| Background:During the latest 20 years, Chinese residents have undergone huge changes in dietary patterns. As a result, the prevalence of nutrition related chronic diseases saw a rapid increase correspondingly. It is therefore meaningful and important to understand the trend and factors of dietary patterns, which will facilitate to identify the sensitive period of intervention and high-risk populations, as well as prevent the development of nutrition related diseases subsequently. However, nowadays, studies on dietary patterns are unusual, besides, most of the accomplished studies mainly on the basis of cross-section surveys. Therefore, studies focus on longitudinal trends of dietary patterns and their factors are needed urgently. Plus, studies on the relationship between dietary patterns and hypertension were reported seldomly.Objective:To understand the characteristics, changing trends, and factors of Chinese dietary patterns, and analysis the influences of diet related factors and non-diet related factors on the development of hypertension. To provide scientific evidences on improving Chinese adults'dietary structures and preventing hypertension.Methods1. SubjectiveThis study was relied on the databank of China Health and Nutrition Survey, which is still ongoing now and conducted corporately by Chinese Centers for Disease Control and Prevention and North Carolina University from 1989-2009. Total 19508 adults aged 18 to 75 years were investigated from 1997-2009. 2. MethodsFirstly, urbanization index(UI) were constructed as the indicator of evaluating the level of community development. Metabolic Equivalent per week and average household income adjusted by Consumer Price Index(CPI) in 2009 were calculated respectively. On the basis of these classification, food and nutrients intake and their trend was analyzed descriptively.Secondly, the exploratory factor analysis was utilized in deriving dietary patterns, as well as the pattern scores were calculated according to gender. Correlation analysis was used to identify the relationship between pattern scores and intake of nutrients. To explore the influence of factors within community level and individual level, two-level random intercept liner regression model was initiated.Subsequently, simplified exploratory factor analysis scores were calculated across total five waves of survey. Three-level random intercept and slope logistic regression model was used to study the relationship within factors at community, individual, and time varying levels.Results1. Intake and changing trend of foodFrom 1997 to 2009, the consumption of cereal declined from 478.3g/d to 400.8g/d, and the consumption of legumes declined from 33.5g/d to 30.1g/d, and the consumption of vegetables declined from 325.8g/d to 323.3g/d, and the consumption of fruits increased from 18.5g/d to 57.5g/d, and the consumption of meats increased from 478.3g/d to 94.6g/d, and the consumption of milk increased from 3.7g/d to 14.1g/d, and the consumption of eggs increased from 24.4g/d to 30.7g/d, and the consumption of aquatic product increased from 26.8g/d to 34.4g/d, and the consumption of edible oils decreased from 40.1g/d to 38.4g/d, as well as the consumption of table salt decreased from 12.7g/d to 8.9g/d。2. Intake and energy and source of energyEnergy intake declined across four areas, namely urban, suburb, town and village. Males in these four areas consumed 2182.9kcal/d,2329.9kcal/d,2276.7kcal/d and 2462.2kcal/d respectively in 2009. Females in these four areas consumed 1812.8kcal/d,1898.0kcal/d,1927.6kcal/d and 2045.2kcal/d respectively in 2009The ratio of energy from fat amounted to 36.0%,34.0%,32.3%and 29.7% among male adults in urban, suburb, town and village respectively in 2009. The ratio of energy from fat amounted to 36.2%,34.7%,33.3% and 29.8% among female adults in urban, suburb, town and village respectively in 2009. The ratio of energy from protein were 13.6%,12.9%,12.4% and 11.7% among male in forementioned four areas and these ratios were 13.8%,13.25,12.5% and 11.9% among female respectively.3. Dietary patterns among adultsTotal five dietary patterns were derived among male. The common pattern was characterized by higher intakes of rice, pork, fished and vegetables. The healthy pattern was characterized by eggs, fruits, fungus, milk and other cereals. The snack-fast-food pattern was characterized by snacks, fast-foods, milk and beverages. The meat pattern was characterized by intake of various red meat and meat products. The alcohol pattern was characterized by beer, wine and other alcohol drinks as well as nuts. Among females, the sober pattern was characterized by flour and eggs. The fruits and milk pattern was characterized by fruits, milk, nuts, beef, mutton, fishes and eggs. The meat pattern was also characterized by different kinds of meat and the meat products. And the common pattern was characterized by vegetables, rice, pork and fishes.4. Correlation between dietary patterns and nutrients intake4.1 Male dietary patternsWith the increase of male common pattern score, the intake of fat, retinol, Vitamin