometric data, one of which died before the study was finalized. Twenty-eight patients (6 living at home and 22 living in residential health care facility) could not answer the questionnaires by themselves, so the information was collected through the main caregiver. Descriptive characteristics of the sample The mean age of the patients was 83, with a higher percentage of females (75.4%). 16.6% came from rural areas, 48.8% were autonomous patients who attended the health centre on their own, 27.4% were living at home but not being autonomous and 23.8% were institutionalized. The average number of chronic diseases detected was 3.9 and the patients took an average of 6.6 different drugs per day. In relation to BMI, a status of overweight was observed in institutionalized (25.9 kg/m2) and not autonomous patients (27.1 kg/m2) while a degree of obesity was presented in autonomous patients (30.9 kg/m2). The commonest pathologies were hypertension (79.4%), arthrosis (46%), hyperlipemia (42.7%) and diabetes (27.8%). From the analytical parameters, the most notable point was the decrease in the number of lymphocytes and albumin in not autonomous and institutionalized patients. As regards their nutritional status, the MNA survey revealed 37.6% malnutrition or risk of malnutrition, while with the Chang method this percentage was 28.7%. According to the place of residence, a higher percentage of cognitive deterioration was observed in Group C (54%). In these, a severe to total degree of dependence for the basic activities of daily life (69.5%) and a moderate to total dependency for instrumental activities (93.3%) was obtained. Group B obtained a 45.6% and 82.3%, respectively. The patient characteristics are described in Table 1. Table 1. Baseline characteristics of 248 patients over 65 years depending on the place of residence and degree of dependence. Data collected between 2014 and 2016   Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3    Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3  BMI, body mass index; MNA, Mini Nutritional Assessment. View Large Table 1. Baseline characteristics of 248 patients over 65 years depending on the place of residence and degree of dependence. Data collected between 2014 and 2016   Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3    Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3  BMI, body mass index; MNA, Mini Nutritional Assessment. View Large Results of reliability, consistency and concordance of the MNA After applying the kappa statistic to assess the intra-observer concordance, a value of 0.870 (95% CI: 0.62–1.12) was obtained, with 0.784 (95% CI: 0.37–1.19) for the inter-observer concordance. The intra-class correlation coefficient was 0.874 (95% CI: 0.76–0.94) for the intra-observer values, and 0.789 (95% CI: 0.61–0.89) for the inter-observer values. Intra- and inter-observer concordance were also evaluated using Bland and Altman plots. The intra-observer measurements, with the exception of one value, follow a normal distribution between the mean and two standard deviations. The inter-observer concordance of the measurements, with the exception of two values, remains between acceptable margins of tolerance. As for the consistency and reliability of the MNA survey, a value of 0.778 was obtained with Cronbach’s alpha for a total of 18 elements (each of the questions that make up MNA), in 218 valid cases (87.9%). Deleting four items from the survey improved the results and gave a value of 0.810. The deleted items were (in order): takes more than three prescription drugs per day (0.792), weight loss during the last 3 months (0.794), selected consumption markers for protein intake (0.807) and how many full meals does the patient eat daily? (0.810). Diagnostic accuracy of the MNA survey according to the Chang method A sensitivity value of 63.2% (95% CI: 51–75.4%) and a specificity value of 72.9% (95% CI: 65.8–80%) were obtained in the total sample. When the analysis by place of residence was carried out, important differences were obtained. The sensitivity for Group A was 18.8% (95% CI: −3.5–41%) and the specificity 94.1% (95% CI: 88.95–99.1%); in Group B was 60.7% (95% CI: 40.8–80.6%) and 56.8% (95% CI: 39.4–74.1%) and in Group C, it was 89.3% (95% CI: 76–102.5%) and 23.3% (95% CI: 6.5–40.1%) respectively. The area under the ROC curve and the remaining validity parameters are shown in Tables 2 and 3. Figure 1 shows the graphical representation of the different curves. Table 2. Diagnostic precision of MNA survey comparing the values of the current MNA with a new cut-off point   ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70    ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. aNew cut-off point. View Large Table 2. Diagnostic precision of MNA survey comparing the values of the current MNA with a new cut-off point   ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70    ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. aNew cut-off point. View Large Table 3. Accuracy of the MNA questionnaire according to population subgroup and comparison with alternative cut-off points.   Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69    Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. View Large Table 3. Accuracy of the MNA questionnaire according to population subgroup and comparison with alternative cut-off points.   Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69    Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. View Large Figure 1. View largeDownload slide Representation of the receiver operating characteristic curve: results of the data obtained with Mini Nutritional Assessment in relation to the Chang method, in the entire sample and by subgroups. The diagonal segments are generated in the case of a tie Figure 1. View largeDownload slide Representation of the receiver operating characteristic curve: results of the data obtained with Mini Nutritional Assessment in relation to the Chang method, in the entire sample and by subgroups. The diagonal segments are generated in the case of a tie After analysing the results from the ROC curve, MNA obtained a better cut-off point to determine malnutrition with an MNA value of 25 and a Youden reading of 0.425, thus increasing the sensitivity to 75% and decreasing the specificity to 67.5%. It was observed that for Group A, the best cut-off point was the one already set in the literature (24). Group B obtained a higher Youden index 0.38 when the cut-off point was 25. In group C, a higher Youden index of 0.38 was obtained when the cut-off point was lowered to 21. Discussion The MNA has been established internationally as a valid, practical tool for measuring the nutritional status of the elderly. The main aim of this study was to assess the validity of the Spanish version of the MNA. To do this, a diverse population sample, over 65 years old, with differing degrees of dependence was recruited, to measure the validation data of the survey for use with any type of elderly population. In addition, the