Development and validation of VTE prevention beliefs questionnaire for neurosurgical hospitalized patients based on HBM: a cross-sectional study Authors: Pan Lin Published: 2024 ABSTRACT AbstractPatients hospitalized for neurosurgical procedures are highly susceptible to venous thromboembolism (VTE). It is crucial to evaluate the prevention beliefs of VTE among neurosurgical inpatients and analyse the influencing factors. This research sought to create an instrument grounded in the Health Belief Model (HBM) to evaluate the beliefs about VTE prevention and the factors influencing these beliefs among neurosurgical inpatients. A questionnaire comprising 28 questions was designed, validated, and distributed offline to 386 neurosurgical inpatients through random sampling. The data analysis was conducted using SPSS 26.0 and Amos 27.0. The measurement model was assessed through reliability, validity, and confirmatory factor analyses. Additionally, multiple linear regression analysis was utilized to identify the factors influencing VTE prevention beliefs. Statistical significance was determined by P values less than 0.05. The questionnaire encompassed six constructs, all of which exhibited strong internal consistency along with solid convergent and discriminant validity. VTE prevention beliefs score among neurosurgical inpatients was 99.56 (SD = 10.11). Among the belief dimensions, the scoring rate of perceived benefits and cues to action ranked highest, while perceived barriers were the lowest. The influencing factors of VTE prevention beliefs were educational attainment (β = 0.343, P < 0.001), receipt of VTE knowledge education from healthcare providers (β = 0.214, P < 0.001), Personal history of VTE (β = 0.175, P < 0.001), surgical history (β = 0.116, P < 0.01), sources of VTE-related information (β = 0.113, P < 0.01) and method of paying medical expenses (β = 0.112, P < 0.05). The questionnaire showed strong reliability and validity, and the results indicating that neurosurgical inpatients hold moderate beliefs about VTE prevention. While they recognize its benefits, barriers to preventive behaviors persist. Addressing these barriers and enhancing self-efficacy through targeted education and interventions is key to improving VTE prevention awareness and actions.Keywords: Venous thormboembolism, Neurosurgical hospitalized patients, Belief, The Health Belief ModelSubject terms: Health care, Risk factors Introduction Among surgical patients, neurosurgical inpatients face the greatest risk of symptomatic venous thromboembolism (VTE)1. VTE, encompassing deep vein thrombosis (DVT) and pulmonary embolism (PE), is a covert yet lethal condition that presents a major public health challenge globally due to its detrimental effects on patient mortality and quality of life2. Patients who undergo craniotomy for brain tumors have a 30% likelihood of developing symptomatic DVT3, whereas the risk of PE in neurosurgical patients can reach 5%, with mortality rates reported between 9 and 50%4. Several factors elevate the risk of VTE in neurosurgical inpatients, with venous stasis due to paralysis and extended periods of coma being particularly significant5. Other factors, such as type of surgery, advanced age, tumour-induced hemostatic changes, and steroid usage, can also increase VTE risk6. Even after discharge, VTE incidence in neurosurgery patients can reach 50–66%7. Previous studies have shown that early VTE prevention can decrease DVT and PE incidence by 84% and 55%, respectively8. Despite the potential benefits of VTE prevention, the implementation rates of VTE prevention behaviours remain low among patients9,10. Therefore, understanding and improving psycho-social variables associated with patients’ VTE prevention beliefs can help health professionals design and implement appropriate VTE prevention interventions. Additionally, behavioral change interventions tend to be more effective when they target theory-based determinants11. The Health Belief Model (HBM), first developed in the 1950s, which emphasises threat perception and behavioural evaluation, is one of the most widespread conceptual frameworks for behavioural interventions12. In China, research has validated the applicability and effectiveness of the Health Belief Model (HBM) for assessing VTE prevention beliefs among orthopedic inpatients and postoperative patients13–15. However, no standardized instrument has been developed and validated to measure VTE prevention beliefs specifically in neurosurgical inpatients. To create a targeted VTE prevention plan for neurosurgical inpatients, it is crucial to gather data on their VTE prevention beliefs and the factors that