- Email: [email protected]
P201 NUTRITIONAL RISK PROFILE OF GERIATRIC, HOME LIVING PATIENTS
Nutritional Epidemiology I P199 SCREENING ON MALNUTRITION IS MANDATORY IN DUTCH HOSPITALS E. Leistra1 , A.M. Evers1 , J.M.J. van den Berg2 , E. van der Heijden1 , C.J.J. Mulder1 , A.C. de Bruijne-Dobben2 , H.M. Kruizenga1 . 1 Dutch Malnutrition Steering Group, Amsterdam, 2 Health Care Inspectorate, Utrecht, Netherlands Rationale: In the past years, the Dutch Malnutrition Steering Group (DMG) gained attention for the problem of disease related malnutrition in The Netherlands. In the period of 2006-2009 DMG performed an implementation project on early screening and treatment of malnutrition in hospitals. The implementation objectives of this project were adapted by the Health Care Inspectorate (HCI) as part of the performance indicators for risk steering supervision. Methods: In 2007 Dutch hospitals obligatory reported on the percentage of patients screened on malnutrition. In addition, as of 2008 hospitals had to report annually on malnutrition treatment (expressed as percentage of patients with an adequate protein intake on the fourth day of admission). This abstract reports on the results of 2007 and the results of a quick evaluation (March 2009). The 2008 HCI-results can be presented during the ESPEN congress. Results: In 2007, 98 out of 100 hospitals reported on malnutrition screening to HCI. 81 hospitals had implemented systematic screening before or during 2007. Of these, 45 hospitals used the SNAQ, 13 used the MUST, and 23 used a combination of tools or another screening tool. 34 hospitals (42%) reported a percentage of screened patients of more than 60%. In total, 52% (range: 0.4 100%) of patients were screened at hospital admission. 7.0% (0.5 26%) were moderately malnourished and 18.4% (0.5 65%) were severlye malnourished. Results from the quick evaluation (n = 100) showed that 96 hospitals have implemented systematic malnutrition screening; 76 use the SNAQ, 19 use the MUST, and one uses its own tool. Conclusion: Screening and treatment of malnutrition have become mandatory performance indicators in Dutch hospitals. Hospitals are eager to score well on this indicator and we will examine the results of the coming years whether early recognition and treatment of malnutrition will further improve in Dutch hospitals. Disclosure of Interest: None declared.
P200 PREVALENCE OF UNDERNUTRITION IN DUTCH HOSPITAL OUTPATIENTS E. Leistra1 , F. Neelemaat1 , A.M. Evers2 , H.W.M. van Zandvoort3 , P.J.M. Weijs1 , M.A.E. van Bokhorst-de van der Schueren1 , M. Visser4 , H.M. Kruizenga1 . 1 Nutrition and Dietetics, VU University Medical Center, 2 Dutch Malnutrition Steering Group, Amsterdam, 3 Nutrition and Dietetics, Leids University Medical Center, Leiden, 4 Health Sciences, Faculty of Earth and Life Sciences, VU University, Amsterdam, Netherlands Rationale: The prevalence of undernutrition in hospital inpatients is high. Early detection and treatment in
109 the hospital outpatient clinic may help to reduce this problem. The aim of this study was to assess the prevalence of undernutrition in hospital outpatients in the Netherlands and to determine high risk departments. Methods: This cross-sectional multicenter study was conducted in 9 Dutch hospitals. Patients who visited the outpatient clinic on one of the study days in the period March May 2008 received a short questionnaire and height and weight were measured. Patients were classified as severely undernourished (BMI < 18.5 kg/m2 and/or unintentional weight loss 5% in the last month or 10% in the last six months), moderately undernourished (5 10% unintentional weight loss in the last six months) or not undernourished. Descriptive statistics and logistic regression analysis were used to determine prevalence and high-risk departments. Results: A total of 2288 patients (47.5% male; mean age 56.5±16.3 years) from 23 different outpatient departments were included in the study, of which 5% were severely undernourished and 2% were moderately undernourished. The prevalence of severe undernutrition was highest in the outpatient departments of oral maxillofacial surgery (17%), oncology (10%), rehabilitation (8%), gastroenterology (7%) and pulmonology (7%). Undernutrition in general (both moderate and severe) was significantly higher in the departments of oral maxillofacial surgery (OR 2.66 (1.00 7.04)) and oncology (OR 2.66 (1.60 4.42)) compared to the other departments. Only 17% of severely undernourished and 4% of moderately undernourished patients reported to receive dietetic treatment. Conclusion: The prevalence of undernutrition in hospital outpatients is generally low but largely undertreated. Future screening and treatment should focus on high risk departments. Disclosure of Interest: None declared.
