A Bayesian hierarchical model for categorical longitudinal data from a social survey of immigrants

The paper investigates a Bayesian hierarchical model for the analysis of categorical longitudinal data from a large social survey of immigrants to Australia. Data for each subject are observed on three separate occasions, or waves, of the survey. One of the features of the data set is that observations for some variables are missing for at least one wave. A model for the employment status of immigrants is developed by introducing, at the first stage of a hierarchical model, a multinomial model for the response and then subsequent terms are introduced to explain wave and subject effects. To estimate the model, we use the Gibbs sampler, which allows missing data for both the response and the explanatory variables to be imputed at each iteration of the algorithm, given some appropriate prior distributions. After accounting for significant covariate effects in the model, results show that the relative probability of remaining unemployed diminished with time following arrival in Australia.

Diagnosis of diabetes mellitus: case of different cutoff values: A data mining approach

Objective: An estimation of cut-off points for the diagnosis of diabetes mellitus (DM) based on individual risk factors. Methods: A subset of the 1991 Oman National Diabetes Survey is used, including all patients with a 2h post glucose load >= 200 mg/dl (278 subjects) and a control group of 286 subjects. All subjects previously diagnosed as diabetic and all subjects with missing data values were excluded. The data set was analyzed by use of the SPSS Clementine data mining system. Decision Tree Learners (C5 and CART) and a method for mining association rules (the GRI algorithm) are used. The fasting plasma glucose (FPG), age, sex, family history of diabetes and body mass index (BMI) are input risk factors (independent variables), while diabetes onset (the 2h post glucose load >= 200 mg/dl) is the output (dependent variable). All three techniques used were tested by use of crossvalidation (89.8%). Results: Rules produced for diabetes diagnosis are: A- GRI algorithm (1) FPG>=108.9 mg/dl, (2) FPG>=107.1 and age>39.5 years. B- CART decision trees: FPG >=110.7 mg/dl. C- The C5 decision tree learner: (1) FPG>=95.5 and 54, (2) FPG>=106 and 25.2 kg/m2. (3) FPG>=106 and =133 mg/dl. The three techniques produced rules which cover a significant number of cases (82%), with confidence between 74 and 100%. Conclusion: Our approach supports the suggestion that the present cut-off value of fasting plasma glucose (126 mg/dl) for the diagnosis of diabetes mellitus needs revision, and the individual risk factors such as age and BMI should be considered in defining the new cut-off value.

Implications of non-response of older women to a short form of the Center for Epidemiologic Studies Depression Scale

The Center for Epidemiologic Studies Depression Scale (CES-D) is frequently used in epidemiological surveys to screen for depression, especially among older adults. This article addresses the problem of non-completion of a short form of the CES-D (CESD-10) in a mailed survey of 73- to 78-year-old women enrolled in the Australian Longitudinal Study on Women's Health. Completers of the CESD-10 had more education, found it easier to manage on available income and reported better physical and mental health. The Medical Outcomes Study Short Form Health Survey (SF-36) scores for non-completers were intermediate between those for women classified as depressed and not depressed using the CESD-10. Indicators of depression had an inverted U-shaped relationship with the number of missing CESD- 10 items and were most frequent for women with two to seven items missing. Future research should pay particular attention to the level of missing data in depression scales and report its potential impact on estimates of depression.

Analysis of Melanoma Onset: Assessing Familial Aggregation by Using Estimating Equations and Fitting Variance Components via Bayesian Random Effects Models

We investigate whether relative contributions of genetic and shared environmental factors are associated with an increased risk in melanoma. Data from the Queensland Familial Melanoma Project comprising 15,907 subjects arising from 1912 families were analyzed to estimate the additive genetic, common and unique environmental contributions to variation in the age at onset of melanoma. Two complementary approaches for analyzing correlated time-to-onset family data were considered: the generalized estimating equations (GEE) method in which one can estimate relationship-specific dependence simultaneously with regression coefficients that describe the average population response to changing covariates; and a subject-specific Bayesian mixed model in which heterogeneity in regression parameters is explicitly modeled and the different components of variation may be estimated directly. The proportional hazards and Weibull models were utilized, as both produce natural frameworks for estimating relative risks while adjusting for simultaneous effects of other covariates. A simple Markov Chain Monte Carlo method for covariate imputation of missing data was used and the actual implementation of the Bayesian model was based on Gibbs sampling using the free ware package BUGS. In addition, we also used a Bayesian model to investigate the relative contribution of genetic and environmental effects on the expression of naevi and freckles, which are known risk factors for melanoma.

