Development and validation of a clinical practice guidelines implementability assessment tool (CPG-IAT) based on the COSMIN framework Authors: Dongmei Zhong, Gregory A Aarons, Yaolong Chen, Nan Yang, Wenjun He, Run Wang, Jiawen Tang, Yimin Wu, Zizhen Huang, Dong Roman Xu Published: 2026 ABSTRACT AbstractIntroductionWe define guideline implementability as the characteristics of the guideline that reflect the extent to which it is likely to be adopted in clinical practice. Improving the intrinsic quality (e.g., context, format, language etc.) of clinical practice guidelines (CPGs) may be a cost-effective and broadly applicable approach. This study was aimed to develop the clinical practice guidelines implementability assessment tool (CPG-IAT) and test its psychometric properties.MethodsThe study used the 2022 CPGs recorded in the STAR guideline repository as the evaluation sample. The evaluation team consisted of 60 members with clinical, guideline development, or prior rating experience, responsible for assessing the included guidelines. Guideline evaluation data were randomly assigned to be utilized for an exploratory factor analysis (n = 131) or for a confirmatory factor analysis (n = 130). Reliability and validity analyses were then conducted with the full sample.ResultsThe exploratory factor analysis resulted in a 16-item tool with four dimensions representing Methodological Rigor and Transparency, Recommendation Clarity and Interpretability, clinical relevance and actionability. Confirmatory factor analysis supported a priori factor structure. The tool demonstrated excellent internal consistency reliability, convergent validity, construct reliability, split-half reliability, test-retest reliability, inter-rater reliability and pragmatic.ConclusionsThe CPG-IAT provides a psychometrically validated instrument for assessing the intrinsic implementability of CPGs both during and after the guideline development process. The CPG-IAT contributes a systematically developed and empirically validated measure of guideline implementability, with potential to inform future research and practice aimed at enhancing the translation of evidence-based guidelines into clinical practice.Trial registrationChina Clinical Trails Registry (ChiCTR2400086931); registered July 15, 2024. https://www.chictr.org.cn/.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s13012-026-01506-2.Keywords: Clinical practice guideline, Implementability, Assessment, Tool This study empirically validates a multidimensional structure of guideline implementability, moving beyond conceptual discussions by identifying four core and measurable domains-Methodological Rigor and Transparency, Recommendation Clarity and Interpretability, Clinical Relevance, and Actionability. This study empirically validates a multidimensional structure of guideline implementability, moving beyond conceptual discussions by identifying four core and measurable domains-Methodological Rigor and Transparency, Recommendation Clarity and Interpretability, Clinical Relevance, and Actionability. By applying rigorous exploratory and confirmatory factor analytic methods, this study demonstrates how theoretically proposed implementability constructs converge and reorganize when examined through empirical data, providing methodological insight for future guideline measurement research. By applying rigorous exploratory and confirmatory factor analytic methods, this study demonstrates how theoretically proposed implementability constructs converge and reorganize when examined through empirical data, providing methodological insight for future guideline measurement research. The resulting implementability assessment tool translates abstract implementation principles into operational items, offering guideline developers and users a practical instrument to identify modifiable features that are potentially relevant to guideline uptake in clinical practice. The resulting implementability assessment tool translates abstract implementation principles into operational items, offering guideline developers and users a practical instrument to identify modifiable features that are potentially relevant to guideline uptake in clinical practice. Introduction Clinical practice guidelines (CPGs) are developed through a rigorous process of evidence evaluation with the aim of facilitating the implementation of evidence and standardizing best practices among practitioners [1, 2]. The number of guidelines is increasing annually, while low adherence remains a significant challenge [3, 4]. Low adherence is due to a variety of constraints that can be categorized as being related to either external environmental factors (e.g., medical personnel, medical institutions, local policies) and/or intrinsic