This study was part of a mailed survey of individuals with postoperative DCM/R (Sapporo Maruyama Study), and a secondary analysis was conducted with objectives different from those previously reported [16]. The sample size required for the multiple regression analysis of this study, according to G*Power 3.1 (Heinrich-Heine-Universität Düsseldorf), was 127 patients with medium effect size (f²=0.15), a significance level of 0.05, power of 0.80, and 12 independent variables described.

The study was conducted following the principles of the Declaration of Helsinki. Approval was obtained from the Ethical Review Committee of Sapporo Maruyama Orthopedic Hospital before conducting the study (approval number: 35). The study’s explanation and a research participation consent form were enclosed with the questionnaire, and the participants were informed to sign the consent form if they were willing to participate.

This study’s inclusion criteria were as follows: (1) individuals aged at least 20 years (since 20 years was the minimum age to provide independent informed consent for participation in studies in Japan at that time); (2) those diagnosed with DCM/R by a spine surgeon based on clinical and radiological findings and who had undergone surgery for DCM/R between December 2017 and June 2021 at Sapporo Maruyama Orthopedic Hospital; and (3) those who did not receive personal care because of cognitive decline. The diagnosis of DCM/R was made by specialists in spine surgery based on clinical and radiological findings. Surgery was indicated for patients whose symptoms did not improve with conservative treatment, or whose motor dysfunction or pain greatly interfered with their daily life. Decompression with fixation was chosen when radiological findings showed dynamic instability of the cervical spine or severe compression of the cervical spinal cord and/or nerve roots, while decompression (laminoplasty) was chosen in other cases. Thus, decompression with fixation was preferred when the disease was severe. Individuals with cervical spine fractures and dislocations, spinal cord tumors, previous cervical spine surgery, and a diagnosis of psychiatric or neurological disease (e.g., multiple sclerosis and cerebrovascular disease) were excluded.

In total, 293 individuals who met the inclusion criteria and did not meet the exclusion criteria were selected from the Sapporo Maruyama Study Database. The questionnaire was mailed to all participants in July 2022, i.e., at least 1 year postoperatively. However, 13 participants (4.4%) had unknown addresses; therefore, we finally mailed the questionnaire to 280 individuals.

Imputing missing values is recommended because listwise deletion of missing values in the analysis reduces the reliability of the estimates [23]. Therefore, missing values were completed using the hot-deck method after excluding respondents who did not answer most of the questions or those who reported having pain on the EQ5D pain domain but did not report a score of ≥1 on the NRS and vice versa. The hot-deck imputation method is a technique where non-respondents (recipients) are matched with similar respondents (donors) and replaced by the observed values of the donors [24]. Hot-deck imputation method is used in various surveys, including the United States Census Bureau [25].

We performed hierarchical multiple regression analysis, which is a multiple regression analysis in which independent variables are entered in several steps and is used to assess the extent to which the additional independent variables contribute to the improvement of the model [26], using the EQ5D index as the dependent variable. In step one, sex (0, male; 1, female), age, surgical technique (0, decompression; 1, decompression and fixation) duration of postoperative days, and NDI were used as the independent variables. In step two, the NRS score was added as an independent variable 2; the TSK-11 score was added in step three; and in step four, paid work, light exercise, walking, strength training, and gardening were added. Each time an independent variable was added, we assessed whether the model improved and subsequently determined the final model. The referred statistics are the changes in F (ΔF) and R² (ΔR²) values; if the F-value is significantly larger, the model is judged to have been improved by the additional independent variables. Finally, the regression coefficients for the final model were obtained.

Statistical analysis was performed using R version 4.2 (R Foundation), with a significance level of p=0.05.

Here, we examined whether neck pain intensity, kinesiophobia, and physical activity are independently associated with HRQOL in individuals with postoperative DCM/R based on the FAM and evaluated their strength if an association exists. This study’s results showed that neck pain, kinesiophobia, and physical activity, such as paid work and walking, were independently associated with HRQOL to the same extent in individuals with postoperative DCM/R, which confirms our hypothesis.

