BACKGROUND: The Oral Health Assessment Tool (OHAT) was a component of the Best Practice Oral Health Model for Australian Residential Care study. The OHAT provided institutional carers with a simple, eight category screening tool to assess residents' oral health, including those with dementia. This analysis presents OHAT reliability and validity results. METHODS: A convenience sample of 21 residential care facilities (RCFs) in urban and rural Victoria, NSW and South Australia used the OHAT at baseline, three-months and six-months to assess intra- and inter-carer reliability and concurrent validity. RESULTS: Four hundred and fifty five residents completed all study phases. Intra-carer reliability for OHAT categories: percent agreement ranged from 74.4 per cent for oral cleanliness, to 93.9 per cent for dental pain; Kappa statistics were in moderate range (0.51-0.60) for lips, saliva, oral cleanliness, and for all other categories in range of 0.61-0.80 (substantial agreement) (p < 0.05). Inter-carer reliability for OHAT categories: percent agreement ranged from 72.6 per cent for oral cleanliness to 92.6 per cent for dental pain; Kappa statistics were in moderate range (0.48-0.60) for lips, tongue, gums, saliva, oral cleanliness, and for all other categories in range of 0.61-0.80 (substantial agreement) (p < 0.05). Intraclass correlation coefficients for OHAT total scores were 0.78 for intra-carer and 0.74 for inter-carer reliability. Validity analyses of the OHAT categories and examination findings showed complete agreement for the lips category, with the natural teeth, dentures, and tongue categories having high significant correlations and percent agreements. The gums category had significant moderate correlation and percent agreement. Non-significant and low correlations and percent agreements were evident for the saliva, oral cleanliness and dental pain categories. CONCLUSION: The Oral Health Assessment Tool was evaluated as being a reliable and valid screening assessment tool for use in residential care facilities, including those with cognitively impaired residents.