influence these beliefs. This study is the first to develop and validate an instrument based on the HBM that specifically measures factors influencing VTE prevention beliefs in neurosurgical inpatients. Theoretical framework The HBM includes six constructs (Fig. 1)16: (1) perceived susceptibility reflects an individual’s assessment of their risk of developing a particular health issue. (2) Perceived severity pertains to one’s anticipation of the potential seriousness of the illness. (3) Perceived benefits indicate a belief in the advantages of engaging in health-promoting behaviors. (4) Perceived barriers represent an individual’s evaluation of the obstacles and costs associated with enacting health behaviors. (5) Cues to action are triggers, both internal and external, that can prompt health behaviors. (6) Self-efficacy refers to an individual’s confidence in their ability to perform a specific health behavior. According to this model, patients who feel vulnerable to VTE or perceive the condition as severe are more likely to engage in VTE prevention behaviors. Theoretical framework of Health Belief Model applied to VTE prevention behaviors. Methods Ethical approval All study materials and procedures were approved by the Institutional Review Board (IRB) of Xiang Ya Nursing School, Central South University under ID: CSU E202151. The research was conducted in accordance with the relevant guidelines and regulations, including the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants, ensuring their privacy and confirming their right to withdraw from the study at any time without consequence. Study design, population and sampling The study was conducted at Xiangya Hospital of Central South University from September 1 to November 6, 2021, using a cross-sectional design to gather data. Using two equations to calculate the sample size: (n = variables × (5 ~ 10) × (1 + 20%) and (n = tα/2+tβSδ2)17,18. The standard deviation (S) of VTE prevention beliefs in neurosurgical inpatients was 15.72, with an allowable error of 0.18S, and an anticipated non-response rate of 15%. Consequently, the first equation suggested a sample size of 168–336, while the second suggested 373. Ultimately, 386 neurosurgical inpatients were randomly selected for the final sample. The study included adults aged 19 and above who were hospitalized in neurosurgical units. Screening was conducted for patients in the 48-bed neurosurgical units based on the following inclusion criteria: admission to the neurosurgery department, age over 18 years, provision of informed consent, and absence of delirium or dementia as indicated in medical charts. Exclusion criteria comprised lack of language proficiency, coma, disturbed consciousness, receipt of palliative care, or terminal illness. Questionnaire design Step 1: literature reviewing Upon reviewing the literature, no validated VTE prevention tools based on the HBM specifically designed for neurosurgical inpatients were found. Therefore, this study created the questionnaire by adapting instruments used in similar contexts. Step 2: Delphi method The second step utilized the Delphi method for expert consultations, which took place between June and August 2021. A consultation questionnaire was developed for this purpose. A panel of 20 experts, including clinical nurses, nursing administrators, physicians, and academics were requested to provide feedback and suggestions on the questionnaire. The questionnaire’s content validity indexes (CVIs) were computed based on their assessments. After two rounds of correspondence and discussions, all experts agreed on minor modifications and the removal of certain items. Step 3: a pilot study and pre-survey A preliminary study was undertaken with a convenient sample of 10 neurosurgical inpatients to verify the questionnaire’s comprehensibility. The questionnaire was then revised and finalized based on expert panel discussions and feedback from voluntary patients. Following the pilot study, a pre-survey was carried out. The pre-survey sample size was established based on the needs of confirmatory factor analysis (CFA), using a subject-to-item ratio ranging from 1:5 to 1:1017. Consequently, the minimum sample size required was set at 200. Using convenience sampling, 210 neurosurgical inpatients from Xiangya Hospital CSU who met the criteria were ultimately included in the pre-survey. Table 1 shows the results of discriminant degree and correlation analyses, which were conducted based on the pre-survey data. Generally, an item critical ratio > 3.000 or P < 0.05 is considered acceptable; otherwise, deleting this item is considered19. An item was deemed poorly functioning