CASE
STUDY ABSTRACT, OBJECTIVES & DESIGN
CASE
The Center for Adherence Support Evaluation
at the New York Academy of Medicine
The challenges of assessing adherence in
community-based research
Issues: Studying adherence (adh) in community-based
settings is important for understanding medication-taking behavior outside
of clinical trials.
Description: A cross-site evaluation was
conducted to prospectively measure adh among participants (pts) in 12
government-funded adh support programs in the US. Data collected through
interviews included standardized 3-day adh self-report and other adh questions
such as time of last dose missed, difficulty taking medications on time,
and difficulty following special medication instructions. HIV RNA levels
and CD4 counts were collected through chart abstraction. Baseline characteristics
of the 606 pts include 42% women, 66% African American, 5% Latino, 29%
with unstable housing, and 46% with a history of mental illness. Pts had
a mean age of 40 yrs, mean CD4 count of 228 + 232 cells/mm3 and mean viral
load of 4.34 + 1.29 log10 copies/mL.
Lessons learned: Issues and challenges in
measuring adh in community-based settings included:
- social desirability bias associated with interview-administered adh self-report (interview necessary because of low literacy in pts);
- dates of laboratory values obtained from chart abstraction did not necessarily correspond to the dates of adh measures, adding complexity to data analysis when trying to correlate variables; and
- several methods can be used to calculate adh rates, e.g.
- taking an average of the mean number of doses taken over doses prescribed for each individual medication, and
- calculating doses taken over doses prescribed for the entire regimen, where a missed pill of any antiretroviral medication would be counted as a missed regimen dose.
Recommendations: A uniform approach to assessing
adherence is urgently needed. Collecting blood for HIV laboratory values
concurrently with interviews would improve analysis. The method for calculating
adh rates need to be clearly specified in adh research, as various methods
can yield different results.
CASE
The New York Academy of Medicine
1216 Fifth Avenue
New York, NY 10029
(212) 822-7237
Last updated: 4/11/02