Protocols for listing threatened species can forecast extinction

Risk-ranking protocols are used widely to classify the conservation status of the world's species. Here we report on the first empirical assessment of their reliability by using a retrospective study of 18 pairs of bird and mammal species (one species extinct and the other extant) with eight different assessors. The performance of individual assessors varied substantially, but performance was improved by incorporating uncertainty in parameter estimates and consensus among the assessors. When this was done, the ranks from the protocols were consistent with the extinction outcome in 70-80% of pairs and there were mismatches in only 10-20% of cases. This performance was similar to the subjective judgements of the assessors after they had estimated the range and population parameters required by the protocols, and better than any single parameter. When used to inform subjective judgement, the protocols therefore offer a means of reducing unpredictable biases that may be associated with expert input and have the advantage of making the logic behind assessments explicit. We conclude that the protocols are useful for forecasting extinctions, although they are prone to some errors that have implications for conservation. Some level of error is to be expected, however, given the influence of chance on extinction. The performance of risk assessment protocols may be improved by providing training in the application of the protocols, incorporating uncertainty in parameter estimates and using consensus among multiple assessors, including some who are experts in the application of the protocols. Continued testing and refinement of the protocols may help to provide better absolute estimates of risk, particularly by re-evaluating how the protocols accommodate missing data.

Multiple imputation for body mass index: lessons from the Australian Longitudinal Study on Women's Health

In large epidemiological studies missing data can be a problem, especially if information is sought on a sensitive topic or when a composite measure is calculated from several variables each affected by missing values. Multiple imputation is the method of choice for 'filling in' missing data based on associations among variables. Using an example about body mass index from the Australian Longitudinal Study on Women's Health, we identify a subset of variables that are particularly useful for imputing values for the target variables. Then we illustrate two uses of multiple imputation. The first is to examine and correct for bias when data are not missing completely at random. The second is to impute missing values for an important covariate; in this case omission from the imputation process of variables to be used in the analysis may introduce bias. We conclude with several recommendations for handling issues of missing data. Copyright (C) 2004 John Wiley Sons, Ltd.

Comparability of skin screening histories obtained by telephone interviews and mailed questionnaires: A randomized crossover study

The comparability of information collected through telephone interviews and information collected through mailed questionnaires has not been well studied. As part of the first phase of a randomized controlled trial of population screening for melanoma in Queensland, Australia, the authors compared histories of skin examination reported in telephone interviews and self-administered mailed questionnaires. A total of 1,270 subjects each completed a telephone interview and a mailed questionnaire 1 month apart in 1999; 564 subjects received the interview first, and 706 received the mailed questionnaire first. Agreement between the two methods was 91.2% and 88.6% for whole-body skin examination by a physician in the last 12 months and the last 3 years, respectively, and 81.9% for whole-body skin self-examination in the last 12 months. Agreement was lower for any skin self-examination. Agreement between the two methods was similar regardless of whether the interview or the questionnaire was administered first. Missing data were less frequent for interviews (0.5%) than for mailed questionnaires (3.8%). Costs were estimated at A$9.55 (US$6.21) per completed interview and A$3.01 (US$1.96) per questionnaire. The similarity of results obtained using telephone interviews and mailed questionnaires, coupled with the substantially higher cost of telephone interviews, suggests that self-administered mailed questionnaires are an appropriate method of assessing this health behavior.

