Background: There is disagreement as to whether patient stratification by a combination of diabetes, smoking, and genetic test results is useful for informing the frequency of dental prophylaxes.
Methods: The authors appeal to basic tenets of clinical study design and statistical analysis of clinical investigations, and highlight how secondary ad hoc analyses, such as those of Diehl and colleagues, are frequently underpowered and inconclusive. They also provide evidence from numerous studies supporting the use of genetics to identify risk.
Results: The authors believe the conclusions reached from their original analyses are valid and the analyses of Diehl and colleagues serve to simply reinforce the authors' specific intent of avoiding such underpowered analyses altogether with the Michigan Personalized Prevention Study.
Conclusions: Until full genome sequencing in many people with highly specified disease phenotypes is feasible, experimental approaches based on biological findings and hypothesis testing should not be summarily discounted.
Practical implications: Stratification of patients to provide "personalized" treatment remains an important, yet elusive, goal. The current debate serves to highlight the need for large, clinical utility studies that can adequately determine how phenotypic and genotypic data can be best used to improve oral health in the US population.
Keywords: Comparative effectiveness research; health care delivery; interleukin polymorphisms; oral health; periodontal disease; personalized medicine.
Copyright © 2015 American Dental Association. Published by Elsevier Inc. All rights reserved.