diagnostic accuracy in the three study subgroups has been evaluated, finding an acceptable validity in the population groups of not autonomous who lived at home and institutionalized. There was a higher percentage of females (75.4%) in the sample, which is common in a population over 65 years of age, since the males have a lower life expectancy (31). This has also been noted in other similar studies performed in a Spanish population (32–34). The reliability of the survey shows excellent results, both when assessing the correctness in determining the state of malnutrition measured by the kappa statistic and with the concordance in the score obtained. The Bland Altman plot places the results within the tolerance limits. These results are similar to those observed previously in a Spanish population using the same survey (11). In relation to the internal consistency of the survey, an acceptable value (Cronbach’s α = 0.7) was obtained. This could be improved to a better value (Cronbach’s α = 0.8) by deleting four questions from the original survey enhancing from 0.778 to 0.810. The precision obtained when performing the diagnostic tests in the entire sample shows a sensitivity and specificity of 63.2% and 72.9%, respectively, for the cut-off point proposed in the original version, which were well below the 96% and 98% obtained in the study where the survey was designed and validated (5). The fact that a lower sensitivity and specificity was obtained has led us to propose alternative cut-off values to establish the limit between normal nutritional status and the risk of malnutrition. When 25 is set as the cut-off value, the sensitivity increases to 75% and the Youden index improves. When these parameters are analysed by subgroups, a low sensitivity (18.8%) and a very high specificity (94.1%) can be seen in Group A. Therefore, the use of this questionnaire would not be recommended in a population which, although they are over 65, has a high degree of autonomy that favours their state of health and presents a lower risk of having a deficient nutritional status. In Group B, the sensitivity was 60.7% with a specificity of 56.8%. In this group, greater sensitivity (89.3%) is obtained by setting 25 as the cut-off point to establish the risk of malnutrition. This would allow those patients at risk of malnutrition to be detected sooner and follow them more thoroughly. Group C had the highest sensitivity 89.3%, although they had only 23.3% of specificity. In this group, a higher Youden index was obtained when setting the cut-off point at 21, which improved the specificity (63.3%) but lost sensitivity (75%). In this case, we consider the original cut-off point more useful. Very little validation research for the MNA has been carried out in Spanish populations. Tarazona et al. (35) performed a study in a sample with cognitive impairment and obtained a sensitivity of 60% and a specificity of 94.7%. In another study in a hospitalized Cuban population (36), different percentages were obtained depending on the gold standard used, with a maximum specificity and sensitivity of 50% and 95.1%, respectively. Another example of this variability can be seen in the validation in its Portuguese version by Santos et al. (37). Our results show a lower sensitivity and specificity than those found in Guigoz’s original work. This can be justified by the reference test used and related to the characteristics of the sample analysed. In our case, the Chang method was selected because it was considered the most objective, has great advantages of reliability, reproducibility and specificity (38), although there is little literature available to support its use, perhaps because it requires analytical and anthropometric variables and the final calculation involves rather complex calculations, which makes its use in normal clinical practice less practical. The mean age of the population can also be related to these differences (79 years in Guigoz’s paper versus 83 in this research). Guigoz’s original work validated the questionnaire in 150 patients who were healthy, frail or seriously ill (39); the results obtained lead us to consider this questionnaire to be suitable only for the frail elderly population, and we would not recommend its use in autonomous/independent patients, due to its low sensitivity in this group. This low validity, in autonomous/independent patients, can be attributed to the fact that the MNA contains several questions closely related to the functional capacity of the patient (40) and, probably, patients with a nutritional status altered but maintaining their functional capacity cannot be detected with this questionnaire. The MNA is the most used method to assess the nutritional status of the elderly, both in hospital admissions for acute pathology and in residential health care facility. However, there is not a protocol that considers the systematic nutritional assessment for elderly population in the National Health System of Spain. It is necessary to establish a care process in patients older than 65 years in which the nutritional status is evaluated periodically and MNA could be a cost-effective tool. In this way, possible assistance, social or nutritional deficits in this population could be prevented or mitigated Limitations It would be advisable to confirm these results with subsequent studies on a larger sample of well-represented patients from primary care, establishing the differences in the validity of the test according to the degree of dependence. The results may not be relevant to patients outside of the Cordova region in Spain. Declaration Funding: The Validation of the Mini Nutritional Assessment (MNA) in Spanish is part of a project supported by the Andalusian Health Service. File number: AP-0064-2016. Conflict of interest: The authors report no relationships that could be construed as a conflict of interest. References 1. Péter S, Saris WH, Mathers JC, et al.   Nutrient status assessment in individuals and populations for healthy aging-statement from an expert workshop. Nutrients   2015; 7: 10491– 500. Google Scholar CrossRef Search ADS PubMed  2. Van Bokhorst-de van der Schueren MA, Lonterman-Monasch S, de Vries OJ, et al.   Prevalence and determinants for malnutrition in geriatric outpatients. Clin Nutr   2013; 32: 1007– 11. Google Scholar CrossRef Search ADS PubMed  3. Ahmed T, Haboubi N. Assessment and management of nutrition in older people and its importance to health. Clin Interv Aging   2010; 5: 207– 16. Google Scholar PubMed  4. Ravasco P, Anderson H, Mardones F. Red Mel-CYTED. Methods of valuation of the nutritional condition. Nutr Hosp   2010; 25( suppl 3): 57– 66 5. Guigoz Y, Vellas B, Garry PJ. Mini Nutritional Assessment: a practical tool for grading the nutritional state of the elderly patients. Facts Res Gerontol   1994; ( suppl 2): 15– 60. 6. Guigoz Y. The Mini Nutritional Assessment (MNA) review of the literature–What does it tell us? J Nutr Health Aging   2006; 10: 466– 85; discussion 485–7. Google Scholar PubMed  7. Guigoz Y, Vellas B. Test d’évaluation de l’état nutritionnel de la personne âgée: le Mini Nutritional Assessment (MNA). Med Hyg   1995; 53: 1965– 69. 