Construction of a Follow-up Evaluation Index System for Patients with Venous Thromboembolism Based on the Donabedian Model: A Modified Delphi Study Authors: Qian Zhang, Qingmei Niu, Yihua Zhang, Yatian Jia Published: 2026 ABSTRACT AbstractBackgroundFollow-up care for patients with venous thromboembolism lacks standardization and a comprehensive quality assessment system. There is an urgent need to improve management measures to enhance the quality of care.ObjectiveThis study developed a comprehensive evaluation index system to standardize and guide clinical nursing practice in the follow-up management of patients with venous thromboembolism, aiming to enhance long-term care quality.MethodsBased on the Donabedian model, we created a preliminary questionnaire through literature review and group discussions. The study employed a modified Delphi method with 20 multidisciplinary experts over two consultation rounds, combining expert feedback with the analytic hierarchy process (AHP) to determine indicators and their weights.ResultsThe consultation achieved a 100% response rate in both rounds. Statistical analysis showed satisfactory consensus with Kendall’s harmony coefficients of 0.32 and 0.33, and high expert authority coefficients of 0.91 and 0.95. The final system comprises 3 primary indicators, 8 secondary indicators, and 37 tertiary indicators.ConclusionIn this study, we developed a comprehensive evaluation index system for the follow-up management of patients with venous thromboembolism based on the Donabedian model. This system provides a structured framework to help standardize follow-up management; however, its feasibility, reliability, and clinical effectiveness require further prospective validation in clinical practice.Keywords: venous thromboembolism, patient follow-up, delphi technique Introduction Venous thromboembolism is the third most common vascular disease worldwide, characterized by high mortality, morbidity, and recurrence. Anticoagulation reduces the risk of venous thromboembolism recurrence.1–3 Most venous thromboembolism patients still need long-term anticoagulation after discharge, but patients generally have poor adherence to anticoagulant medication after discharge, and up to 30% of post-discharge venous thromboembolism patients develop post-thrombotic syndrome. This not only imposes a substantial financial burden on patients but also seriously affects their quality of life.4–6 Most developed countries have established well-established discharge follow-up systems, with various hospital-based and community-based follow-up management service models.7,8 Studies have shown that various follow-up methods are currently used in hospitals in China. Among them, the most widely used follow-up methods include discharge education, telephone follow-up, Internet platform follow-up, and specialty outpatient follow-up.9–11 Although follow-up management has been implemented to some extent, it has not yet been systematized, and the mechanism for evaluating service effectiveness is inadequate.12–14 A comprehensive evaluation of the quality of care is essential to ensure the quality of follow-up care for patients with venous thromboembolism. Quality-of-care evaluation is a systematic process and an important tool for quality control.15 The establishment of a scientific and systematic quality assessment index system for the follow-up of patients with venous thromboembolism is an important issue of common concern to nurse managers. At present, there is no standardized quality evaluation index system for the follow-up of patients with venous thromboembolism, and there is a lack of unified guidance on the content and method of the visit, as well as a lack of standardized continuity-of-care processes, documentation methods, and outcome evaluation standards for patients with venous thromboembolism. Existing VTE follow-up pathways and management recommendations mainly focus on clinical management, such as anticoagulant therapy, recurrence prevention, and complication screening; however, they do not provide a structured and quantifiable framework for evaluating the quality of follow-up management itself. In contrast, the proposed evaluation index system is designed to systematically assess the structure, process, and outcome of VTE follow-up care, which enables the quality of follow-up to be measured, compared across institutions, and continuously improved in a standardized manner. This lack of standardization and evaluation hinders the improvement of follow-up management quality of patients with venous thromboembolism.15 Therefore, there is an urgent need to establish a comprehensive quality evaluation system to monitor the quality of follow-up care for patients with venous thromboembolism. The Donabedian model is a widely used international quality evaluation model, first proposed by Avedis Donabedian in 1966, consists of three dimensions—structure, process, and outcome—which are interrelated.16 Although this model was introduced to China in the 1980s, it has only begun to be applied in clinical nursing quality evaluation in recent years, expanding from general hospital nursing quality assessment