Unequal Selection Probabilities Research Articles

Dentists' Attitudes About Their Role in Addressing Obesity in Patients: A national survey

Reducing the incidence of obesity requires coordination among primary health care providers. Because of their frequent contact with patients, dentists are positioned to recognize patients at risk of developing obesity. The authors conducted a study to assess dentists' interest in and barriers to providing obesity counseling to patients. The authors surveyed a random sample of 8,000 American Dental Association members by mail, stratified according to census region (West, Midwest, South, Northeast) and dentist type (general, pediatric). The authors weighted respondents' data to account for the unequal probability of selection and nonresponse rates among regions and dentist types. In all, 2,965 dentists responded. Overall, 4.8 percent of respondents offered a form of counseling services and 50.5 percent reported that they were interested in offering obesity-related services. More than one-half of the respondents cited fears of offending patients (53.8 percent) and appearing judgmental (52 percent) as major barriers, followed by a paucity of trained personnel (46.3 percent) and patients' rejection of weight-loss advice (45.7 percent). Eighty-two percent of respondents agreed that dentists would be more willing to intervene if obesity were linked definitively to oral disease. Given continued increases in obesity in the United States and the willingness of dentists to assist in preventive and interventional efforts, experts in obesity intervention, in conjunction with dental educators, should develop models of intervention within the scope of dental practice. Educating dentists about obesity and counseling may reduce barriers for those interested in addressing obesity in their practices.

Prevalence of Chlamydia trachomatis: results from the first national population-based survey in France

BackgroundFew studies have estimated Chlamydia trachomatis (CT) prevalence in the general population, most prevalence studies being based on people already attending healthcare settings.ObjectivesTo estimate the prevalence of CT in France,...

Greenways: a sustainable leisure experience concept for both communities and tourists

This paper reviews the development of greenways/car-free trails as an alternative environmentally friendly communication system and assesses their strengths and weaknesses. It gives the results of a questionnaire survey of 1261 users along a 106-km greenway in Spain linking the Pyrenees with the Mediterranean Sea. Unlike previous studies, sampling was random and unequal probabilities of selection were properly accounted for. The results show a complex range of user profiles, tourist and non-tourist, their perceptions of the trail and some of the direct and indirect impacts of the greenway on the communities through which it passes. Management recommendations flowing from the survey are presented with the aim of increasing the use of accommodation services and businesses in the towns and villages near the greenway and helping create a more sustainable tourism system. The paper illustrates the potential importance of greenways in a future low-carbon tourism strategy adapting to climate change.

エリア・サンプリングの問題点

This article discusses two important problems of area sampling in public opinion surveys.The first revolves around the weighting that should be done to obtain unbiased estimates by correcting for the unequal selection probabilities in households of different size. In practice, many Japanese surveys omit such weighting. When the response rate is low, analyses showed that weighted data had an effect on the proportions of other demographics.The second revolves around the residential map itself. The currency of the residential map is extremely important because of its effect on the response rate and survey quality. Despite the relatively low population mobility in the area of the present survey, there were some changes due to persons moving in and out. Therefore, the residential map must be updated with current information prior to the fieldwork.

A generalized nonlinear mixed-effects height–diameter model for Eucalyptus globulus L. in northwestern Spain

A generalized height–diameter model was developed for Eucalyptus globulus Labill. stands in Galicia (northwestern Spain). The study involved a variety of pure stands ranging from even-aged to uneven-aged. Data were obtained from permanent circular sample plots in which trees were sampled within different radii according to their diameter at breast height. A combination of weighted regression, to take into account the unequal selection probabilities of such an inventory design, and mixed model techniques, to accommodate local random fluctuations in the height–diameter relationship, were applied to estimate fixed and random parameters for several models reported in the relevant literature. The models that provided the best results included dominant height and dominant diameter as fixed effects. These models explained more than 83% of the observed variability, with mean errors of less than 2.5 m. Random parameters for particular plots were estimated with different tree selection options. Height–diameter relationships tailored to individual plots can be obtained by calibration of the height measurements of the three smallest trees in a plot. An independent dataset was used to test the performance of the model with data not used in the fitting process, and to demonstrate the advantages of calibrating the mixed-effects model.