A, and Calcium went up and Vitamin E went down. Intake of energy within the fourth quartiles and first quartiles was 2617.8kcal/d and 2389 kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 34.4% and 27.8% respectively.There was significant positive correlation between health pattern scores and nutrients, so did the intake of energy. Intake of energy within the fourth quartiles and first quartiles was 2511.7kcal/d and 2233.9 kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 32.5% and 30.6% respectively.With the increase of male snack and fast-food pattern score, the intake of energy, protein, fat, carbohydrate, retinol, Vitamin C, Vitamin E, and Potassium decreased. Intake of energy within the fourth quartiles and first quartiles was 2254.2kcal/d and 2699.7kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 33.8% and 28.7% respectively.With the increase of meat pattern score, the intake of energy, protein, fat increased too, whereas the intake of carbohydrate decreased. Intake of energy within the fourth quartiles and first quartiles was 2545.5kcal/d and 2460.7kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 35.1% and 28.2% respectively.With the increase of alcohol pattern score, the intake of energy, protein, fibre, Vitamin C, Vitamin E, Potassium and Sodium increased too. Intake of energy within the fourth quartiles and first quartiles was 2622.3kcal/d and 2393.7kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 31.9% and 32.1% respectively.4.2 Female dietary patternWith the increase of sober pattern score, the intake of fat, retinol, Vitamin C and Calcium decreased too. Intake of energy within the fourth quartiles and first quartiles was 2036.0kcal/d and 2165.9kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 28.2% and 33.0% respectively.With the increase of fruit and milk pattern score, the intake of energy, protein, fat, fibre, Vitamin C, Vitamin E, Potassium, Sodium and Fe increased too, whereas intake of Carbohydrate decreased. Intake of energy within the fourth quartiles and first quartiles was 2083.8kcal/d and 1900.5kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 34.0% and 28.9% respectively.With the increase of meat pattern score, the intake of fat increased, whereas the intake of Carbohydrate and Vitamin E decreased. Intake of energy within the fourth quartiles and first quartiles was 2027.1kcal/d and 2069.8kcal/d respectively. The ratio of energy from fat within the fourth quartiles and first quartiles was 34.9% and 28.6% respectively.With the increase of common pattern score, the intake of energy, protein, fat, carbohydrate, and retinol, Vitamin C, Vitamin E, Potassium and Fe increased too. Intake of energy within the fourth quartiles and first quartiles was 2291.2kcal/d and 1763.4kcal/d. The ratio of energy from fat within the fourth quartiles and first quartiles was 31.6% and 32.1% respectively.5. Factors of dietary patternsAdjusted by other variables, male common dietary pattern manifested positive correlation with urbanization index(UI) and household income and adversely correlated with age. Health pattern score positively correlated with UI, household income, education and total intake of energy, whereas negatively correlated with age. Snack and fast-food pattern score correlated positively with UI and work-related physical activity (WPA) as well as negatively with age and total intake of energy. The meat pattern score was correlated positively with UI, household income, drink, WPA and total intake of energy but negatively with age. And the alcohol pattern score was correlated positively with age, household income and total energy intake.Among the females, adjusted by other variables, the sober pattern score was correlated with education positively, whereas the score was correlated with UI and total energy intake adversely. The fruits and milk pattern score correlated positively with UI, household income, education, WPA and total intake of energy. The meat pattern score was correlated positively with UI, education, occupation, and total intake of energy, as well as negatively with age. And the common pattern score was correlated with UI, household income, education, WPA and total intake of energy.6. The relationship between dietary patterns and hypertensionAmong the dietary patterns, the common pattern and health pattern were protective factors for hypertension. For male common pattern, compared with the first quartiles, the forth quartiles had 0.841 probability of suffering from hypertension (95%CI:0.686-0.953), and the third quartiles had 0.787 probability as well (95%CI: 0.583-1.063). For the health pattern, take the first quartiles as reference group, the third quartiles had 0.852 probability of suffering from hypertension (95%CI: 0.755-0.962), and the forth quartiles had 0.746 probability of suffering from hypertension (95%CI:0.632-0.881)For the females, the fruit and milk pattern was benefit for prevention of hypertension. Compared with those in the first quartiles of fruit and milk pattern, female within the third quartiles had a lower probability of suffering from hypertension (OR=0.770,95%CI:0.626-0.948), meanwhile, the probability of hypertension of female in the forth quartiles was 75% of those in the first quartiles (OR=0.754,95%CI:0.597-0.932). For female common pattern, compared with those in the first quartiles, the females in the forth quartiles had 76% probability of suffering from hypertension (95%CI:0.589-0.977). For the sober pattern, the females in the higher quartiles were more probably suffer from hypertension. Compared with those in the first quartile, the females in forth quartile had more probability of suffering hypertension(OR=1.209,95%CI:1.014-1.443). So did the female in the third quartile(OR=1.236,95%CI:1.016-1.503).7. The non-dietary factors and hypertensionAdjusted by other factors, from 1997 to 2009, both male and female had higher probability of hypertension with the time was gone. The odd ratio of male and female was 1.280 (95%CI:1.200-1.366) and 1.165 (95%CI:1.086-1.251) respectively.The probability of hypertension went up with the increase of age. For male, the odd ratio was 1.064 (95%CI:1.052-1.077). For the females, this odd ratio was 1.095 (95%CI:1.078-1.113)There was no significant difference of the probability of suffering from hypertension among the females between any two levels of household income. However, the probability of suffering from hypertension among the females in high level of household income was 1.4 times more than that of low level of household income.The males graduated from junior middle school had 0.7 probability of suffering from hypertension than those graduated only from primary school. However, this result was not found with females.Both male and female, there was not any significant difference among the various levels of occupations.Females with low level of work-related physical activities had a higher probability of suffering from hypertension than those with middle level of work-related physical activities (OR=0.689,95%CI:0.517-0.917). There was no difference for males across level of work-related physical activities.Male drinker had higher probability of hypertension than those never drink alcohol (OR=1.219,95%CI:1.012-1.468),whereas this result was not found in female.Smoke had no correlation with hypertension within both male and female.BMI correlated to hypertension positively. Those who were overweight or obese had a higher probability than those with normal body weight, and the odd ratio were 2.421(95%CI:1.970-2.974) and 6.456(95%CI:4.563-9.133) respectively. For the females, the result were similar and the odd ratio was 2.090(95%CI:1.678-2.603) and 5.755(95%CI:4.156-7.967) respectively.Urbanization index had no relation with male probability of hypertension. However, for those female lived in community of UI>80 had a lower probability of suffering from hypertension than those lived in community of UI＜50.Adults from different community had different overall probability of suffering from hypertension((?)male=0.293, P＜0.001, (?)ofemale=0.433, P＜0.001), and the slope of change for male and female was different. The lower of overall hypertension risk, the bigger change of slope was found with time.(male:(?)2(?)-time=-0.177,P<0.001; female:(?)-time.=-0.248, P<0.001)。The probabilities of suffering from hypertension among the males across communities were different. The odd ratio of males whose intake of table salt was>12g and<6g in the community(UI was within 50～65) was 1.9 times higher than that of males who lived in community(UI<50).The probability of suffering from hypertension increased with the incline of urbanized index. The probability of suffering from hypertension among females from community of over 80 was 1.1 times more than that among females from community of less than 50 (95%CI:1.078-1.113). And together with the changing of time, the changing rate of probability of hypertension was higher among females from community of over 80 UI than that among females from community of less than 50UI (OR=1.401,95%CI:1.020-1.052)ConclusionChinese adult aged 18～75 had better dietary quality during 1997-2009 than before. However, the outstanding problem of higher intake of animal food and lower intake of plant food, as well as higher ratio of energy from fat to total energy.There were different characteristics within dietary patterns among male and female and the factors of dietary patterns and their trend were various as well. Therefore, different situation should be considered when dietary interventions were conducted for male and female.Dietary pattern had close relationship with hypertension. The occurrence of hypertension presented the characteristic of accumulation within communities or areas. The hypertension was correlated to age, education, household income, work-related physical activity and drink.Compared with single nutrients analysis, dietary pattern analyses have more advantages in understanding the real dietary intake and their relationship with diseases. Multilevel model is useful in analyzing hierarchal data, and the result can benefit the formulation of nutrition improvement policies and strategies.
|
PDF Full Text Request