Abstract
This study evaluates the effectiveness of predictive nursing in preventing lower extremity deep vein thrombosis (DVT) in severe trauma patients compared with conventional care. This retrospective study included severe trauma patients hospitalized between October 2023 and January 2024. Patients were divided into an observation group (51 cases, predictive nursing) and a control group (69 cases, conventional care). Propensity score matching was used to balance baseline characteristics. Differences in DVT incidence, hospital stay, hematological indicators, complications, and rehabilitation outcomes were analyzed. DVT incidence was significantly lower in the observation group (8.90%) compared with the control group (26.70%, P < .05), with a relative risk reduction of 67%. Hospital stays were shorter in the observation group (14.2 ± 4.1 vs 16.5 ± 5.2 days, P = .03). Improvements in D-dimer levels, prothrombin time, and international normalized ratio were significant in the observation group (P < .05), indicating lower thrombosis risk. Hematocrit (HCT) and platelet count (PLT) remained stable, showing significant differences compared with the control group (P < .05). Rehabilitation outcomes, including lower limb function and activity capacity, were significantly better in the observation group (P < .05). The control group had 4 cases of pulmonary embolism, while none were observed in the observation group (P = .03). Predictive nursing based on Caprini risk assessment effectively reduces DVT incidence, improves hematological profiles, shortens hospital stays, and enhances recovery in severe trauma patients. This personalized care model shows excellent clinical potential.
Keywords: Caprini risk assessment, deep vein thrombosis, predictive nursing, prevention, rehabilitation
1. Introduction
Deep vein thrombosis (DVT) is a common and potentially life-threatening complication in patients with severe trauma, particularly those with fractures, soft tissue injuries, or spinal injuries. DVT refers to the formation of thrombi in deep veins, usually in the lower extremities. If untreated, it can lead to serious consequences such as pulmonary embolism, which remains a leading cause of mortality and disability among hospitalized trauma patients. Therefore, developing effective prevention strategies is critical to reducing the incidence of DVT and improving patient outcomes.[1–5]
Current strategies for DVT prevention primarily include pharmacological and non-pharmacological approaches. While conventional care often involves anticoagulant therapy and mechanical prophylaxis (e.g., compression stockings and intermittent pneumatic compression (IPC) devices), individualized predictive nursing plans are gaining attention. Based on effective risk assessment tools such as the Caprini or Padua risk assessment models, predictive nursing aims to identify high-risk patients and implement personalized interventions from the start of hospitalization. This approach encompasses not only anticoagulation and mechanical devices but also the promotion of physical activity, early mobilization, and monitoring of hematological parameters.[6–8]
Despite the existing evidence highlighting the importance of risk assessment and early intervention in reducing DVT incidence, several gaps remain in current research. First, although multiple studies have shown that risk assessment can effectively predict the risk of DVT, comparative studies on the effectiveness of individualized care based on risk assessment for different types of trauma patients are relatively limited. Second, existing research mainly focuses on evaluating the effects of pharmacological interventions, while the role of comprehensive interventions (such as the combination of pharmacological and non-pharmacological treatments) in DVT prevention has not been systematically explored. Furthermore, many studies fail to consider individual patient differences, such as age, underlying diseases, and comorbidities, which may impact DVT incidence.[9–11] Therefore, this study aims to address these gaps by evaluating the effectiveness of predictive nursing based on the Caprini risk assessment model in preventing DVT in severe trauma patients and comparing it with conventional care.
This study hypothesizes that predictive nursing based on the Caprini risk assessment model can significantly reduce the incidence of postoperative DVT and improve clinical outcomes. Specifically, we anticipate significant differences between the observation group (receiving predictive nursing based on Caprini risk assessment) and the control group (receiving conventional care) in terms of DVT incidence, hospital stay duration, rehabilitation progress, and overall quality of life. Through this study, we hope to provide more scientific and effective DVT prevention strategies for clinical practice and promote the broader application of predictive nursing based on individual risk assessment in trauma patients.
2. Materials and methods
2.1. Study subjects
This study was approved by the Ethics Committee of the First Affiliated Hospital of Hebei North University. This retrospective case-control study aimed to evaluate the effectiveness of predictive nursing in preventing lower extremity DVT in patients with severe trauma. Medical records of 120 patients treated for severe trauma at our hospital between October 2023 and January 2024 were collected. Patients were divided into a predictive nursing group (45 patients) and a conventional care group (75 patients) based on prior nursing interventions. To minimize the impact of confounding factors, propensity score matching (PSM) was employed to select a suitable control group for the predictive nursing group, resulting in 45 patients in each group for outcome evaluation.
Inclusion criteria: Patients aged 18 to 80 years, meeting the diagnostic criteria for severe trauma (e.g., ISS score ≥ 16), with no prior history of lower extremity DVT at admission, a minimum hospital stay of 7 days, and complete clinical records (including laboratory indicators, imaging examinations, and nursing records). Additionally, patients must have undergone lower extremity DVT screening (e.g., color Doppler ultrasound) with confirmed results.
