Pain intensity characteristics of referrals to national dementia behavior support programs in Australia

BACKGROUND: Recognizing pain in people with advanced dementia who cannot effectively communicate is difficult. As such, pain is underdetected and undermanaged in this group and can lead to behaviors and psychological symptoms of dementia, impaired quality of life, and significant caregiver distress. Further, little is known about the clinical experience of pain intensity in dementia. This study aims to describe the pain intensity characteristics of referrals to Dementia Support Australia, a national dementia-specific behavior support program in Australia. METHODS: Pain in a sample of referrals (period: January-November 2020) was evaluated during various activities (rest vs movement) using PainChek® , a medical device application (app)(Figure 1) linked to a web admin portal (WAP)(Figure 2) for non-verbal adults including those living with dementia. PainChek® consists of six domains (Face, Voice, Movement, Behavior, Activity, and Body) that have collectively a total of 42 items. The PainChek® app uses artificial intelligence to detect pain-indicative facial action units (e.g., brow lowering) in combination with a digital checklist of non-facial indicators to produce a pain intensity score as follows: 0-6 (no pain), 7-11 (mild pain), 12-15 (moderate pain), ≥16 (severe pain). The difference in pain scores according to the activity status (rest vs movement) was computed using Cohen's d values (e.g., 0.01-