Estimates of Lifetime Extradyadic Sex Using a Hybrid of Randomized Response Technique and Crosswise Design.

The prevalence of extradyadic sex (EDS) tends to be underestimated due to underreporting in national surveys, which use direct questioning. Self-administered questionnaires can reduce this response bias but may cause the anxiety of privacy exposure. Randomized Response Technique (RRT) can ensure participants' confidentiality under the randomized design of indirect questions to and was found to yield more valid estimates of sexual or other sensitive behaviors than direct questions. This study estimated the EDS rate among Taiwanese aged 18 years and over, using a hybrid of Randomized Response Technique and Crosswise Design (RRTCD) and the Weighted Conditional Likelihood (WCL) estimator. The data analyzed were from the 2012 Taiwan Social Change Survey, in which the answer to the innocuous question from the unrelated-question RRT of Greenberg, Abul-Ela, Simmons, and Horvitz (1969) was obtained indirectly from a demographic question related to the innocuous question. This RRTCD provided more information on the innocuous question to effectively improve the efficiency of the unrelated-question RRT of Greenberg et al. The WCL estimator was found to be more efficient than the Greenberg et al. estimator for estimating the EDS rate in terms of smaller standard errors and smaller differences in the levels of EDS across sociodemographics and extramarital-sex attitudes. Similar to those suggested in the literature, the estimated rates of EDS in two subsamples were higher among men, homosexuals, those who have or had wages, and those who accepted extramarital sex. The levels of EDS varying with sociodemographics were different between the married and the unmarried.

An alternative to unrelated randomized response techniques with logistic regression analysis.

The randomized response technique (RRT) is an important tool that is commonly used to protect a respondent's privacy and avoid biased answers in surveys on sensitive issues. In this work, we consider the joint use of the unrelated-question RRT of Greenberg et al. (J Am Stat Assoc 64:520-539, 1969) and the related-question RRT of Warner (J Am Stat Assoc 60:63-69, 1965) dealing with the issue of an innocuous question from the unrelated-question RRT. Unlike the existing unrelated-question RRT of Greenberg et al. (1969), the approach can provide more information on the innocuous question by using the related-question RRT of Warner (1965) to effectively improve the efficiency of the maximum likelihood estimator of Scheers and Dayton (J Am Stat Assoc 83:969-974, 1988). We can then estimate the prevalence of the sensitive characteristic by using logistic regression. In this new design, we propose the transformation method and provide large-sample properties. From the case of two survey studies, an extramarital relationship study and a cable TV study, we develop the joint conditional likelihood method. As part of this research, we conduct a simulation study of the relative efficiencies of the proposed methods. Furthermore, we use the two survey studies to compare the analysis results under different scenarios.

Design and sample characteristics of the 2005-2008 Nutrition and Health Survey in Taiwan.

The Nutrition and Health Survey in Taiwan (NAHSIT) 2005-2008 was funded by the Department of Health to provide continued assessment of health and nutrition of the people in Taiwan. This household survey collected data from children aged less than 6 years and adults aged 19 years and above, and adopted a three-stage stratified, clustered sampling scheme similar to that used in the NAHSIT 1993-1996. Four samples were produced. One sample with five geographical strata was selected for inference to the whole of Taiwan, while the other three samples, including Hakka, Penghu and mountainous areas were produced for inference to each cultural stratum. A total of 6,189 household interviews and 3,670 health examinations were completed. Interview data included household information, socio-demographics, 24-hour dietary recall, food frequency and habits, dietary and nutritional knowledge, attitudes and behaviors, physical activity, medical history and bone health. Health exam data included anthropometry, blood pressure, physical fitness, bone density, as well as blood and urine collection. Response rate for the household interview was 65%. Of these household interviews, 59% participated in the health exam. Only in a few age subgroups were there significant differences in sex, age, education, or ethnicity distribution between respondents and non-respondents. For the health exam, certain significant differences between participants and non-participants were mostly observed in those aged 19-64 years. The results of this survey will be of benefit to researchers, policy makers and the public to understand and improve the nutrition and health status of pre-school children and adults in Taiwan.

Nutrition and Health Survey of Taiwan Elementary School Children 2001-2002: research design, methods and scope.

