Cardiovascular‐Kidney‐Metabolic Syndrome: Conceptualising an Approach to Health Economic Modelling Authors: Volker Foos, Jieling Chen, Phil McEwan, Marc Evans, Pardeep S. Jhund, Paul R. Kalra, Hongtu Zhu, David C. Wheeler, Geraint Roberts, Martin R. Cowie, Andrew H. Briggs Published: 2026 ABSTRACT ABSTRACTAimsCardiovascular‐kidney‐metabolic (CKM) syndrome has developed into a global public health crisis. Although integrated care is increasingly used to treat patients with CKM, current health economic models fail to capture the value of considering the CKM syndrome as a unified condition. We present a conceptual framework to guide development of a new health economic model that captures value holistically across interconnected CKM domains.Materials and MethodsFollowing best‐practice guidelines, we undertook a review of published literature to identify and connect the key biological drivers of the CKM syndrome. These insights were synthesised into a biological framework, which was simplified into a streamlined influence diagram of sufficient complexity to guide model development. Both framework and diagram were iteratively reviewed and developed based on feedback from clinicians, health economists and modelling specialists.ResultsA biological framework and influence diagram were developed to illustrate how the CKM syndrome's biological drivers interconnect and link to key clinical outcomes. We detail the components of these diagrams, explaining the rationale behind connections. We discuss how the proposed framework could inform a model for future health economic evaluation.ConclusionThere is a clear unmet need for comprehensive modelling of the CKM syndrome to support evidence‐based decision‐making and guide management of this highly prevalent, far‐reaching disease. As more evidence has become available, the mathematical development, in the near future, of a long‐term, holistic and flexible health economic model has become feasible, despite some gaps. This work seeks to shape a shared understanding of the challenges and opportunities in this developing field. Introduction Increasing understanding of the associations between cardiovascular disease (CVD, including heart failure across the ejection fraction spectrum, atrial fibrillation, coronary heart disease, stroke and peripheral artery disease), chronic kidney disease (CKD) and metabolic diseases led the American Heart Association (AHA) to define the cardiovascular‐kidney‐metabolic (CKM) syndrome [1, 2]. CKM is conceptualised as a progressive syndrome, with excess/dysfunctional adipose tissue as the most common underlying pathophysiology (stage 1). A complex system of multidirectional relationships drives individuals through stage 2 (metabolic risk factors, CKD), stage 3 (subclinical CVD) and stage 4 CKM (clinical CVD). Obesity (body mass index [BMI] ≥ 30 kg/m2) is a well‐established risk factor for type 2 diabetes (T2D) [3], CKD [4] and CVD [5]. Dysfunction in either the heart or kidneys may cause or exacerbate dysfunction in the other organ system (the bidirectional ‘cardiorenal syndrome’) [6]. However, pathologic adiposity alone does not fully explain the burden of the CKM syndrome [1]; other disease processes (e.g., inflammation, oxidative stress, hypertension), as well as genetics and environmental factors, all contribute [1]. Epidemiological trends [7, 8, 9, 10, 11] and resource constraints have made optimal management of the CKM syndrome a public health priority. In response to the interconnected nature of the diseases within the CKM syndrome, many disease‐specific guidelines emphasise holistic management [6, 12, 13, 14]. Pharmacotherapeutic advances have yielded drug classes with efficacy across multiple CKM components, with hundreds of novel therapies in development [15]. However, while clinical practice has advanced, health economic modelling has lagged behind. Current reimbursement decisions rely largely on single‐indication models. These may treat cardiometabolic conditions either completely separately, as sequentially linked but without feedback loops, for example, the FINE‐CKD model, which omitted the possibility of cardiovascular events influencing CKD progression and vice versa [16]; or may not consider other conditions at all. Illustrating this latter point, only two of the five obesity‐focused models that engaged in the 2025 Mt. Hood challenges (an annual health economics/simulation modelling conference focused on diabetes and obesity modelling) modelled all CKD stages [17, 18]. In addition, single‐indication models may not appropriately capture heterogeneity across the CKM syndrome, nor how multimorbidity affects event risk and quality‐of‐life, and may only model individuals from a relatively late disease stage. Such models fail to capture the full spectrum of therapeutic benefit across organ systems in people with CKM. This has implications with respect to cost‐effectiveness estimates for therapies with multiple treatment effects, but also impacts modelling done in support of broader health policy. There is an unmet need for a credible multi‐indication model that holistically reflects the CKM syndrome (including bidirectional relationships, feedback