Factor analysis of the interrelationships between clinical variables in horses with colic.

The study examines the relationship between 17 clinical variables in horses with colic to potentially uncover multicollinearity and further determine these factors’ association with the horses’ overall survival.

Research Context

• The research was a prospective survey conducted on horses with colic that were referred to the Large Animal Hospital at the Royal Veterinary and Agricultural University of Copenhagen, Denmark, from August 1994 to December 1997.
• It studied the interrelationships between 17 clinical variables through factor analysis, a statistical method that determines the structural variability in data, thereby revealing multicollinearity (where variables highly correlate with each other).

Research Sample and Method

• Out of the total 528 horses involved in the research, 16 were excluded owing to miscellaneous conditions, leaving 205 horses that had observations for all 17 variables.
• The researchers used these observations as a representative subset since no significant change occurred in the main diagnostic categories.

Research Findings

• The factor analysis practised in this study confirmed a correlation between variables, although multicollinearity was not prevalent.
• Four factors were extracted, and these accounted for 51% of the total variance. These factors were interpreted by integrating previously reported clinical research.
• The first factor was endotoxaemia, characterized by capillary refill time, mucus-membrane color, degree of pain, heart rate, packed-cell volume, and abdominal sounds.
• The second factor, named cecal tympany, was characterized by cecal decompression, admission month, and gastric reflux.
• The third factor, represented by gender, age and temperature, was classified as age.
• The fourth factor could not be easily categorized, even if breed had a significant influence on its formation.

Association with Outcome

• The extracted factors were used in a logistic-regression analysis to evaluate their relationship with the outcome (survival or death).
• The factors of “endotoxaemia” and “age” turned out to be related to the outcome.