8. Rubenstein LZ, Harker JO, Salvà A, Guigoz Y, Vellas B. Screening for undernutrition in geriatric practice: developing the short-form mini-nutritional assessment (MNA-SF). J Gerontol A Biol Sci Med Sci   2001; 56: M366– 72. Google Scholar CrossRef Search ADS PubMed  9. Kaiser MJ, Bauer JM, Ramsch C, et al.  ; MNA-International Group. Validation of the Mini Nutritional Assessment short-form (MNA-SF): a practical tool for identification of nutritional status. J Nutr Health Aging   2009; 13: 782– 8. Google Scholar CrossRef Search ADS PubMed  10. Salvá A, Bolívar I, Muñoz M, Valvanera S. Un nuevo instrumento para la valoración nutricional en geriatría: el ‘Mini Nutricional Assessment’ (MNA). Rev Gerontol   1996; 6: 319– 28. 11. Bleda MJ, Bolibar I, Parés R, Salvà A. Reliability of the mini nutritional assessment (MNA) in institutionalized elderly people. J Nutr Health Aging   2002; 6: 134– 7. Google Scholar PubMed  12. Donini LM, de Felice MR, Tassi L, et al.   A ‘proportional and objective score’ for the mini nutritional assessment in long-term geriatric care. J Nutr Health Aging   2002; 6: 141– 6. Google Scholar PubMed  13. Ferreira LS, Nascimento LF, Marucci MF. Use of the mini nutritional assessment tool in elderly people from long-term institutions of southeast of Brazil. J Nutr Health Aging   2008; 12: 213– 7. Google Scholar CrossRef Search ADS PubMed  14. Kuzuya M, Kanda S, Koike T, Suzuki Y, Satake S, Iguchi A. Evaluation of Mini-Nutritional Assessment for Japanese frail elderly. Nutrition   2005; 21: 498– 503. Google Scholar CrossRef Search ADS PubMed  15. Sarikaya D, Halil M, Kuyumcu ME, et al.   Mini nutritional assessment test long and short form are valid screening tools in Turkish older adults. Arch Gerontol Geriatr   2015; 61: 56– 60. Google Scholar CrossRef Search ADS PubMed  16. Epidat: programa para análisis epidemiológico de datos. Versión 4.2 . Colombia: Consellería de Sanidade, Xunta de Galicia, España; Organización Panamericana de la salud (OPS-OMS); Universidad CES, 2016. 17. Martínez de la Iglesia J, Dueñas Herrero R, Onís Vilches MC, et al.   Cross-cultural adaptation and validation of Pfeiffer’s test (Short Portable Mental Status Questionnaire [SPMSQ]) to screen cognitive impairment in general population aged 65 or older . Med Clin (Barc)   2001; 117: 129– 34. 18. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist   1969; 9: 179– 86. Google Scholar CrossRef Search ADS PubMed  19. Mahoney FI, Barthel DW. Functional evaluation: the Barthel index. Md State Med J   1965; 14: 61– 5. Google Scholar PubMed  20. Chang RW. Nutritional assessment using a microcomputer. 1. Programme design. Clin Nutr   1984; 3: 67– 73. Google Scholar CrossRef Search ADS PubMed  21. Chang RW, Richardson R. Nutritional assessment using a microcomputer. 2. Programme evaluation. Clin Nutr   1984; 3: 75– 82. Google Scholar CrossRef Search ADS PubMed  22. Stewart A, Marfell-Jones M, Olds T, de Ridder H. International standards for anthropometric assessment. International Society for de Advancement of Kinanthropometry . Lower Hutt, New Zealand: ISAK, 2011. 23. Dunnett CW. A multiple comparisons procedure for comparing several treatments with a control. J Am Stat Assoc   1955; 50: 1096– 121. Google Scholar CrossRef Search ADS   24. Kruskal WH, Wallis WA. Use of ranks in one-criterion variance analysis. J Am Stat Assoc   1952; 47: 583– 621. Google Scholar CrossRef Search ADS   25. Lilliefors H. On the Kolmogorov–Smirnov test for normality with mean and variance unknown. J Am Stat Assoc   1967; 62: 399– 402. Google Scholar CrossRef Search ADS   26. Cohen J. Weighted kappa: nominal scale agreement with provision for scaled disagreement or partial credit. Psychol Bull   1968; 70: 213– 20. Google Scholar CrossRef Search ADS PubMed  27. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet   1986; 1: 307– 10. Google Scholar CrossRef Search ADS PubMed  28. Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika   1951; 16: 297– 334. Google Scholar CrossRef Search ADS   29. Youden WJ. Index for rating diagnostic tests. Cancer   1950; 3: 32– 5. Google Scholar CrossRef Search ADS PubMed  30. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology   1982; 143: 29– 36. Google Scholar CrossRef Search ADS PubMed  31. Abellán García A, Pujol Rodríguez R. Un perfil de las personas mayores en España, 2016. Indicadores estadísticos básicos. Madrid, Informes Envejecimiento en red n º 14. 2016: 3–7. 32. Jiménez Sanz M, Sola Villafranca JM, Pérez Ruiz C, et al.   Study of the nutritional status of elders in Cantabria. Nutr Hosp   2011; 26: 345– 54. Google Scholar PubMed  33. Cuerda C, Álvarez J, Ramos P, et al.   Prevalence of malnutrition in subjects over 65 years of age in the Community of Madrid. The DREAM + 65 Study. Nutr Hosp   2016; 33: 263– 69. Google Scholar CrossRef Search ADS PubMed  34. Cuervo M, Ansorena D, García A, Astiasarán I, Martínez JA. Food consumption analysis in Spanish elderly based upon the mini nutritional assessment test. Ann Nutr Metab   2008; 52: 299– 307. Google Scholar CrossRef Search ADS PubMed  35. Tarazona FJ, Belenguer A, Doménech JR, et al.   Validation of score in MNA scale like nutritional risk factor in institutionalized geriatric patients with moderate and severe cognitive impairment. Nutr Hosp   2009; 24: 724– 31. Google Scholar PubMed  36. Cuyac Lantigua M, Santana Porbén S. The Mini Nutritional Assessment of the elderly in the practice of a hospital geriatrics service: inception, validation and operational characteristics. Arch Latinoam Nutr   2007; 57: 255– 65. Google Scholar PubMed  37. Santos Pereira Machado R, Santa Cruz Coelho MA, Peixoto Veras R. Validity of the Portuguese version of the mini nutritional assessment in Brazilian elderly. BMC Geriatrics   2015; 15: 132. Google Scholar CrossRef Search ADS PubMed  38. Linares Guerra ME, Bencomo JF, Santana S, Barreto J, Ruiz ML. Nutritional evaluation of HIV-infected patients by Chang’s method. DST—J bras Doenças Sex Transm   2005; 17: 259– 64. 39. Vellas B, Villars H, Abellan G, et al.   Overview of the MNA—its history and challenges. J Nutr Health Aging   2006; 10: 456– 63. Google Scholar PubMed  40. Muñoz Díaz B, Arenas de Larriva AP, Molina-Recio G, Moreno-Rojas R, Martínez de la Iglesia J; grupo de investigación Nutrianco. Study of the nutritional status of patients over 65 years included in the home care program in an urban population. Aten Primaria   2018; 50: 88– 95. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Family Practice Oxford University Press

Validation (in Spanish) of the Mini Nutritional Assessment survey to assess the nutritional status of patients over 65 years of age

Loading next page...
 
/lp/ou_press/validation-in-spanish-of-the-mini-nutritional-assessment-survey-to-MCgj2mbgfy
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0263-2136
eISSN
1460-2229
D.O.I.
