Diabetes Research and Clinical Practice - Volume 120 - Supplement 1 - October 2016

89 / 244

numbers of MetS component. We would like investigate UA
according to the number of MetS components.
Methods: We randomly selected 36,169 subjects whowere over
65 years old for 10 years and 18,907 subjects, 9,732 male and
9,175 female participants had MetS, were eligible finally.
Subjects were grouped by age to be young-old (65 –74 years
old), old-old (75 –84 years old), and oldest-old (85–94 years
old). Correlations between MetS components and UA were
evaluated and multivariate linear regression analysis was
performed to confirm if UA is independently related to MetS
components.
Results: Positive correlation between UA levels and variable
numbers of MetS component in the female and male young-
old and old-old groups but oldest-old groups. There were
trends of correlations between UA levels and variable numbers
of MetS component in young-old and old-old groups, but
reversed relationships in both genders of oldest-old groups.
Conclusion: In this cross-sectional study, we found that serum
UA is correlated with the elder age with MetS and age is one of
the risk factors for increasing UA in the elder groups, but
reversed correlation in the male oldest-old group.
PA-20
The correlation between carbohydrate food intake and
nutritional knowledge, attitude, behavior and self-efficacy
among community people
Mei Yuan LIU 1*, Mei-Yu TU1. 1Chi Mei Medical Center, Taiwan
Objectives: The purpose of this study is to understand the
correlation between carbohydrate food intake and nutritional
knowledge, attitude, behavior and self-efficacy among com-
munity people and participants enrolled in hospital reduction
program. With the health intervention strategies to maintain
more balanced diet, it is anticipated that subjects can reach
long-term weight reduction goal.
Methods: Subjects who randomly selected from community
activity and participants who joined Chimei weight reduction
programwere issued a self-reporting questionnaire to fill after
a single instruction. Questionnaire included questions cover-
ing food frequency, nutrition knowledge, attitude and self-
efficacy.
Results: A final 169 valid questionnaires (response rate 84.5%)
were analyzed. The results showed a significant positive
correlation between body weight and waist circumference,
hip circumference, blood pressure, pasta-based intake, meal
frequency and fried food frequency. There were significant
negative correlations between body weight and vegetable
intake, eating attitudes, as well as eating behaviors. There
were significant positive correlation between nutrition knowl-
edge, attitude, health self-efficacy, but not between nutrition
knowledge and dietary behaviors. Eating behaviors were also
positively correlated with health self-efficacy. Among women
group, better nutrition knowledge, attitudes and behaviors
were shown along with age increase. Healthy eating behaviors
were negatively associated with frequency of dining out or
purchase of half-cooked food. We concluded that more
frequent dining out owing to busy work and school could be
a major factor hindering health eating behaviors.
PA-21
Comparison of diabetes conversion rates among Chinese
prediabetic population using different diagnosis criteria
Zhenping ZHAO1, Tengfei MAN2 , Mei ZHANG1,
Tongtong WANG3, Zhengjing HUANG1, Shusen LIU2,
Yichong LI 1, Chun LI1, Samuel S. ENGEL3, R. Ravi SHANKAR3,
Yingmei TU2, Linhong WANG1, Kimberly BRODOVICZ3,
Limin WANG 1 *. 1National Center for Chronic and Non-
communicable Disease Control and Prevention, Chinese Center for
Disease Control and Prevention, 2Merck Sharp & Dohme (China) R&D
Co., Ltd., Beijing, China; 3Merck Research Laboratories, Merck & Co.,
Inc., Kenilworth, NJ, USA
Background: The American Diabetes Association (ADA), the
World Health Organization (WHO) and the National Institute
for Health and Care Excellence (NICE) have different criteria for
diagnosing prediabetes. However, what needs to be explored is
whether these criteria result in different conversion rates to
diabetes in Chinese population in order to determine which
criteria should be used in the early screening or intervention of
prediabetes.
Objective: The study aims to compare the diabetes conversion
rates in a Chinese prediabetic population diagnosed using
different criteria.
