Introduction

Orthopedic Trauma refers to injuries to the musculoskeletal system caused by external forces, including fractures, dislocations, ligament/tendon tears, and soft tissue damage1. Trauma patients have high rates of death and disability because of their severe and complex conditions. The World Health Organization reports that trauma causes more than 5.8 million deaths globally each year2. From 2008 to 2017, trauma accounted for 9.44% of the all-cause mortality in Anhui Province, ranked as the fifth leading cause of death in the whole population3.

Nursing service quality refers to the nursing staff’s proficiency in delivering vital life services and technical nursing care to patients as well as the degree to which the staff meet the needs of patients. This encompasses both technical quality attributes and service quality attributes4. Monitoring nursing service quality is very important for new hospitals, which is not only the key to ensuring patient safety and satisfaction, but also an important measure to standardize the process and promote continuous improvement. Orthopaedic trauma mainly includes limb injury and spinal fractures. And the patients have severe conditions, rapid clinical deterioration, extended convalescence periods, and limited activity5,6,7. So high-quality care is essential for these patients to achieve good effects of treatment and rehabilitation.

From the perspective of health service providers, the important indicators of the quality of medical services are the incidence of adverse events, length of hospital stay, etc.8. However, “patient participation” has been clearly listed as a key goal by the Global Action Plan for Patient Safety (2021–2030), and the slogan "Nothing About Me Without Me" was put forward. Nursing professionals should try their best to meet the demands of patients9. So identifying the key points of nursing quality that need to be improved from the perspective of patients deserves nursing managers’ attention10.

To date, most studies evaluated the nursing service quality for traumatic orthopedic patients by assessing patient satisfaction11 and service needs12. However, they all unilaterally know the perception or needs of patients, it can’t simultaneously tell us what needs to be improved and which to improve first. Therefore, it has limited significance for clinical practice.

The SERVQUAL model and the Kano model are two representative methods in the field of quality management. These two models are complementary in terms of quality feedback and control13. The SERVQUAL model focuses on late control to test service capabilities, and the Kano model is mainly used in product development and focuses on early control. Both models are based on patient perception and are closer to patient needs. This study is the first to combine the application of the SERVQUAL model and the Kano model in the quality management of nursing care for orthopaedic trauma patients. The SERVQUAL model was used to screen for weaknesses in nursing services from the perspective of patients. The Kano model combined with factor selection line and sensitivity were used to clarify the specific order of nursing quality improvement after preliminary screening. The joint application of these two models provides new ideas for exploring patients ’ views on nursing quality.

Methods

Design and medical ethics

This study used a cross-sectional design and was reported according to the EQUATOR STROBE checklist14 (see Related file 1). This study was conducted in accordance with the Declaration of Helsinki and relevant national laws, and in accordance with the protocol approved by the Ethics Committee of Hefei BOE Hospital to protect the health and rights of the subjects. This study passed ethical review by the Hospital Ethics Committee (NO. 2021008, see Related file 2). Informed consent was obtained from all participants in this study, participation was voluntary, and participants could withdraw at any time without repercussions.

Study setting and participants

In this study, convenience sampling was used to select trauma patients who were hospitalized in the Department of Orthopaedics of a hospital in Hefei, South China. The hospital is a prominent new tertiary comprehensive private hospital that integrates emergency treatment, medical treatment, teaching, scientific research, health management and intelligent health care. The hospital commenced full operation in March 2019 and passed JCI international medical certification in November 2021. The hospital has 1,000 open beds and aims to provide high-quality. The Department of Orthopaedics is the key development discipline of the hospital and receives, with the clinical team encompassing trauma, spine, microsurgery, sports medicine, and joint specialties. The ward admitted both emergency and outpatient patients, approximately 1,080 patients each year, of which emergency patients accounted for about half, and two-thirds of them were orthopaedic trauma patients. The age of trauma patients ranged from 18 to 91 years, with a mean age of 49 years. The male-to-female ratio was approximately 2.2:1. The average length of hospital stay (LOS) varied by injury type: 10 days for extremity trauma, 10.5 days for joint injuries, 7 days for hand and foot injuries, and 12 days for lumbar spine injuries. During the questionnaire collection period, the doctor‒nurse ratio was 17:15, the bed-nurse ratio was 4.85:1, and the nurse‒patient ratio was 1:7.31.

