TY - JOUR
T1 - Assessment of adherence to antiretroviral therapy in perinatally HIV-infected children and youth using self-report measures and pill count
AU - Farley, John J.
AU - Montepiedra, Grace
AU - Storm, Deborah
AU - Sirois, Patricia A.
AU - Malee, Kathleen
AU - Garvie, Patricia
AU - Kammerer, Betsy
AU - Naar-King, Sylvie
AU - Nichols, Sharon
PY - 2008/10
Y1 - 2008/10
N2 - BACKGROUND:: Parent/caregiver or child/youth self-report and pill counts are commonly used methods for assessing adherence to antiretroviral therapy among children and youth with HIV. The purpose of this study was to compare these different methods with one another and with viral load. METHODS:: Randomly selected parent/caregiver and child/youth dyads were interviewed using several adherence self-report measures and an announced pill count was performed. Adherence assessment methods were compared with one another and their relative validity was assessed by comparison with the child's viral load close to the time of the interview or pill count, adjusting for primary caregiver, child age, and child disclosure of the diagnosis. RESULTS:: There were 151 evaluable participants. Adherence rate by pill count was > 90% in 52% of participants, was significantly associated with log(RNA) viral load (p = .032), and had significant agreement with viral load <400 copies/mL. However, pill count data were incomplete for 26% of participants. With similar proportions considered adherent, a variety of self-report adherence assessment methods also were associated with log(RNA) viral load including: "no dose missed within the past 1 month" (p = .054 child/youth interview, p = .004 parent/caregiver interview), and no barrier to adherence identified (p = .085 child/youth interview, p = .015 parent/caregiver interview). Within-rater and inter-rater agreement was high among self-report methods. Three day recall of missed doses was not associated with viral load. CONCLUSION:: Findings demonstrate the validity of adherence assessment strategies that allow the parent/caregiver or child/youth to report on adherence over a longer period of time and to identify adherence barriers. Adherence assessed by announced pill count was robustly associated with viral load, but there was incomplete data for many participants.
AB - BACKGROUND:: Parent/caregiver or child/youth self-report and pill counts are commonly used methods for assessing adherence to antiretroviral therapy among children and youth with HIV. The purpose of this study was to compare these different methods with one another and with viral load. METHODS:: Randomly selected parent/caregiver and child/youth dyads were interviewed using several adherence self-report measures and an announced pill count was performed. Adherence assessment methods were compared with one another and their relative validity was assessed by comparison with the child's viral load close to the time of the interview or pill count, adjusting for primary caregiver, child age, and child disclosure of the diagnosis. RESULTS:: There were 151 evaluable participants. Adherence rate by pill count was > 90% in 52% of participants, was significantly associated with log(RNA) viral load (p = .032), and had significant agreement with viral load <400 copies/mL. However, pill count data were incomplete for 26% of participants. With similar proportions considered adherent, a variety of self-report adherence assessment methods also were associated with log(RNA) viral load including: "no dose missed within the past 1 month" (p = .054 child/youth interview, p = .004 parent/caregiver interview), and no barrier to adherence identified (p = .085 child/youth interview, p = .015 parent/caregiver interview). Within-rater and inter-rater agreement was high among self-report methods. Three day recall of missed doses was not associated with viral load. CONCLUSION:: Findings demonstrate the validity of adherence assessment strategies that allow the parent/caregiver or child/youth to report on adherence over a longer period of time and to identify adherence barriers. Adherence assessed by announced pill count was robustly associated with viral load, but there was incomplete data for many participants.
UR - http://www.scopus.com/inward/record.url?scp=58149157496&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=58149157496&partnerID=8YFLogxK
U2 - 10.1097/DBP.0b013e3181856d22
DO - 10.1097/DBP.0b013e3181856d22
M3 - Article
C2 - 18714204
AN - SCOPUS:58149157496
SN - 0196-206X
VL - 29
SP - 377
EP - 384
JO - Journal of Developmental and Behavioral Pediatrics
JF - Journal of Developmental and Behavioral Pediatrics
IS - 5
ER -