In this study, the mean NRS score for neck pain and EQ5D index of the participants were 3.0 and 0.738 points, respectively. A study that followed the postoperative course of patients who underwent cervical spine surgery showed that the NRS score for neck pain and EQ5D index were approximately 3 and 0.5 points, respectively, at 2 years postoperatively, with stable values at 1 and 2 years [26]. HRQOL was better among our study participants, although the neck pain intensity was similar between our study participants and those of Revesz et al.’s [27] study. HRQOL was also better in our study participants than in those in other reports 1 year after cervical spine surgery [28]. However, the HRQOL of our study participants was lower than that of the general population, as the general population’s mean EQ5D index was 0.83 points, even for individuals aged >75 years [29]. The factors contributing to better postoperative HRQOL of the participants in our study are unknown, and one factor for this may be our inclusion of individuals with DCM/R for >1 year; therefore, we included those with DCM/R for up to 3 years.

We demonstrated that neck pain intensity, kinesiophobia, and physical activity were independently associated with HRQOL in patients with postoperative DCM/R. This finding supports the use of the FAM to better understand the relationship between HRQOL and neck pain in patients with postoperative DCM/R.

A report on contemporaneous improvements in neck pain and HRQOL postoperatively [4] suggested an association between neck pain and HRQOL. We hypothesized that pain interacts with various issues, such as sleep disruption, emotional disturbance, and reduced labor productivity (presenteeism), which directly or indirectly reduce HRQOL. Therefore, the HRQOL model that focuses on kinesiophobia and physical activity was developed in this study; however, there is room to improve and expand the model.

Kinesiophobia and physical activity were independently associated with HRQOL and neck pain in this study. A systematic review indicated that kinesiophobia is a psychosocial factor involved in the HRQOL of patients with chronic low back pain [30]. Kinesiophobia has also been reported as a risk factor for poor HRQOL in patients after undergoing hip surgeries [31]. Therefore, generalizing that kinesiophobia caused by pain resulting from bone and joint diseases, regardless of the disease, may have a negative impact on HRQOL. However, the relationship between pain intensity and kinesiophobia, a psychosocial variable, is not necessarily unidirectional.

Kinesiophobia is the largest factor associated with HRQOL, followed by paid work, walking, neck pain intensity, and light exercise. Paid work is a measure to improve physical activity because performing one's job duties or commuting to work involves physical activity, i.e., work-related physical activity. However, work-related physical activity was reported to differ between white- and blue-collar workers [32]; therefore, as in this study, the frequency of performing paid work alone could not accurately represent the amount of physical activity. Even if eliminating neck pain is difficult, interventions to reduce kinesiophobia and promote physical activity, such as paid work and walking, can be modified, indicating that these factors may be effective in improving HRQOL. This study did not measure the amount of physical activity using a tri-axial accelerometer or an international standard questionnaire but only asked about the modalities and the frequencies of physical activity they usually perform; therefore, whether paid work and walking are related to HRQOL dose-dependently remains unclear. Assessing the quantitative and qualitative impact of physical activity on HRQOL will contribute to further understanding and effective support of HRQOL in individuals with DCM/R who underwent surgery. Finally, contrary to the hypothesis, the conduct of light exercise was found to be negatively associated with HRQOL. The etiology of this cannot be addressed in this study. However, it was hypothesized that those individuals who had neck complaints tried to move their necks more than those who did not have any complaint.

This is the first study to examine factors associated with postoperative HRQOL in individuals with DCM/R based on the FAM framework. The knowledge obtained through this study to better understand the relationship between postoperative HRQOL and neck pain can contribute to developing postoperative rehabilitation strategies.

However, this study had some limitations. First, the rate of questionnaire collection was low. Consequently, a low response rate can lead to sampling bias, raising concerns regarding the robustness of the results. Second, as previously mentioned, the physical activity levels were not quantified. Therefore, reporting past performance does not rule out recall or social desirability biases, and the quantitative relationship between physical activity and HRQOL cannot be determined. Because of these methodological limitations, this study’s results should interpreted with caution, and our research should be developed to clarify the quantitative and qualitative relationship between physical activity and HRQOL. Third, this was a cross-sectional study. Although this study hypothesized a causal relationship based on the FAM, determining the causal relationship between HRQOL and its associated factors was difficult. Therefore, future interventional studies on kinesiophobia and physical activity should be designed to establish a causal relationship. Finally, HRQOL was assessed as a unidimensional construct. We used the EQ5D index value that operationally defines scores of 0 as “a state as bad as being dead” and 1 as “full health.” The 36-item Short Form Survey uses a two-factor model of physical and mental health [33]. Therefore, further analysis using HRQOL substructures may provide additional information that requires validation.

In conclusion, this study reveals that the reduction of neck pain, reduction of kinesiophobia, and promotion of physical activity, such as paid work and walking, may be effective intervention strategies to improve HRQOL in individuals with postoperative DCM/R.