Methods: One hundred and nineteen random patients were screened according to the NRS-2002 system and the energy requirements were calculated. The type of food ordered for the patients and the energy intake were determined. Results: Eighty had an NRS score of 0 2, i.e. not at nutritional risk and 36 a score 3, i.e. at nutritional risk. Among patients with NRS score 3 only 50% were ordered the correct menu. Only 50% of the patients had an energy intake <75% of their requirements. In the group of patients, who took <50% of energy requirement most of the energy came from the main courses and very little from snacks, Table 1. Table 1: The median energy intake from main courses and from snacks Energy intake
% of energy requirement from
% of calculated requirement
median, % of the three main requirement courses
>75% (n = 56) 50 75% (n = 39) <50% (n = 24)
97 64 36
21 11 6
76 54 31
Conclusion: In-patients at nutrtional risk focus should be on ordering the correct type of food for the main courses and especially on increasing the intake from snacks. References  Freil M et al. Reorganisation of a hospital catering systm increases food intake in patients with inadequate intake. Scand J Nutr 2005;50:83 8. Disclosure of Interest: None declared.
P192 SUBJECTIVE GLOBAL ASSESSMENT IS CLINICALLY MORE USEFUL THAN MINI-NUTRITIONAL ASSESSMENT IN HOSPITALISED OLDER ADULTS Y.P. Lim1 , W.S. Lim2 , T.L. Tan2 , L. Daniels3 . 1 Nutrition and Dietetics, 2 Geriatric Medicine, Tan Tock Seng Hospital, Singapore, Singapore; 3 Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia Rationale: Malnutrition is prevalent among hospitalized older adults. None of the nutritional assessment tools had been evaluated for use in Singapore. The study aimed to compare the use of Subjective Global Assessment (SGA) and Mini-Nutritional Assessment (MNA) on hospitalized older adults in Singapore. Methods: Newly admitted patients aged 60 years, who were not critically or terminally ill, were recruited from the geriatric medicine unit in Tan Tock Seng Hospital (TTSH). Nutritional status was assessed using SGA and MNA upon admission by a singe Dietitian. Nutritional status was analysed against clinical outcomes: length of hospital stay (LOS), discharge to higher level care, readmission at 3-month, modiﬁed Barthel Index (MBI) and mortality at 6-month, before and after adjustment for age, gender, race, depression, dementia, severity of illness, Charlson’s comorbidity index, number of prescribed drugs and admission MBI, using regression analysis.
Results: The sample comprised 281 patients with mean age 81.3±7.6 years; 44% male; 83% Chinese; median LOS 9 days. SGA and MNA were completed in 100% and 84% of patients, respectively. 35% and 23% were identiﬁed as malnourished by SGA and MNA, respectively. SGA-determined malnutrition was associated with LOS 11 days (OR 1.94), readmission at 3-month (OR 2.42), mortality at 6-month (OR 4.30), and MBI < 50 at 6-month (OR 2.08, all p < 0.05). MNA-determined malnutrition was associated with readmission at 3-month (OR 2.15), mortality at 6-month (OR 2.97), and MBI < 50 at 6-month (OR 5.80, all p < 0.05). After adjustment for covariates, only SGA-determined malnutrition remained predictive of LOS 11days (OR 2.45, p < 0.05). Conclusion: SGA has a higher completion rate and is better associated with clinical outcomes than MNA. Therefore SGA is a more useful nutritional assessment tool for assessing the nutritional status of hospitalized older adults in Singapore. Disclosure of Interest: None declared.
P193 EVALUATING THE VALIDITY OF FOUR NUTRITIONAL SCREENING TOOLS IN HOSPITALISED OLDER ADULTS Y.P. Lim1 , W.S. Lim2 , T.L. Tan2 , L. Daniels3 . 1 Nutrition and Dietetics, 2 Geriatric Medicine, Tan Tock Seng Hospital, Singapore, Singapore; 3 Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia Rationale: Hospitalised older adults are at risk of malnutrition. There are limited nutritional screening tools which are validated in this high risk population in Singapore. The study aimed to compare the diagnostic validity of four nutritional screening tools on hospitalised older adults in Singapore. Methods: Newly admitted patients aged 60 years, who were not critically or terminally ill, were recruited from 3 wards of a geriatric medicine unit in Tan Tock Seng Hospital (TTSH). Nutritional screening was performed by a single Diet Technician on admission using four screening tools: TTSH Nutrition Screening Tool (NST), Nutrition Risk Screening 2002 (NRS 2002), Short Nutrition Assessment Questionnaire (SNAQ), and Mini Nutritional Assessment Short Form (MNA-SF). Nutritional status was assessed using Subjective Global Assessment (SGA) as the reference standard by a single Dietitian. ROC analysis and diagnostic performance of the tools were compared with SGA to determine AUC, sensitivity, speciﬁcity, positive and negative predictive values. Results: The sample comprised 281 patients with the following characteristics: mean age 81.3±7.6 years; 44% male; 83% Chinese; 33% dementia; 35% malnourished. The TTSH NST, NRS, SNAQ and MNA-SF identiﬁed 42%, 37%, 6% and 81% as at risk of malnutrition, respectively. The sensitivity, speciﬁcity, positive and negative predictive values of the screening tools against SGA were: TTSH NST (84%, 79%, 68%, 90%, AUC = 0.87); NRS 2002 (69%, 79%, 64%, 83%, AUC = 0.78); SNAQ (17%, 100%, 100%, 69%, AUC = 0.76); MNA-SF (100%, 27%, 39%, 100%, AUC = 0.16). The optimal cut-off for the TTSH NST remained unchanged even for patients aged >85 years (AUC = 0.85).