We tested the usability, sensitivity and validity of the Health of the Nation Outcome Scales (HoNOS) in routine clinical practice in North America. Three pilot sites provided ratings on all inpatient and outpatient referrals over 4 months using versions covering children and adolescents (HoNOSCA), working-age adults and the over-65s. Data were entered using the routine administrative data system. Sixty-one percent of eligible patients had at least one HoNOS rating (n = 485). Following the initial rating, subsequent completion rates reached 80%. Ratings were sensitive to time and setting, with significantly higher scores in inpatients than outpatients. Individual diagnoses had different patterns of scores, further supporting validity.