characteristics (e.g., guideline context, format, language etc.), with the former referring to guideline users, and the latter referring to inherent features associated with the guidelines themselves [5, 6]. Although both external and intrinsic factors are important when seeking to strengthen guideline implementation, many scholars have argued that a focus on improving the intrinsic quality may be a more cost-effective and broadly applicable approach [3, 7]. It is particularly important to identify intrinsic factors, because in many cases they can be ameliorated or fully remedied by guideline authors while the guideline is being developed [8, 9]. Shiffman et al. defined a set of characteristics that predict the relative ease of implementation of guideline recommendations as implementability [8]. We define guideline implementability as the characteristics of the guideline that reflect the extent to which it is likely to be adopted in clinical practice [3]. More attention is needed in regard to clinical practice guidelines implementability as a key step in effective guideline implementation. Contemporary implementation science frameworks have further articulated the role of innovation characteristics in shaping adoption. For example, the Consolidated Framework for Implementation Research (CFIR) includes ‘Relative Advantage,’ ‘Adaptability,’ ‘Trialability,’ and ‘Complexity’ as key innovation-level determinants of implementation [10, 11]. Similarly, the Theoretical Domains Framework (TDF) and Behaviour Change Wheel have highlighted how the characteristics of a healthcare intervention, including clinical guidelines, interact with individual, social, and environmental factors to influence uptake [12, 13]. These frameworks underscore the importance of systematically assessing and optimizing the intrinsic features of CPGs as a mechanism for facilitating implementation. In recent decades, tools and frameworks were developed for clinicians and methodology experts to evaluate the implementability of CPGs. Our recent systematic review identified 15 concepts, theories, models, frameworks, or scales related to guideline implementability [14], including AGREE (Appraisal of Guidelines, REsearch and Evaluation) [15], GLIA (Guideline Implementability Appraisal) [8], GUIDE-IT (Guideline Implementability Tool) [16], GUIDE-M (Guideline Implementability Decision Excellence Model) [17], etc. Based on the results of the systematic review, combined with external consultations and natural language processing tool-assisted, we created a Comprehensive Framework for Guideline Implementability (CFGI) [14]. Furthermore, Kun Zou et al. developed a generalised tool for evaluating the success of the implementation of clinical practice guidelines (A-GIST) [18]. However, few existing tools for evaluating guideline implementability have a clear theoretical basis or have a systematic reliability and validity verification [5]. Given the limitations of existing guideline evaluation tools, guideline implementation can be advanced by integrating new knowledge from implementation science to develop and validate a new guideline implementability evaluation tool underpinned by guideline implementability theory [5]. Previously, we developed the CFGI, which provides a systematic theoretical basis for the implementability of guidelines [14]. Through a factorial experiment, the validity of the constructs within the CFGI framework were validated. The present study aimed to develop a clinical practice guidelines implementability assessment tool (CPG-IAT) based on CFGI that has been verified through empirical research, and systematically evaluate its reliability and validity based on the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) checklist [19, 20]. Methods Item generation Item generation and dimension identification proceeded in five phases. First, based on the previously established CFGI framework [14], an initial item pool covering seven dimensions was developed by integrating evidence from an extensive literature review [10]. The initial item pool was provided as Table S1 in Supplementary 1. The literature review at this stage focused on articles addressing barriers and facilitators of guideline implementation. Second, rapid qualitative interviews were conducted with 15 guideline stakeholders, including 4 nurses, 3 physicians, 4 organizational managers, and 4 guideline methodologists, to gain a deeper understanding of the attributes and characteristics of guideline implementability. Third, a two-round Delphi process was conducted to screen and refine the items [21, 22] (the Delphi form was presented in Supplementary 2). In total, 90 stakeholders from Asia, Africa, Europe, and North America participated, including 38 healthcare professionals, 29 