India is ageing rapidly. Persons aged 60 years and above made up 8.6% of the population at the 2011 Census, were estimated at about 138 million (10.1%) in 2021, and are projected to reach 194 million by 2031.¹ Uttar Pradesh holds one of the largest absolute elderly populations in the country, although its proportional share is among the lowest of the major states and remains below the national figure.¹ Ageing and undernutrition travel together in this setting: in the Longitudinal Ageing Study in India (LASI) Wave 1, 26.7% of Indians aged 60 years and above were underweight, and Uttar Pradesh, at 37%, was among the states with the highest prevalence.²

A meta-analysis of Indian studies using the Mini Nutritional Assessment (MNA) estimated pooled malnutrition among elderly Indians at 18.3% and pooled risk of malnutrition at 48.2%.³ The primary studies vary widely and are almost entirely community-based: 17.9% malnourished with 58.8% at risk in rural Puducherry,⁴ 20.8% and 49.2% in rural Manipur,⁵ possible malnutrition in 49.3% of elderly residents of urbanized villages in East Delhi,⁶ 15% and 55% in rural Assam,⁷ and 18% undernourished with 38% at risk in rural Gautam Buddha Nagar, the only recent MNA survey from Uttar Pradesh.⁸ The correlates that recur through these surveys are advancing age, female sex, functional and financial dependency and low socio-economic status.^6–8 Community surveys, however, describe elderly people living at home rather than those admitted acutely with anorexia, enforced fasting and inflammation-driven catabolism added on top.

Admission to hospital both exposes and worsens this burden. Depending on the setting, diagnosis and criteria applied, malnutrition is reported in 30 to 50% of hospitalized patients, and it impairs wound healing, blunts immunity and compromises organ function, which in turn lengthens stay and raises readmission, mortality and cost.⁹ When MNA data were pooled by healthcare setting, prevalence of malnutrition in hospital (22.0%) was several times that in the community (3.1%) or among outpatients (6.0%).¹⁰ Multicentre studies from Korea¹¹ and Colombia,¹² where the economic penalty was quantified alongside the clinical one, and a comprehensive review of malnutrition in hospitalized older patients¹³ reach the same conclusion. The most directly relevant study is a Chinese national study of 5,516 older inpatients screened with the MNA-SF, in which malnutrition carried close to a six-fold increase in the odds of in-hospital death, and both malnutrition and risk of malnutrition independently lengthened stay and increased 90-day readmission.¹⁴ Indian hospital data are sparse. A Bangalore teaching hospital linked malnutrition in 200 elderly inpatients to cognitive and functional decline, depression and longer stay,¹⁵ a study of 177 hospitalized elderly patients at a North Indian tertiary-care centre found MNA-defined malnutrition to predict three-month mortality,¹⁶ and a nutritionDay audit in five Indian hospitals documented a high prevalence of malnutrition risk and poor food intake among older inpatients.¹⁷

What makes this worth screening for is that the risk is modifiable. The EFFORT trial showed that individualized nutritional support given to medical inpatients at nutritional risk reduced adverse outcomes including mortality,¹⁸ its secondary analysis established the nutritional risk score as a modifiable predictor rather than a passive prognostic label,¹⁹ and an updated systematic review and meta-analysis reached the same conclusion.²⁰ The MNA-SF is well suited to the task in older adults. It was derived from the full 18-item MNA²¹ and later revalidated with calf circumference standing in for body mass index where height or weight cannot be obtained.²² Six items are scored to a total of 0 to 14, placing the patient in one of three groups: normal nutritional status (12–14), at risk of malnutrition (8–11) or malnourished (0–7). It is regarded as a practical first-line screen for hospitalized older people,^23,24 and the 2019 ESPEN guideline on clinical nutrition and hydration in geriatrics asks that all older persons be screened with a validated tool and that those who screen positive be assessed against the Global Leadership Initiative on Malnutrition (GLIM) criteria.^25,26

Despite Uttar Pradesh carrying one of the country's highest burdens of late-life underweight, and despite Lucknow hosting several high-volume tertiary-care teaching institutions and a designated Regional Geriatric Centre under the National Programme for Health Care of the Elderly,²⁷ our structured search of the indexed literature for 2018 to 2023 did not retrieve any hospital-based study using the MNA or MNA-SF among elderly inpatients in Lucknow, or elsewhere in Uttar Pradesh, that also reported hospital outcomes. The region with the greatest need therefore has the least hospital-based evidence, and the present study was planned to narrow that gap.

Aim and objectives

The aim was to determine the prevalence of malnutrition and of risk of malnutrition among patients aged 60 years and above admitted to a tertiary-care teaching hospital in Lucknow using the MNA-SF, and to examine whether the nutritional category assigned at admission is associated with the outcome of that admission.

The objectives were:

1. To estimate the prevalence of malnutrition (MNA-SF 0–7) and of risk of malnutrition (MNA-SF 8–11), with 95% confidence intervals, among patients aged 60 years and above admitted to the medicine and allied wards (primary objective).
2. To describe the distribution of scores on the six MNA-SF items and identify the items accounting for most of the score lost.
3. To identify the socio-demographic, dental, pharmacological, functional and comorbidity correlates of nutritional risk.
4. To measure the association between MNA-SF category at admission and length of stay, hospital-acquired complication, in-hospital mortality and readmission within 30 days of discharge, unadjusted and then adjusted for age, sex and comorbidity burden.
5. As a post-hoc comparison, to count how many patients at nutritional risk a body mass index threshold of 18.5 kg/m² used on its own would have missed.