loops and the impact of multimorbidity) and supports evidence‐based public health decision‐making. The primary objective of this study was to develop a conceptual health economic model capable of holistically describing CKM syndrome pathophysiology and predicting the long‐term outcomes of management strategies across the syndrome [1, 2] to provide insight, support public health decision‐making and guide clinical guideline development. To ensure practical relevance, the model must be flexible, allowing for the integration of additional outcomes and complexity where biologically plausible and supported by evidence. This flexibility is essential for accommodating emerging therapies and evolving clinical needs. Methods We undertook a two‐step process (‘problem conceptualisation’ and ‘model conceptualisation’) aligned with best‐practice guidelines from the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) [19, 20]. An overview of our methodology for developing, refining and validating the conceptual CKM syndrome health economic model is summarised in Figure 1. Overall flow diagram of conceptual model development. Abbreviations: CKM, cardiovascular kidney metabolic syndrome. *Two‐stage process (‘problem conceptualisation’ and ‘model conceptualisation’), based on Roberts et al. [20]. Step One: Problem Conceptualisation and a Biological Framework ‘Problem conceptualisation’ involves describing the nature of the problem and disease process under consideration, stating project objectives, converting real‐world healthcare processes into simplified representations, and clarifying the decision problem to be modelled [20]. The diseases considered within the CKM syndrome have all been modelled previously, but not as a set of inter‐related comorbidities. Accordingly, deriving a biological framework of the CKM syndrome was identified as a priority. Conceptualisation started by reviewing existing literature to inform the modelling team's understanding of the CKM syndrome. An extensive review of previously published evidence focused on the 2023 AHA definition [2] and its evidence base [1], from which key components were pragmatically shortlisted for use as search terms, for example, ‘metabolic syndrome’, ‘type 2 diabetes’, ‘chronic kidney disease’, ‘cardiovascular disease’. For each component condition, evidence relating to epidemiology, pathophysiology, interconnectivity with other CKM components, and outcomes was gathered by handsearching in Google Scholar and PubMed. Narrative reviews and primary studies were examined, with further literature identified by backwards snowballing of cited references. The hierarchy of evidence [21] informed the prioritisation of systematic reviews and meta‐analyses where possible. This approach was supplemented with additional evidence highlighted by experts with relevant disease expertise. Existing model implementations that addressed similar research problems were also reviewed [22, 23]. Key findings from this phase were synthesised into a graphical representation and were subject to iterative review and validation by experts from relevant disciplines (see below) with discussion to reach consensus. Separately, a written statement of the decision problem, modelling objective and scope for the subsequent health economic model was developed following round‐table discussions attended by an expert working group (see ‘Critical review, validation and exploratory extension of the CKM syndrome’ below). Step Two: Model Conceptualisation and Influence Diagram ‘Model conceptualisation’ involved converting the disease problem and biological framework into an appropriate model structure via an explicit process, whereas recognising the need to focus only on aspects relevant to the decision problem, regardless of data availability. An influence diagram was deemed the most appropriate way of translating the biological complexity of the CKM syndrome into a simplified but practically useful tool for future model developers that retains the essence of the disease pathology. Within this diagram, person‐centred risk factors would influence the onset and progression of chronic conditions. Initial disease diagnosis would link to key clinical outcomes, such as disease progression or condition‐specific clinical events that are important to stakeholders (e.g., due to costs or impact on quality‐of‐life). Connections between chronic conditions, based on the underlying CKM biological framework, would reflect how comorbidities lead to or exacerbate each other. Biological concepts (risk factors, conditions and clinical outcomes) would be linked to humanistic (e.g., quality‐of‐life, life years lost) and financial outcomes (e.g., clinical event costs) to enable economic assessment. A draft influence diagram was derived and reviewed iteratively. Critical Review, Validation and Exploratory Extension of the CKM Syndrome Working draft versions of the CKM biological framework and influence diagram were refined iteratively over 6 months, based on the experience of a team of modellers, health economists and researchers working in the related disease areas of T2D, obesity and CVD. A multidisciplinary working group of experts, including