As mentioned above, the investigation was conducted in two phases. The inclusion criteria for the two surveys were as follows. Subjects enrolled at admission ① met the clinical diagnostic criteria for orthopaedic trauma15; ②were ≥ 18 years old; ③ were in stable condition; ④ had the ability to understand, express and communicate; ⑤ provided informed consent;⑥ had an orthopaedic hospitalization certificate. Subjects enrolled before discharge ① had received the first questionnaire survey; ②were ready for discharge; ③ were hospitalized ≥ 3 days and received a full experience of care. Patients with psychiatric disorders were excluded from both rounds of the study.

The researcher estimated the minimum sample size required for this study based on the sample size formula. The specific formula is n = \({\left({\mu }_{\alpha /2}\sigma \right)}^{2}/{\delta }^{2}\),where α = 0.05, Z = 1.96, and d = 0.3. According to a cross-sectional study by LanYang16. SD was estimated to be 2.43,and considering a 20% sample attrition rate, it was estimated that at least 300.

SERVQUAL model

The SERVQUAL model, proposed by Parasuraman, serves as a tool for evaluating service quality from the perspective of customers17. As shown in Table 1, the model categorizes the factors that affect service quality into five dimensions: empathy, tangibility, assurance, responsiveness and reliability. The SERVQUAL model analyses the differences between the actual perceptions and expectations of customers with regard to service quality in these five dimensions. Expected service quality refers to the user’s psychological expectation of service quality before experiencing the service18. Perceived service quality refers to the actual level of service quality experienced by users in the process of receiving services18. Therefore, two rounds of surveys are required when applying the SERVQUAL model to measure service quality. When the difference between the perceived and expected service quality is negative, it is considered to indicate a low-level service item18.

Table 1 Dimensions and implications.
Full size table

The SERVQUAL model has been widely used in the evaluation of service quality in the fields of logistics, transportation, retail, library management, medical health and other service management domains19, and it is the most widely used and most typical and effective method for evaluating service quality20. In the field of nursing, the model has also been applied in the evaluation of nursing service quality in internal medicine, surgery, outpatient departments, emergency departments, intensive care units and other departments and has shown good reliability and validity21. Liao et al22.utilized the SERVQUAL model to investigate the nursing service quality in orthopaedic wards. The results confirmed that the SERVQUAL model had good reliability and validity, highlighting its efficacy in evaluating the nursing service quality in orthopaedic departments.

Kano model

In the 1980s, Professor Noriaki Kano of Tokyo University of Technology, Japan, was inspired by the "two-factor theory" and introduced the Kano model, which aimed to accurately identify and prioritize the service attributes required by customers23. According to the nonlinear relationship between customer satisfaction and demand fulfilment, the Kano model divides demand service attributes into five categories, must-be attributes (M), one-dimensional attributes (O), attractive attributes (A), indifferent attributes (I) and reverse attributes (R), as shown in online Supplementary Fig. 1.

Must-be attributes are requirements that must be satisfied23. The degree of satisfaction with one-dimensional attributes is directly proportional to customer satisfaction23. Attractive attributes are unexpected services whose satisfaction will surprise customers23. Indifferent attributes have no effect on customer satisfaction23. Reverse attributes refer to services that reduce customer satisfaction23. The Kano model approaches customer expectation attributes by developing attractive quality characteristics and has been applied in various research fields, such as quality management, product design and customer demand detection24.

Instrument

General information questionnaire

The self-designed questionnaire included gender, age, ethnicity, occupation, education level, religious belief, trauma hospital was our hospital, and the primary diagnosis at admission , etc.(see Supplementary file 1).

The evaluation questionnaire of nursing service quality for orthopaedic trauma inpatients