Disaster Preparedness of Households with Special Needs in Southeastern Pennsylvania

Individuals with disabilities are disproportionately vulnerable in evacuation emergencies, and they face numerous challenges accessing resources for response and recovery. The aim of this study was to compare the preparedness behaviors of households with and without special-needs members. A random-digit-dial telephone survey was conducted of 501 adults in southeastern Pennsylvania in 2008. The survey instrument gathered data on sociodemographic characteristics, disability status/functional limitations, and preparedness behaviors related to an evacuation emergency. Analyses were conducted in 2009 and included multiple logistic regressions. Data were weighted to correct for unequal probabilities of selection and response. Nineteen percent (n=95) of respondents reported living in a household in which someone had a transportation-related special need requiring assistance in case of an emergency. Households with a special-needs member had greater odds of having arranged a place to meet (OR=2.2; 95% CI=1.26, 3.88); located a shelter (OR=1.8; 95% CI=1.05, 3.24); or packed a bag (OR=1.8; 95% CI=1.02, 3.21). No significant differences were identified with respect to awareness of evacuation routes, purchasing of food and water, or creation of an emergency plan to guide evacuation decision-making. Despite both greater vulnerability to disaster and numerous messages by governmental and nongovernmental organizations, households with special-needs members are not more likely to engage in time-consuming preparedness behaviors such as planning and buying. Understanding the motivations and behaviors of special-needs groups is critical to designing informed risk-communication messages and interventions to bolster the preparedness of the most vulnerable populations.

Employment Among Persons With Past and Current Mood and Anxiety Disorders in the Israel National Health Survey

Employment Among Persons With Past and Current Mood and Anxiety Disorders in the Israel National Health Survey

Proper analysis of secondary phenotype data in case‐control association studies

Case-control association studies often collect extensive information on secondary phenotypes, which are quantitative or qualitative traits other than the case-control status. Exploring secondary phenotypes can yield valuable insights into biological pathways and identify genetic variants influencing phenotypes of direct interest. All publications on secondary phenotypes have used standard statistical methods, such as least-squares regression for quantitative traits. Because of unequal selection probabilities between cases and controls, the case-control sample is not a random sample from the general population. As a result, standard statistical analysis of secondary phenotype data can be extremely misleading. Although one may avoid the sampling bias by analyzing cases and controls separately or by including the case-control status as a covariate in the model, the associations between a secondary phenotype and a genetic variant in the case and control groups can be quite different from the association in the general population. In this article, we present novel statistical methods that properly reflect the case-control sampling in the analysis of secondary phenotype data. The new methods provide unbiased estimation of genetic effects and accurate control of false-positive rates while maximizing statistical power. We demonstrate the pitfalls of the standard methods and the advantages of the new methods both analytically and numerically. The relevant software is available at our website.

A modified unrelated question randomized response device for complex surveys

Singh et al. (Stat Trans 6(4):515–522, 2003) proposed a modified unrelated question procedure and they also demonstrated that the modified procedure is capable of producing a more efficient estimator of the population parameter πA, namely, the proportion of persons in a community bearing a sensitive character A when πA < 0.50. The development of Singh et al. (Stat Trans 6(4):515–522, 2003) is based on simple random samples with replacement and on the assumption that πB, namely, the proportion of individuals bearing an unrelated innocuous character B is known. Due to these limitations, Singh et al.’s (Stat Trans 6(4):515–522, 2003) procedure cannot be used in practical surveys where usually the sample units are chosen with varying selection probabilities. In this article, following Singh et al. (Stat Trans 6(4):515–522, 2003) we propose an alternative RR procedure assuming that the population units are sampled with unequal selection probabilities and that the value of πB is unknown. A numerical example comparing the performance of the proposed RR procedure under alternative sampling designs is also reported.