clinicians from CKM‐relevant specialities alongside health economists, modellers and biostatisticians, was convened; a description of the relevant disciplinary expertise of the contributors is given in Table S1. Final versions of the biological framework and influence diagram were validated over two 90‐min workshop discussions held in early 2025. Informal feedback was gathered throughout, with additional insight provided in writing. Agreement and consensus were confirmed verbally throughout the workshop discussions and noted in the relevant discussion minutes. Clinician input ensured the relevance of the developing approach to real‐world clinical practice and future applications. The relative importance of chronic conditions and the feasibility of extending the CKM syndrome's definition to include additional chronic conditions were discussed. Where consensus was reached, the framework and diagram were revised to integrate new concepts. Where disagreements arose, contributors outlined their perspectives to the working group, which then discussed the issue to reach consensus. Results Problem Conceptualisation and a Biological Framework The decision problem guiding the development of the CKM syndrome health economic model is described in Table 1. A model capable of evaluating health interventions across the CKM syndrome would serve a broad range of stakeholders; therefore, multiple perspectives are accommodated. Given the range of health outcomes potentially linked to the CKM syndrome, we focused on those imposing substantial clinical and societal burden, for example, multimorbidity, acute events associated with high economic and/or humanistic costs, and premature mortality; examples are shown in Table 1, with additional details in Table S2. Regarding strategies/comparators, the model should support treatment effects across various CKM syndrome components, particularly for multi‐indication pharmacotherapies. The model's primary objective is to support public health decision‐making and clinical guideline development. TABLE 1Decision problem for the CKM syndrome model, based on ISPOR‐SMDM modelling good research practices task‐force guidance [20].Decision problemTo evaluate novel health interventions within the entirety of the Cardiovascular‐Kidney‐Metabolic (CKM) syndromeObjectiveTo guide public health decision‐making (e.g., the holistic impact of initiatives encouraging patient interaction with healthcare); to guide clinical practice and guideline development (e.g., the assessment of inter‐specialty clinics and decision‐making)PerspectiveHealthcare payer, guidelines (medical society), health policy makers, inform clinical decision making, patient advocacyDiseaseCKM—It is necessary to understand the complexity of the CKM syndrome, and to translate this understanding into a credible conceptual model structureTarget populationPeople with the CKM syndromeHealth outcomesProgression to multiple chronic disease flags (e.g., chronic kidney [CKD] stages 4–5, cardiovascular disease [CVD]), incidence of acute events (e.g., myocardial infarction [MI], liver transplant), mortality, quality of lifeStrategies/comparatorsTreatments with multiple effects across the CKM syndrome (SGLT2i, GLP‐1RA, aldosterone inhibitors), versus standard of care (SoC)Data sourcesDue to the complex nature of disease progression in the CKM spectrum, it is useful to provide a clear overview of the epidemiological data required to inform the modelling frameworkAbbreviations: CKD, chronic kidney disease; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular disease; GLP‐1RA, glucagon‐like‐peptide 1 receptor agonist; MI, myocardial infarction; SGLT2i, sodium‐glucose co‐linked transporter‐2 inhibitor; SoC, standard of care. Decision problem for the CKM syndrome model, based on ISPOR‐SMDM modelling good research practices task‐force guidance [20]. Abbreviations: CKD, chronic kidney disease; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular disease; GLP‐1RA, glucagon‐like‐peptide 1 receptor agonist; MI, myocardial infarction; SGLT2i, sodium‐glucose co‐linked transporter‐2 inhibitor; SoC, standard of care. A consensus biological framework depicting the conceptual components of the CKM syndrome is given in Figure 2. Diagram elements represent relevant clinical and biological concepts, whereas labelled directional arrows depict interrelationships; each label (letter/number) is detailed in Table 2. Solid arrows show ‘X is directly associated with Y’ and create causal pathways. Dashed arrows indicate feedback mechanisms, for example, link ‘a’ reflects that metabolic dysfunction‐associated steatotic liver disease/metabolic dysfunction‐associated steatohepatitis (MASLD/MASH) contributes to a proinflammatory and prooxidative state, and ultimately to further morbidity. Risk factors for hypertension, dyslipidaemia and reduced insulin sensitivity show substantial overlap. Diagnosis of metabolic syndrome (MetS) requires the presence of several chronic conditions, including obesity, impaired glucose tolerance, hypertension and dyslipidaemia [25]. Multiple pathological pathways connect MetS with T2D, CKD, MASLD/MASH and CVD, reflecting the evidence collated by Ndemele