P201 NUTRITIONAL RISK PROFILE OF GERIATRIC, HOME LIVING PATIENTS E. Leistra1 , O.J. de Vries2 , H.M. Kruizenga1 , M.A.E. van Bokhorst-de van der Schueren1 . 1 Nutrition and Dietetics, 2 Internal Medicine, VU University Medical Center, Amsterdam, Netherlands Rationale: Little information is available on the nutritional status of un-institutionalized, home living, elderly patients with multi-morbidity. This study describes nutritional status of this patient group in relation to sociodemographics, functional and physical health. Methods: All newly referred patients with complete data on weight and height who visited our outpatient geriatric assessment unit between 2004 and 2008 were retrospectively included. Patients were defined severely undernourished with a BMI 20 and moderately undernourished with a BMI 20.1 22. Risk profile included unintentional weight loss, B-vitamin status, age, gender, living arrangement, marital status, changes in appetite, functional status (Barthel Index and handgrip strength) and cognition (MMSE score). Results: Out of 179 new referrals, 159 had complete data on BMI (32% male, mean age 80 y); 87 were referred
110
Poster presentations
with fall incidents and 72 with general geriatric (mostly cognitive) problems. Eleven patients (7%) had a BMI < 20, another 13 (8%) had a BMI 20.1 22. Although exact data on weight loss were not documented reliably, 33 patients (21%) reported serious unintentional weight loss. Vit B1, B6 and B12 were deficient in 5.2%, 12.2% and 8.4% of the population. Significantly more unmarried people, widows/widowers or divorced had a BMI < 22 (29%) compared to married ones (9%, p = 0.022), with the highest prevalence among widows/widowers (38%). BMI < 22 was significantly associated with low handgrip strength (p = 0.047), but not with MMSE and Barthel score. Conclusion: This study indicates that undernutrition is highly prevalent among un-institutionalized, home living elderly with multi-morbidity. Being undernourished is associated with not/no longer being married. Undernourished patients had lower grip strength, indicating worse functionality. We advise to start more intensive nutritional screening and support to these patients at high risk. Disclosure of Interest: None declared.
P202 Outstanding abstract BLOODSTREAM INFECTION RATES IN PARENTERAL NUTRITON: ARE THERE DIFFERENCES FOR COMPOUNDED AND MANUFACTURED PREMIX? G. Zaloga1 , R.S. Turpin2 , K. Williams3 , R. Stephen3 . 1 Medical Affairs, 2 Health Economics, Baxter Healthcare, Deerfield IL, 3 Epidemiology, United BioSource, Lexington MA, United States Rationale: Bloodstream infection rates (BSI) with parenteral nutrition (PN) are reported at 7.2 39%, and it is unknown whether PN type impacts BSI rates. Our objective was to identify if BSI rates differ for compounded versus premix PN. Methods: We performed a retrospective analysis of data from 197 US hospitals using Premier’s database; including all patients >18 years, receiving PN Jan 05 Dec 07 (n = 69,093). Patients were either PRE (premix PN in a 2-chamber bag with glucose & amino acids; lipids & micronutritients added; n = 4692) or COM (hospital or outsource compounded PN; n = 64,401). A sub-group analysis compared PRE to patients receiving only hospitalcompounded PN (HCOM; n = 979). Chi-square, t-tests & logistic regression examined the impact of hospital (teaching, location, size) and patient characteristics (age, diagnosis, comorbidities, surgical status, ICU stay, APR-DRG severity, IV lines, days on PN) on BSI rates (ICD-9 codes 038.x, 995.91 2 & 790.7). Table: BSI rates
Unadjusted BSI rates Adjusted BSI rates
COM (n = 64,401)
PRE (n = 4692)
26.6%* 25.8%*
17.4% 20.6%
*Significantly different from PRE at p < 0.01
Results: COM patients were more likely than PRE to have major/extreme illness severity (81 v. 72%), receive more
days PN (8.1 v. 5.3), have a longer ICU (6.3 v. 3.3 days) & hospital length of stay (LOS; 19.1 v. 14.9 days); p < 0.01. The unadjusted BSI rate was 53% higher for COM v. PRE (see Table). After accounting for baseline differences, the adjusted probability for BSI was 25% higher for COM v. PRE; odds = 1.43; 95% CI = 1.56 & 1.30). The subgroup adjusted probability for BSI was 29.0% for HCOM & 16.7% for PRE (odds = 2.38: 95% CI =3.03 & 1.85). Conclusion: The COM group was more severely ill, with greater LOS & a higher unadjusted BSI rate than the PRE group. Once illness severity and other variables were accounted for, the adjusted probability of BSI remained significantly lower for premix PN than compounded PN, particularly for PN compounded in hospital facilities. Disclosure of Interest: G. Zaloga, Baxter Healthcare, Employee. R. Turpin, Baxter Healthcare, Employee.