healthcare organization managers, 15 implementation science experts, and 8 guideline methodologists. In Round 1, the initial pool of 46 items across 7 dimensions was sent to experts, who rated each item on five criteria (relevance, importance, clarity, feasibility, and necessity) using a 5-point Likert scale and provided open-ended suggestions. The response rate was 80.4% (90/112), the expert authority coefficient (Cr) was 0.89, and Kendall’s W ranged from 0.287 to 0.522 (all P < 0.001). Items were retained if the mean score was ≥ 3.5 and the full-score ratio was ≥ 0.2; items with a coefficient of variation (CV) > 0.25 were flagged for deletion. Based on quantitative thresholds and 139 expert suggestions, the tool was reduced from 7 dimensions and 46 items to 6 dimensions and 36 items. In Round 2, the revised tool was redistributed using the same rating criteria; the response rate was 63.3% (57/90), all CV values fell below 0.25, and Kendall’s W improved to 0.844–0.855 (all P < 0.001), indicating strong consensus. Based on 41 additional expert suggestions, the tool was further reduced to 5 dimensions and 25 items. No systematic differences in rating patterns were identified across participant types; divergent qualitative feedback was adjudicated by the research team. Detailed Delphi results, including item-level statistics for both rounds, are presented in Supplementary 3. Fourth, potential items were reviewed by the investigative team and program managers for face validity and content validity. Twenty-six items were developed that represented five potential content dimensions of guidelines implementability: clarity, applicability, reliability, readability, and transparency. Fifth, a pilot survey using the tool to rate 30 CPGs was conducted to estimate the item discrimination, dispersion, correlation coefficients, and internal consistency [23]. Items with low relevance or poor consistency were removed to generate a pragmatic and optimized version of the tool [24]. The purpose was to refine items and improve the tool structure based on feedback from experts and guideline users. During the pilot survey, we removed the items with a resolution coefficient < 0.50, a correlation coefficient < 0.40, and a CITC value < 0.20. Finally, the first version of the CPG-IAT with five dimensions and 21 items, were developed (see Supplementary 4). Guidelines and participants The Scientific, Transparent, and Applicable Rankings (STAR) tool [25] has applied to evaluate Chinese CPGs and consensus published in 2020, 2021, and 2022 [26–28]. CPGs included in the most recent STAR assessment cycle served as the sampling frame for this study [29]. As of October 24, 2024, the STAR Working Group had completed evaluations of guidelines published in 2022. To identify eligible guidelines and consensus, the STAR methodological rating group conducted systematic searches on August 24, 2022, and January 15, 2023. Chinese-language searches were performed in the China National Knowledge Infrastructure (CNKI), Wanfang Data Knowledge Service Platform, the Chinese Biomedical Literature Database (CBM), and the Chinese Medical Journals Database, while English-language searches were conducted in PubMed and Web of Science. The corresponding search periods were January 1 to July 31, 2022, and January 1 to December 31, 2022. In addition, supplementary searches were carried out in Hong Kong Journals Online, the Hong Kong and Macao Journals Network, the Macao Journal Paper Index, and the Taiwan Journal Paper Index [25]. The evaluation working group consisted of a Steering Committee, a Secretariat, an Evaluation Panel, and a Quality Control Group. The Steering Committee comprised two members and was responsible for developing the evaluation framework and designing specific assessment procedures, overseeing the evaluation process, and providing methodological consultation and guidance. The Secretariat included two members and was responsible for supporting the Steering Committee in the organization and management of the evaluation activities, as well as for collecting, organizing, and archiving evaluation materials. The Evaluation Panel was composed of 60 individuals with clinical, guideline development, and/or prior rating experience, responsible for assessing the included guidelines. The Quality Control Group consisted of three authors and was responsible for reviewing and approving the evaluation results. Procedure The study was approved by the appropriate Institutional Review Boards prior to participants recruitment and informed consent was obtained prior to administering surveys. Members of the specialty committees within the working group completed an online training course and reviewed training materials on the CPG-IAT. Following the training, participating