A Re-analysis of a Caries Clinical Trial by Survival Analysis

Abstract

The decline in caries prevalence, the increases in the level of fluoride exposure, and the lack of placebo control subjects have complicated caries clinical trials in recent times. There has been a substantial increase in the numbers of subjects required for the detection of statistically significant differences between dental products, and hence, the cost of these trials has grown enormously. This study uses a new statistical approach to the analysis of the data from these trials with the ultimate aim of providing a more sensitive method of analysis. The new approach uses survival analysis, where the outcome measure is the survival time of an individual tooth surface. It exploits recent developments in the analysis of clustered survival data where survival times within the same cluster or subject are correlated. To illustrate, the new method of analysis was used for the North Wales, UK, caries clinical trial. It is concluded that survival analysis uses most of the data available in a caries clinical trial, an outcome measure that is easily understood and may lead to a more sensitive method of analysis.

The Anticariogenic Effect of Sugar-Free Gum Containing CPP-ACP Nanocomplexes on Approximal Caries Determined Using Digital Bitewing Radiography

“…Traditional methods for analysing the data obtained from these studies have involved calculating a summary measure, such as the DMF index or DMF increment, for each subject and then comparing the summary measure between study groups. The use of subject-level summary measures for comparison of efficacy or effectiveness may mask important information, such as the differences in caries susceptibility or intervention effectiveness between the various tooth surfaces [Hannigan et al, 2001;Burnside et al, 2007]. Analysing these data at the tooth surface level allows both subject-level and surfacelevel risk factors to be incorporated into the analysis.…”

Chlorhexidine and Preservation of Sound Tooth Structure in Older Adults

“…The primary test statistics for evaluating treatment effects was based on the proportional hazards model [Robins et al, 1995;Schafer, 1997]. Using a time-to-event analysis increases the efficiency and sensitivity of the statistical analysis over analysis of variance methods [DeRouen et al, 1995;Beck et al, 1997;Spencer, 1997;Hannigan et al, 2001Hannigan et al, , 2004Mancl et al, 2004]. The SAS 'phreg' and 'genmod' procedures were used to estimate hazard ratios and rates of transition from sound to nonsound surfaces.…”

Classical and Modern Methods in Caries Epidemiology

“…al., 2002), because many cariesprone teeth were left-censored. Since then, survival analysis methods have been applied to reanalyzing the longitudinal caries data in many other clinical trials, too (Hannigan et al, 2001, Hujoel et al, 2003, Baelum et al, 2003. All these large-scale time dependent analyses of caries data have employed fully parametric tooth-specific methods in permanent dentitions.…”

“…Fully parametric regression models are widely used for the sound-carious transitions inside the oral cavity, where tooth surfaces are clustered within children, and such methods are preferred in many caries data analyses (Hannigan et al, 2001, Baelum et al, 2003, Leroy et al, 2005a,b, Stephenson et al, 2010. In particular for the interval-censored data, parametric modeling is preferred to non-fully parametric models such as the Cox proportional hazards model (Leroy et al, 2005a), and parametric models may lead to more precise inferences than those arising from semiparametric methods (Baelum et al, 2003).The log-logistic distribution selected for the parametric competing risk models was parameterized in an accelerated failure time setting and was found to describe caries data well in permanent dentition (Hannigan et al, 2001). Leroy et al, (2005a,b) used multivariate survival analysis for the identification of factors, like caries in primary molars, with cavity formation of first permanent molars in a follow-up study with the EBD-fashion.…”

“…Survival modeling has mainly used in the renewed statistical treatment of the findings of previous caries trials, the results showing that the differences existed also when using more sophisticated survival analysis. Hujoel et al, (2003) established the benefits of xylitol with Poisson regression models and Hannigan et al, (2001) who reanalyzed the effect of diet and tooth brushing habits by the log-logistic distribution selected for the parametric competing risk models, showed a good fit to the caries data. She concluded that the marginal model approach may be more suitable than the frailty model for multivariate survival data from caries clinical trials, since most of the source of variability is due to different anatomic susceptibilities of the tooth surfaces (caries may be regarded as a local disease in different teeth) rather than to the specific subject -specific frailty (caries is a general disease in caries prone subjects, like is osteoporosis with fractures).…”