SEARCH · PubMed Health
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Multiple data sources are sometimes available as potential sampling frames for population surveys, and in some situations the use of a multiple frame sample design is more advantageous than using a single sampling frame. The use of multiple sampling frames, however, has variance and bias implications, as well as sampling, data collection, and logistical considerations. These issues are addressed for a proposed dual frame sampling approach in the National Health Interview Survey (NHIS). The results of an investigation of the sampling efficiencies and operational issues in supplementing the NHIS area frame sample with a sample of elderly African and Hispanic Americans and persons with disabilities selected from Social Security Administration files are presented.
Kinships composed of twin parents, their spouses and children, offer a robust and flexible sampling design for research in genetic epidemiology. Families-of-twins designs circumvent some of the sampling problems that arise when independent data sets are combined, and these designs provide unique evaluations of maternal influences, assortative mating and X-linkage. Unfortunately, empirical studies of families of twin parents have been limited by relatively small samples and by the self-selection biases intrinsic in ascertainment of families from volunteer twin registries. A large and representative cohort of monozygotic and dizygotic twin parents, drawn from a population-based twin registry, provides the optimal sampling frame for twin-family research. This paper reviews the sampling considerations underlying the initial family study based on the Finnish Twin Cohort and evaluates the representativeness of the sampled twins. Spouses and adult children (over 18 years) of 236 pairs of twins, about equally divided by gender and zygosity, were evaluated by a postal questionnaire. Individual response rates exceeded 86% and in 464 of the 472 nuclear families (98.3%), at last one member of the twin's family completed the questionnaire. The sampled twins, selected for fecundity to maximize statistical power of the obtained data, were broadly representative of non-selected twins drawn from the Cohort, with whom they were matched on age, gender, and zygosity. Such results suggest that the Finnish Cohort has excellent potential for extended twin-family research designs.
Immature density and population size of the biting midge Culicoides belkini (Wirth & Arnaud) were estimated for habitats on Moorea Island, French Polynesia, by means of random, 2- and 3-stage sampling designs. Samples were taken in March 1993 from 5 strata of a large larval habitat: a sandy-mud surface of approximately 5,000 m2 (stratum 1) in which approximately 12,000 land crab burrows (stratum 2) were counted, a small pond surrounded by approximately 300 m2 of muddy bank (stratum 3), and a high organic muddy area (Kopara) of approximately 1,200 m2 (stratum 4) with approximately 3,500 crab burrows (stratum 5). Larval density was usually higher in the mud of crab burrows, especially those in the Kopara stratum. Larval density was significantly lower in the sediment of the sandy area as compared with pond banks or Kopara surface. The sampling designs and techniques were logistically adequate, statistically relevant, and were recommended for future studies on C. belkini larval density. Larval habitats were characterized by means of multivariate analysis. Comparison of larval densities with selected environmental variables indicated that larvae density was higher in wet sediments with high levels of organic matter (approximately 8% of dry weight of sediment) and low salinity (approximately 0.5-1.5% NaCl equivalents). These variables were considered significant if larval control by means of habitat modification has to be achieved. Nevertheless, C. belkini can tolerate a broad spectrum of variation in the other environmental variables measured and breed in a variety of ecological situations. Therefore, it has a high potential for colonizing new habitats.
The "activities of daily living," or ADLs, are the basic tasks of everyday life, such as eating, bathing, dressing, toileting, and transferring. Reported estimates of the size of the elderly population with ADL disabilities differ substantially across national surveys. Differences in which ADL items are being measured and in what constitutes a disability account for much of the variation. Other likely explanations are differences in sample design, sample size, survey methodology, and age structure of the population to which the sample refers. When essentially equivalent ADL measures are compared, estimates for the community-based population vary by up to 3.1 percentage points; and for the institutionalized population, with the exception of toileting, by no more than 3.2 percentage points. As small as these differences are in absolute terms, they can be large in percent differences across surveys. For example, the National Medical Expenditure Survey estimates that there are 60 percent more elderly people with ADL problems than does the Supplement on Aging.
A symposium on Human Tissue Monitoring and Specimen Banking: Opportunities for Exposure Assessment, Risk Assessment, and Epidemiologic Research was held from 30 March to 1 April 1993 in Research Triangle Park, North Carolina. There were 117 registered participants from 18 states and 5 foreign countries. The first 2 days featured 21 invited speakers from the U.S. Environmental Protection Agency, the Centers for Disease Control and Prevention, the National Institute of Environmental Health Sciences, various other government agencies, and universities in the United States, Canada, Germany, and Norway. The speakers provided a state-of-the-art overview of human exposure assessment techniques (especially applications of biological markers) and their relevance to human tissue specimen banking. Issues relevant to large-scale specimen banking were discussed, including program design, sample design, data collection, tissue collection, and ethical ramifications. The final group of presentations concerned practical experiences of major specimen banking and human tissue monitoring programs in the United States and Europe. The symposium addressed the utility and research opportunities afforded by specimen banking programs for future research needs in the areas of human exposure assessment, risk assessment, and environmental epidemiology. The third day of the symposium consisted of a small workshop convened to discuss and develop recommendations to the U.S. Environmental Protection Agency regarding applications and utility of large-scale specimen banking, biological monitoring, and biological markers for risk assessment activities.
