OBJECTIVE:  From a Canadian societal perspective, little is known about the impact of Attention Deficit Hyperactivity Disorder (ADHD) in adults.  A methodology for the design and conduct of a web-based survey used to populate a Canadian economic model is described. METHODS: An IRB approved, wed-based survey was administered to evaluate the impact of ADHD in pre-identified Canadian subjects with this condition.  There were 3 subgroups of interest: an ADHD group with self-reported controlled symptoms, an ADHD group with self-reported uncontrolled symptoms, and a non-ADHD reference group. Eligible participants received a 28-item questionnaire (English or French) which evaluated socio-demographic characteristics, ADHD treatment, co-morbidities, health care resource utilization, functional status (Sheehan Disability Scale), productivity (Work Productivity Assessment and Activity Impairment; WPAI), and health-related quality of life (EQ-5D). Validation rules were pre-programmed to optimize data quality and survey completion at the point of data entry. EQ-5D utilities and productivity losses from WPAI were used to inform an economic model. Other data were used as supportive information. RESULTS: The targeted number of completed questionnaires, n=174, was reached within 2 weeks of study launch after 1,878 survey invitations were circulated.  More specifically, 69 controlled, 70 uncontrolled and 35 non-ADHD subjects responded. Among participants, 56% were male, mean age was 35 years, 43% were diagnosed with ADHD in adulthood, and the majority responded in English. Functional status and utility values were significantly lower for subjects with uncontrolled versus controlled ADHD.  Employment rates were lower (59% vs. 74%, p=0.06), and overall work impairment was significantly higher for uncontrolled versus controlled ADHD subjects, respectively. CONCLUSION: Web-based surveys are a cost and time efficient methodology for evidence gathering in support of economic evaluations. Despite some inherent limitations of online studies, such as generalizability, survey variables and outcomes can be tailored to collect data from populations and sub-groups of interest.