A comparison between the design-based and model-based approaches using longitudinal survey data

Survey data analysis using complex sampling designs ought to account for clustering, stratification and unequal probability of selection. Design-based and model-based methods are two commonly used routes taken to account for such survey designs. Several studies of cross-sectional survey designs have shown that these two approaches provide similar results when the model fits the data well. The present paper aims at comparing these two approaches for longitudinal survey design using the National Population Health Survey (NPHS) dataset. A marginal modeling approach proposed by Rao and modified bootstrap method for longitudinal data were used by way of a design-based method. The Generalized Estimating Equation (GEE) method, proposed by Liang and Zeger was used as a typical model-based approach. The parameter estimates obtained using the design-based and model-based methods were similar. However, the standard errors and the 95% confidence interval were different. Rao's method produced the most conservative standard errors. In conclusion, design-based methods should be preferred over model-based methods, as this method provides reliable results.

Accounting for Sampling Design in the SHIW

This paper analyses how sampling design affects variance estimates and inference using the data collected by the Survey on Household Income and Wealth (SHIW). The SHIW combines three basic features: stratification, clustering, and weighting to correct for unequal probabilities of selection among sampling units. A model to assess variance is presented and a Jackknife Repeated Replication method is suggested to estimate variance. Empirical evidence shows that: 1) simple random sampling formula for variance underestimates by a factor of between 3 and 2 the estimates that take into account all the design features; 2) the bias of unweighted estimates may be fairly substantial; 3) all these factors can seriously mislead inference based on SHIW data.

Non‐Response in a Birth Cohort Study: The Case of the Millennium Cohort Study

The advantages that birth cohort data offer researchers interested in the measurement and explanation of change across the life course are tempered by the problem of non‐response that becomes progressively larger as cohorts age. This article sets out the extent of this problem for the first two waves of the fourth in the series of UK birth cohorts: the Millennium Cohort Study. The response rate at Wave 1 is 72%, declining to 58% at Wave 2. Sample loss between Waves 1 and 2 was due to the failure to trace families who had moved, to contact families at a known address and to refusal. The correlates of these three kinds of non‐response are different. Although non‐respondents are systematically different from respondents at Waves 1 and 2, these differences in the propensity to respond are small compared with the unequal selection probabilities built into the sample design. It is, therefore, unlikely that weighting adjustments will have a substantial effect, over and above the effect of the sample design, on longitudinal analyses based on the first two waves of the study.

The Impact of Immigration Status on Tobacco Use among Chinese-American Adults in Texas

This study analyzed the impact of immigration status on current tobacco use among adult Chinese-Americans living in Texas. A survey was administered in Chinese and English in 2004 to assess tobacco use among Chinese-American adults using a stratified probability sample from two large metropolitan areas in Texas. Data were adjusted for unequal probability of selection and weighted to provide state-wide estimates for Chinese-Americans in Texas. The study sample was comprised of 1,054 Chinese-American adults. The overall current smoking rate was 11.1% with men's rates much higher (16.1%) than women's (6.7%). Lower household income and education increased smoking among males but more educated females had a tendency to smoke more. Although overall smoking rate among Chinese-Americans was significantly lower than the general Texas population (20.6%), smoking rate among recent immigrant men (<5 years in the U.S.) was alarmingly higher (28.0%). U.S.-born Chinese-American men's smoking rate (25.0%) is similar to that of their U.S.-born counterparts (23.7%). U.S.-born Chinese-American men initiated smoking 4 years earlier (13.8 years) than their immigrant counterparts. Although Chinese-Americans in Texas had overall lower smoking rates than the general population, the high smoking rates among recently immigrated men emphasize the need for cessation activities targeting this group.

