Search

Clinical dental research generates large amounts of data with a potentially complex correlation structure from measurements recorded at several sites throughout the mouth. Clinical attachment level (CAL) is one such measure popularly used to assess the periodontal disease (PD) status. We model the proportion of sites for each tooth-type (i.e., incisor, canine, pre-molar and molar) per subject that exhibit moderate to severe PD. Disease free and highly diseased tooth-sites cause these proportion responses to lie in the closed interval [0, 1]. In addition, PD may be spatially referenced, i.e., the disease status of a site is influenced by its neighbours. While beta regression can assess the covariate-response relationship for proportion data, its support in the interval (0, 1) impairs its ability to account for the observed proportions at zero and one. In contrast to ad hoc transformations that confine responses to (0, 1), we develop a framework that augments the beta density with non-zero masses at zero and one while also controlling for spatial referencing. Our approach is Bayesian and is computationally amenable to available software. A simulation study evaluates estimation of regression effects in scenarios of varying sample size, degree of spatial dependence and response transformations. Application to real PD data provide insights into assessing covariate effects on proportion responses.

Background Various dose-finding clinical trial designs, including the continual reassessment method (CRM), dichotomize toxicity outcomes based on prespecified dose-limiting toxicity (DLT) criteria. This loss of toxicity information is particularly inefficient due to the small sample sizes in phase I trials, especially when Common Terminology Criteria for Adverse Events (CTCAE v4.0) are an established ordinal toxicity grading classification already used in the clinical practice. Purpose The purpose of this simulation study is to incorporate ordinal toxicity grades as specified by CTCAE v4.0 using a continuation ratio (CR) model in the likelihood-based CRM. Methods This simulation study compares the CR model design to the dichotomous CRM as well as an ordinal CRM that implements the proportional odds (PO) model. We compare six scenarios for model performance based on various safety and efficiency criteria and consider a range of dose–toxicity relationship models, including CR models, PO models, and models that violate the PO assumption. Results The ordinal CRM performs as well as the dichotomous CRM in all scenarios considered, especially in situations where the starting dose is overly toxic, the ordinal designs show slight improvement in the estimation of the maximum tolerated dose (MTD) and fewer median patients exposed to excessively toxic dose levels as compared to the binary CRM. We also find slight discrepancies in the performance between the PO model and CR model; however, the differences were not substantial enough to strongly recommend one model over the other. Limitations The CR model design does require slightly more input from clinical investigators prior to the start of the trial as compared to the dichotomous CRM. Investigators must specify the distribution of toxicity grades at the expected dose levels for a 10% and 90% DLT rate in this CR design. However, an R package will help with the implementation of this ordinal design. Conclusions While the ordinal designs did not perform significantly better than the binary counterpart, we were able to incorporate maximal toxicity information available into a feasible dose-finding design without compromising overall design performance.

One of the most important indicators of dental caries prevalence is the total count of decayed, missing or filled (DMF) surfaces in a tooth. These count data are often clustered in nature (several count responses clustered within a subject), over-dispersed, as well as spatially referenced (a diseased tooth might be positively influencing the decay process of a set of neighboring teeth). In this paper, we develop a multivariate spatial Beta-Binomial (BB) model for these data that accommodates both over-dispersion as well as latent spatial associations. Using a Bayesian paradigm, the re-parameterized marginal mean (as well as variance) under the BB framework are modeled using a regression on subject/tooth-specific co-variables and a conditionally autoregressive (CAR) prior that models the latent spatial process. The necessity of exploiting spatial associations to model count data arising in dental caries research is demonstrated using a small simulation study. Real data confirms that our spatial BB model provides a superior estimation and model fit as compared to other sub-models that do not consider modeling spatial associations.