Using patients' and rheumatologists' opinions to specify a short form of the WOMAC function subscale

Background: The WOMAC ( Western Ontario and McMaster Universities) function subscale is widely used in clinical trials of hip and knee osteoarthritis. Reducing the number of items of the subscale would enhance efficiency and compliance, particularly for use in clinical practice applications. Objective: To develop a short form of the WOMAC function subscale based on patients' and experts' opinions ( WOMAC function short form). Methods: WOMAC function subscale data ( Likert version) were obtained from 1218 outpatients with painful hip or knee osteoarthritis. These patients and their rheumatologists selected the five items that they considered most in need of improvement. The rheumatologists were asked to select the five items for which patients in general are the most impaired. Items that were least important to patients and experts, those with a high proportion of missing data, and those with a response distribution showing a floor or ceiling response were excluded, along with one of a pair of items with a correlation coefficient >0.75. Results: The WOMAC function short form included items 1, 2, 3, 6, 7, 8, 9, and 15 of the long form. The short form did not differ substantially from the long form in responsiveness ( standardised response mean of 0.84 v 0.80). Conclusions: A short form of the WOMAC function subscale was developed according to the views of patients and rheumatologists, based on the responses of 1218 patients and 399 rheumatologists. The clinical relevance and applicability of this WOMAC function subscale short form require further evaluation.

Differences in mail and telephone responses to self-rated health: use of multiple imputation in correcting for response bias

Objectives: To estimate differences in self-rated health by mode of administration and to assess the value of multiple imputation to make self-rated health comparable for telephone and mail. Methods: In 1996, Survey 1 of the Australian Longitudinal Study on Women's Health was answered by mail. In 1998, 706 and 11,595 mid-age women answered Survey 2 by telephone and mail respectively. Self-rated health was measured by the physical and mental health scores of the SF-36. Mean change in SF-36 scores between Surveys 1 and 2 were compared for telephone and mail respondents to Survey 2, before and after adjustment for socio-demographic and health characteristics. Missing values and SF-36 scores for telephone respondents at Survey 2 were imputed from SF-36 mail responses and telephone and mail responses to socio-demographic and health questions. Results: At Survey 2, self-rated health improved for telephone respondents but not mail respondents. After adjustment, mean changes in physical health and mental health scores remained higher (0.4 and 1.6 respectively) for telephone respondents compared with mail respondents (-1.2 and 0.1 respectively). Multiple imputation yielded adjusted changes in SF-36 scores that were similar for telephone and mail respondents. Conclusions and Implications: The effect of mode of administration on the change in mental health is important given that a difference of two points in SF-36 scores is accepted as clinically meaningful. Health evaluators should be aware of and adjust for the effects of mode of administration on self-rated health. Multiple imputation is one method that may be used to adjust SF-36 scores for mode of administration bias.

Interagency collaboration between child protection and mental health services: Practices, attitudes and barriers

Objective: The aim of this paper is to examine some of the factors that facilitate and hinder interagency collaboration between child protection services and mental health services in cases where there is a parent with a mental illness and there are protection concerns for the child(ren). The paper reports on agency practices, worker attitudes and experiences, and barriers to effective collaboration. Method: A self-administered, cross-sectional survey was developed and distributed via direct mail or via line supervisors to workers in statutory child protection services, adult mental health services, child and youth mental health services, and Suspected Child Abuse and Neglect (SCAN) Teams. There were 232 completed questionnaires returned, with an overall response rate of 21%. Thirty-eight percent of respondents were statutory child protection workers. 39% were adult mental health workers, 16% were child and youth mental health workers, and 4% were SCAN Team medical officers (with 3% missing data). Results: Analysis revealed that workers were engaging in a moderate amount of interagency contact, but that they were unhappy with the support provided by their agency. Principle components analysis and multivariate analysis of variance (MANOVA) on items assessing attitudes toward other workers identified four factors, which differed in rates of endorsement: inadequate training, positive regard for child protection workers, positive regard for mental health workers, and mutual mistrust (from highest to lowest level of endorsement). The same procedure identified the relative endorsement of five factors extracted from items about potential barriers: inadequate resources, confidentiality, gaps in interagency processes, unrealistic expectations, and professional knowledge domains and boundaries. Conclusions: Mental health and child protection professionals believe that collaborative practice is necessary; however, their efforts are hindered by a lack of supportive structures and practices at the organizational level. (c) 2005 Published by Elsevier Ltd.