10.1093/fampra/cmy051
Publisher site
See Article on Publisher Site

Abstract

Abstract Aim To validate the Mini Nutritional Assessment (MNA) in a Spanish population over 65 years of age with varying degrees of independence. Design This cross-sectional validation study used the Chang nutritional assessment method as a reference test. Participants 248 subjects (75.4% female), with a mean age of 83, completed the study. They were classified into three groups: (i) autonomous patients who were able to take part in activities outside their home; (ii) patients who require help with daily self-care; (iii) patients living in a residential health care facility. Setting . Three health centres and three residential care homes situated in Cordoba (Spain). Results The kappa values for intra-observer and inter-observer agreement were 0.870 and 0.784, respectively. The intra-class correlation coefficient intra-observer was 0.874 and the inter-observer was 0.789. The sensitivity and specificity readings for the diagnostic accuracy of MNA were 63.2% and 72.9% in the total sample, respectively. The area under the curve was 0.726. For patients in the Group A, B and C, the sensitivity was 89.3%, 60.7% and 18.8%, and the specificity was 23.3%, 56.8% and 94.1%, respectively. Conclusion The results for the reliability of the survey were excellent, and its internal consistency was acceptable. The diagnostic accuracy, as measured by the sensitivity and specificity readings, was lower than that obtained with the original survey. It can therefore be considered more suitable for a population with limited autonomy, and less appropriate for independent patients. The results may not be relevant to patients outside of the Cordova region in Spain. Aging, geriatrics, nutrition, risk assessment, screening Introduction Nutritional status is a determining factor when assessing health status, especially in the elderly, where it has been shown that it often leads to acute and chronic diseases, which are, in turn, linked to high morbidity and mortality (1,2). It is a high priority, therefore, to diagnose and treat complications associated with poor nutritional status. There are several ways of carrying out nutritional assessment, including dietary history, anthropometric measurements, body composition analysis and biochemical measurements (3,4). This methodology may require prior training. Therefore, nutritional surveys are considered an easier and useful evaluation tool in daily clinical practice. The Mini Nutritional Assessment (MNA) survey, in particular, has a high sensitivity and specificity, which allows the risk of malnutrition to be identified early on (5,6). The survey contained 18 questions, divided into four nutritional areas (anthropometric measurements, a dietary questionnaire, a global assessment and a subjective assessment) carried out during the validation process. The final score classifies patients into three groups: satisfactory nutritional status (score over or equal to 24), risk of malnutrition (between 17 and 23.5) and malnutrition (below 17). The original MNA questionnaire was created and validated in a French population in 1994 (7). Since then, a number of studies have translated and validated the questionnaire into several languages. Subsequently, six questions of this questionnaire were selected to develop the short version of the MNA (MNA-SF), used as a screening method (8,9). Despite its frequent use in Spain among the elderly population, the MNA has not yet been fully validated in our language. In our view, the data available for reliability and validity of the MNA in Spanish are scant. Only one adaptation study (10) and one reliability study (11) have been carried out and validity data (the extent to which the results of the MNA test correlate with the diagnosis of malnutrition) are unknown in different types of outpatient population (autonomous patients, patients being looked after at home and those living in a residential health care facility). Several validations in different languages have been carried out from the development of MNA, highlighting the research from Donini et al. (12) in Italy, Ferreira et al. (13) in Brazil, Kuzuya et al. (14) in Japan or Sarikaya et al. (15) in Turkey. The aim of the present study is, therefore, to validate the MNA in a Spanish population over 65 years old with different degrees of independence, and to establish not only the reliability (internal consistency and concordance), but also the criterion validity. Material and methods Study design and settings This cross-sectional study was carried out in three health centres in Cordoba (Spain), two of which were in the city and another in a village and three residential care homes for the elderly. Participants A random sample of 248 patients over 65 were stratified by place of residence and degree of dependence into three groups: autonomous patients who were able to take part in activities outside their home (Group A), patients who require help with daily self-care (not autonomous; Group B), patients living in a residential health care facility (institutionalized; Group C). Inclusion and exclusion criteria The inclusion criteria were as follows: the age of the patients should be over 65 years, to be able to obtain information from the subject and/or their primary care-giver and to have given their informed consent to participate in the study. The exclusion criteria were: the existence of an underlying illness (dementia, stroke, etc.) that constituted a disability and prevented us from obtaining the informed consent or the information required for the study in cases where a responsible family member/caregiver was not available to supply these details. Sample size calculation The epidemiological-statistical package Epidat (version 4.2) (16) was used to calculate the sample size. For an accuracy of 3% and an alpha error of 5%, the necessary sample size was 248 subjects. Based on the resulting losses in a pilot study, it was decided to increase the sample size by 3% to get a final sample of 255 subjects. Study variables and measurements Patients were characterized on sociodemographics, anthropometrics, clinimetric scales, haematological, biochemical variables and therapeutic variables. The variables for the validation and the description of the sample are described below. Sociodemographic: age (years), gender (male/female), education level (illiterate, able to read, primary, secondary and superior), residence (autonomous in health centres, not autonomous in health centres, institutionalized patients). Anthropometric: size (cm), weight (kg), body mass index (BMI; kg/m2), triceps fold (mm), calf circumference (cm), arm circumference (cm) and knee-heel length (cm). Clinimetric scales: cognitive impairment (Pfeiffer) (17), instrumental dependence (Lawton y Brody) (18) and dependence in basic activities of daily life (Barthel) (19), nutritional evaluation (Chang) (20,21). Haematological and biochemical variables: Haemoglobin (g/dl), Nº lymphocytes (103/µl), total proteins (g/dl), albumin (g/dl). Therapeutic variables: Nº illnesses, Nº drugs taken, degree