P203 Outstanding abstract UNDERNUTRITION IN COMMUNITY-DWELLING OLDER MEN AND WOMEN: COMPARING DOSE-RESPONSE ASSOCIATIONS OF DIFFERENT ANTHROPOMETRIC MEASURES WITH MORTALITY H.A.H. Wijnhoven1 , M.A.E. van Bokhorst-de van der Schueren2 , M.W. Heymans3 , H.C.W. de Vet4 , H.M. Kruizenga5 , J.W. Twisk5 , M. Visser5 . 1 Department of Health Sciences and the EMGO Institute for Health and Care Research, VU University, 2 Department of Nutrition and Dietetics, VU University Medical Center, Amsterdam, Netherlands; 3 Department of Health Sciences and the EMGO Institute for Health and Care Research, VU University, Amsterdam, 4 the EMGO Institute for Health and Care Research, 5 Department of Health Sciences and the EMGO Institute for Health and Care Research, Faculty of Earth, VU University Medical Center, Amsterdam, Netherlands Rationale: Although low body mass index (BMI) is widely used as an indicator of undernutrition, its use in older persons is hampered by practical and validity problems. Other simple anthropometric measures that better reflect muscle mass may be stronger related to mortality and thus could be better indicators of chronic undernutrition. However, a systematic comparison is lacking. The objective of this study is to examine dose-response associations of BMI, calf, arm, hip and waist circumference with mortality in a population-based sample of community dwelling older men and women. Methods: The study population consisted of noninstitutionalized persons aged 65 y and older (n = 1677 in 1992 93) from the Longitudinal Aging Study Amsterdam. Anthropometric measures were assessed every 3-y. Generalized estimating equations with restricted cubic spline functions were used to examine the dose responseassociations of anthropometric measures with 3-y mortality (across four measurement rounds). Cox regression models with restricted cubic spline functions were used to study 15-y mortality. The influence of excluding 3-y mortality (from 15-y) and excluding (former) smokers, obstructive lung disease or cancer was examined. Results: Low mid upper arm circumference (MUAC) was most consistently associated with 3-y and 15-y mortality in both men and women. Calf circumference showed a
P200 PREVALENCE OF UNDERNUTRITION IN DUTCH HOSPITAL OUTPATIENTS E. Leistra1 , F. Neelemaat1 , A.M. Evers2 , H.W.M. van Zandvoort3 , P.J.M. Weijs1 , M.A.E. van Bokhorst-de van der Schueren1 , M. Visser4 , H.M. Kruizenga1 . 1 Nutrition and Dietetics, VU University Medical Center, 2 Dutch Malnutrition Steering Group, Amsterdam, 3 Nutrition and Dietetics, Leids University Medical Center, Leiden, 4 Health Sciences, Faculty of Earth and Life Sciences, VU University, Amsterdam, Netherlands Rationale: The prevalence of undernutrition in hospital inpatients is high. Early detection and treatment in
109 the hospital outpatient clinic may help to reduce this problem. The aim of this study was to assess the prevalence of undernutrition in hospital outpatients in the Netherlands and to determine high risk departments. Methods: This cross-sectional multicenter study was conducted in 9 Dutch hospitals. Patients who visited the outpatient clinic on one of the study days in the period March May 2008 received a short questionnaire and height and weight were measured. Patients were classified as severely undernourished (BMI < 18.5 kg/m2 and/or unintentional weight loss 5% in the last month or 10% in the last six months), moderately