Epidemiologic studies utilizing surveys: accounting for the sampling design.

BACKGROUND. Since large-scale health surveys usually have complicated sampling schemes, there is often a question as to whether the sampling design must be considered in the analysis of the data. A recent disagreement concerning the analysis of a body iron stores-cancer association found in the first National Health and Nutrition Examination Survey and its follow-up is used to highlight the issues. METHODS. We explain and illustrate the importance of two aspects of the sampling design: clustering and weighting of observations. The body iron stores-cancer data are reanalyzed by utilizing or ignoring various aspects of the sampling design. Simple formulas are given to describe how using the sampling design of a survey in the analysis will affect the conclusions of that analysis. RESULTS. The different analyses of the body iron stores-cancer data lead to very different conclusions. Application of the simple formulas suggests that utilization of the sample clustering in the analysis is appropriate, but that a standard utilization of the sample weights leads to an uninformative analysis. The recommended analysis incorporates the sampling weights in a nonstandard way and the sample clustering in the standard way. CONCLUSIONS. Which particular aspects of the sampling design to use in the analysis of complex survey data and how to use them depend on certain features of the design. We give some guidelines for when to use the sample clustering and sample weights in the analysis.     

National health interview surveys in Europe: an overview.

In order to study the value of national health interview surveys for national and international research and policy activities, this paper examines the existence and content of recent and future health interview surveys in the 15 member states of the European Union (EU), Norway, Iceland and Switzerland. National health interview surveys are performed in most countries, but not in Greece (only regional surveys), Luxembourg, Ireland and Iceland (only multi-purpose surveys). The health interview surveys in the other 14 countries provide regular data on the main health topics. Of the 14 health topics that are examined in this inventory seven are measured in all countries. Questions on health status (e.g. self-assessed health, long-term physical disability, and height and weight) and medical consumption (e.g. consultations with the general practitioner, GP) are often included. Lifestyle topics are less often included, except smoking habits, information about which is sought in all countries. Topics like diet and drugs/narcotics are more often included in special surveys than in general health interview surveys. Despite differences in the content, frequency and methodology of national health interview surveys in different countries, these surveys are a valuable source of information on the health of Europeans.     

Don't spin the pen: two alternative methods for second-stage sampling in urban cluster surveys

In two-stage cluster surveys, the traditional method used in second-stage sampling (in which the first household in a cluster is selected) is time-consuming and may result in biased estimates of the indicator of interest. Firstly, a random direction from the center of the cluster is selected, usually by spinning a pen. The houses along that direction are then counted out to the boundary of the cluster, and one is then selected at random to be the first household surveyed. This process favors households towards the center of the cluster, but it could easily be improved. During a recent meningitis vaccination coverage survey in Maradi, Niger, we compared this method of first household selection to two alternatives in urban zones: 1) using a superimposed grid on the map of the cluster area and randomly selecting an intersection; and 2) drawing the perimeter of the cluster area using a Global Positioning System (GPS) and randomly selecting one point within the perimeter. Although we only compared a limited number of clusters using each method, we found the sampling grid method to be the fastest and easiest for field survey teams, although it does require a map of the area. Selecting a random GPS point was also found to be a good method, once adequate training can be provided. Spinning the pen and counting households to the boundary was the most complicated and time-consuming. The two methods tested here represent simpler, quicker and potentially more robust alternatives to spinning the pen for cluster surveys in urban areas. However, in rural areas, these alternatives would favor initial household selection from lower density (or even potentially empty) areas. Bearing in mind these limitations, as well as available resources and feasibility, investigators should choose the most appropriate method for their particular survey context.     

Post-coital contraception: an overview of published studies

Information from published studies testing the effect of hormonal and mechanical interceptive methods in post-coital contraception was pooled with the aim of assessing the efficacy of various regimens. The pooled failure rates were 0.6% (19/3168 based on four studies) for high-dosage ethinylestradiol, 1.1% (11/975, based on two studies) for other high-dose estrogens, 1.8% (69/3802 based on 11 studies) for combined estro-progestin preparations, 2.0% (20/998 based on three studies) for danazol and 0.1% (1/879 based on nine studies) for intra-uterine devices (IUD). Although this overview has several limitations in terms of over-representation of optimistic results in the literature or absence of direct comparison between treatments in several studies, it suggests that IUD and high-dose ethinylestradiol are more efficacious forms of post-coital contraception than danazol or combined contraceptives.     