This study examined whether South Carolina’s sex offender registration and notification (SORN) policy was associated with changes in judicial decision making concerning adult sex crime cases. Outcomes pertained to reduced charges (e.g., sex to nonsex crimes) and final case dispositions. Statewide crime data from 1990 to 2004 corresponded with three time periods of interest: the 5 years immediately preceding enactment of SORN (Year Group 1; 1990—1994), the first 4 years of SORN implementation (Year Group 2; 1995—1998), and the subsequent 6 years of SORN implementation, which included Internet notification (Year Group 3; 1999—2004). Univariate and generalized estimating equations methods were used to model patterns of charging and disposition decisions with respect to year groups. Results indicated that defendants were more likely to have charges reduced from sex to nonsex crimes over time, with a 9% predicted probability of reduced charges in Year Group 1, a 15% predicted probability in Year Group 2 (corresponding with initial implementation of SORN), and a 19% predicted probability in Year Group 3 (corresponding with Internet notification). Results also indicated that the probability of a guilty disposition changed at each year group, with a predicted probability in Year Group 1 of 55%, increasing to 65% in Year Group 2, and then declining to 60% in Year Group 3. This final decline was more pronounced when pleaded cases were removed from analyses. South Carolina’s SORN policy shares several characteristics with the federal Adam Walsh Act (AWA). Thus, both state and national policy implications of these findings are discussed.

This study examined whether South Carolina’s sex offender registration and notification (SORN) policy was associated with a general deterrent effect on adult sex crimes. Using adult arrest data from 1990 through 2005, trend analyses modeled the intervention effects of 1995 (the year South Carolina’s SORN policy was initially implemented) and 1999 (the year the policy was revised to include online registration). Results supported a significant deterrent effect for the 1995 intervention year, with an approximately 11% reduction in first-time sex crime arrests in the post-SORN period (1995-2005) relative to the pre-SORN period (1990-1994). This decline equated to averting approximately three new sex crime arrests per month. Comparison analyses with serious nonsex offenses against persons (assault and robbery) failed to identify similar effects, suggesting that the 1995 effect is attributable to sex offense—specific legislation. Findings are compared with the existing literature on general deterrent effects of SORN and discussed in the context of research examining other effects of SORN.

This study examined whether South Carolina’s sex offender registration and notification (SORN) policy was associated with a general deterrent effect on juvenile sex crimes. Using juvenile justice data from 1991 through 2004, trend analyses modeled the intervention effects of 1995 (the year South Carolina’s SORN policy was initially implemented) and 1999 (the year the policy was revised to include online registration). Initial results suggested a significant deterrent effect of SORN on first-time juvenile sex crimes. However, comparison analyses with nonsex offenses identified a similar effect on first-time robbery crimes. Follow-up analyses indicated that the apparent declines identified for first-time sex and robbery offenses were due to another legislative change, also enacted in 1995, that moved the prosecution of 16-year-old defendants from juvenile to adult court. When these cases were included in the database, follow-up analyses indicated no significant effect for the 1995 intervention year. Thus, South Carolina’s SORN policy was not associated with a general deterrent effect on juvenile sex crimes. Specific policy changes are suggested regarding the application of registration and notification requirements to juveniles.

Some sex offender registration and notification (SORN) policies subject all registered sex offenders to Internet notification. The present study examined the effects of one such broad notification policy on sex crime recidivism. Secondary data were analyzed for a sample of 6,064 male offenders convicted of at least one sex crime between 1990 and 2004. Across a mean follow-up of 8.4 years, 490 (8%) offenders had new sex crime charges and 299 (5%) offenders had new sex crime convictions. Cox’s relative risks and competing risks models estimated the influence of registration status on risk of sexual recidivism while controlling for time at risk. Registration status did not predict recidivism in any model. These results cast doubt on the effectiveness of broad SORN policies in preventing repeat sexual assault. Policy implications, particularly with respect to the federal Adam Walsh Child Protection and Safety Act, which requires broad notification, are discussed.

This study examines the influence of South Carolina's sex offender registration policy on juvenile offender recidivism risk. Juvenile male sexual offender ( N = 1,275) risk of recidivism was examined with an average 9-year follow-up. Survival analysis was used to examine the influence of covariates, including the primary time-dependent covariate registration status at time, on risk of new sexual offense charges and adjudications. A competing risks model was used to explain the effects of covariates on different types of recidivism events (sexual, assault, and other). Registration status at time had a significant effect on risk of new “other” offense charges and a marginal ( p < .10) effect on risk of sexual offense charges (survival analysis) but no statistical evidence of affecting risk of new adjudications. Results suggest a surveillance effect leading to unnecessary charges for registered (vs. nonregistered) youth. State and national policy implications are discussed.