Assessment of Malnutrition in Community Chinese Elderly : A Hidden Problem in a Developed Society

Background. Malnutrition can lead to chronic disease especially in vulnerable population. is study is to explore the extent of the malnutrition problem in community Chinese elderly and its associating factors in Hong Kong. Method. is is a cross-sectional study using structured interview with 7-day food frequency questionnaires. 1960 men and 1954 women aged 65 or above were recruited from the community. Results. Less than half of the participants ful�lled the recommended intake level of the others. Less than 10% of participants met the daily requirement of �bre, Calcium, Vitamin D, Iodine, and Copper. 22.32% of male participants and 6.21% of female participants have cholesterol intake exceeding the recommended upper level. 41.19% men and 28.99% women exceeded the upper level of fat percentage of total energy. Educational level was demonstrated to have a consistent positive association with the adequate intake in most nutrients. Single or divorced marital status is a risk factor in over-intake of saturated fatty acid inmen and under-intake in Zinc andmeat in women.Conclusion. Our study has shown the seriousmalnutrition problem in a wide range of nutrients in community Chinese elderly in Hong Kong.


Introduction
Nutrition is vital for healthy ageing.Inappropriate diet is associated with different illnesses such as cancer, stroke, ischaemic heart disease, diabetic mellitus, and osteoporosis irrespective of the sex and ethnicity [1].Previous studies have shown that the Mediterranean diet helped to reduce risk of cancer, which suggested that increasing fruit and vegetable consumption and decreasing meat intake would have a protective effect against cancer [2].Another study also showed a strong protective effect of �bre, Vitamin C, and carotene as well as positive association between dietary fat intake and breast cancer amongst the Chinese [3].A systematic review showed that diets with nonhydrogenated unsaturated fat, as the major forms of fat intake, whole grain, as the major form of carbohydrate along with high intake of fruits and vegetables, as well as adequate omega-3 fatty acid had a signi�cant protection effect against cardiovascular diseases [4].Another study in China showed that hyperlipidemia was positively associated with total energy, total fat, and saturated fatty acid intake and negatively associated with �bre intake [5].A previous local study in Hong Kong found reduced mortality with increased �sh uptake [6].Inappropriate nutrient intake is not only con�ned to developing countries, but is also common amongst elderly in high income countries [7].It is a global health issue closely related to socioeconomic transition and social disparity [7,8].Both over-and undernutrition can exist at the same time in the same country among different socioeconomic classes [7,8].is leads to differentials in mortality and morbidity in a population especially in the vulnerable groups such as elderly.
Although the importance of an appropriate diet is well documented, the prevalence of poor nutrition including a detailed account of speci�c nutrient intake is not well described in Chinese elderly populations.Appropriate health education and policies cannot be formulated without this important information.Such studies as have been carried out have found variable results.For example, a small survey done in Singapore showed that elderly do not conform to healthy eating guidelines for the major food categories [9].e only study assessing the protein intake of elderly people in a lower socioeconomic group in Hong Kong was done 20 years ago [10].
us, this study was to assess the health need in elderly people by documenting the extent of the malnutrition problem in a community of Chinese elderly and study the associated factors in Hong Kong.

Study Design and Participants.
Two thousand men and two thousand women aged 65 or above were invited from the community by self-referral in health talks and posting recruitment notice in community centers for the elderly and housing estates in Hong Kong.All participants are Chinese and community dwelling.ey were recruited and followed up for 5 years with questionnaires and health checks including blood taking, blood pressure monitoring, cognitive function assessment, and bone density measurement.

