Towards more effective, efficient, and person-centered care for patients with chronic conditions in Spain through the CARABELA initiative

Abstract

Introduction

Chronic multimorbidity, aging and care fragmentation challenge health systems. In Spain, strengthening coordination between primary and hospital care may support more effective chronic disease management.

Objective

The CARABELA PC-HC initiative proposes an integrated, patient-centered approach to strengthen chronic care continuity across levels.

Methods

This cross-sectional analysis included five CARABELA initiatives (from May 2019 to October 2024) supported by ten Spanish scientific societies and AstraZeneca. Centers were selected for territorial representativeness, structural diversity, and population size. The CARABELA methodology included: design of the ideal care process based on clinical guidelines and real-world practice; identification of improvement areas (IAs) and solutions; validation of proposals and quality indicators (QIs) through a national Delphi consensus; and development of a Whitebook incorporating patient input and dissemination via the CARABELA Playbook platform.

Results

Chronic care delivery was mapped across 124 Spanish health areas, covering 1337 centers and 10,696 professionals. Globally, 82 IAs, 282 solutions, and 121 QIs were identified. Prioritized IAs included poor care continuity, lack of shared protocols, limited patient education, and uneven digital integration. Professionals emphasized coordination, training, and clinical protocols; patients prioritized autonomy, information, and emotional support. A core set of 54 cross-cutting solutions (including shared electronic health records and multidisciplinary pathways) and 46 transversal indicators was defined to guide implementation and evaluation.

Conclusions

CARABELA PC-HC provides a scalable framework to optimize chronic care integration. By aligning processes and patient-centered professional collaboration, it supports coordinated care models that enhance clinical outcomes, efficiency and equity, with potential applicability to other health systems.

Resumen

Introducción y objetivos

La multimorbilidad crónica, el envejecimiento y la fragmentación asistencial suponen un desafío para los sistemas sanitarios de todo el mundo. En España, la falta de coordinación entre la atención primaria y la hospitalaria dificulta un manejo eficaz de las enfermedades crónicas. La iniciativa CARABELA AP-AH propone un enfoque integrado y centrado en el paciente para reforzar la continuidad asistencial en enfermedades crónicas entre niveles y especialidades.

Métodos

Presentamos un análisis transversal (2019-2024) de 5 iniciativas CARABELA desarrolladas con el apoyo de 10 sociedades científicas españolas y AstraZeneca. Los centros participantes fueron seleccionados por su representatividad territorial, diversidad estructural y tamaño de la población atendida. La metodología CARABELA incluyó: i) diseño del proceso asistencial ideal basado en guías clínicas y práctica real; ii) identificación de áreas de mejora y soluciones específicas; iii) validación de propuestas e indicadores de calidad mediante un consenso nacional Delphi; y iv) desarrollo de un Libro Blanco que incorporó la perspectiva del paciente y su difusión a través de la plataforma CARABELA Playbook.

Resultados

CARABELA AP-AH caracterizó la atención a las enfermedades crónicas en 124 áreas sanitarias españolas, abarcando 1.337 centros y 10.696 profesionales. En conjunto, se identificaron 82 áreas de mejora, 282 soluciones y 121 indicadores de calidad. Las áreas de mejora priorizadas incluyeron la necesidad de mejorar la continuidad asistencial, ausencia de protocolos compartidos, educación limitada al paciente y una integración digital desigual. Los profesionales destacaron la necesidad de coordinación, formación y protocolos clínicos; los pacientes priorizaron la autonomía, la información y el apoyo emocional. La historia clínica electrónica compartida y las rutas multidisciplinares fueron elementos facilitadores clave. Se definió un conjunto central de 54 soluciones transversales y 46 indicadores transversales para guiar la implementación y la evaluación.

Conclusiones

CARABELA AP-AH proporciona un marco escalable y pragmático para optimizar la integración de la atención a las enfermedades crónicas en España. Al alinear los procesos, la colaboración profesional y las prioridades centradas en el paciente, respalda la transformación desde enfoques específicos por enfermedad hacia modelos asistenciales transversales y coordinados que mejoran los resultados clínicos y la eficiencia y equidad del sistema, sirviendo como referencia para otros sistemas sanitarios.

