Methods: Data were obtained from a behavioral/educational intervention trial in inner city children with poorly controlled asthma. Structured questionnaires elicited demographics, MARS-5 (range=5-25), caregiver-reported ICS use, and caregiver stress (PSS4, CRISYS), quality of life (Juniper-QOL), and depression (CES-D). Participants reporting ICS use at enrollment were included. Caregiver-reported daily, scheduled (not prn) ICS use over the previous two weeks were classified as Adherent. Statistical analysis included Chi-square (categorical variables), and t-test or Wilcoxon rank sum test for continuous variables. MARS-5 score was categorized as high/low using median split.
Results: Our 230 child sample was mostly black (96.1%), male (58.3%), age 5.6+/-2.1 (range=3-10) and Medicaid-insured (92.2%). High MARS-5 was reported by 49.6% of caregivers (Median MARS-5=23). Although MARS-5 was associated with self-reported ICS adherence (p=0.02), it was concordant in only 55.2% of the sample. Neither CES-D (p=0.63), Juniper-QOL (p=0.21), PSS4 (p=0.10), nor CRISYS (p=0.25) was associated with caregiver-reported ICS adherence.
Conclusions: MARS-5 was concordant with caregiver-reported adherence in some inner city caregivers, but was not valid in a large proportion of the sample. Further study is needed to evaluate child/caregiver characteristics affecting the predictive value of MARS-5 as a caregiver-reported adherence measure.