The purpose of this study was to examine the influence of cognitive functio
n and other biopsychosocial factors on test-retest agreement, four-week var
iability, and intensity of self-reported pain using the verbal 0 to 10 scal
e and a pain thermometer in 115 nursing home residents over four weeks. Pai
n was assessed twice on three days during week I, and once each during week
s 2, 3 and 4. A forward stepwise regression procedure was used to examine t
he influence of biopsychosocial parameters (age, race, gender, educational
status, marital status, comorbidity, cognitive function, depression, social
support, physical function and sell-rated health) on pain intensity, test-
retest agreement and variability. There was a quadratic association between
cognitive function and test-retest agreement with the 0-10 scale; resident
s with Folstein scores of 22-26 were more likely to show disagreement (50%
of 34) than residents with scores <22 or >26 (7% of 71). Higher Folstein sc
ores were also associated with greater pain intensity for both pain scales
(p<0.001). Baseline pain intensify was significantly related to pain variab
ility (0-10 scale only). The clinician should be cognizant of these relatio
nships when interpreting verbalizations of pain in long-term care facilitie
s. (C) 1998, Editrice Kurtis.