PC program extending the Potthoff-Roy longitudinal data analysis model to allow missing data: Kleinbaum's method

Area of Science:

• Biostatistics
• Longitudinal Data Analysis
• Growth Curve Modeling

Background:

• The Potthoff-Roy generalized multivariate analysis of variance (MANOVA) model is valuable for longitudinal growth curve data.
• A key limitation of the original model is the requirement for complete data at all time points, restricting its application.
• Missing data in longitudinal studies can arise randomly or by design, posing analytical challenges.

Purpose of the Study:

• To present a user-friendly PC program for analyzing longitudinal growth curve data with incomplete measurement sequences.
• To implement Kleinbaum's extension of the Potthoff-Roy model, which accommodates missing data.
• To facilitate the application of advanced growth curve analysis in the presence of missing data.

Main Methods:

• The study utilizes Kleinbaum's extension of the Potthoff-Roy model to handle incomplete longitudinal data.
• A user-friendly, interactive PC program is developed and described.
• The program is illustrated with examples, demonstrating its practical application.

Main Results:

• The developed PC program effectively implements the extended Potthoff-Roy model for growth curve analysis.
• The software allows for the analysis of longitudinal data where measurements are missing at certain time points.
• The program supports scenarios with missing data arising from random attrition or planned study designs (e.g., mixed longitudinal studies).

Conclusions:

• The introduced PC program overcomes the limitations of the original Potthoff-Roy model by enabling analysis with incomplete data.
• This tool enhances the applicability of growth curve modeling in biostatistics and related fields.
• The program provides a practical solution for researchers dealing with missing data in longitudinal studies.