Construct validation of the Japanese Core Outcome Measures Index and the impact of diseases on patient-reported outcome measures in preoperative patients with lumbar spinal stenosis and disk herniation: a single-center observational study

This cross-sectional study compared the QOL of preoperative patients aged 60 years and above with LSS or LDH. The study adhered to the STROBE statement and received approval from the Institutional Review Board of Saitama Medical Center, Saitama Medical University (No. 1969-III). The study was conducted between April 2018 and May 2023. In this study, we classified individuals aged 60 and above as elderly, drawing from prior research on QOL assessments [15]. Patients aged 60 years and older, seeking consultation at our center and scheduled for lumbar spine surgery based on the assessment of surgical eligibility for LSS or LDH, were enrolled in the study. In this study, cases of LSS did not involve a prolapsed nucleus pulposus contributing to lower limb neurological symptoms, and surgical procedures on the intervertebral disk were not conducted in these instances. Specifically, decompression was achieved by removing vertebral arch, partial facet, and/or ligamentum flavum. Conversely, cases of LDH were diagnosed based on the identification of compression on the cauda equina and/or nerve roots caused by a prolapsed nucleus pulposus through preoperative imaging, such as MRI. Furthermore, intraoperative observations confirmed the presence of herniated disks, leading to the performance of herniotomy and discectomy. Among the 251 patients scheduled for surgery, 79.2% (199 individuals, including 147 diagnosed with LSS and 52 with LDH) completed all necessary PROMs and were consequently included in the final analysis.

The participants’ baseline demographic and clinical characteristics were collected, including age, sex, body mass index (BMI: kg/m²), ASA classification, smoking status, history of diabetes, hypertension, hyperlipidemia, and arrhythmia (Table 1). The variables, including these comorbidities, were extracted from the medical records of each patient as documented by the attending physician.

To assess the preoperative QOL of patients, the following PROMs were utilized: Japanese COMI, Japanese ODI, EQ-5D-3L, and SF-12v2. The patients filled out these questionnaires at the point when the decision for lumbar spine surgery was determined. The severity of lumbar symptoms was assessed using the COMI questionnaire, which covers five domains: pain, function, symptom-specific well-being (SSWB), general QOL, and disability [6].

The COMI encompasses seven elements: back pain, leg/buttock pain, function, symptom-specific well-being, general quality of life, social disability, and work disability. All assessments pertain to the preceding week, except for the two disability items, which consider the past four weeks. Back and leg/buttock pain are evaluated on distinct 10-point graphic rating scales, while the remaining items are rated on a 5-point scale. A higher score indicates a more unfavorable status. Calculations are performed for individual domains and a summary score [16]. The higher score between back or leg/buttock pain is initially chosen as the pain domain score for the latter. Subsequently, the scores for other items are converted from their 5-point scales to a 0 to 10-point range using 2.5-point increments (0, 2.5, 5.0, 7.5, 10.0). Social and work disability scores are averaged to create a unified disability domain score. The summary score results from averaging the five domain scores (now each scored 0–10)—pain, function, symptom-specific well-being (SSWB), general quality of life, and disability—providing a comprehensive score ranging from 0 to 10 (representing the best to worst health status) [17, 18]. In this study, we utilized the Japanese version of the COMI, which has been previously validated for cross-cultural adaptation [6]. The ODI questionnaire evaluated the impact of lumbar pain on daily life [19]. The ODI comprises ten items related to daily activities, sleep, and social functioning. Each item is scored on a scale of 0 to 5, with higher total scores indicating more significant disability. The ODI score is calculated by dividing the sum of the scores by 50 and expressing it as a percentage, ranging from 0 to 100. EQ-5D-3L is a questionnaire that comprehensively assesses HRQOL based on five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension is scored on a scale of 1 to 3, with higher scores indicating more problems [20, 21]. The utility score is derived from these five dimensions and ranges from 0 (or negative) to 1, where 1 represents perfect health and lower scores indicate poorer health. The SF-12v2 questionnaire provides a comprehensive measurement of patients’ physical and mental health status, comprising eight health domains: physical functioning, role physical, bodily pain, general health, vitality, social functioning, emotional role limitations, and mental health [22]. Summary scores can be calculated from these domains, including physical component summary (PCS), mental component summary (MCS), role component summary (RCS), and norm-based scores. Considering biases due to fatigue and stress, the order of responding to these PROMs was randomized. To capture patients’ responses to these PROMs more accurately, they completed them in a separate room away from the physician’s presence. In cases of any questions or concerns, dedicated administrative staff assisted.

We calculated the sample size using the G*Power software for power analysis (http://​www.​gpower.​hhu.​de/​). Since there have been no previous studies of this nature, we estimated the effect size (d) for the independent t test to compare the results of each PROM between the LSS and LDH groups as 0.5 (Medium, α = 0.05, 1-β = 0.8). Furthermore, considering the ratio of elderly LSS to LDH patients in our hospital (3:1), the number of subjects needed for the t test was determined to be at least 170 (LSS: 127 cases, LDH: 43 cases). Therefore, the number of cases in this study can be considered sufficient.

Relying on prior research [5], we evaluated the construct validity of the preoperative COMI score and other PROMs using Spearman’s correlation coefficient.

The Mann–Whitney U test and χ² test were employed for continuous and categorical variables, respectively, to compare baseline PROMs and demographic information between the LSS and LDH groups. Additionally, a multivariate covariance analysis (MANCOVA) was performed to examine whether diseases (LSS and LDH) influenced the composite scores of each PROM and subscales. The dependent variables were the combined scores of each PROM or subscales of PROMs; the fixed factor was diseases, and covariates included gender, ASA classification, smoking status, history of diabetes, hypertension, hyperlipidemia, and arrhythmia. By conducting this analysis, we can demonstrate whether covariates influence the comparison results between the two groups or reflect of the impact of the diseases (LSS or LDH). Post hoc power analysis was conducted using G*Power 3.1.9.7.

Correlations between the COMI core index score, COMI subscales, and various PROMs are shown in Table 2. A positive correlation was observed between the COMI core index score and ODI (ρ = 0.73, p < 0.001), while negative correlations were noted with EQ-5D-3L (ρ = −0.71, p < 0.001) and SF-12v2 (PCS: ρ = −0.29, p < 0.001; MCS: ρ = −0.280, p < 0.001; RCS: ρ = −0.43, p < 0.001).

The results of the comparison of PROMs between the two groups are presented in Table 3. In comparing PROMs, the LDH group exhibited higher values in the COMI summary score, as well as in the function and disability subscales of COMI. The LDH group also showed higher values in the ODI score, along with the personal care, sexual life, and social life subscales of ODI. The EQ-5D-3L utility score was lower in the LDH group, with no significant differences observed in the subscales. While no significant differences were found in the component scores of SF-12v2, the LDH group had lower values in the bodily pain subscale.

The results of the MANCOVA are presented in Table 4. There was a significant overall impact of diseases (LSS and LDH) on PROMs and their subscales. Even after adjusting for covariates such as gender and comorbidities, diseases had a significant effect on COMI Core Index, ODI score, EQ-5D utility score, COMI-function, COMI-SSWB, COMI-disability, ODI-personal care, ODI-social life, and SF-12v2-bodily pain.