Study design, setting and period This was a hospital-based observational cross-sectional study with prospective follow-up of in-hospital outcomes to discharge and of readmission to 30 days after discharge. It was carried out in the medicine and allied inpatient wards of Career Institute of Medical Sciences & Hospital, Lucknow, Uttar Pradesh, a tertiary-care teaching hospital, over a six-month period from October 2024 to March 2025. Reporting follows the STROBE statement for observational studies.28 Participants Consecutive patients aged 60 years and above admitted to the medicine and allied wards during the study period were eligible, provided they could be assessed within 48 hours of admission and gave written informed consent personally or through a legally acceptable representative where capacity was impaired. Patients were excluded if the admission lasted less than 24 hours, including planned day-care and diagnostic admissions; if they were admitted directly to critical care with intubation or sedation that made MNA-SF administration impossible; if death was anticipated within 48 hours of admission; if consent was refused; or if neither body mass index nor calf circumference could be obtained, one of the two being mandatory for item F of the MNA-SF. Sample size Sample size was calculated for the estimation of a single proportion by the formula n = Z2pq/d2, where Z is 1.96 for 95% confidence, p is the anticipated proportion, q is 1 − p and d is the absolute precision.29 Taking the pooled prevalence of risk of malnutrition among Indian elderly of 48% reported by Kushwaha et al.3 as the anticipated proportion, with 5% absolute precision, n = (1.96)2 × 0.48 × 0.52 / (0.05)2 ≈ 384. A minimum of 384 participants was therefore required; 384 were enrolled and analysed. Study instrument and data collection All assessments were carried out within 48 hours of admission by trained investigators using a pre-tested structured proforma. The MNA-SF was administered in the local language in its validated six-item form,21,22,30 without modification of items or cut-offs. The six items cover decline in food intake over the past three months (0–2), weight loss during the last three months (0–3), mobility (0–2), psychological stress or acute disease in the past three months (0–1), neuropsychological problems (0–2) and body mass index (0–3), the last replaced by calf circumference (0 below 31 cm, 3 at 31 cm or more) where height or weight could not be obtained reliably.22 The total ranges from 0 to 14 and is read as 12–14 for normal nutritional status, 8–11 for risk of malnutrition and 0–7 for malnutrition. Weight was taken on a calibrated digital scale without footwear or heavy outer garments to the nearest 0.1 kg, using a bed or chair scale for non-ambulant patients. Height was measured with a stadiometer where the patient could stand and otherwise estimated from knee height or half-arm-span. Body mass index was calculated as weight in kilograms divided by the square of height in metres and interpreted using the World Health Organization cut-offs for Asian populations.31 Calf circumference was measured on the non-dominant leg at the maximum girth with the knee flexed to 90 degrees, to the nearest 0.1 cm. Age, sex, residence (urban or rural), living arrangement, educational attainment and economic dependency were recorded by interview. Comorbidity burden was quantified with the Charlson comorbidity index32 and functional status with the Katz index of independence in activities of daily living, scored 0 to 6.33 Polypharmacy was defined in advance as the use of five or more regular prescribed medications. Dentition was recorded as significant dentition loss or edentulism where chewing capacity was judged to be materially impaired. Outcome variables Four hospital outcomes were defined in advance and ascertained prospectively. Length of hospital stay was counted in completed days from admission to discharge or death. Prolonged stay was defined as a stay of 7 days or more, a threshold chosen because nutritional status itself deteriorates measurably in admissions of that duration.34 Hospital-acquired complication was defined as any new nosocomial infection, pressure injury, fall or delirium episode arising after 48 hours of admission. The remaining two outcomes were in-hospital mortality and readmission within 30 days of discharge, the latter ascertained from hospital records and by telephone follow-up. For the primary analysis of readmission the denominator was all enrolled patients, with patients who died in hospital counted as not readmitted; a sensitivity analysis restricted the denominator to patients discharged alive. Statistical analysis Data were analysed in Python 3 using the pandas, SciPy and statsmodels libraries. Continuous variables are summarized as mean ± standard deviation (SD) with median and interquartile range, and categorical variables as frequency and percentage. Prevalence estimates carry 95% confidence intervals (CI) computed by the normal approximation to the binomial distribution. Associations between MNA-SF category and categorical variables were tested by the Pearson chi-square test. Differences in continuous variables across the three MNA-SF categories were tested by one-way analysis of variance (ANOVA), with the Kruskal–Wallis test run alongside every comparison as a distribution-free confirmation. Length of stay in the combined at-risk or malnourished group was compared with the well-nourished group by Welch's unequal-variance t test with Mann–Whitney U confirmation. Linear association of the continuous MNA-SF score with clinical variables was quantified by Pearson correlation, with Spearman rank correlation reported beside it. For each of the four binary outcomes a multivariable binary logistic regression model was fitted, with MNA-SF category entered as two indicator variables (at risk, malnourished) against normal nutritional status as the reference, and adjusted for age in years, sex and Charlson comorbidity index per point. Adjusted odds ratios (aOR) are reported with 95% CI and Wald p values, together with McFadden pseudo-R2 and the likelihood-ratio test p value. The screening yield of the MNA-SF was then compared with a body mass index rule, taking an MNA-SF score of 11 or below as the reference standard and a body mass index below 18.5 kg/m² as the comparator test, with sensitivity, specificity, predictive values, overall agreement and Cohen's kappa reported. A two-tailed p value below 0.05 was taken as statistically significant throughout. Ethical considerations The protocol was approved by the Institutional Ethics Committee of Career Institute of Medical Sciences & Hospital, Lucknow, before the study began. Written informed consent was obtained from every participant, or from a legally acceptable representative where decision-making capacity was impaired. The study followed the Declaration of Helsinki as revised in 201335 and the Indian Council of Medical Research National Ethical Guidelines for Biomedical and Health Research Involving Human Participants, 2017.36 Data were anonymised at entry and held in password-protected form. Patients found to be malnourished or at risk were referred to the hospital dietetics service for formal nutritional assessment and care planning in line with the ESPEN geriatric guideline,25 and no participant was denied any standard-of-care intervention by reason of participation. 