Improving RBS estimates - effects of the auxiliary variable, stratification of the crown, and deletion of segments on the precision of estimates

Randomized Branch Sampling (RBS) is a multistage sampling procedure using natural branching in order to select samples for the estimation of tree characteristics. The existing variants of the RBS method use unequal selection probabilities based on an appropriate auxiliary variable, and selection with or without replacement. In the present study, the effects of the choice of the auxiliary variable, of the deletion of segments, and of the stratification of the tree crown on the sampling error were analyzed. In the analysis, trees of three species with complete crown data were used: Norway spruce (&lt;i&gt;Picea abies&lt;/i&gt; [L.] Karst.), European mountain ash (&lt;i\&gt;Sorbus aucuparia&lt;/i&gt; L.) and Monterey pine (&lt;i&gt;Pinus radiata&lt;/i&gt; D. Don). The results clearly indicate that the choice of the auxiliary variable affects both the precision of the estimate and the distribution of the samples within the crown. The smallest variances were achieved with the diameter of the segments to the power of 2.0 (Norway spruce) up to 2.55 (European mountain ash) as an auxiliary variable. Deletion of great sized segments yielded higher precision in almost all cases. Stratification of the crown was not generally successful in terms of a reduction of sampling errors. Only in combination with deletion of stem segments, a clear improvement in the precision of the estimate could be observed, depending on species, tree, target variable, and definition and number of strata on the tree. For the trees divided into two strata, the decrease in the coefficient of variation of the estimate lies between 10% (European mountain ash) and 80% (old pine) compared with that for unstratified trees. For three strata, the decrease varied between 50% (European mountain ash) and 85% (old pine).

Inverse Sampling with Unequal Selection Probabilities

In some real situations the population of interest is divided into two groups, of which one contains only a few units. In other cases, the population may be considered as subdivided into two group', for example, if only a few units display a value of the variable of interest which is highly different from zero, while all the other units show a value equal to or near zero. In both cases, inverse sampling is more efficient than classical fixed sample-size designs to obtain the parameter estimators for the whole population as well as for its groups (e.g., Salehi and Seber, 2004). In fact, in this design the procedure selection continues until a prefixed number of units with the characteristic of interest is sampled. Since it is not known a priori to which group the population units belong, the sample size is a random variable. Christman and Lan (2001) and Salehi and Seber (20012004) considered inverse sampling designs when all the population units have equal selection probabilities. In this article, we consider the general case in which the units may have unequal probabilities of being included in the sample. In fact, in many real situations different units may have different selection probabilities because of some inherent features of the sampling procedure, or in order to obtain better estimates. We derive unbiased estimators of the totals of the two groups, their variance and the corresponding unbiased variance estimators in inverse sampling with replacement. Finally, we derive similar results for more complex designs, where the selection procedure stops before observing the prefixed number of units from the rare group.

Association between psychiatric disorders and smoking stages within a representative clinic sample of Puerto Rican adolescents

Objective We examined the prevalence of smoking behaviors and their association with specific psychiatric disorders in a representative sample of youth from behavioral health clinics in Puerto Rico. Method A complex sampling design was used to select the sample, and analyses were conducted to account for the unequal selection probability, stratification, and clustering. All analyses were weighted back to the clinical population from which they were drawn. Psychiatric and substance use disorders were assessed using the parent and youth versions of the Diagnostic Interview Schedule for Children, Version 4.0. Results More than one third of the sample reported experience with cigarette smoking, and approximately one quarter reported smoking at least once per week (23.4%). As expected, the alcohol and drug use disorders demonstrated some of the strongest associations with individual smoking stages. These were the only disorders that remained significantly associated with nicotine dependence after controlling for comorbidity. Conclusions Our findings confirm the need for screening of smoking behavior and nicotine dependence in treatment settings and the integration of psychiatric/substance use treatments with smoking cessation.