The survey consisted of two parts, the expectation questionnaire (see Supplementary file 1) and the perception questionnaire (see Supplementary file 2), on the nursing service quality for orthopaedic trauma inpatients. The expectation questionnaire was designed according to the following steps. First, taking the five dimensions of the SERVQUAL model as the theoretical framework, the measurement item were designed according to the characteristics of orthopaedic trauma patients, and the initial item version of the expectation questionnaire was developed after discussion among the research group. Second, a clinical treatment expert in emergency trauma orthopaedics, 2 clinical nursing experts and 2 nursing management experts were invited to evaluate the content validity of the questionnaire. After two rounds of expert consultation, a formal questionnaire was developed. The expert authority coefficient was 0.91, and the content validity index of each item was 0.8-1. The questionnaire included 5 dimensions and 26 items,The perception questionnaire was developed according to the expectation questionnaire. The two questionnaires had the same measurement items and used the standard questioning method of the SERVQUAL model. All questions were scored on a 5-point Likert scale ranging from 1 point (very dissatisfied) to 5 points (very satisfied). For example, for the item "Orthopaedic nurses introduce themselves to you and your family members", the item in the expectation questionnaire was"Should orthopedic nurses introduce themselves to you and your family?", while in the perception questionnaire, it was "Orthopedic nurses could introduce themselves to you and your family members?". Subsequently, a presurvey was conducted with 50 people. The patients who were surveyed did not give valid feedback when assessing face validity. The results of the reliability test showed that the overall Cronbach’s α coefficients for the expectation questionnaire and the perception questionnaire were 0.860 and 0.869, respectively. The Cronbach’s α coefficients for each dimension of the expectation questionnaire and the perception questionnaire were 0.744-0.830 and 0.670-0.909, Among them, The smallest Cronbach’s α of the specific reactive dimension in the perception questionnaire was 0.670. It might have been related to the small sample size (n=50) and the fact that there were only three questions in the response dimension in the pre-experiment, however, the overall Cronbach’s α of the perception questionnaire was 0.869, so in the exploration stage, the Cronbach’s α of the specific dimension in the questionnaire was acceptable at least 0.67025. Finally, to test the construct validity of the questionnaire, 150 orthopaedic trauma inpatients were selected for exploratory factor analysis. The Kaiser–Meyer–Olkin (KMO) test was used to compare the simple correlation coefficient and partial correlation coefficient between variables26. The results showed that the KMO values of the expectation questionnaire and the perception questionnaire were 0.780 and 0.745, respectively, and the results of Bartlett’s test of sphericity were all less than 0.001. Five common factors were extracted from the expectation questionnaire and the perception questionnaire, and the cumulative contribution rates to the total variance were 61.933 and 55.133, respectively. Because the loads of items 11, 19 and 24 of the expectation questionnaire were greater than 0.5 for two common factors, item 11 was placed in the "Reliability" dimension, item 19 was placed in the "Assurance" dimension, and item 24 was placed in the "Empathy" dimension, considering the clinical significance and large factor load27. Finally, the factor loading ranged between 0.415 and 0.906, which indicated that the questionnaire had good validity.

Kano model questionnaire of nursing service qualityfor orthopaedic trauma inpatients

The Kano model classifies the needs of customers according to their satisfaction, which is analysed with a pair of performance (positive) and non-performance (negative) questions28.

The researcher administered both positive and negative questions to each subject who submitted an expectations questionnaire to form the Kano Model Questionnaire for the Quality of Trauma Orthopaedic Inpatient Care Services (hereafter referred to as the Kano model Questionnaire; see Supplementary file 2). For example, for the item "Orthopaedic nurses introduce themselves to you and your family members", the positive question was "How would you feel if the orthopaedic nurses introduced themselves to you and your family members? ", and the negative question was "How would you feel if the orthopaedic nurses did not introduce themselves to you and your family members?" The questions were scored on a 5-point Likert scale ranging from 1 point (very dissatisfied) to 5 points (very satisfied). The Kano model questionnaire included a total of 26 items. The overall Cronbach’s α coefficient for the positive part of the Kano model questionnaire was 0.891, and the dimensions ranged from 0.680 to 0.838. The overall Cronbach’s α coefficient for the negative part was 0.848, and the dimensions ranged from 0.649 to 0.848.

Data collection

Questionnaires were collected from January to November 2022. The investigators included an orthopaedic nurse and two trauma physicians. Firstly, general information and expectations questionnaires were administered by 2 trauma surgeons to patient in the emergency department and orthopedic outpatient department while they were waiting for nucleic acid results before admission. Secondly, on the day of discharge, nurses distributed perception questionnaires and Kano model questionnaires to patients in the orthopedic ward. Before distribution, the investigators explained the purpose and method of filling out the questionnaires.The criteria for invalid questionnaires were as follows:①incomplete information;②patterned or contradictory responses;③duplicate submissions. Following on-site verification, the questionnaires were collcted in two phases, yielding a final sample of 300 valid responses with an effective responserate of 89.82%. The Expectation Questionnaire scores (M = 22.59, SD = 1.62) were significantly higher than those of the Perception Questionnaire (M = 19.73,SD = 1.85), P < 0.05,with a statistical power of 1.0, indicating a large effect size (Cohen’s d = 1.75,95% CI [1.52,1.98]). The general characteristics of the patients are shown in online Supplementary Table 1.