Exclusion criteria: Patients with a history of lower extremity DVT or other venous thrombotic diseases prior to admission; those with severe cardiovascular or cerebrovascular diseases (e.g., heart failure, stroke) or malignancies; pregnant or lactating women; patients with incomplete data due to death or transfer to another hospital; those with severe infections or autoimmune diseases; and patients who did not undergo standardized lower extremity DVT screening.
2.2. Nursing procedures
2.2.1. Predictive nursing workflow
2.2.1.1. Early risk assessment upon admission
Within 24 hours of admission, the nursing team conducts a comprehensive DVT risk assessment using standardized tools such as the Caprini or Padua scoring systems. The assessment includes factors such as the type of trauma (e.g., fractures or soft tissue injuries), ISS score, medical history (e.g., prior DVT or venous thromboembolism), laboratory indicators (e.g., D-dimer, prothrombin time [PT]), and the patient’s mobility status. Based on the evaluation, patients are categorized into low-risk, moderate-risk, and high-risk groups, with individualized nursing intervention plans developed accordingly.
2.2.1.2. Individualized interventions
Low-Risk Patients: Encourage early active mobilization with at least 3 bedside joint exercises daily, each lasting over 10 minutes. Assist with repositioning every 4 hours and elevate the lower limbs to promote venous return. Ensure daily fluid intake of 2000 to 2500 mL unless contraindicated.
Moderate-Risk Patients: Apply graduated compression stockings for at least 12 h/d. Utilize continuous passive motion devices for lower limb exercises, 20 minutes per session, 2 to 3 times daily. Provide psychological support through daily communication to alleviate anxiety.
High-Risk Patients: Use IPC devices 2 to 3 times daily, each session lasting 30 minutes. Perform the first color Doppler ultrasound on the third day of admission, followed by repeat scans every 5 days. Dynamically monitor coagulation parameters to detect early signs of thrombosis.
2.2.1.3. Health education
The nursing team conducts daily one-on-one health education sessions for patients and their families, lasting approximately 10 minutes. The sessions cover high-risk factors for DVT, early symptoms (e.g., calf swelling, skin redness, pain), proper use and precautions for compression stockings, and dietary guidance (e.g., high-protein, low-fat diets). Upon discharge, patients receive personalized written instructions detailing follow-up activity plans, dietary recommendations, and methods for monitoring thrombosis symptoms. A follow-up visit or ultrasound screening is scheduled 3 weeks post-discharge.
2.2.1.4. Team collaboration and quality control
The nursing team collaborates closely with trauma physicians, rehabilitation therapists, and ultrasound specialists to develop comprehensive prevention plans for patients. Weekly team meetings are held to analyze the effectiveness of nursing interventions, identify issues, and optimize processes. Nursing staff document daily care measures, monitoring indicators, and feedback to ensure procedural standardization and traceable quality control.
2.2.2. Conventional nursing workflow
2.2.2.1. Basic care upon admission
Within 24 hours of admission, nursing staff conduct a basic assessment, primarily documenting the type of trauma, the patient’s mobility, and vital signs.
2.2.2.2. Routine interventions
General positioning management is performed, with position adjustments twice daily without a specific schedule or emphasis on elevating the lower limbs. Patients are encouraged to perform in-bed activities, but there is no structured plan regarding the duration or intensity of these activities. Physical prevention measures are inconsistently applied; some patients receive gradient compression stockings or IPC devices, but their use is not systematic or standardized.
2.2.2.3. Health education
Health education is provided only when patients actively ask questions, focusing on basic nursing knowledge during hospitalization. There is no dedicated guidance on the importance of DVT prevention or specific preventive measures. At discharge, simple recommendations are provided, such as rest and general dietary advice, with no emphasis on DVT prevention-related precautions.
2.2.2.4. Monitoring and follow-up
DVT screening is performed only when patients exhibit obvious symptoms (e.g., lower limb swelling or pain) and is not part of a routine dynamic monitoring plan. No systematic follow-up or reevaluation is arranged after discharge, leaving patients to return for consultations on their own, which may result in missed or delayed diagnoses of early thrombosis.
2.3. Data collection
2.3.1. Baseline characteristics
Baseline characteristics were collected on the day of admission (day 0). The collected data included the patient’s age, gender, body mass index (BMI), Injury Severity Score (ISS), type of trauma (e.g., fracture or soft tissue injury), presence of multiple fractures, spinal injuries, smoking and alcohol consumption history, and comorbidities such as diabetes, hypertension, or malnutrition. All information was recorded in detail from the patient’s electronic medical records and admission medical examination records.
2.3.2. Incidence of deep vein thrombosis
The incidence of DVT was recorded on day 28 of hospitalization. Data included whether each patient developed DVT, confirmed by ultrasound examination. The incidence of DVT was documented by clinical physicians based on regular ultrasound results and analyzed in relation to the patient’s adherence to preventive nursing interventions. DVT diagnosis was confirmed using standardized imaging protocols in accordance with the 2023 American Society of Hematology guidelines. Diagnostic criteria for acute DVT included: non-compressibility of the vein in the transverse plane at defined anatomical landmarks (common femoral vein at the inguinal ligament, popliteal vein at the knee crease, and posterior tibial veins at the ankle); absence of spontaneous or phasic Doppler flow signals within the thrombosed segment; 3. Visualization of intraluminal thrombus on grayscale imaging.