et al. [1]. Our framework also includes ‘cognitive impairment’, reflecting that several elements of the CKM syndrome (obesity, hypertension, diabetes) are modifiable risk factors for all‐cause dementia [31]. Biological framework for the CKM syndrome. Abbreviations: CKD, chronic kidney disease; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular; MASH, metabolic dysfunction‐associated steatohepatitis; MASLD, metabolic dysfunction‐associated steatotic liver disease; RF, risk factor; T2D, type 2 diabetes. *CVD encompasses multiple diseases, including heart failure across the ejection fraction spectrum, atrial fibrillation, coronary heart disease, stroke and peripheral artery disease [1, 2]. TABLE 2Tabulated explanation of relationships on biological framework diagram and supporting literature (when applicable).AssociationExplanationReferences1Obesity (body mass index [BMI] > 30 kg/m2) is characterised by excess adipose tissue, with or without abnormal distribution or function[24]2Risk factors for reduced insulin sensitivity influence risk of impaired glucose toleranceN/A3Risk factors for hypertension influence risk of hypertensionN/A4Risk factors for dyslipidaemia influence risk of dyslipidaemiaN/A5The individual components of Metabolic syndrome (MetS) are interconnected and impact on risk for each other[2, 25]6Excess and dysfunctional adipose tissue is associated with the presence of proinflammatory and prooxidative products…[26]7…leading to the development of elements of MetS[26, 27]8Components of MetS (e.g., obesity, impaired glucose tolerance) increase the risk of developing type 2 diabetes (T2D)[28, 29]9T2D increases the risk of developing cardiovascular disease (CVD)[30]10Components of MetS (e.g., obesity, impaired glucose tolerance and hypertension) are associated with increased risk of various aetiologies of cognitive impairment[31]11Excess adiposity is associated with ectopic fat deposition[32]12Proinflammatory and prooxidative products contribute to pathological processes within the body that damage arterial, cardiac and kidney tissues[1, 26, 32]13Proinflammatory and prooxidative products are associated with increased risk of CVD[26, 33]14Proinflammatory and prooxidative products are associated with increased risk of chronic kidney disease (CKD)[33]15MetS and its components are associated with increased risk of CVD[34]16MetS potentiates the cardiorenal bidirectional relationship[1]17MetS and its components are associated with increased risk of CKD[35]18T2D is a leading cause of CKD and end‐stage kidney disease (ESKD)[36]19Ectopic fat deposition leads to metabolic dysfunction‐associated steatotic liver disease/metabolic dysfunction‐associated steatohepatitis (MASLD/MASH)[37]20Ectopic fat deposition is associated with CKD[38]21Ectopic fat deposition is a risk factor for CVD[32]22Pathological processes contribute to atherosclerosisN/A23Pathological processes contribute to myocardial injuryN/A24Pathological processes contribute to CVDN/A25Pathological processes contribute to glomerulosclerosis, kidney fibrosis, kidney tubular inflammation, and CKDN/A26Glomerulosclerosis, kidney fibrosis, kidney tubular inflammation contribute to CKDN/A27Atherosclerosis is associated with CVDN/A28Myocardial injury is associated with CVDN/A29CVD and CKD share a bidirectional relationship; one condition increases risk of and exacerbates the other[6, 39] Feedback mechanisms aMASLD/MASH contributes to a proinflammatory and prooxidative state[40, 41, 42, 43]bCVD contributes to a proinflammatory and prooxidative state[44]cCKD contributes to a proinflammatory and prooxidative state[1, 33]dT2D contributes to a proinflammatory and prooxidative state[45, 46]Abbreviations: BMI, body mass index, CKD, chronic kidney disease; CVD, cardiovascular disease; ESKD, end stage kidney disease; HDL, high density lipoprotein; MASH, metabolic dysfunction‐associated steatohepatitis; MASLD, metabolic dysfunction‐associated steatotic liver disease; MetS, metabolic syndrome; SBP, systolic blood pressure; T2D, type 2 diabetes. Tabulated explanation of relationships on biological framework diagram and supporting literature (when applicable). Abbreviations: BMI, body mass index, CKD, chronic kidney disease; CVD, cardiovascular disease; ESKD, end stage kidney disease; HDL, high density lipoprotein; MASH, metabolic dysfunction‐associated steatohepatitis; MASLD, metabolic dysfunction‐associated steatotic liver disease; MetS, metabolic syndrome; SBP, systolic blood pressure; T2D, type 2 diabetes. Model Conceptualisation and Influence Diagram Key elements of the real‐world decision problem were translated into a qualitative, structured influence diagram (Figure 3), designed to provide a comprehensive representation regardless of current data availability; ISPOR guidance emphasises that the structure of conceptual models should be guided by the decision problem rather than by data availability [20]. The influence diagram encompasses the broad spectrum of conditions within the CKM syndrome. It distinguishes between the early stages of chronic conditions and their sequelae, including key clinical outcomes like cardiovascular events and condition‐specific downstream