Measurements.
Sociodemographics including age, educational level, marital status, and medical history were recorded by structured questionnaires administered by trained staff.Socioeconomic status was measured by a selfrated 10-point ladder scale.eir body mass index (BMI) was measured.
Dietary intake was assessed by a 7-day food frequency questionnaire with mean nutrients per day calculated using the table derived from McCance and Widdowson and the Chinese Medical Science Institute [11,12].e food frequency questionnaire consisted of seven categories of food items including bread/pasta/rice, vegetables, fruits, meat, beverages, dimsum/snacks, soups, and oil/salt/sauce.Food items in each category were chosen based on the Chinese diet in the Hong Kong population.Methods of cooking were also recorded.Since a previous study had found that variation between seasons was not a signi�cant factor, we conducted interviews throughout the year regardless of the season [13].Nutrients and food items of interest are Vitamin A, Vitamin B1, B2, and B3, Vitamin C, Vitamin D, Calcium, Phosphorus, Iron, Zinc, Iodine, Copper, protein, carbohydrate, fruit, vegetable, meat, �bre, cholesterol, fat percentage of total energy, saturated fatty acid, monosaturated fatty acid and polyunsaturated fatty acid, and total energy.e subjects were asked to �ll in the food items consumed, portion size, and number of times consumed per day, per week, or per year for less frequently consumed items.Portion size was explained to the participants by pictures and photographs.e amount of oil used in cooking is estimated by the formula used in another local study depending on the method of cooking: 0.2 tablespoon (tbsp) for steaming �sh or stir frying half a portion of vegetables; 1 tbsp for stir frying one portion of meat or one portion of vegetables [14].Validity of this food frequency questionnaire was validated by comparing the 24-hour intake of total energy, lipid, sodium, and potassium, with estimated values of total energy expenditure, plasma lipid level, and 24-hour urinary sodium, and potassium outputs.ere were no signi�cant differences observed in energy intake but there was an underestimation of sodium and potassium.e lipid intake is highly associated with the plasma lipid level [14].
2.3.Statistical Analysis.e mean dietary intake of each nutrient was calculated and compared with the intake level recommended by Department of Health, Hong Kong, and we have used the United States recommended levels when local standards were not available.e percentage of participants under or exceeding the recommended level was calculated.e adjusted odds ratio by multiple logistic regression was used to identify the nutritional risk factor of various sociodemographics and BMI of our subjects in different nutrients.

Results
One thousand nine hundred and sixty men and one thousand nine hundred and ��y four women aged 65 or above were recruited from the community.eir mean BMI was not signi�cantly different from the Hong Kong population average in this age group (23 kg/m 2 ).e characteristics were shown in Table 1.e mean nutrient intakes of the subjects were shown in Tables 2 and 3. Most participants met the recommended requirement of Vitamin A; Vitamin C, Phosphorus, Iron, protein, and carbohydrate.ere was no signi�cant difference in men and women.A small percentage of participants exceeded the upper recommended level, most commonly for Vitamin A for which 3.3% of men and 2.3% of women exceeded the upper intake level of 3000 g.
Less than half of the participants ful�lled the recommended intake level of the other nutrients.e most serious ones were Vitamin D, Iodine, and Copper for which none of the participants met the recommended daily intake.Moreover, less than 5% of subjects have taken enough Calcium and less than 10% of participants met the daily requirement of �bre.Less than half of the participants met the recommended daily intake of the following nutrients: Vitamin B1, Vitamin B2, Vitamin B3, Zinc, fruit, vegetable, and meat.ese �ndings show that in most cases, more elderly women are undernourished than men.
Vitamin B3 intake showed the greatest variation, and despite less than half of the participants reaching the recommended level, up to 10% of subjects exceeded the upper recommended level (12.66% in men and 6.01% in women).
Fat intake was also suboptimal.Hong Kong elderly overconsumed fat.e situation in men is worse than in women.22.32% of male participants and 6.21% of female participants had cholesterol intakes exceeding the recommended upper level.41.19% men and 28.99% women exceeded the upper level of fat as a percentage of total energy.e picture is better for saturated fatty acid as a percentage of total energy.Only 5.11% of men and 2.4% of women exceeded the upper recommended level.
Logistic regression analysis of socioeconomic factors such as age, socioeconomic status, educational level, marital status, and BMI found that only educational level had a consistently positive association with adequate intake of most nutrients (Table 4).Single or divorced marital status is a risk factor in over-intake of saturated fatty acid of total energy in men.It is also a risk factor of under-intake in zinc and meat in women (Table 5).ere is no association between BMI, age, living and socioeconomic status, and nutritional status.erefore, body weight cannot re�ect whether the subject is having the appropriate diet or not.e undernutrition was more serious when age advanced in both men and women (Tables 6 and 7).

