PARO evidence from the Queensland trial
What the Griffith University PARO trial across 28 Queensland aged care facilities found in 2017, what it did not find, and how to cite it in a proposal.
Anyone writing a proposal for PARO in Australia ends up citing one study: the Griffith University cluster randomised trial published in 2017. This page summarises it, including the results that tend to get left out.
Who ran the trial and how?
Moyle and colleagues at Griffith University ran the trial in 28 long-term care facilities in South-East Queensland, with 415 residents living with dementia. Facilities were assigned to one of three groups:
- PARO, switched on.
- A plush version of PARO, switched off, looking the same but not moving or responding.
- Usual care.
Residents in the two active groups had 15 minute sessions three times a week for 10 weeks. The results were published in the Journal of the American Medical Directors Association (2017, volume 18, pages 766 to 773).
What did it find?
- On video observation, PARO reduced agitation compared with usual care.
- PARO also increased observed pleasure compared with usual care.
- Residents engaged more with PARO, verbally and visually, than with the plush toy.
What did it not find?
On the staff-rated Cohen-Mansfield Agitation Inventory short form, there was no difference between the three groups. A proposal that quotes only the video results overstates the trial.
A cost study from the same trial (Mervin and colleagues, 2018) found both PARO and the plush toy were cost-effective ways to reduce agitation, and the plush toy was slightly better value for money.
What do later reviews add?
A 2025 meta-analysis of 15 randomised trials with 705 participants (Fan and colleagues, Journal of Medical Internet Research) found robot interventions reduced agitation and anxiety in people with dementia. It found no significant effect on cognition, depression or quality of life. Longer programs did better, and animal-shaped robots did as well as humanoid ones.
Describing the evidence in a proposal
- Say PARO reduced agitation compared with usual care in an Australian trial, and improved engagement.
- Do not claim it reduces medication, improves memory or works for every resident.
- Include the null result on the staff-rated scale. Reviewers who know the study will look for it.
- Plan to record each person’s response during your own use.
The trial was in residential aged care. For wards, see hospital and palliative use. For the price side of a proposal, see the PARO price guide.