to specialty-specific nursing quality evaluation.17–19 It can be used to assess the quality and effectiveness of the implementation of follow-up management, and to provide theoretical guidance for the improvement of follow-up management. In summary, given the limited number of studies on the follow-up management of patients with venous thromboembolism and the comprehensiveness of the Donabedian model in quality assessment. However, existing VTE follow-up pathways and management recommendations primarily focus on clinical management aspects such as anticoagulant therapy, recurrence prevention, and complication screening, but do not provide a structured and quantifiable framework for evaluating the quality of follow-up management itself. The proposed index system differs by systematically assessing the structure, process, and outcome of VTE follow-up care, enabling follow-up quality to be measured, compared, and improved. This study used the Donabedian model as the theoretical framework, adopted the modified Delphi method, and constructed a follow-up management index system for patients with venous thromboembolism after hospital discharge through a multidisciplinary approach. This study aims not only to meet the post-discharge care needs of patients with venous thromboembolism, reduce the risk of complications, and improve patients’ quality of life, but also to provide a basis for further exploring follow-up management pathways that align with chronic disease prevention and control needs in China. Methods The Delphi method is commonly employed to gather expert opinions on real-world issues. This technique addresses the limitations of individual expert judgment by aggregating collective expert opinions.20 The Delphi process involves sending multiple rounds of questionnaires to a panel of experts. In this study, a modified Delphi approach was utilized to seek expert input on follow-up management indicators for patients with venous thromboembolism.21 The research was divided into two stages: (1) Preliminary study: A draft of follow-up management indicators for venous thromboembolism patients was created through a literature review combined with clinical experience. (2) Delphi stage: A Delphi questionnaire was designed based on the initial draft, and two rounds of surveys were conducted to achieve consensus among the experts. This study was reported in line with the Delphi Study Implementation and Reporting (CREDES) recommendations22 (see Supplementary Table 1). Supplementary Figure S1 shows the flow chart of the Delphi process. Preliminary Formulation of Index Entries A group consisting of a chief physician, an associate chief nurse, three charge nurses, and three graduate students was established. Relevant systematic reviews, meta-analyses, and original articles were retrieved from PubMed, Medline, Web of Science, Cochrane Library, China National Knowledge Infrastructure and Wanfang databases to determine the keywords for the literature search. Based on the literature search, the synonymous search terms were continuously expanded, and the main search terms were “venous thromboembolism”, “deep vein thrombosis”, “pulmonary thromboembolism” or “pulmonary embolism” and “follow-up”, “evaluation index”. Studies published in English or Chinese on the follow-up management of patients with venous thromboembolism were included, and journal articles published between 2013–2023 were searched, including original research papers and review articles. The flowchart for searching literature and screening care indicators is shown in Supplementary Figure S2. Initially, 5052 records were identified through the database search. After removing duplicates and screening titles and abstracts, 857 articles remained for full-text review. Following full-text assessment, 80 articles were included for indicator extraction. Based on these 80 articles, the research group extracted relevant follow-up management indicators and, combined with the Donabedian model, clinical experience, and relevant guidelines, developed the preliminary index system comprising 3 first-level, 8 second-level, and 40 third-level indicators.23 Development of the Expert Consultation Questionnaire We designed the consultation questionnaire, which was divided into 3 parts, including the questionnaire description, the basic information questionnaire for experts, and the consultation form for the follow-up management of patients with venous thromboembolism. The questionnaire description explains the study background, purpose, and completion instructions. The expert background questionnaire collects experts’ demographic information, the basis for their judgments, and their degree of familiarity with the indicators. The consultation form for follow-up management of patients with venous thromboembolism contains 3 levels of indicators, a 5-point Likert scale and a column for modification (Supplementary Table 2). Delphi Process This study covered 20 experts from 12 tertiary-level A hospitals in 7 provinces and autonomous regions, including Beijing, Shanghai, Guangdong, Shandong, Hebei, Shanxi, and Inner Mongolia, who participated in 2 rounds of expert