Convergent and Discriminative Validity of the PROMIS Physical Function 4 Questionnaire for Assessing Pain-related Disability in Low Back Pain Patients Seeking Chiropractic Care.

Abstract

A prospective cohort study. To investigate construct validity by examining the convergent and discriminative validity of the Patient-Reported Outcomes Measurement Information System Physical Function 4a (PROMIS-PF4) questionnaire in low back pain (LBP) patients seeking care from Danish chiropractors. The Roland-Morris Disability Questionnaire (RMDQ) is often used to assess physical functioning in LBP. However, it consists of 24 items, which is time consuming to complete in clinical practice. The PROMIS-PF4 questionnaire has only four items and may be more applicable for clinical use. Patients with LBP seeking care from chiropractors in Denmark completed the PROMIS-PF4, RMDQ, Subgroups for Targeted Treatment (STarT) Back screening tool, and Numeric Pain Rating Scale (NPRS) in advance of their first appointment with the chiropractor. Convergent validity was assessed using Spearman correlation coefficients between the PROMIS-PF4 and RMDQ and NPRS, respectively. Discriminative validity of the PROMIS-PF4 was assessed by calculating the area under the receiver operating characteristic curve (AUC) when plotting the PROMIS-PF4 t score against a reference case for RMDQ; NPRS; and high and low risk groups using the STarT Back screening tool. Among 356 patients the PROMIS-PF4 questionnaire had strong convergent validity with the RMDQ ( r =-0.76) and moderate convergent validity with the NPRS ( r =-0.42). The PROMIS-PF4 had good and acceptable discriminative validity for disability (AUC=0.88) and high risk of persisting disability (AUC=0.72), and poor or no discriminative validity for pain intensity (AUC=0.66) and low-risk of persisting disability (AUC=0.26), respectively. As hypothesized, for convergent validity the PROMIS-PF4 has stronger correlation with the RMDQ than the NPRS and good discriminative validity for identifying patients with pain-related disability and at high risk of persisting disability but not for identifying pain intensity or low-risk of persisting disability. Consequently, the PROMIS-PF4 has adequate construct validity for measuring pain-related disability in an LBP population in chiropractic care.