Background/aim: Randomised trials have demonstrated that occupational therapy can delay functional decline, improve quality of life and increase leisure participation in people with dementia. However, surveys conducted with occupational therapists suggest that clinical practice does not reflect the type of intervention shown to be effective in research studies. Case note audits can be used to quantify practice and demonstrate how and where provision of care could improve without the potential bias associated with self‐report.

Method: A total of 87 occupational therapy case notes were audited from different service contexts in two states in Australia. The case notes were reviewed against criteria including duration of service, assessments conducted and interventions used. Descriptive statistics were used to present the data and examine associations between intervention, age of the person with dementia and whether or not the person lived alone.

Results: Services tended to be short‐term with an average of 2.1 consultations per referral. The most common assessments related to home safety, falls risk and function. Intervention most commonly focussed on referrals to other services, environmental modification advice and assistive device prescription.

Conclusion: This audit reveals that current occupational therapy practice for people with dementia focusses on assessment and management of risk and is usually limited to a couple of consultations. Future work should seek to evaluate if such approaches to care are effective for people with dementia.