Correcting for numerator/denominator bias when assessing changing inequalites in occupational class mortality, Australia 1981-2002

Objective Comparisons of the changing patterns of inequalities in occupational mortality provide one way to monitor the achievement of equity goals. However, previous comparisons have not corrected for numerator/denominator bias, which is a consequence of the different ways in which occupational details are recorded on death certificates and on census forms. The objective of this study was to measure the impact of this bias on mortality rates and ratios over time. Methods Using data provided by the Australian Bureau of Statistics, we examined the evidence for bias over the period 1981-2002, and used imputation methods to adjust for this bias. We compared unadjusted with imputed rates of mortality for manual/non-manual workers. Findings Unadjusted data indicate increasing inequality in the age-adjusted rates of mortality for manual/non-manual workers during 1981-2002, Imputed data suggest that there have been modest fluctuations in the ratios of mortality for manual/non-manual workers during this time, but with evidence that inequalities have increased only in recent years and are now at historic highs. Conclusion We found that imputation for missing data leads to changes in estimates of inequalities related to social class in mortality for some years but not for others. Occupational class comparisons should be imputed or otherwise adjusted for missing data on census or death certificates.

Customised birthweight: Coefficients for an Australian population and validation of the model

Background: Published birthweight references in Australia do not fully take into account constitutional factors that influence birthweight and therefore may not provide an accurate reference to identify the infant with abnormal growth. Furthermore, studies in other regions that have derived adjusted (customised) birthweight references have applied untested assumptions in the statistical modelling. Aims: To validate the customised birthweight model and to produce a reference set of coefficients for estimating a customised birthweight that may be useful for maternity care in Australia and for future research. Methods: De-identified data were extracted from the clinical database for all births at the Mater Mother's Hospital, Brisbane, Australia, between January 1997 and June 2005. Births with missing data for the variables under study were excluded. In addition the following were excluded: multiple pregnancies, births less than 37 completed week's gestation, stillbirths, and major congenital abnormalities. Multivariate analysis was undertaken. A double cross-validation procedure was used to validate the model. Results: The study of 42 206 births demonstrated that, for statistical purposes, birthweight is normally distributed. Coefficients for the derivation of customised birthweight in an Australian population were developed and the statistical model is demonstrably robust. Conclusions: This study provides empirical data as to the robustness of the model to determine customised birthweight. Further research is required to define where normal physiology ends and pathology begins, and which segments of the population should be included in the construction of a customised birthweight standard.

On the statistical analysis of the GS-NS0 cell proteome: Imputation, clustering and variability testing

We have undertaken two-dimensional gel electrophoresis proteomic profiling on a series of cell lines with different recombinant antibody production rates. Due to the nature of gel-based experiments not all protein spots are detected across all samples in an experiment, and hence datasets are invariably incomplete. New approaches are therefore required for the analysis of such graduated datasets. We approached this problem in two ways. Firstly, we applied a missing value imputation technique to calculate missing data points. Secondly, we combined a singular value decomposition based hierarchical clustering with the expression variability test to identify protein spots whose expression correlates with increased antibody production. The results have shown that while imputation of missing data was a useful method to improve the statistical analysis of such data sets, this was of limited use in differentiating between the samples investigated, and highlighted a small number of candidate proteins for further investigation. (c) 2006 Elsevier B.V. All rights reserved.

Small-scale randomized controlled trials need more powerful methods of mediational analysis than the Baron-Kenny method

Objective: To devise more-effective physical activity interventions, the mediating mechanisms yielding behavioral change need to be identified. The Baron-Kenny method is most commonly used. but has low statistical power and May not identify mechanisms of behavioral change in small-to-medium size Studies. More powerful statistical tests are available, Study Design and Setting: Inactive adults (N = 52) were randomized to either a print or a print-plus-telephone intervention. Walking and exercise-related social support Were assessed at baseline, after file intervention, and 4 weeks later. The Baron-Kenny and three alternative methods of mediational analysis (Freedman-Schatzkin; MacKinnon et al.: bootstrap method) were used to examine the effects of social support on initial behavior change and maintenance. Results: A significant mediational effect of social support on initial behavior change was indicated by the MacKinnon et al., bootstrap. and. marginally. Freedman-Schatzkin methods, but not by the Baron-Kenny method. No significant mediational effecl of social support on maintenance of walking was found. Conclusions: Methodologically rigorous intervention studies to identify mediators of change in physical activity are costly and labor intensive, and may not be feasible with large samples. The Use of statistically powerful tests of mediational effects in small-scale studies can inform the development of more effective interventions. (C) 2006 Elsevier Inc. All rights reserved.