Lowest Bayesian Information Criterion Research Articles

Impact of air pollution on cause-specific mortality in Korea: Results from Bayesian Model Averaging and Principle Component Regression approaches

Health effects related to air pollution are a major global concern. Related studies based on reliable exposure assessment methods would potentially enable policy makers to propose appropriate environmental management policies. In this study, integrated Bayesian Model Averaging (BMA) and Principle Component Regression (PCR) were adopted to assess the severity of air pollution impacts on mortality related to circulatory, respiratory and skin diseases in 25 districts of Seoul, South Korea for the years 2005–2015. These methods were consistent in determining the best regression models and most important pollutants related to mortality in those highly susceptible to poor air quality. Specifically, the results demonstrated that pneumonia was highly associated with air pollution, with a large determination coefficient (BMA: 0.46, PCR: 0.51) and high model's posterior probability (0.47). The most reliable prediction model for pneumonia was indicated by the lowest Bayesian Information Criterion. Among the pollutants, particulate matter with an aerodynamic diameter of 10 μm or less (PM10) was associated with serious health risks on evaluation, with the highest posterior inclusion probabilities (range, 80.20 to 100.00%) and significantly positive correlation coefficients (range, 0.14 to 0.34, p < 0.05). In addition, excessive PM10 concentration (approximately 2.54 times the threshold) and a continuous increase in mortality due to respiratory diseases (approximately 1.50-fold in 10 years) were also exhibited. Overall, the results of this study suggest that currently, socio-environmental policies and international collaboration to mitigate health effects of air pollution is necessary in Seoul, Korea. Moreover, consideration of uncertainty of the regression model, which was verified in this research, will facilitate further application of this approach and enable optimal prediction of interactions between human and environmental factors.

Road traffic accidents prediction modelling: An analysis of Anambra State, Nigeria

One of the major problems in the world today is the rate of road traffic crashes and deaths on our roads. Majority of these deaths occur in low-and-middle income countries including Nigeria. This study analyzed road traffic crashes in Anambra State, Nigeria with the intention of developing accurate predictive models for forecasting crash frequency in the State using autoregressive integrated moving average (ARIMA) and autoregressive integrated moving average with explanatory variables (ARIMAX) modelling techniques. The result showed that ARIMAX model outperformed the ARIMA (1,1,1) model generated when their performances were compared using the lower Bayesian information criterion, mean absolute percentage error, root mean square error; and higher coefficient of determination (R-Squared) as accuracy measures. The findings of this study reveal that incorporating human, vehicle and environmental related factors in time series analysis of crash dataset produces a more robust predictive model than solely using aggregated crash count. This study contributes to the body of knowledge on road traffic safety and provides an approach to forecasting using many human, vehicle and environmental factors. The recommendations made in this study if applied will help in reducing the number of road traffic crashes in Nigeria.

Data-driven identification of endophenotypes of Alzheimer\u2019s disease progression: implications for clinical trials and therapeutic interventions

BackgroundGiven the complex and progressive nature of Alzheimer’s disease (AD), a precision medicine approach for diagnosis and treatment requires the identification of patient subgroups with biomedically distinct and actionable phenotype definitions.MethodsLongitudinal patient-level data for 1160 AD patients receiving placebo or no treatment with a follow-up of up to 18 months were extracted from an integrated clinical trials dataset. We used latent class mixed modelling (LCMM) to identify patient subgroups demonstrating distinct patterns of change over time in disease severity, as measured by the Alzheimer’s Disease Assessment Scale—cognitive subscale score. The optimal number of subgroups (classes) was selected by the model which had the lowest Bayesian Information Criterion. Other patient-level variables were used to define these subgroups’ distinguishing characteristics and to investigate the interactions between patient characteristics and patterns of disease progression.ResultsThe LCMM resulted in three distinct subgroups of patients, with 10.3% in Class 1, 76.5% in Class 2 and 13.2% in Class 3. While all classes demonstrated some degree of cognitive decline, each demonstrated a different pattern of change in cognitive scores, potentially reflecting different subtypes of AD patients. Class 1 represents rapid decliners with a steep decline in cognition over time, and who tended to be younger and better educated. Class 2 represents slow decliners, while Class 3 represents severely impaired slow decliners: patients with a similar rate of decline to Class 2 but with worse baseline cognitive scores. Class 2 demonstrated a significantly higher proportion of patients with a history of statins use; Class 3 showed lower levels of blood monocytes and serum calcium, and higher blood glucose levels.ConclusionsOur results, ‘learned’ from clinical data, indicate the existence of at least three subgroups of Alzheimer’s patients, each demonstrating a different trajectory of disease progression. This hypothesis-generating approach has detected distinct AD subgroups that may prove to be discrete endophenotypes linked to specific aetiologies. These findings could enable stratification within a clinical trial or study context, which may help identify new targets for intervention and guide better care.

