S14 9th Congress of the EUGMS / European Geriatric Medicine 4 (2013) S7–S19

to assess the prevalence of inadequate serum vitamin D levels inEuropean women aged over 80 years.Material and methods.– Assessment of 25-hydroxyvitamin D[25(OH)D] was performed in 8532 European women withosteoporosis or osteopenia, of which 1984 were aged over80 years. European countries included France, Belgium, Denmark,Italy, Poland, Hungary, United Kingdom, Spain and Germany.Two cut-offs of 25(OH)D inadequacy were fixed: < 75 nmol/Land < 50 nmol/L.Results.– Mean (SD) age of the patients was 83.4 (2.9) years, bodymass index was 25.0 (4.0) kg/m2. Level of 25(OH)D was 53.3(26.7) nmol/L. There was a highly significant difference of 25(OH)Dlevel across European countries (P < 0.0001) with the lowest level of25(OH)D found in France [47.7 (26.4) nmol/L]. In these women agedover 80 years, the prevalence of 25(OH)D inadequacy was 88.6% and53.4% when considering cut-offs of 75 and 50 nmol/L, respectively.In the 397 (20.0%) patients taking supplemental vitamin D with orwithout supplemental calcium, the mean serum 25(OH)D level wassignificantly higher than in the other patients (65.2 (29.2) nmol/Lvs 50.3 (25.2) nmol/L; P < 0.001).Conclusion.– This study indicates a high prevalence of vitamin D[25(OH)D] inadequacy in old European women. The prevalencecould be even higher in some particular countries.

http://dx.doi.org/10.1016/j.eurger.2013.07.030

O 023Effects of CGA and multicomponentintervention on mobility of pre-frailand frail community-dwelling olderpeopleP. Tikkanen , E. Lönnroos , S. Sipilä , I. Nykänen ,R. Sulkava , S. HartikainenUEF University of Eastern Finland, JYU, Lapinlahti, Finland

Introduction.– Frailty is a condition with a high risk of disabil-ity progression, including mobility limitations. The aim of thepresent study was to assess the effects of comprehensive geriatricassessment (CGA)-based individually targeted intervention on theability to walk 400 m in pre-frail or frail and non-frail community-dwelling older people.Methods.– This population-based comparative study the Geri-atric Multidisciplinary Strategy for the Good Care of the Elderly(GeMS) was conducted in the city of Kuopio, Finland, from 2004 to2007. Present analysis included 605 community-dwelling personsaged ≥ 76 years (mean 80.9, 70% women), 314 in the interventionand 291 in the control group in 2005. Frailty status was assessed in2005 by the criteria of Fried et al. with slight modifications. Mobilitywas assessed annually by the ability to walk 400 m. The generalizedestimating equation model with binary logistic regression was usedto assess the treatment effect of the intervention on the ability towalk 400 m between 2005 and 2007.Results.– In the intervention group 55% (n = 173) and in the con-trol group 64% (n = 187) of participants were pre-frail/frail in 2005.The intervention prevented the loss of ability to walk 400 m inde-pendently among pre-frail/frail persons (OR 0.74, 95% CI: 0.59 to0.93, P = 0.01). The treatment effect was not statistically significantamong non-frail participants (OR 0.99, 95% CI: 0.68 to 1.42, P = 0.94).Key conclusion.– CGA-based individually targeted intervention waseffective to prevent mobility limitation among pre-frail/frail olderpeople.

http://dx.doi.org/10.1016/j.eurger.2013.07.031

O 024Which factors are associated withhealth related quality of life one yearafter an acute hospitalisation in anelderly patient population?

J.L. Parlevliet a, J.L. MacNeil-Vroomen a,b,J.E. Bosmans b, S.E. de Rooij a, B.M. Buurman a

a Academic Medical Center, Department of Internal Medicine,Geriatrics section, Amsterdam, The Netherlandsb VU University Amsterdam, Faculty of Earth and Life Sciences andEMGO Institute of Health and Care Research, Department of HealthSciences, Section of Health Economics and Health TechnologyAssessment, Amsterdam, The Netherlands