Introduction

Advances in medicine and public health have significantly increased life expectancy and improved survival for people with chronic conditions. Yet, despite these achievements, healthcare systems continue to struggle with the real-world implementation of these innovations, especially in the management of long-term care needs. The aging of the population and the growing prevalence of multimorbidity are placing unprecedented pressure on the healthcare system worldwide.^1–4 This tension is particularly evident in Spain: currently, 20% of the population is aged 65 or older, and ≥40% live with multiple chronic conditions,⁵ which reflects the growing complexity of care needs and the high pressure this places on the health system. This demographic and clinical shift calls for a comprehensive reorganization of care delivery, focusing on the efficient and integrated management of chronic conditions.^6–8

One of the most critical barriers faced by patients with multiple chronic conditions is the need to improve continuity across care levels, particularly between primary care (PC) and hospital care (HC).^9–11 Fragmented care results in repeated testing, uncoordinated treatment, delays in referrals, and overall inefficient resource use. These organizational discontinuities negatively affect care quality, patient experience, and health outcomes, while also contributing to clinician overload.^12,13 A longitudinal, coordinated approach to chronic care, grounded in multidisciplinary collaboration and shared decision-making, is essential.^6,14–16 While disease-specific care pathways can improve clinical outcomes, advancing toward a truly integrated model between specialties and care levels is indispensable to ensure comprehensive, efficient, and continuous care. Integration not only enables earlier diagnoses and reduces unnecessary procedures, but also supports better treatment adherence and long-term follow-up.^10,17

Overcoming persistent gaps in continuity, especially in the PC–HC interface, requires sustainable, shared, context-sensitive organizational models adapted to real-world practice. In response to this need, the CARABELA initiative was launched in 2019. To date, ten scientific societies, six patient associations, and AstraZeneca (AZ) have collaborated in this project. Based on lean health principles and multidisciplinary co-creation, CARABELA aims to serve as a catalyst for transforming chronic care delivery. Its methodology addresses clinical, territorial, and organizational challenges in managing chronic diseases and redesigning care pathways with a patient-centered perspective through the integration of patient and professional perspectives. CARABELA applies a cross-cutting approach, placing care continuity at the center of chronic disease management. It implements structured assessments to identify improvement areas, propose scalable solutions, and define quality indicators (Qis) tailored to the specific context of each territory to monitor progress and drive system transformation.¹⁸ Since its inception, this methodology has been successfully applied to high-burden chronic diseases such as severe asthma (SA),^19,20 heart failure (HF),^21,22 chronic kidney disease (CKD),^23–25 chronic obstructive pulmonary disease (COPD),^26,27 and diabetes mellitus (DM).²⁸ Each disease-focused CARABELA initiative has yielded integrated, evidence-based, territorially adapted care models, offering validated tools and actionable insights to improve care delivery.

Building on this experience, CARABELA has evolved to adopt a broader and more integrated perspective on care continuity between PC and HC. This article presents the results of CARABELA PC-HC, a national, cross-sectional analysis of chronic care delivery across care levels and specialties in Spain. This initiative has identified priority improvement areas, validated high-impact solutions, and defined shared QIs to build a practical and scalable framework that will enhance care continuity in chronic diseases and support system transformation.

This study offers a foundational reference framework with actionable insights for clinicians, healthcare managers and decision-makers committed to improving chronic care in the Spanish healthcare system.

Methods

Evaluation tools and analytical procedures

Between May 2019 and October 2024, we conducted a cross-sectional analysis of five CARABELA initiatives focused on chronic disease management in real-world clinical practice in Spain (CARABELA-SA, CARABELA-COPD, CARABELA-HF, CARABELA-CKD, and CARABELA-PC, the latter encompassing asthma, DM, COPD, HF, and CKD). Each CARABELA initiative was coordinated by a Scientific Committee and supported by several Spanish scientific societies, in partnership with AstraZeneca and clinical experts. Each committee was appointed by the participating scientific societies, which validated the different methodological phases and outputs of each initiative. Participating centers (primary care centers and hospitals involved in the management of chronic diseases within the Spanish National Health System) were selected based on predefined criteria aimed at ensuring territorial representativeness (coverage of the 17 autonomous communities), operational feasibility, structural diversity, and population size served. These centers were identified and invited as part of a jointly conducted initiative endorsed by the participating scientific societies, with the logistical coordination of invitations managed by AstraZeneca. For descriptive purposes, results are also presented by health areas, defined as administrative healthcare territories that integrate one or more hospitals and their associated primary care centers within the Spanish National Health System. Selection criteria included: (i) healthcare capacity and resources (e.g., availability of specialized nursing, dedicated patient units, and clinical registries enabling data extraction), (ii) existing professional training and patient education plans, (iii) level of coordination between primary and specialized care, and (iv) population profile served. These criteria were defined to capture heterogeneity in organizational contexts and levels of development across the Spanish healthcare system and were not applied as exclusion requirements; the availability of specific resources was therefore characterized as part of the analysis rather than used to select centers with an optimal organizational profile.