Seasonal behavior and forecasting trends of tuberculosis incidence in Holy Kerbala, Iraq.

Pulmonary tuberculosis (PTB) remains major public health problem over the world. Cities witnessing religious event throughout of the year like Kerbala/Iraq require great efforts to minimize the incidence of deadly communicable diseases like TB. The aim of this study is to model the monthly incidence rates of PTB cases in Kerbala/Iraq. This is a retrospective study in which records of confirmed PTB patients whom they referred to the chest and respiratory illnesses center of Holy Kerbala governorate were obtained. Monthly registered new smear-positive PTB cases from January 2010 to December 2016 were analyzed. Seasonal autoregressive integrated moving average (SARIMA), SARIMA-exponential smoothing method (ETS), SARIMA-neural network autoregressive, and SARIMA-adaptive neuro-fuzzy inference system (SARIMA-ANFIS) were used for forecasting monthly incidence rate of TB in Kerbala, Iraq. Mean absolute percentage error, root mean square error, and mean absolute square error were used to compare the models, and Akaike information criterion (AIC) and Bayesian information criterion (BIC) were used to selected best model. The trend of PTB incidence showed a seasonal characteristic, with peaks in spring and winter. Predicted estimates using all models proposed to forecast the number of PTB cases from 2016 to 2018 showed that the PTB cases indicated marginal decrease trends and best forecasted in SARIMA-ANFIS model (the lower AIC and BIC values, 712.69 and 731.05, respectively). Seasonal characteristic of PTB incidence was observed with peaks during spring and winter. Forecasting of PTB incidence between the period 2016 and 2018 showed marginal decrease trends, and the best forecasting model was SARIMA-ANFIS model.

Nonstationary extreme value analysis of temperature extremes in China

In a changing climate, the common assumption of stationarity of climate extremes has been increasingly challenged, raising the need to incorporate non-stationarity in extreme value modeling. In this study, quantile regression is used to identify the trends of annual temperature extremes and their correlations with two large climate patterns, the western Pacific subtropical high (WPSH) and the Arctic Oscillation (AO) at 357 stations in China. Statistical significant positive trends and correlations between warm (or cold) temperature extremes and WPSH (or AO) have been detected at most stations. The influence of WPSH on warm extremes is significant in southern China, while the AO mainly affects the cold extremes in northern and eastern China. Then, annual temperature extremes are fitted to generalized extreme value (GEV) distributions with time-varying parameters. The summer (or winter) mean daily maximum (or minimum) temperatures and two climate indices, the WPSH index and the AO index, are chosen as covariates. In total, 16 candidate GEV distribution models are constructed, and the best fitting model with the lowest Bayesian information criterion (BIC) is selected. The 20-year return levels of annual warm (or cold) extremes in the period 1961–1980 and 1991–2010 are computed and compared. The changes of 20-year return levels of annual warm and cold extremes are jointly determined by trend and distributional changes of annual temperature extremes. Analysis of large scale atmospheric circulation changes indicate that a strengthening anticyclonic circulation and increasing geopotential height in recent decades may have contributed to the changes in temperature extremes in China.