consultation. The inclusion criteria of the experts were: engaged in venous thromboembolism prevention, health care, treatment, rehabilitation, or nursing for more than 15 years, with extensive clinical experience; hospital health service managers with extensive management experience; holding a master’s degree or higher; holding the academic rank of associate chief nurse or higher; familiar with the Delphi method, interested in and willing to participate in this study, and able to complete all rounds of consultation within the study timeframe. Based on the study’s requirements, 20 experts were selected who completed both rounds of consultation. The sample size of 20 experts is consistent with the Delphi methodological guidelines, which indicate that the number of participants in Delphi studies commonly ranges from 15 to over 60; a panel of this size is therefore considered sufficient to achieve reliable consensus in modified Delphi studies.24 Consultation questionnaires were distributed via email, WeChat, or in paper form, and experts were asked to respond within 2 weeks. After the first round of consultation questionnaires was collected and summarized, based on the experts’ opinions, the group revised the indicator system after discussion and formed the next round of consultation questionnaires. The same method was used for the next round of consultation until the experts’ opinions converged. Through two rounds of consultation, consistent results were obtained. Indicators were included based on the following criteria:18 (1) a mean importance rating greater than 4.0 on the 5-point Likert scale; (2) a coefficient of variation less than 0.25; and (3) a full-score frequency greater than 20%. An indicator was retained only when it satisfied all three criteria simultaneously. Indicators that failed to meet one or more of these criteria were not automatically excluded; instead, they were reviewed by the research group together with the experts’ written comments, and a final decision on whether to retain, delete, or merge each indicator was made through collective discussion that took into account both the statistical criteria and the experts’ opinions.25 When Kendall’s coefficient of agreement (W) reaches statistical significance (p < 0.05), it indicates a high degree of consistency among the expert group, and thus consensus is considered to have been reached. After determining the final indicators, the AHP was applied to determine the relative importance of the first- and second-level indicators. The 20 participating experts were invited to compare each pair of indicators at the same level using Saaty’s 1–9 scale26 (Table 1), in which a score of 1 indicates equal importance and 9 indicates that one indicator is extremely more important than the other. The pairwise comparisons formed a positive reciprocal judgment matrix A = (aij)n×n, where aij denotes the relative importance of indicator i over indicator j, satisfying aij × aji = 1 and aii = 1. The judgment matrices were analyzed using SPSSAU software, and the weight of each indicator was calculated as the normalized eigenvector corresponding to the maximum eigenvalue (λmax) of each judgment matrix. To assess the consistency of each judgment matrix, the consistency index (CI) was calculated as CI = (λmax − n)/(n − 1), and the consistency ratio (CR) was calculated as CR = CI/RI, where RI is the random consistency index determined by the matrix order n; a CR value below 0.1 indicates acceptable consistency. In this study, the CR values of all first-level judgment matrices were 0, and the CR of the second-level judgment matrices ranged from 0 to 0.07, indicating acceptable consistency for all matrices. The weights of the three first-level indicators (structure, process, and outcome) were calculated and are presented in the Results section. For the third-level indicators, the rank-order weighting method was used. The 20 participating experts ranked the third-level indicators under each second-level indicator in descending order of importance and assigned scores according to their ranking positions (the most important indicator receiving the highest score and the least important receiving 1). The scores assigned by all experts were then summed for each third-level indicator, and the weight of each third-level indicator was calculated as its summed score divided by the total summed score of all third-level indicators under the same second-level indicator. The composite (overall) weight of each indicator was calculated by multiplying its own weight by the weights of its corresponding higher-level indicators, so that the composite weights of all indicators at the same level sum to 1.0; higher composite weights indicate greater importance of the indicator.27 Table 1Characteristics of Experts (N=20)Characteristicsn (%)Age 35~39 years7 (35.0) 40~60 years13 (65.0)Position Chairman of section4 (20.0) Head nurse and head nurse of the department16 (80.0)Title Senior level6 (30.0) Associate senior level14 (70.0)Educational background Doctoral degree4 (20.0) Master’s degree16 (80.0)Working experience 16~20 years4 (20.0) 21~30 years13 (65.0) 30 years above3 (15.0) Characteristics of Experts (N=20) All completed questionnaires were double-checked for completeness and accuracy. Following data entry and statistical analysis, the results were checked for logical and practical significance to avoid errors in data entry or statistical analysis and to identify other operational errors. To further ensure the content quality and rigor of the final index system, the final version was reviewed by 2 external experts in nursing education. Data Analysis Microsoft Excel 2013 was used for data entry and SPSS 26.0 was used for statistical analysis, and the indicators were descriptively analyzed using means, full-score frequencies, proportions, coefficients of variation, and standard deviations. Experts’ demographic characteristics were described using frequencies and proportions. Expert motivation was reflected by the questionnaire response rate. The degree of authority of experts was expressed using the authority coefficient. The degree of agreement among experts was assessed using Kendall’s coefficient of concordance (W), and p <0.05 was used to indicate that the difference was statistically significant. The SPSSAU software was used to calculate the weights of the indicators and to perform the consistency test. Matrices were considered acceptable when the consistency ratio was < 0.1. Ethical Consideration This study was approved by the Medical Ethics Committee of Shanxi University of Chinese Medicine (Approval Number: 2023LL068). All participants signed the informed consent form. The research team clearly informed them that throughout their participation, personal information and related data would be processed in an anonymized manner, and the confidentiality of information and data would be fully guaranteed throughout the process. Participants were informed that the data would only be used for academic purposes of this study, that they had the right to refuse to participate or withdraw from the study at any time, and that they would not be subject to any adverse effects as a result. Results Sample Characteristics The general information about the experts is shown in Table 2. Table 2Scale for the Basis of Expert JudgmentBasis of JudgmentDegree of ImpactHighMediumLowPractical experience0.50.40.3Theoretical analysis0.30.20.1Knowledge of literature0.10.10.1Intuitive feeling0.10.10.1 Scale for the Basis of Expert Judgment Motivation of Experts and Harmonization of Opinions A total of 20 questionnaires were distributed in each of the two rounds, and all were returned, yielding a 100% response rate in both rounds. This indicates that the experts were highly motivated for this study. The degree of agreement among experts was assessed using Kendall’s coefficient of concordance (W); the higher the coefficient, the greater the agreement. The Kendall’s harmony coefficient of the first round of expert consultation was 0.32 (χ2 = 320.54, p <0.001), and the Kendall’s harmony coefficient of the second round of expert consultation was 0.33 (χ2 = 305.10, p <0.001), indicating an acceptable degree of agreement among experts. Two rounds of expert consultation were conducted, as specified a priori in the study protocol. The Delphi process was considered complete when the indicator set had stabilized and expert consensus had been achieved. After the first round, the indicator set was revised from 40 to 37 tertiary indicators—six were deleted and three were added—based on the experts’ written comments and the predefined consensus criteria. In the second round, no further deletions, additions, or modifications were proposed, and the Kendall’s coefficient of concordance was statistically significant in both rounds (W = 0.32 and 0.33, respectively; both p < 0.001), indicating meaningful agreement among the experts. The process was therefore terminated after two rounds.24 Expert’s Authority The authority coefficient was calculated using the formula: authority coefficient = (judgment coefficient + familiarity coefficient)/2, with values greater than 0.7 indicating high expert credibility.26 The judgment coefficient reflects the basis of the expert’s judgment; Supplementary Table 2 shows the weighting assignments of the expert judgment coefficients. The familiarity coefficient represents the degree of familiarity of the expert with the content of the study, scored as 1.0 (very familiar), 0.8 (relatively familiar), 0.5 (moderately familiar), 0.2 (slightly familiar), or 0.0 (unfamiliar). The judgment coefficient of the experts in round 1 was 0.94, the familiarity coefficient was 0.87, and the authority coefficient was 0.91. The judgment coefficient of the experts in round 2 was 0.97, the familiarity coefficient was 0.94, and the authority coefficient was 0.95. This indicates that the experts who participated in this study had a high level of authority in the field. Relative Importance of Quality Indicators Indicator weights were calculated using the analytic hierarchy process and percentage weighting. The consistency ratio values for both primary and secondary indicators were <0.1, indicating acceptable weighting results. The weights of the primary indicators were 0.34 (“structural quality”), 0.37 (“process quality”), and 0.30 (“outcome quality”). The weights of the secondary indicators ranged from 0.07 to 0.22. For the tertiary indicators, the weights ranged from 0.01 to 0.06. Results of Two Rounds of Expert Enquiry The preliminary questionnaire comprised 3 first-level, 8 second-level, and 40 third-level indicators (Supplementary Table 3). After the first round of expert consultation, the initial indicator system was revised to 3 first-level, 8 second-level, and 37 third-level indicators. In the first round of expert consultation, a total of 11 experts proposed changes, which involved the deletion of 6 and the addition of 3 third-level indicators (as shown in Table 3). Table 3Revisions to the First Round of Expert Correspondence and Results of the discussionSerial NumberCorrespondence for AmendmentsResults1Deletion of the tertiary indicator at the structural level, “Green lanes for follow-up of patients discharged with VTE” (Experts 1, 12)Accept2Delete the tertiary indicator “Video outreach: multimedia, mobile TV” at the process level (Expert 10)Accept3Deletion of tertiary indicator “health lectures: health lecture halls” at the process level (Experts 1, 9)Accept4Deletion of the tertiary indicator “Community education: involvement of community members” at the process level (Expert 3)Accept5Addition of a tertiary indicator at the process level “Motivational interviewing: post-motivational interviewing of patients for sensitization” (Experts 5, 11)Accept6Addition of a tertiary indicator at the process level: “Teach-back method: teaching back knowledge of patient health management” (Expert 6)Accept7Delete the tertiary indicator “Out-of-hospital Self-Management Behavior Scale for Patients with VTE (self-administered)” from the outcome level (Experts 4, 8)Accept8Deletion of the tertiary indicator “Self-Assessment Scale of Anxiety (SAS)” from the outcome level (Experts 1, 2)Accept9Addition of “medication adherence scale” to the tertiary indicators at the outcome level (Experts 1, 2)Accept Revisions to the First Round of Expert Correspondence and Results of the discussion Based on the selection criteria for the indicator entries, combining expert opinions and group discussions, an indicator system containing 3 first-level, 8 second-level, and 37 third-level indicators was finalized. The importance values, weights, and coefficients of variation of each indicator after the second round of expert consultation are presented in Table 4. Table 4Quality Management Indicator System for Follow-up Management of Patients with Venous ThromboembolismIndicators I, II and III (Weights)SDMCVPercentage ofthe Full ScoreComposedWeightI Structural level (0.34)0.414.800.0980% II-1 Environmental assessment (0.24)0.414.200.1020%0.08 III-1 The hospital has departments related to VTE diagnosis and treatment such as respiratory, vascular surgery, emergency medicine, imaging, and testing (0.28)0.444.750.0975%0.02 III-2 Independent or affiliated anticoagulation and thrombosis clinics (0.29)0.474.700.1070%0.02 III-3 Fixed follow-up offices (0.23)0.554.750.1280%0.02 III-4 Fixed places for health education (0.21)0.414.200.1020%0.02 II-2 Resource security (0.60)0.514.450.1145%0.20 III-5 Establishment of a follow-up management team with specialized follow-up staff (0.24)0.414.200.1020%0.05 III-6 Disease education materials (bulletin boards, brochures, boards, etc.) (0.24)0.444.750.0975%0.05 III-7 Follow-up information system (0.23)0.474.700.1070%0.05 III-8 Follow-up Service Record Form (0.30)0.444.750.0975%0.06 II-3 Patient follow-up files (0.16)0.494.650.1165%0.05 III-9 Patient general information assessment (diagnosis, site of thrombus, typing, staging, factors predisposing to thrombosis) (0.18)0.224.950.0595%0.01 III-10 Hospitalization programs (medication, surgery) (0.17)0.414.800.0980%0.01 III-11 Date and number of follow-up visits (0.15)0.504.600.1160%0.01 III-12 Follow-up patient checkups (0.16)0.494.650.1165%0.01 III-13 Results of tests and examinations for follow-up patients (0.17)0.374.850.0885%0.01 III-14 Recommendations for follow-up instructions (0.18)0.444.750.0975%0.01I Process level (0.37)0.224.950.0595% II-4 Mode of discharge follow-up (0.25)0.414.800.0980%0.09 III-15 Telephone follow-up (0.35)0.414.800.0980%0.03 III-16 WeChat, SMS follow-up (0.36)0.444.750.0975%0.03 III-17 Appointments, follow-up visits to follow-up clinics (0.30)0.554.100.1320%0.03 II-5 Follow-up management content (0.52)0.554.250.1330%0.19 III-18 Symptom monitoring and management (0.11)0.494.350.1135%0.02 III-19 Complication monitoring and management (0.13)0.504.600.1160%0.02 III-20 Guidance on the use of medication (0.11)0.504.600.1160%0.02 III-21 Guidelines for care of the affected limb (0.13)0.444.750.0975%0.02 III-22 sports coaching (0.10)0.374.850.0885%0.02 III-23 Lifestyle guidance (diet, clothing, etc.) (0.11)0.494.650.1165%0.02 III-24 Antithrombotic compression stockings instructions for use (0.12)0.374.850.0885%0.02 III-25 psychological counselling (0.08)0.224.950.0595%0.02 III-26 Guidance for follow-up consultations (0.11)0.524.200.1225%0.02 II-6 Health promotion methods (0.23)0.494.650.1165%0.09 III-27 Graphic communication: