66. Defining PROMIS MCID for spine patients with three legacy PROs: an anchor-based approach

BACKGROUND CONTEXT Patient Reported Outcome Measurement Information System (PROMIS) assesses patients’ perception of their well-being, which may be used to gauge the effectiveness of an intervention. Identifying the PROMIS score change that represents a clinically important difference is important. The minimum clinically important difference (MCID) describes this threshold and has been determined for legacy PRO tools. Our objective is to define MCID values for PROMIS domains using SF-12, EQ-5D, and Global Rating of Change (GRC) as anchors. METHODS Patient selection prospectively collected data of patients 18 years or older referred through a single institution's orthopedic spine clinic for any spine pathology from November 2018 to January 2021 were reviewed. Patients were included if they underwent surgery and completed both a preoperative and at least one postoperative PRO survey. PROMIS Physical Function (PF), Pain Interference (PI), and Depression (DE) domain responses were assessed in addition to SF-12 Physical (PCS) and Mental Component Scores (MCS), EQ-5D Visual Analogue Scores (VAS), and Global Ratings of Change (GRC) scores. Anchor definitions for the SF12 and EQ-5D domains, differences between rescaled pre and postoperative scores of 0, 1, 2, or 3-4 were classified as “no change”, “minimal”, “moderate”, and “marked” improvement. GRC responses of -1 to 1, 2-3, 4-5, and 6-7 were defined as “no change”, “minimal”, “moderate”, and “marked” improvement respectively. MCID Calculation and Statistics Only patients that reported improvements in both the PROMIS score and the anchor question were included. Mean PROMIS improvement score was then defined as the MCID cutoff for a given anchor question. Improvement categories were then compared to the “no change” category. RESULTS The final cohort comprised 184 patients. The average age was 60.6 (SD 15.4) years, 44.2% were female, and 86.96% identified as White. Depending on the anchors used, MCID PROMIS PF thresholds detecting “minimal improvement” ranged 4.77-6.76, “moderate improvement” 5.74-7.74, and “marked improvement” 8.96-11.60 with most values statistically significantly different from “no change” (-3.09- -0.40). PROMIS PI thresholds for “minimal improvement” ranged 5.61-10.42, “moderate improvement” ranged 6.67-9.44, and “marked improvement” ranged 11.69-14.09 were statistically significantly different from “no change” (-0.44-2.4) (all p<0.001). PROMIS DE threshold for “minimal improvement” ranged 6.04-7.97, “moderate improvement” ranged 7.34-9.93, and “marked improvement” ranged 9.96-15.17 with all values significantly different from “no change” (1.59-2.39) except minimal improvement GRC mood (p=0.217) and minimal improvement GRC general (p=0.204). CONCLUSIONS This study describes MCIDs for PROMIS PF, PI, and DE using three established, unique PROs as anchors. Threshold values to detect a set degree of change for PROMIS PF, PI, and DE were similar across the selected anchors, with tighter ranges than previously described in literature. Commonly accepted MCID thresholds obtained by distribution methods are limited by a lack of external validation, which anchor-based methods used here provide. To our knowledge, we are among the first to derive MCIDs for various levels of improvement (minimal/moderate/marked) following spine surgery. Future large-scale studies validating these thresholds with additional legacy PROMs are warranted. FDA Device/Drug Status This abstract does not discuss or include any applicable devices or drugs. Patient Reported Outcome Measurement Information System (PROMIS) assesses patients’ perception of their well-being, which may be used to gauge the effectiveness of an intervention. Identifying the PROMIS score change that represents a clinically important difference is important. The minimum clinically important difference (MCID) describes this threshold and has been determined for legacy PRO tools. Our objective is to define MCID values for PROMIS domains using SF-12, EQ-5D, and Global Rating of Change (GRC) as anchors. Patient selection prospectively collected data of patients 18 years or older referred through a single institution's orthopedic spine clinic for any spine pathology from November 2018 to January 2021 were reviewed. Patients were included if they underwent surgery and completed both a preoperative and at least one postoperative PRO survey. PROMIS Physical Function (PF), Pain Interference (PI), and Depression (DE) domain responses were assessed in addition to SF-12 Physical (PCS) and Mental Component Scores (MCS), EQ-5D Visual Analogue Scores (VAS), and Global Ratings of Change (GRC) scores. Anchor definitions for the SF12 and EQ-5D domains, differences between rescaled pre and postoperative scores of 0, 1, 2, or 3-4 were classified as “no change”, “minimal”, “moderate”, and “marked” improvement. GRC responses of -1 to 1, 2-3, 4-5, and 6-7 were defined as “no change”, “minimal”, “moderate”, and “marked” improvement respectively. MCID Calculation and Statistics Only patients that reported improvements in both the PROMIS score and the anchor question were included. Mean PROMIS improvement score was then defined as the MCID cutoff for a given anchor question. Improvement categories were then compared to the “no change” category. The final cohort comprised 184 patients. The average age was 60.6 (SD 15.4) years, 44.2% were female, and 86.96% identified as White. Depending on the anchors used, MCID PROMIS PF thresholds detecting “minimal improvement” ranged 4.77-6.76, “moderate improvement” 5.74-7.74, and “marked improvement” 8.96-11.60 with most values statistically significantly different from “no change” (-3.09- -0.40). PROMIS PI thresholds for “minimal improvement” ranged 5.61-10.42, “moderate improvement” ranged 6.67-9.44, and “marked improvement” ranged 11.69-14.09 were statistically significantly different from “no change” (-0.44-2.4) (all p<0.001). PROMIS DE threshold for “minimal improvement” ranged 6.04-7.97, “moderate improvement” ranged 7.34-9.93, and “marked improvement” ranged 9.96-15.17 with all values significantly different from “no change” (1.59-2.39) except minimal improvement GRC mood (p=0.217) and minimal improvement GRC general (p=0.204). This study describes MCIDs for PROMIS PF, PI, and DE using three established, unique PROs as anchors. Threshold values to detect a set degree of change for PROMIS PF, PI, and DE were similar across the selected anchors, with tighter ranges than previously described in literature. Commonly accepted MCID thresholds obtained by distribution methods are limited by a lack of external validation, which anchor-based methods used here provide. To our knowledge, we are among the first to derive MCIDs for various levels of improvement (minimal/moderate/marked) following spine surgery. Future large-scale studies validating these thresholds with additional legacy PROMs are warranted.