To inform decisions about mental health resource allocation, planners require reliable estimates of people who report service demand (i.e. people who use or want mental health services) according to their level of possible need.

Using data on 6915 adults aged 16–64 years in Australia’s 2007 National Survey of Mental Health and Wellbeing, we examined past-year service demand among respondents grouped into four levels of possible need: (a) 12-month mental disorder; (b) lifetime but no 12-month mental disorder; (c) any other indicator of possible need (12-month symptoms or reaction to stressful event, or lifetime hospitalisation); (d) no indicator of possible need. Multivariate logistic regression analyses examined correlates of service demand, separately for respondents in each of levels 1–3.

Sixteen per cent of Australian adults reported service demand, of whom one-third did not meet criteria for a 12-month mental disorder (equivalent to 5.7% of the adult population). Treatment patterns tended to follow a gradient defined by level of possible need. For example, service users with a 12-month disorder received, on average, 1.6–3.9 times more consultations than their counterparts in other levels of possible need, and had 1.9–2.2 times higher rates of psychologist consultation. Service users with a lifetime but not 12-month disorder or any other indicator of need consumed a similar average number of services to people with mild 12-month mental disorders, but received relatively fewer services involving the mental health sector. Service demand was associated with increased suicidality and psychological distress in all levels of possible need examined, and with poorer clinical and functional status for those with 12-month or lifetime disorders.

Many Australians reporting service demand do not meet criteria for a current mental disorder, but may require services to maintain recovery following a past episode or because they are experiencing symptoms and significant psychological distress.