members underwent an assessment and certification process; only those who passed the certification were eligible to participate in the evaluation. Members of the Evaluation Panel independently assessed each included guideline using the CPG-IAT. Based on the full text of the guideline and supplementary materials, evaluators judged whether each item’s criteria were met, selected an item score on a 5-point Likert scale (ranging from 1 = strongly disagree to 5 = strongly agree), and documented the rationale for each rating in the evidence form. Upon completion of all evaluations, members of the Quality Control Group, with assistance from the Secretariat, reviewed the submitted materials to check the consistency between scores and supporting rationales, as well as to identify any internal inconsistencies or contradictions. If issues were identified, the materials were returned to the evaluators for revision. After independent evaluations were completed, the Secretariat cross-checked the ratings from two evaluators and calculated the intraclass correlation coefficient (ICC) to assess inter-rater agreement. When the ICC was ≤ 0.75, the guideline was assigned to a third evaluator by the Secretariat, and the ICC was recalculated based on the three sets of ratings. When the ICC exceeded 0.75, the final score for the guideline was calculated as the mean of the three evaluation results [30]. Statistical analyses Data were randomized to be utilized for either the EFA (exploratory factor analysis) (n = 131) or CFA (confirmatory factor analysis) (n = 130). Exploratory factor analysis was used to derive and evaluate the factor structure of the tool using IBM SPSS. Principal axis factoring was selected for factor extraction because it allows for consideration of both systematic and random error and Promax oblique rotation was utilized for factor rotation as we assumed that derived factors would be correlated [31]. Item inclusion or exclusion was based on an iterative process in which items with relatively low primary loadings (e.g., < 0.40) or high cross-loadings (e.g., > 0.30) were removed [31]. The number of factors to be retained was determined based on parallel analysis, factor loadings, and interpretability of the factor structure as indicated in the rotated solution. Parallel analysis is among the better methods for determining the number of factors based on simulation studies [32]. Parallel analysis was based on estimation of 1000 random data matrices with values that correspond to the 95th percentile of the distribution of random data eigenvalues [33, 34]. The random values were then compared with derived eigenvalues to determine the number of factors. Confirmatory factor analysis was performed to validate the factor structure of the CPG-IAT developed in exploratory factor analysis using AMOS 26.0 (IBM Corp., Armonk, NY, USA). Model fit was assessed using several empirically supported indices: the comparative fit index (CFI), the Tucker-Lewis index (TLI), the root mean square error of approximation (RMSEA), and the standardized root mean square residual (SRMR). CFI and TLI values greater than 0.90, RMSEA values less than 0.10, and SRMR values less than 0.08 indicate acceptable model fit [35–38]. If necessary, bootstrap Bollen-Stine P would be utilized to assist confirmatory factor analysis when the model fit was not ideal [39, 40]. The average variance extracted (AVE) was calculated using the output of the confirmatory factor analysis [41, 42]. Internal consistency, construct reliability, Spearman-Brown split-half reliability and test-retest reliability were analysed for the total sample during the formal investigation stage [5]. Internal consistency was evaluated using Cronbach’s alpha along with McDonald’s omega, with values ≥ 0.70 considered acceptable [43]. The test-retest reliability was calculated by randomly selecting 50 guidelines from the total sample, and having different evaluators conduct the evaluation two weeks later. The inter-rater reliability was calculated by randomly selecting 2 evaluators to assess 10 guidelines. The psychometric and pragmatic evidence rating scale (PAPERS) was used to evaluate pragmatic of the CPG-IAT, and the evaluation content included normativity, cost of use, language, training cost, and length [44]. Results The information of guidelines A total of 291 guidelines were included in both the pilot survey and the formal investigation stage. Regarding specialty distribution, the included guidelines covered 30 medical specialties. Oncology had the highest number of included guidelines (n = 36), while pathology, reproductive medicine, and cardiothoracic surgery had the fewest, with one guideline each (see Fig. 1). The medical specialties covered by the guidelines included