Sample size requirements are provided for designs of studies in which clusters are randomized within each of several strata, where cluster size itself may be a stratifying factor. The approach generalizes a formula derived by Woolson et al., which provides sample size requirements for the Cochran-Mantel-Haenszel statistic. Issues of data analysis are also discussed.
Data obtained from national probability sample surveys provide important information on the prevalence of various health conditions and distributions of physical and biochemical characteristics of the U.S. population. The sample design of a survey specifies how sampling from a designated population over a stated period is to be accomplished. A survey's analytical objectives and interests--in particular subpopulations--affect the sample design strategy. Selected subdomains of the population often must be oversampled so that estimates can be made with acceptable precision. This article addresses sample design considerations for a national probability sample for human tissue monitoring and specimen banking. Among the sampling issues addressed are the oversampling of special populations e.g., minority groups and at-risk groups such as low income or elderly persons; geographic coverage; and sample size considerations. The sample design for a major health survey, the Third National Health and Nutrition Examination Survey (NHANES III), is used to illustrate a complex, multistage probability sample design and to highlight some of the sampling issues discussed in this article.
To achieve its primary objectives, the Study on the Efficacy of Nosocomial Infection Control (SENIC Project) focused its attention on a target population of patients referred to as SENIC-eligible admissions in a target population of hospitals referred to as the "SENIC Universe." SENIC thus required a design for sampling hospitals and patients within these hospitals and a valid procedure for projecting sample results to the target population. This paper presents the details of the sampling design used, describes the actual process of selecting hospitals and patients for the surveys, explains the procedure used to project sample results to the target population, and examines the possibility of bias in the design and hospital selection process. As with most large-scale sample surveys, the design and sample selection processes for the surveys in Phases II and III of SENIC were complicated by incomplete frame, nonresponse and measurement problems. Nevertheless, adjustments to reduce the effects of some of these problems have been made through the development of a valid procedure for projecting sample results to the target population, and it appears unlikely that practically important nonsampling biases will result from the estimation procedures applied to this sample of hospitals.
Legislative measures restricting cigarette smoking have the potential to influence whether a person begins or continues to smoke and to affect the impact of passive smoking. We surveyed a representative sample of the adult population in Ontario on their knowledge of existing legislation and of the adverse effects of primary and secondary smoking on health. We also assessed their attitudes toward a range of restrictions and changes in legislation as well as their views on the enactment and enforcement of such legislation. This paper reports on the sample design, methods, response rates, and representativeness of the respondents. We used a three-stage stratified cluster design, covering both urban areas (with or without existing smoking bylaws) and rural areas and incorporating telephone interviews using random-digit dialing. The total number of respondents was 1,383, for an overall response rate of 67.5 percent. Despite attempts to ensure anonymity and to convey the importance of participation, we did not achieve total representativeness in sex ratio, age distribution, and certain educational and occupational categories. A companion paper reports on the population estimates of the variables under study.
In this report, we consider the situation in which one wishes to identify a cohort of a specified number of individuals within each of several domains for future follow-up studies based on a single-stage cluster sampling design. We develop sample size formulae relevant to this situation and introduce a variation of single-stage cluster sampling that seems more suitable in this situation than is ordinary single-stage cluster sampling. The basis for this variation is the concept that the definition of eligible respondents is not the same for all clusters. The use of this modified respondent rule (which we call telescopic) enables one to meet specified sample sizes in all domains of interest without the need to sample extra individuals in some domains. We used a version of this sampling design successfully in the field with a survey of elderly persons conducted in Shanghai, People's Republic of China.
Three different geometrically spaced sampling schedule designs, a theoretically optimal design, and a design that included all sampling times were evaluated by comparing parameter estimates, half-life, R2, and an indicator of variance-covariance space. Alfalfa and oat hays were tested using nylon bags placed in the rumen of a fistulated, non-lactating cow, and the amount of NDF remaining was measured at specified times. Parameters were estimated from f(t, phi) = Ae-K(t-lag) + U, where f (t, phi) = NDF at time t (h), A = degradable NDF, U = undegradable NDF, lag = time before digestion, and K = rate constant (h). A, U, and f(t, phi) are expressed as a fraction of DM at time 0. Estimates A and U did not fluctuate, whereas K and lag varied across designs. All R2 were over .96 and did not vary across designs. Comparison of designs that had the same number of observations showed that the indicator of the variance-covariance space was statistically similar across designs, although the optimal design was ranked best. Parameter estimates were similar when using different sampling schedule designs, but some estimates differed by 29%. The optimal design sampling schedule provided sufficient information to estimate parameters without loss of accuracy when compared with other designs.