Data analysis

EpiData 3.1 was used for data entry, and SPSS 24.0 software was used for statistical analysis. The general data of the respondents were analysed by descriptive statistics, and the count data were expressed as frequencies and constituent ratios. The scores for each dimension and specific item of the Nursing Service Quality Questionnaire are presented as the mean and standard deviation. Because the low-quality items were all negative, only the mean was used for description. Paired sample t tests were used to compare the expected and perceived differences in service quality. The level of significance was P < 0.05. Single factor and multiple linear regression were used to analyze the influence of each factor on the quality of nursing service.

Service quality

The difference between the perceived questionnaire score (P) and the expected questionnaire score € represented the quality of service (SQ=P-E). A negative service quality indicated that there was a service quality gap.

Requirements classification

The Kano attribute classification of each need was determined based on the responses of the study subjects to the positive and negative questions. There were 25 possible outcomes for each requirement, each corresponding to a Kano attribute, as shown in Table 2. The frequency of each demand on the Kano quality characteristic distribution was counted, and the quality attribute with the highest frequency was taken as the final quality attribute of the demand item23. "A" stands for an attractive attribute, "M" for a must-be attribute, "O" for a one-dimensional attribute, "I" for an indifferent attribute, "R" for a reverse attribute, and "Q" for a suspicious problem.

Table 2 Attribute classification methods of the Kano model.
Full size table

The traditional Kano classification method cannot effectively distinguish attributes with similar or equal frequencies, so it needs to be combined with the better-worse method26. In the better-worse method, the satisfaction coefficient (SI) is also called the better coefficient. Its value is usually a positive number representing the impact of providing a certain demand on the improvement of user satisfaction. The closer the value is to 1, the greater the impact of requirements on user satisfaction. The satisfaction coefficient was calculated using the formula SI= (A+O)/(A+O+M+I). The dissatisfaction coefficient (DSI) is also called the worse coefficient. Its value is usually negative, representing the impact of not providing a certain demand on the decrease in user satisfaction. The closer its value is to -1, the greater the impact of the demand on the decrease in user satisfaction. The formula DSI=-1*(M+O)/(A+O+M+I) is used to calculate the coefficient of dissatisfaction. The matrix graph was constructed with the absolute value of the SI as the abscissa and the absolute value of the DSI as the ordinate, and the mean values of the ordinate and ordinate were used to divide the matrix graph into four quadrants. Each of the four quadrants corresponds to four attributes, as shown in online Supplementary Fig. 2. The different needs of the |SI| and |DSI| values determinedthe position of each demand in the matrix graph, as shown in online Supplementary Fig. 2.

Screening elements for improving nursing service quality

As a two-dimensional model, the Kano model can classify demand attributes. According tothe quadrant in the matrix diagram, the trend of user satisfaction after demand is met can be roughly evaluated. However, demand trends alone cannot determine which demand needs to be addressed for improvement. Therefore, this study adopted the method of screening factors described by Liu29 and added the sensitivity "R" and the factor selection line to representhe degree of demand improvement on the basis of the better-worse coefficient. As shown in online Supplementary Fig. 2, the distance ST (ST = \(\sqrt {\overline{SI} *\overline{SI} + \overline{DSI} *\overline{DSI} }\)) from the origin to the centre of the matrix graph was the criterion for the degree of sensitivity. The improved elements weredistinguished from the other elements by the element selection line (a quarter arc of radius ST). Sensitivity "R" (R = \(\sqrt {SI*SI + DSI{*}DSI} - ST\)) was used to represent the degree of improvement of the element, that is, the distance between the point on the right side of the element selection line and the line. The improvement elements were sorted according to sensitivity. On the right of the element selection line are the elements that need to be improved. The greater the R value of the element is, the more sensitive the patient experience, which should be improved. The factors on the left side were less sensitive to patient experience and cannot be considered.

Results

Dimension scores and total scores of the expectation and perception questionnaire

The total score and each dimension score of perceived quality were significantly lower than the expected quality score (Table 3). Among the five dimensions of nursing service quality, the largest gap was related to the dimension of reliability, while the smallest gap was related to the dimension of tangibility.