complications. ‘Chronic conditions’ and ‘key clinical outcomes’ are connected to ‘humanistic and financial outcomes’. For each diagnosis/clinical event, there would be an impact on quality‐of‐life (QoL), life expectancy and costs. To accommodate multiple perspectives, diverse costs could be incorporated, including direct healthcare costs (e.g., therapy costs, surgery/acute care costs), indirect costs (e.g., lost income/economic productivity for the person with poor CKM health; broader impacts on family and informal caregivers) and social care costs. The impact of intervention, via a ‘disease management module’ is demonstrated on the right‐hand side of the graphic; interventions in the CKM disease space would be introduced based on condition‐specific eligibility criteria (not shown) evaluated at each cycle, and would affect person‐centred risk factors, the progression of chronic conditions, the incidence of key clinical outcomes, humanistic outcomes and diverse cost outcomes. Influence diagram (model framework) for the CKM syndrome. For clarity, we do not show the explicit connections between risk factors and particular chronic conditions, nor the explicit connections between each clinical condition or clinical outcome and QoL, life‐years lost and costs. Abbreviations: BMI, body mass index; CKD, chronic kidney disease; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular; eGFR, estimated glomerular filtration rate; ESKD, end‐stage kidney disease; HbA1c, glycated haemoglobin; MASH, metabolic dysfunction‐associated steatohepatitis; MASLD, metabolic dysfunction‐associated steatotic liver disease; QoL, quality of life T2D, type 2 diabetes. *CVD encompasses multiple diseases, including heart failure across the ejection fraction spectrum, atrial fibrillation, coronary heart disease, stroke and peripheral artery disease [1, 2]. Discussion We present a conceptual approach to modelling the CKM syndrome. The syndrome's biological complexity (Figure 2) was abstracted and simplified into an influence diagram (Figure 3) for a future health economic model. Feedback mechanisms and multidirectional relationships are critical to understanding CKM pathophysiology; by capturing interconnectivity, the model would allow a more complete evaluation of the impact of multi‐indication therapies or interventions. In terms of implementing the influence diagram, we propose a person‐level microsimulation model that allows simultaneous modelling across the entire CKM syndrome as the most appropriate method. There is a growing preference for microsimulation models implemented in high‐performance computing environments, due to their scalability, flexibility and greater realism [47], in addition to maintaining model transparency and interpretability. By simulating individuals, microsimulations allow for a greater appreciation of the effects of heterogeneity in terms of risk factors and disease progression. This is particularly important when characteristics that vary across patients, for example, age or individual surrogate markers, exhibit non‐linear relationships with outcomes. In such cases, state transition Markov models which evaluate at the mean can lead to biassed estimates. Microsimulations track history and are well suited to modelling time‐dependent risk. In contrast, Markov models are memoryless; defining numerous, mutually exclusive health states to reflect the different chronic conditions within CKM and their combinations would result in excessive complexity. Given sufficient data, a CKM microsimulation model would allow the journeys of simulated individuals to more closely resemble the clinical reality of CKM multi‐morbidity than is achievable with single‐indication, state transition Markov models; this can be illustrated with hypothetical case studies (see Supporting Information). The availability of robust data is key; we propose that modellers implement any CKM syndrome model in a modular fashion that enables the exclusion of components, either where data are not available or where a more narrowly bounded disease context is demanded, for example, where the research question or stakeholder expectations favour restricting the number of comorbidities considered. In our conceptual model, simulated individuals would be followed over time to estimate the onset of chronic conditions and their sequelae (downstream or condition‐related events). Proposed definitions and thresholds for chronic conditions are shown in Table S3. The risks for onset of specific chronic conditions and incidence of related complications would be calculated using risk equations that incorporate an individual's characteristics, such as biomarkers and comorbidities. The evolution of biomarkers over time would be modelled using time‐path equations, with due consideration of how biomarkers covary. Connections between chronic conditions in Figures 2 and 3 reflect that for many conditions within the CKM syndrome, diagnosis with one is associated with increasing risk of others, which would be captured by the application of hazard ratios to condition‐onset equations and risk‐factor progression equations. The model's primary focus would be what we define as ‘key clinical