Participants and baseline characteristics

Three hundred and eighty-four patients aged 60 years and above were enrolled and analysed. All were followed to discharge or death, and 30-day readmission status was ascertained for every patient discharged alive, so there was no loss to follow-up. Mean age was 71.3 ± 7.3 years, with a median of 70 years and a range of 60 to 97. There were 211 men (54.9%) and 173 women (45.1%). By age band, 181 patients (47.1%) were 60–69 years, 150 (39.1%) were 70–79 years and 53 (13.8%) were 80 years or above. Mean body mass index was 22.1 ± 3.4 kg/m², and 53 patients (13.8%) had a body mass index below 18.5 kg/m². The mean MNA-SF score was 9.7 ± 3.0, with a median of 11 and a range of 0 to 13. Baseline characteristics overall and by MNA-SF category are shown in Table 1.

Table 1: Baseline characteristics of 384 hospitalized elderly patients, overall and by MNA-SF category.

Percentages for categorical rows are column percentages within each MNA-SF category. Chi-square tests compare the three MNA-SF categories (df 2 unless stated). Overall mean ± SD for calf circumference, Charlson index and Katz score are pooled from the category values. Kruskal–Wallis testing confirmed every ANOVA result at p < 0.001. Education dichotomised as illiterate versus any schooling, and living arrangement as living alone versus living with others.

Prevalence of malnutrition and of malnutrition risk

Eighty-seven patients (22.7%, 95% CI 18.5–26.8) were malnourished on MNA-SF scoring and a further 154 (40.1%, 95% CI 35.2–45.0) were at risk of malnutrition, while 143 (37.2%, 95% CI 32.4–42.1) had normal nutritional status. Combining the two abnormal strata, 241 patients (62.8%, 95% CI 57.9–67.6) carried some degree of nutritional risk at admission (Figure 1).

Figure 1: Prevalence of MNA-SF categories among 384 hospitalized elderly patients. Error bars denote 95% confidence intervals of the proportion.

Correlates of nutritional risk

Nutritional category was associated with age band (χ² = 26.01, df 4, p < 0.001). Within each age band, the proportion malnourished rose from 14.4% at 60–69 years to 25.3% at 70–79 years and 43.4% at 80 years and above (Figure 2). Edentulism or significant dentition loss was the strongest non-demographic correlate (χ² = 12.59, p = 0.002): 72.5% of patients with poor dentition were at nutritional risk against 56.6% of those with intact dentition. Economic dependency (χ² = 7.00, p = 0.030) was also associated with MNA-SF category, and rural residence showed a borderline association (χ² = 5.97, p = 0.050), whereas sex (p = 0.759), illiteracy (p = 0.349), living alone (p = 0.180) and polypharmacy (p = 0.183) were not (Table 1). These correlate analyses are univariate.

Figure 2: Distribution of MNA-SF category within each age band (percentages are row-wise within the age band). χ² = 26.01, df 4, p < 0.001.

On continuous measures, mean age, body mass index, calf circumference, Charlson comorbidity index and Katz activities-of-daily-living score all differed across the three MNA-SF categories, with every ANOVA at p < 0.001 and each confirmed by Kruskal–Wallis testing (Table 1). The MNA-SF total score correlated positively with body mass index (r = 0.73), calf circumference (r = 0.61) and functional independence (r = 0.24), and inversely with length of hospital stay (r = −0.44), age (r = −0.32) and comorbidity burden (r = −0.28), all at p < 0.001 (Table 2).

Table 2: Correlation of MNA-SF total score with clinical variables (n = 384).

Item-level analysis

Item-level analysis showed where the score was lost (Figure 3). The worst affected item was D, recent psychological stress or acute disease, at 39.8% of its maximum attainable score (mean 0.40 of 1), followed by F, body mass index, at 60.2% (mean 1.80 of 3). Items reflecting intake and weight history scored better, A at 79.4% (mean 1.59 of 2) and B at 81.2% (mean 2.43 of 3), as did mobility at 84.4% (mean 1.69 of 2), and neuropsychological problems were least affected at 88.5% (mean 1.77 of 2).

Figure 3: Mean MNA-SF item score as a percentage of each item's maximum (n = 384). A lower percentage denotes greater impairment in that domain. Items below 65% of maximum are highlighted.

