An R-package for the Estimation and Testing of Multiple Covariates and Biomarker Interactions for Survival Data Based on Local Partial Likelihood

When we study the variables that a ffect survival time, we usually estimate their eff ects by the Cox regression model. In biomedical research, e ffects of the covariates are often modi ed by a biomarker variable. This leads to covariates-biomarker interactions. Here biomarker is an objective measurement of the patient characteristics at baseline. Liu et al. (2015) has built up a local partial likelihood bootstrap model to estimate and test this interaction e ffect of covariates and biomarker, but the R code developed by Liu et al. (2015) can only handle one variable and one interaction term and can not t the model with adjustment to nuisance variables. In this project, we expand the model to allow adjustment to nuisance variables, expand the R code to take any chosen interaction terms, and we set up many parameters for users to customize their research. We also build up an R package called "lplb" to integrate the complex computations into a simple interface. We conduct numerical simulation to show that the new method has excellent fi nite sample properties under both the null and alternative hypothesis. We also applied the method to analyze data from a prostate cancer clinical trial with acid phosphatase (AP) biomarker.

MENTAL HEALTH AND CRIMINALITY AMONG PSYCHIATRIC OFFENDERS IN SOUTHEASTERN ONTARIO

OBJECTIVES: (1) Describe the population of mentally ill offenders over whom Ontario Review Board (ORB) held jurisdiction. (2) Assess the influences of psychopathology and criminal factors on criminal career. METHOD: This study was a retrospective case series design that reviewed all offenders who were court ordered for psychiatric evaluation at Mental Health Services Site of Providence Care in Kingston, Ontario from 1993 to 2007 (N=347). Eighty five subjects were found not criminally responsible on the account of mental disorder and were included in statistical analysis (n=85). Bivariate associations between five key variables and two outcome variables, seriousness of crime and recidivism, were examined. Logistic regressions were conducted to test the role of the predictor variables on the outcome variables. RESULTS: Age and change in principal psychiatric diagnosis over time were shown to be associated with seriousness of crime. Timing of psychiatric onset, early signs of deviance and change in diagnosis were shown to be associated with recidivism. On the whole, study population did not markedly vary in their distribution of variables by the outcome variables. Regression model included timing of psychiatric onset; psychiatric history; existence of criminal associate; child abuse history; and early signs of deviance. Recidivism was shown to be predicted by early signs of deviance (OR=8.154, p<0.05). Existence of criminal associates was shown to have substantial values of odds ratio at marginal significance (OR=7.577, p=0.13). CONCLUSION: Seriousness of crime is a complex factor that could not be sufficiently predicted by any one or combinations of study variables. Recidivism is better predicted by criminality factors than psychopathology. In the future, an exploratory analysis that more broadly examines the psychopathology and criminal factors in Canadian forensic population is needed. Findings from this study have important clinical and legal implications.