in the evaluation Exploratory factor analysis An iterative approach was taken to conducting the factor analyses and item reduction. In the first iteration and consistent with our hypotheses, five factors were specified and all 21 items were included. The EFA results indicated that only one item met the factor loading criteria for the proposed fifth factor (loading > 0.40 on the primary factor and < 0.30 loading on any secondary factors) [31]. Factors defined by a single item are considered poorly defined and psychometrically unstable [31]; therefore, the fifth factor was not retained. The parallel analysis also suggested a four factor solution. Thus, we conducted the next EFA specifying four factors. The results suggested the removal of 5 items because of low primary factor loadings and/or high cross loadings. Thus, 16 items and 4 factors were retained in the final EFA. Table 1 displays the factor means, item means, initial eigenvalues, variance accounted for by each factor, internal consistency reliabilities, and rotated factor loadings. Internal consistencies were high, ranging from 0.746 to 0.919. All items demonstrated acceptable corrected item-total correlations (CITC ≥ 0.30). Deletion of any individual item did not substantially improve Cronbach’s alpha, supporting the retention of all items for subsequent analyses. As shown in Table 2, factor correlations ranged from 0.14 to 0.64, suggesting that the dimensions represent related but relatively distinct aspects of guideline implementability. Subscale labels were created based on an examination of the items and factor loadings presented in Table 1. Factor 1 was labeled ‘Methodological Rigor and Transparency’ as it indicated the methodological foundations of guideline development, including transparent definition of target users and populations, explicit evidence selection criteria, and systematic processes for formulating and updating recommendations.Factor 2, ‘Recommendation Clarity and Interpretability’ reflects how recommendations are structured, articulated, and presented to end users, encompassing clarity of recommendation identification, evidence linkage, language, and layout. Factor 3, ‘Clinical Relevance’ represents the extent to which guidelines address clearly defined clinical questions and provide recommendations that directly respond to those questions. Factor 4, ‘Actionability’ indicates whether guideline recommendations offer concrete implementation steps and clearly defined application contexts to support real-word use. The item explanations and examples of the final version of the CPG-IAT are presented in Supplementary 5. Table 1Clinical practice guidelines implementability assessment tool and item statisticsTool, items, and totalMeansdevvαEFA factor loadings1234 1. Methodological Rigor and Transparency 14.77 7.21 7.6746.00.919G01. Clearly identifies the intended users and target population3.101.34 0.702 -0.1220.0120.085G02. Clearly describes the criteria for evidence selection2.341.39 0.879 0.0020.002-0.014G03. Evidence has been systematically searched2.211.44 0.916 0.022-0.077-0.020G04. Mentions plans or processes for guideline updating2.291.55 0.604 0.0410.1320.113G05. Provides a detailed description of the process used to develop recommendations2.371.42 0.913 -0.002-0.041-0.015G06. Discloses key steps in the guideline development process2.461.39 0.807 0.0800.013-0.056 2. Recommendation Clarity and Interpretability 23.87 5.62 2.1111.40.858G07. Recommendations in the guideline text are clearly identifiable3.501.430.076 0.722 0.047-0.096G08. There is a clear linkage between recommendations and supporting evidence3.981.260.094 0.622 0.1880.061G09. Provides the quality of evidence and strength of recommendation for each item3.391.750.326 0.484 -0.0420.018G10. The wording of recommendations is clear and concise4.430.86-0.098 0.767 0.0060.278G11. The content is complete and logically organized4.240.910.026 0.677 0.201-0.196G12. The layout is well-designed and facilitates quick reference and use4.330.87-0.074 0.870 -0.097-0.052 3. Clinical Relevance 8.12 2.22 1.235.40.868G13. Clearly defines the specific clinical questions3.971.210.0730.056 0.807 -0.026G14. Recommendations adequately address the clinical questions4.151.16-0.0830.067 0.888 0.078 4. Actionability 8.54 1.77 0.964.30.746G15. Recommendations provide practical steps for application4.211.050.055-0.2770.206 0.778 G16. The guideline specifies the implementation settings or applicable contexts4.340.940.0020.244-0.177 0.800 Tool total 55.3113.720.920Note: N = 261 for means and standard deviations; n = 131 for other EFA derived statistics; sd= Standard deviation; ev= initial eigenvalue; v= variance accounted for before rotation; α = Cronbach’s alpha; bold font for means and sd indicate