# SMARTp: A SMART design for non-surgical treatments of chronic periodontitis with spatially-referenced and non-randomly missing skewed outcomes

## TL;DR

This paper introduces a SMART design tailored for non-surgical treatments of chronic periodontitis, addressing challenges with spatially-referenced, non-randomly missing, and skewed outcomes to optimize personalized treatment strategies.

## Contribution

It develops a novel SMART framework specifically designed for complex periodontal data with missing and skewed outcomes, including an accessible R implementation.

## Key findings

- Proposes a new sample size formula for SMART in periodontal studies
- Addresses spatially-referenced and non-random missing data issues
- Provides R code for practical implementation

## Abstract

This paper proposes dynamic treatment regimes for choosing individualized effective treatment strategies of chronic periodontal disease. R codes for implementing the proposed sample size formula are available in GitHub.