Evaluation of a Periodontal Disease Risk Assessment Tool

Objectives: Severe periodontitis is the sixth most prevalent condition affecting about 11% of people worldwide. At present, many at-risk patients are not identified at routine dental exams because they appear to be in good periodontal health, despite underlying risk factors that increase their risk for periodontal disease. Our objective was to study the extent to which the Previser Risk Calculator (PRC) (PreViser Corp., Mount Vernon, WA), a commercial risk assessment tool can help us predict an individual’s periodontal disease risk.
Methods: First, we created a limited data set of 300 patient records data of individuals 18 years or older who underwent comprehensive dental care at the Indiana University School of Dentistry (IUSD) between January 1, 2011 and January 1, 2012. The data (see Table 1) were entered into the PRC to estimate the each patient’s disease state and output recorded in MS Excel. We then estimated the percent agreement between the PRC s’ and IUSD student-providers’ diagnosis.
Results: The agreement between PRC and IUSD student-providers’ diagnosis was 70% (209 patients). Specificity for disease severity (mild versus moderate periodontitis), was much lower, at about 47%. Although the IUSD faculty evaluated the student-providers’ data entries, the use of free text may have introduced subjectivity and vagueness to certain factors such as assessing bone loss. Further, insufficient explanation of weightage assigned to risk factors in estimating risk in PRC makes quantitative assessment of factors that influence the overall patient risk score difficult.
Conclusions: This study provided a preliminary assessment of PRC’s capability to assess an individual’s periodontal disease risk. We plan to test whether the PRC systematically over- or under-estimates disease state using Wilcoxon rank sum tests to detect statistically significant differences between the disease state and risk produced by the PRC and that documented by the student provider.