the overall tool mean and sd; bold font for EFA factor loadings indicates the tool on which the items load Clinical practice guidelines implementability assessment tool and item statistics Note: N = 261 for means and standard deviations; n = 131 for other EFA derived statistics; sd= Standard deviation; ev= initial eigenvalue; v= variance accounted for before rotation; α = Cronbach’s alpha; bold font for means and sd indicate the overall tool mean and sd; bold font for EFA factor loadings indicates the tool on which the items load Table 2Clinical practice guidelines implementability assessment tool factor intercorrelationsFactor12341. Methodological Rigor and Transparency1.002. Recommendation Clarity and Interpretability0.621.003. Clinical relevance0.510.641.004. Actionability0.140.420.331.00Note: N = 131; All correlations, p < 0.001 Clinical practice guidelines implementability assessment tool factor intercorrelations Note: N = 131; All correlations, p < 0.001 Confirmatory factor analysis Confirmatory factor analysis was used with a sample independent of the EFA sample to evaluate the factor structure identified in the EFA. The model was modified using the bootstrap Bollen–Stine P method [39, 40]. The factor model demonstrated good fit as indicated by multiple fit indicators (n = 130; CFI = 0.99; TLI = 0.99; RMSEA = 0.03). The SRMR value was 0.11, indicating that the model did not fully reproduce the observed correlation structure. Nevertheless, given the satisfactory performance of incremental fit indices such as CFI, TLI and RMSEA, the overall model fit was considered acceptable. Figure 2 displays the standardized factor loadings for the factor model, ranging from 0.373 to 0.904 and all factor loadings were statistically significant (p’s < 0.001). McDonald’s omega coefficients indicated good to excellent internal consistency for all four factors, ranging from 0.74 to 0.92, with Cronbach’s alpha values similarly ranging from 0.746 to 0.920. Confirmatory factor analysis factor loadings for the clinical practice guidelines implementability assessment tool. Note: n = 130; All factor loadings are standardized and are statistically significant, p < 0.001; CFI = 0.99; TLI = 0.99; RMSEA = 0.03; SRMR = 0.11. F1 Methodological Rigor and Transparency, F2 Recommendation Clarity and Interpretability, F3 Clinical relevance, F4 Actionability Convergent validity Confirmatory factor analysis was performed to validate the factor structure obtained from exploratory factor analysis. Convergent validity, expressed as average variance extracted (AVE), ranged from 0.530 to 0.743 across dimensions. Reliability analysis Construct reliability (CR) for each dimension ranged from 0.617 to 0.935, all above the minimum threshold of 0.6. The coefficient of the split-half reliability was 0.810, indicating good split-half reliability. For the 50 guidelines assessed, ICC values ranged from 0.825 to 0.999, all > 0.75, demonstrating good test-retest reliability. Two randomly selected raters independently evaluated 10 guidelines using the tool, yielding a Pearson correlation of 0.794 (P = 0.006), indicating significant correlation and high inter-rater agreement, confirming good inter-rater reliability. Pragmatic evaluation The tool demonstrated good normativity, showing concentration and distribution characteristics when applied to samples of ≥ 100 guidelines (see Table 1). The tool will be freely available to all researchers. It is comprehensible to individuals with at least a bachelor’s degree. Training is required but minimal, indicating acceptable assessor burden. The tool produces quantifiable scores with clear cut-offs, handles missing data via assessor follow-up, and can automatically calculate scores, reflecting minimal workload for evaluators. Finally, the tool contains 16 items, indicating an appropriate length. Discussion The current study describes the development of the measure of guideline implementability for guideline implementation, the CPG-IAT. We used an iterative process to develop items representing guideline implementability and then used quantitative data reduction techniques to develop a brief measure that may be easily and efficiently used for research and applied purposes. The CPG-IAT has the potential to optimize guidelines from within, by providing criteria and evaluation tools for assessing implementability before, during and after the guideline development process. Although we originally proposed five factors of guideline implementability, quantitative analyses supported a four-factor model. Items related to evidence handling, methodological processes, and transparency clustered into a single Methodological Rigor and Transparency factor, while items concerning readability and usability converged into