PRIORITIZING CAREGIVERS' CONCERNS WITH INITIATING ADHD MEDICINE FOR THEIR CHILD: A BEST WORST SCALING PREFERENCE ELICITATION STUDY

Purpose: Caregiver decision-making about child attention-deficit/hyperactivity disorder (ADHD) treatment is strongly influenced by the caregiver's preferences, concerns, and priorities for their child. However, little is known about the heterogeneity of concerns among caregivers seeking treatment for their child. This study sought to elicit caregiver concerns with child ADHD treatment and to examine heterogeneity in how caregivers prioritize these concerns.

Methods: Caregivers of a child aged 4-14 years old and in care for ADHD were recruited from outpatient mental health and pediatric primary care clinics, support groups, and family support services in public schools throughout Maryland. A 16-item discrete-choice experiment, using Best-Worst Scaling preference elicitation method, was conducted to measure caregivers' concerns.  Participants evaluated 16 concerns, displayed six at a time in a profile, that were associated with initiating ADHD medication. Concerns spanned four domains: short-term concerns, long-term concerns, supportive network concerns, and societal concerns. Conditional logit models were used to estimate utility scores, using effects coding and under the assumption of sequential best-worst responses. Scores were ranked to assess relative importance. Heterogeneity was investigated via latent class analysis.   

Results: The 108 caregivers were on average 42 years old, primarily the biological mother, 65% Caucasian, and 27% African-American. Children were, on average, 10 years old, about one-third were diagnosed with ADHD for three years or less, and the majority took ADHD medication. The attributes with relatively higher scores included their child becoming a successful adult (0.474, p<0.0001), a doctor addressing their concerns (0.346, p<0.0001), and their child's school behavior improving (0.347, p<0.0001). The latent class analysis yielded two segments: outcome oriented and medication risk-averse. Outcome oriented caregivers made up 60% of the sample and identified important concerns as their child becoming a successful adult (p<0.0001), behaving better in school (p<0.0001), and receiving better grades (p<0.0001).  Medication-risk oriented caregivers made up 40% of the sample and selected as important concerns a doctor addressing their concerns (p<0.0001), medication side effects outweighing its benefits (p<0.0001), and medication risks affecting their child's future health (p<0.0001).

Conclusions: This work sheds light on the relative importance of concerns that can be antecedents or barriers to engaging in evidence-based treatment.  Identifying concerns that influence caregivers' treatment decisions can be usefully applied to enhance opportunities for shared-decision making between healthcare providers and caregivers.