Method: A 4-year retrospective cohort (recruting1397 sub-
jects aged 18 –89 at 4 sites) was built to estimate the rates of
diabetic conversion in subjects with prediabetes compared to
normoglycemic at baseline. Diabetes (physician diagnosis)
conversion rates were calculated in seven mutually exclusive
prediabetic subgroups, including ADA specific isolated
impaired fasting glucose (IFG; 5.6≤ FPG < 6.1 and 2hPG < 7.8);
WHO isolated IFG (6.1≤ FPG < 6.9 and 2hPG < 7.8); ADA specific
combined IFG and impaired glucose tolerance (IGT; 5.6 ≤ FPG
< 6.1 and 7.8 ≤ 2hPG < 11.1); WHO Combined IFG and IGT
(6.1 ≤ FPG < 6.9 and 7.8≤ 2hPG < 11.1); ADA isolated IGT
(FPG < 5.6 and 7.8 ≤ 2hPG < 11.1); Rest captured by ADA specific
HbA1c (5.7% ≤ HbA1c < 6.0%); and Rest captured by NICE HbA1c
(6.0%≤ HbA1c < 6.5%). These subgroups were compared
to normoglycemia group (FPG < 5.6 and 2hPG < 7.8 and
HbA1c < 5.7%) to assess the relative risk among these groups.
Results: The study found that the 4-year diabetes conversion
rate in subjects with normoglycemiawas 6.9%. Compared with
normoglycemia, the rates in ADA isolated IFG, WHO isolated
IFG, ADA specific combined IFG and IGT, WHO combined
IFG and IGT, ADA isolated IGT, ADA specific HbA1c and
NICE HbA1c subgroups were 13.8% (risk ratio, RR = 1.8; 95%CI:
1.2– 2.7), 22.5% (RR = 2.8; 95%CI: 1.6–4.7), 23.9% (RR = 2.7; 95%CI:
1.5– 4.8), 35.3% (RR = 4.2; 95%CI: 2.4–7.1), 22.6%(RR = 2.8; 95%
CI: 1.7–4.8), 8.7% (RR = 1.1; 95%CI: 0.7–1.9), and 19.5% (RR = 2.2;
95%CI: 1.3–4.0) respectively, after adjusting for age, gender, low
HDL, binge drinking and BMI.
Conclusion: The 4-year diabetes conversion rates were lower
in the ADA groups than in theWHO or NICE groups. Other than
the ADA specific HbA1c group, the rates in all the other groups
were significantly higher than normoglycemia group over 4
years. Diabetes conversion rates were lower using ADA criteria
and the ADA specific criterion diagnosing prediabetes based
on HbA1c did not identify a group at higher risk converting to
diabetes compared to normoglycemia group, possibly due to
lower IFG or HbA1c cutoff used.
PA-22
Improvement of hemoglobin A1c by a diet adherence
monitoring for diabetic inpatients ordering diabetes meals
Ling-Tsen CHIU1*, Zhuo-Tsan BAO2, Wei-Sheng CHUNG 3,
Wan-Yi CHIANG3, Hsueh-Chen LEE3, Jing-Shi LU1, Ya-Hui YU 1.
1Diet and Nutrition Department, Taichung Hospital, Ministry of
Health and Welfare, 2Endocrine and Metabolism, Taichung Hospital,
Ministry of Health and Welfare, 3Nursing, Taichung Hospital,
Ministry of Health and Welfare, Taiwan
Background: Diabetes and its associated comorbidities
increase healthcare resources utilization. Diabetes control
requires long-term management, including lifestyle changes,
effective self-care, and optimal glycemic control, to prevent
acute and chronic complications. Eating healthful meals is
pivotal for maintaining glycemic control.
Objectives: Utilize PRECEDE model as a framework to explore
the behavior of compliance and relevant parameters on the
impact of HbA1c among hospitalized diabetic patients.
Methods: We conducted a cross-sectional study on diabe-
tic inpatients by using hospital therapeutic diet and a diet
adherence monitoringin a teaching hospital between October
2015 and March 2016. In tota, 46 patients were enrolled. The
Poster Presentations / Diabetes Research and Clinical Practice 120S1 (2016) S65–S211 S71