Recommendation Clarity and Interpretability. In contrast, items addressing the definition of clinical questions and the extent to which recommendations address those questions formed a distinct Clinical Relevance factor, and items focusing on implementation steps and applicable contexts were retained as Actionability. This four-factor solution demonstrated acceptable model fit and internal consistency in CFA, offering a more parsimonious and empirically supported representation of guideline implementability while preserving its key conceptual components. Transparency did not emerge as an independent factor because items related to transparency were empirically inseparable from methodological rigor, reflecting that, in practice, transparent reporting of development processes is closely intertwined with perceptions of evidence quality and rigor rather than representing a distinct construct. The CPG-IAT demonstrated strong internal consistency reliability, convergent validity, construct reliability, split-half reliability, test-retest reliability, inter-rater reliability and pragmatic. The inter-rater reliability findings (ICC ≥ 0.75) reflect agreement on verifiable, document-level characteristics of guidelines rather than implying that implementability is a context-free property. We acknowledge that implementability is partly constituted by the interaction between a guideline’s intrinsic features and the conditions of the target implementation setting, including healthcare system structures, resource availability, and professional norms. As such, some degree of variation in CPG-IAT scores across jurisdictions or healthcare contexts may reflect genuine context-sensitivity rather than measurement error, a perspective consistent with implementation science frameworks such as CFIR [10, 11], which recognizes the fit between an innovation and its context as a key determinant of implementation success. Future research should examine whether CPG-IAT scores vary systematically across international or health-system contexts, and whether certain domains are more sensitive to contextual interpretation than others. Given that the CPG-IAT is brief (i.e., 16 items), administration and use in guideline implementability evaluation can be very efficient with little respondent burden. The practicality of this brief scale is consistent with calls for measures that can be utilized in real-world settings where the efficiency of the research process is paramount [45]. The CPG-IAT shares a common objective with existing implementability appraisal instruments such as GLIA [8], GUIDE-IT [16], GLAFI (Guideline Language and Format Instrument) [46], and tool developed by Jin et al. [47], all of which aim to identify factors that influence the uptake of CPGs in clinical practice. However, the CPG-IAT advances the field in several important ways. First, different with GLIA, which is primarily grounded in Shiffman’s implementability framework and focuses on decidability and executability at the recommendation level [8], the CPG-IAT is built upon a more comprehensive theoretical foundation that integrates the CFGI framework, and extensive literature review, and multi-round expert consensus [14]. Second, different with many previous instruments that have not undergone comprehensive psychometric validation, the CPG-IAT has been systematically tested for reliability, validity, and pragmatic utility across diverse guideline types. In addition, compared with A-GIST [18], which evaluates the actual outcomes of guideline implementation based on the Reach Effectiveness Adoption Implementation Maintenance (RE-AIM) framework [48], the CPG-IAT focuses on assessing the intrinsic implementability of guidelines before, during, and after development. Together, these two tools form a complementary methodological landscape: the CPG-IAT addresses “how to make guidelines easier to implement,” whereas A-GIST examines “how successfully guidelines have been implemented.” The CPG-IAT also complements the AGREE Recommendations EXcellence (AGREE-REX) tool, which evaluates the quality of guideline recommendations with respect to clinical applicability, values and preferences, and implementability at the recommendation level [49]. However, AGREE-REX focuses on the quality of individual recommendations, the CPG-IAT provides an assessment of implementability across the entire guideline document, enabling a more integrated appraisal of whether a guideline as a whole is ready for real-world uptake. There are several strengths of CPG-IAT. First, the tool is grounded in a comprehensive theoretical framework that integrates conceptual, empirical, and expert-based evidence, ensuring its content is both systematic and aligned with established implementation science principles. Second, the