Introduction.– Acute hospitalization is a hazardous event for olderpeople. The impact on their perceived health related quality of life(HRQOL) remains unclear. We explored the association betweencomorbidity, geriatric conditions, and baseline HRQOL on HRQOLoutcomes one year after an acute hospitalization in older patients.Methods.– A prospective multicenter cohort study conductedbetween 2006 and 2008 in three teaching hospitals in the Nether-lands. Medical patients ≥ 65 years, acutely hospitalized ≥48 h, wereeligible. Eighteen geriatric conditions were assessed at admission,and three HRQOL outcomes (utility scores of the EuroQuol-5D (EQ-5D), visual analogue scale (VAS) and quality adjusted life years(QALY)) were also assessed at three months and one year afteradmission. Linear regression analysis was conducted to identifyvariables associated with HRQOL outcomes. Utility score at baselinewas a covariate in all analyses.Results.– In total, 535 patients were included, mean age 77.4 years,46.2% male. The number of geriatric conditions was associated withutility and VAS at one year (respectively B –0.036, 95% CI –0.05–0.02and B –1.197, 95% CI –2.01–0.38) and this had a stronger effect onHRQOL than individual geriatric conditions. Utility at baseline wasassociated with utility at one year and QALY (respectively B 0.254,95% CI 0.15–0.36 and B 0.355, 95% CI 0.23–0.48). Co-morbiditymeasured at discharge and premorbid ADL functioning were alsoassociated with QALY (respectively B –0.042, 95% CI –0.06–0.03 andB –0.019, 95% CI –0.03–0.01). People who were alive at one year hadbetter utility and VAS scores at baseline (P for both < 0.001).Conclusion.– HRQOL one year after admission is associated withthe number of geriatric conditions and utility at baseline. BaselineHRQOL measured by the EQ-5D could guide patient, family andprofessionals in the decision making process determining goals toachieve by hospitalisation.

http://dx.doi.org/10.1016/j.eurger.2013.07.032

O 025Identifying predictors for hospitaloutcomes of older patients admittedto an acute care ward: Amulti-component approach. Resultsfrom the CRIME studyS.L. De Buyser a,b,c, M. Petrovic a,b,c, Y.E. Taes a,b,c,D.L. Vetrano a,b,c, G. Onder a,b,c

a Department of Geriatrics, Ghent University Hospital, Ghent, Belgiumb Department of Endocrinology and Unit for Osteoporosis andMetabolic Bone Diseases, Ghent University Hospital, Ghent, Belgiumc Centro Medicina dell’Invecchiamento, Università Cattolica del SacroCuore, Rome, Italy

Introduction.– The identification of older patients at risk for poorhospital outcome (e.g. longer hospital stay, in-hospital mortality,institutionalization) is important to provide an effective healthcareservice. Aim of the present study is to identify factors associatedwith poor hospital outcomes in older patients admitted to an acutecare ward.Methods.– The Criteria to assess appropriate medication use amongelderly complex patients project was an observational study of1123 older patients, consecutively admitted to geriatric and inter-nal medicine acute care wards of seven Italian hospitals. Dataon demographics, anthropometrics, social factors, cognitive status(Mini Mental State Examination), psychological status (15 items

9th Congress of the EUGMS / European Geriatric Medicine 4 (2013) S7–S19 S15

Geriatric Depression Scale), medical diagnoses, geriatric conditions,and physical function (walking speed, grip strength, ADL) wererecorded. Hospital outcomes were length of stay, in-hospital mor-tality, and institutionalization.Results.– Mean age of participants was 81 years, 56% werewomen. Median length of stay was 10 (IQR 7–14) days, 41patients died during hospital stay and 70 were institutionalizedafter discharge. Multivariate linear regression analyses identi-fied metastasized cancer, renal failure, infection, number of drugsbefore admission, history of falls, and walking speed as indepen-dent predictors for length of stay. Malnutrition (BMI < 18 kg/m2),total dependence on ADL, and pressure sores were inde-pendent predictors for in-hospital mortality. Malnutrition andtotal dependence in ADL were also independent predictors forinstitutionalization.Key conclusions.– Besides medical diagnoses, physical function andmalnutrition are two important aspects that are independently pre-dictive for poor hospital outcome. Assessment of these factors mayinfluence decisions about the optimal management of healthcareresources.

http://dx.doi.org/10.1016/j.eurger.2013.07.033

O 026A disease management programintervention in elderly patients withhigh comorbidity: Results of arandomized-controlled trial(HF-GERIATRICS)C. Rodriguez-Pascual , E. Paredes-Galan ,A.L. Ferrero-Martinez , J.L. Gonzalez-Guerrero ,M. Hornillos , P. Abizanda , F. Suarez ,R. Menendez-Colino ,F. Rodriguez-Artalejo , in representation of the HF-GERIATRICSstudy groupServicio de Geriatria, Hospital Meixoeiro, Vigo (Pontevedra), Spain