Data were collected using a mixed-methods approach, comprising structured workshops, interviews, documentary review, and questionnaires, applied at both the national and regional phases of the CARABELA initiatives. In addition, during the regional phase, the CARABELA Playbook was used as a digital tool to facilitate institutional self-assessment and to generate individualized feedback reports for each center.

Three analytical phases were applied: (i) assessment of current chronic care management in Spanish centers, using program-wide ad hoc questionnaires in SA, COPD, HF, and CKD; (ii) definition of an ideal, cross-cutting care pathway for multiple chronic diseases, integrating insights from all 5 CARABELA initiatives; (iii) identification and prioritization of transversal improvement areas and solutions, using ad hoc questionnaires and including those validated during regional dissemination workshops from the 5 CARABELA initiatives under study.

All analyses were descriptive in nature, in line with the cross-sectional design and the exploratory objectives of the study. Distinct analytical procedures were applied depending on the objective of each analysis, including frequency-based characterization of current care management, ranking of improvement areas and solutions based on perceived impact, and normalization procedures to enable comparability across domains, care levels, and pathologies (Supplementary Table 1).

The opinions of 1800 healthcare professionals were incorporated into the process. Healthcare professionals contributing to the analyses included PC staff (family physicians and community nurses), hospital-based specialists (e.g., pulmonology, allergy, cardiology, endocrinology, nephrology, internal medicine, and specialized nursing), and clinical managers. In both the prioritization and the current management assessment phases, data were collected through structured, multidisciplinary workshops at each center. Questionnaire responses were completed and agreed by the attending local multidisciplinary group.

The questionnaires used in the prioritization and assessment phases were developed ad hoc for the CARABELA program. Content validity was ensured through iterative development and review by the CARABELA Scientific Committees (members from the involved scientific societies), who defined domains and items to reflect routine practice needs and care standards for chronic disease management.

For the prioritization analysis, 213 centers in Spain were analyzed using impact-based ranking and normalization procedures to enable comparative interpretation across domains, care levels, and pathologies, as detailed in Supplementary Table 1. The impact of each improvement area was calculated by aggregating the individual impact scores of associated solutions and adjusting by the number of solutions linked to each improvement area to avoid overestimation. Prioritization was performed within each of the six domains in which the improvement areas and solutions were organized: resources, protocols, training, patient education, technology, and organization.

For the assessment analysis of current chronic patient management, 143 centers across Spain participated using pathology-specific questionnaires. Across all pathology-specific questionnaires, a total of 149 items were included and grouped into six thematic domains previously defined. Responses were qualitatively classified as “frequent,” “very frequent,” or “infrequent,” based on their relative frequency, as described in Supplementary Table 1. Both analyses were conducted at the centers listed in the Appendix.

All previously conducted CARABELA initiatives applied a shared, structured methodological framework. Briefly, the CARABELA methodology comprises the following sequential phases¹⁸: (i) meetings with the CARABELA Scientific Committees to design the ideal patient care processes based on clinical guidelines and real-world clinical practice for the management of the study diseases. A comprehensive analysis of care processes was conducted in selected pilot centers (centers involved in the initial exploratory phase of the CARABELA methodological framework) in order to identify areas for improvement and propose specific solutions addressing each identified area; (ii) validation of key improvement areas, potential solutions and QIs at a national meeting of multidisciplinary experts using a national Delphi-based consensus; (iii) drafting of a document that compiles all the initiative proposals and incorporates the patient voice (Whitebook); and (iv) dissemination and implementation of the validated proposals that optimize the management of the care pathway across Spain beyond the pilot centers, leveraging the CARABELA Playbook digital platform.