All imaging findings were independently reviewed by 2 radiologists, with discrepancies resolved through consensus. Asymptomatic DVT (detected via screening without clinical signs) and symptomatic DVT were recorded separately.
2.3.3. Hematological indicators
Hematological indicators were collected through blood tests on days 0, 7, 14, and 28 of hospitalization. These included D-dimer, international normalized ratio (INR), PT, and hematocrit (HCT). The data were obtained from routine blood tests and used to evaluate the patients’ coagulation function and its trends, providing insights into the effectiveness of nursing interventions.
2.3.4. Length of hospital stay
The total length of hospital stay, measured from admission to discharge, was automatically recorded by the hospital information system. Changes in the patient’s condition, treatment progress, and rehabilitation during hospitalization influenced this duration. This metric reflects the impact of nursing interventions on reducing hospital stay.
2.3.5. Rehabilitation progress
Rehabilitation progress was assessed on day 14 of hospitalization. The evaluation included activity recovery scores (0–10), lower limb function recovery scores (0–10), and the Barthel Index. Activity recovery and lower limb function scores were determined by physicians based on the patient’s actual recovery. The Barthel Index was used to assess the patient’s daily living abilities, particularly self-care. All evaluations adhered to standardized scoring scales to ensure objectivity and consistency.
2.3.6. Quality of life assessment
Quality of life was assessed both before the intervention (at admission) and after the intervention (day 28). The SF-36 or EQ-5D questionnaire was used to evaluate multiple dimensions of quality of life, including physical functioning, role functioning, bodily pain, general health, vitality, social functioning, emotional roles, and mental health. Quality of life was assessed through self-reported questionnaires, with patients completing the forms based on their own experiences. These data were analyzed to determine the effectiveness of predictive nursing interventions in improving patients’ quality of life.
2.4. Statistical methods
Data analysis was conducted using SPSS 25.0 statistical software. Descriptive statistics for baseline characteristics, changes in hematological indicators, and rehabilitation progress were expressed as mean ± standard deviation for continuous variables, or as median and interquartile range for non-normally distributed data. Categorical variables were presented as frequencies and percentages. For group comparisons, independent-sample t-tests were used for continuous variables with normal distribution, and chi-square tests were applied for categorical variables. Paired t-tests were used to assess changes before and after interventions in paired data. To address baseline differences between the intervention and control groups, PSM was used to ensure comparability across multiple baseline characteristics. The relative risk (RR) and odds ratio (OR) were calculated to evaluate the risk of DVT incidence between groups. For repeated measures data, such as hematological indicators, mixed-effects models or analysis of variance (ANOVA) were employed to assess the effects of the intervention over time and between groups. All statistical tests were 2-tailed, with the significance level set at P < .05. Differences were considered statistically significant when P < .05.
3. Results
3.1. Baseline characteristics of predictive nursing and conventional care groups before and after propensity score matching
Significant differences were observed between the predictive nursing group and the conventional care group before PSM, as shown in Table 1. Patients in the predictive nursing group were younger (43.1 ± 15.3 years vs 50.4 ± 12.7 years, P = .01), had a higher BMI (26.4 ± 3.2 vs 24.8 ± 2.5 kg/m2, P = .03), and higher ISSs (21.3 ± 3.2 vs 18.7 ± 2.5, P < .01). The group also exhibited higher rates of fractures (P = .04), multiple fractures (P = .03), greater intraoperative blood loss (P = .03), and earlier initiation of anticoagulation therapy (P < .01).
Table 1.
Baseline characteristics of patients before and after PSM.
| Variable | Pre-matching predictive care group (n = 45) | Pre-matching conventional care group (n = 75) | P-value | Post-matching predictive care group (n = 45) | Post-matching conventional care group (n = 45) | P-value |
|---|---|---|---|---|---|---|
| Age (yr, mean ± SD) | 43.1 ± 15.3 | 50.4 ± 12.7 | .01* | 43.1 ± 15.3 | 43.5 ± 14.9 | .87 |
| Gender (male/female, n) | 27/18 | 50/25 | .35 | 27/18 | 28/17 | .84 |
| BMI (kg/m2, mean ± SD) | 26.4 ± 3.2 | 24.8 ± 2.5 | .03* | 26.4 ± 3.2 | 26.2 ± 3.0 | .72 |