oral health communication using communication materials (0.18)0.414.200.1020%0.02 III-28 Motivational interviewing: preaching to patients after motivational interviewing (0.19)0.494.150.1220%0.02 III-29 Teaching back method: teaching back the patient’s knowledge of health management (0.19)0.604.400.1445%0.02 III-30 Peer education: inviting friends with VTE for evangelization (0.16)0.614.500.1355%0.01 III-31 Family education: inviting the patient’s family for family education (0.15)0.514.550.1155%0.01 III-32 “Key-to-button” education: using the Internet (microblogs, public numbers, etc.) for education (0.13)0.664.300.1540%0.01I Outcome level (0.30)0.414.800.0980% II-7 Disease progression outcomes (0.76)0.224.950.0595%0.22 III-33 Rate of unplanned readmission of patients (0.55)0.444.750.0975%0.13 III-34 Villalta score (post-thrombotic syndrome classification score) (0.45)0.444.750.0975%0.10 II-8 Patient self-report (0.24)0.604.400.1445%0.07 III-35 Venous Disease Specific Quality of Life Questionnaire (CIVIQ) (0.38)0.504.600.1160%0.03 III-36 VTE knowledge questionnaire (0.33)0.474.300.1130%0.02 III-37 Medication Adherence Scale (0.28)0.314.900.0690%0.02Abbreviations: SD, Standard Deviation; M, Mean; CV, Coefficient of Variation; VTE, Venous Thromboembolism; I, Primary Indicator; II, Secondary Indicator; III, Tertiary Indicator; SMS, Short Message Service; App, Application; PTS, Post-Thrombotic Syndrome; CIVIQ, Chronic Venous Insufficiency Quality of Life Questionnaire. Quality Management Indicator System for Follow-up Management of Patients with Venous Thromboembolism Abbreviations: SD, Standard Deviation; M, Mean; CV, Coefficient of Variation; VTE, Venous Thromboembolism; I, Primary Indicator; II, Secondary Indicator; III, Tertiary Indicator; SMS, Short Message Service; App, Application; PTS, Post-Thrombotic Syndrome; CIVIQ, Chronic Venous Insufficiency Quality of Life Questionnaire. Discussion Venous thromboembolism is the third most common vascular disease worldwide, and patients require long-term anticoagulation after hospital discharge. However, poor medication adherence leads to post-thrombotic syndrome in many post-discharge patients, imposing a financial burden and seriously affecting quality of life. Although developed countries have established well-established post-discharge follow-up systems, follow-up methods in China remain diverse and the evaluation mechanism is inadequate, hindering improvements in follow-up management quality. Therefore, there is an urgent need to establish a comprehensive quality evaluation system for VTE follow-up management. The structure, process and outcome indicator content of this study were developed based on Chinese venous thromboembolism management standards and relevant expert consensus. Two rounds of expert consultation were conducted in strict accordance with the modified Delphi method. A response rate of 80% or higher is generally considered acceptable for Delphi studies.27 The high response rate and the experts’ detailed suggestions for questionnaire modification indicate that the experts actively engaged with and supported this study, and the results are reliable. In this study, 20 expert consultants were selected from a total of 12 tertiary hospitals in 7 provinces and autonomous regions of China, all of whom participated in 2 rounds of expert consultation. The experts mainly consisted of vascular surgery nurse leaders and directors of nursing departments with extensive experience in the field of venous thromboembolism follow-up care and management. The authority coefficients of the experts were 0.91 and 0.95, indicating that the experts’ opinions were well represented, and Kendall’s W was statistically significant in both rounds (p <0.001), indicating reliable consultation results. Furthermore, this study used the Donabedian “structure–process–outcome” model as the framework to ensure the comprehensiveness of the VTE follow-up management index system, and quantified the subjective judgments of experts by combining the modified Delphi consultation with the analytic hierarchy process to enhance the scientific rigor of the study findings.28 Based on Donabedian’s “structure-process-outcome” model, this study, through literature analysis and clinical experience, developed an assessment index system for the follow-up management of patients with venous thromboembolism. It covers 3 main indicators, 8 secondary indicators, and 37 tertiary indicators. This system clearly defines the key points and key links in the follow-up management of patients with venous thromboembolism, and may have potential for clinical application, although its feasibility and effectiveness require further validation. The indicator weights reflect the relative priorities assigned by the expert panel, rather than direct evidence that higher-weight indicators are objectively more effective or clinically more important, in the follow-up management of venous thromboembolism in clinical practice. The Donabedian “structure–process–outcome” model was selected because it is the most widely used international framework for health care quality assessment and comprehensively covers the three