Objective.– To assess the efficacy of a disease management pro-gramme (DMP) for very old patients with heart failure (HF) andsignificant comorbidity.Methods.– A multicentre randomized trial in 630 patients with HF,aged over 75 years, admitted to the acute-care units of the Geri-atrics Departments in 6 hospitals. Patients were randomly allocatedto a DMP or to usual care. The DMP was conducted by a case man-ager, and included three main components:– patient education to improve disease’ knowledge and self-care;– monitoring of clinical status;– therapeutic adherence.Main statistical analyses was performed according to the intention-to-treat principle, and used Cox regression models to examine theassociation of a DMP with hospital readmission, quality-of-life, andmortality over 12 months.Results.– Mean age was 85.6 + 5 years with a 63% of women. Inter-vention group included 279 and control group 351 patients. Therewas no differences between groups in relation to age, gender,functional or cognitive status, presence of different co-morbidities,including depression and dementia, NYHA functional class, aeti-ology, HF drugs at discharge, previous admissions due to HF.Educational intervention significantly improved disease knowl-edge and shelf care behaviour in patients allocated to interventiongroup in relation to control group. There was no significant differ-ences between intervention and control group in one-year survival,any cause or HF readmissions and quality of life.Conclusion.– A DMP that included an educational interven-tional program did not improved survival, readmissions orhealth-related quality of life in very elderly patients with highcomorbidity.

http://dx.doi.org/10.1016/j.eurger.2013.07.034

O27The Multidimensional PrognosticIndex (MPI) predicts mortality inelderly cancer patients (ECP) betterthan the traditional ComprehensiveGeriatric Assessment (CGA)V. Giantin a, C. Falci b, E. De Luca a, E. Valentini a,M. Iasevoli a, P. Siviero c, S. Maggi c, B. Martella d,G. Orrù a, G. Crepaldi c, S. Monfardini e,O. Terranova d, E. Manzato a

a Department of Medicine-DIMED, Geriatrics Section, University ofPadova, Italyb Medical Oncology Unit II, Istituto Oncologico Veneto-IRCCS, Padova,Italyc CNR Aging Section, Institute of Neuroscience, Padova, Italyd Geriatric Surgery Clinic, Department of Surgical andGastroenterological Sciences, University of Padova, Padova, Italye Program of Geriatric Oncology, Istituto Palazzolo, Fondazione DonGnocchi, Milan, Italy

Introduction and aims.– Although it is strongly recommended thatECP receive a CGA before any oncological treatment decision, thereis no consensus about the best form of evaluation. The present anal-ysis was conducted to compare the Balducci’s CGA1 (MGA) and theMPI2 as predictor of 12-months mortality in ECP.Patients and methods.– In the context of a prospective trial reportedelsewhere 2160 patients aged 69 years or more with inopera-ble/metastatic cancer received MGA1 (Balducci, Cancer Control2001), from which the MPI3 was calculated. Cohen’s kappa coef-ficient and proportion of overall agreement for MGA and MPIcategories were constructed. The area under the ROC curves (AUC)was assessed to compare MGA and MPI performance using logisticregression models, sex and age adjusted.Results.– Between April 2008 and April 2010, 160 patients, 88females (55%), mean age 79.4 years, range 69–93, entered the study.The overall mortality rate was 46.9% (75 patients) at 12 months. AtMGA, 30 patients (18.7%) were classified as fit, 35 (21.9%) as vulner-able and 95 (59.4%) as frail. One hundred and two patients (63.8%)had a low MPI, 46 (28.7%) a moderate MPI and 12 (7.5%) a severeMPI.The agreement between MGA and MPI was low, with a weightedCohen’s K of 0.17 (95% CI = 0.11–0.23, P < 0.0001).The AUC for MPI was statistically greater than AUC for MGA (0.73,95% CI 0.65–0.81 vs 0.65, 95% CI 0.56–0.73, respectively; P = 0.019).Conclusions.– The present study showed that the MPI performs bet-ter than MGA and could be used to correctly classifying 12 monthsmortality in ECP.

http://dx.doi.org/10.1016/j.eurger.2013.07.035

O 028Is there a need for a Europeangraduate exam in geriatric medicine:A cross-national surveyJ. Petermans , T. Fruehwald , J.P. Michel , D. Luttje ,R.E. Roller , on behalf of the UEMS/EUGMS working groupon postgraduate training in geriatricsUEMS G section/EUGMS, Belgium

Background.– Training of young colleagues builds the basis fordevelopment of Geriatric Medicine within the European Union.Differences between countries in the postgraduate training are sus-pected.Method.– To gather information concerning training standards,assessment methods and quality control in that matter, a stan-dardized and structured questionnaire was sent to the 28 membercountries of the UEMS-Geriatric Section.