| ISS score (mean ± SD) | 21.3 ± 3.2 | 18.7 ± 2.5 | <.01* | 21.3 ± 3.2 | 21.0 ± 3.4 | .68 |
| Trauma type (fracture/soft tissue, n) | 35/10 | 50/25 | .04* | 35/10 | 34/11 | .82 |
| Multiple fractures (yes/no, n) | 25/20 | 29/46 | .03* | 25/20 | 24/21 | .89 |
| Spinal injury (yes/no, n) | 10/35 | 15/60 | .28 | 10/35 | 11/34 | .78 |
| Smoking history (yes/no, n) | 12/33 | 20/55 | .22 | 12/33 | 13/32 | .79 |
| Alcohol history (yes/no, n) | 18/27 | 15/60 | <.01* | 18/27 | 17/28 | .85 |
| Diabetes (yes/no, n) | 8/37 | 16/59 | .19 | 8/37 | 7/38 | .78 |
| Hypertension (yes/no, n) | 10/35 | 18/57 | .14 | 10/35 | 11/34 | .81 |
| Malnutrition (yes/no, n) | 5/40 | 12/63 | .05* | 5/40 | 4/41 | .81 |
| Surgical intervention (yes/no, n) | 40/5 | 52/23 | .02* | 40/5 | 39/6 | .9 |
| Surgery duration (min, mean ± SD) | 120.5 ± 35.2 | 135.8 ± 38.4 | .04* | 120.5 ± 35.2 | 122.3 ± 34.9 | .78 |
| Intraoperative blood loss (mL, mean ± SD) | 320.4 ± 120.5 | 275.2 ± 115.6 | .03* | 320.4 ± 120.5 | 315.8 ± 123.4 | .84 |
| Anticoagulant therapy (yes/no, n) | 30/15 | 30/45 | <.01* | 30/15 | 31/14 | .9 |
| Anticoagulant start time (d, mean ± SD) | 2.3 ± 0.8 | 2.8 ± 0.7 | <.01* | 2.3 ± 0.8 | 2.4 ± 0.6 | .79 |
| Bed rest duration (d, mean ± SD) | 9.8 ± 3.2 | 10.4 ± 3.5 | .32 | 9.8 ± 3.2 | 10.0 ± 3.1 | .89 |
| Time from admission to intervention (d) | 1.8 ± 0.6 | 2.4 ± 0.7 | <.01* | 1.8 ± 0.6 | 1.9 ± 0.5 | .68 |
PSM = propensity score matching.
The P-value is statistically significant.
After PSM, these differences were eliminated, ensuring comparability between groups for evaluating the effectiveness of predictive nursing in preventing DVT.
3.2. Relationship between adherence levels and DVT incidence in predictive nursing and conventional care groups
Analysis of DVT incidence based on adherence levels revealed significant differences between the predictive nursing group and the conventional care group, as shown in Table 2. In the predictive nursing group, the DVT incidence was 4.0% among high-adherence patients, 13.3% in the moderate-adherence group, and 20.0% in the low-adherence group. In the conventional care group, the DVT incidence was 8.0% in the high-adherence group, 50.0% in the moderate-adherence group, and 70.0% in the low-adherence group.
Table 2.
DVT incidence according to compliance levels in predictive care and conventional care groups.
| Care group | Compliance level | Number of patients (n) | DVT cases (n) | DVT incidence (%) | Relative risk (RR) | Odds ratio (OR) | P-value |
|---|---|---|---|---|---|---|---|
| Predictive care | High compliance | 25 | 1 | 4.00% | Reference | Reference | <.05* |
| Predictive care | Medium compliance | 15 | 2 | 13.30% | 3.33 | 4.67 | <.05* |
| Predictive care | Low compliance | 5 | 1 | 20.00% | 5 | 7 | <.05* |
| Conventional care | High compliance | 25 | 2 | 8.00% | Reference | Reference | <.05* |
| Conventional care | Medium compliance | 10 | 5 | 50.00% | 6.25 | 8 | <.05* |
| Conventional care | Low compliance | 10 | 7 | 70.00% | 8.75 | 15 | <.05* |
DVT = deep vein thrombosis.
The P-value is statistically significant.
These findings highlight that higher adherence to preventive measures, especially in predictive nursing, is closely associated with a significantly lower risk of DVT.
3.3. Effectiveness of predictive nursing in reducing DVT incidence
Compared with the conventional care group, the predictive nursing group demonstrated a significantly lower incidence of DVT, as shown in Table 3. The incidence of DVT was 8.90% (4/45) in the predictive nursing group, compared to 26.70% (12/45) in the conventional care group, with a statistically significant difference (P < .05).
Table 3.
Comparison of DVT incidence between predictive care and conventional care groups.
| Group | Number of patients (n) | DVT cases (n) | Incidence rate (%) | Relative risk (RR) | Odds ratio (OR) | P-value |
|---|---|---|---|---|---|---|
| Predictive care | 45 | 4 | 8.90% | Reference | Reference | |
| Conventional care | 45 | 12 | 26.70% | 0.33 | 0.27 | <.05* |
DVT = deep vein thrombosis.
The P-value is statistically significant.
The RR of developing DVT in the predictive nursing group was 0.33, indicating a 67% reduction in risk. The OR was 0.27, showing a 73% reduction in the likelihood of DVT compared with the conventional care group. These results confirm the effectiveness of predictive nursing in reducing the incidence of DVT in patients with severe trauma.
3.4. Effectiveness of predictive nursing in modulating hematological indicators during hospitalization
This study compared changes in hematological indicators, including D-dimer, INR, PT, and HCT, between the 2 groups during hospitalization, as shown in Table 4. At admission (day 0), there were no significant differences in hematological indicators between the predictive nursing and conventional care groups.
Table 4.
