Psychometric validation of the electronic chronic pain questions (eCPQ) in a primary care setting

• Published in: Current medical research and opinion 2017-01, Vol.33 (1), p.137-148
• Main Authors: Coyne, Karin S., Currie, Brooke M., Donevan, Sean, Brodsky, Marina, Asmus, Michael J., Krichbaum, Daniel W., Cappelleri, Joseph C., Hegeman-Dingle, Rozelle, Sadosky, Alesia, Whipple, Samuel Z., Burbridge, Claire, Mulhem, Elie, Hillenberg, J. Bruce
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Chronic pain; Chronic Pain - psychology; eCPQ; Female; Humans; Male; Middle Aged; Pain Measurement; primary care; Primary Health Care; psychometric; Psychometrics; reproducibility; Reproducibility of Results; validity
• ISSN: 0300-7995; 1473-4877
• DOI: 10.1080/03007995.2016.1240666

Objective: Collecting data that helps evaluate different types of pain may improve physicians' decision-making with regard to treatment selection and on-going monitoring of patients. To date, no chronic pain assessments have been widely implemented in primary care. The aim of this study was to psychometrically validate the electronic Chronic Pain Questions (eCPQ) in a primary care setting. Research design and methods: All men and women ≥18 years arriving at two similar primary care clinics in southeastern Michigan were invited to participate. Clinic staff verbally administered the eCPQ to patients and recorded their answers into the electronic medical record (EMR) prior to physician consultation with results available for physician review. Concurrent validity was assessed using Spearman correlations between eCPQ and patient-completed ancillary measures. Known-group validity was assessed by stratifying patients on self-reported chronic pain as well as by pain diagnosis (i.e. ICD-9 codes). To compare patients with chronic pain versus no chronic pain t-tests and chi-square tests were performed. Reproducibility was assessed between interviewer- and self-administration over time. Results: A total of 534 patients were invited to participate and 455 patients consented to take part in the study (85.2% response rate); 395 patients had analyzable eCPQ data; 70.1% were Caucasian; 68.1% female; mean age was 43.4; 52.7% (n = 208) self-reported chronic pain. Correlations between eCPQ and ancillary measures supported concurrent validity. Excellent discrimination between groups was evidenced based on self-reported chronic pain and ICD-9 diagnosis. Patients with self-reported chronic pain reported significantly (p