outcomes’, which are associated with substantial burden and are of greatest significance to stakeholders. These were the development of CVD, progression of CKD to later stages and end‐stage kidney disease (ESKD), progression of early to late‐stage liver disease, development of cognitive impairment and condition‐specific clinical events (see Table S2). Although some of our key clinical outcomes align with the later stages of the AHA's CKM definition [2], we do not apply this staging more generally as expert opinion concluded that the early stages did not adequately represent clinical reality for many people, and would artificially constrain the possible ‘medical histories’ of simulated individuals. Given the complexity and heterogeneity of therapy pathways for patients with CKM syndrome, we propose the inclusion of a ‘disease management module’ which considers multiple criteria, including CKM stage, to determine therapy eligibility and intervention. Simulated patients would be assessed against condition‐specific eligibility criteria per model cycle. Interventions would be introduced, switched or stopped to update treatment in a way that reflects guidelines and/or clinical practice. An example of this approach in a T2D model has been published recently [48]. Treatment effects would propagate through the model's risk equation structure, ultimately driving predicted outcomes. Where biomarkers feed into multiple risk equations, the potential for double‐counting of effects must be evaluated. Expert opinion identified that it would be critical to consider chronic conditions and key clinical outcomes as distinct constructs within the model. Chronic conditions represent the underlying disease states of simulated individuals, whereas key clinical outcomes capture the downstream events and complications of those conditions. Separating these concepts would allow the model flexibility to evaluate interventions across the disease continuum and across interconnected treatment pathways. Critically, this would allow the model to assess strategies across primary, secondary and tertiary prevention. Although both the framework and diagram represent the consensus opinion of a working group of expert clinicians and researchers, and are based on both the AHA definition of the CKM syndrome [1, 2] and on a review of the literature, neither is comprehensive; they incorporate only what we consider to be the most important concepts. The AHA acknowledge that ‘nearly every major organ system is affected’ by the CKM syndrome [2], and there is an emerging recognition within the clinical community of the need to expand the AHA definition and include other outcomes and conditions (e.g., MASLD/MASH [49] and certain aetiologies of dementia [50]) caused or exacerbated by the CKM syndrome. Accordingly, our conceptual model includes some elements not currently emphasised by the AHA (e.g., cognitive impairment). Further expansion is possible (e.g., to incorporate the relationship between the CKM syndrome and respiratory health [51, 52, 53]). Our biological framework and influence diagram provide a foundation for aspiring CKM modellers. As they seek to translate the concepts shown into code, modellers may use these figures to structure their approach to model development. They may start by creating a template for patient‐centred risk factors to be considered within the model, then capturing their individual relationships with a single disease within CKM syndrome. Researchers working with our influence diagram must then consider multiple risk factors, disease definitions and outcomes, and must capture substantial interconnectivity. A rigorous approach to data gathering would be essential to ensure the robustness and validity of any novel CKM model, particularly regarding risk equations. Modellers would need to critically review equations to ensure they adequately capture the treatment effects of multi‐indication therapies, which may require the calibration of existing equations, the application of hazard ratios, or the derivation of de novo risk equations from patient‐level data [54]. De novo risk equations could be guided by clinical insight to ensure that covariates are included that make the equations sensitive across the entire CKM syndrome. However, equation derivation would require long‐term, harmonised datasets that link multiple biomarker trajectories with clinical events, or, if trial‐based, would require careful calibration to account for heterogeneity between trial populations and designs. Where bidirectional relationships exist between conditions, the order in which risk equations are evaluated is important; modellers must determine the most appropriate sequence. Modellers would also need to account for potential double‐counting of treatment effects (changes to surrogate risk factors considered in multiple risk equations, or the application of hazard ratios to risk equations, may result in unintended multiplicative effects) while ensuring that utility data used appropriately captures QoL in people with multiple conditions. Similarly, modellers would need to consider the potential for interactions between modern therapies given in combination across