of mobility (bedridden, bed-chair, able to walk with help, independent). The resulting variable produced after completing the MNA survey (5) was considered as ordinal (the total score obtained for each patient) and categorical (being normal/malnourished or the risk thereof). All measurements were taken by specially trained medical and nursing staff to minimize the coefficients of variation, with three measurements for each anthropometric variable and choosing the mean as the final value. The set recommendations were followed when carrying out the anthropometric measurements (22). To carry out the nutritional assessment, the Chang method was used as the reference test. This method is based on anthropometric (percentage of weight loss compared with the ideal, brachial perimeter and triceps fold), biochemical (serum albumin level) and immunological measurements (peripheral blood lymphocyte count), and it produces a nutritional score that allows us to quantify the type and degree of malnutrition suffered by the patient. To perform the statistical analysis, the variables were dichotomized, labelling those with a normal score (equal to or above 24 in the MNA and a score of 1 in the Chang method) as ‘normally nourished’ patients, and those who presented some degree of malnutrition or risk of malnutrition (a score below 24 in the MNA and above 1 in the Chang method) as ‘malnourished’ patients. Statistical analysis Student’s parametric t-test was used to compare the two means, according to the normality of the data. The analysis of variance (ANOVA) was used as a parametric test to compare three or more means, using the Bonferroni method (23) for post hoc contrasts and for non-parametric tests, Kruskal–Wallis (24). A Kolmogorov–Smirnov test (n > 50) was carried out to find the goodness of fit for a normal distribution of the quantitative variables, with the Lilliefors test correction (25), as well as graphic representation tests such as a histogram or QQ (quantil–quantil) and PP (percent–percent) plots. Inter-observer reliability was evaluated through data collected by two different observers who used the MNA independently, and in the same time period, to evaluate 32 subjects. To establish intra-observer reliability, 31 patients were re-evaluated by the same observer conducting the survey again, 15–20 days after the first interview. Both inter-observer and intra-observer reliability were finally established by calculating the Cohen kappa index (26), the intra-class correlation coefficient and the Bland Altman score (27) (represents the pairs of results of the measurements made with the same method or instrument, to check the reproducibility of the test results). Internal consistency was assessed using Cronbach’s alpha coefficient (28). The sensitivity, specificity and likelihood ratios for the positive and negative results, as well as the Youden index (29) and the validity index to validate the criterion or the practice were calculated. A receiver operating characteristic (ROC) curve (30) was used to correlate the score obtained from the MNA survey and the diagnosis of malnutrition measured by the Chang method. These statistical tests were carried out first on the total population and later on each type of sample according to the place of residence and degree of autonomy. The best cut-off point for each type of population was obtained through ROC curves. For all these contrast tests, the statistical significance level was set for a two-tailed alpha error equal to or less than 5% and confidence intervals (CI) were set at 95%. SPSS Version 22 statistical software (IBM Corp, New York, USA) was used to carry out the statistical analysis. Bioethical concerns The research protocol was approved by the Cordoba Research Ethics Committee, Public Andalusian Health Service (Spain). All the patients and/or the main caregiver were informed verbally and in writing of the aims of the study and, after they had agreed to take part, they gave their informed consent in writing, as set down by Law 41/2002 on patient autonomy and their right to decide. The research was carried out in agreement with the Declaration of Helsinki, backed by the World Medical Association for conducting research studies with human beings. Results Of the 255 patients selected, 248 completed the study, with 7 losses due to lack of analytical and/or anthropometric data, one of which died before the study was finalized. Twenty-eight patients (6 living at home and 22 living in residential health care facility) could not answer the questionnaires by themselves, so the information was collected through the main caregiver. Descriptive characteristics of the sample The mean age of the patients was 83, with a higher percentage of females (75.4%). 16.6% came from rural areas, 48.8% were autonomous patients who attended the health centre on their own, 27.4% were living at home but not being autonomous and 23.8% were institutionalized. The average number of chronic diseases detected was 3.9 and the patients took an average of 6.6 different drugs per day. In relation to BMI, a status of overweight was observed in institutionalized (25.9 kg/m2) and not autonomous patients (27.1 kg/m2) while a degree of obesity was presented in autonomous patients (30.9 kg/m2). The commonest pathologies were hypertension (79.4%), arthrosis (46%), hyperlipemia (42.7%) and diabetes (27.8%). From the analytical parameters, the most notable point was the decrease in the number of lymphocytes and albumin in not autonomous and institutionalized patients. As regards their nutritional status, the MNA survey revealed 37.6% malnutrition or risk of malnutrition, while with the Chang method this percentage was 28.7%. According to the place of residence, a higher percentage of cognitive deterioration was observed in Group C (54%). In these, a severe to total degree of dependence for the basic activities of daily life (69.5%) and a moderate to total dependency for instrumental activities (93.3%) was obtained. Group B obtained a 45.6% and 82.3%, respectively. The patient characteristics are described in Table 1. Table 1. Baseline characteristics of 248 patients over 65 years depending on the place of residence and degree of dependence. Data collected between 2014 and 2016   Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3    Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3  BMI, body mass index; MNA, Mini Nutritional Assessment. View Large Table 1. Baseline characteristics of 248 patients over 65 years depending on the place of residence and degree of dependence. Data collected between 2014 and 2016   Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3    Autonomous in health centres (Group A) (n = 121)  Not autonomous in health centres (Group B) (n = 68)  Institutionalized patients (Group C) (n = 59)  Age (years)  75.7 (74.5–76.9)  86.5 (84.7–88.3)  86.7 (84.6–88.7)  Nº illnesses  3.7 (3.4–4)  4 (3.7–4.4)  4.2 (3.7–4.6)  Nº drugs taken  5.9 (5.4–6.4)  7.3 (6.5–8)  6.7 (5.9–7.6)  BMI (kg/m2)  30.9 (30.1–31.6)  27.1 (25.5–28.6)  25.9 (24.2–27.5)  Triceps fold (mm)  16.1 (14.9–17.3)  18.3 (16.4–20.2)  