if its correlation coefficient with the subscale score was less than 0.3020. Table 1Item-total correlation and internal consistency of Health Belief Model Scale for VTE prevention.Items of subscaleMeanSDScoring rate (%)Item critical ratio (t-value)Item-total correlationCronbach’s α (n = 210)Split-half reliabilityTest–retest intraclass correlation (n = 30)VTE prevention beliefs questionnaire (28 items)99.5610.1171.110.8940.7490.855**1. Perceived susceptibility (11 items)38.835.1570.60.8770.8250.778** SUS1 Immobility or bed rest for more than 3 days may increase the risk of VTE3.660.638− 7.359**0.607** SUS2 Trauma may increase the risk of VTE3.590.579− 8.411**0.623** SUS3 Surgery may increase the risk of VTE3.620.693− 7.316**0.596** SUS4 Obesity may increase the risk of VTE3.590.668− 7.124**0.607** SUS5 Use dehydrating drugs may increase the risk of VTE3.460.632− 6.101**0.620** SUS6 Deep vein catheterization may increase the risk of VTE3.460.616− 6.844**0.627** SUS7 Having high blood pressure or diabetes may increase the risk of VTE3.630.652− 9.284**0.638** SUS8 Infection may increase the risk of VTE3.510.637− 10.483**0.637** SUS9 Cancer/Chemotherapy/Radiotherapy may increase the risk of VTE3.550.667− 9.913**0.666** SUS10 Personal history of VTE may increase the risk of VTE3.670.663− 10.246**0.657** SUS11 I’m at risk for VTE3.080.885− 7.350**0.579**2. Perceived severity (4 items)14.822.9374.10.9200.9170.770** SEV1 I think VTE will affect my health3.720.768− 9.084**0.654** SEV2 I think VTE will cause me psychological stress3.650.838− 8.398**0.676** SEV3 I think VTE will affect my future work and life3.700.807− 10.099**0.718** SEV4 I think VTE will be a financial burden on my family3.750.816− 8.796**0.693**3. Perceived benefits (3 items)12.061.8880.40.9240.9280.830** BEN1 VTE prevention can reduce the occurrence of VTE3.990.674− 9.197**0.715** BEN2 VTE prevention can increase my sense of security4.010.674− 7.850**0.683** BEN3 VTE prevention is good for my health4.060.648− 8.408**0.697**4. Perceived barriers (4 items)10.352.1951.80.6390.5090.655** BAR1 Even with VTE prophylaxis, I could still get VTE2.580.683− 2.762*− 0.213** BAR2 I am concerned about the adverse effects of VTE prophylaxis (e.g. bleeding)2.510.673− 3.152*− 0.266** BAR3 VTE prophylaxis takes a lot of time2.610.782− 3.557*− 0.218** BAR4 The economic cost of VTE prophylaxis is high2.660.755− 1.996− 0.136*5. Self-efficacy (2 items)7.621.3076.20.5390.5770.515** EFF1 I can take the initiative to understand VTE related information3.600.859− 7.080**0.567** EFF2 I can cooperate with the medical staff to take preventive measures for VTE4.020.642− 8.481**0.617**6. Cues to action (4 items)15.862.2579.30.7740.6970.824** CUE1 The medical staff explained VTE knowledge to me, which makes me more focused on preventing VTE4.140.640− 9.804**0.682** CUE2 Others have experienced VTE, which makes me more focused on preventing VTE4.090.669− 10.120**0.644** CUE3 Obtaining VTE knowledge from books, magazines, TV networks, which makes me more focused on preventing VTE3.900.686− 5.822**0.563** CUE4 The current state of my body reminds me that VTE prevention is extremely necessary3.740.766− 6.654**0.525****P < 0.001 (two-sided test), *P < 0.01 (two-sided test). Item-total correlation and internal consistency of Health Belief Model Scale for VTE prevention. **P < 0.001 (two-sided test), *P < 0.01 (two-sided test). The results indicated that 27 out of 28 items (96.43%) had a critical ratio greater than 3.000 and P < 0.05, suggesting that most items in this questionnaire demonstrated a high degree of discrimination. Pearson correlation analysis revealed that the correlation coefficients ranged from -0.136 to 0.755, indicating a generally positive correlation; the Pearson correlation coefficients of items BAR1, BAR2, BAR3 and BAR4 were − 0.213, − 0.266, − 0.218 and − 0.136, respectively; however, the differences of all items were statistically significant (P < 0.01). Therefore, all questionnaire items were reserved to form a formal questionnaire. Before issuing the formal questionnaire, its reliability and validity were thoroughly tested. Questionnaire structure The survey was divided into two main sections. The first section comprised 17 questions that gathered demographic and socioeconomic data about the participants, such as age, education, residence, marital status, and income. The second section consisted of 28 questions in six dimensions measuring the components of HBM (Table 1), including (1) perceived susceptibility (SUS-11 items: e.g., “Prolonged immobility or bed rest exceeding three days can elevate the risk of VTE”); (2) perceived severity (SEV-4 items: e.g., ‘I