In the context of the present work, outputs derived from these phases were integrated and analyzed transversally. From the initial pool of QIs identified across previous CARABELA projects, a subset of transversal indicators was selected based on alignment with the validated transversal solutions. This selection was informed by a comparative analysis across the five initiatives, considering clinical relevance, applicability across different care settings, and alignment with established quality standards. The final set of transversal indicators underwent methodological validation by an independent expert in healthcare quality indicators to ensure consistency and appropriateness of the selected indicators.

Patient perspectives

The patient voice was integrated in CARABELA-PC, taking into account the full PC–HC pathway and addressing five conditions (SA, DM, COPD, CKD, and HF). Six patient associations were involved in patient identification and recruitment to ensure representation of the five conditions analyzed and diversity of profiles. In total, 30 patients (13 women and 17 men, aged 15–80 years, 7 of whom had SA, 6 had DM, 5 had COPD, 6 had CKD, and 6 had HF), participated via 5 structured focus groups, with telephone interviews for those unable to attend. Disease duration was classified as <10 years (14 patients), 10–20 years (10), and >20 years (5). Data were collected using a semi-structured guide including qualitative and quantitative questions, organized into four domains: (1) communication and care coordination, (2) disease-specific experiences, (3) patient education, and (4) emotional impact and adaptation. These inputs were analyzed separately and later contrasted with professionals’ perspectives within the overall analytical framework to identify areas of alignment and divergence.

Ethics declaration

All participants provided informed consent to participate and authorized the processing of their data. Data were collected and analyzed in anonymized and aggregated form, with no identifiable information recorded. These procedures complied with current Spanish data protection legislation, specifically Organic Law 3/2018 of December 5 on the Protection of Personal Data and Guarantee of Digital Rights.

Given the non-interventional, qualitative nature of the patient activities, and in accordance with Spanish biomedical research regulations (Law 14/2007 and Royal Decree 1090/2015), formal approval by a Research Ethics Committee was not required. All procedures were conducted in accordance with the principles of the Declaration of Helsinki.

Results

Population-level impact of the CARABELA PC-HC initiative

Since its inception, the CARABELA initiative has achieved broad national implementation across Spain. As of November 25, 2025, as a descriptive overview of implementation, 124 health areas have been characterized using the CARABELA PC-HC framework, providing care coverage to approximately 27,486,000 inhabitants, representing 56.55% of the Spanish population. These characterizations ensure that at least one joint evaluation between PC and HC has been conducted in each health area across all 17 autonomous communities (Fig. 1A).

Figure 1.

National scope and professional engagement of the CARABELA PC-HC initiative. (A) Map of Spain showing the number of health areas (124) characterized using the CARABELA PC-HC model per autonomous community. (B) Summary of the centers, healthcare professionals, improvement areas, proposed solutions, and quality indicators generated through the 5 CARABELA initiatives. CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; HF, heart failure; PC, primary care; SA, severe asthma.

A total of 1337 healthcare centers have been characterized across the five CARABELA initiatives: 43 in CARABELA-SA, 45 in CARABELA-HF, 77 in CARABELA-CKD, 43 in CARABELA-COPD, and 1129 in CARABELA-PC (Fig. 1B), involving 10,696 healthcare professionals and representing all 17 autonomous communities.

Current management of chronic diseases in Spain

Before identifying areas for improvement, solutions, and QIs, the management of chronic patients in Spain was assessed, and to facilitate analysis and readability, the data sourced from the Playbook were organized into 6 thematic domains (resources, protocols, professional training, patient education, technology, and organization).

While the implementation of certain tools and items was reported consistently across disease areas, the penetration of other elements into routine clinical pathways was limited. The use of case management programs, multidisciplinary committees, and remote monitoring tools was frequently or very frequently reported by professionals involved in the management of SA, COPD, CKD, and HF. Similarly, multidisciplinary committees and digital platforms such as e-consultations were widely available in several settings, particularly in SA and COPD (Supplementary Fig. 1). In contrast, several items were reported as infrequently available. These included standardized referral protocols in SA and HF, home follow-up, gamified lifestyle programs in CKD, and initial diagnostic evaluations in HF.

Based on the description of how chronic patients are managed within the national health system, an ideal care process has been defined that can help us identify areas for improvement, cross-sectional solutions, and QIs across care levels and pathologies. The ideal care process outlines the sequential phases of care (suspicion, diagnosis, treatment, and follow-up) in line with the 5 CARABELA initiatives (Fig. 2).

