SUN-PO049 IMPACT OF NUTRITIONAL STATUS ON DISCHARGE DESTINATION IN OLDER HOSPITALIZED PATIENTS K. Miyakoshi 1* , K. Sato 1 , Y. Ueta 2 . 1 Department of Rehabilitation Medicine, 2 Department of Nutrition, Kameda General Hospital, Chiba, Japan * Corresponding author. Rationale: Prediction of discharge destination is important to provide effective treatment plan. But the prediction is sometimes difficult because various factors can affect the outcome. Malnutrition is considered to be one of important factors affecting the outcome. We report the relationship between the nutritional status and discharge destination by retrospective chart review. Methods: The subjects were consecutive patients who were admitted to our hospital, from April 2017 to march 2018. Study inclusion criteria were age 65 years or older. Our hospital is an 857-bed acute care community hospital accredited by Joint Commission International. We collected the following variables from the medical records retrospectively. The candidates for prognostic factors were age, gender, hospitalized ward (medical or surgical), diagnosis (cancer or others), and mini nutritional assessment short-form (MNA-SF) on admission. The outcome measure was discharge destination (home or nursing facilities). Results: The participants were 8315 patients. The mean agewas 76.4 years, mean MNA-SF was 9.9 and mean length of hospital stay was 14.6 days. A total of 86.6% of patients were discharged home. All variables were statistically significant in univariate analysis. In the logistic regression analysis, following factors were statistically significant: diagnosis (non-cancer) (OR6.800, 95%CI 5.801–7.972), hospitalized ward (OR2.951, 95%CI 2.512–3.467), neuropsychological problems (OR1.996, 95%CI 1.734–2.297), food intake decline (OR1.442, 95%CI 1.291–1.611) and bodymass index (OR1.122, 95%CI1.050–1.1.198). Conclusions: Statistically significant correlation exists between MNA- SF and discharge destination. Nutritional status should be taken into consideration in order to provide an effective treatment plan. Disclosure of Interest: None declared. SUN-PO050 TWO FEASIBILITY STUDIES AMONG GERIATRIC PATIENTS: (FS1) PHOTO DOCUMENTATION OF FOOD AND FLUID TO ASSESS ENERGY & PROTEIN INTAKE – (FS2) CONTENT OF REFRIGERATOR AS PREDICTOR FOR READMISSION M.O. Cramon 1,2* , I. Raben 1,2 , A.M. Beck 3 , J.R. Andersen 1,1 . 1 Department of Nutrition, Exercise and Sports, University of Copenhagen, Copenhagen, 2 Department of Medicine, Zealand University Hospital, Køge, 3 Institute for Nursing and Nutrition, Faculty of Health, University College Copenhagen, Copenhagen, Denmark * Corresponding author. Rationale: (FS1) It can be a challenge for patients to register diet and fluid. Geriatric patients are particularly challenged due to low functional ability. Objectives: To investigate if geriatric patients had the desire and ability to take photos and if photo could be used to assess energy and protein intake. (FS2) An older study showed an empty refrigerator increased the risk of readmission 1 . Objective: To investigate, by means of photo, if refrigerator contents 3 days after discharge were predictive for readmission. Methods: (FS1) In a RCT the intervention group of geriatric patients (n = 21) were asked if they would take photos prior to and after food and fluid intake 2 × 2 days and SMS/mail the photos prior to home visits. Control: a 48-hour diet interview conducted on visiting day. (FS2) Permission to take photos of the refrigerator contents was asked at the first home visit. The contents were evaluated adequately with ≥3 perishable durable food items. Statistical analysis (FS1): Bland- Altmann plot, median. Results: (FS1) 15(71%) had a smartphone,11(73%) took photos and 7 (33%) sent photos. Bland-Altman plot and percentage median difference indicated a tendency to underestimate when using the photo method compared to dietary interviews with 14%(-25%>6%) energy intake and 13% protein intake (-31%>8%). Probably due to missing photos of snacks/drinks. (FS2) 17 photos of refrigerator contents, all evaluated adequate and not predictive of readmission. Conclusions: (FS1) Geriatric patients with smartphone seem to have the desire, but in some cases lack the ability to take/send photos. Photo documentation seems to underestimate energy and protein intake compared to diet interviews. (FS2) The refrigerator content 3 days after discharge was not predictive of readmission. Reference 1. Boumendjel (2000) Disclosure of Interest: None declared. SUN-PO051 BODY COMPOSITION, ENERGY AND PROTEIN INTAKE IN INSTITUTIONALIZED PORTUGUESE OLDER ADULTS M.A. Marques 1,2* , A. Faria 3 , M. Cebola 4 . 1 Faculdade de Medicina da Universidade de Lisboa, Lisboa, 2 Santa Casa da Misericórdia de Alvaiázere, Alvaiázere, 3 Nutrition and Dietetics, Escola Superior de Tecnologia da Saúde de Coimbra, Instituto Politécnico de Coimbra, Coimbra, 4 (H&TRC) – Centro de Investigação em Saúde e Tecnologia; (ESTeSL) – Dietética e Nutrição, Escola Superior Tecnologia da Saúde de Lisboa; (IPL) – Instituto Politécnico de Lisboa, Lisboa, Portugal * Corresponding author. Rationale: Adequate nutritional intake (particularly energy and protein) in older adults is crucial to maintain and prevent loss of muscle mass and further deterioration of nutritional and health status. This investigation aimed to assess body composition and its relationship with energy and protein intake in institutionalized older adults. Methods: Sociodemographic and anthropometric data was collected. Body composition was assessed through bioimpedance analysis (BodyStat 1500). Energyand protein intake was estimated through a food frequency questionnaire validated in the Portuguese population. Results: One-hundred and forty-six older adults were included (63.3% female), with a mean age of 83 years old. Mean Body Mass Index was 25.2 ± 4.8 kg/m 2 (min 14.6; max 37.5). Mean Free Fat Mass Index was 15.4 ± 3.2 kg/m 2 (min 8.2; max 22.2). Food intake was estimated, obtaining a mean energy intake of 2011.9 ± 275.7 kcal (32.4 ± 6.9 kcal/ kg Body Weight (BW)). As for protein, a mean daily intake of 76.1 ± 18.8 g (1.2 ± 0.3 g/kg BW). A strong positive correlation was found between BMI and FFMI (r = 0.670; p = 0.000). A moderate positive correlation was found between BMI and daily energy intake (r = 0.217; p = 0.037) and between BMI and daily protein intake (r = 0.363; p = 0.000). Protein intake below 1.5 g/kg BW and below 1.0g/kg was found in 89.2% and 28.4% of the sample, respectively. Conclusions: In this sample, a strong positive correlation was found between BMI and FFMI, meaning that a free fat mass height adjusted index is a better indicator of nutritional status and body composition, than percentage of free fat mass. Free fat mass declines with age, and even when a higher BMI is detected, nutritional intervention is still needed to optimize energy and protein intake, preventing further decline of nutritional and functional status of older adults. Disclosure of Interest: None declared. S76 Abstracts / Clinical Nutrition 38 (2019) S59–S296