Applying Stochastic Process Models for the Disease Progression, Implication for Prevention and Cost-Effectiveness Analysis for Periodontal Disease

Project: A - Government Institutionb - Ministry of Science and Technology

Description

As the majority of PD cases are silent the manifestation of clinically-detected PD cases are often too late to retain teeth. The most consequences leading to endentulous condition would require enormous costs for overdenture and dental implant. Preventive strategies like the empowered PD care and screening for averting loss of tooth are necessary. However, evaluation of those prevention programs requires the understanding of the natural history of the evolution of PD from PD-free, silent PD, symptomatic PD, and tooth loss. Little is known about the quantification of the disease natural history of PD. It is also of great interest to elucidate factors affecting each transition, particularly between silent states, in order to develop individually-tailored preventive care for PD. Information for supporting such personalized PD care is also scanty. Quantifying the temporal disease natural history also forms the basis for decision analysis and cost-effectiveness analysis of preventive strategies of PD. In spite of several progressive models for PD in the previous studies, the methodology is intractable and the empirical data are not available while taking into account the forward and backward transitions between mild, moderate, and severe states of silent PD cases, which play important role of early detection of silent PD cases before subsequent progression to symptomatic PD and further loss of teeth. The objectives of this 3-year project are list as follows. Year 1 (Aug 2015-July 2016) (1) To analyze data form community for those attending periodontal examination (2) To apply the Coxian phase-type distribution to estimating transition parameters based on the empirical data of repeated measurements of community periodontal index (CPI) for detecting silent PD Year 2 (Aug 2016-July 2017) (3) To elucidate the effect of covariates, after controlling for the baseline transitions between states, on the dynamic change of each state, particularly the forward and backward transitions of silent PD cases by using the proportional odds model. Year 3 (Aug 2017-July 2018) (4) Developing the decision analysis model for preventing periodontal disease (5) To estimate the accumulated cost and effectiveness in terms of life years of PD-related tooth loss averted to evaluate preventive strategies of “empowered PD care” and “mass screening” compared to “no intervention”.
StatusFinished
Effective start/end date8/1/157/31/16

Keywords

  • Multi-state proportional odds model
  • community
  • periodontal disease