think VTE will be a financial burden on my family’); (3) perceived benefits (BEN-3 items: e.g., ‘VTE prevention is good for my health’); (4) perceived barriers (BAR-4 items: e.g., ‘the economic cost of VTE prophylaxis is high’); (5) self-efficacy (EFF-2 items; e.g., ‘I can take the initiative to understand VTE related information’); and (6) cues to action (CUE-4 items: e.g., ‘The medical staff explained VTE related knowledge to me, which makes me more focused on preventing VTE’). Responses were assessed on a five-point Likert scale: respondents are asked to rate on a scale of 1 to 5, ranging from strongly disagree to strongly agree. In the perceived barriers dimension, scoring was reversed. Data collection Prior to the formal investigation, the investigator underwent standardized training to ensure consistency in the questionnaire distribution process. This training emphasized the importance of not influencing respondents’ answers through voice tone or facial expressions that might suggest a “correct” response. The investigators examined participants’ medical records to determine their diagnoses and length of hospital stay. Patients were assessed for eligibility based on inclusion criteria and excluded if they exhibited cognitive impairment, using data from medical records and evaluations by the unit nurses. The questionnaires were administered by the investigators in the privacy of the patients’ rooms. To mitigate privacy concerns and ensure data authenticity, patients completed the questionnaire anonymously. The survey took approximately 10–20 min to complete, and investigators were available throughout to assist with any questions and ensure consistent survey administration. Participants received toothpaste as an incentive. After collecting the completed questionnaires, invalid responses were excluded, including those completed in less than 300 s and those with inconsistent answers21. Data analyses Data analysis utilized SPSS V.26.0 and AMOS V.26.0. Reliability was evaluated using Cronbach’s alpha coefficients and split-half reliability. Stability was assessed through test–retest reliability analysis, employing intraclass correlation coefficients (ICC). Validity was assessed using Item Content Validity Index (I-CVI) and Scale Content Validity Index (S-CVI/Ave). Confirmatory factor analysis (CFA) was conducted to assess the consistency of the data and the structural model. Common indicators for evaluating the structural equation model included Chi-square statistics (χ2), degrees of freedom (df), normed chi-square (χ2/df), Incremental Fit Index (IFI), Goodness of Fit Index (GFI), Comparative Fit Index (CFI), Adjusted Goodness of Fit Index (AGFI), Root Mean Square Error of Approximation (RMSEA), and root mean square residual (RMR). According to established research, χ2/df should be < 3, GFI > 0.90, IFI > 0.90, TLI > 0.90, CFI > 0.90, RMSEA < 0.06 (less than 0.08 are generally acceptable), AGFI > 0.90, and RMR < 0.10 to indicate a good fit with the data22,23. These standards are widely accepted in statistics. Sociodemographic characteristics were presented as numbers and percentage distributions. The mean ± standard deviation and scoring rate were employed to characterize the scores related to VTE prevention beliefs. The scoring rate was determined by dividing the average score by the total possible score, then multiplying by 100%. Scores below 60%, between 60 and 79%, and above 80% were categorized as low, medium, and high levels, respectively13. Differences in VTE prevention belief scores among neurosurgical inpatients based on age, sex, education level, and economic status were analyzed and compared using t-tests, analysis of variance and Kruskal–Wallis H test. Multiple regression analyses were conducted to ascertain factors impacting VTE prevention beliefs among neurosurgical inpatients. A significance level of P < 0.05 was used to determine statistical significance. Table 2 details the coding of independent variables. Table 2Independent variable coding table.Independent variableVariableAssignment instructionsAge (years)X1The original valuePlace of residenceX2Rural areas = 0, Urban areas = 1Education LevelX3Setting dummy variables (Use ‘Primary and below’ as controls)OccupationX4Setting dummy variables (Use ‘other occupations’ as controls)Family per capita monthly income (US dollars)X5Less than 1500 = 0, 1501–3000 = 1, 3001–5000 = 2, more than 5000 = 3Payment manner of the medical expensesX6Setting dummy variables (Use ‘other manners’ as controls)SurgeryX7No = 0, Yes = 1CaregiversX8Setting dummy variables (Use ‘other caregivers’ as controls)Sources of information related to VTEX9Setting dummy