European Geriatric Medicine - Supplement 1 - Volume 7 - October 2016

10 / 290

Results: At baseline near vision loss was present in 17.4% of subjects
with DM, 21.4% of those with undiagnosed DM, 17.6% with IFG and
12.6% in others. In cross-sectional analyses controlled for socio-
economic, lifestyle and comorbidity variables (fully adjusted model),
the risk of near vision loss was higher in subjects with diagnosed DM
and with IFG (OR = 1.38, 95% CI 1.08–1.76, p = 0.01 and OR = 1.50, 95% CI
1.07 –2.11, p = 0.02) compared to others. DM was also significantly
associated with a higher risk of incidence of near vision loss when
controlling for socio-economic factors (RR = 1.29; 95% CI 1.05 –1.56,
p = 0.02) but not in the fully adjusted model (RR = 1.21; 95% CI 0.98–
1.49, p = 0.08). In a sub-sample of subjects with diagnosis of major eye
diseases recorded at the 7-year follow-up visit, people with DM
presented more often retinopathy (p < 0.001) and near uncorrected
refractive errors than their counterparts (p = 0.01).
Conclusion: Association between DM and near vision impairment is
confirmed in the older people. Ophthalmological care has to be
maintained or reinforced in the oldest people with diabetes.
O-012
Malnutrition and hospital healthcare costs: The FRADEA Study
M. Martínez-Reig 1, I. Aranda-Reneo2, L.M. Peña-Longobardo2, J. Oliva-
Moreno2, N. Barcons-Vilardell3, P. Abizanda1. 1Geriatrics Department of
the Complejo Hospitalario Universitario de Albacete, Albacete,
2Department of Economic Analysis and Finance, Universidad de Castilla-
La Mancha, Toledo, 3Medical Affairs Department. Nestlé Health Science,
Barcelona, Spain
Objectives: To analyze the relationship between nutritional markers
and hospital healthcare costs.
Methods: Population-based prospective cohort study. 827 partici-
pants aged 70 years or older from Albacete city (Spain), stratified by
age and sex. Body mass index (BMI), biochemical parameters (total
cholesterol, total proteins and albumin), and the Mini Nutritional
Assessment®-Short Form (MNA®-SF) were collected at baseline
(2007–2009). Follow-up visits were conducted in 2012 –2013.
Generalized Linear Models adjusted for age, sex, comorbidity,
disability, and polypharmacy, were constructed to estimate the
impact of malnutrition on hospital healthcare costs per person and
year of follow-up ( €).
Results: Mean age 78 (Range 70–102), 492 women (59.5%). Mean
follow-up 1,044 days (Range 115 –2007). Mean annual total healthcare
cost per person was 1,922€. The cost associated to hospitalization was
1,292€/year of follow-up, associated to emergency visits 83€/year, and
associated with specialist visits 544€/year. Lower MNA ®-SF scores
(p < 0.001), lower albumin values (p < 0.001) and lower cholesterol
levels (p < 0.05) were associated with higher hospital costs. Older
adults with malnutrition or at nutritional risk (MNA ® -SF < 12)
presented a mean adjusted healthcare cost 714€/year (95%CI 155–
1,274 €/year; p < 0.05) higher with respect to those with normonutri-
tion. Those with albumin < 4.0 gr/dL presented a mean adjusted
healthcare cost 1,159 €/year (95%CI 407–1,910 €/year; p < 0.05) higher.
No relationship was found with cholesterol, proteins and BMI.
Conclusions: Worse nutritional status measured with the MNA® -SF,
and low albumin levels are associated with higher hospital healthcare
costs in older adults.
O-013
The efficacy on the prevention of cardiovascular attack of the
guidelines for elderly diabetics: lessons from9.2 years study of 4014
diabetic patients including 1016 elderly older than 75 y.o.
T. Hayashi1, A. Araki 2, S. Kawashima 3, H. Sone 4, H. Watanabe5, T. Ohr6,
K. Yokote7, M. Takemoto7, K. Kubota8. 1Nagoya Univ. Grad. Sch. of Med.,
2Tokyo Metropolitan Geriatric Hosp., 3Osaka Saiseikai Nakatsu Hosp.,
4Endocrinol. & Metabol. Niigata Grad. Sch. of Med., 5Dept. of Clinical
Pharmacol. & Therap., Hamamatsu Univ. Sch. of Med., 6Dept. of Geriat.