LO17: A comparative evaluation of ED crowding metrics and associations with patient mortality

Introduction: Over 700 different input, throughput and output metrics have been used to quantify ED crowding. Of these, only ED length-of-stay (ED LOS) has been shown to be associated with mortality. No comparative evaluation of ED crowding metrics has been performed to determine which ones have the strongest association with patient mortality. The objective of this study was to compare the strength of association of common ED input, throughput and output metrics to patient mortality. Methods: Administrative data from five years of ED visits (2011-2014) at three urban EDs were linked to develop a database of over 900,000 ED visits with patient demographics, electronic time stamps for care processes, dispositions and outcomes. The data were randomly divided into three partitions of equal size. Here we report the findings from one partition of 253,938 ED visits. The remaining two data partitions will be used to validate these findings. Commonly-used crowding metrics were quantified and aggregated by day or by shift (0800-1600, 1600-2400, 2400-0800), and the shift-specific metrics assigned to each patient. The primary outcome was 7-day all-cause mortality. Multilevel logistic regression models were developed for 7-day mortality, with selected ED crowding metrics and a common set of confounders as predictors. The strength of association between the crowding metrics and mortality was compared using Akaike’s Information Criterion (AIC) and the Bayesian Information Criterion (BIC): ED crowding metrics with lower AIC and BIC have stronger associations with 7-day mortality. Results: Of 909,000 ED encounters, 124,679 (16.5%) arrived by EMS, 149,233(19.7%) were admitted, and 3,808 patients (0.5%) died within 7 days of ED arrival. Of input metrics, the model with ED wait-time was better (i.e. had a smaller AIC and BIC) than models for daily census, ED occupancy or LWBS proportion for predicting 7-day mortality. Of throughput metrics, the model with mean ED LOS was better than the model for mean MD care time. Of output metrics, the model with daily inpatient hospital occupancy was better than the model with mean boarding time. Conclusion: Based on one data partition, regression models based on the average wait-time, ED LOS and inpatient occupancy best predicted 7-day mortality. These results will be validated in the two other data partitions to confirm the best-performing ED input, throughput and output metrics.

English

The phylogenetic utility of the 16S gene in cichlids is assessed. Eighty-six (86) partial sequences belonging to 37 genera of cichlids from the Genbank was analyzed. The alignment had four hundred and sixty three (463) basepairs with 337 conserved sites and 126 variable sites. Base compositional bias is similar to that found in higher organism with Adenine having the highest average of 30.3%, followed by cytosine, guanine and thiamine with the average values of 26.1, 21.9 and 21.7% respectively. The most suitable evolutionary model is the K2+G+I model as this had the lowest Bayesian Information Criterion. There were 4 major indels at basepair positions 328 which is unique to the Heterotilapia buttikoferi, position 369 unique to Gramatoria lemarii, position 396 which is shared by Tilapia sparrmanii, T. guinasana and T. zilli. The indel at position 373 was found in all tested species except the Oreochromis mossambicus. The Tilapine general is the basal group in Cichlids. The 16S gene separates the Tilapia genera without any ambiguity but there were phylogenetic overlaps in the Sarotherodon and Oreochromis. More finite molecular and statistical methodology may be needed to distinguish the Sarotherodon and Oreochromis. The diversity of cichlids is generally very low due to a common ancestry with little differentiation genetically. The grouping of the Oreochromis and Sarotherodon genera together in the same clade is not unconnected with the preservation of genetic beacons that the group retained as it evolved. Key words: Cichlids, 16S gene, phylogeny, evolutionary model, monophyletic, conserved segment, speciose, indels

Time Series Models for Short Term Prediction of the Incidence of Japanese Encephalitis in Xianyang City, P R China△.