Figure 2.

The ideal chronic care process model is defined by the CARABELA initiatives. HC, hospital care; PC, primary care.

Identified improvement areas and solutions

Based on the previously defined ideal care process and the evaluation of the data collected during the CARABELA initiatives, a total of 82 improvement areas and 282 solutions were initially discussed in the CARABELA Playbooks and refined through a structured analysis. As a result of this process, a set of 13 cross-sectional improvement areas (Fig. 3) and 54 transversal solutions (Supplementary Table 2) with a focus on care continuity were selected and grouped into the 6 thematic areas (resources, protocols, training, education, technology, and organization). This provided a practical framework for subsequent analysis and prioritization.

Figure 3.

Identified cross-sectional improvement areas grouped into 6 thematic areas.

Improvement areas across care levels

Healthcare professionals from PC and HC prioritized the identified improvement areas based on their relative impact scores, derived from normalized ordinal ratings assigned during multidisciplinary workshops, and used to enable comparative visualization across domains and care levels (Fig. 4A). Among these, continuity of care, non-standardized protocols, and training were identified as the highest impact improvement areas overall, followed by nursing/group consultations, lack of health education, and digital tools, among others. When stratified by care level, notable differences emerged (Fig. 4B and C). In PC, the highest priorities were digital tools, lack of health education, and system overload, reflecting operational and educational needs. In contrast, professionals in HC emphasized continuity of care, protocol harmonization, and training.

Figure 4.

Prioritization of improvement areas according to participating professionals. (A) Overall prioritization across all respondents; (B) comparative prioritization by care level: PC vs. HC; (C) summary of the top 3 priority improvement areas identified for each care level. IAs, improvement areas; PC, primary care; HC, hospital care.

Transversal solutions

Following the identification of improvement areas, professionals were invited to prioritize the 54 proposed solutions grouped into the six thematic areas. The prioritization of proposed solutions revealed marked differences across thematic areas, as those related to organization, protocols, and training were most frequently selected by professionals as having the greatest potential impact on the healthcare system.

The evaluation of the 54 proposed solutions, grouped into 6 thematic areas, revealed clear differences in perceived impact.

Under Resources, the most frequently prioritized solutions across all professionals were the improvement of data recording, the availability of diagnostic testing resources in PC, and the use of patient-reported outcome measures (PROMs), patient-reported experience measures (PREMs) and clinician-reported outcome measures (CROMs) (Supplementary Fig. 2A and Fig. 5A). When analyzed by care level (Supplementary Fig. 2A), PC professionals assigned the highest impact to diagnostic testing resources, followed by the definition and registration of PREMs, PROMs, and CROMs, followed by the recruitment and retention of talent, highlighting the need to ensure not only material resources for diagnostic testing but also the standardization of these procedures. In HC, the top-rated solutions were data recording, diagnostic testing resources, and rehabilitation resources (as shown in Fig. 5B for each setting).

Figure 5.

Prioritization of solutions according to thematic areas and care levels. (A) The top 3 priority solutions identified across all professionals, grouped by thematic area: resources, protocols, training, education, technology, and organization. (B) The top 3 solutions prioritized at each care level are shown, highlighting differences and overlaps in perceived impact between settings. CP, community pharmacy; CROMs, clinician-reported outcome measures; ED, emergency department; EHR, electronic health record; HC, hospital care; HIS, health information systems; HP, hospital pharmacy; PC, primary care; PREMs, patient-reported experience measures; PROMs, patient-reported outcome measures.

Under Protocols, the most frequently prioritized solution across all professionals was the development of multidisciplinary referral and coordination protocols between PC and HC, followed by the design of care and follow-up protocols for PC, and specific protocols for HC (Supplementary Fig. 2B and Fig. 5A). When stratified by level of care (Supplementary Fig. 2B), PC professionals gave the highest priority to multidisciplinary protocols, PC care protocols and protocols for HC referrals. In the hospital setting, the most selected solutions were care and follow-up protocols for HC, followed by multidisciplinary protocols and protocols for PC. These results are reflected in the top solutions listed in Fig. 5B. Differences were also noted regarding protocols related to emergency care, nursing, and day hospitals, which were less frequently prioritized by PC professionals but gained relevance among HC professionals.