undernourished (5 10% unintentional weight loss in the last six months) or not undernourished. Descriptive statistics and logistic regression analysis were used to determine prevalence and high-risk departments. Results: A total of 2288 patients (47.5% male; mean age 56.5±16.3 years) from 23 different outpatient departments were included in the study, of which 5% were severely undernourished and 2% were moderately undernourished. The prevalence of severe undernutrition was highest in the outpatient departments of oral maxillofacial surgery (17%), oncology (10%), rehabilitation (8%), gastroenterology (7%) and pulmonology (7%). Undernutrition in general (both moderate and severe) was significantly higher in the departments of oral maxillofacial surgery (OR 2.66 (1.00 7.04)) and oncology (OR 2.66 (1.60 4.42)) compared to the other departments. Only 17% of severely undernourished and 4% of moderately undernourished patients reported to receive dietetic treatment. Conclusion: The prevalence of undernutrition in hospital outpatients is generally low but largely undertreated. Future screening and treatment should focus on high risk departments. Disclosure of Interest: None declared.
P201 NUTRITIONAL RISK PROFILE OF GERIATRIC, HOME LIVING PATIENTS E. Leistra1 , O.J. de Vries2 , H.M. Kruizenga1 , M.A.E. van Bokhorst-de van der Schueren1 . 1 Nutrition and Dietetics, 2 Internal Medicine, VU University Medical Center, Amsterdam, Netherlands Rationale: Little information is available on the nutritional status of un-institutionalized, home living, elderly patients with multi-morbidity. This study describes nutritional status of this patient group in relation to sociodemographics, functional and physical health. Methods: All newly referred patients with complete data on weight and height who visited our outpatient geriatric assessment unit between 2004 and 2008 were retrospectively included. Patients were defined severely undernourished with a BMI 20 and moderately undernourished with a BMI 20.1 22. Risk profile included unintentional weight loss, B-vitamin status, age, gender, living arrangement, marital status, changes in appetite, functional status (Barthel Index and handgrip strength) and cognition (MMSE score). Results: Out of 179 new referrals, 159 had complete data on BMI (32% male, mean age 80 y); 87 were referred
110
Poster presentations
with fall incidents and 72 with general geriatric (mostly cognitive) problems. Eleven patients (7%) had a BMI < 20, another 13 (8%) had a BMI 20.1 22. Although exact data on weight loss were not documented reliably, 33 patients (21%) reported serious unintentional weight loss. Vit B1, B6 and B12 were deficient in 5.2%, 12.2% and 8.4% of the population. Significantly more unmarried people, widows/widowers or divorced had a BMI < 22 (29%) compared to married ones (9%, p = 0.022), with the highest prevalence among widows/widowers (38%). BMI < 22 was significantly associated with low handgrip strength (p = 0.047), but not with MMSE and Barthel score. Conclusion: This study indicates that undernutrition is highly prevalent among un-institutionalized, home living elderly with multi-morbidity. Being undernourished is associated with not/no longer being married. Undernourished patients had lower grip strength, indicating worse functionality. We advise to start more intensive nutritional screening and support to these patients at high risk. Disclosure of Interest: None declared.