Objective To construct a model of Seasonal Autoregressive Integrated Moving Average (SARIMA) for forecasting the epidemic of Japanese encephalitis (JE) in Xianyang, Shaanxi, China, and provide valuable reference information for JE control and prevention. Methods Theoretically epidemiologic study was employed in the research process. Monthly incidence data on JE for the period from Jan 2005 to Sep 2014 were obtained from a passive surveillance system at the Center for Diseases Prevention and Control in Xianyang, Shaanxi province. An optimal SARIMA model was developed for JE incidence from 2005 to 2013 with the Box and Jenkins approach. This SARIMA model could predict JE incidence for the year 2014 and 2015. Results SARIMA (1, 1, 1) (2, 1, 1)12 was considered to be the best model with the lowest Bayesian information criterion, Akaike information criterion, Mean Absolute Error values, the highest R2, and a lower Mean Absolute Percent Error. SARIMA (1, 1, 1) (2, 1, 1)12 was stationary and accurate for predicting JE incidence in Xianyang. The predicted incidence, around 0.3/100 000 from June to August in 2014 with low errors, was higher compared with the actual incidence. Therefore, SARIMA (1, 1, 1) (2, 1, 1)12 appeared to be reliable and accurate and could be applied to incidence prediction. Conclusions The proposed prediction model could provide clues to early identification of the JE incidence that is increased abnormally (≥0.4/100 000). According to the predicted Results in 2014, the JE incidence in Xianyang will decline slightly and reach its peak from June to August.

Analysis of significant factors for dengue fever incidence prediction.

BackgroundMany popular dengue forecasting techniques have been used by several researchers to extrapolate dengue incidence rates, including the K-H model, support vector machines (SVM), and artificial neural networks (ANN). The time series analysis methodology, particularly ARIMA and SARIMA, has been increasingly applied to the field of epidemiological research for dengue fever, dengue hemorrhagic fever, and other infectious diseases. The main drawback of these methods is that they do not consider other variables that are associated with the dependent variable. Additionally, new factors correlated to the disease are needed to enhance the prediction accuracy of the model when it is applied to areas of similar climates, where weather factors such as temperature, total rainfall, and humidity are not substantially different. Such drawbacks may consequently lower the predictive power for the outbreak.ResultsThe predictive power of the forecasting model-assessed by Akaike’s information criterion (AIC), Bayesian information criterion (BIC), and the mean absolute percentage error (MAPE)-is improved by including the new parameters for dengue outbreak prediction. This study’s selected model outperforms all three other competing models with the lowest AIC, the lowest BIC, and a small MAPE value. The exclusive use of climate factors from similar locations decreases a model’s prediction power. The multivariate Poisson regression, however, effectively forecasts even when climate variables are slightly different. Female mosquitoes and seasons were strongly correlated with dengue cases. Therefore, the dengue incidence trends provided by this model will assist the optimization of dengue prevention.ConclusionsThe present work demonstrates the important roles of female mosquito infection rates from the previous season and climate factors (represented as seasons) in dengue outbreaks. Incorporating these two factors in the model significantly improves the predictive power of dengue hemorrhagic fever forecasting models, as confirmed by AIC, BIC, and MAPE.

Abstract B12: To screen or not to screen: Examining components of the Extended Parallel Process Model in a tailored-risk communication intervention to promote colorectal cancer screening