13.9 (12.1–15.6)  Arm circumference (cm)  30.2 (29.5–30.7)  27.7 (26.6–28.7)  28.1 (26.7–29.4)  Calf circumference (cm)  36.2 (35.5–36.8)  33 (31.7–34.2)  31.8 (30.3–33.2)  Nº lymphocytes (103/µl)  2.4 (1.9–2.9)  1.8 (1.7–2)  1.7 (1.5–1.8)  Total proteins (g/dl)  6.7 (6.7–6.8)  6.4 (6.2–6.5)  6.5 (6.3–6.6)  Albumin (g/dl)  4.2 (4.1–4.2)  3.7 (3.5–3.7)  3.5 (3.4–3.6)  Gender  Female (n = 179)  61.2% (74) 52.5–69.8  85.3% (58) 76.9–93.7  79.7% (47) 67.2–89  Male (n = 69)  38.8% (47) 30.2–47.5  14.7% (10) 6.3–23.1  20.3% (12) 11–32.8  Pfeiffer’s test (Cognitive Impairment)  Major (n = 26)  0.8% (1) −0.8–2.4  22.6% (14) 12.2–33  29.7% (11) 15–44.5  Moderate (n = 21)  2.5% (3) −0.3–5.3  14.5% (9) 5.7–23.3  24.3% (9) 10.5–38.1  Normal-light (n = 173)  96.7% (117) 93.5–99.9  62.9% (39) 50.9–74.9  45.9% (17) 29.9–62  Barthel index (Dependence in Basic Activities of Daily Life)  Total dependence(n = 43)  0.8% (1) −0.8–2.4  22.1% (15) 12.9–33.8  45.8% (27) 33.1–58.5  Severe (n = 31)  0.8% (1) −0.8–2.4  23.5% (16) 12.2–31.9  23.7% (14) 12.9–34.6  Low-Indep. (n = 173)  98.4% (118) 96–100  54.4% (37) 42.6–66.2  30.5% (18) 18.8–42.6  Lawton scale (Instrumental Dependence)  Total dependence(n = 84)  2.5% (3) −0.3–5.5  52.9% (36) 41.1–64.8  76.3% (45) 65.4–87.1  Severe–moderate (n = 43)  11% (13) 5.5–16.9  29.4% (20) 18.6–40.2  17% (10) 7.4–26.5  Low-Indep. (n = 118)  86.5% (102) 82–93.9  17.7% (12) 8.6–26.7  6.7% (4) 0.4–13.1  MNA (Nutritional Evaluation)  Malnutrition (n = 22)  0% (0)  10.3% (7) 3.1–17.5  25.9% (15) 19.6–37.1  Risk of malnutrition (n = 70)  7.6% (9) 2.8–12.3  41.2% (28) 29.5–52.9  56.9% (33) 44.2–69.6  Normal nutrition (n = 153)  92.4% (110) 87.7–97.2  48.5% (33) 36.7–60.4  17.2% (10) 7.5–27  Chang method (Nutritional Evaluation)  Severe malnutrition (n = 2)  0% (0)  3.3% (2) −1.2–7.7  0% (0)  Moderate malnutrition (n = 19)  1.8% (2) −0.6–4.1  11.4% (7) 3.5–19.5  17.3% (10) 7.5–27  Slight malnutrition (n = 47)  11.8% (14) 6.1–18  26.3% (16) 15.2–37.3  29.3% (17) 17.6–41  Normal nutrition (n = 169)  86.4% (102) 82–93.9  59% (36) 46.7–71.4  53.4% (31) 40.6–66.3  BMI, body mass index; MNA, Mini Nutritional Assessment. View Large Results of reliability, consistency and concordance of the MNA After applying the kappa statistic to assess the intra-observer concordance, a value of 0.870 (95% CI: 0.62–1.12) was obtained, with 0.784 (95% CI: 0.37–1.19) for the inter-observer concordance. The intra-class correlation coefficient was 0.874 (95% CI: 0.76–0.94) for the intra-observer values, and 0.789 (95% CI: 0.61–0.89) for the inter-observer values. Intra- and inter-observer concordance were also evaluated using Bland and Altman plots. The intra-observer measurements, with the exception of one value, follow a normal distribution between the mean and two standard deviations. The inter-observer concordance of the measurements, with the exception of two values, remains between acceptable margins of tolerance. As for the consistency and reliability of the MNA survey, a value of 0.778 was obtained with Cronbach’s alpha for a total of 18 elements (each of the questions that make up MNA), in 218 valid cases (87.9%). Deleting four items from the survey improved the results and gave a value of 0.810. The deleted items were (in order): takes more than three prescription drugs per day (0.792), weight loss during the last 3 months (0.794), selected consumption markers for protein intake (0.807) and how many full meals does the patient eat daily? (0.810). Diagnostic accuracy of the MNA survey according to the Chang method A sensitivity value of 63.2% (95% CI: 51–75.4%) and a specificity value of 72.9% (95% CI: 65.8–80%) were obtained in the total sample. When the analysis by place of residence was carried out, important differences were obtained. The sensitivity for Group A was 18.8% (95% CI: −3.5–41%) and the specificity 94.1% (95% CI: 88.95–99.1%); in Group B was 60.7% (95% CI: 40.8–80.6%) and 56.8% (95% CI: 39.4–74.1%) and in Group C, it was 89.3% (95% CI: 76–102.5%) and 23.3% (95% CI: 6.5–40.1%) respectively. The area under the ROC curve and the remaining validity parameters are shown in Tables 2 and 3. Figure 1 shows the graphical representation of the different curves. Table 2. Diagnostic precision of MNA survey comparing the values of the current MNA with a new cut-off point   ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70    ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. aNew cut-off point. View Large Table 2. Diagnostic precision of MNA survey comparing the values of the current MNA with a new cut-off point   ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70    ROC CI 95%  S  Sp  PPV  NPV  PLR  NLR  J  VI  MNA  0.726 (0.647–0.805)  63.2%  72.9%  48.8%  82.8%  2.25  0.51  0.36  0.70  MNAa  75%  67.5%  48.6%  86.8%  2.31  0.37  0.43  0.70  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. aNew cut-off point. View Large Table 3. Accuracy of the MNA questionnaire according to population subgroup and comparison with alternative cut-off points.   Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69    Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. View Large Table 3. Accuracy of the MNA questionnaire according to population subgroup and comparison with alternative cut-off points.   Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69    Cut-off point MNA  ROC CI 95%  S (%)  Sp (%)  PPV (%)  NPV (%)  PLR  NLR  J  VI  Autonomous in health centres (Group A)  24  0.439 (0.263–0.616)  18.8  94.1  33  87.9  3.18  0.86  0.13  0.84  Not autonomous in health centres (Group B)  24  0.708 (0.579–0.836)  60.7  56.8  51.5  65.6  1.4  0.69  0.17  0.58  25  89.3  48.6  56.8  85.7  1.74  0.22  0.38  0.66  Institutionalized patients (Group C)  24  0.691 (0.552–0.830)  89.3  23.3  52  70  1.16  0.46  0.13  0.55  21  75  63.3  65.6  73  2.04  0.39  0.38  0.69  MNA, Mini Nutritional Assessment; ROC, receiver operating characteristic curve; S, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; J, Youden index; VI, validity index. View Large Figure 1. View largeDownload slide Representation of the receiver operating characteristic curve: results of the data obtained with Mini Nutritional Assessment in relation to the Chang method, in the entire sample and by subgroups. The diagonal segments are generated in the case of a tie Figure 1. View largeDownload slide Representation of the receiver operating characteristic curve: results of the data obtained with Mini Nutritional Assessment in relation to the Chang method, in the entire sample and by subgroups. The diagonal segments are generated in the case of a tie After analysing the results from the ROC curve, MNA obtained a better cut-off point to determine malnutrition with an MNA value of 25 and a Youden reading of 0.425, thus increasing the sensitivity to 75% and decreasing the specificity to 67.5%. It was observed that for Group A, the best cut-off point was the one already set in the literature (24). Group B obtained a higher Youden index 0.38 when the cut-off point was 25. In group C, a higher Youden index of 0.38 was obtained when the cut-off point was