Premium
Evaluating the merits of option‐3 methods in relative attachment level measurement in a healthy population
Author(s) -
Johnson P.,
Yang M. C. K.,
Magnusson I.,
Kang T.,
Ramirez J.,
Hefti A. F.
Publication year - 1997
Publication title -
journal of periodontal research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.31
H-Index - 83
eISSN - 1600-0765
pISSN - 0022-3484
DOI - 10.1111/j.1600-0765.1997.tb00532.x
Subject(s) - outlier , statistics , consistency (knowledge bases) , variance (accounting) , clinical attachment loss , observational error , scheme (mathematics) , medicine , mathematics , psychology , econometrics , dentistry , periodontal disease , computer science , artificial intelligence , mathematical analysis , accounting , business
Attachment loss is the principal clinical indicator for periodontal disease progression. However, the attachment measurement can be inaccurate due to outliers. A remedial scheme, the option‐3 scheme, has been suggested in the literature. This scheme involves first measuring the relative attachment twice at any given site. A third measurement is required if there is a large disagreement between the first two probings. When the third measurement is taken, there are several possible ways to make the final assessment of the true attachment level. Among them are the average of the 3 measurements, the average of the closest 2, and the median of the 3. This report evaluates the merits of the option‐3 scheme by comparing it with other schemes to reduce measurement variance and outliers. Since true attachment levels are unknown, we chose to check accuracy through measurement consistency between visits in healthy subjects using the fact that such subjects should have little real change in attachment levels. 12 healthy subjects participated in the study. They visited the clinic 4 times in 4 months. Two types of the Florida stent probe were used by 2 calibrated examiners. It was found that option‐3 schemes were more accurate than a single probing or duplicate probings in estimating the relative attachment level. The 3 final assessment schemes performed almost equally well from a practical standpoint, but the median was the best among them in a statistical sense.