This paper describes a two-stage sampling design for obtaining probability samples of the elderly and other specialized populations. The first stage enumerates a sample of elderly individuals residing in a probability, sample of households; the second stage involves the mailing of a questionnaire to the eligible respondents located in the first stage. The results of this method for a study of noninstitutionalized elderly in Washington State are reported. These results indicate that (1) elderly individuals will respond in the two-stage sampling design does not appear to increase sample bias above that expected in normal mailed-questionnaire studies, and (3) the method is extremely inexpensive. Use of the two-stage design by social gerontologists is recommended.
An experiment design procedure is proposed for nonlinear parameter estimation studies that formally incorporates prior parameter uncertainty. The design criterion derives from information theory considerations and involves an asymptotic interpretation of the expected posterior information provided by an experiment. A pharmacokinetic sample schedule design problem is used to illustrate and evaluate this information theoretic design strategy. The model considered is commonly used to describe the plasma concentration of a drug following its oral administration. The limitations and advantages of the proposed design procedure are discussed in relation to other previously reported design techniques for incorporating parameter uncertainty.
The basis of sample size calculations is usually needed in protocols for clinical trials and when publishing results in respected journals. Although a large amount of research has been undertaken on sample size estimation for different trial designs, in practice the methods are rarely used. This paper describes some useful theory that has practical relevance.
This paper presents a simple method of calculating sample sizes for unequal-sample-size designs with use of published tables applicable to equal-sample-size design. The method applies to both the logrank test and the t-test. For the power of logrank test, this paper compares the proposed method with existing methods and with the Monte Carlo simulation.
We develop approximate methods to compare the efficiencies and to compute the power of alternative potential designs for sampling from a cohort before beginning to collect exposure data. Our methods require only that the cohort be assembled, meaning that the numbers of individuals Nkj at risk at pairs of event times tk and tj greater than or equal to tk are available. To compute Nkj, one needs to know the entry, follow-up, censoring, and event history, but not the exposure, for each individual. Our methods apply to any "unbiased control sampling design," in which cases are compared to a random sample of noncases at risk at the time of an event. We apply our methods to approximate the efficiencies of the nested case-control design, the case-cohort design, and an augmented case-cohort design, compared to the full cohort design, in an assembled cohort of 17,633 members of an insurance cooperative who were followed for mortality from prostatic cancer. The assumptions underlying the approximation are that exposure is unrelated both to the hazard of an event and to the hazard for censoring. The approximations performed well in simulations when both assumptions held and when the exposure was moderately related to censoring.
OBJECTIVE: There is growing interest in the extent to which body composition, both short- and long-term, differs in infants and children born at the extremes of birth weight. This is because a growing number of studies have linked low birth weight and fetal growth restriction to the chronic diseases in adulthood that often are obesity-related, and there is also evidence to suggest that heavy infants may be at increased risk for obesity in later life, again with the attendant obesity-related chronic diseases. Our objective was to compare anthropometric indices of body composition of infants and young children born small-for-gestational-age (SGA, <10th percentile) or large-for-gestational age (LGA, >/=90th percentile) with those of normal birth weight status (appropriate-for-gestational-age, AGA) in a US sample. DESIGN: National sample of US-born non-Hispanic white, non-Hispanic black, and Mexican-American infants and young children, 2 to 47 months of age, examined in the third National Health and Nutrition Examination Survey (NHANES III, 1988-1994), for whom birth certificates were obtained. The primary outcomes were normalized anthropometric indices (z scores or standard deviation units [SDU]) of nutritional status and body composition (mid-upper arm circumference, triceps and subscapular skinfolds, mid-upper arm muscle and mid-upper arm fat areas (UFA), and the arm fat index). The outcomes thus were scaled to permit comparison across chronologic ages. RESULTS: The prevalence of SGA was 8.6%, appropriate-for-gestational-age 80.9%, and LGA 10.5%. From ages 2 to 47 months, for infants and young children born SGA, there was a persistent overall deficit in muscularity (mid-upper arm circumference and mid-upper arm muscle area) of approximately -0.50 SDU, but less of a deficit in fatness, particularly at the youngest ages. For infants and young children born LGA, there was a surfeit in muscularity of approximately 0.45 SDU, with less of a surfeit in fatness, particularly at the youngest ages. Across all ages, the mean UFA showed a statistically significant deficit for SGA children (-0.27 +/- 0.10 SDU) and surfeit for LGA children (0.24 +/- 0.08 SDU). At individual ages for UFA and at individual and all ages combined for skinfold thicknesses, there were no significant differences in level of subcutaneous fatness in the three birth-weight-for-gestational-age groups. There was a tendency in the first year for the arm fat index (% arm fat) to be significantly higher for SGA infants, but the effect did not persist after the first year. CONCLUSION: SGA infants remain smaller and LGA infants larger in size through early childhood, but the discrepancies in weight are primarily attributable to differences in lean body mass (muscularity). Fatness is less affected. Thus, based on the fatness indicators used, at any given weight for infants and children 2 to 47 months of age, percent body fat appears to be relatively higher for children who were SGA at birth and lower in those who were LGA at birth. These differences in body composition for SGA infants support the evidence documenting a link between disturbances in intrauterine growth and chronic disease associated with subsequent adiposity in adulthood.