Finite mixture models with concomitant information: assessing diagnostic criteria for diabetes

The World Health Organization (WHO) diagnostic criteria for diabetes mellitus were determined in part by evidence that in some populations the plasma glucose level 2 h after an oral glucose load is a mixture of two distinct distributions. We present a finite mixture model that allows the two component densities to be generalized linear models and the mixture probability to be a logistic regression model. The model allows us to estimate the prevalence of diabetes and sensitivity and specificity of the diagnostic criteria as a function of covariates and to estimate them in the absence of an external standard. Sensitivity is the probability that a test indicates disease conditionally on disease being present. Specificity is the probability that a test indicates no disease conditionally on no disease being present. We obtained maximum likelihood estimates via the EM algorithm and derived the standard errors from the information matrix and by the bootstrap. In the application to data from the diabetes in Egypt project a two-component mixture model fits well and the two components are interpreted as normal and diabetes. The means and variances are similar to results found in other populations. The minimum misclassification cutpoints decrease with age, are lower in urban areas and are higher in rural areas than the 200 mg dl^-1 cutpoint recommended by the WHO. These differences are modest and our results generally support the WHO criterion. Our methods allow the direct inclusion of concomitant data whereas past analyses were based on partitioning the data.     

Curved Chaotic Map Time Series Models and Their Stochastic Reversals

This paper considers two types of chaotic map time series models, including the well-known tent, logistic and binary-shift maps as special cases; these are called curved tent and curved binary families. Deterministic behaviour is investigated by invariant distributions, Lyapunov exponents, and by serial dependency. Stochastic time reversal of the families is shown to produce models which have a broader range of stochastic and chaotic properties than their deterministic counterparts. The marginal distributions may have concentrations and restricted supports and are shown to be a non-standard class of invariant distribution. Dependenc y is generally weaker with the reversed stochastic models. The work gives a broad statistical account of deterministic and stochastically reversed map models, such as are emerging in random number generation, communica tion systems and cryptography     

GENDER, EARNINGS, AND THE ENGLISH SKILL ACQUISITION OF HISPANIC WORKERS IN THE UNITED STATES

Using the 1980 and 1990 Public Use Microdata Samples, we find that labor market outcomes associated with English proficiency vary with respect to gender. For example, a synthetic cohort analysis provides evidence of gender-related differences in Hispanic workers' English skill acquisition. Moreover, we observe that Hispanic women face a lower English deficiency earnings penalty that rises more sharply with education than the penalty obtained by their otherwise similar male peers. Finally, English fluency appears to serve as a stronger occupational sorting mechanism for women than men. ( JEL J3, J1)     

Practicing managed care in the academic medical center

The purpose of this article is to outline the contrasts between the traditional AMC and an organization oriented toward the delivery of population-based managed care. Academic medical centers differ from one another considerably in the extent to which they serve as quaternary care community resources, the degree to which they emphasize primary care in training and care delivery, and the amount of research undertaken. Nor is there a single organizational structure for managed care; successful managed care is practices in IPAs, multispecialty groups, PHOs, and staff-model HMOs. Nonetheless, the contrasts outlined here between AMCs and managed care organizations (MCOs) are valid in most cases.     

The implementation of the healthy cities project in developing countries: lessons from Chittagong

The Chittagong Healthy City Project was carried out in late 1994 in Chittagong, Bangladesh. This paper presents findings of an evaluation of the project based upon internationally generated process indicators related to the institutional aspects of the project. The following issues are discussed with regard to project implementation: the institutional organization of local authorities, institutions' conceptual understanding of the project, formal insertion of the project into public authorities' activities, institutional leadership of the project, central-local relations, the lack of interministerial coordination, the project's office, international projects, and community organization. Giving consideration to these issues may help program planners detect problems in forthcoming projects prior to their implementation.     

The majority legal status of women in southern Africa: implications for women and families

Women in many countries of southern Africa do not have majority status or have only recently gained this right. Majority status grants individuals adult legal status and the right to bring matters to court, own and administer property, have legal custody of children, and contract for marriage. This article summarizes the legal status of women in Botswana, Lesotho, South Africa, Swaziland, Zambia, and Zimbabwe. Lack of majority status contributes to the ongoing risk of poverty for women and makes them overly dependent on men. Compounding the situation in these countries is the presence of a dual legal system. Improving the situation of women and their families involves targeting changes in the legal system, influencing implementation of laws, educating women about their rights, and giving women needed support to seek their legal rights. The legal status of women must be viewed in the context of historical changes in the economic, educational, political, and cultural developments of society.The research for this paper was conducted during her previous faculty affiliation with the University of Michigan, School of Social Work. Her research interests include the impact of social change on women and families, rural mental health services, and health and mental health linkages. She received her Ph.D. from Rutgers University and her M.S.W. from the Columbia University School of Social Work.Her research interests are gender, work and family, culture and power, social transformations in rural and peri-urban communities, critical education, women and community politics, history of women, family and kinship issues, and community-based program development and evaluation. She received her M.S.W. from Dalhousie University, Halifax.     

Statistics in epidemiology: the case-control study

This article presents a general review of the major trends in the conceptualization, development, and success of case-control methods for the study of disease causation and prevention. "Recent work on nested case-control, case-cohort, and two-stage case control designs demonstrates the continuing impact of statistical thinking on epidemiology. The influence of R. A. Fisher's work on these developments is mentioned wherever possible. His objections to the drawing of causal conclusions from observational data on cigarette smoking and lung cancer are used to introduce the problems of measurement error and confounding bias."     

Characterizations,length-biasing,and infinite divisibility

SupposeL(X) is the law of a positive random variableX, andZ is positive and independent ofX. Admissible solution pairs (L(X),L(Z)) are sought for the in-law equation $\hat X \cong X o Z$ °Z, where $L\left( {\hat X} \right)$ is a weighted law constructed fromL(X), and ° is a binary operation which in some sense is increasing. The class of weights includes length biasing of arbitrary order. When ° is addition and the weighting is ordinary length biasing, the class of admissibleL(X) comprises the positive infinitely divisible laws. Examples are given subsuming all known specific cases. Some extensions for general order of length-biasing are discussed.