The study was approved by the Ethics Committee of West China Hospital, Sichuan University (No. 971; ref: 2021).

Sample size calculation was performed using PASS 15 software with a two-sided significance level of α = 0.05 and statistical power of β = 0.2. Based on preliminary study results, the Neck Disability Index (NDI) was selected as the primary outcome measure with a standard deviation of 2 and a margin of error of 1.4. The sample size was estimated, and n1 = n2 = 35 was obtained. Considering a dropout rate of 20%, we estimated that a total of approximately 88 patients would be required (44 per group).

A total of 88 patients diagnosed with confirmed cervical spondylosis were assigned to undergo cervical spondylosis surgery. The patients were discharged between June 2, 2021, and July 7, 2022. Inclusion criteria for the study were: (1) age ≥ 18 years; (2) eligibility for cervical spondylosis surgery; and (3) ability to provide informed consent, communicate effectively, and understand/read Chinese language. All discharged patients were required to meet uniform discharge criteria based on standard clinical care.

Patients with a history of previous cervical spine surgery, diagnosed dementia, blindness, or deafness, cervical malformation or significant instability, a history of severe cervical trauma, pregnancy, rheumatoid arthritis, malignancy, active infection, or other systemic diseases, as well as patients who are not recommended for follow-up, and those who faced difficulty in attending in-person visits for diagnostics or treatment or declined to participate, were excluded from the study.

This study utilized convenience sampling as the sampling method. Patient recruitment was conducted prior to hospital admission. The head of the medical team explained the purpose and process of the program to patients who were about to undergo cervical spine surgery and invited them to participate in the study. Patients who agreed to participate were required to sign an informed consent form, confirming their voluntary participation in the research. Since the terms of the multidisciplinary follow-up contract needed to be personally confirmed and signed by the patients, blinding was not applicable in this study. Patients were non-randomly separated into two groups based on whether they signed a multidisciplinary team follow-up contract: patients who signed the contract were placed in the intervention group, while those who did not sign were placed in the control group. The sequential enrollment process started with the first eligible patient who signed the informed consent and continued until the last patient completed the 3-month follow-up study.

The study period was from June 2, 2021, when the first patient was enrolled, to July 7, 2022, when the last patient completed the follow-up. The study was conducted at the spinal surgery ward of West China Hospital, Sichuan University.

This study utilized the continuity of care model proposed by Sarah et al. (20). The model suggests that there are overlapping and hierarchical relationships among the three dimensions of continuity of care: patient-provider relationship, communication, and collaboration. The theoretical dimensions mentioned involve the three types of continuity as identified by Haggerty et al. (23): relational continuity, informational continuity, and coordination continuity. Additionally, continuity is only required when there are changes in time and settings, which serves as the foundation for these three types of continuity. Our multidisciplinary follow-up pattern primarily consists of five components: Management of the multidisciplinary team, objectives, and content, settings, follow-up pathways, and patient experience. According to the TIDier checklist, we provided a detailed description of the intervention measures for the multidisciplinary follow-up pattern (Supplementary material 1).

Patients in the control group received standard preoperative and postoperative nursing care following the established protocols for cervical spondylosis upon their admission to the hospital. One day before discharge, the responsible nurse and medical team leader provide the patient with pre-discharge health education, including postoperative skin and wound management, common discomfort symptoms after surgery, discharge instructions, postoperative rehabilitation exercise plan for cervical spondylosis, methods and duration of wearing a cervical collar, guidance on pain medication, hospital health consultation hotline, outpatient follow-up, and reexamination schedule, methods for appointment and registration, as well as distribution of postoperative health education materials. After discharge, the patient needs to make an appointment for postoperative follow-up visits for cervical spondylosis, at 1 month, 3 months, 6 months, and 1 year after the surgery. During the postoperative outpatient follow-up visits, the patient’s appointed physician will assess their health condition, provide medical advice based on the patient’s health needs, and offer corresponding health guidance.