Hematological indices changes between predictive care and conventional care groups at different time points.
| Parameter | Time point | Predictive care group (n = 45, Mean ± SD) | Conventional care group (n = 45, Mean ± SD) | P-value (group) | P-value (time) | P-value (group × time) |
|---|---|---|---|---|---|---|
| D-dimer (mg/L) | Admission (day 0) | 3.6 ± 0.8 | 3.6 ± 0.7 | .89 | <.01 | <.01 |
| Day 7 | 3.0 ± 0.7 | 3.2 ± 0.8 | .32 | |||
| Day 14 | 2.4 ± 0.6 | 2.8 ± 0.7 | .04* | |||
| Day 28 | 1.6 ± 0.4 | 2.6 ± 0.6 | <.01* | |||
| INR | Admission (day 0) | 1.15 ± 0.10 | 1.14 ± 0.11 | .72 | <.05 | .03 |
| Day 7 | 1.12 ± 0.09 | 1.13 ± 0.10 | .65 | |||
| Day 14 | 1.08 ± 0.08 | 1.10 ± 0.09 | .03* | |||
| Day 28 | 1.02 ± 0.07 | 1.08 ± 0.08 | <.01* | |||
| PT (s) | Admission (day 0) | 14.5 ± 1.5 | 14.4 ± 1.4 | .81 | .03 | .02 |
| Day 7 | 13.9 ± 1.3 | 14.1 ± 1.3 | .42 | |||
| Day 14 | 13.2 ± 1.2 | 13.6 ± 1.3 | .04* | |||
| Day 28 | 12.7 ± 1.0 | 13.7 ± 1.2 | <.01* | |||
| HCT (%) | Admission (day 0) | 38.2 ± 3.4 | 38.1 ± 3.2 | .92 | .01 | .02 |
| Day 7 | 38.8 ± 3.3 | 38.7 ± 3.5 | .85 | |||
| Day 14 | 39.2 ± 3.2 | 38.5 ± 3.6 | .03* | |||
| Day 28 | 39.5 ± 3.1 | 38.0 ± 3.5 | <.01* |
The P-value is statistically significant.
However, as the hospitalization period progressed, the predictive nursing group demonstrated significantly better improvements. On day 14, the predictive nursing group had a lower D-dimer level (2.4 ± 0.6 mg/L vs 2.8 ± 0.7 mg/L, P = .04) and INR (1.08 ± 0.08 vs 1.10 ± 0.09, P = .03) compared with the conventional care group. By day 28, the D-dimer level in the predictive nursing group had decreased to 1.6 ± 0.4 mg/L, INR to 1.02 ± 0.07, PT to 12.7 ± 1.0 seconds, and HCT increased to 39.5 ± 3.1%, all significantly better than those in the conventional care group. These results indicate that predictive nursing has a significant positive impact on improving coagulation function and hemorheological parameters in patients with severe trauma.
3.5. Impact of predictive nursing on hospital stay and rehabilitation outcomes
Analysis of hospital stay and rehabilitation progress showed that the predictive nursing group had a significantly shorter hospital stay (14.2 ± 4.1 days) compared to the conventional care group (16.5 ± 5.2 days, P = .03), as shown in Table 5. Additionally, the predictive nursing group demonstrated better recovery in activity levels (8.3 ± 1.2 vs 7.1 ± 1.5, P = .02) and lower limb function (7.9 ± 1.3 vs 6.5 ± 1.8, P = .01). However, no significant difference was observed in functional independence as measured by the Barthel Index (P = .15). These results suggest that predictive nursing contributes to shorter hospital stays and improved recovery outcomes, particularly in activity levels and lower limb functionality.
Table 5.
Comparison of hospital stay and rehabilitation progress between predictive care and conventional care groups.
| Variable | Predictive care group (n = 45) | Conventional care group (n = 45) | P-value |
|---|---|---|---|
| Hospital stay (d) | 14.2 ± 4.1 | 16.5 ± 5.2 | .03* |
| Activity recovery score (0–10) | 8.3 ± 1.2 | 7.1 ± 1.5 | .02* |
| Functional independence (Barthel Index) | 85.2 ± 15.4 | 80.3 ± 18.7 | .15 |
| Lower limb recovery score (0–10) | 7.9 ± 1.3 | 6.5 ± 1.8 | .01* |
The P-value is statistically significant.
3.6. Impact of predictive nursing on quality of life
Analysis of quality of life dimensions revealed significant improvements in the predictive nursing group compared to the conventional care group, as shown in Table 6. After the intervention, the predictive nursing group demonstrated marked increases in physical functioning (75.8 ± 9.5 vs 55.6 ± 12.3, P < .05), role functioning (60.5 ± 13.2 vs 42.3 ± 14.1, P = .02), general health (70.4 ± 8.9 vs 58.2 ± 10.4, P = .03), vitality (63.2 ± 10.1 vs 48.9 ± 11.7, P = .04), social functioning (69.5 ± 13.2 vs 53.7 ± 15.6, P = .01), emotional role (63.6 ± 12.5 vs 45.1 ± 14.2, P = .03), and mental health (65.4 ± 12.9 vs 50.2 ± 13.8, P = .02). While the conventional care group also showed some improvements, these were generally less pronounced. These findings indicate that predictive nursing significantly enhances patients’ quality of life, especially in physical and emotional health dimensions.
Table 6.
