Applicability of multiple covariance analysis in caries studies

Area of Science:

• Dental Research
• Biostatistics
• Clinical Trials

Background:

• Dental caries clinical trials often measure changes in decayed, missing, or filled surfaces (DMFS).
• Standard analysis assumes a linear relationship between baseline and post-treatment DMFS scores.
• This assumption may not fully capture the data, potentially overlooking important information.

Purpose of the Study:

• To investigate the relationship between pre- and post-treatment DMFS measures in dental trials.
• To determine if the relationship is linear or quadratic.
• To highlight the limitations of traditional DMFS increment measurements.

Main Methods:

• Statistical analysis of DMFS data from dental clinical trials.
• Exploration of linear and quadratic regression models.
• Consideration of polynomial regression for nonlinear relationships.

Main Results:

• The relationship between pre- and post-treatment DMFS scores may be quadratic, not strictly linear.
• Even in linear relationships, the slope might deviate from one.
• Traditional DMFS increment calculations do not leverage baseline DMFS information effectively.

Conclusions:

• Nonlinear relationships necessitate advanced statistical approaches.
• Analysis of covariance with polynomial regression parameters is indicated for nonlinear DMFS data.
• Accurate interpretation of dental trial outcomes requires acknowledging potential nonlinearities.