The assessor (follow-up nurse), who had completed relevant professional courses, filled out a follow-up questionnaire. The assessor is aware of the distribution of patients in both groups. To ensure survey quality, a second researcher was invited to assist in the removal of invalid questionnaires. The primary outcome measure was neck pain-related disability, assessed using the NDI (0–100). Secondary outcome measures included the Visual Analog Scale (VAS), Self-Efficacy for Managing Chronic Disease 6-item Scale (SECD-6), and Study 12-Item Short-Form Health Survey (SF-12), which were used to assess neck pain, self-efficacy, and health-related quality of life. Additional measures included complications, patient satisfaction, and economic indicators. Patient characteristics, such as age, gender, marital status, qualifications, occupation, telephone number, length of stay, employment status, number of people living together, family income, body mass index (BMI), number of previous hospitalizations, admission patterns, cervical spondylosis classification, number of surgical segments, American Society of Anesthesiologists (ASA) physical status classification of I, II, or III, and convenience of medical treatment, were obtained from medical records and patient responses at baseline (the day of discharge). The NDI, VAS, SECD-6, and SF-12 were collected on the day of discharge, 24–72 h, 1 month, and 3 months after discharge. Patient satisfaction, complications, and economic indicators were recorded at 3 months after discharge. Data on the day of discharge for both groups of patients were collected in the ward. Data for the intervention group after discharge were collected through the Cervical Spondylosis Surgery Follow-up Platform, while data for the control group after discharge were collected either in the outpatient department or through telephone follow-up.

The NDI (24) is a 10-item questionnaire that measures the impact of neck symptoms on functional activities, with each item scored from 0 to 5, a higher score indicates a higher disability. The NDI covers pain, personal care, lifting, reading, headaches, concentration, work, driving, sleeping, and recreation. The NDI had acceptable responsiveness and construct validity to assess self-perceived disability (25).

The VAS (26) is a 100 mm horizontal line used to measure neck pain. The left end of the scale represents “no pain” and the right end represents the “most severe pain imaginable.” Patients mark the location that best represents their pain intensity based on their own condition and degree of pain.

Six items were included in the SECD-6 (27) to help assess how confident patients are in doing certain activities. The scale ranges from 1 (not at all confident) to 10 (totally confident). The score for the scale is the mean of the 6 items, and high scores indicate high self-efficacy.

The SF-12 (28), a measure of Health-related Quality of Life, is a shortened version of the SF-36 with 12 items, covering eight dimensions–general health (GH), physical functioning (PF), role-physical (RP), bodily pain (BP), vitality, social functioning (SF), role-emotional (RE), and mental health (MH). The eight dimensions can be consolidated into two summary scores using scoring algorithms, a physical component score (PCS) and a mental component score (MCS), ranging from 0 to 100, with higher scores representing a better health-related quality of life.

The follow-up satisfaction questionnaire consists of 10 items, covering satisfaction with the follow-up staff, health education, rehabilitation guidance, medical services, outpatient consultation, appointment registration process, auxiliary examination process, physical health, mental health, and costs related to cervical spondylosis after discharge. Responses are graded from 1 (not at all satisfied) to 5 (very satisfied), with higher scores representing higher satisfaction.

Economic indicators encompass direct medical costs, direct non-medical costs, indirect costs, and total costs. The estimation of direct medical costs and direct non-medical costs is conducted using the micro-costing method (29). These cost data are derived from hospital systems and patient interviews. Direct medical costs consist of post-discharge registration fees, medication expenses, treatment fees, examination fees, costs related to the treatment of complications, and other medical expenditures directly associated with the disease. Direct non-medical costs comprise transportation and accommodation expenses incurred by patients during medical visits. Total transportation costs are calculated based on patients’ reported round-trip transportation expenses and the number of visits made after discharge. Accommodation costs are estimated based on patients’ self-reported expenditures. Indirect costs pertain to income loss following cervical spine surgery and are estimated based on the patient’s lost work time post-discharge, converted into costs using daily wages as a basis. For employed individuals or those with fixed monthly incomes, indirect costs are represented by the total hours or days of productivity lost (seeking care and productivity time lost) multiplied by the average hourly or daily earnings. Total costs encompass both direct and indirect costs.