Hospital outcomes by MNA-SF category

Mean length of stay increased steadily across the three MNA-SF categories, from 5.52 ± 1.99 days in patients with normal nutritional status to 7.04 ± 2.33 days in at-risk patients and 9.15 ± 3.96 days in malnourished patients (ANOVA F = 49.54, p < 0.001; Kruskal–Wallis H = 76.37, p < 0.001). Malnourished patients stayed a mean 3.6 days longer than the well-nourished. Taking the two abnormal strata together, patients with any nutritional risk stayed 7.80 ± 3.18 days against 5.52 ± 1.99 days for the well-nourished, a mean excess of 2.28 days, or about 41% (Welch t = 8.63, p < 0.001; Mann–Whitney U = 25,434, p < 0.001).

All four binary outcomes showed a stepwise gradient across MNA-SF categories (Table 3). Prolonged stay of 7 days or more occurred in 25.2% of well-nourished patients, 51.3% of at-risk patients and 70.1% of malnourished patients (χ² = 47.10, p < 0.001). Hospital-acquired complications occurred in 12.6%, 17.5% and 26.4% respectively (χ² = 7.13, p = 0.028). In-hospital mortality was 2.1%, 4.5% and 13.8% respectively (χ² = 14.37, p < 0.001), a 6.6-fold higher crude death rate in malnourished than in well-nourished patients. Readmission within 30 days occurred in 9.1%, 10.4% and 21.8% respectively (χ² = 9.08, p = 0.011). When the readmission denominator was restricted to the 362 patients discharged alive, the corresponding proportions were 9.3%, 10.9% and 25.3% (χ² = 12.15, p = 0.002).

Table 3: Hospital outcomes by MNA-SF category (n = 384).

F, one-way ANOVA; H, Kruskal–Wallis; IQR, interquartile range. ^a Denominator is all enrolled patients, with in-hospital deaths counted as not readmitted.

Multivariable analysis

Four multivariable logistic regression models, each adjusted for age, sex and Charlson comorbidity index and each taking normal nutritional status as the reference, are summarized in Table 4 and Figure 4.

Malnutrition independently predicted three of the four outcomes. The adjusted odds of in-hospital death were 8.08-fold higher in malnourished patients (95% CI 2.01–32.45, p = 0.003), of prolonged stay 4.28-fold higher (95% CI 2.27–8.06, p < 0.001) and of 30-day readmission 2.78-fold higher (95% CI 1.20–6.46, p = 0.017). For hospital-acquired complications the point estimate was raised but did not reach significance (aOR 1.80, 95% CI 0.85–3.81, p = 0.127), and in that model comorbidity burden was the significant predictor (aOR 1.25 per Charlson point, p = 0.018).

Being at risk already carried an independent penalty for stay: at-risk patients had 2.74-fold higher adjusted odds of prolonged stay (95% CI 1.65–4.55, p < 0.001). For mortality, complications and readmission the at-risk point estimates were raised but their confidence intervals included unity. Age (aOR 1.06 per year, p = 0.001) and Charlson index (aOR 1.37 per point, p < 0.001) were themselves independent predictors of prolonged stay, indicating that the nutritional signal is additional to, and not a proxy for, chronological age or comorbidity burden.

Table 4: Multivariable logistic regression: adjusted odds of adverse hospital outcomes (reference = normal nutritional status, MNA-SF 12–14).

All models adjusted for age, sex and Charlson comorbidity index; only predictors reaching p < 0.05 other than MNA-SF category are shown. Bold entries are statistically significant at p < 0.05. OR, odds ratio; CI, confidence interval.

Figure 4: Adjusted odds ratios for adverse hospital outcomes by MNA-SF category, reference = normal nutritional status. The horizontal axis is logarithmic and the vertical line marks an odds ratio of 1. Filled markers denote statistical significance at p < 0.05.

MNA-SF against body mass index as a screening rule

Taking an MNA-SF score of 11 or below as the reference standard, a rule of body mass index below 18.5 kg/m² identified only 53 of the 241 nutritionally at-risk patients and missed 188 of them (78.0%). All 53 underweight patients were at risk on the MNA-SF, so the rule had a specificity of 100% and a positive predictive value of 100%, but its sensitivity was 22.0% and its negative predictive value 43.2%. Overall agreement was 51.0% and Cohen's kappa 0.17, indicating poor agreement beyond chance (Figure 5).

Figure 5: Screening yield of the MNA-SF against a body mass index threshold. Of 241 patients at nutritional risk on the MNA-SF, a rule of body mass index below 18.5 kg/m² would have flagged 53 and missed 188.

