OBJECTIVES: To systematically identify and review published evidence on health-state utility weights in paediatric and adolescent ADHD.  METHODS: Utility weights were identified as part of a wider systematic review, which was performed to identify input parameter values for an economic model in ADHD. PubMed, EMBASE, Cochrane National Health Service’s Economic Evaluation Database, EconLit, and Health Economic Evaluations Database were searched from 1960 to May 2011. The review focused on articles reporting utility weights in ADHD by level of treatment response.  RESULTS: A total of 1,648 unique titles and abstracts were retrieved; 1,630 records were excluded upon title (or abstract) review and 12 upon full-text review. The review included 6 studies reporting utilities by level of response, 2 of which further stratified utilities by treatment type. One study reported utilities for 4 health states (‘normal’, ‘borderline to mildly ill’, ‘moderately to markedly ill’, and ‘severely ill’). One study used the EQ-5D instrument completed by parents; 2 assessed parents’ preferences using standard gamble; 1 used statistical mapping (for responder and nonresponder states) and time trade-off (TTO) interviews and visual analogue scales (for 4 disease-severity health states); 1 calculated scores from other utility studies; and 1 did not report methods explicitly. Utilities for responders ranged between 0.98 and 0.80; utilities for non-responders ranged between 0.93 and 0.70. TTO-derived utilities for the 4 health states were rated by members of the general public and ranged from 0.839 ('normal') to 0.444 ('severely ill'). The review assessed the methodological compliance of each included study with NICE Reference Case.  CONCLUSIONS: Our review identified utility estimates potentially suitable for use in ADHD economic evaluations. Appropriateness of these for inclusion in a particular analysis needs to be assessed in terms of the modelled heath states and the clinical outcomes used to define these.