Abstract The Extended Parallel Process Model (EPPM) has been effectively used to promote healthy behaviors, but less is known about the effectiveness of the EPPM in promoting screening adherence in individuals at familial risk for colorectal cancer (CRC). The original EPPM model posits that an additive relationship exists between the constructs perceived severity and susceptibility (“threat”) and between perceived self- and response efficacy (“efficacy”). However, this model may not be as effective in predicting message acceptance in the context of CRC screening as these constructs may independently contribute to the model. For example, with CRC threat, there may be little variance in the perceived severity score as perceptions of cancer severity have been shown to be universally high. Likewise, one may have high levels of response efficacy (“Colonoscopy can save my life”) but low self-efficacy (“Without insurance I can't pay for a colonoscopy”) thus influencing efficacy scores. To determine the effectiveness of the EPPM in promoting CRC screening, we used data from the Family Colorectal Cancer Awareness and Risk Education Project (Family CARE), that tested a remote tailored-risk communication intervention aimed at individuals with familial risk for CRC. Intention-to-screen and medical record-verified colonoscopy (CS) were collected on 218 participants who received the personalized intervention. Using structural equation modeling (SEM), we examined associations of CRC threat and efficacy with greater message acceptance in a full additive model, and in an apriori alternative model in which perceived severity, susceptibility, self-efficacy and response efficacy acted as independent predictors. We considered indicators of good fit to be Root Mean Square Error of Approximation (RMSEA) &amp;lt;.08, Standardized Root Mean Residual (SRMR) &amp;lt;.08, and Comparative Fit Index (CFI) &amp;gt;.90. We compared the two models based on model fit indices, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), with the preferred model having lower AIC and BIC. SEM results indicated poor fit for the full additive EPPM model (RMSEA=0.109; SRMR=0.137; CFI 0.800) while the alternative model showed good fit (RMSEA=0.068; SRMR=0.063; CFI=0.933). Additionally, comparison of the two models indicated that the alternative model was a better fit than the full additive model (RMSEA=0.071; SRMR=0.108; CFI 0.918; AIC=11021; BIC=11332). Further, we found perceptions of susceptibility (B =0.321, p &amp;lt;.001) and self-efficacy (B=0.365, p&amp;lt;0.001) to be significantly associated with intention to screen, while severity (B=0.059, p=0.454) and response efficacy (B=-0.189, p=0.101) showed no significant association. Importantly, intention to screen was significantly associated with medically verified CS behavior (B=0.728, p&amp;lt;0.001). Relatives of CRC patients are at increased risk for the disease, yet screening rates remain low among these individuals despite the effectiveness of CS in both early detection and cancer prevention. Theory-based communication such as the EPPM that addresses multiple determinates of behavior may help promote CRC screening yet few studies have tested conceptual models with regard to an intervention's effect on psychosocial pathways and actual cancer screening behavior. Our results indicate that focusing on individuals' perceptions of CRC susceptibility and screening self-efficacy may more effectively contribute to CRC screening behavior. Noteworthy, we measured both intentions to screen and actual screening behavior. These findings uniquely contribute to the literature by providing evidence of the underlying theoretical psychosocial pathways of how a personalized health promotion intervention works in motivating CRC screening among individuals at increased familial risk. Citation Format: Wendy C. Birmingham, Man Hung, Watcharaporn Boonyasiriwat, Scott T. Walters, Antoinette M. Stroup, Marc D. Schwartz, Jeremy D. Franklin, Anita Y. Kinney. To screen or not to screen: Examining components of the Extended Parallel Process Model in a tailored-risk communication intervention to promote colorectal cancer screening. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr B12.

Dimensionality of the 9-item Utrecht Work Engagement Scale revisited: A Bayesian structural equation modeling approach.

The aim of this study was to reexamine the dimensionality of the widely used 9-item Utrecht Work Engagement Scale using the maximum likelihood (ML) approach and Bayesian structural equation modeling (BSEM) approach. Three measurement models (1-factor, 3-factor, and bi-factor models) were evaluated in two split samples of 1,112 health-care workers using confirmatory factor analysis and BSEM, which specified small-variance informative priors for cross-loadings and residual covariances. Model fit and comparisons were evaluated by posterior predictive p-value (PPP), deviance information criterion, and Bayesian information criterion (BIC). None of the three ML-based models showed an adequate fit to the data. The use of informative priors for cross-loadings did not improve the PPP for the models. The 1-factor BSEM model with approximately zero residual covariances displayed a good fit (PPP>0.10) to both samples and a substantially lower BIC than its 3-factor and bi-factor counterparts. The BSEM results demonstrate empirical support for the 1-factor model as a parsimonious and reasonable representation of work engagement.