Discussion
Our results showed that malnutrition exists amongst the elderly in Hong Kong Chinese elderly.ey have a low intake of �bre, Calcium, and Vitamin D. On the other hand, they have an overconsumption of fat and cholesterol.eir diets are not appropriate for their needs, possibly contributing to the burden of chronic disease and/or poor control of existing illnesses such as osteoporosis, cardiovascular disease, and high fat diet-related cancer.is is re�ected by the ongoing upward trend in age standardized cancer rate in breast cancer and male colorectal cancer which are most closely contributed by inappropriate dietary intake [15].e International Osteoporosis Foundation also pointed out that the hip fracture rate in Hong Kong has increased by 300% from the 1960s to the 1990s and now costs more than 17 million USD a year in acute hospital care [16].e coronary heart disease has also risen from 38.6% in 1972 to 59.6% in 1992 [17].It is also now the second commonest cause of death in Hong Kong.Comparison with another local study done 20 years ago showed that the protein intake has greatly improved in both men and women re�ecting the increasing a�uence of our society [10].
Our study showed that people with lower education level, single/divorced/widowed, or male gender are at higher nutritional risk.ese �ndings were compatible with another study of the Chinese population in Singapore [9].However, there is no increased risk with advanced age and or with lower socioeconomic level which is at variance with other studies, possibly re�ecting the greater socioeconomic homogeneity in Hong Kong.Another study that showed BMI has a positive association with the nutritional status.However, in our study, there is no association.is is perhaps because the energy from fat is quite high in our subjects and therefore the BMI cannot be used as a proxy measurement of overall nutrition.
Factors contributing to the nutritional imbalance are complex.A previous study has demonstrated that Chinese elderly have poor nutritional knowledge and they held strong traditional Chinese or food-texture-related dietary restriction attitudes [18].is study also showed elderly of younger age and higher educational level had better nutrition knowledge.erefore, health and nutrition liturgy is very important in strategies for promoting health in the elderly.We did not ask the source of food.However, in Chinese culture, elderly people usually cook and eat at home.erefore, a health promotion programme targeted at healthy home or healthy recipe can help solving this problem in elderly.
Our study has its own limitations.e use of food frequency questionnaires has the merit of alleviating the memory burden amongst the elderly and can be used to assess long-term diet and is less in�uenced by daily variation.e tool has also been validated and used in nutrition studies in Chinese populations.However, it cannot cover all the individualized items of food and therefore does not capture some extreme diet range.Our study recruitment was not done by random sampling although we tried stratifying our subjects to cover a wider range of age and their BMIs are no different from the mean values of Hong Kong elderly.
Our subjects are all ambulatory and therefore they cannot represent the elderly who live in institutions.Moreover, recruitment by self-referral usually will attract more health conscious and healthier subjects.
Our study has shown that there are nutritional dietary imbalances including under-and overconsumption of a wide range of nutrients in Hong Kong elderly.Diets have become westernized and rates of noncommunicable diseases increased.Our study has identi�ed the health need of malnutrition problem in the elderly, and it also revealed that this problem existed across different socioeconomic classes.Education has played much in�uence on this health problem and therefore health promotion strategy should be adjusted according to the different educational levels.Health services should be targeted at promoting better diets and tackling barriers to healthier diet decisions.
��n��c� �f �n�eres�s e authors declare that they have no con�ict of interests.
T 1: Characteristics of subjects.
*  value of -test for continuous and  2 for categorical variables.
T 2: Median (IQR) nutrient intake of the male participants.T 4: Crude odds ratio and adjusted odds ratio of educational level in meeting nutrient recommendation.T 5: Marital status as a risk factor in overintake of saturated and monounsaturated fatty acid of total energy and underintake of zinc and meat.
c Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids, Aug. 3, 2000.d Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride, Jan. 1, 1997.e 2330 for age 65-74, 2100 for age over 74, dietary reference values for food, energy, and nutrients for united kingdom, 1991.f Dietary reference intakes for energy, carbohydrate, �bre, fat, fatty acids, cholesterol, protein, and amino acids, sep.5, 2002.Note: e reports for references a-f may be accessed via http://www.nap.edu/.Copyright 2002 by the National Academies.All rights reserved.g Eat Smart!Follow the Food Pyramid!, Central Health Education Unit of the Department of Health, Hong Kong, 2005.h Fats and oils in human nutrition, Food and Agriculture Organization of the United Nations and the World Health Organization, Rome, 19-26 Oct. 1993.c Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids, Aug 3, 2000.d Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride, Jan 1, 1997.e 1900 for age 65-74, 1810 for age over 74, Dietary Reference Values for Food, Energy and Nutrients for United Kingdom, 1991.f Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids, Sep 5, 2002.Note: e reports for references a-f may be accessed via http://www.nap.edu/.Copyright 2002 by e National Academies.All rights reserved.g Eat Smart!Follow the Food Pyramid!, Central Health Education Unit of the Department of Health, Hong Kong, 2005.h Fats and oils in human nutrition, Food and Agriculture Organization of the United Nations and the World Health Organization Rome, 19-26 October 1993.