The Weighting Process in the SHIW

The design of a probability sample jointly determines the method used to select sampling units from the population and the estimator of the population parameter. If the sampling fraction is constant for all the units in the sample, then the unweighted sampling mean is an unbiased estimator. In the Survey on Household Income and Wealth (SHIW), units included in the sample have unequal probabilities of selection and each observation is weighted using the inverse of the proper sampling fraction (design weight) adjusted for the response mechanism (non-response weight) and for other factors such as imperfect coverage. In this paper we present the weighting scheme of the SHIW and assess its impact on bias and variance of selected estimators. Empirical evidence shows that the increasing variability induced by using weighted estimators is compensated by the bias reduction even when performing analysis on sample domains. A set of longitudinal weights is also proposed to account for the selection process and the attrition of the SHIW component. These weights, given their enhanced description of the panel population, should be better suited to perform longitudinal analysis; nevertheless their higher variance implies that they are not always preferable in terms of mean square error.

Competing Impact of Excess Weight Versus Cardiorespiratory Fitness on Cardiovascular Risk

Obesity is a risk factor for cardiovascular disease, whereas high cardiorespiratory fitness (CRF) is cardioprotective. This study evaluated the competing effect of weight and fitness on biomarkers of cardiovascular risk in a nationally representative sample of 2,112 adults (20 to 49 years of age; body mass index [BMI] > or =18.5 kg/m(2)) without previously diagnosed cardiovascular disease from the National Health and Nutrition Examination Survey 1999 to 2002. CRF levels were assigned using age- and gender-specific reference points of estimated maximal oxygen consumption calculated from submaximal graded exercise treadmill testing. CRF was also categorized by sample-specific tertiles of maximal oxygen consumption. Weight was categorized using BMI. Fasting insulin level >12.2 mU/L, C-reactive protein level > or =3.0 mg/L, and total cholesterol/high-density lipoprotein ratio (TC/HDL) >5 characterized increased cardiovascular risk. CRF and BMI were independently associated with increased fasting insulin and C-reactive protein (p <0.05). When patients with low, moderate, and high CRF were further stratified as normal, overweight, or obese, weight remained significantly associated with increased fasting insulin, C-reactive protein, and TC/HDL (p <0.001), but CRF did not. Logistic regressions evaluating increased fasting insulin, C-reactive protein, and TC/HDL demonstrated no significant differences in overweight/obese patients by CRF level after adjustment for other factors. Significant differences were present between normal-weight and overweight or obese patients regardless of fitness level. Analyses using tertiles of CRF yielded similar results. In conclusion, patients who are "fat but fit" require weight-loss interventions to improve their cardiovascular risk profiles. Future interventions should emphasize weight control, even for those with high CRF.

Prevalence and risk factors of intestinal parasites in Cuban children

To determine the prevalence of intestinal parasite infections and their risk factors in children in urban and rural settings in two Cuban municipalities. A total of 1320 Cuban schoolchildren aged 4-14 were tested by stool examination for intestinal parasite infections and evaluated by parental questionnaire for a number of common environmental, sanitary, socioeconomic and behavioural risk factors. Multivariate regression was applied to examine the relationship between the respective parasite infections and the risk factors. Prevalences of intestinal parasite infections were 58% in Fomento and 45% in San Juan y Martínez; for helminth infections, these were 18% and 24% and for protozoa infections, 50% and 29%, respectively. Helminth infections were associated with high parental education (maternal: OR 0.68, CI 0.50-0.93; paternal: OR 0.71, CI 0.52-0.96), absence of toilet (OR 1.57, CI 1.12-2.19), consumption of water from a well or river (OR 0.56, CI 0.41-0.77) and eating unpeeled/unwashed fruit (OR 1.37, CI 1.01-1.87); protozoa infections were only associated with high maternal education (OR 0.72, CI 0.57-0.91). Paediatric intestinal parasite infections are still prevalent in certain areas in Cuba and associated with a number of common environmental, socioeconomic and sanitary risk factors.

