This study develops a preliminary indicator system framework for the county-level integration of HIV medical and preventive services, guided by the Valentijn Rainbow Model and refined through focus group discussions (FGDs) [12]. The Rainbow Model provides a comprehensive theoretical foundation, emphasizing person-centered and population-centered integration. It comprises 6 dimensions: system integration (macrolevel), organizational and professional integration (mesolevel), clinical integration (microlevel), and normative and functional integration, which facilitate coordination across all levels [13]. This model aligns well with the practical need for seamless medical-preventive service integration in HIV care, ensuring a holistic and patient-centric approach.

The Rainbow Model provides a structured perspective on integrated care, emphasizing that successful service integration requires multilevel coordination. System integration ensures policy alignment and resource allocation at the macro level; organizational and professional integration at the meso level enhances interdisciplinary collaboration and institutional efficiency, while clinical integration at the micro level focuses on seamless patient care. Additionally, normative integration (shared values and goals) and functional integration (harmonized workflows and IT systems) serve as critical connecting mechanisms across levels. By adopting this model, the study acknowledges the complexity of health care systems and provides a structured method to evaluate HIV service integration at the county level [14].

Textbox 1 outlines the framework of the indicator system for the county-level integration of HIV medical and preventive services, based on the Valentijn Rainbow Model. This model highlights the coordination across multiple levels—system, organizational, professional, and clinical—and categorizes indicators into primary dimensions such as policy support, health care accessibility, and service coordination. These indicators are crucial for evaluating the effectiveness of integrated HIV services in rural, county-level regions.

To refine the framework, FGDs were conducted, engaging 3 public health professors, 10 HIV health care professionals, and 7 graduate researchers. Participants engaged in 3 rounds of discussion, with graduate students systematically reviewing PubMed, Web of Science, China National Knowledge Infrastructure (CNKI), and Wanfang to gather evidence on HIV policies and intervention [33]. The discussion process facilitated an evidence-driven and consensus-based refinement of the tertiary indicators, ensuring that the evaluation framework is both empirically grounded and contextually relevant. The resulting tertiary indicator system is detailed in Textbox 1.

The application of FGDs enhances the practical relevance and stakeholder alignment of the indicator system [33,34]. By incorporating experts with diverse backgrounds, this study ensures that the framework reflects real-world health care challenges and policy priorities. The rationality of the theoretical approach lies in its ability to bridge theory and practice, using the Rainbow Model structured framework to guide integration while refining the indicators through expert consensus. This approach enhances the validity, feasibility, and applicability of the proposed indicator system, making it a valuable tool for evaluating and improving HIV service integration at the county level.

The Delphi method, developed by the RAND Corporation [35] in the 1950s, is a structured communication technique that gathers and refines expert opinions through multiple rounds of questionnaires, aiming to achieve consensus on complex issues [36]. In public health, this method is instrumental in areas where empirical evidence is scarce or inconclusive, facilitating the development of guidelines and evaluation frameworks [37,38]. For instance, it has been used to define criteria for selecting peer educators in HIV management in China, ensuring that the chosen individuals are well suited to current practices.

In the context of HIV prevention and treatment in rural China, the Delphi method has been used to develop performance evaluation systems for the CDC at various administrative levels [39]. By engaging experts in iterative consultations, this approach ensures that the resulting evaluation indicators are both contextually relevant and scientifically robust, addressing the unique challenges faced in underdeveloped regions [40].

Participants engaged in 3 rounds of discussion, with graduate students systematically reviewing PubMed, Web of Science, CNKI, and Wanfang to gather evidence on HIV/AIDS policies and intervention strategies, including those related to HIV service delivery during the COVID-19 pandemic, with a focus on key populations in diverse settings. The search results were systematically analyzed, and experts with a high volume of publications in high-impact journals and a research focus aligned with the study objectives were selected as consulting experts. In addition, we included professionals with more than 5 to 10 years of experience from the Health Commission and CDC [41], as well as medical personnel involved in AIDS prevention and treatment (Textbox 1). Two rounds of consultation questionnaires were then distributed via email. Upon receipt of the feedback, comprehensive data analysis and consistency tests were performed to validate the responses.

The Delphi questionnaire is designed to develop a 3-tier indicator system for HIV prevention and treatment in China’s counties and rural areas. Its purpose is to achieve consensus on key themes and indicators and to assess the degree of consistency across various focus areas.

Recruitment for this study will involve sending emails and letters to 100 targeted individuals. Additionally, recruitment efforts will be extended through networks across China, patient-oriented research support groups, and academic institutions via word-of-mouth and strategic email campaigns. The primary participants will include staff from health and wellness committees, disease control centers, university professors and researchers, and medical personnel involved in AIDS prevention and control.

The use of financial incentives (500 RMB [US $30]) in this study is intended to motivate expert participation and compensate for their time and effort, especially given the multiple rounds of feedback required by the Delphi method. This approach helps ensure a high response rate and minimizes participant attrition. However, there is a potential risk of bias, as experts may feel subtly influenced by the incentive. To mitigate this, it is important to emphasize that participation is voluntary, with a clear assurance that responses will remain anonymous and independent.

