Gwendolyn E. P. Zahner

Goodness‐of‐fit for GEE: an example with mental health service utilization

Suppose we use generalized estimating equations to estimate a marginal regression model for repeated binary observations. There are no established summary statistics available for assessing the adequacy of the fitted model. In this paper we propose a goodness-of-fit test statistic which has an approximate chi-squared distribution when we have specified the model correctly. The proposed statistic can be viewed as an extension of the Hosmer and Lemeshow goodness-of-fit statistic for ordinary logistic regression to marginal regression models for repeated binary responses. We illustrate the methods using data from a study of mental health service utilization by children. The repeated responses are a set of binary measures of service use. We fit a marginal logistic regression model to the data using generalized estimating equations, and we apply the proposed goodness-of-fit statistic to assess the adequacy of the fitted model.

Multivariate Logistic Models for Incomplete Binary Responses

Abstract In this article we describe a likelihood-based regression model appropriate for analyzing incomplete multivariate binary responses. We focus on “marginal models”; that is, models where the marginal mean or expectation of the binary response is related to a set of covariates. The association between the binary responses is modeled in terms of conditional log odds ratios. When the nonresponse mechanism is ignorable, it is not necessary to specify a nonresponse model, and valid inferences can be obtained provided that the likelihood for the responses has been correctly specified. But when the nonresponse mechanism is nonignorable, valid inferences can only be obtained by incorporating a model for nonresponse. An unresolved issue with nonignorable models concerns the identifiability of the parameters. So far, no general and practically useful necessary and sufficient conditions for identifiability are available. Here we suggest some simple procedures for examining the identifiability status of nonign...

Loss in childhood: Anxiety in adulthood

Recent research, especially in Great Britain, has attracted interest by reporting on the relationship between maternal loss and vulnerability to depression among women. Several studies in the United States that included men have not received equal attention. The present study expands on the US work by reporting findings from the Queensbrook Study in New York City, a cross-sectional survey that provides information about the relationships between the family environment of childhood and the prevalence of psychiatric illness in adulthood. The Queensbrook survey was conducted in the mid 1960 as an urban counterpart to the Stirling County Study in rural Atlantic Canada. The data from the urban sample described here were not published earlier, and for this report we used DSM-111 criteria to develop scoring algorithme to identify depression and anxiety. We investigated several types of adverse childhood losses, not solely the death of a mother, and related them to depression and anxiety in both men and women. None of the childhood experience was significant associated with these disorders among women, nor was the death of a parent related to either type of disorder among men. However, boys who left home before 16 years of age, whose parents were divorced or separated, or who were placed in an adopted family had a threefold increase in rates of anxiety as adults. This finding of a positive association between the divorce of parents and later anxiety in men is supported by several of the other population surveys carried out in the United States.