The "Nutrition and Health Survey of Taiwan's Elementary School Children (2001-2002)" was to investigate the nutritional status, influential dietary and non-dietary factors, health and development, and school performance, as well as the inter-relationships among these factors. The survey adopted a two-staged stratified, clustered probability sampling scheme. Towns and districts in Taiwan with particular ethnic and geographical characteristics were designated into 13 strata including Hakka areas, mountain areas, eastern Taiwan, the Penghu Islands, 3 northern regions, 3 central regions and 3 southern regions. Eight schools were selected from each stratum using the probabilities proportional to sizes method. Twenty-four pupils were randomly selected within each school. The survey included face-to-face interviews and health examinations. Taking seasonal effects into consideration, the face-to-face interviews were evenly allocated into each of the two semesters. A total of 2,419 face-to-face interviews and 2,475 health examinations were completed. Interview data included household information, socio-demographics, 24-hour dietary recall, food frequency, dietary and nutritional knowledge, attitudes and behaviors, physical activity, medical history, oral health, pubertal development, and bone health. Health exam data included anthropometry, blood pressure, physical fitness, bone density, dental health, and blood and urine collection. SUDAAN was used to adjust sampling design effect. There were no significant differences in sibling rank and parental characteristics between respondents and non-respondents, which indicates that our survey is representative and unbiased. The results of this survey will increase our understanding on the nutrition and health status of schoolchildren and can be used to shape public health policy in Taiwan.

Bone mass status of school-aged children in Taiwan assessed by quantitative ultrasound: the Nutrition and Health Survey in Taiwan Elementary School Children (NAHSIT Children 2001-2002).

Bone health status in childhood and adolescence may be important factors influencing the attainment of peak bone mass. The Nutrition and Health Survey in Taiwan Elementary School Children 2000-2001 was carried out to evaluate the overall nutrition and health status of school children aged between 6 and 13 years. The survey was conducted using a multi-stage complex sampling scheme. Townships and city districts in Taiwan were classified into 13 strata. Bone mass measured as broadband ultrasound attenuation was taken at heel by quantitative ultrasound bone densitometry. A total of 1164 boys and 1016 girls who had complete physical examination data with ultrasound bone scan were included in the current analysis. There were no apparent differences in BUA across all strata for both genders. In both boys and girls, age, height, body weight, BMI, and intake frequencies of vegetables and fruits/juices were significantly related to BUA. Results of multivariate regression showed that age (beta=1.36, p=0.0002) and body weight (beta=0.40, p<0.0001) were significant predictors for BUA in boys, whereas in girls body weight (beta=0.47, p<0.0001), height, (beta=0.20, p=0.01), dietary phosphorus intake (beta=-0.002, p=0.038), and frequency of fruit/juice intake (beta=0.15, p=0.029) remained statistically significant. The differential effects dietary intake variables on BUA in boys and girls may be in part due to the development of puberty. It would be necessary to include levels of physical activity in future analyses for better understanding factors influencing the development of peak bone mass in Taiwanese children.

Association between unhealthful eating patterns and unfavorable overall school performance in children.

The objective of this article is to evaluate the relationship between children's unhealthful eating patterns and overall school performance. The Nutrition and Health survey in Taiwan Elementary School Children, 2001-2002, was carried out by using a multistaged complex sampling design. A total of 2,222 elementary school children who had complete data on demographics, anthropometrics, diet and lifestyle, and overall school performance were included in the analyses. Differences in characteristics between children with favorable and unfavorable overall performance were compared using t test and chi(2) test. Using factor analysis, food frequency of 22 food groups was grouped into five factors, which were used to construct dietary patterns. The association between dietary patterns and unfavorable overall performance was assessed by multiple logistic regression after adjustment for known risk factors. Prevalence of unfavorable overall performance in Taiwanese elementary school children was 7.1%. Unfavorable overall school performance was positively associated with unhealthful eating patterns, which included high intake of low-quality foods (eg, sweets and fried foods) and low intake of dairy products and highly nutrient-dense foods (eg, vegetables, fruit, meat, fish, and eggs). Children with a greater number of unhealthful eating patterns were more at risk for unfavorable overall performance in school. The study shows that children with unfavorable overall school performance were more likely to eat sweets and fried foods, and were less likely to eat foods rich in protein, vitamins, and minerals. A potential relationship between eating patterns and unfavorable overall school performance is supported by a positive relationship between frequency of food intake and food preferences in our study.

Elderly Nutrition and Health Survey in Taiwan (1999-2000): research design, methodology and content.