core dimensions of follow-up management (resource allocation, service delivery, and patient outcomes). Compared with other quality frameworks, this model provides a clear, measurable logical chain from structure to process to outcome and can potentially be operationalized into clinical nursing quality indicators, which made it particularly suitable for constructing the follow-up evaluation system for patients with VTE. The high-weight indicators at the structural level focus on the content related to resource guarantee, among which the establishment of a professional follow-up team and the standardization of follow-up record forms are the highest priority. This weight distribution feature essentially stems from the fact that the structural level is the fundamental support for ensuring the quality of follow-up. A professional VTE follow-up management team is the key component for delivering high-quality follow-up services. It can provide specialized guidance for key needs such as the use of anticoagulant drugs, limb care, and lifestyle adjustments of patients, directly addressing the key clinical challenge of poor treatment adherence.29 The standardized follow-up record forms are also an important prerequisite for achieving continuity and individualization of follow-up. By systematically recording key information of patients’ diagnosis and treatment, they provide precise basis for subsequent adjustments of follow-up plans and avoid fragmentation of follow-up work. Moreover, the consensus of experts on the assessment of patients’ basic information is also evident. This is because comprehensively understanding patients’ individual circumstances is a prerequisite for delivering precise follow-up assessments.30,31 Insufficient basic information results in less targeted follow-up guidance. The low-weight indicators at the structural level are represented by fixed health education venues. This result is closely related to the current development trend of clinical health education models. With the increasing popularity of online education methods, conducting health education through online platforms such as WeChat official accounts and online courses has become the mainstream. It has the advantages of convenience and strong accessibility, significantly reducing the reliance on physical venues. Although fixed health education venues can still meet some offline education needs, in the current diversified education model, they are no longer a priority for resource allocation. Therefore, their weight is relatively low. Post-discharge follow-up methods are the highest-weight indicators at the process level, with telephone, WeChat, and short message service follow-up as core components—making process quality the most valued of the three dimensions. This characteristic underscores the critical role of timely and convenient follow-up communication in venous thromboembolism management. Venous thromboembolism is associated with high recurrence and mortality rates; long-term post-discharge anticoagulation therapy is key to preventing recurrence, yet poor patient adherence to anticoagulation remains a widespread clinical issue. Telephone, WeChat, and short message service follow-up can overcome time and geographical barriers to provide real-time monitoring and guidance. They not only remind patients to take medications and attend regular check-ups but also promptly detect changes in symptoms and adverse reactions, strengthen nurse–patient communication, and build trust—thereby enhancing overall follow-up effectiveness.32,33 Peer education also garners relatively high recognition among health education methods. Patients with similar disease experiences are more likely to connect emotionally, and the practical management insights they share are more readily accepted—effectively improving the relevance and effectiveness of health education.34 Psychological counseling and follow-up visit guidance constitute the primary low-weight indicators at the process level. This weight distribution stems from the specific nature of these indicators: they are not universal needs for all venous thromboembolism follow-up patients, but rather apply to specific groups—such as patients with severe post-thrombotic syndrome, those experiencing significant psychological stress, or individuals unclear about subsequent diagnosis and treatment processes. Experts generally agree that these indicators have a limited scope of application and are not core universal links in follow-up management, leading to their relatively low weight and varying expert priorities regarding their importance. Disease progression-related outcomes dominate the high-weight indicators at the outcome level, with unplanned patient readmission rates and post-thrombotic syndrome grading assessments as the core focus. This reflects clinical consensus on the primary goal of follow-up management: improving patient disease prognosis. Unplanned readmission rates directly reflect follow-up effectiveness—insufficient follow-up increases the risk of emergency hospitalization due to disease recurrence or deterioration,35 making this indicator a key benchmark for evaluating