variables (Use ‘never heard of VTE’ as controls)Whether there is a doctor/nurse in the familyX10No = 0, Yes = 1Personal history of VTEX11No = 0, Yes = 1Family history of VTEX12No = 0, Yes = 1Whether the health-care providers provided the VTE knowledge educationX13No = 0, Yes = 1 Independent variable coding table. Results Descriptive statistics of the sample The questionnaire was completed by 386 neurosurgical inpatients. Table 3 presents the demographic information of the patients. The participants had an average age of 51.45 years (SD = 13.65), with ages ranging from 19 to 86 years. Among them, 200 (51.8%) were female, and over a third identified as farmers (139, 36.0%). A total of 168 participants (43.5%) had completed either high school or college education, with intracranial tumors being the primary reason for admission in 232 cases (60.1%). Table 3Characteristics of Patients in neurosurgical hospitalized patients and univariate analysis (n = 386).FactorsPatients N (%)Beliefs score (x¯ ± s)t/FPFactorsPatients N (%)Beliefs score (x¯ ± s)t/FPAge (years)3.9380.002Reason for Admission0.7790.605 Less than 2515 (3.9)105.47 ± 13.37Craniocerebral injury17 (4.4)104.94 ± 14.42 26–3543 (11.1)107.74 ± 12.43Intracranial tumor232 (60.1)102.28 ± 11.96 36–4551 (13.2)103.65 ± 12.23Cerebrovascular disease72 (18.7)105.83 ± 11.66 46–55121 (31.3)102.86 ± 11.40Diseases of the scalp and skull5 (1.3)100.20 ± 10.40 56–6597 (25.1)103.58 ± 11.93Intracranial infection3 (0.8)104.00 ± 12.17 More than 6559 (15.3)97.46 ± 12.19Spinal cord diseases15 (3.9)102.67 ± 15.41Gender1.1440.253Functional disease35 (9.1)100.34 ± 13.14 Male186 (51.8)103.66 ± 12.21Other7 (1.8)106.71 ± 5.53 Female200 (48.2)102.32 ± 12.18Surgery3.4610.001Place of residence5.1540.000Yes195 (50.5)104.87 ± 12.70 Rural areas208 (53.9)100.17 ± 11.36No191 (49.5)101.02 ± 11.36 Urban areas178 (46.1)106.23 ± 12.36Length of hospitalization3.9110.271Education level76.858#0.000Less than 7 days272 (70.5)102.03 ± 11.75 Primary and below90 (23.2)94.57 ± 10.328–14 days69 (17.9)104.51 ± 13.92 Junior high school128 (33.2)102.54 ± 10.0015–21 days23 (6.0)107.00 ± 12.20 High school/Technical secondary school93 (24.1)105.31 ± 11.03More than 21 days22 (5.7)105.50 ± 10.87 College49 (12.7)110.86 ± 13.16Caregivers4.5350.004 Bachelor degree or above26 (6.7)110.85 ± 12.72Spouse213 (55.2)103.88 ± 12.47Marital status1.6570.176Child or son/daughter-in-law122 (31.6)100.11 ± 11.21 Married343 (88.9)102.83 ± 12.19Parents34 (8.8)107.12 ± 11.67 Spinsterhood32 (8.3)104.88 ± 12.81Other (Rotate care/hire carers)17 (4.4)92.00 ± 4.24 Divorce5 (1.3)108.20 ± 12.64Sources of information related to VTE75.921#0.000 Widowed6 (1.6)95.83 ± 6.18Doctors84 (21.8)108.55 ± 11.85Occupation39.247#0.000Nurses51 (13.2)108.86 ± 10.05 Farmer139 (36.0)99.35 ± 10.69Family member/friend40 (10.4)106.68 ± 11.92 Worker49 (12.7)103.78 ± 12.78Internet/TV47 (12.2)101.87 ± 11.29 Self-employed21 (5.4)101.14 ± 9.28other patient14 (3.6)101.00 ± 13.96 Professional34 (8.8)111.74 ± 12.17Pamphlet/poster18 (4.7)105.83 ± 8.20 Administrator16 (4.1)112.00 ± 13.69Professional books2 (0.5)103.50 ± 16.26 Retiree58 (15.0)103.67 ± 13.13Never heard of VTE130 (33.7)102.96 ± 12.20 Other69 (17.9)103.20 ± 11.21Whether there is a doctor/nurse in the family2.7680.006Family per capita monthly income (US dollars)29.093#0.000Yes35 (9.1)108.43 ± 10.26 Less than $23523 (6.0)96.13 ± 13.01No351 (90.9)102.42 ± 12.25 $235—$470133 (34.5)99.69 ± 10.31Personal history of VTE4.7220.000 $470—$783140 (36.3)105.62 ± 12.81Yes11 (2.8)119.64 ± 10.66 More than $78390 (23.3)105.41 ± 11.95No375 (97.2)102.47 ± 11.90Payment manner of the medical expenses35.193#0.000Family history of VTE1.8200.077 Out-of-pocket medical31 (8.0)101.26 ± 10.73Yes30 (7.8)107.67 ± 9.88 Medical services at state expense14 (3.6)107.14 ± 17.22No356 (92.2)102.62 ± 13.32 Medical insurance for urban workers74 (19.2)109.69 ± 12.24Whether the health-care providers provided the VTE knowledge education7.0140.000 The medical insurance for urban residents81 (21.0)103.81 ± 12.17Yes130 (33.7)108.55 ± 11.56 New rural cooperative medical insurance170 (44.0)99.80 ± 10.52No256 (66.3)100.13 ± 11.53 Other16 (4.1)100.81 ± 14.06#: H-value. Characteristics of Patients in neurosurgical hospitalized patients and univariate analysis (n = 386). #: H-value. Reliability and validity of the study constructs A content validity index (CVI) > 0.75 was deemed acceptable24. Findings in this study showed that both item-level and scale-level CVIs ranged from 0.70 to 1.00 and were 0.92, respectively, indicating robust content validity24. A Cronbach’s alpha value of ≥ 0.70 indicates good internal consistency reliability25. In this study, the