Aggregate Australian Takeovers: A Review of Markov Regime Switching Models

This article reviews the case of modeling merger waves in the Australian market for the period 1972–2004. Three Markov switching models are examined, the Gaussian AR(1), Poisson AR(1), and State-Space autoregressive moving average (ARMA) (1,1), to find which gives the best fit. The State-Space Markov switching ARMA(1,1) model is found to be the best for describing Australian takeover activity as estimation results based on it have a lower Bayesian information criterion score than the other two models. Each model's ability to predict a ‘wave’ is then tested by including its estimated probability in a macroeconomic model to explain merger activity. The State-Space model also performs better because the inclusion of its estimated probability substantially increases the explanatory power of the regression model (measured by the regression adjusted R2). In addition, it predicted a takeover wave in 2009, which was closer to the actual incidents of takeover activity in the market at that time than the predictions of the other two models. The results are robust when the measure of takeover activity is changed from the number of takeover bids to the proportion of takeover bids relatively to the number of exchange-listed companies. JEL classification: G34, C32.

A trajectory-based approach to understand the factors associated with persistent depressive symptoms in primary care

BackgroundDepression screening in primary care yields high numbers. Knowledge of how depressive symptoms change over time is limited, making decisions about type, intensity, frequency and length of treatment and follow-up difficult. This study is aimed to identify depressive symptom trajectories and associated socio–demographic, co-morbidity, health service use and treatment factors to inform clinical care. Methods789 people scoring 16 or more on the CES-D recruited from 30 randomly selected Australian family practices. Depressive symptoms are measured using PHQ-9 at 3, 6, 9 and 12 months. ResultsGrowth mixture modelling identified a five-class trajectory model as the best fitting (lowest Bayesian Information Criterion): three groups were static (mild (n=532), moderate (n=138) and severe (n=69)) and two were dynamic (decreasing severity (n=32) and increasing severity (n=18)). The mild symptom trajectory was the most common (n=532). The severe symptom trajectory group (n=69) differed significantly from the mild symptom trajectory group on most variables. The severe and moderate groups were characterised by high levels of disadvantage, abuse, morbidity and disability. Decreasing and increasing severity trajectory classes were similar on most variables. LimitationsAdult only cohort, self-report measures. ConclusionsMost symptom trajectories remained static, suggesting that depression, as it presents in primary care, is not always an episodic disorder. The findings indicate future directions for building prognostic models to distinguish those who are likely to have a mild course from those who are likely to follow more severe trajectories. Determining appropriate clinical responses based upon a likely depression course requires further research.

Skew-t Fits to Mortality Data—Can a Gaussian-Related Distribution Replace the Gompertz–Makeham as the Basis for Mortality Studies?

Gompertz-related distributions have dominated mortality studies for 187 years. However, nonrelated distributions also fit well to mortality data. These compete with the Gompertz and Gompertz-Makeham data when applied to data with varying extents of truncation, with no consensus as to preference. In contrast, Gaussian-related distributions are rarely applied, despite the fact that Lexis in 1879 suggested that the normal distribution itself fits well to the right of the mode. Study aims were therefore to compare skew-t fits to Human Mortality Database data, with Gompertz-nested distributions, by implementing maximum likelihood estimation functions (mle2, R package bbmle; coding given). Results showed skew-t fits obtained lower Bayesian information criterion values than Gompertz-nested distributions, applied to low-mortality country data, including 1711 and 1810 cohorts. As Gaussian-related distributions have now been found to have almost universal application to error theory, one conclusion could be that a Gaussian-related distribution might replace Gompertz-related distributions as the basis for mortality studies.

Estimation of genetic parameters for body weights of Nile tilapia Oreochromis niloticus using random regression models

