Boosting multi-state models

One important goal in multi-state modelling is to explore information about conditional transition-type-specific hazard rate functions by estimating influencing effects of explanatory variables. This may be performed using single transition-type-specific models if these covariate effects are assumed to be different across transition-types. To investigate whether this assumption holds or whether one of the effects is equal across several transition-types (cross-transition-type effect), a combined model has to be applied, for instance with the use of a stratified partial likelihood formulation. Here, prior knowledge about the underlying covariate effect mechanisms is often sparse, especially about ineffectivenesses of transition-type-specific or cross-transition-type effects. As a consequence, data-driven variable selection is an important task: a large number of estimable effects has to be taken into account if joint modelling of all transition-types is performed. A related but subsequent task is model choice: is an effect satisfactory estimated assuming linearity, or is the true underlying nature strongly deviating from linearity? This article introduces component-wise Functional Gradient Descent Boosting (short boosting) for multi-state models, an approach performing unsupervised variable selection and model choice simultaneously within a single estimation run. We demonstrate that features and advantages in the application of boosting introduced and illustrated in classical regression scenarios remain present in the transfer to multi-state models. As a consequence, boosting provides an effective means to answer questions about ineffectiveness and non-linearity of single transition-type-specific or cross-transition-type effects.     

Cox regression with missing covariate data using a modified partial likelihood method

Missing covariate values is a common problem in survival analysis. In this paper we propose a novel method for the Cox regression model that is close to maximum likelihood but avoids the use of the EM-algorithm. It exploits that the observed hazard function is multiplicative in the baseline hazard function with the idea being to profile out this function before carrying out the estimation of the parameter of interest. In this step one uses a Breslow type estimator to estimate the cumulative baseline hazard function. We focus on the situation where the observed covariates are categorical which allows us to calculate estimators without having to assume anything about the distribution of the covariates. We show that the proposed estimator is consistent and asymptotically normal, and derive a consistent estimator of the variance–covariance matrix that does not involve any choice of a perturbation parameter. Moderate sample size performance of the estimators is investigated via simulation and by application to a real data example.     

A Proportional Hazards Regression Model for the Subdistribution with Covariates‐adjusted Censoring Weight for Competing Risks Data

With competing risks data, one often needs to assess the treatment and covariate effects on the cumulative incidence function. Fine and Gray proposed a proportional hazards regression model for the subdistribution of a competing risk with the assumption that the censoring distribution and the covariates are independent. Covariate‐dependent censoring sometimes occurs in medical studies. In this paper, we study the proportional hazards regression model for the subdistribution of a competing risk with proper adjustments for covariate‐dependent censoring. We consider a covariate‐adjusted weight function by fitting the Cox model for the censoring distribution and using the predictive probability for each individual. Our simulation study shows that the covariate‐adjusted weight estimator is basically unbiased when the censoring time depends on the covariates, and the covariate‐adjusted weight approach works well for the variance estimator as well. We illustrate our methods with bone marrow transplant data from the Center for International Blood and Marrow Transplant Research. Here, cancer relapse and death in complete remission are two competing risks.     

Staying strong: gender ideologies among African-American adolescents and the implications for HIV/STI prevention

This paper explores adolescents' definitions of what it means to be a man and a woman, the psycho-social context surrounding the formation of gender ideologies and their relationship to HIV/STI prevention. Semi-structured, in-depth interviews were conducted with fifty African-American adolescents living in Baltimore, Maryland. Female gender ideologies included economic independence, emotional strength and caretaking. Male gender ideologies emphasized financial responsibility, toughness and sexual prowess. Findings suggest that stronger adherence to male gender ideologies related to toughness and sexual prowess is influenced by male participants' perceived inability to fulfill their primary gender role as economic providers and the importance of gaining approval from male peers in the absence of adult male role models. Stronger adherence to female gender ideologies related to emotional strength and caretaking may be linked to a heightened desire for male intimacy and tolerance of male sexual risk behavior. Implications of the gender ideologies documented and their commonalities are discussed in terms of HIV/STI prevention.     