tool has undergone rigorous psychometric testing-including validity, reliability, and pragmatic assessments-ensuring it is both methodologically sound and applicable across diverse types of CPGs. Third, its design supports use throughout the guideline development lifecycle, enabling developers, reviewers, and implementers to systematically identify potential implementability gaps at multiple stages of guideline development. Together, these strengths position the CPG-IAT as an evidence-based instrument with broad potential applicability that can inform the development of high-quality, implementable CPGs. Despite the strengths, there are still some limitations. First, although the tool was developed through a rigorous process-including literature review, expert consensus, and psychometric testing-the evaluation relied primarily on CPGs published in 2022 from a single guideline repository. This may limit the generalizability of the findings to guidelines produced in different years, or development systems. Second, the assessment focuses on intrinsic features of guideline documents and development processes rather than contextual or system-level determinants of implementation. As a result, while the tool effectively evaluates document-level implementability, it does not replace organizational- or system-level readiness assessments. Third, although reliability, validity, and pragmatic utility were systematically examined, further testing in real-world implementation projects and across international guideline settings is needed to confirm cross-cultural applicability and predictive validity. Finally, as with most rating tools, the CPG-IAT involves a degree of subjectivity, and inter-rater consistency may vary depending on assessors’ experience and familiarity with guideline methodology. In summary, the CPG-IAT provides a systematically developed and psychometrically validated instrument for assessing the implementability of CPGs at the guideline document level. Although further cross-context validation is needed, the CPG-IAT represents a robust foundation for strengthening guideline quality and supporting future efforts to enhance the translation of evidence into practice. Conclusion A validated CPG-IAT, grounded in the COSMIN framework and informed by the CFGI, has been developed to appraise the intrinsic implementability of CPGs. CPG-IAT provides a structured and balanced assessment of guideline Methodological Rigor and Transparency, Recommendation Clarity and Interpretability, clinical relevance, and actionability, offering a document-level evaluation that precedes real-world implementation. With its solid theoretical foundation, rigorous psychometric validation, and broad applicability across different types of CPGs, the CPG-IAT equips guideline developers, researchers, implementers, and policymakers with a practical tool to identify potential barriers embedded within guideline documents, inform efforts to strengthen guideline quality before dissemination, and contribute to the evidence base for strategies aimed at improving future uptake in clinical practice. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Supplementary Material 5 Acknowledgements We acknowledge all members in Acacia Lab for Implementation Science for their cooperation in this study. Author contributions ZDM conceptualized the paper, assisted with data analysis, developed the study methods, and drafted, reviewed, and edited the manuscript. GAA developed the study methods, reviewed and edited the manuscript. CYL reviewed and edited the manuscript. YN reviewed and edited the manuscript. HWJ reviewed and assisted with data analysis. WR reviewed and edited the manuscript. TJW reviewed and edited the manuscript. WYM assisted data collection. HZZ reviewed the manuscript. XD conceptualized the overall study, obtained funding, and edited and reviewed the manuscript. All authors read and approved the final manuscript. Funding This project is funded through the following competitive grants: The Swiss Agency for Development and Cooperation (#81067392) and the China Postdoctoral Science Foundation (Grant No. 2025M780687). Data availability All data generated or analysed during this study are included in this published article and its supplementary information files. Declarations Ethics approval and consent to participate This study was approved by the Institutional Review Board (IRB) of Southern Medical University (#Southern Medical Audit (2024) No. 012). All the participants signed the informed consent form. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Supplementary Material 5 Data Availability Statement All data generated or analysed during this study are included in this published article and its supplementary information files.