PSY82 - The Pain Assessment for Lower Back Symptoms (Pal-S): Refinement of A New Pro Instrument Through A Mixed Methods Approach

Abstract

BACKGROUND & OBJECTIVE Objectives The Pain Assessment in Low Back Pain Symptoms (PAL-S) is a Patient Reported Outcome (PRO) instrument being developed to assess the key symptoms of chronic low back pain (cLBP). Qualitative development included both concept elicitation and cognitive interviews. As part of the ongoing development of the instrument, we further evaluated and refined the PAL-S using a mixed methods approach. Methods Adults self-reporting a clinical diagnosis of cLBP were recruited from an existing US-based commercial survey panel to participate in a pilot quantitative study. Qualifying participants completed a web-based survey consisting of the 14-item PAL-S and items assessing clinical, treatment, and demographic characteristics. Study data was analyzed to assess itemand scale-level performance of the PAL-S using Rasch Measurement Theory analyses. Following analysis and modification, two waves of cognitive interviews were conducted to evaluate respondent understanding of the revised PAL-S. Results The dataset included 598 respondents (mean age: 55.5±12.6; 67.9% female; 88.0% white; and 54.0% married) who had cLBP for mean of 15.2±11.5 years. The Rasch analyses item threshold maps showed only two items having ordered thresholds, suggesting that respondents experienced increased difficulty distinguishing between options at the lower levels of the 0-10 scale. Simulations collapsing the responses to a best-fit 4-point response scale resulted in improved ordering of thresholds, suggesting a more optimal response option structure. Based on these findings, the numeric response scale of the PAL-S items was replaced with a 4-point verbal rating scale incorporating response choices such as not at all, slight, moderate, and severe. Findings from eight cognitive interviews confirmed patient comprehension and relevance of the revised instrument. Conclusion The mixed-methods approach proved valuable to the ongoing development of the PAL-S, as Rasch analyses identified a need for refinement of the response scale. The measurement properties of the revised PAL-S will be evaluated in additional web-based and clinic-based quantitative studies. Although a number of patient-reported outcome (PRO) instruments are available in low back pain (LBP), none of them were developed based on the most recent FDA PRO Guidance (2009) and would not be acceptable to the FDA to support labeling claims. Consequently, Forest Research Institute and partner Grunenthal in collaboration with experts have developed the Pain Assessment for Lower Back-Symptoms (PAL-S) instrument. The PAL-S was developed using state of the science best practices and in accordance with the FDA PRO Guidance (2009) to ensure evidence of content validity and measurement appropriate to support a labeling claim using PRO data. The early development work to create the PALS has included: • Completion of systematic reviews of the LBP literature and existing PRO instruments • The formation of an expert panel of clinical and methodological experts to provide advice during the development process • The completion of forty-three qualitative concept elicitation interviews conducted to identify the LBP-related concepts that are most important and relevant to the patients’ experience • A formal item-generation process in which evidence from the concept elicitation interviews, systematic literature reviews, and expert input was used to develop the content of the PAL-S • Four waves of qualitative cognitive interviews with patients with LBP to evaluate and refine the draft instruments This process has resulted in a 14-item developmental version ready for further quantitative testing. The primary aim is to collect initial quantitative data to further evaluate the performance of the PAL-S as appropriate PRO measures for use as endpoints in clinical studies of low back pain. The information gathered in this study will be used in the process of instrument refinement (item response theory, item reduction and/or modification), assessment of measurement characteristics, and finalization of the scoring approach to support the use of the PAL-S in subsequent studies.