Principal findings Nutritional risk in this cohort was the majority state rather than the exception. Nearly two-thirds of elderly inpatients (62.8%) screened positive and 22.7% were malnourished. Nutritional category then graded every hospital outcome we examined in a stepwise fashion, and that gradient survived adjustment for age, sex and comorbidity: malnutrition raised the adjusted odds of in-hospital death roughly eight-fold, of prolonged stay four-fold and of 30-day readmission almost three-fold. Being only at risk was not a benign state either, since those patients already carried 2.74-fold higher adjusted odds of a prolonged stay, and it is precisely this stratum that an approach based on body mass index fails to see. Comparison with published studies The prevalence figures sit inside the published range, which is reassuring for a single-centre study (Table 5). Kushwaha et al. pooled 18.3% malnourished and 48.2% at risk among elderly Indians, mostly from community settings,3 and our hospital figures of 22.7% and 40.1% are consistent with that while showing the upward shift in malnutrition expected in an inpatient population. Cereda et al. demonstrated this setting gradient when pooling MNA data, with hospital prevalence several times that of the community.10 Our combined burden of 62.8% is of the same order as the 68% reported from a Bangalore teaching hospital15 and the 72% (22.6% malnourished, 49.7% at risk) reported from a North Indian tertiary-care centre,16 above the 46.2% of the Chinese national inpatient study,14 and consistent with the general statement that malnutrition affects 30 to 50% of hospitalized patients depending on setting and criteria.9 Regionally the finding is unsurprising, since LASI identified 37% of the Uttar Pradesh population aged 60 and above as underweight against 26.7% nationally.2 Our mean age in the low seventies and slight male preponderance are close to the Chinese MNA-SF inpatient cohort, in which mean age was 72.5 ± 5.8 years.14 The outcome findings reproduce the international literature closely. Liu et al., in 5,516 older inpatients screened with the same instrument, reported odds ratios for in-hospital death of 5.74 (95% CI 3.47–9.48) for malnutrition and 3.28 (95% CI 2.13–5.07) for risk of malnutrition, with increases of about 26 to 28% in length of stay and about 36 to 39% in 90-day readmission.14 Our adjusted odds ratio of 8.08 for mortality in malnourished patients is of the same order and direction, with a wider confidence interval that reflects a far smaller cohort and only 22 deaths; the two intervals overlap substantially. Our at-risk estimate for mortality (aOR 2.26) lies within their confidence interval even though our own interval crosses unity, which is what would be expected with so few events. Our excess stay of 2.28 days, or about 41%, is larger than the 26 to 28% increase in that study but consistent in direction, and the absolute stay of about nine days in our malnourished group matches the 9.13 days reported from Bangalore.15 Ruiz et al. showed that inpatient malnutrition carries an economic as well as a clinical penalty,12 of the kind modelled much earlier by Correia and Waitzberg.37 That the readmission signal persisted to 30 days is in keeping with Sharma et al., who found associations with readmission and death both within 7 days and from 8 to 180 days after discharge,38 and the mortality association agrees with Işık et al., who linked nutritional status to early and mid-term mortality in geriatric emergency admissions,39 with Söderström et al., who showed malnutrition to predict death in older adults regardless of cause,40 and with Mukundan et al., who found MNA-defined malnutrition to predict three-month mortality in hospitalized Indian elderly (OR 3.80).16 Table 5: Principal findings of the present study compared with published studies. Parameter Present study (Lucknow, n = 384) Published comparison Source Malnourished 22.7% 18.3% pooled (Indian, mainly community) Kushwaha et al.3 At risk of malnutrition 40.1% 48.2% pooled (Indian, mainly community) Kushwaha et al.3 Any nutritional risk, community 62.8% 76.7% rural Puducherry; 70.0% rural Manipur; 56% rural Uttar Pradesh Krishnamoorthy et al.;4 Joymati et al.;5 Gohar et al.8 Any nutritional risk, inpatients 62.8% 46.2% Chinese inpatients; 68% Bangalore inpatients; 72% Rishikesh inpatients Liu et al.;14 Manjaly et al.;15 Mukundan et al.16 Mean age of cohort 71.3 ± 7.3 years 72.5 ± 5.8 years Liu et al.14 Excess length of stay with nutritional risk +2.28 days (≈ +41%) +26 to +28% Liu et al.14 Length of stay, malnourished 9.15 ± 3.96 days 9.13 days Manjaly et al.15 Adjusted OR, in-hospital death, malnourished 8.08 (2.01–32.45) 5.74 (3.47–9.48) Liu et al.14 Adjusted OR, in-hospital death, at risk 2.26 (0.57–8.99) 3.28 (2.13–5.07) Liu et al.14 Mortality, malnourished aOR 8.08 (in-hospital) OR 3.80 (3-month) Mukundan et al.16 30-day readmission, malnourished 21.8% Raised readmission or death at 0–7 and 8–180 days Sharma et al.38 Correlates of nutritional risk Age, poor dentition, economic dependency (rural residence borderline) Age, female sex, functional and financial dependency; age, tenancy, financial dependency Agarwalla et al.;7 Vaish et al.6 Agreement of BMI < 18.5 kg/m² with screening κ = 0.17 Different definitions capture different patients Sánchez-Rodríguez et al.41 OR, odds ratio; aOR, adjusted odds ratio; BMI, body mass index; κ, Cohen's kappa. Correlates and item-level findings The correlates we found, namely advancing age, poor dentition and economic dependency, with a borderline association with rural residence, fit the multifactorial model of late-life malnutrition in which dentition, drugs, depression, dependency and disease act together.42 Age and financial dependency recur through the Indian community literature,6–8 and the socio-economic axis picked out by dependency and, less firmly, rural residence matches the pattern described from rural Puducherry and East Delhi.4,6 Poor dentition emerging as the strongest non-demographic correlate is clinically useful in a way that age is not, because impaired chewing can be addressed through texture modification of the diet and dental referral, and it is routinely overlooked in medical admission assessment. That sex, education and living arrangement were not significant here does not contradict the wider literature so much as reflect the composition of this cohort and the statistical power available; the pooled Indian data show these particular associations to be inconsistent between regions.3 The item-level analysis adds something of practical value. The two most depressed items were D, recent psychological stress or acute disease, and F, body mass index. A low score on item D is intrinsic to an inpatient population, since hospitalization itself amounts to acute disease, and this partly