lowered to 21. Discussion The MNA has been established internationally as a valid, practical tool for measuring the nutritional status of the elderly. The main aim of this study was to assess the validity of the Spanish version of the MNA. To do this, a diverse population sample, over 65 years old, with differing degrees of dependence was recruited, to measure the validation data of the survey for use with any type of elderly population. In addition, the diagnostic accuracy in the three study subgroups has been evaluated, finding an acceptable validity in the population groups of not autonomous who lived at home and institutionalized. There was a higher percentage of females (75.4%) in the sample, which is common in a population over 65 years of age, since the males have a lower life expectancy (31). This has also been noted in other similar studies performed in a Spanish population (32–34). The reliability of the survey shows excellent results, both when assessing the correctness in determining the state of malnutrition measured by the kappa statistic and with the concordance in the score obtained. The Bland Altman plot places the results within the tolerance limits. These results are similar to those observed previously in a Spanish population using the same survey (11). In relation to the internal consistency of the survey, an acceptable value (Cronbach’s α = 0.7) was obtained. This could be improved to a better value (Cronbach’s α = 0.8) by deleting four questions from the original survey enhancing from 0.778 to 0.810. The precision obtained when performing the diagnostic tests in the entire sample shows a sensitivity and specificity of 63.2% and 72.9%, respectively, for the cut-off point proposed in the original version, which were well below the 96% and 98% obtained in the study where the survey was designed and validated (5). The fact that a lower sensitivity and specificity was obtained has led us to propose alternative cut-off values to establish the limit between normal nutritional status and the risk of malnutrition. When 25 is set as the cut-off value, the sensitivity increases to 75% and the Youden index improves. When these parameters are analysed by subgroups, a low sensitivity (18.8%) and a very high specificity (94.1%) can be seen in Group A. Therefore, the use of this questionnaire would not be recommended in a population which, although they are over 65, has a high degree of autonomy that favours their state of health and presents a lower risk of having a deficient nutritional status. In Group B, the sensitivity was 60.7% with a specificity of 56.8%. In this group, greater sensitivity (89.3%) is obtained by setting 25 as the cut-off point to establish the risk of malnutrition. This would allow those patients at risk of malnutrition to be detected sooner and follow them more thoroughly. Group C had the highest sensitivity 89.3%, although they had only 23.3% of specificity. In this group, a higher Youden index was obtained when setting the cut-off point at 21, which improved the specificity (63.3%) but lost sensitivity (75%). In this case, we consider the original cut-off point more useful. Very little validation research for the MNA has been carried out in Spanish populations. Tarazona et al. (35) performed a study in a sample with cognitive impairment and obtained a sensitivity of 60% and a specificity of 94.7%. In another study in a hospitalized Cuban population (36), different percentages were obtained depending on the gold standard used, with a maximum specificity and sensitivity of 50% and 95.1%, respectively. Another example of this variability can be seen in the validation in its Portuguese version by Santos et al. (37). Our results show a lower sensitivity and specificity than those found in Guigoz’s original work. This can be justified by the reference test used and related to the characteristics of the sample analysed. In our case, the Chang method was selected because it was considered the most objective, has great advantages of reliability, reproducibility and specificity (38), although there is little literature available to support its use, perhaps because it requires analytical and anthropometric variables and the final calculation involves rather complex calculations, which makes its use in normal clinical practice less practical. The mean age of the population can also be related to these differences (79 years in Guigoz’s paper versus 83 in this research). Guigoz’s original work validated the questionnaire in 150 patients who were healthy, frail or seriously ill (39); the results obtained lead us to consider this questionnaire to be suitable only for the frail elderly population, and we would not recommend its use in autonomous/independent patients, due to its low sensitivity in this group. This low validity, in autonomous/independent patients, can be attributed to the fact that the MNA contains several questions closely related to the functional capacity of the patient (40) and, probably, patients with a nutritional status altered but maintaining their functional capacity cannot be detected with this questionnaire. The MNA is the most used method to assess the nutritional status of the elderly, both in hospital admissions for acute pathology and in residential health care facility. However, there is not a protocol that considers the systematic nutritional assessment for elderly population in the National Health System of Spain. It is necessary to establish a care process in patients older than 65 years in which the nutritional status is evaluated periodically and MNA could be a cost-effective tool. In this way, possible assistance, social or nutritional deficits in this population could be prevented or mitigated Limitations It would be advisable to confirm these results with subsequent studies on a larger sample of well-represented patients from primary care, establishing the differences in the validity of the test according to the degree of dependence. The results may not be relevant to patients outside of the Cordova region in Spain. Declaration Funding: The Validation of the Mini Nutritional Assessment (MNA) in Spanish is part of a project supported by the Andalusian Health Service. File number: AP-0064-2016. Conflict of interest: The authors report no relationships that could be construed as a conflict of interest. References 1. Péter S, Saris WH, Mathers JC, et al.   Nutrient status assessment in individuals and populations for healthy aging-statement from an expert workshop. Nutrients   2015; 7: 10491– 500. Google Scholar CrossRef Search ADS PubMed  2. Van Bokhorst-de van der Schueren MA, Lonterman-Monasch S, de Vries OJ, et al.   Prevalence and determinants for malnutrition in geriatric outpatients. Clin Nutr   2013; 32: 1007– 11. Google Scholar CrossRef Search ADS PubMed  3. Ahmed T, Haboubi N. Assessment and management of nutrition in older people and its importance to health. Clin Interv Aging   2010; 5: 207– 16. Google Scholar PubMed  4. Ravasco P, Anderson H, Mardones F. Red Mel-CYTED. Methods of valuation of the nutritional condition. Nutr Hosp   2010; 25( suppl 3): 57– 66 5. Guigoz Y, Vellas B, Garry PJ. Mini Nutritional Assessment: a practical tool for grading the nutritional state of the elderly patients. Facts Res Gerontol   1994; ( suppl 2): 15– 60. 