Comparison of quality of life dimensions before and after intervention between predictive care and conventional care groups.
| Dimension | Predictive care group (pre-intervention, Mean ± SD) | Predictive care group (post-intervention, Mean ± SD) | Routine care group (pre-intervention, Mean ± SD) | Routine care group (post-intervention, Mean ± SD) | P-value (post-intervention between groups) |
|---|---|---|---|---|---|
| Physical function | 55.6 ± 12.3 | 75.8 ± 9.5 | 53.4 ± 11.2 | 60.2 ± 13.7 | <.05* |
| Role function | 42.3 ± 14.1 | 60.5 ± 13.2 | 40.5 ± 12.3 | 45.6 ± 14.8 | .02 |
| Bodily pain | 55.4 ± 15.3 | 70.1 ± 13.4 | 54.2 ± 16.2 | 58.3 ± 14.5 | .06 |
| General health | 58.2 ± 10.4 | 70.4 ± 8.9 | 56.6 ± 11.5 | 60.1 ± 10.0 | .03 |
| Vitality | 48.9 ± 11.7 | 63.2 ± 10.1 | 47.5 ± 12.0 | 50.6 ± 13.1 | .04 |
| Social function | 53.7 ± 15.6 | 69.5 ± 13.2 | 51.3 ± 14.3 | 55.8 ± 15.6 | .01 |
| Emotional role | 45.1 ± 14.2 | 63.6 ± 12.5 | 43.3 ± 12.7 | 49.2 ± 13.0 | .03 |
| Mental health | 50.2 ± 13.8 | 65.4 ± 12.9 | 48.7 ± 14.3 | 53.4 ± 15.2 | .02 |
The P-value is statistically significant.
4. Discussion
This study evaluated the effectiveness of predictive nursing in preventing lower extremity DVT in patients with severe trauma and compared it with conventional care. By using PSM, baseline differences were minimized, ensuring comparability between the 2 groups across multiple variables. The results demonstrated that predictive nursing significantly reduced DVT incidence and improved coagulation function, rehabilitation progress, and quality of life in patients. This study demonstrates the effectiveness of predictive nursing guided by the Caprini risk assessment model in reducing DVT incidence among severe trauma patients. However, the generalizability of the Caprini model across diverse patient subgroups warrants careful consideration. While the model integrates broad risk factors (e.g., age, immobility, surgery type), its performance in specific trauma populations – such as those with polytrauma, spinal injuries, or penetrating trauma – remains underexplored in this study. For instance, our cohort predominantly included patients with lower limb fractures (68%), potentially limiting insights into its applicability for visceral or neurovascular trauma.
The study found that the incidence of DVT in the predictive nursing group was 8.9%, significantly lower than the 26.7% observed in the conventional care group (P < .05). This indicates that predictive nursing effectively prevented DVT in patients with severe trauma through early risk assessment, individualized interventions, and enhanced adherence to preventive measures. In contrast, the conventional care group, which lacked systematic DVT prevention strategies, showed a higher incidence of DVT. This finding aligns with previous studies, reinforcing the importance of regular risk assessments and interventions in reducing thrombosis formation.
In this study, the predictive nursing group demonstrated superior improvements in hematological indicators during hospitalization compared to the conventional care group. Notably, significant improvements in D-dimer, PT, and INR were observed on days 14 and 28 in the predictive nursing group. These results suggest that predictive nursing, through systematic coagulation monitoring and targeted interventions, effectively optimized coagulation function and reduced thrombosis risk. Furthermore, the significant increase in HCT in the predictive nursing group may reflect the potential role of nursing interventions in improving blood viscosity.[12–17]
This study also found that the hospital stay in the predictive nursing group was significantly shorter than that in the conventional care group (14.2 days vs 16.5 days, P = .03). This result is likely associated with the effectiveness of predictive nursing in preventing complications and reducing the incidence of DVT. A shorter hospital stay not only alleviates patient suffering but also reduces healthcare costs. Additionally, the predictive nursing group achieved significantly higher scores in activity recovery and lower limb function recovery compared to the conventional care group (P < .05). These findings suggest that predictive nursing promotes early rehabilitation, particularly in lower limb function, which is crucial for patients with severe trauma.[18–20]
In terms of quality of life, the predictive nursing group scored significantly higher than the conventional care group across various dimensions, including physical functioning, role functioning, vitality, social functioning, and emotional roles. This indicates that predictive nursing not only improves physical health but also has a positive impact on emotional and psychological well-being. Notably, the intervention significantly enhanced emotional health, improving patients’ mental states.[21,22]
The study also revealed a close relationship between adherence to care and DVT incidence. In the predictive nursing group, patients with high adherence had a DVT incidence of 4.0%, while those with low adherence showed a significantly higher incidence of 20.0%. Similarly, in the conventional care group, lower adherence was associated with a higher DVT incidence. This finding underscores the critical role of adherence in preventing DVT, especially when using individualized, systematic nursing interventions.[23–25]
Despite its strengths, this study has several limitations that may affect the generalizability and reliability of the findings. First, as a retrospective analysis, it is susceptible to selection bias and information bias, and residual confounding factors cannot be entirely excluded. Second, the sample size was relatively small and limited to patients with lower limb fractures hospitalized between October 2023 and January 2024, which may affect the external validity of the findings and their applicability to other types of trauma or long-term follow-up patients. The follow-up period was relatively short, and the long-term health effects of nursing interventions were not evaluated. Although PSM was applied to balance baseline characteristics between groups, unmeasured confounding factors – such as genetic predispositions, variations in anticoagulant medication adherence, or subtle differences in preexisting comorbidities (e.g., undiagnosed hypercoagulable states) – may still influence the results. Retrospective designs also limit the ability to establish causality, as unobserved variables (e.g., subtle differences in rehabilitation protocols or undocumented patient behaviors) could contribute to outcome disparities.