Under Training, the most frequently prioritized actions overall were training and awareness sessions for PC, nursing training, and a standardized training plan for emergency care (Supplementary Fig. 3A and Fig. 5A). By care level, PC professionals emphasized the need for training in PC, followed by sessions for nursing staff and for hospital professionals. Prioritized solutions in HC were training for PC and nursing staff, and the development of a structured training plan for Emergency Departments (ED) (Fig. 5B). Differences arose in the prioritization of training for emergency and hospital-based professionals, which was valued more highly by HC professionals compared to PC.

The most frequently prioritized solutions across all professionals within the Education domain were the development of patient education programs and materials, followed by health promotion initiatives and Expert Patient workshops (Supplementary Fig. 3B and Fig. 5A). Both PC and HC professionals consistently highlighted the importance of structured and standardized educational tools aimed at patients with chronic conditions and their caregivers, to promote correct treatment use and strengthen adherence. By care level, PC professionals prioritized Expert Patient workshops, health education materials, and autonomy-focused education programs. In contrast, HC professionals placed a higher priority on health education materials, health promotion programs, and Expert Patient workshops. While both settings recognize patient education and empowerment as key strategies, the emphasis on fostering patient autonomy was perceived as more impactful in PC than in HC. These findings are mirrored in the top 3 solutions identified by thematic area and by care level (Fig. 5B).

Under Technology, the solutions most frequently prioritized across all professionals were the implementation of a shared electronic health record (EHR) between PC and HC, together with EHR alerts for treatment adherence and follow-up, and incorporation of standardized checklists in health information systems (HIS) for screening and discharge (Supplementary Fig. 4A and Fig. 5A). This general alignment highlights the perceived importance of tools that strengthen coordination, optimize care processes, and ensure continuity in the management of chronic conditions. When disaggregated by care level (Supplementary Fig. 4A), PC professionals placed the greatest emphasis on shared EHRs, standardized checklists, and EHR alerts. In contrast, HC professionals emphasized shared EHRs, EHR alerts, and remote communication tools. Differences were also observed regarding the digitalization of diagnostic tests, which was valued more highly in HC than in PC. These results are consistent with the top-rated solutions listed for each setting (summarized in Fig. 5B).

Under Organization, the most impactful solutions across all professionals were the creation of multidisciplinary consultations, the establishment of interdisciplinary committees, and the implementation of fast-track circuits and pathways (Supplementary Fig. 4B and Fig. 5A). When analyzed by care level (Supplementary Fig. 4B), PC professionals assigned the highest priority to fast-track circuits from ED to PC/HC, followed by defining timelines for patient follow-up and implementing adherence control programs across care levels. In contrast, HC professionals most frequently selected multidisciplinary consultations and the development of interdisciplinary committees, together with the incorporation of specialized nursing staff. These results, summarized in Fig. 5B, show that while PC professionals prioritized improving coordination, timeliness, and follow-up in patient care, HC professionals emphasized the importance of multidisciplinary collaboration for managing chronic conditions.

Quality indicators for care model assessment

To evaluate the implementation and impact of the CARABELA proposals, a set of 203 QIs was initially compiled from previous CARABELA initiatives. From this pool, 46 transversal QIs were selected based on their alignment with the 54 validated transversal solutions. These QIs were classified into three categories: 27 structural, 14 process, and five outcome indicators, providing a comprehensive framework for monitoring progress, identifying improvement areas, and supporting decision-making based on objective and actionable metrics (Table 1).

Table 1.

List of quality indicators for characterizing the care model for patients with prevalent chronic conditions, classified by structure, process, and outcomes dimensions.

