OBJECTIVES: A systematic literature review was performed to examine economic modeling approaches utilized in published economic evaluations (EEs) of health interventions for attention deficit hyperactivity disorder (ADHD).

METHODS: Key biomedical databases (EMBASE^®, MEDLINE^®, NHS EED) and ISPOR database were searched to identify EEs published in the ADHD population, in the English language over the last 10 years. Citation snowballing and grey literature searches were conducted to gather comprehensive evidence. Predefined extraction forms were used to capture (i) study characteristics, (ii) modeling approaches; (iii) costs, utilities, and benefits; and (iv) discounting.

RESULTS: A total of 30 studies from 37 publications met the inclusion criteria. Most of the EEs (n=22) were conducted in children and adolescent populations, whereas, eight assessed adult (≥18 years) patients with ADHD. Overall, cost-utility analysis (n=18) was the most reported EE followed by cost-effectiveness analysis (n=5), and cost-minimization analysis (n=4). The EEs were conducted from the perspective of a third party/payer in most of the studies (n=22), while five were conducted from the societal perspective. The time horizon ranged from 12 weeks to a lifetime with a 1-year horizon utilized by almost 50% of studies (n=14). Few studies reported cycle length (n=4; 1 day, 1 week, and 1 month) and discounting on cost and benefits (n=6, range 0%-5%) details. Markov models were the most frequently utilized (n=13) followed by the decision tree approach (n=5). The health states in the Markov models ranged between 2 to 18 with treatment initiation, response, no-response, and treatment discontinuation being the most common health states.

CONCLUSIONS: We recommend using parsimonious Markov cohort models to assess the cost-effectiveness of future treatments in ADHD due to their versatility in adopting different types of health states that patients can potentially experience.