There are several measures of coping strategies in response to nociception. These measures all correlate highly both with each other and with symptom intensity and magnitude of disability in patients with upper limb illness. This study aims to determine if distinct measures of coping strategies in response to nociception address the same underlying aspect of human illness behavior. Our primary study question was: is there one common aspect of human illness behavior measured by (1) the Pain Catastrophizing Scale (PCS); (2) the Psychological Inflexibility in Pain Scale (PIPS); (3) the Patient-Reported Outcomes Measurement Information System-Pain Interference (PROMIS-PI) Computer Adaptive Test (CAT); and (4) the Pain Self-Efficacy Questionnaire (PSEQ)? Secondarily, we aimed to determine which of the four questionnaires is most psychometrically sound. We measured correlations among questionnaires, coverage, reliability, completion time, and collinearity of these questionnaires when entered together in a multivariable model with the shortened version of the Disabilities of the Arm, Shoulder and Hand (QuickDASH) upper extremity disability questionnaire. In this prospective study, 138 consecutive new or followup English-speaking patients aged 18 years or older presenting to a tertiary care referral center with traumatic and nontraumatic upper extremity conditions were invited to participate between March and May 2014. One hundred thirty-four (97%) patients agreed to participate and completed the four questionnaires in random order before their visit with the physician. We used exploratory factor analysis to assess whether there was a single common trait-an underlying aspect of human illness behavior-measured by these questionnaires. Interquestionnaire correlation was assessed using Spearman rank correlation coefficients; coverage by assessing floor and ceiling effect (proportion of scores at lower and upper limit); reliability by Cronbach's alpha measure of internal consistency; completion time in seconds using Kruskal-Wallis analysis; and collinearity statistics through a regression model with QuickDASH. Exploratory factor analysis identified a common trait measured by these four measures-coping strategies in response to nociception-indicated by a substantial correlation of every individual questionnaire with the underlying trait (PCS: 0.74, PIPS: 0.84, PROMIS-PI: 0.83, PSEQ: -0.86). All interquestionnaire correlations were also large to substantial and were highest for PROMIS-PI with PSEQ (rho = -0.84, p < 0.001) and lowest for PROMIS-PI with PCS (rho = 0.67, p < 0.001). Internal consistencies were high (PCS: 0.93, PIPS: 0.88, PSEQ: 0.92, and not determined for the PROMIS-PI as a result of its CAT administration). PROMIS-PI was the quickest to complete (30 seconds [interquartile range, 24-44]) compared with the others (PCS: 91 seconds [66-122], p < 0.001; PIPS: 105 seconds [82-141], p < 0.001; PSEQ: 78 seconds [60-101], p < 0.001). The four coping questionnaires had a low partial r(2) and a relatively high variation inflation factor, indicating multicollinearity. PROMIS-PI was found to have the strongest correlation with QuickDASH (beta coefficient: 0.63; standard error: 0.10; p < 0.001). There is evidence that the four widely used measures of coping strategies in response to nociception address a single common aspect of human illness behavior, which negatively impacts upper extremity disability. Future studies assessing functional outcome should incorporate a measure of human illness behavior as it strongly relates to disability. Given that all of these measures address the same important aspect of human illness behavior, we recommend the PROMIS-PI CAT as the most efficient measure.