Outcome-dependent sampling: an efficient sampling and inference procedure for studies with a continuous outcome

To characterize the relation between an exposure and a continuous outcome, the sampling of subjects can be done much as it is in a case-control study, such that the sample is enriched with subjects who are especially informative. In an outcome-dependent sampling design, observations made on a judiciously chosen subset of the base population can provide nearly the same statistical efficiency as observing the entire base population. Reaping the benefits of such sampling, however, requires use of an analysis that accounts for the outcome-dependent sampling. In this report, we examine the statistical efficiency of a plain random sample analyzed with standard methods, compared with that of data collected with outcome-dependent sampling and analyzed by either of 2 appropriate methods. In addition, 3 real datasets were analyzed using an outcome-dependent sampling approach. The results demonstrate the improved statistical efficiency obtained by using an outcome-dependent sampling, and its applicability in a wide range of settings. This design, coupled with an appropriate analysis, offers a cost-efficient approach to studying the determinants of a continuous outcome.     

U.S. and German case studies in chronic care management: an overview

In December 2004, the Bertelsmann Stiftung, The Commonwealth Fund, and AcademyHealth jointly sponsored a conference in Berlin, Germany on "Case Studies in Chronic Care Management." This conference assembled representatives from the U.S. and German Governments, several organizations that had developed chronic care management initiatives in the two countries, and other health care and policy experts to discuss clinical, organizational, and financing issues. The case studies discussed at the conference are presented in this issue of the Review; this article describes some relevant considerations that are common to the U.S. and German health systems.     

The landscape of community health insurance in India: an overview based on 10 case studies

The Indian health system is mainly funded by out-of-pocket payments. More than 80% of health care expenditure is borne by individual households. Only about 3% of the population, mostly those in the formal sector, benefit from some form of health insurance. Several Indian Non-Governmental Organisations (NGOs) have initiated Community Health Insurance (CHI) schemes within their existing development programmes. This article describes the principal features of the design and functioning of a selection of 10 CHI schemes and presents a brief overview of the current landscape of CHI in India. The schemes explicitly target the poorest and most vulnerable households in Indian society-scheduled tribes, scheduled castes and poor women. Three CHI management models can be distinguished. The first model consists of local NGOs acting as both insurer and provider. In the second model, the NGO is the insurer but does not itself provide care, which is then purchased from a private provider. In the third model, the NGO neither does provide health care nor acts as an insurer: the NGO, on behalf of a community, links with an insurer and purchases health care from a provider. The benefit packages generally include both primary and secondary care and most of the providers are in the private sector. Most of the schemes require external resources for financial sustainability. There is currently little information on the impact of CHI schemes on the performance of local health systems and more research is warranted in that respect.     

Child anthropometry in cross-sectional surveys in developing countries: an assessment of the survivor bias.

In cross-sectional surveys, the sample of children with anthropometric measurements is not representative of all children in a birth cohort, since only children surviving to the survey date are measured. This survivor bias may have implications for studies of trends and differentials in anthropometric indicators. In this paper, the effects of the survivor bias on the estimates of child anthropometric indicators are assessed by 1) reviewing evidence from longitudinal studies on the prevalence of malnutrition among deceased children and among surviving children and by 2) analyzing retrospective data on child mortality and cross-sectional data on child anthropometry in 17 national surveys that are part of the Demographic and Health Surveys Program. It is concluded that comparisons of anthropometric data across geographic units, population subgroups, and calendar time are marginally affected by the survivor bias, unless mortality differences between the birth cohorts are very large (e.g., well over 50 per 1,000 births).     

Missing data in the 2 x 2 table: patterns and likelihood-based analysis for cross-sectional studies with supplemental sampling

Standard measures of crude association in the context of a cross-sectional study are the risk difference, relative risk and odds ratio as derived from a 2x 2 table. Most such studies are subject to missing data on disease, exposure, or both, introducing bias into the usual complete-case analysis. We describe several scenarios distinguished by the manner in which missing data arise, and for each we adjust the natural multinomial likelihood to properly account for missing data. The situations presented allow for increasing levels of generality with regard to the missing data mechanism. The final case, quite conceivable in epidemiologic studies, assumes that the probability of missing exposure depends on true exposure and disease status, as well as upon whether disease status is missing (and conversely for the probability of missing disease information). When parameters relating to the missing data process are inestimable without strong assumptions, we propose maximum likelihood analysis subsequent to collecting supplemental data in the spirit of a validation study. Analytical results give insight into the bias inherent in complete-case analysis for each scenario, and numerical results illustrate the performance of likelihood-based point and interval estimates in the most general case. Adjustment for potential confounders via stratified analysis is also discussed.     

Health-related quality of life in patients with myalgic encephalomyelitis/chronic fatigue syndrome: an Australian cross-sectional study

Quality of Life Research - Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a serious and debilitating disorder associated with significant disruptions in daily life including. This...     