P202 Outstanding abstract BLOODSTREAM INFECTION RATES IN PARENTERAL NUTRITON: ARE THERE DIFFERENCES FOR COMPOUNDED AND MANUFACTURED PREMIX? G. Zaloga1 , R.S. Turpin2 , K. Williams3 , R. Stephen3 . 1 Medical Affairs, 2 Health Economics, Baxter Healthcare, Deerfield IL, 3 Epidemiology, United BioSource, Lexington MA, United States Rationale: Bloodstream infection rates (BSI) with parenteral nutrition (PN) are reported at 7.2 39%, and it is unknown whether PN type impacts BSI rates. Our objective was to identify if BSI rates differ for compounded versus premix PN. Methods: We performed a retrospective analysis of data from 197 US hospitals using Premier’s database; including all patients >18 years, receiving PN Jan 05 Dec 07 (n = 69,093). Patients were either PRE (premix PN in a 2-chamber bag with glucose & amino acids; lipids & micronutritients added; n = 4692) or COM (hospital or outsource compounded PN; n = 64,401). A sub-group analysis compared PRE to patients receiving only hospitalcompounded PN (HCOM; n = 979). Chi-square, t-tests & logistic regression examined the impact of hospital (teaching, location, size) and patient characteristics (age, diagnosis, comorbidities, surgical status, ICU stay, APR-DRG severity, IV lines, days on PN) on BSI rates (ICD-9 codes 038.x, 995.91 2 & 790.7). Table: BSI rates
Unadjusted BSI rates Adjusted BSI rates
COM (n = 64,401)
PRE (n = 4692)
26.6%* 25.8%*
17.4% 20.6%
*Significantly different from PRE at p < 0.01
Results: COM patients were more likely than PRE to have major/extreme illness severity (81 v. 72%), receive more
days PN (8.1 v. 5.3), have a longer ICU (6.3 v. 3.3 days) & hospital length of stay (LOS; 19.1 v. 14.9 days); p < 0.01. The unadjusted BSI rate was 53% higher for COM v. PRE (see Table). After accounting for baseline differences, the adjusted probability for BSI was 25% higher for COM v. PRE; odds = 1.43; 95% CI = 1.56 & 1.30). The subgroup adjusted probability for BSI was 29.0% for HCOM & 16.7% for PRE (odds = 2.38: 95% CI =3.03 & 1.85). Conclusion: The COM group was more severely ill, with greater LOS & a higher unadjusted BSI rate than the PRE group. Once illness severity and other variables were accounted for, the adjusted probability of BSI remained significantly lower for premix PN than compounded PN, particularly for PN compounded in hospital facilities. Disclosure of Interest: G. Zaloga, Baxter Healthcare, Employee. R. Turpin, Baxter Healthcare, Employee.
P203 Outstanding abstract UNDERNUTRITION IN COMMUNITY-DWELLING OLDER MEN AND WOMEN: COMPARING DOSE-RESPONSE ASSOCIATIONS OF DIFFERENT ANTHROPOMETRIC MEASURES WITH MORTALITY H.A.H. Wijnhoven1 , M.A.E. van Bokhorst-de van der Schueren2 , M.W. Heymans3 , H.C.W. de Vet4 , H.M. Kruizenga5 , J.W. Twisk5 , M. Visser5 . 1 Department of Health Sciences and the EMGO Institute for Health and Care Research, VU University, 2 Department of Nutrition and Dietetics, VU University Medical Center, Amsterdam, Netherlands; 3 Department of Health Sciences and the EMGO Institute for Health and Care Research, VU University, Amsterdam, 4 the EMGO Institute for Health and Care Research, 5 Department of Health Sciences and the EMGO Institute for Health and Care Research, Faculty of Earth, VU University Medical Center, Amsterdam, Netherlands Rationale: Although low body mass index (BMI) is widely used as an indicator of undernutrition, its use in older persons is hampered by practical and validity problems. Other simple anthropometric measures that better reflect muscle mass may be stronger related to mortality and thus could be better indicators of chronic undernutrition. However, a systematic comparison is lacking. The objective of this study is to examine dose-response associations of BMI, calf, arm, hip and waist circumference with mortality in a population-based sample of community dwelling older men and women. Methods: The study population consisted of noninstitutionalized persons aged 65 y and older (n = 1677 in 1992 93) from the Longitudinal Aging Study Amsterdam. Anthropometric measures were assessed every 3-y. Generalized estimating equations with restricted cubic spline functions were used to examine the dose responseassociations of anthropometric measures with 3-y mortality (across four measurement rounds). Cox regression models with restricted cubic spline functions were used to study 15-y mortality. The influence of excluding 3-y mortality (from 15-y) and excluding (former) smokers, obstructive lung disease or cancer was examined. Results: Low mid upper arm circumference (MUAC) was most consistently associated with 3-y and 15-y mortality in both men and women. Calf circumference showed a