Beverage Intake Among Preschool Children and Its Effect on Weight Status

OBJECTIVE. The obesity epidemic in the United States continues to increase. Because obesity tends to track over time, the increase in overweight among young children is of significant concern. A number of eating patterns have been associated with overweight among preschool-aged children. Recently, 100% fruit juice and sweetened fruit drinks have received considerable attention as potential sources of high-energy beverages that could be related to the prevalence of obesity among young children. Our aim was to evaluate the beverage intake among preschool children who participated in the National Health and Nutrition Examination Survey 1999–2002 and investigate associations between types and amounts of beverages consumed and weight status in preschool-aged children. METHODS. We performed a secondary analysis of the data from the National Health and Nutrition Examination Survey 1999–2002, which is a continuous, cross-sectional survey of a nationally representative sample of the noninstitutionalized population of the United State. It included the collection of parent reported demographic descriptors, a 24-hour dietary recall, a measure of physical activity, and a standardized physical examination. The 24-hour dietary recall was obtained in person by a trained interviewer and reflected the foods and beverages that were consumed by the participant the previous day. The National Health and Nutrition Examination Survey food groups were classified on the basis of the US Department of Agriculture9s Food and Nutrient Database for Dietary Studies. We reviewed the main food descriptors used and classified all beverages listed. One hundred percent fruit juice was classified as only beverages that contained 100% fruit juice, without sweetener. Fruit drinks included any sweetened fruit juice, fruit-flavored drink (natural or artificial), or drink that contained fruit juice in part. Milk included any type of cow milk and then was subcategorized by percentage of milk fat. Any sweetened soft drink, caffeinated or uncaffeinated, was categorized as soda. Diet drinks included any fruit drink, tea, or soda that was sweetened by low-calorie sweetener. Several beverages were removed from the analysis because of low frequency of consumption among the sample. Water was not included in the analysis because it is not part of the US Department of Agriculture9s Food and Nutrient Database categories. For the purposes of this analysis, the beverages were converted and reported as ounces, rather than grams, as reported by the National Health and Nutrition Examination Survey, to make it more clinically relevant. The child9s BMI percentile for age and gender were calculated on the basis of Centers for Disease Control and Prevention criteria and used to identify children9s weight status as underweight ( RESULTS. All children who were aged 2 to 5 years were identified (N = 1572). Those with missing data were removed from additional analysis, resulting in a final sample of 1160 preschool children. Of the 1160 children analyzed, 579 (49.9%) were male. White children represented 35%, black children represented 28.3%, and Hispanic children represented 36.7% of the sample. Twenty-four percent of the children were overweight or at risk for overweight (BMI ≥85%), and 10.7% were overweight (BMI ≥95%). There were no statistically significant differences in BMI between boys and girls or among the ethnicities. Overweight children tended to be older (mean age: 3.83 years) compared with the normal-weight children (mean age: 3.48 years). Eighty-three percent of children drank milk, 48% drank 100% fruit juice, 44% drank fruit drink, and 39% drank soda. Whole milk was consumed by 46.5% of the children, and 3.1% and 5.5% of the children consumed skim milk and 1% milk, respectively. Preschool children consumed a mean total beverage volume of 26.93 oz/day, which included 12.32 oz of milk, 4.70 oz of 100% fruit juice, 4.98 oz of fruit drinks, and 3.25 oz of soda. Weight status of the child had no association with the amount of total beverages, milk, 100% fruit juice, fruit drink, or soda consumed. There was no clinically significant association between the types of milk (percentage of fat) consumed and weight status. In analysis of covariance, daily total energy intake increased with increased consumption of milk, 100% fruit juice, fruit drinks, and soda. However, there was not a statistically significant increase in BMI on the basis of quantity of milk, 100% fruit juice, fruit drink, or soda consumed. CONCLUSIONS. On average, preschool children drank less milk than the 2005 Dietary Guidelines for Americans recommendation of 16 oz/day. Only 8.6% drank low-fat or skim milk, as recommended for children who are older than 2 years. On average, preschool children drank