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.     

Going to Pieces: Paul Monette and the Fragments of the Tribe

Author of the highly acclaimed Borrowed Time: An AIDS Memoir (1988) and winner of the 1992 National Book Award for his autobiography Becoming a Man: Half a Life Story, Paul Monette died of complications from AIDS in Los Angeles on February 12, 1995. His most recent work, Last Watch of the Night: Essays Too Personal and Otherwise (1994), is his third volume in a trilogy of autobiographical writing on growing up gay, coming-out, and living with AIDS. A graduate of Phillips Academy and Yale University, Monette received a number of awards for his writing and his work as an outspoken voice in the battle against AIDS, including the National Book Award, three Lambda Literary awards, the PEN Center West Freedom to Write Award, and three honorary doctorate degrees, from Wesleyan University, the City University of New York, and the State University of New York at Oswego. A memorial service was held for Monette on Sunday, February 19, 1995, at the Directors Guild of America on Sunset Boulevard in Los Angeles. Paul Monette was 49.     

Ability, disability, and the oldest old

A measure of range of ability is used to profile the 85-years-old-and-older (oldest old) population, including the highly disabled institutional population. This new measure uses two new questions available in the 1990 Decennial Census concerning a self-care limitation and a mobility limitation as well as the usual question concerning a work limitation. In addition to examining the extent of disability among the oldest old, the article examines the extent of care potentially available in the household as well as the economic characteristics of this age group. It is also profiled in terms of relevant personal characteristics, including age, gender, marital status, race, ethnicity, rural residence, education, and employment. A key question addressed is the need for help or care among the oldest old and how various long-term care proposals would meet such needs. A careful analysis of this unique and growing population is necessary to both allay fears of the cost of care or help as well as to dispel stereotypes of this age group as frail and dependent, and in need of institutional care.     

Planning for productivity

In the 1980s the subject of productivity will receive increasing attention as the old macro economic policies are replaced by micro economic policies. Everyone in various ways will discover the changes which will be involved in learning the real economic facts of life. The need to improve productivity and business performance is urgent and yet it is an area which is being neglected by management.     

The certainty of judgments in health evaluations

Most evaluators seem to presume that all evaluative studies should result in conclusive evidence. Often, however, less than conclusive evidence may not only suffice, but be the only real alternative under conditions of limited evaluation resources. Three levels of evidence common to evaluation studies are discussed: suggestive evidence, preponderant evidence, and conclusive evidence. Ten factors are also presented which can be used to determine the level of certainty appropriate for a given evaluation study. Consideration of these factors enables the evaluator to specify when less than conclusive evidence is acceptable.     

Public opinion and the privatization of welfare: some theoretical implications

The social policy of the Thatcher government is characterized by an abrupt shift in the direction of the private sector. To what extent does this reflect what people want? The Institute of Economic Affairs conclude from the only suitable national opinion survey that such a move is strongly supported. Our reanalysis of their data shows that this strand in public opinion can coexist with, and need not contradict, an equal public enthusiasm for state welfare. Such results have important implications for our understanding of social policy. We conclude, therefore, with a discussion of contrasting marxist and liberal accounts which seeks to show that the evidence of ambivalence in popular attitudes about the welfare state supports particular developments in theory.     

Sensitivity of the Optimum to Perturbations of the Profit or Weight of an Item in the Binary Knapsack Problem

In the binary single constraint Knapsack Problem, denoted KP, we are given a knapsack of fixed capacity c and a set of n items. Each item j, j = 1,...,n, has an associated size or weight w_j and a profit p_j. The goal is to determine whether or not item j, j = 1,...,n, should be included in the knapsack. The objective is to maximize the total profit without exceeding the capacity c of the knapsack. In this paper, we study the sensitivity of the optimum of the KP to perturbations of either the profit or the weight of an item. We give approximate and exact interval limits for both cases (profit and weight) and propose several polynomial time algorithms able to reach these interval limits. The performance of the proposed algorithms are evaluated on a large number of problem instances.