Additionally, there may be heterogeneity in the implementation of nursing interventions, as differences in execution standards across hospitals and nursing teams could influence the outcomes. The diagnosis of DVT relied on imaging and clinical symptoms, and variations in diagnostic criteria may have led to misdiagnoses or underdiagnoses. Furthermore, the influence of patient adherence on outcomes was not fully evaluated, even though it may be a critical factor in the effectiveness of interventions. Lastly, as a single-center study, the findings lack multi-center validation and do not consider the impact of other postoperative complications on patient health. To address these limitations, prospective multicenter randomized controlled trials (RCTs) with larger sample sizes are warranted. RCTs would allow stricter control of confounding variables, standardized implementation of predictive nursing protocols, and longitudinal follow-up to assess sustained benefits. Moreover, Prospective studies should stratify analyses by trauma type, age decades, and comorbidity clusters. Multicenter collaborations could pool data to validate Caprini thresholds in underrepresented groups, such as pediatric trauma or obstetric patients. Adaptive machine learning approaches, incorporating dynamic biomarkers (e.g., serial D-dimer trends) and injury-specific factors (e.g., ISS sub-scores), may refine risk prediction beyond static Caprini criteria. And future studies should incorporate extended follow-up periods (e.g., 6–12 months) to assess critical long-term outcomes such as DVT recurrence, quality of life, long-term recovery, and more.
5. Conclusion
In summary, this study demonstrated that predictive nursing effectively prevents the occurrence of DVT in patients with severe trauma and significantly improves coagulation function, hospitalization recovery progress, and quality of life. Based on these findings, it is recommended to promote predictive nursing interventions in clinical practice, particularly for DVT prevention in patients with severe trauma, to enhance clinical outcomes and improve patients’ quality of life.
Author contributions
Conceptualization: Shuwei Huang, Chunyan Liu.
Data curation: Shuwei Huang, Juan Cui, Lilin Cao, Bo Wang, Ruyue Wang, Chunyan Liu.
Formal analysis: Shuwei Huang, Juan Cui, Lilin Cao, Bo Wang, Ruyue Wang, Chunyan Liu.
Investigation: Shuwei Huang, Juan Cui, Liqun Jia, Ruyue Wang.
Methodology: Shuwei Huang, Liqun Jia, Ruyue Wang.
Software: Lilin Cao.
Supervision: Chunyan Liu.
Validation: Chunyan Liu.
Visualization: Juan Cui, Lilin Cao, Bo Wang, Liqun Jia.
Writing – original draft: Shuwei Huang, Juan Cui, Lilin Cao, Bo Wang, Liqun Jia, Chunyan Liu.
Writing – review & editing: Shuwei Huang, Chunyan Liu.
Abbreviations:
- BMI
- body mass index
- CPM
- continuous passive motion
- DVT
- Deep vein thrombosis
- HCT
- Hematocrit
- INR
- international normalized ratio
- IPC
- intermittent pneumatic compression
- ISS
- Injury Severity Score
- OR
- odds ratio
- PE
- pulmonary embolism
- PLT
- platelet count
- PSM
- propensity score matching
- PT
- prothrombin time,
- RR
- relative risk
Zhangjiakou Science and Technology Bureau 2023 Municipal Science and Technology Plan Project (2322150D).
The authors have no conflicts of interest to disclose.
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
How to cite this article: Huang S, Cui J, Cao L, Wang B, Jia L, Wang R, Liu C. Predictive nursing in the prevention of lower extremity deep vein thrombosis in patients with severe trauma. Medicine 2025;104:22(e42446).
Contributor Information
Shuwei Huang, Email: shuwei19821012@163.com.
Juan Cui, Email: 631087239@qq.com.
Lilin Cao, Email: Cy15127370416@163.com.
Bo Wang, Email: 847726336@qq.com.
Liqun Jia, Email: 773930433@qq.com.
Ruyue Wang, Email: 847726336@qq.com.