the CKM syndrome, where evidence may not exist. Evidence gaps would constrain the structure of any CKM model, and further research is needed. As part of the current work, we identified some prospective candidate equations which may be suitable for use in a CKM model (e.g., the PREVENT risk equations for risk of CVD events in patients without established CVD [55]; the UKPDS 82 equations for outcomes in patients with T2D [56]). We also identified several hazard ratios that could be applied to reflect the role of chronic conditions in increasing overall risk. However, collecting candidate equations and related evidence was not our primary objective, hence we do not specify possible data sources to be used in a future CKM model in Table 1. Because we do not have a complete appreciation of the evidence available, it would be premature to give concrete recommendations as to model implementation, beyond identifying that microsimulation is likely to be the most appropriate model‐type, and noting the most relevant conditions and events to be considered. This will be the subject of a future manuscript. In addition, validation of model outcomes is a critical aspect of good modelling practice [57]. There are five main types of validation: face validity, verification/internal validity, cross‐validity, external validity and predictive validity. Only confirmation of face validity is possible with our conceptual framework; the remaining four types require a fully developed model. Such a model could then be subjected to verification procedures and internal and external validation exercises. In addition, cross‐validation against established single‐ or multi‐indication models could be conducted. Predictive validation (assessing whether a model can accurately forecast outcomes) would require an opportunity to compare model outcomes with data from a study or real‐world source that is still ongoing at the time of prediction. At present, we are unable to discuss plans for validation further, but these will also be addressed in a subsequent publication. Several therapeutic agents with efficacy across multiple indications are available to clinicians treating the CKM syndrome. Given that clinicians treat holistically wherever possible, modelling the treatment of CKM‐related indications in isolation does not reflect clinical reality. Any model intended to inform decision‐making in this space must deliver a rigorous and clinically relevant estimation of the value of multi‐indication treatments, and only a holistic, integrated approach can achieve this fully. The utility of such a model to stakeholders would be substantial. Clinicians could leverage a CKM model to predict what management strategies would optimise clinical outcomes, such as the economic assessment of inter‐specialty workstreams compared against multi‐disciplinary team discussions for ‘joined‐up’ decision‐making. A patient‐facing interface would clearly communicate risk/benefit to people living with chronic conditions (similar to QRISK3 [58]), and would empower shared decision‐making. Additional example use cases for a CKM syndrome model are given in Table S4. Although we believe that CKM syndrome modelling will allow health economists, payers and policymakers greater insight into the real‐world value of multi‐indication interventions, increasing the relevance of cost‐effectiveness estimates and ultimately improving patient outcomes, we recognise that reimbursement and HTA bodies typically consider product‐specific single indication models. However, a holistic, integrated CKM syndrome model would provide valuable disease insights that the more traditional, product‐focused health economic models are less able to address. Hence, the models could be applied flexibly, depending on stakeholders' needs. The two models would provide complementary insights that, when taken together, could inform optimal reimbursement, access and policy considerations. For example, use of an integrated CKM syndrome model may be particularly valuable in the context of population‐level, holistic primary prevention or early disease management strategies, which product‐centric HTA‐focused models do not typically address. Conversely, where specific interventions at the later stages of a given disease are under assessment, the greater focus of a single‐indication model may be preferred. Even in such cases, a pragmatic incorporation of some CKM syndrome modelling concepts may be beneficial, providing additional context and evidence to support decision‐making. Our study has several strengths. We followed best‐practice guidelines for modelling transparency [20]. We leveraged expert insight in support of pragmatic literature research, with the intention of balancing the CKM syndrome's complexity, the simplicity required of a useful and transparent model, and the fast‐moving nature of this disease area. However, we acknowledge several limitations. First, we did not employ a systematic approach to literature review, and important areas of research may have been missed. However, we see our CKM syndrome influence diagram as a live project, and welcome comments and development. Second, expert opinion was collected