44. Does recall bias exist in lumbar spine surgery patients as reported through PROMIS questionnaires?

• Published in: The spine journal 2022-09, Vol.22 (9), p.S22-S23
• Main Authors: Arpey, Nicholas, Barrett, Joshua, Gerlach, Erik B., Alvandi, Bejan, Divi, Srikanth, Hsu, Wellington K., Patel, Alpesh A.
• Format: Article
• Language: English
• ISSN: 1529-9430; 1878-1632
• DOI: 10.1016/j.spinee.2022.06.058

Surgical outcomes are increasingly judged on patients' perceptions of the results. Despite the improved quantitative data provided by patient-reported outcomes (PROs), they remain susceptible to confounding factors on patients' interpretations such as recall bias: the inability to accurately recall prior impairment. No studies to date have reported the accuracy of patient recall using PROMIS outcomes after lumbar spine surgeries. The purpose of this study is to determine the presence and extent of recall bias in adult patients after elective lumbar spine surgery. Subgroup analysis of ongoing prospective study. This study included 98 patients who had undergone either lumbar decompression or lumbar decompression and fusion procedures. Baseline preoperative PROMIS Physical Function (PF), PROMIS Pain Interference (PI) and recalled preoperative PROMIS PF and PI. Patients, at least 2 years after their index surgery, completed a recall questionnaire comprised of the PF CAT and PI CAT as though it was a time immediately before their surgery. All patients had prospectively completed PROMIS PF CAT and PI CAT prior to surgery and at 3 months, 6 months, 1 year and 2 years. T tests were used to compare recalled PROs with actual baseline PROs. Correlation coefficients were calculated to evaluate the agreement between recalled PROs and baseline PROs. Regression analysis was performed to determine the impact of patient characteristics or clinical factors. Patient recollection of preoperative status at a minimum of 2 years postoperatively was significantly more severe than baseline preoperative status with mean difference in PF of -1.5 (p < 0.05) and PI of 2.1 (p < 0.01). There was only moderate correlation between recalled and baseline scores with regards to PF (r=0.52) and PI (r=0.38). No significant differences on recalled PROs were found based on age, gender, time between surgery and recalled outcomes, and duration of symptoms before surgery. Our data indicate that patient recall of preoperative status after lumbar surgery is significantly different than actual baseline preoperative status as measured by PROMIS PF and PI scores with only moderate correlation existing between the two. Patients' interpretations of improvement after surgery may be impacted by significant recall bias. This may have an impact on patient satisfaction and perceived value of lumbar spine surgery. This abstract does not discuss or include any applicable devices or drugs.