Descriptive statistics, mean and standard deviation (SD) for normally distributed continuous variables, median and interquartile range (IQ) for non-normally distributed continuous variables, and absolute numbers and percentages for categorical variables were calculated to summarize the data. Since the data included repeated measurements and did not follow a normal distribution, we used Generalized Estimating Equations (GEE) to examine the effects of interventions on disability, pain intensity, self-efficacy, and health-related quality of life. In our analysis, we considered the intervention group (contract follow-up vs. routine follow-up), visit number (as a categorical variable), and the interaction between the intervention group and visit number as independent variables. We used robust standard errors and specified an exchangeable working correlation structure. We utilized all available time-point data. Differences between medians and their 95% confidence intervals were calculated using the Hodges-Lehmann estimator. For normally distributed categorical variables, we performed chi-square tests. For cases suitable for Fisher’s exact test, we used Fisher’s exact test. For normally distributed continuous variables, we conducted independent samples t-tests. For non-normally distributed continuous variables and ordered ordinal data, we used Mann–Whitney U tests. Results were considered significant at a 5% level of significance (p < 0.05). Data analysis was performed using licensed STATA 16 software, and graphs were generated using licensed GraphPad Prism 9.0.

The comparison between groups revealed a significant difference in the NDI scores after discharge, specifically at 24 to 72 h, 1 month, and 3 months (p < 0.001), with the intervention group scoring notably lower than the control group. In addition, the intervention group exhibited sustained improvement in NDI scores at 24 to 72 h, 1 month, and 3 months after discharge according to within-group comparisons. However, there was no statistically significant difference in NDI scores among the control group during the first 24 to 72 h after discharge compared to initial baseline (Figure 2; Table 2).

In this study, we observed that patients in intervention group had significant improvements in various secondary outcome measures, including SECD-6, VAS, PCS, and MCS, at the 1-month mark (p < 0.05, Table 2). Furthermore, at the 3-month mark, the intervention group exhibited a greater increase in SECD-6, PCS, and MCS scores compared to the control group, demonstrating statistical significance (p < 0.001). However, no significant differences were observed between the two groups within the 24-to-72-h period. Additionally, the VAS scores showed a statistically significant difference between the two groups at both the 24–72 h and 1-month time points (p < 0.05).

The findings demonstrate that the intervention group (mean 42.52, SD 8.2) is significantly more effective in meeting the needs of patients compared to control group (mean 41.56, SD 7.34) (p < 0.001, Table 3). Moreover, when assessing various aspects of the follow-up satisfaction questionnaire, patients in the intervention group consistently expressed significantly higher levels of satisfaction (p < 0.001, Table 3).

During the study, a total of 11.4% of patients in the intervention group experienced complications, while 40.9% were in the control group (p < 0.01; Supplementary Table S1). Importantly, it is worth noting that no patients discontinued the study due to complications. When patients experience complications, they have been duly reported to the ethics committee. These complications are recognized as being caused by the patient’s underlying condition, rather than the intervention measures.

The findings of this study revealed significant differences in direct medical costs between the intervention group and the control group. The median direct medical cost in the intervention group was 1821.2 yuan, whereas it was 2421.7 yuan in the control group (p < 0.001; Supplementary Table S2; Figure 2). Furthermore, notable distinctions were observed in the median direct non-medical costs. The intervention group had a median direct non-medical cost of 110 yuan, whereas the control group had a median cost of 270 yuan (p = 0.035). Regarding indirect costs, no significant differences were found between the intervention group (median cost of RMB 4750) and the control group (median cost of RMB 5000) (p = 0.59). Moreover, the total cost of the intervention was significantly lower in the intervention group (7621.2 yuan) compared to the control group (8725.2 yuan) (p = 0.04) (Figure 3).

There are several limitations associated with this non-randomized control study design. Firstly, due to the requirement of voluntary patient consent in contractual terms, a randomized study design and blinding was not feasible. Non-randomized controlled designs may suffer from selection bias, raising doubts about the reliability and validity of study findings. To mitigate potential biases arising from non-randomization, this study rigorously enforced inclusion criteria for participants and ensured that both groups of patients originated from the same medical group. The lack of blinding may cause researchers to exhibit heightened concern for the patients, potentially impacting the objectivity of the observed outcomes. Secondly, our follow-up period was relatively short, lasting only 3 months after discharge. A short follow-up time may result in insufficient assessment of patients’ post-discharge conditions and intervention effects. Future studies should incorporate long-term follow-ups to ascertain the long-lasting benefits of multidisciplinary contractual continuing care. Thirdly, this study was conducted at a single center, and the findings exclusively reflect the performance of that specific center. To enhance external validity, future research endeavors could encompass various geographical regions and diverse tiers of hospitals. Finally, our study protocol has not been previously published, which may have an impact on the credibility and transparency of the study. However, we have included a detailed research protocol (Supplementary material 1) in this study to facilitate reproducibility by other researchers.