Optimizing measurement of self-reported adherence with the ACTG Adherence Questionnaire: a cross-protocol analysis

Area of Science:

• Clinical Trials
• HIV/AIDS Research
• Pharmacological Adherence

Background:

• The AIDS Clinical Trials Group (ACTG) Adherence Questionnaire is widely used in HIV research.
• Investigators often rely solely on the first item (4-day recall) for adherence assessment.
• This limited use may not fully capture patient adherence to antiretroviral therapy.

Purpose of the Study:

• To evaluate the validity and reliability of all five items of the ACTG Adherence Questionnaire.
• To compare the common 4-day recall method with advanced statistical approaches for adherence measurement.
• To determine if a comprehensive approach using the full questionnaire enhances adherence assessment.

Main Methods:

• Secondary analysis of data from ACTG 370 and ACTG 398 trials.
• Principal component (PC) analysis was employed to derive adherence measures from the full questionnaire.
• Logistic regression was used to correlate adherence measures with plasma HIV RNA levels.

Main Results:

• The first principal component (PC1) significantly correlated with plasma HIV RNA outcomes.
• PC1 demonstrated a stronger correlation with viral load than the average 4-day recall.
• An adherence index derived from PC1 showed greater score variability and compared favorably with medication event monitoring system data.

Conclusions:

• Utilizing the full ACTG Adherence Questionnaire with principal component analysis offers a superior method for assessing antiretroviral adherence.
• This comprehensive approach provides a more nuanced and reliable measure of patient adherence.
• The findings support a revised methodology for adherence data collection and analysis in HIV clinical trials.