The accuracy of predicted breeding values used for selection in animal breeding programs is dependent upon the amount and quality of information, the adequacy of statistical models and the analytical methodology employed. Random regression models more adequately describe growth than repeatability and multitrait models; because they account for continuous changes in phenotype and for related genetic and environment effects while using all available information without pre-adjustment of the data. In this experiment, longitudinal genetic studies of body weight of Nile tilapia Oreochromis niloticus recorded between 106 and 245days of age were conducted by fitting random regression models. Four models, inclusion or exclusion of a family effect and the assumption of homogeneity vs. heterogeneity of residual variance, were compared by Bayesian Information Criteria (BIC). The lowest BIC (best fit) was observed for the model considering heterogeneous residual variance and including a family effect. Heritability estimates for live body weight ranged from .02 to .52 during the age interval examined, and all estimates beyond 190days of age were greater than .30. Part of the variance among families was attributable to direct additive genetic, common environmental, maternal and other heritable genetic effects. Thus, the inclusion of a family effect in the model would result in lower estimates of additive genetic variance, particularly at the beginning of the growth period, close to when fish were individually identified. Until that point, fish within family groups were kept in a common environment, which may explain the low heritability estimates obtained at the beginning of the experiment. Genetic correlations between body weights at different ages all were greater than .60, suggesting correlated response in body weight at older ages (245days) could be achieved by selecting fish at earlier ages (106days).

The effects of congestion charging on road traffic casualties: A causal analysis using difference-in-difference estimation

This paper aims to identify the impacts of the London congestion charge on road casualties within the central London charging zone. It develops a full difference-in-difference (DID) model that is integrated with generalized linear models, such as Poisson and Negative Binomial regression models. Covariates are included in the model to adjust for factors that violate the parallel trend assumption, which is critical in the DID model. The lower Bayesian Information Criterion value suggests that the full difference-in-difference model performs well in evaluating the relationship between road accidents and the London congestion charge as well as other socio-economic factors. After adjusting for a time trend and regional effects, the results show that the introduction of the London congestion charge has a significant influence on the incidence of road casualties. The congestion charge reduces the total number of car accidents, but is associated with an increase in two wheeled vehicle accidents.

Comparison of the performance of the CMS Hierarchical Condition Category (CMS-HCC) risk adjuster with the charlson and elixhauser comorbidity measures in predicting mortality

BackgroundThe Centers for Medicare and Medicaid Services (CMS) has implemented the CMS-Hierarchical Condition Category (CMS-HCC) model to risk adjust Medicare capitation payments. This study intends to assess the performance of the CMS-HCC risk adjustment method and to compare it to the Charlson and Elixhauser comorbidity measures in predicting in-hospital and six-month mortality in Medicare beneficiaries.MethodsThe study used the 2005-2006 Chronic Condition Data Warehouse (CCW) 5% Medicare files. The primary study sample included all community-dwelling fee-for-service Medicare beneficiaries with a hospital admission between January 1st, 2006 and June 30th, 2006. Additionally, four disease-specific samples consisting of subgroups of patients with principal diagnoses of congestive heart failure (CHF), stroke, diabetes mellitus (DM), and acute myocardial infarction (AMI) were also selected. Four analytic files were generated for each sample by extracting inpatient and/or outpatient claims for each patient. Logistic regressions were used to compare the methods. Model performance was assessed using the c-statistic, the Akaike's information criterion (AIC), the Bayesian information criterion (BIC) and their 95% confidence intervals estimated using bootstrapping.ResultsThe CMS-HCC had statistically significant higher c-statistic and lower AIC and BIC values than the Charlson and Elixhauser methods in predicting in-hospital and six-month mortality across all samples in analytic files that included claims from the index hospitalization. Exclusion of claims for the index hospitalization generally led to drops in model performance across all methods with the highest drops for the CMS-HCC method. However, the CMS-HCC still performed as well or better than the other two methods.ConclusionsThe CMS-HCC method demonstrated better performance relative to the Charlson and Elixhauser methods in predicting in-hospital and six-month mortality. The CMS-HCC model is preferred over the Charlson and Elixhauser methods if information about the patient's diagnoses prior to the index hospitalization is available and used to code the risk adjusters. However, caution should be exercised in studies evaluating inpatient processes of care and where data on pre-index admission diagnoses are unavailable.