Table 3 The total score and dimension score of nursing service quality
Full size table

Item scores of the expectation and perception questionnaire

Among the 26 nursing service items, 21 items were negative and 5 items were positive (Table 4). According to the paired sample t-test results, except for the 23rd and 26th items in the empathy dimension, the scores of the other items were significantly different in terms of perceived value and expected value.

Table 4 The comparisons of item score of expectation questionnaire and perception questionnaire.
Full size table

Classification of attributes of negative nursing service quality entries by the Kano model

The 21 nursing service items with negative scores were classified by the Kano model (Table 5).The results showed that there were 1 must-be need, 5 one-dimensional attribute needs, 12 attractive needs and 3 indifferent needs. All items were ranked from high to low according to the sensitivity "R".

Table 5 Classification and sensitivity of patient needs improvement items.
Full size table

Priority items for improvement

The study revealed that the factor selection line was a quarter arc with a radius of 0.859, and 10 items appeared on the right side of the factor selection line ( online Supplementary Fig. 3). According to the descending order of sensitivity "R" in Table 5, the specific improvement priorities of these 10 items were as follows: 14, 9, 13, 10, 12, 5, 22, 24, 8, and 16. The top three items were emergency response ability, precision nursing and professional practice ability.

The influencing factors of nursing service quality for orthopedic trauma patients

Single factor analysis of variance showed that gender, marital status, per capita monthly household income, first admission to our hospital, and the primary diagnosis of this admission were the influencing factors of the total score of nursing service quality(online Supplementary Table 2). Multiple linear regression analysis showed that gender, per capita monthly household income, whether it was the first time to be admitted to the hospital, hand and foot injury, and spinal fracture had statistical significance on the score of nursing service quality(online Supplementary Table 3). Among them, the primary diagnosis of this admission had the most significant impact on the score of nursing service quality.

Discussion

The strengths of nursing care in the eyes of patients

There were some advantages pertained to tangible aspects of nursing service from the perspective of patients, including maintaining clean clinical environment and neat professional attire, and proactive self-introduction. These reflected that the nurses had strong professional identity and self-confidence , which could made a good first impression on the patients30.

Causal analysis of low-level nursing service quality in each dimension

The results showed that the perceived nursing service quality was lower than the expected in the eyes of patients, covering five dimensions. The score gap ranged from the highest to the lowest was reliability (-0.78), assurance (-0.63), empathy (-0.56), responsiveness (-0.51), and tangibility (-0.34). So the dimension of reliability was the most in need of enhancement.

Reliability

This dimension contains a total of 6 items, all of which were low-level nursing items. The key items need to be improved were item 9 (provide safe and accurate nursing), item 10 (ensure informed consent), item 12 (have professional knowledge and rich experience), item 13 (have professional nursing practice ability for preoperative preparation and postoperative rehabilitation), and item 14 (have the ability to deal with emergencies and disease changes). All of the items were associated with one-dimensional attributes, because they were directly related to the treatment and rehabilitation of patients and were positively correlated with satisfaction. The possible reasons for the low scores of these items were as follows. First, all the patients were divided into five subspecialties in the orthopedic ward, where the ratio of doctors and nurses was 19:15, and part of the nurses were transferred nurses. Especially undering the COVID-19 epidemic, the hospital could not allocate enough experienced nurses and provide standardized training for transferred nurses, such as professional knowledge and skills, ethic rules, and communication skills in a short time.

Assurance

The low-level nursing service items included items 18–21. In which, item 18(proper communication to obtain their understanding) was a must-be attribute, item 19(being able to answer patiently) was an attractive attribute. During the COVID-19 pandemic, nursing staff were required to frequently inform patients and their families to comply with hospital regulations, such as reducing accompanying and collecting nucleic acid on time31. So most of the patients felt bored and intent to break rules, which was easy to result in the conflicts with the nurses, and were probably not satisfied with nursing staff.

Empathy

The low-level nursing service items in this dimension included items 22, 24 and 25. Items 22 (pain relief) and 24 (protect privacy) were one-dimensional attributes, which need to be focused on. Pain was difficult for trauma patients to endure. To relieve pain,nurses should be able to assess and manage pain dynamically. Clinical nurse specialists for pain may be competent. However, the hospital did not. In addition, patients with hand and foot injury, and spinal fracture were mainly bedridden, nurses in the ward need to observe skin changes in the sacrococcygeal region of patients in bed. However, in the study, male patients accounted for 68.70%, while most of the nurses were female. In China, patients will feel embarrassed when private parts are observed by nurses of different genders32. So, failure to meet the humanistic care needs of patients will directly reduce their satisfaction with the services.