6. Guigoz Y. The Mini Nutritional Assessment (MNA) review of the literature–What does it tell us? J Nutr Health Aging   2006; 10: 466– 85; discussion 485–7. Google Scholar PubMed  7. Guigoz Y, Vellas B. Test d’évaluation de l’état nutritionnel de la personne âgée: le Mini Nutritional Assessment (MNA). Med Hyg   1995; 53: 1965– 69. 8. Rubenstein LZ, Harker JO, Salvà A, Guigoz Y, Vellas B. Screening for undernutrition in geriatric practice: developing the short-form mini-nutritional assessment (MNA-SF). J Gerontol A Biol Sci Med Sci   2001; 56: M366– 72. Google Scholar CrossRef Search ADS PubMed  9. Kaiser MJ, Bauer JM, Ramsch C, et al.  ; MNA-International Group. Validation of the Mini Nutritional Assessment short-form (MNA-SF): a practical tool for identification of nutritional status. J Nutr Health Aging   2009; 13: 782– 8. Google Scholar CrossRef Search ADS PubMed  10. Salvá A, Bolívar I, Muñoz M, Valvanera S. Un nuevo instrumento para la valoración nutricional en geriatría: el ‘Mini Nutricional Assessment’ (MNA). Rev Gerontol   1996; 6: 319– 28. 11. Bleda MJ, Bolibar I, Parés R, Salvà A. Reliability of the mini nutritional assessment (MNA) in institutionalized elderly people. J Nutr Health Aging   2002; 6: 134– 7. Google Scholar PubMed  12. Donini LM, de Felice MR, Tassi L, et al.   A ‘proportional and objective score’ for the mini nutritional assessment in long-term geriatric care. J Nutr Health Aging   2002; 6: 141– 6. Google Scholar PubMed  13. Ferreira LS, Nascimento LF, Marucci MF. Use of the mini nutritional assessment tool in elderly people from long-term institutions of southeast of Brazil. J Nutr Health Aging   2008; 12: 213– 7. Google Scholar CrossRef Search ADS PubMed  14. Kuzuya M, Kanda S, Koike T, Suzuki Y, Satake S, Iguchi A. Evaluation of Mini-Nutritional Assessment for Japanese frail elderly. Nutrition   2005; 21: 498– 503. Google Scholar CrossRef Search ADS PubMed  15. Sarikaya D, Halil M, Kuyumcu ME, et al.   Mini nutritional assessment test long and short form are valid screening tools in Turkish older adults. Arch Gerontol Geriatr   2015; 61: 56– 60. Google Scholar CrossRef Search ADS PubMed  16. Epidat: programa para análisis epidemiológico de datos. Versión 4.2 . Colombia: Consellería de Sanidade, Xunta de Galicia, España; Organización Panamericana de la salud (OPS-OMS); Universidad CES, 2016. 17. Martínez de la Iglesia J, Dueñas Herrero R, Onís Vilches MC, et al.   Cross-cultural adaptation and validation of Pfeiffer’s test (Short Portable Mental Status Questionnaire [SPMSQ]) to screen cognitive impairment in general population aged 65 or older . Med Clin (Barc)   2001; 117: 129– 34. 18. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist   1969; 9: 179– 86. Google Scholar CrossRef Search ADS PubMed  19. Mahoney FI, Barthel DW. Functional evaluation: the Barthel index. Md State Med J   1965; 14: 61– 5. Google Scholar PubMed  20. Chang RW. Nutritional assessment using a microcomputer. 1. Programme design. Clin Nutr   1984; 3: 67– 73. Google Scholar CrossRef Search ADS PubMed  21. Chang RW, Richardson R. Nutritional assessment using a microcomputer. 2. Programme evaluation. Clin Nutr   1984; 3: 75– 82. Google Scholar CrossRef Search ADS PubMed  22. Stewart A, Marfell-Jones M, Olds T, de Ridder H. International standards for anthropometric assessment. International Society for de Advancement of Kinanthropometry . Lower Hutt, New Zealand: ISAK, 2011. 23. Dunnett CW. A multiple comparisons procedure for comparing several treatments with a control. J Am Stat Assoc   1955; 50: 1096– 121. Google Scholar CrossRef Search ADS   24. Kruskal WH, Wallis WA. Use of ranks in one-criterion variance analysis. J Am Stat Assoc   1952; 47: 583– 621. Google Scholar CrossRef Search ADS   25. Lilliefors H. On the Kolmogorov–Smirnov test for normality with mean and variance unknown. J Am Stat Assoc   1967; 62: 399– 402. Google Scholar CrossRef Search ADS   26. Cohen J. Weighted kappa: nominal scale agreement with provision for scaled disagreement or partial credit. Psychol Bull   1968; 70: 213– 20. Google Scholar CrossRef Search ADS PubMed  27. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet   1986; 1: 307– 10. Google Scholar CrossRef Search ADS PubMed  28. Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika   1951; 16: 297– 334. Google Scholar CrossRef Search ADS   29. Youden WJ. Index for rating diagnostic tests. Cancer   1950; 3: 32– 5. Google Scholar CrossRef Search ADS PubMed  30. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology   1982; 143: 29– 36. Google Scholar CrossRef Search ADS PubMed  31. Abellán García A, Pujol Rodríguez R. Un perfil de las personas mayores en España, 2016. Indicadores estadísticos básicos. Madrid, Informes Envejecimiento en red n º 14. 2016: 3–7. 32. Jiménez Sanz M, Sola Villafranca JM, Pérez Ruiz C, et al.   Study of the nutritional status of elders in Cantabria. Nutr Hosp   2011; 26: 345– 54. Google Scholar PubMed  33. Cuerda C, Álvarez J, Ramos P, et al.   Prevalence of malnutrition in subjects over 65 years of age in the Community of Madrid. The DREAM + 65 Study. Nutr Hosp   2016; 33: 263– 69. Google Scholar CrossRef Search ADS PubMed  34. Cuervo M, Ansorena D, García A, Astiasarán I, Martínez JA. Food consumption analysis in Spanish elderly based upon the mini nutritional assessment test. Ann Nutr Metab   2008; 52: 299– 307. Google Scholar CrossRef Search ADS PubMed  35. Tarazona FJ, Belenguer A, Doménech JR, et al.   Validation of score in MNA scale like nutritional risk factor in institutionalized geriatric patients with moderate and severe cognitive impairment. Nutr Hosp   2009; 24: 724– 31. Google Scholar PubMed  36. Cuyac Lantigua M, Santana Porbén S. The Mini Nutritional Assessment of the elderly in the practice of a hospital geriatrics service: inception, validation and operational characteristics. Arch Latinoam Nutr   2007; 57: 255– 65. Google Scholar PubMed  37. Santos Pereira Machado R, Santa Cruz Coelho MA, Peixoto Veras R. Validity of the Portuguese version of the mini nutritional assessment in Brazilian elderly. BMC Geriatrics   2015; 15: 132. Google Scholar CrossRef Search ADS PubMed  38. Linares Guerra ME, Bencomo JF, Santana S, Barreto J, Ruiz ML. Nutritional evaluation of HIV-infected patients by Chang’s method. DST—J bras Doenças Sex Transm   2005; 17: 259– 64. 39. Vellas B, Villars H, Abellan G, et al.   Overview of the MNA—its history and challenges. J Nutr Health Aging   2006; 10: 456– 63. Google Scholar PubMed  40. Muñoz Díaz B, Arenas de Larriva AP, Molina-Recio G, Moreno-Rojas R, Martínez de la Iglesia J; grupo de investigación Nutrianco. Study of the nutritional status of patients over 65 years included in the home care program in an urban population. Aten Primaria   2018; 50: 88– 95. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Family PracticeOxford University Press

Published: Jun 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off