Using observational facial descriptors to infer pain in persons with and without dementia

BMC Geriatrics

Table 3 Step 2 - Which facial descriptors can best predict the self-report of pain (criterion: self-report of pain)

Videos:	All videos showing facial responses to noxious 5 kg pressure stimuli			All videos showing facial responses to noxious 5 kg pressure stimuli		Healthy elderly (“5 kg”)	Patients with dementia (“5 kg”)
Participant group:	All participants			Nurses	Controls	All participants
Overall model fit:	R² = 0.093 (p < .001)			R² = 0.091 (p < .001)	R² = 0.109 (p < .001)	R² = 0.160 (p < .001)	R² = 0.116 (p < .001)
	Beta weight	Zero-order r	Product measure*	Product measure	Product measure	Product measure	Product measure
Opened mouth	.332	.212	.070	.054	.090	.109	.047
Raising upper lip	.177	.082	.015	.019	.009	.001	.056
Frowning	.084	.064	.006	.014	.001	.044	.003
Narrowing eyes	.065	.058	.004	.011	.001	.009	.003
Looking frightened	.024	.042	.001	<.001	.004	<.001	.002
Looking tense	.005	.018	<.001	<.001	.002	.007	.006
Tightened lips	.069	−.021	< −.001	−.001	.006	−.008	<.001

Values are given separately for the whole sample and the whole videos presented, as well as separately for the videotaped groups of healthy older individuals and patients with dementia and separately for the two subject groups (nurses, controls)
*The product measure is a combination of beta weights and zero order correlations and uniquely reflects both direct and total effects of each predictor variable

ISSN: 1471-2318