Responsiveness

In this dimension, item 16 (skilled nursing operation standard) and item 17 (rapid entry and exit process) needed to be focused on. During COVID-19 epidemic, emergency patients ’ waiting time before admission was prolonged due to nucleic acid testing, therefore, patients compensated for the expectation that ward nurses could deal with it in time, and considered that skilled operation of nursing staff was the basic requirement, and patients had difficulty accepting this deficiency.

Tangibility

The low-level nursing service items included item 4 (convenient service facilities), item 5 (good emergency escape and safety facilities), item 6 (eye-catching service logo), and item 8 (reasonable department layout). Item 5 and 8 were one-dimensional attributes. Firstly, in the hospital, emergency escape signs and safety facilities were mostly distributed in corridors and stairwells, it were less likely to be seen by trauma patients who mainly moved within wards. Second, the average length of stay for trauma patients was long, and the ward did not have a dry area, but Chinese people preferred the sun to dryers. So patients’sense of safety and comfort could not be ensured, which directly led to a decrease in patient satisfaction.

Optimization strategy for ensuring the nursing service quality for orthopaedic trauma inpatients

According to the results of the selected priority improvement projects, the items that need to be prioritized for improvement were closely related to patients ’ sense of security and comfort during hospitalization. This underlines the significance of safe and comfortable diagnosis and treatment services, which were the main aspects of patient concern. Based on the results of our study, we propose the following service quality improvement strategies.

First, a training system should be established for orthopaedic nurses to improve their competency. The nursing department formulates the nursing training programme, defines the training content and objectives needed to be completed by the department around the core system, and strengthens the supervision and management of the daily training quality of nurses at different levels. The key training objects are newly recruited nurses without orthopaedic work experience33. The proposed training system includes five modules: theoretical knowledge of orthopaedics, professional skills in orthopaedics, emergency response, interpersonal communication and humanistic care, and health education. The training period is at least one year34. According to the shortcomings of specialties in the department, nurses are sent to study regularly to enrich their vision.To ensure the competency of nursing staff, managers need relatively fixed manpower and a reduction in frequent post-transfers.

Second, the hospital needs to optimize the nursing service process. Standardized operating procedures (SOP)35 will be formulated to clarify all aspects of nursing services, such as patient admission assessment, nursing plan formulation, implementation, and record, so as to ensure that the nursing staff can operate in accordance with the unified standard. Hospitals should improve the professional operation ability of nursing staff through process optimization.

Third, strengthen informatization and technology upgrading. Hospitals provide patient-side apps or online platforms, so patients can view their test results and medication information at any time. They also provide patients with disease prevention, rehabilitation guidance, and other information. Finally, the use of the Pneumatic Tube System (PTS)36was strengthened to facilitate specimen examination and drug delivery so as to timely administer drugs to relieve pain and improve patient safety and trust.

The last but not least, consistent evaluation of orthopaedic nursing service quality coupled with the dynamic adjustment of improvement strategies is vital. The purpose of evaluating the nursing service quality is to better serve patients. Individual differences in patients, hospital hardware conditions, and the ability of nursing staff have an impact on the nursing service quality. Therefore, when adhering to the "all patient-centred" service tenet, nursing managers regularly evaluate the nursing service quality and dynamically adjust improvement strategies according to the results. This approach ensures alignment with patients ’ expectations and the efficacy of nursing care.

Limitations

First, the evaluation of service quality in this study considered only the evaluations of patients and did not solicit input from hospital managers or frontline nurses. Future research could conduct qualitative research on the basis of this quantitative research and explore nursing service quality improvement strategies from the perspective of service providers. Second, the Kano model cannot reflect the dynamic changes in patients’ needs. Third, under the influence of harmonious culture, patients may pay more attention to “harmonious relationship” than objective feedback, which leads to the deviation of evaluation results from reality. Finally, this study was conducted in a newly established private general hospital during the COVID-19 pandemic using convenience sampling. Therefore, the findings may have limited applicability to other hospital settings in China.

Conclusions

The quality of care provided to orthopaedic trauma inpatients is average. There is a negative gap between patients ’ actual perceptions and expectations of five aspects of nursing services. Using the Kano model and the factor screening method, we sing service quality in orthopaedic trauma wards.