questionnaire achieved a coefficient of 0.921, demonstrating strong internal consistency. Split-half reliability was calculated at 0.815. Intraclass correlation coefficients (ICC) values ranging from 0.75 to 0.90 indicate strong stability26. For this study, 30 patients were reassessed after 7–10 days, yielding a test–retest reliability coefficient of 0.853 (P < 0.01). These results collectively affirm the questionnaire’s strong reliability and stability. Structural model results In this study, the fit indices for the final 28-item beliefs model were evaluated as follows: χ2 = 617.898, df = 335, P < 0.001, χ2/df = 1.844, AGFI = 0.787, RMSEA = 0.064, GFI = 0.824, RMR = 0.035, IFI = 0.905, CFI = 0.904. All indices met the prescribed thresholds (Table 4), suggesting a satisfactory model fit to the data (Fig. 2). Table 4Fit indices.Fit Indexχ2/dfGFIAGFIRMSEARMRIFICFIMeasurement model1.8440.8240.7870.0640.0350.9050.904Recommended value (good)< 3.00> 0.90> 0.90< 0.05< 0.05> 0.90> 0.90Recommended value (medium)< 5.000–10–10.80–0.100–1> 0.900–1 Fit indices. Standardization six-factors structural equation model of VTE prevention belief questionnaire. Beliefs level of VTE prevention Table 5 shows the status quo of VTE prevention beliefs. The survey yielded a total questionnaire score of 140, with the highest score being 129 and the lowest 71. The average VTE prevention belief score among all participants was 99.56 (SD = 10.11). The scoring rate was 71.11%, indicating a moderate level of VTE prevention belief among neurosurgical inpatients. The scoring rate of each dimension from high to low was perceived benefits (of the VTE prevention behaviours) (80.4%), cues to action (79.3%), self-efficacy (76.2%), perceived seriousness (of VTE) (74.10%), perceived susceptibility (to VTE) (70.60%), perceived barriers (to adopting VTE prevention behaviours) (51.75%). Figure 3 shows the histogram for the total beliefs score. Table 5Beliefs level of VTE prevention in neurosurgical hospitalized patients.DimensionMinimum valueMaximum valueAverage scoreScoring rate (%)Total score7112999.56 ± 10.1171.11Perceived benefits31512.06 ± 1.8880.40Cues to action82015.86 ± 2.2579.30Self-efficacy4107.62 ± 1.3076.20Perceived seriousness42014.82 ± 2.9374.10Perceived susceptibility305538.83 ± 5.1570.60Perceived barriers41610.35 ± 2.1951.75 Beliefs level of VTE prevention in neurosurgical hospitalized patients. The histogram for the total beliefs score. Multiple linear regression analysis Table 3 presents the findings from the univariate analysis. Age, residential location, educational attainment, employment status, monthly household income per capita, method of paying medical expenses, personal history of VTE, surgical history, caregiver presence, sources of VTE-related information, presence of healthcare professionals in the family, and receiving education on VTE prevention from healthcare providers were all significantly associated with VTE prevention beliefs among neurosurgical inpatients (P < 0.05). Table 6 presents the outcomes derived from the multiple linear stepwise regression analysis. Six factors were found to enter the regression equation: education (β = 0.343, P < 0.001), provision of VTE knowledge education by healthcare providers (β = 0.214, P < 0.001), undergoing surgery (β = 0.116, P < 0.01), history of VTE (β = 0.175, P < 0.001), receiving VTE information from family members or friends (β = 0.113, P < 0.01), and payment method for medical expenses (medical insurance for urban workers) (β = 0.112, P < 0.05). Collectively, these variables explained 32.8% of the overall variance. The model exhibited tolerances ranging from 0.740 to 0.979 and variance inflation factors ranging from 1.021 to 1.352, suggesting no substantial collinearity concerns. Table 6Multivariate linear stepwise regression analysis of VTE prevention beliefs scores among neurosurgical hospitalized patients (n = 386).ModelUnstandardized coefficientsStandardized coefficientstPBStd. Errorβ(Constant)88.2981.11179.4820.000Education2.9580.4220.3437.0120.000Whether the health-care providers provided the VTE knowledge education4.5700.9510.2144.8050.000Personal history of VTE10.6082.6190.1754.0510.000Surgery2.3520.8650.1162.7190.007Sources of information related to VTE-Family member/friend3.7461.4100.1132.6580.008Payment manner of the medical expenses-Medical insurance for urban workers2.8821.2480.1122.3100.021Dependent variable: Total VTE beliefs score, the model R = 0.572, R2 = 0.328, revise R2 = 0.317, F = 30.778, P = 0.000 < 0.001, input = 0.05, output = 0.10. Multivariate linear stepwise regression analysis of VTE prevention beliefs scores among neurosurgical hospitalized