To mitigate potential respondent loss, the sample size will be expanded to 200 during the first round of recruitment.

The selection criteria for experts in this study were based on the recommendations proposed in 1975 by Delbecq, Van de Ven, and Gustafson regarding the selection of Delphi research participants. These recommendations include (1) senior management decision-makers who will use the outcomes of the Delphi study, (2) professionals and their support teams, and (3) respondents to the Delphi questionnaires whose opinions are being solicited [42].

Accordingly, we selected 3 key categories of experts in the field of HIV prevention and control: government officials, health care professionals, and academic researchers. Among them, government officials from disease prevention and control agencies were required to encompass all administrative levels, including the Chinese Center for Disease Control and Prevention (China CDC), provincial CDCs, municipal CDCs, and county-level CDCs. Health care professionals were drawn from tertiary hospitals, township health centers, and community health service centers, ensuring representation across different tiers of the health care system. Efforts were made to achieve a balanced distribution of experts across these various levels [40].

To enhance the representativeness and authority of the expert panel, we established specific inclusion criteria tailored to each category of experts. Determining the appropriate sample size in Delphi studies is crucial for achieving a reliable consensus. While there is no universally prescribed number, research suggests that a sample size of 10 to 15 participants may be sufficient if the expert panel is homogeneous [43]. However, for multistakeholder Delphi surveys, a minimum of 60 to 80 participants is recommended to ensure high replicability of results. Considering the scope and diversity of our study, we aimed to include 60 to 75 experts, balancing feasibility with the need for diverse and comprehensive input [44].

This structured and deliberate approach to expert selection is designed to ensure that the Delphi process yields robust, credible, and actionable insights for advancing HIV prevention and control efforts (Table 1)

Subsequent validation of the evaluation framework will include pilot testing in 2 hospitals in Yunnan Province, involving 5 physicians in each hospital. This small-scale, real-world application will test the framework’s feasibility and allow for the collection of feedback from health care professionals. Presurveys will also be conducted to assess the initial perceptions and readiness of the health care staff. Based on the pilot results and feedback, adjustments will be made before broader implementation. The framework will then be evaluated through field implementation across additional county-level areas to assess its effectiveness, applicability, and impact on improving health care service integration.

The project was reviewed and approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology (ID: 21YJA630062). This study strictly adheres to the ethical principles of the Declaration of Helsinki. Full voluntary informed consent was obtained from all participants, with clear disclosure of data collection purposes, scope, and secondary use rights, as well as the right to withdraw without prejudice. Collected data was de-identified and encrypted to protect privacy and the study. Participants were compensated up to 500 RMB (US $30).

The research findings will be presented at international conferences and published in reputable international academic journals. Additionally, a policy report on the “China County AIDS Prevention and Control Evaluation Effect Indicator System” will be submitted to the relevant stakeholders.

Based on the WHO guidelines for HIV, we identified key research themes during each round of our Delphi expert consultation. As shown in Figure 1, our process involved 3 rounds of letter-based consultations that ensured active expert participation and refined the indicators until a practical, consensus-based index system was achieved.

The study will be conducted from April 1, 2025, to January 31, 2026. The first to third rounds of questionnaires will be distributed using Tencent Documents, from May to September 2025, and qualitative interviews will be conducted using Zoom (Zoom Communications, Inc) and telephone consultation. Respondents who wish to receive a printable version of the questionnaire can choose to receive a PDF version of the questions. The members of the research group will submit a reply through Tencent Documents. Quantitative and qualitative data analysis will be carried out between the 3 rounds of surveys, and the development of the indicator system and the dissemination of the final report will be completed by January 2026.

The first round of the Delphi study aims to refine the evaluation framework for the county-level integration of HIV medical and preventive services by eliminating nonessential indicators and dimensions. A panel of 60 to 75 experts is invited to assess the preliminary indicator framework, providing suggestions for additions, deletions, and modifications based on practical experience, theoretical analysis, peer knowledge, and intuitive judgment. Experts rate the importance of each indicator on a 1 to 5 scale, with higher scores indicating greater importance. Their feedback is used to calculate engagement, authority, coordination, and consensus levels, guiding the selection of indicators and ensuring iterative refinement.

Statistical analysis is conducted using SAS (SAS Institute Inc) to compute the mean, SD, coefficient of variation, and Kendall W for each indicator, assessing the degree of expert agreement. Nonparametric tests are applied to determine statistical significance (P<.05), indicating whether expert opinions are significantly divergent. Additionally, R (R Foundation for Statistical Computing) is used for analytic hierarchy process (AHP) to derive indicator weights. The consistency ratio is calculated to validate expert agreement in the AHP process, with CR<0.1 indicating acceptable consistency.

Findings from this round will be used to refine the evaluation framework, ensuring that the selected indicators are scientifically robust and relevant to assessing the integration of medical and preventive services for HIV. Expert feedback and statistical validation contribute to the systematic development of a comprehensive and evidence-based assessment system.