explains why hospital MNA-SF prevalence structurally exceeds community prevalence; the point needs to be kept in mind whenever figures are compared across settings. It also illustrates the mechanism by which acute inflammation raises resting energy expenditure, promotes muscle catabolism and suppresses appetite at the same time, so that requirement rises just as intake falls.9 Item F being the second most affected indicates that a genuine anthropometric deficit was present and not merely inferred. Why body mass index cannot substitute for a structured screen The most operationally consequential result is the comparison of the MNA-SF with a body mass index threshold. A rule of body mass index below 18.5 kg/m², the one most often applied informally at the bedside, would have flagged 53 patients and missed 188 of the 241 at nutritional risk, a sensitivity of only 22% and a kappa of 0.17 that is barely better than chance agreement. This is a consequence of how nutritional risk arises. Falling intake, recent weight loss, immobility and acute illness all carry risk well before body mass index crosses the underweight threshold, and a patient can be overweight or obese and malnourished at the same time through weight loss or sarcopenic muscle loss.9,26 Serum proteins offer no rescue: albumin and prealbumin behave as negative acute-phase reactants, falling with inflammation whatever the nutritional state, and are therefore not recommended as indicators of malnutrition.9 Sánchez-Rodríguez et al. showed directly that different nutritional definitions capture materially different patients in geriatric care.41 Calf circumference, by contrast, gives a simple bedside approximation of skeletal muscle mass that has been validated clinically,9 which is why the MNA-SF embeds it,22 and where muscle function rather than muscle mass is of interest the Asian Working Group for Sarcopenia 2019 consensus supplies region-appropriate thresholds.43 A six-item structured screen is therefore not a bureaucratic addition to the admission process but the only dependable way of identifying the majority of at-risk patients. Clinical and programmatic implications MNA-SF screening within 48 hours of admission should be routine rather than discretionary for every inpatient aged 60 and above, which is what the ESPEN geriatric guideline recommends for all older persons using a validated tool.25 A positive screen then has to trigger a defined pathway of GLIM-based diagnostic assessment,26 dietetics referral and an individualized nutrition care plan, since the evidence that nutritional risk is modifiable comes specifically from trials of individualized support in screen-positive inpatients,18,20,44 and the choice of screening tool itself propagates into the measured benefit of intervention.45 The at-risk stratum must be treated as actionable, being the single largest group in this cohort at 40.1% and already carrying independent excess risk of prolonged stay. Because nutritional status can deteriorate during admissions of 7 days or more,34 re-screening during longer admissions ought to be built into the pathway rather than relying on one snapshot at admission. Programmatically the scaffolding already exists: the National Programme for Health Care of the Elderly provides geriatric clinics, geriatric wards and Regional Geriatric Centres, one of which operates in Lucknow, so embedding a six-item screen needs a protocol change rather than new infrastructure.27 Strengths and limitations The strengths of this study are: • Use of a validated, internationally standardized instrument with its published cut-offs unmodified. • Consecutive enrolment to limit selection bias, with the calculated sample size reached and no loss to follow-up. • Measurement of calf circumference alongside body mass index, so that item F could be scored even in non-ambulant patients. • Four distinct outcome domains examined rather than a single composite, with every parametric test confirmed by a distribution-free equivalent and all outcome analyses adjusted for age, sex and comorbidity burden. • Reporting to the STROBE standard,28 addressing a documented geographical gap in the Indian hospital literature. The limitations need to be stated plainly: 1. Single-centre design in a tertiary-care teaching hospital, which over-represents referred and more severely ill patients and limits generalizability to secondary-level and community hospitals. 2. Screening at a single time point. The MNA-SF was administered once, within 48 hours of admission, so deterioration during the stay was not captured even though it is known to occur in admissions of 7 days or more.34 3. Screening is not diagnosis. The MNA-SF categorizes nutritional risk and does not constitute a GLIM-based diagnosis of malnutrition, which additionally needs a phenotypic and an etiologic criterion,26 so the figures reported here are screening prevalences. 4. Item D is inflated by design in an inpatient population, which complicates direct comparison with community settings. 5. Residual confounding. Disease severity at admission, inflammatory markers such as C-reactive protein, detailed socio-economic status and diagnosis category were not modelled, the correlate analyses were univariate, and reverse causality cannot be excluded, since severe acute illness both depresses MNA-SF items and independently prolongs stay. 6. Limited power for rare outcomes. With 22 deaths against five model parameters (about four events per variable) the mortality model is at risk of overfitting and its confidence intervals are wide, so the non-significant at-risk estimates for mortality, complications and readmission should be read as inconclusive rather than as evidence of no effect, particularly as larger cohorts have found them significant.14 7. Readmission was modelled on all enrolled patients with in-hospital deaths counted as not readmitted; the survivor-only proportions in Table 3 show the same gradient. 8. No intervention arm. This study identifies risk but does not show that acting on the screen improves outcomes in this population, although randomized evidence from elsewhere supports that inference.18,20 9. Sarcopenia was not formally assessed with handgrip strength or performance measures against the Asian Working Group for Sarcopenia 2019 thresholds,43 so muscle function as distinct from muscle mass was not characterized. What is needed next is multicentre prospective work across Uttar Pradesh, with serial MNA-SF screening during admission, GLIM-based diagnostic confirmation of screen-positive patients, formal sarcopenia assessment using region-specific thresholds, inflammatory marker profiling, costing of the excess length of stay, and ultimately an interventional arm testing whether a protocolized screen-and-refer pathway shortens stay and reduces readmission in the Indian tertiary-care setting.