Med., Tohoku Univ. Sch. of Med., 7Div. of Diabetes, Metabo. and
Endocrinol., Chiba Univ. Hosp., 8Dept. of Pharmacoepidemiol, Faculty of
Med., Univ
Background: Elderly diabetic individuals are drastically increasing all
over the world, and the guideline for them were porposed by major
diabetic academic societies such as IDF (International Diabetes
Federation) and JDS (Japan Diabetes Society). However, those are
recently decided, and the efficacy has not been determined. Further,
elderly’ s risk of stroke is not well-known.
Methods: We performed a prospective cohort study (Japan Cholesterol
and Diabetes Mellitus Study). A total of 4,014 type 2 diabetics without
previous IHD or CVA (1,936 women; 30– 80 years, 67.4 ± 9.5 years, ≥75
years, n = 1,016) were recruited from 40 Japanese hospitals in 2004.
Lipids, glucose and other risk factors were investigated annually.
Results: Two hundred eighteen IHD cases and 138 CVAs (7.8 and 5.7/
1,000 people/year) occurred over 9.2 years. 134 patients died.
Hemoglobin A1C(HbA1C) on registration was correlated with IHD in
patients <75 years (p < 0.05), LDL-C in patients <65 and >75 years
(HR:1.028, 1.014), and HDL-C in all patients >65 years. Contrastly,
fasting plasma glucose (FPG) and HDL-C were correlated with CVA in
all generations. HbA1C was correlated with CVA in patients <75 years.
For severe nephropathy and proliferative retinopathy, FPG and HDL-C
were risk for patients >75 years. Guidelines by IDF was applied. The
patient controlled of their blood glucose within IDF guideline
significantly decreased IHD between 65 and 75 years and stroke
older than 65 years, however there are no evident guideline’ s effect of
for IHD older than 75 years. The results on JDS is almost same as IDF.
Conclusion: IHD and CVA in late elderly diabetics were predicted by
LDL-C or HDL-C. HDL-C also affects microangiopathies in elderly. These
age-dependent differences in risk and the guideline by IDF and/or JDS
are important for developing individualized strategies to prevent
diabetic complications.
Trial Registration: UMIN-CTR:UMIN00000516
O-014
Exploring psychosocial factors associated with nutritional status
among elderly living in nursing homes: Preliminary results from
the PEN-3S project
T. Madeira1, C. Peixoto1, N. Santos1, A. Bergland2, A. Bye2, T. Amaral3,
C. Lopes4, V. Alarcão1, B. Goulão 5, N. Mendonça6, P. Nicola1, J. Gorjão
Clara1. 1Instituto de Medicina Preventiva e Saúde Pública da Faculdade de
Medicina, Universidade de Lisboa, Portugal; 2Oslo and Akershus
University College of Applied Sciences, Norway; 3Faculdade de Ciências da
Nutrição e Alimentação, Universidade do Porto, 4EPIUnit–Institute of
Public Health, University of Porto, Portugal; 5Health Services Research
Unit, University of Aberdeen, 6Institute for Ageing, Newcastle University,
UK
Introduction: A comprehensive assessment of nutritional status
comprises psychosocial dimensions and these are essential for the
promotion and maintenance of healthy aging. Understanding the
interactions between nutritional status and psychosocial factors can
support the development of health protection policies and health care
equity in this age group.
Methods: This nationally representative cross-sectional study col-
lected data through face-to-face structured interviews and anthropo-
metric measurements. Nursing homes were randomly selected. All
older adults (65 years and over) without severe dementia and not
bedridden were interviewed. Nutritional status was assessed with the
Mini Nutritional Assessment (MNA®), depression with the Geriatric
Depression Scale 15, instrumental activities (functionality) with the
Lawton Scale and loneliness through the UCLA Loneliness Scale.
Results: 654 individuals (mean age 84.3 ± 6.9 years) have already
taken part in the study (60% of the target sample size). Undernutrition
is present in 5.6% (95% CI: 3.9– 7.4%) and 38.9% (95% CI: 35.2–42.7%) are
at risk of undernutrition. Nutritional status is associated with
depression, loneliness perception and functionality (p-values
<0.001). In multiple linear regression analysis, lower functionality
and depression (but not loneliness) are predictors of worse nutritional
status, adjusting for age (β = 0.367 and 0.272 respectively, R2 = 0.261, p-
values <0.001).
Oral presentations / European Geriatric Medicine 7S1 (2016) S1–S27 S4