Oral Presentation Australian and New Zealand Society for Geriatric Medicine Annual Scientific Meeting 2025

What digital decoding can tell us about how people living with moderate to severe dementia experience pain (119539)

Jeffery Hughes 1 2 , Paola Chivers 3 , Ipsit Vahia 4 , Kreshnik Hoti 2 5
  1. Curtin University, Perth, Western Australia, Australia
  2. PainChek Ltd, Sydney, New South Wales, Australia
  3. Child and Adolescent Health , Western Australian Department of Health, Perth, Western Australia, Australia
  4. Harvard Medical School, Harvard University, Boston, Massachusetts , USA
  5. Faculty of Medicine, University of Prishtina, Pristina, Kosovo

Aims: To evaluate how people living with dementia unable to reliably self-report experience pain across different levels of pain.

Methods: This was a data-driven retrospective observational study. The study utilized de-identified PainChek® pain assessment data for a total of 89,230 residents living with moderate to severe dementia. Participants were from Australia, New Zealand and the United Kingdom.  The primary outcome was to determine the occurrence of pain indicators across various levels of pain and explore their predictive contributions on pain intensity. Outcomes of pain assessments conducted at rest and post-movement were analyzed using a binary logistic generalized estimating equation.

Results: 2,125,022 pain assessments were conducted on 89,230 residents over a period of 5.3 years. During this period, a total of 47,158 (52.8%) residents had at least one recorded pain episode. Independent of sex (χ2=2.2 p=0.139) or age (χ2=2.7 p=0.101) residents were more likely to have higher levels of pain when assessed post-movement (OR=2.5 CI 2.4-2.7, χ2=1129.6 p<0.001).  All 42 PainChek® pain indicators were predictive of pain, confirming the value of multidimensional pain assessment. The number and presence of individual pain indicators increased incrementally as pain intensity increased from mild to severe, with the presence of some indicators (e.g. restlessness, resisting care) predictive of higher pain.

Conclusion: Our findings provide real world evidence on how people with moderate to severe dementia display their pain experience across different levels of pain. Further, they support movement-based assessment of pain and confirm the role of multidimensional pain assessment.