The purpose of the Elderly Nutrition and Health Survey in Taiwan (1999-2000) was to assess the diet, nutrition and health of persons aged 65 and above in Taiwan. A multi-staged, stratified, clustered probability sampling scheme was used in the survey. The survey population was stratified into a total of 13 strata. The four strata of "Hakka areas", "Mountain areas", " Eastern areas", and "PengHu islands" were unique in their ethnicity or geographic locations. The remaining areas of Taiwan were stratified into "Northern", "Central", and "Southern" parts with these 3 strata, then each subdivided into a further 3 strata based on population density. The household interview of the survey was arranged such that effect of seasonal variation was taken into account. A total of 1,937 persons completed the interview and 2,432 persons completed the health exam. The following data were collected: (1) Interview data : household information, basic demographics, 24 hour dietary recall, food frequency and habit, knowledge, attitudes and practice, medical history, 36-item Short Form for generic health status, and physical activity. (2) Health exam data: blood sample for measurement of nutritional biochemical indicators and complete clinical chemistry profile, urine sample for urinary electrolytes, anthropometric measurements, ECG, blood pressure, body temperature, pulmonary function, and an osteoporosis assessment. Data from the survey were analyzed using SUDAAN to adjust for the design effect and to obtain unbiased estimates of the mean, standard error and confidence intervals. Survey respondents were slightly younger compared to non-respondents; however, after weighting and adjustment with SUDAAN, the education levels and ethnicity of respondents and non-respondents were similar indicating lack of bias. We anticipate that the results of this survey will be of benefit in understanding the nutritional status of the elderly, the relationship between nutrition and health, and factors influencing elderly persons' nutritional status. Furthermore, this information could be used in the development of public health nutrition policy aimed at improving the nutrition and health of the elderly in Taiwan.

Bone health status of the elderly in Taiwan by quantitative ultrasound.

Bone density of the elderly in Taiwan was assessed by quantitative ultrasound bone densitometry of the heel in the Elderly Nutrition and Health Survey in Taiwan (1999-2000). Broadband ultrasound attenuation (BUA) was measured, and the corresponding Z-score was calculated. Physical examination data of 1123 males and 1034 females were included in the current analysis, and data of 752 males and 721 females who also had complete questionnaire information were used for analysis of relationships between several risk factors and BUA status. The results show that in elderly Taiwanese males, higher BMI and intake of dietary calcium is positively associated with a higher BUA Z-score. Advancing age and living in the second stratum in the southern areas appeared to be negatively associated with BUA Z-score in elderly females. BMI, height, years of education, and intake of dietary calcium were positive predictors of BUA Z-score. Further analysis was performed by grouping subjects according to the gender-specific medians of intake levels of dietary calcium, protein, and sodium. The results revealed that for both genders, those in the "high calcium/high protein" group had a higher mean BUA Z-score. The results of the current analysis show that in Taiwan, BMI and dietary calcium intake are positive predictors of BUA Z-score in elderly males, whereas BMI, height, years of education, and dietary calcium intake are positively associated with BUA Z-score in elderly females. The effects of dietary calcium intake may be influenced by the intake of other nutrients such as sodium and protein.

Obesity in the elderly and its relationship with cardiovascular risk factors in Taiwan.

OBJECTIVES: The obese elderly are at increased risk of mortality, morbidity, and functional disability. In this study, we examined the prevalence of obesity and relationship between various anthropometric indices (AI) and cardiovascular disease (CVD) risk factors in the elderly. RESEARCH METHODS AND PROCEDURES: A stratified multistage clustered sampling scheme was used in the Elderly Nutrition and Health Survey in Taiwan during 1999 to 2000. 2432 non-institutionalized subjects (age, 72.8+/-9.4 years; BMI, 23.6+/-6.4 kg/m2) were recruited. The receiver operating characteristic analysis was used to compare predictive validity of CVD risk factors among various AI, including BMI, waist circumference (WC), and waist-to-hip ratio (WHR). RESULTS: The prevalence of obesity was 29.0% in men and 36.8% in women by obesity criteria for Asians (BMI>or=25 kg/m2) and 13.3% in men and 21.0% in women by the Taiwanese definition (BMI>or=27 kg/m2). Odds ratios of acquiring various CVD risk factors increased significantly with increment of WC, WHR, and BMI. The areas under the curve predicting metabolic syndrome were all <0.8. The cut-off values of WC corresponding to the highest sensitivity and the highest specificity in predicting various CVD risk factors were 86.2-88.0 cm in men and 82.0-84.0 cm in women, respectively. DISCUSSION: Obesity was prevalent in the Taiwanese elderly. WC was related to CVD risk factors to a greater extent than BMI and WHR. However, none of them alone was a good screening tool for CVD risk factors. Therefore, how to apply AI prudently to screen elderly for CVD risk factors needs further research.