Approximating Model Probabilities in Bayesian Information Criterion and Decision-Theoretic Approaches to Model Selection in Phylogenetics

A priori selection of models for use in phylogeny estimation from molecular sequence data is increasingly important as the number and complexity of available models increases. The Bayesian information criterion (BIC) and the derivative decision-theoretic (DT) approaches rely on a conservative approximation to estimate the posterior probability of a given model. Here, we extended the DT method by using reversible jump Markov chain Monte Carlo approaches to directly estimate model probabilities for an extended candidate pool of all 406 special cases of the general time reversible + Γ family. We analyzed 250 diverse data sets in order to evaluate the effectiveness of the BIC approximation for model selection under the BIC and DT approaches. Model choice under DT differed between the BIC approximation and direct estimation methods for 45% of the data sets (113/250), and differing model choice resulted in significantly different sets of trees in the posterior distributions for 26% of the data sets (64/250). The model with the lowest BIC score differed from the model with the highest posterior probability in 30% of the data sets (76/250). When the data indicate a clear model preference, the BIC approximation works well enough to result in the same model selection as with directly estimated model probabilities, but a substantial proportion of biological data sets lack this characteristic, which leads to selection of underparametrized models.

Survival of Striped Bass Entrained into the Cooling Systems of Two Hudson River Power Stations

AbstractData from entrainment survival sampling programs conducted in 1979, 1980, and 1988 at the Indian Point generating stations and in 1977, 1978, and 1980 at the Roseton generating station, both located on the tidal Hudson River estuary, were reanalyzed using a series of failure time models to estimate the survival of entrained striped bass Morone saxatilis based on sampling stress, mechanical entrainment stress, and thermal stress. The model with the lowest Akaike information criterion value was the most complex model examined, with year‐ and station‐specific parameters for both sampling and entrainment. However, the lowest Bayesian information criterion was achieved with a model that had year‐ and station‐specific sampling parameters but only station‐specific entrainment parameters. The models indicated that survival was positively related to fish length and salinity and negatively related to discharge temperatures above 30°C. Parameter uncertainty was addressed via a bootstrap resampling procedure to estimate parameters for the station‐specific model, and the results were applied to entrainment abundance data from each facility from 1983 to 1987. The distribution of the estimated mean annual entrainment survival after 1 d for Indian Point had modes at 0.48, 0.60, 0.70, and 0.81. For Roseton, the distribution of mean annual entrainment survival was normal, with a mean of 0.41 and a variance of 0.0028. Survival differences between the two stations were due to inherent differences in entrainment stresses, as indicated by differences in the parameters for the model as well as differences in the length of entrained striped bass, discharge temperature, and salinity.

Genetic parameters for post weaning growth of Nellore cattle using polinomyals and trigonometric functions in random regression models

Covariance functions and random regression models have been considered as an alternative for data adjustment, in sequence, stemming from the same animal along time and which presents a structured pattern of covariance. Aiming to evaluate the performance of random regression models based on the Legendre, modified Jacobi and trigonometric functions, data concerning the weights of Nellore breed animals were used from birth to the 800th day of life, in models that assumed direct additive and animal permanent environmental effects coefficients. The Schwarz Bayesian information criterion (BIC) led to the selection of the models Legendre of order six (ML6), Jacobi of order five (MJ5) and trigonometric of order six (MT6), the ML6 model presenting the lowest BIC. At the extremity of the interval, the MJ5 model presented lower variance of component estimates than those obtained through the ML6 model, however the estimates were in accordance to the medium part of the interval; while the estimates from the MT6 model were oscillating and different from those obtained through the other models. At the extremity of the interval, the heritability coefficient estimates (&lt;img src="/img/revistas/sa/v66n4/h4_circ.gif" align="absmiddle"&gt;2) obtained through the MJ5 model were lower than those obtained through the ML6 model, however, in the medium part of the interval, they were in accordance, remaining between 0.2 and 0.3. The values obtained through the MT6 model were different from those obtained through the other models, remaining between 0.35 and 0.40 on the first 285th days and then dropping to 0.01 on the 800th days of life. The means of the estimated growth curves started to distance from the data mean tendency from the 470th days on, and in this interval, the MT6 model was the most suitable.