References
- [1].Jia CY, Dai DD, Bi XY, Zhang X, Wang YN. Advancements in the interventional therapy and nursing care on deep vein thrombosis in the lower extremities. Front Med (Lausanne). 2024;11:1420012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Abate LG, Bayable SD, Fetene MB. Evidence-based perioperative diagnosis and management of pulmonary embolism: a systematic review. Ann Med Surg (Lond). 2022;77:103684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Ruan Y, Wang F, Du X, Sun S. Rehabilitation nursing after lower limb fracture: preventing deep vein thrombosis and enhancing quality of life. Medicine (Baltimore). 2023;102:e36180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Avila M, Bhogadi SK, Nelson A, et al. The long-term risks of venous thromboembolism among non-operatively managed spinal fracture patients: a nationwide analysis. Am J Surg. 2023;225:1086–90. [DOI] [PubMed] [Google Scholar]
- [5].Zhang H, Dou Y. Precision nursing and intermittent pneumatic compression significantly reduce perioperative deep vein thrombosis in post-surgical ovarian cancer patients. Am J Transl Res. 2024;16:2533–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Yamashita Y, Morimoto T, Amano H, et al. ; COMMAND VTE Registry Investigators. Deep vein thrombosis in upper extremities: clinical characteristics, management strategies and long-term outcomes from the COMMAND VTE Registry. Thromb Res. 2019;177:1–9. [DOI] [PubMed] [Google Scholar]
- [7].Lv J. Risk factors and nursing measures of lower extremity deep vein thrombosis in ICU patients. Electron J Pract Clin Nurs. 2020;5:71–9. [Google Scholar]
- [8].Zhu Y, Duan X. Predictive nursing helps improve treatment efficacy, treatment compliance, and quality of life in unstable angina pectoris patients. Am J Transl Res. 2021;13:3473–9. [PMC free article] [PubMed] [Google Scholar]
- [9].Yu Y, Hu L, Chen X, Ge M, Zhu H, Yan Y. The impact of the predictive nursing education process on degree of comfort and quality of life for patients in the oncology department. Iran J Public Health. 2017;46:1231–6. [PMC free article] [PubMed] [Google Scholar]
- [10].Gong L, Ruan C, Yang X, Lin W. Effects of predictive nursing intervention among patients with acute stroke. Iran J Public Health. 2021;50:1398–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Lanmin Z. Application analysis of predictive nursing guidance inorthopedic nursing. Int J Med Data. 2022;6:81–3. [Google Scholar]
- [12].Liu Y, Liu X. Current situation of application of predictive nursing model in clinical nursing practice. J Gen Nurs. 2015;13:1815–7. [Google Scholar]
- [13].Zhao F, Zhang L, Chen X, et al. Building and verifying a prediction model for deep vein thrombosis among spinal cord injury patients undergoing inpatient rehabilitation. World Neurosurg. 2024;194:123451. [DOI] [PubMed] [Google Scholar]
- [14].Sun J, Li M, Zhou H. The impact of health education based on the pender health promotion model on patients with deep vein thrombosis of the lower limbs treated with rivaroxaban [published online ahead of print June 14, 2024]. Altern Ther Health Med. [PubMed] [Google Scholar]
- [15].Xu D, Zhang W, Wang X, Geng X. Application of the Caprini risk model in preventing deep venous thrombosis formation in patients with nephrotic syndrome. Am J Transl Res. 2024;16:5409–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Chen J, Guo L, Li S, Shi Y. Efficacy and safety of postural intervention on prevention of deep venous thrombosis of lower extremity in postpartum women with pregnancy-induced hypertension: a protocol for systematic review and meta-analysis. Medicine (Baltimore). 2021;100:e24959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Fan C, Jia L, Fang F, et al. Adjunctive intermittent pneumatic compression in hospitalized patients receiving pharmacologic prophylaxis for venous thromboprophylaxis: a systematic review and meta-analysis. J Nurs Scholarsh. 2020;52:397–405. [DOI] [PubMed] [Google Scholar]
- [18].Rappold JF, Sheppard FR, Carmichael Ii SP, et al. Venous thromboembolism prophylaxis in the trauma intensive care unit: an American Association for the Surgery of Trauma Critical Care Committee Clinical Consensus Document. Trauma Surg Acute Care Open. 2021;6:e000643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Min SK, Kim YH, Joh JH, et al. Diagnosis and treatment of lower extremity deep vein thrombosis: Korean practice guidelines. Vasc Specialist Int. 2016;32:77–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Liew NC, Alemany GV, Angchaisuksiri P, et al. Asian venous thromboembolism guidelines: updated recommendations for the prevention of venous thromboembolism. Int Angiol. 2017;36:1–20. [DOI] [PubMed] [Google Scholar]
- [21].Minet C, Potton L, Bonadona A, et al. Venous thromboembolism in the ICU: main characteristics, diagnosis and thromboprophylaxis. Crit Care. 2015;19:287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Lichte P, Kobbe P, Almahmoud K, et al. ; Trauma Register DGU. Post-traumatic thrombo-embolic complications in polytrauma patients. Int Orthop. 2015;39:947–54. [DOI] [PubMed] [Google Scholar]
- [23].Zhu Y, Chen Y. Effect of predictive nursing combined with intermittent inflatable pressure pump on the incidence of deep vein thrombosis in ICU patients. Thromb Hemost. 2021;27:514–5. [Google Scholar]
- [24].Li H, Bao H, Shao L, et al. Application of predictive phased intervention in the prevention of deep venous thrombosis of lower limbs in ICU patients. Chin J Mod Nurs. 2017;23:2361–4. [Google Scholar]
- [25].Barbar S, Noventa F, Rossetto V, et al. A risk assessment model for the identification of hospitalized medical patients at risk for venous thromboembolism: the Padua Prediction Score. J Thromb Haemost. 2010;8:2450–7. [DOI] [PubMed] [Google Scholar]