Among 384 patients aged 60 years and above admitted to a tertiary-care teaching hospital in Lucknow, 62.8% carried nutritional risk on MNA-SF screening and 22.7% were malnourished. Nutritional status graded length of stay, hospital-acquired complications, in-hospital mortality and 30-day readmission in a stepwise manner, and malnutrition remained an independent predictor of death, prolonged stay and readmission after adjustment for age, sex and comorbidity burden. These estimates agree in magnitude and direction with Indian and international studies using the same instrument. Since a body mass index rule on its own missed 78% of at-risk patients, clinical impression and anthropometry cannot stand in for a structured screen. Routine MNA-SF screening of every elderly inpatient within 48 hours of admission, with a defined referral and nutrition-care pathway for those who screen positive, is a low-cost and high-yield measure that can be implemented within the existing National Programme for Health Care of the Elderly framework.

DECLARATIONS

Ethical approval and consent to participate. The study was approved by the Institutional Ethics Committee of Career Institute of Medical Sciences & Hospital, Lucknow. Written informed consent was obtained from all participants or their legally acceptable representatives.

Conflicts of interest. None.

1. National Statistical Office. Elderly in India 2021. New Delhi: Ministry of Statistics and Programme Implementation, Government of India; 2021. 2. International Institute for Population Sciences (IIPS), National Programme for Health Care of Elderly (NPHCE), Ministry of Health and Family Welfare, Harvard T.H. Chan School of Public Health, University of Southern California. Longitudinal Ageing Study in India (LASI) Wave 1, 2017–18, India Report. Mumbai: IIPS; 2020. 3. Kushwaha S, Khanna P, Srivastava R, Jain R, Singh T, Kiran T. Estimates of malnutrition and risk of malnutrition among the elderly (≥60 years) in India: a systematic review and meta-analysis. Ageing Res Rev. 2020;63:101137. 4. Krishnamoorthy Y, Vijayageetha M, Kumar SG, Rajaa S, Rehman T. Prevalence of malnutrition and its associated factors among elderly population in rural Puducherry using mini-nutritional assessment questionnaire. J Family Med Prim Care. 2018;7(6):1429–33. 5. Joymati O, Ningombam M, Rajkumari B, Gangmei A. Assessment of nutritional status among elderly population in a rural area in Manipur: community-based cross-sectional study. Int J Community Med Public Health. 2018;5(7):3125–9. 6. Vaish K, Patra S, Chhabra P. Nutritional status among elderly: a community-based cross-sectional study. Indian J Public Health. 2020;64(3):266–70. 7. Agarwalla R, Saikia AM, Baruah R. Assessment of the nutritional status of the elderly and its correlates. J Family Community Med. 2015;22(1):39–43. 8. Gohar M, Barua K, Srivastava S, Mahajan H, Pawaiya AS, Nagar SK. A cross-sectional study to assess the under-nutrition amongst the elderly population in a rural area of district Gautam Buddha Nagar, Uttar Pradesh. J Family Med Prim Care. 2024;13(2):780–6. 9. Lew CCH, Compher CW. Malnutrition and nutritional assessment. In: Loscalzo J, Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL, editors. Harrison's principles of internal medicine. 22nd ed. New York: McGraw Hill; 2025. Chapter 345. 10. Cereda E, Pedrolli C, Klersy C, Bonardi C, Quarleri L, Cappello S, et al. Nutritional status in older persons according to healthcare setting: a systematic review and meta-analysis of prevalence data using MNA®. Clin Nutr. 2016;35(6):1282–90. 11. Kang MC, Kim JH, Ryu SW, Moon JY, Park JH, Park JK, et al. Prevalence of malnutrition in hospitalized patients: a multicenter cross-sectional study. J Korean Med Sci. 2018;33(2):e10. 12. Ruiz AJ, Buitrago G, Rodríguez N, Gómez G, Sulo S, Gómez C, et al. Clinical and economic outcomes associated with malnutrition in hospitalized patients. Clin Nutr. 2019;38(3):1310–6. 13. Bellanti F, Lo Buglio A, Quiete S, Vendemiale G. Malnutrition in hospitalized old patients: screening and diagnosis, clinical outcomes, and management. Nutrients. 2022;14(4):910. 14. Liu H, Jiao J, Zhu M, Wen X, Jin J, Wang H, et al. Nutritional status according to the Short-Form Mini Nutritional Assessment (MNA-SF) and clinical characteristics as predictors of length of stay, mortality, and readmissions among older inpatients in China: a national study. Front Nutr. 2022;9:815578. 15. Manjaly SP, Sebastian SA, Alex A. Malnutrition in elderly – a predictor of cognitive, functional decline, depression and prolonged hospital stay. J Med Sci Health. 2022;8(1):28–34. 16. Mukundan M, Kashyap K, Dhar M, Muralidharan A, Agarwal D, Saxena Y. Nutritional and functional status as a predictor of short-term mortality in hospitalized elderly patients in a tertiary care hospital. Cureus. 2022;14(2):e22576. 17. Wright C, Shankar B, Marshall S, Pearcy J, Somani A, Agarwal E. Prevalence of malnutrition risk and poor food intake in older adults in Indian hospitals: a prospective observational nutritionDay study with novel mapping of malnutrition risk to the Malnutrition Screening Tool. Nutr Diet. 2021;78(2):135–44. 