patients (n = 386). Dependent variable: Total VTE beliefs score, the model R = 0.572, R2 = 0.328, revise R2 = 0.317, F = 30.778, P = 0.000 < 0.001, input = 0.05, output = 0.10. Discussion The current study developed and validated a tool for evaluating VTE prevention beliefs among neurosurgical inpatients using the HBM framework. The HBM, pioneered by a group of psychologists, represents one of the earliest models in behavioral science, elucidating the link between health beliefs and intentional health behaviors27,28. It encompasses six key dimensions: perceived susceptibility, perceived barriers, perceived benefits, perceived severity, cues to action, and self-efficacy12. The instrument developed in this study encompasses all these dimensions, each reflecting aspects of motivation and capability to prevent VTE. The study participants consisted of individuals at heightened risk for VTE, with approximately 71.7% aged over 45 years and nearly two-thirds reporting intracranial tumors. In China, neurosurgical inpatients have reported a DVT incidence as high as 31.1%29. Hence, enhancing understanding and enhancing VTE prevention beliefs among neurosurgical inpatients is crucial, given the significant impact of VTE on patient safety. Improvements in VTE prevention behaviors can be optimized by enhancing positive perceptions across HBM components. In this study, the final 28-item questionnaire was rigorously tested for reliability and stability. Strong internal consistency were observed, test–retest reliability and split-half reliability further affirmed the questionnaire’s stability. Subsequently, construct validity was assessed via CFA to evaluate how well the data aligned with the model. Among the seven model fit indices examined, four met the criteria for good fit (χ2/df, RMR, IFI, CFI), one fell within the acceptable range (RMSEA), and two met the standard threshold (GFI, AGFI), indicating satisfactory fit of the 28-item questionnaire to the six-domain conceptual beliefs model. The results indicated that neurosurgical inpatients had a positive attitude towards VTE prevention, with an average score of 99.56 ± 10.11 and a scoring rate of 71.11%, suggesting a moderate level. This score was slightly lower than a previous study, which reported an average VTE prevention belief score of 31.05 ± 14.56 and a scoring rate of 77.63%13. These results indicate that most neurosurgical inpatients agree on the importance of early VTE prevention. The scores for each dimension, ranked from highest to lowest, were as follows: perceived benefits of healthy behaviour (80.40%), cues to action (79.30%), self-efficacy (76.20%), perceived severity of disease (74.10%), perceived susceptibility to disease (70.60%), perceived obstacles to healthy behaviour (51.75%). A study utilizing the Health Belief Model (HBM) to investigate DVT health belief levels in spinal cord injury patients found that the highest scoring dimension was cues to action, while perceived health behavior disorder scored the lowest30. The finding aligns with the results of this study. For the questionnaire item, “I am concerned about the adverse effects of VTE prophylaxis (e.g., bleeding),” 52.9% of patients in this study selected 4 (agree) or 5 (strongly agree) on the scale, demonstrating that neurosurgical inpatients have high expectations and positive attitudes toward VTE prophylaxis; however, they are still worried about VTE prophylaxis’ adverse effects. This may be due to the increased potential risk of bleeding caused by postoperative prophylactic use of anticoagulants in neurosurgical patients31. In an open-ended question study, most patients after orthopaedic surgery were concerned about anticoagulant drugs’ side effects32. In addition, the perceived susceptibility and severity scores were relatively low, suggesting that the neurosurgical inpatients were largely unaware of VTE severity and risk. These results suggest that medical staff should pay attention to the cognitive status of patients when implementing VTE prevention interventions, especially to help neurosurgical inpatients correctly understand VTE prevention’s adverse effects and VTE’s severity and risk factors. Multiple linear regression analysis revealed that patients with higher education levels had significantly greater VTE prevention belief levels (P < 0.001). This result is consistent with previous research findings, which similarly indicated that individuals who have experienced VTE in the past tend to report higher scores on VTE prevention beliefs compared to those without a history of VTE (P < 0.01)33,34. The patients were considered to have received health education related to VTE. VTE