Home study methods for evaluating prospective resource families: history, current challenges, and promising approaches

Every state requires a home study before the placement of foster children for adoption. This article examines the history of home studies, presents results from expert interviews on the changing processes and purposes of home studies, and explores current challenges for the field. The article also introduces the Structured Analysis Family Evaluation (SAFE), a uniform home study format that encourages consistent family evaluations across workers, agencies, and jurisdictions. The article clarifies how SAFE may address challenges facing foster care and adoption practice.     

Relationship of interpersonal behaviors and health-related control appraisals to patient satisfaction and compliance in a university health center

OBJECTIVE: The authors' aim was to evaluate patient-provider relationships in a college health center. PARTICIPANTS: Eighty student patients and their health-care providers. METHODS: Patients completed a measure of perceived health competence before a consultation and measures of provider participatory behavior and interpersonal behavior before and after the consultation. They evaluated their satisfaction with care and compliance after the consultation and again 2 weeks later. Providers completed measures of their participatory behavior and patients' interpersonal behavior after the consultation. RESULTS: Patients preferred to be well informed and to have their preferences taken into account, and generally felt competent at managing their own health affairs. They indicated they obtained the high level of participation they desired. Patients desired and actually experienced friendly and submissive providers. Degree of match between patients' desired and actual level of involvement in their care was associated with greater satisfaction. A greater match between the extent to which they desired the provider to be affiliative and the provider's actual affiliative behavior was associated with more satisfaction. No variables were predictive of patient compliance. CONCLUSION: The authors discuss results in terms of the influence of situational factors characteristic of a college health center.     

Unplanned sexual activity as a consequence of alcohol use: a prospective study of risk perceptions and alcohol use among college freshmen

OBJECTIVE: The authors' goal was to show how risk perceptions regarding unplanned sexual activity following alcohol use are prospectively related to subsequent alcohol consumption. PARTICIPANTS: Undergraduate students (N = 380) completed questionnaires at 2 time points during their freshman year. METHODS: In the middle of the academic year (T1), students estimated their risk of engaging in unplanned sex and reported their alcohol use during the previous term. Four months later (T2), they again reported alcohol use and indicated whether they had engaged in unplanned sex since T1. RESULTS: Students who consumed more alcohol at T1 rated their risk of unplanned sex more highly, suggesting relative accuracy. Those with higher risk perceptions consumed more alcohol at T2 (controlling for T1 use), suggesting that they maintained the high-risk behavior. Last, those who were unrealistically optimistic (ie, estimated low risk at T1 yet had unplanned sex by T2) reported greater alcohol use at T2. CONCLUSIONS: These findings highlight the role that risk perceptions regarding sexual activity may play in college students' alcohol use.     

A family systems-based model of organizational intervention

Employee assistance professionals are expected to be proficient at intervening in organizations and creating meaningful behavioral change in interpersonal functioning. Because of their training in family systems theories and concepts, marriage and family therapists (MFTs) are well suited to serve organizations as "systems consultants." Unfortunately, the authors were unable to identify any family systems-based models for organizational intervention that have been empirically tested and supported. In this article, the authors present a family systems-based model of intervention that they developed while working in an employee assistance program (EAP). They also present research that was used to refine the model and to provide initial support for its effectiveness.     

Mixture of latent trait analyzers for model-based clustering of categorical data

Model-based clustering methods for continuous data are well established and commonly used in a wide range of applications. However, model-based clustering methods for categorical data are less standard. Latent class analysis is a commonly used method for model-based clustering of binary data and/or categorical data, but due to an assumed local independence structure there may not be a correspondence between the estimated latent classes and groups in the population of interest. The mixture of latent trait analyzers model extends latent class analysis by assuming a model for the categorical response variables that depends on both a categorical latent class and a continuous latent trait variable; the discrete latent class accommodates group structure and the continuous latent trait accommodates dependence within these groups. Fitting the mixture of latent trait analyzers model is potentially difficult because the likelihood function involves an integral that cannot be evaluated analytically. We develop a variational approach for fitting the mixture of latent trait models and this provides an efficient model fitting strategy. The mixture of latent trait analyzers model is demonstrated on the analysis of data from the National Long Term Care Survey (NLTCS) and voting in the U.S. Congress. The model is shown to yield intuitive clustering results and it gives a much better fit than either latent class analysis or latent trait analysis alone.