follow-up quality. Post-thrombotic syndrome is a common long-term complication of venous thromboembolism that significantly impacts patients’ quality of life; grading assessments of this condition directly reflect the effectiveness of follow-up management in preventing complications and enhancing long-term patient quality of life. Both indicators are considered particularly important by the expert panel, reflecting their high weights.36 Patient-reported outcomes—including quality of life scores, venous thromboembolism knowledge assessments, and medication adherence scales—make up the majority of low-weight indicators at the outcome level.37 These indicators carry relatively low weight due to their high subjectivity: patient self-reports are easily influenced by individual cognitive levels, emotional states, and reporting biases, making it difficult to accurately and objectively reflect the true effectiveness of follow-up management. Therefore, in outcome assessments, these indicators serve more as supplementary to objective outcome measures rather than core evaluation criteria, leading to their relatively low weights. Strengths and Limitations This study has some distinct advantages. First, this study constructed quality of care indicators for venous thromboembolism patients discharged for follow-up rather than for hospitalized venous thromboembolism patients, which may contribute to improving the quality of posthospital continuity of care for venous thromboembolism patients. In addition, the modified Delphi method enhances the reliability of the consensus process. The high response rates and authority coefficients in both rounds demonstrate strong expert engagement and support the reliability of the findings for vascular surgery nursing practice. However, this study still has some limitations. Due to limited research funding and time, questionnaires were distributed to 20 experts from different regions of China. The number of experts was sufficient, and the expert group was reliable and representative, but the findings may not be fully representative of all of China. Second, the indicators were generated entirely within the context of Chinese clinical practice, based on Chinese venous thromboembolism management standards and the opinions of Chinese experts. Therefore, the applicability of both the indicator content and its weights to other healthcare settings or cultural populations remains uncertain and requires further evaluation. Third, the evaluation index system was developed entirely based on expert consensus and has not yet been tested in routine clinical practice. It has also not been evaluated against patient outcomes (eg, VTE recurrence or unplanned readmission rates) or service outcomes (eg, follow-up completion rates or patient satisfaction). Therefore, its feasibility, reliability, and effectiveness in improving follow-up quality in real-world practice require prospective validation in future studies. Conclusion In this study, an evaluation index system for the follow-up management of patients with venous thromboembolism was constructed under the guidance of Donabedian’s “structure-process-outcome” framework, based on literature review, expert consultation, and the analytic hierarchy process. The indicator system demonstrated satisfactory expert consensus and reliability. This evaluation system may provide a structured framework for standardizing the follow-up management of patients with venous thromboembolism. However, as the index system constructed in this study is still at the theoretical stage and has not been validated in real-world clinical practice, future prospective studies are needed to evaluate its feasibility, reliability, and clinical applicability, and to further improve the system to enhance its clinical value. Implications for Nursing Management Patients with venous thromboembolism require long-term anticoagulation therapy after discharge; however, their quality of life is significantly impacted by poor adherence to treatment. Although various follow-up approaches for venous thromboembolism patients are employed in China, the absence of a standardized and comprehensive quality evaluation system underscores the urgent need for improvements in management practices and enhancements in nursing quality. The follow-up management indicator system developed in this study is designed to meet this need and may serve as a preliminary reference framework for assessing and improving the quality of VTE follow-up management. Acknowledgments We acknowledge the respondents who completed the questionnaire in this study. Funding Statement This work was supported by the “136 Medical Promotion Project Research Fund” of Shanxi Province (Grant No. 2019XY023). The funder had no role in the design of the study, collection, analysis and interpretation of data, writing of the manuscript, or the decision to submit the paper for publication. Data Sharing Statement The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure The authors report no conflicts of interest in this work. References Associated Data Data Availability Statement The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.