prevention belief was highest in neurosurgical inpatients whose VTE knowledge was mainly from relatives and friends (P < 0.01), indicating that health education from family members is more effective. A previous study showed that family participation in nursing could significantly improve patients’ self-efficacy after orthopaedic surgery35. The occurrence of surgery is another factor influencing VTE prevention beliefs. Results showed that the patients before neurosurgery (104.87 ± 12.70) had weaker VTE prevention beliefs than those after surgery (101.02 ± 11.36) (P < 0.01). This could be attributed to the heightened risk of VTE among post-neurosurgical patients, prompting nursing staff to emphasize the significance of VTE prevention and to repeatedly provide health education to ensure patients remain attentive to the issue36. Patients who explicitly indicated that healthcare providers had educated them about VTE showed higher beliefs in VTE prevention (P < 0.001). This suggests the importance and effectiveness of VTE health education in improving patients’ belief level of VTE prevention and also puts forward certain requirements for medical staff’s ability in VTE health education. Previous studies showed that neurosurgical nursing staff had the lowest VTE prevention cognition levels compared to other high-risk departments of surgery for VTE (including orthopaedics, general surgery, thoracic surgery, and intensive care medicine)37,38. Therefore, strengthening the professional training of neurosurgical medical staff on VTE prevention and improving VTE health education to enhance patients’ VTE prevention beliefs are necessary. There were several limitations noted in this study. To begin with, the study recruited participants exclusively from a single tertiary-level hospital, potentially restricting the broader applicability of the findings;This questionnaire warrants validation across diverse institutions and with a larger sample size to enhance its generalizability. Secondly, respondents may potentially misconstrue VTE as cerebral thrombosis, leading to inaccuracies in their responses regarding VTE prevention beliefs. Last, owing to the study’s cross-sectional design, recall bias cannot be ruled out. Conclusions Based on the results, the questionnaire assessing VTE prevention beliefs among neurosurgical inpatients demonstrated strong psychometric properties, with all six constructs exhibiting high internal consistency, as well as robust convergent and discriminant validity. The overall VTE prevention belief score was moderate, suggesting room for improvement in patient awareness and understanding. Key factors influencing VTE prevention beliefs included educational attainment, receipt of VTE knowledge education from healthcare providers, personal history of VTE, surgical history, sources of VTE-related information, and method of paying medical expenses. These findings highlight the need for targeted interventions, particularly focusing on enhancing patient education, especially for those with lower educational levels or limited access to healthcare information. Among the belief dimensions, the scoring rate of perceived benefits and cues to action ranked highest, while perceived barriers were the lowest. These findings suggest that while patients recognize the benefits of VTE prevention, barriers to adopting preventive behaviors remain a significant concern. Targeted interventions should focus on reducing these barriers and enhancing self-efficacy to improve overall VTE prevention awareness and actions among neurosurgical inpatients. Abbreviations Venous thromboembolism Health Belief Model Standard deviation Deep venous thrombosis Pulmonary embolism Content Validity Index Statistical package for the social sciences Analysis of moment structure Analysis of variance Confirmatory factor analysis Exploratory factor analysis Perceived susceptibility Basic prophylaxis behaviors Perceived severity Mechanical prophylaxis behaviors Perceived benefits Anticoagulant prophylaxis behaviors Perceived barriers Professional prophylaxis behaviors Cues to action Self-efficacy Central South University Author contributions A was responsible for conceptualization, questionnaire design, data collection and input, data curation and analysis, and writing the original draft as well as reviewing and editing the manuscript. Funding This study was supported by grants from the Science and Technology Research Project of Jiangxi Provincial Department of Education (No. GJJ2401613). Data availability Data is provided within the manuscript files. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement Data is provided within the manuscript files.