18. Schuetz P, Fehr R, Baechli V, Geiser M, Deiss M, Gomes F, et al. Individualised nutritional support in medical inpatients at nutritional risk: a randomised clinical trial. Lancet. 2019;393(10188):2312–21. 19. Hersberger L, Bargetzi L, Bargetzi A, Tribolet P, Fehr R, Baechli V, et al. Nutritional risk screening (NRS 2002) is a strong and modifiable predictor risk score for short-term and long-term clinical outcomes: secondary analysis of a prospective randomised trial. Clin Nutr. 2020;39(9):2720–9. 20. Gomes F, Baumgartner A, Bounoure L, Bally M, Deutz NE, Greenwald JL, et al. Association of nutritional support with clinical outcomes among medical inpatients who are malnourished or at nutritional risk: an updated systematic review and meta-analysis. JAMA Netw Open. 2019;2(11):e1915138. 21. Rubenstein LZ, Harker JO, Salva 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(6):M366–72. 22. Kaiser MJ, Bauer JM, Ramsch C, Uter W, Guigoz Y, Cederholm T, et al. Validation of the Mini Nutritional Assessment short-form (MNA-SF): a practical tool for identification of nutritional status. J Nutr Health Aging. 2009;13(9):782–8. 23. Dent E, Hoogendijk EO, Visvanathan R, Wright ORL. Malnutrition screening and assessment in hospitalised older people: a review. J Nutr Health Aging. 2019;23(5):431–41. 24. Guigoz Y, Vellas B. Nutritional assessment in older adults: MNA® 25 years of a screening tool and a reference standard for care and research; what next? J Nutr Health Aging. 2021;25(4):528–83. 25. Volkert D, Beck AM, Cederholm T, Cruz-Jentoft A, Goisser S, Hooper L, et al. ESPEN guideline on clinical nutrition and hydration in geriatrics. Clin Nutr. 2019;38(1):10–47. 26. Cederholm T, Jensen GL, Correia MITD, Gonzalez MC, Fukushima R, Higashiguchi T, et al. GLIM criteria for the diagnosis of malnutrition – a consensus report from the global clinical nutrition community. Clin Nutr. 2019;38(1):1–9. 27. Ministry of Health and Family Welfare, Government of India. National Programme for Health Care of the Elderly (NPHCE) [Internet]. New Delhi: MoHFW [cited 2026 Oct 13]. Available from: https://www.mohfw.gov.in/node/575 28. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4(10):e296. 29. Daniel WW, Cross CL. Biostatistics: a foundation for analysis in the health sciences. 10th ed. Hoboken (NJ): John Wiley & Sons; 2013. 30. Nestlé Nutrition Institute. Mini Nutritional Assessment MNA® [Internet]. Vevey: Nestlé Nutrition Institute [cited 2026 Oct 13]. Available from: https://www.mna-elderly.com/ 31. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363(9403):157–63. 32. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–83. 33. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA. 1963;185:914–9. 34. Allard JP, Keller H, Jeejeebhoy KN, Laporte M, Duerksen DR, Gramlich L, et al. Decline in nutritional status is associated with prolonged length of stay in hospitalized patients admitted for 7 days or more: a prospective cohort study. Clin Nutr. 2016;35(1):144–52. 35. World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191–4. 36. Indian Council of Medical Research. National ethical guidelines for biomedical and health research involving human participants. New Delhi: ICMR; 2017. 37. Correia MITD, Waitzberg DL. The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr. 2003;22(3):235–9. 38. Sharma Y, Miller M, Kaambwa B, Shahi R, Hakendorf P, Horwood C, et al. Malnutrition and its association with readmission and death within 7 days and 8–180 days postdischarge in older patients: a prospective observational study. BMJ Open. 2017;7(11):e018443. 39. Işık AC, Karagöz A, Öcal A, Akın S, Böyük B, Keskin Ö. The relationship between nutritional status and early- and mid-term mortality of geriatric patients admitted to the emergency internal medicine unit. Med Bull Haseki. 2022;60(1):57–63. 40. Söderström L, Rosenblad A, Thors Adolfsson E, Bergkvist L. Malnutrition is associated with increased mortality in older adults regardless of the cause of death. Br J Nutr. 2017;117(4):532–40. 41. Sánchez-Rodríguez D, Annweiler C, Ronquillo-Moreno N, Tortosa-Rodríguez A, Guillén-Solà A, Vázquez-Ibar O, et al. Clinical application of the basic definition of malnutrition proposed by the European Society for Clinical Nutrition and Metabolism (ESPEN): comparison with classical tools in geriatric care. Arch Gerontol Geriatr. 2018;76:210–4. 42. Agarwal E, Miller M, Yaxley A, Isenring E. Malnutrition in the elderly: a narrative review. Maturitas. 2013;76(4):296–302. 43. Chen LK, Woo J, Assantachai P, Auyeung TW, Chou MY, Iijima K, et al. Asian Working Group for Sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment. J Am Med Dir Assoc. 2020;21(3):300–7.e2. 44. Deutz NE, Matheson EM, Matarese LE, Luo M, Baggs GE, Nelson JL, et al. Readmission and mortality in malnourished, older, hospitalized adults treated with a specialized oral nutritional supplement: a randomized clinical trial. Clin Nutr. 2016;35(1):18–26. 45. Marshall S, Young A, Bauer J, Isenring E. Malnutrition in geriatric rehabilitation: prevalence, patient outcomes, and criterion validity of the scored Patient-Generated Subjective Global Assessment and the Mini Nutritional Assessment. J Acad Nutr Diet. 2016;116(5):785–94.