Factors affecting attrition in a longitudinal smoking prevention study

Area of Science:

• Public Health
• Behavioral Science
• Epidemiology

Background:

• Evaluating treatment effects in longitudinal smoking prevention studies is challenging due to participant attrition.
• Understanding predictors of attrition is crucial for accurate assessment of program efficacy.

Purpose of the Study:

• To identify significant predictors of attrition in smoking prevention studies.
• To suggest strategies for reducing attrition bias and improving future study designs.

Main Methods:

• Survival analysis was employed to examine factors associated with participant dropout.
• Attrition was assessed at various time points throughout the longitudinal study.

Main Results:

• Lower academic achievement, reduced tobacco/health knowledge, and lower social influence/resistance skills were linked to higher attrition.
• Current smokers, marijuana users, and Black participants exhibited increased dropout rates.
• Dropout rates varied significantly across different treatment conditions; gender was not a significant predictor.

Conclusions:

• Participant attrition in smoking prevention studies is non-random and predictable.
• Identifying dropout characteristics allows for the implementation of procedures to mitigate bias.
• Targeted strategies can be developed to reduce attrition rates in future smoking prevention research.