ESTIMATING THE NUMBER OF VISITS TO THE DOCTOR

The frequency of doctor consultations has direct consequences for health care budgets, yet little statistical analysis of the determinants of doctor visits has been reported. We consider the distribution of the number of visits to the doctor and, in particular, we model its dependence on a number of demographic factors. Examination of the Australian 1995 National Health Survey data reveals that generalized linear Poisson or negative binomial models are inadequate for modelling the mean as a function of covariates, because of excessive zero counts, and a mean‐variance relationship that varies enormously over covariate values. A negative binomial model is used, with parameter values estimated in subgroups according to the discrete combinations of the covariate values. Smoothing splines are then used to smooth and interpolate the parameter values. In effect the mean and the shape parameters are each modelled as (different) functions of gender, age and geographical factors. The estimated regressions for the mean have simple and intuitive interpretations. However, the dependence of the (negative binomial) shape parameter on the covariates is more difficult to interpret and is subject to influence by extreme observations. We illustrate the use of the model by estimating the distribution of the number of doctor consultations in the Statistical Local Area of Ryde, based on population numbers from the 1996 census.     

Linear models with an inverse gaussian poisson error distribution

The inverse Gaussian-Poisson (two-parameter Sichel) distribution is useful in fitting overdispersed count data. We consider linear models on the mean of a response variable, where the response is in the form of counts exhibiting extra-Poisson variation, and assume an IGP error distribution. We show how maximum likelihood estimation may be carried out using iterative Newton-Raphson IRLS fitting, where GLIM is used for the IRLS part of the maximization. Approximate likelihood ratio tests are given.     

On-Line Learning for the Infinite Hidden Markov Model

We develop a sequential Monte Carlo algorithm for the infinite hidden Markov model (iHMM) that allows us to perform on-line inferences on both system states and structural (static) parameters. The algorithm described here provides a natural alternative to Markov chain Monte Carlo samplers previously developed for the iHMM, and is particularly helpful in applications where data is collected sequentially and model parameters need to be continuously updated. We illustrate our approach in the context of both a simulation study and a financial application.     

Supporting Critical Multicultural Teacher Educators: Transformative teaching,social justice education,and perceptions of institutional support

ABSTRACT

In most teacher education programmes in Canada and the United States, educators’ opportunities to develop equity-related skills are concentrated into single ‘multicultural’ courses. These courses tend to have a conservative or liberal orientation, focused on appreciating diversity or cultural competence, rather than a critical orientation, focused on preparing teachers to address inequity. In this study, based on a survey of instructors of multicultural and intercultural teacher education courses in Canada and the US (N = 186), we examined the relationship between the criticality of their multicultural teacher education courses and their perceptions of institutional support for the values they teach. We found a negative relationship between the two – the more critical the instructors’ approaches, the less institutional support they perceived.     

Targeting the worst-off for free health care: a process evaluation in Burkina Faso

Effective mechanisms to exempt the indigent from user fees at health care facilities are rare in Africa. A State-led intervention (2004-2005) and two action research projects (2007-2010) were implemented in a health district in Burkina Faso to exempt the indigent from user fees. This article presents the results of the process evaluation of these three interventions. Individual and group interviews were organized with the key stakeholders (health staff, community members) to document the strengths and weaknesses of key components of the interventions (relevance and uptake of the intervention, worst-off selection and information, financial arrangements). Data was subjected to content analysis and thematic analysis. The results show that all three intervention processes can be improved. Community-based targeting was better accepted by the stakeholders than was the State-led intervention. The strengths of the community-based approach were in clearly defining the selection criteria, informing the waiver beneficiaries, using a participative process and using endogenous funding. A weakness was that using endogenous funding led to restrictive selection by the community. The community-based approach appears to be the most effective, but it needs to be improved and retested to generate more knowledge before scaling up.     

Moving Beyond Basic Emotions in Aging Research

There is consistent evidence that older adults have difficulties in perceiving emotions. However, emotion perception measures to date have focused on one particular type of assessment: using standard photographs of facial expressions posing six basic emotions. We argue that it is important in future research to explore adult age differences in understanding more complex, social and blended emotions. Using stimuli which are dynamic records of the emotions expressed by people of all ages, and the use of genuine rather than posed emotions, would also improve the ecological validity of future research into age differences in emotion perception. Important questions remain about possible links between difficulties in perceiving emotional signals and the implications that this has for the everyday interpersonal functioning of older adults.     

Human Interindividual Variability in Susceptibility to Airborne Particles

Part of the explanation for the persistent epidemiological findings of associations between mortality and morbidity with relatively modest ambient exposures to airborne particles may be that some people are much more susceptible to particle-induced responses than others. This study assembled a database of quantitative observations of interindividual variability in pharmacokinetic and pharmacodynamic parameters likely to affect particle response. The pharmacodynamic responses studied included data drawn from epidemiologic studies of doses of methacholine, flour dust, and other agents that induce acute changes in lung function. In general, the amount of interindividual variability in several of these pharmacodynamic response parameters was greater than the variability in pharmacokinetic (breathing rate, deposition, and clearance) parameters. Quantitatively the results indicated that human interindividual variability of breathing rates and major pharmacokinetic parameters-total deposition and tracheobronchial clearance-were in the region of Log(GSD) = 0.1 to 0.2 (corresponding to geometric standard deviations of 10(.1)-10(.2) or 1.26-1.58). Deposition to the deep lung (alveolar region) appeared to be somewhat more variable: Log(GSD) of about 0.3 (GSD of about 2). Among pharmacodynamic parameters, changes in FEV1 in response to ozone and metabisulfite (an agent that is said to act primarily on neural receptors in the lung) were in the region of Log(GSD) of 0.2 to 0.4. However, similar responses to methacholine, an agent that acts on smooth muscle, seemed to have still more variability (0.4 to somewhat over 1.0, depending on the type of population studied). Similarly high values were suggested for particulate allergens. Central estimates of this kind of variability, and the close correspondence of the data to lognormal distributions, indicate that 99.9th percentile individuals are likely to respond at doses that are 150 to 450-fold less than would be needed in median individuals. It seems plausible that acute responses with this amount of variability could form part of the mechanistic basis for epidemiological observations of enhanced mortality in relation to ambient exposures to fine particles.