The second round of the study focused on identifying the key themes and indicators prioritized by the participants in the first round. During this round, participants were tasked with selecting a predetermined number of topics across multiple dimensions, including prevention and health education for potential patients, diagnosis and treatment, health intervention for patients, and the interface between prevention and treatment. The specific number of topics selected for each category was determined based on insights from the first round, and a weighting calculation was applied to determine the final selection of topics and indicators in similar categories. Topics and indicators that received the highest participant selection advanced to the next round.

Cronbach α was used to assess the internal consistency of the responses in each round. Additionally, 2 members of the research team analyzed the free-text comments provided in the questionnaires to identify relevant changes in the survey topics. These findings were reported in detail to the advisory group, which evaluated the need to add or modify topics and indicators accordingly.

In the third round of the Delphi expert consultation, efforts were made to ensure that the evaluation index system for regional HIV medical and preventive integration was comprehensive, practical, and refined. The research team conducted a collective review of the primary and secondary topics related to HIV integration at the county level, aiming to minimize bias and capture all key concepts to enhance the scientific rigor and practical applicability of the indicator system.

First, indicators and dimensions that received little attention from experts in the first and second rounds were eliminated. Second, the questions were revised to include clear definitions of medical terms and, where possible, to use simpler language, facilitating the implementation of the indicator system in practical evaluations. Finally, participants were informed that they could complete the round in multiple sessions, save their progress, and would receive a reminder if their questionnaire remained incomplete after 1 week.

The purpose of the third round is to accept suggestions on the modification of the index system already obtained by experts in the first 2 rounds, accept feedback, and resolve the uncertainty of the index system determined by the first 2 rounds of expert review. The research team analyzes the advice provided by 2 master’s and PhD experts in order to inform any final adjustments if necessary. These changes will be reviewed and finalized by the research team of 7 master and doctoral students and 2 associate professors. The task force will use these findings to develop a countywide AIDS prevention and control indicator system that best matches actual needs.

This study aims to systematically develop an evaluation framework for county-level HIV service integration, contributing to both theoretical and methodological advancements in the field. While previous research has explored various metrics for assessing HIV-related health outcomes [45], diagnosis, and treatment effectiveness, no prior study has provided a comprehensive, empirical framework specifically for evaluating the integration of HIV prevention and medical care at the regional level [46]. The key contributions of this study are as follows.

First, this study uses the Rainbow Model as the theoretical foundation to construct an integrated health care service framework, developing a 3D indicator system encompassing medical services, public health interventions, and social support [47]. This model emphasizes a multilayered integration of prevention and treatment, facilitating a more comprehensive assessment of HIV service effectiveness at the country level. Second, the study engages a diverse group of stakeholders, including policymakers, researchers, and health care professionals, ensuring that the evaluation framework incorporates multidisciplinary perspectives and remains relevant to real-world implementation [48]. Finally, this study uses a mixed-methods approach by integrating FGD, the Delphi method, and AHP, which helped reduce participant attrition while enhancing the interpretability and practical applicability of the findings.

Despite its contributions, this study has several limitations. First, the developed framework is designed to assess the overall effectiveness of integrated HIV service delivery at a regional level rather than evaluating the impact of specific health education or intervention programs. Therefore, its primary utility lies in policy and system-level evaluations rather than in assessing individual interventions. Second, the literature review is conducted primarily using Chinese-language databases, and the recruited experts are predominantly from China, which may limit the generalizability of the framework to other geographical contexts. Future research should incorporate international datasets and expert input to improve the global applicability of the framework. Third, although the Delphi method and AHP enhance consensus-building, expert judgments remain inherently subjective, potentially influencing the selection and weighting of indicators. Selection bias in expert recruitment may also limit the comprehensiveness of the identified indicators, which is a known limitation of the Delphi process.

The effectiveness of the framework will be measured through both qualitative and quantitative assessments. During the pilot testing phase, we will gather feedback from health care professionals and administrators regarding the framework’s feasibility, ease of use, and impact on service integration. Additionally, key performance indicators such as patient satisfaction, service coordination, and health care outcomes will be tracked during the field implementation phase. We will use statistical methods, including pre- and postimplementation comparisons, to evaluate the framework’s effectiveness in improving HIV service delivery at the county level. We have now explicitly included this in the methods section for clarity.

The findings of this study will have significant implications for HIV prevention and treatment policies and implementation science. First, this study refines the conceptual framework of HIV service integration, providing a structured approach to enhancing the continuity of care and comprehensive lifecycle services for people living with HIV [48]. Second, by identifying key indicators, this study develops a robust evaluation framework specifically tailored for resource-limited settings, offering policymakers a practical tool for assessing the effectiveness of integrated HIV interventions. Finally, by engaging 75 leading HIV prevention and treatment experts in China, this study contributes to evidence-informed policymaking and promotes cross-disciplinary collaboration in HIV service delivery.