Recanting of Previous Reports of Alcohol Consumption within a Large-Scale Clustered Randomised Control Trial

Area of Science:

• Public Health
• Epidemiology
• Adolescent Health

Background:

• Adolescent alcohol use is a significant public health concern.
• Prevention trials rely on accurate self-reported data, but inconsistencies (recanting) can occur.
• Understanding recanting is vital for interpreting trial outcomes.

Purpose of the Study:

• To assess the prevalence of recanting (inconsistent reporting of lifetime alcohol use) in an adolescent alcohol prevention trial.
• To determine the impact of recanting on primary outcomes: heavy episodic drinking and alcohol-related harms.
• To evaluate the influence of recanting on intervention effectiveness.

Main Methods:

• A large-scale randomized controlled trial involving 12,738 secondary school students.
• Data collected via self-report questionnaires at multiple time points over 33 months.
• Recanting defined as a negative report of lifetime alcohol use contradicting a prior positive report.

Main Results:

• Approximately 9.9% of students recanted earlier alcohol consumption between 12 and 33 months post-baseline.
• Recanting was negatively associated with primary outcomes, but adjusting for it did not change the overall intervention effects.
• Males recanted at a higher rate than females.

Conclusions:

• Recanting is identifiable and manageable in adolescent alcohol prevention trials.
• Adjusting for recanting is recommended as a sensitivity analysis to ensure robust outcome assessment.
• While differential recanting rates exist, they do not fundamentally undermine trial findings when properly addressed.