informally. A formal process, for example, a Delphi panel, may have improved clarity and transparency. Third, although ‘model design’, where final model construction and data sources are defined, is a critical ‘next step’ [59], we do not address this: concepts can only be included given sufficient data, and the syndrome's complexity is such that until candidates are found for all data needs, piecemeal presentation would not be useful. Data identification is ongoing; final model design, implementation and validation will be addressed in subsequent publications. Fourth, our conceptual model includes some elements not currently emphasised by the AHA. Where evidence supporting inter‐connectivity within the CKM syndrome may be immature, the inclusion of such additional components may introduce uncertainty into any later model and affect the robustness of projections. Modellers should be aware of this potential, which may be mitigated by implementing CKM syndrome models in a modular fashion, as described. Uncertainty should highlight where future research is most needed. Finally, we present ‘intervention’ in our model framework at a conceptual level. This however does not reflect the complexity of managing CKM syndrome. In reality, therapy choices would need to be governed by multiple, clearly defined eligibility criteria within a ‘disease management module’, and would require explicit prioritisation of risk factors to support appropriate clinical decision‐making in heterogeneous populations with complex health problems. A comprehensive discussion of this module will be necessary following acceptance of formalised CKM treatment pathways. Conclusion There is significant value in developing a multi‐indication health economic model for the CKM syndrome that reflects biological reality and clinical practice. This paper is the first step towards a shared understanding of the modelling challenges and opportunities. We present a foundation for model development: a clearly defined decision problem, a biologically grounded framework and an influence diagram. Models built on this foundation would support robust evidence‐based decision‐making regarding the adoption and funding of emerging therapies with multiple disease‐modifying effects across this highly prevalent syndrome. Author Contributions All authors conceptualised and designed the study. All authors contributed to interpretation of the results, preparation and review of the manuscript, and gave approval of the final manuscript for publication. Funding This work was supported by AstraZeneca, who provided support for model conceptualisation, model development and medical writing for this study. Conflicts of Interest V.F., P.M. and G.R. are employees of Health Economics and Outcomes Research Ltd., who received fees from AstraZeneca in relation to this study. J.C. is an employee of AstraZeneca. M.E. has received consulting fees or honoraria from AstraZeneca, Novo Nordisk, Napp, MSD, Sunovion and Novartis, and lecture/speaker bureau fees from AstraZeneca, Novo Nordisk, Napp, Mundipharma, Sunovion and Novartis. D.C.W. has received consulting fees or honoraria from Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Eledon, Galderma, GlaxoSmithKline, Janssen, Menarini, Merck, Pharmacosmos, ProKidney, Tricida and Vifor. P.S.J. reports speakers' fees from AstraZeneca and ProAdWise Communications, advisory board fees from AstraZeneca and Bayer AG, and research funding from AstraZeneca, Boehringer Ingelheim, Analog Devices Inc., and Roche Diagnostics. P.S.J.'s employer, the University of Glasgow, has been remunerated for clinical trial work from AstraZeneca, Bayer AG, Novartis and Novo Nordisk. P.S.J. is a Director of Global Clinical Trial Partners Ltd. P.R.K. has received research grants from Pharmacosmos, and has received consulting and lecturing honoraria for Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, FIRE1, Novartis, Pfizer, Pharmacosmos, Servier, and CSL Vifor. H.Z. has received consulting fees or honoraria from AstraZeneca and Lilly. D.C.W. has received consulting fees or honoraria from Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Eledon, Galderma, GlaxoSmithKline, Janssen, Menarini, Merck, Pharmacosmos, ProKidney, Tricida and Vifor. M.R.C. is employed by AstraZeneca, and holds stock and stock options of that company. A.H.B. has received consulting fees from AstraZeneca in relation to the issues covered in this manuscript, as well as consulting fees from the following companies in the last 3 years: Astellas; BMS; Boehringer Ingelheim; Chiesi; Daiichi Sankyo; Evidera; Gilead; Galderma; Genmab; GSK; Idorsia; Novartis; Pfizer; Pharmacosmos; Precision Advisors; Rhythm; Sanofi; Sobi; Takeda; Teofarma. Supporting information Table S1: Areas of expertise for working group contributors. Table S2: Proposed definitions and thresholds for chronic conditions considered within the biological framework of the CKM syndrome. Table S3: Clinical events related to chronic conditions within the CKM syndrome that may* be modelled in a CKM model. Table S4: Example real‐world use cases for a CKM syndrome modelling approach. Figure S1: Patient* time trajectory within a hypothetical CKM model.