Oral drug self-administration: an overview of laboratory animal studies.

Many abused drugs can be established as orally delivered reinforcers for rhesus monkeys and other animals. Benzodiazepines, barbiturates, opioids, psychomotor stimulants, dissociative anesthetics, and ethanol can come to serve as reinforcers when taken by mouth. The principal problems in establishing drugs as reinforcers by the oral route of administration are (1) aversive taste, (2) delay in onset of central nervous system effects, and (3) consumption of low volumes of drug solution. Strategies have been devised to successfully overcome these problems, and orally delivered drugs can be established as effective reinforcers. Reinforcing actions are demonstrated by consumption of greater volumes of drug solution than volumes of the water vehicle, and supporting evidence for reinforcing effects consists of the maintenance of behavior under intermittent schedules of reinforcement and the generation of orderly dose-response functions. This article presents an overview of studies of behavior reinforced by oral drug reinforcement. Factors that control oral drug intake include dose, schedule of reinforcement, food restriction, and alternative reinforcers. Many drugs, administered by the experimenter, can alter oral drug reinforcement. Relative reinforcing effects can be assessed by choice procedures and by persistence of behavior across increases in schedule size. In general, reinforcing effects increase directly with dose. Rhesus monkeys prefer combinations of reinforcing drugs to the component drugs. The taste of drug solutions may act as a conditioned reinforcer and a discriminative stimulus. Consequences of drug intake include tolerance and physiological dependence. Findings with orally self-administered drugs are similar to many findings with other positive reinforcers, including intravenously self-administered drugs.     

Differences between studies in reported relative risks associated with smoking: an overview.

Reported Relative Risks associated with smoking differ between studies; these differences may reflect true biological differences between populations or may be research artifacts introduced by differences in factors such as amount smoked or smoking duration. The authors reviewed the literature published before June 1992 on relative risks associated with smoking for heart disease, stroke, lung cancer, and chronic obstructive lung disease. They quantified the effect of variables such as age, amount smoked, and smoking duration on reported relative risks. The main reasons for the variation in reported relative risks were: misclassification of former smokers as never smokers, the use of mortality rate ratios rather than incidence rate ratios, a possible period effect suggesting increasing relative risks over time, and differences in the amounts smoked. It is far more likely that these factors are responsible for the observed variation between studies than that the variations reflect true biological differences between populations. Using relative risks from other studies is therefore justified in calculating a population attributable risk if the studies are carefully selected and address factors such as amount smoked and period effects.     

Analysis of cluster randomized trials with repeated cross-sectional binary measurements

Analytical techniques appropriate for cluster randomized trials that utilize a repeated cross-sectional design have not been extensively evaluated. This paper compares methods that can be used to evaluate the impact of an intervention on dichotomous outcomes. The methods are applied to data from a study on the implementation of Cochrane review evidence, in which 25 hospital obstetric units were randomized. Assessments were made for 30 pregnancies in each obstetric unit at baseline, and for 30 separate pregnancies at follow-up. The principal issues addressed are how best to take clustering into account and to allow for baseline imbalance. We compare cluster level analyses, the clustered Woolf method, marginal models based on generalized estimating equations, multilevel models, and methods based on random effects meta-analysis. Analyses which ignored the baseline assessments showed no effect of the intervention. There was substantial baseline imbalance, however, so that analyses taking into account the baseline were necessary. Yet, while analyses of change from baseline showed evidence of an effect of intervention, adjusting for baseline using analysis of covariance did not. Analysis of covariance required the use of cluster level rather than individual level responses, since different pregnancies were evaluated at baseline and follow-up. Also, when analysing change from baseline, we show it is important to allow for variation in the effect of secular trend between clusters in a multilevel model, or use robust variance estimates in a marginal model, for otherwise confidence intervals for the effect of intervention will be too narrow. We conclude however that analyses of change from baseline can be misleading since they are affected too much by baseline results, and that analysis of covariance approaches are preferable. To prevent difficulties in interpreting the results from repeated cross-sectional cluster trial designs, one should either attempt to achieve baseline balance by careful stratification of the clusters prior to randomization, or have sufficiently large samples for precise estimation of the effects of imbalance.     

抗高血压药物与老年痴呆关系的流行病学研究现状

阿尔茨海默病是老年痴呆症的主要类型,约占全部痴呆病例的60％～70％。其次是血管性痴呆,约占15％～20％。除年龄和遗传因素外,其病因尚不清楚,临床上亦无确切有效的治疗方法。近年来流行病学研究倾向于支持阿尔茨海默型痴呆亦是一种血管性疾病,即血管性危险因素