Naslov
CONCORDANCE OF NEWLY DESIGNED CANCER-SPECIFIC MALNUTRITION SCREENING TOOL (GNP) AND WELL ESTABLISHED STANDARD NRS 2002 ; CROATIAN NESTED CROSS-SECTIONAL STUDY

Autori
Šeparović, R ; Tečić Vuger, A ; Jurić, A ; Pavlović, Mirjana ; Bajić, Ž

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, ostalo

Izvornik
Book of abstracts of 12th Central European Oncology Congress, A Best of ASCO® Meeting / - , 2016, 56-56

Skup
12th Central European Oncology Congress, A Best of ASCO® Meeting

Mjesto i datum
Opatija, Hrvatska, 22.06.2016. - 25.06.2016

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
nutritional risk screening, cancer patients

Sažetak
Prevalence of malnutrition in cancer patients can be as high as 87%. Malnutrition may affect a cancer treatment and prognosis from the very beginning , through the course of illness. Good nutritional screening has to meet two conflicting requirements: it has to be accurate (valid, reliable - with low inter-observer variability, sensitive and specific) while feasible (speedy, inexpensive and non-invasive). Complexity of this demand may be one of the reasons why nutritional assessment is routine part of cancer care in minority of hospitals, why only 50% of malnourished cancer patients are recognized and only 58% receive nutritional support. The aim of our study was to check the concordance of Good Nutrition Practice (GNP) screening tool, newly designed for continuous tracking of cancer patients, and well established and currently recommended by European Society of Parenteral and Enteral Nutrition (ESPEN) but not cancer specific Nutritional Risk Screening - 2000 (NRS 2002), designed for ad - hoc usage only. This one – center nested cross - sectional study was done on the consecutive sample of general population of patients hospitalized for at least two days at Department of Medical Oncology, University Hospital for Tumors, University Hospital Center "Sestre milosrdnice", Zagreb, Croatia, between December 2015 and February 2016. Power analysis revealed that 42 subjects were needed to achieve 80% power at significance level of 0.05 using a one – sided non - inferiority test of correlated proportions when expected, reference NRS 2000 accuracy is 88%. The maximum allowed difference in GNP accuracy of predicting NRS 2000 result, that was still considered non - inferior was set at 15%. Anticipating up to 15% of missing data, the initial sample of 50 was recruited. Total of 50 patients were included, 26 (52%) of them female, 24 (48%) male, with mean (SD) age of 58 (11.3) years, and median (IQR) baseline body mass index (kg/m2) of 25 (22 - 28). By NRS 2000 we identified 17/50 (34%) and by GNP we identified 15/48 (31%) of patients as being in need for nutritional support. Overall percentage of agreement between two assessments (true positives + true negatives) was 40/48 (83% ; 95% CI 69-92%). Overall agreement indicated by Cohen’s Kappa was 0.63 ; 95% CI 0.30- 0.82 ; p<0.001, and McNemar test didn’t indicate significant differences between two screener’s results (p=0.727). In conclusion, newly designed, cancer-specific screening tool (GNP) is concordant with NRS 2000 standard in detecting cancer patients with need for nutritional support.

Izvorni jezik
Engleski

Znanstvena područja
Kliničke medicinske znanosti

POVEZANOST RADA