Premium
Restricted latin square design in a plaque disclosant study
Author(s) -
Varma A. A. O.,
Fertig J. W.,
Chilton N. W.,
Mandel I. D.
Publication year - 1974
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.1600-0765.1974.tb00695.x
Subject(s) - latin square , correlation , medicine , analysis of variance , interaction , geometric mean , saline , dispersion (optics) , mathematics , gastroenterology , statistics , chemistry , geometry , biochemistry , rumen , fermentation , physics , optics
Twenty subjects were used in a clinical trial of the effect of 3 disclosants (erythrosine, fast green, sodium fluorescein and normal saline) on plaque score, plaque weight and EDTA dispersion. Five 4×4 Latin Squares of a restricted type were employed, so that carryover effects can be estimated most efficiently. Each of the 20 subjects used each of the 4 test agents for a week at a time. The details of the design and of the statistical analysis are presented. There were no significant carryover effects for any of the study solutions on plaque score, plaque weight or EDTA dispersion. The 4 disclosants did not differ significantly from each other in their effect on the 3 variables. In the case of plaque score, the fourth week gave significantly higher average values than the first three weeks. In the case of plaque weight, the values in the first week were significantly elevated. The correlation coefficients between the 3 indices of plaque accumulation were only modest in size. The relatively low correlation between the 3 variables suggest that in a clinical trial the results with the 3 variables will not necessarily confirm each other.