Structure indicators		Process indicators	Outcome indicators
1. Existence of training programs in PC on the diagnosis of prevalent chronic conditions, their coding, the use and interpretation of diagnostic tests, treatment, and follow-up.	2. Existence of an alert system in PC for requesting and implementing diagnostic tests and medication changes/upon admission of patients with exacerbations of a prevalent chronic condition.	1. Number of patients waiting <30 days for the first consultation.	1. Percentage of patients with prevalent chronic diseases with avoidable hospital admissions.
3. Availability of diagnostic checklists and follow-up protocols in PC for prevalent chronic conditions.	4. Availability of resources in PC for conducting diagnostic tests/rapid tests/treatment/follow-up (e.g., SMS appointment reminders) for patients with prevalent chronic conditions.	2. Waiting list (in the last 6 months) for the first diagnostic confirmation visit with a hospital specialist.	2. Percentage of readmissions within 15/30 days after discharge due to prevalent chronic diseases.
5. Annual number of diagnostic tests performed and waiting time for the following tests.	6. Existence of scheduled consultations adapted to the follow-up of patients with prevalent chronic conditions.	3. Use of updated referral protocols (within the last 2 years) for PC, Emergency Departments, and other specialties.	3. Percentage of patients with prevalent chronic diseases who are treatment-adherent.
7. Existence of priority access for severe patients to internal medicine/specialized units (fast-track).	8. Existence of emergency protocols for patients with exacerbations.	4. Percentage of patients meeting protocol and referral criteria from PC to HC due to suspected chronic conditions.	4. Percentage of patients with prevalent chronic diseases with good disease control.
9. Existence and use of treatment adherence programs.	10. Existence of a specific care pathway for hospitalized patients.	5. Percentage of patients referred from PC with a correct diagnosis.	5. Percentage of patients with prevalent chronic diseases satisfied with the care received in PC.
11. Existence of a specific nursing consultation for each prevalent chronic condition.	12. Existence of a designated nurse responsible for coordinating care transitions for patients with prevalent chronic conditions (case manager).	6. Use of an integrated care pathway updated in the last 2 years for patients with prevalent chronic diseases.
13. Availability of tools that facilitate continuity of care between PC, HC, and continuity of care professionals (e-consultation, shared EHR).	14. Existence of a day hospital and/or specialized unit for patients with prevalent chronic conditions.	7. Percentage of patients assessed by a multidisciplinary clinical case committee.
15. Availability of Expert Patient workshops/health education activities and educational materials for patients and/or caregivers, including digital tools.	16. Availability of questionnaires for measuring and analyzing QoL and patient-/caregiver-reported outcomes (PROMs).	8. Monitoring of hospital-approved protocols for referral to other centers when additional services or technology are required.
17. Existence of a multidisciplinary unit involving different specialties.	18. Availability of specific care delivery models such as day hospital, home hospitalization, or telemonitoring.	9. Percentage of patients with prevalent chronic conditions receiving self-care guidance including medication instructions and recognition of exacerbation symptoms.
19. Number of training courses aimed at PC professionals.	20. Percentage of nurses trained and specialized in the care of patients with prevalent chronic conditions, assigned to specialized units or internal medicine.	10. Use of educational support materials on the disease following diagnosis, adapted to patients/caregivers, in >70% of cases.
21. Existence of referral protocols from PC to other specialties.	22. Presence of rehabilitation programs.	11. Percentage of patients with prevalent chronic diseases who participated in a disease and treatment education program within the past year.
23. Existence of up-to-date support materials for patients or caregivers validated by the healthcare center or scientific societies.	24. Existence of a digital communication channel between PC and HC, and among different specialties (e-consultation, video calls, and other telemedicine tools).	12. Percentage of patients with comorbidities who received lifestyle recommendations.
25. Number of patients in the chronic condition unit.	26. Number of healthcare professionals trained and exclusively dedicated to prevalent chronic conditions.	13. Percentage of patients with prevalent chronic diseases assessed for social risk.
27. Number of trained nurses exclusively dedicated to prevalent chronic conditions.		14. Percentage of patients with prevalent chronic diseases evaluated for treatment adherence.

EHR, electronic health records; HC, hospital care; PC, primary care; PROMs, patient-reported outcome measures; QoL, quality of life; SMS, shorth message service.

Patient perspective: perception of care continuity

Patient feedback revealed several key aspects of care continuity across PC and HC. Responses to the disease-specific questionnaires highlighted the need for early diagnosis and reducing diagnostic delays, particularly when diagnosis followed an acute episode. Patients emphasized the importance of improving training for professionals in PC to enhance the detection and management of chronic diseases, as well as reinforcing chronic care training among emergency services staff. Across all pathologies, participants underscored the value of receiving clear, detailed information tailored to both the disease and the patient profile, especially at the time of diagnosis. Overall, participants expressed a shared need for supplementary written or digital information to reinforce medical guidance.

The psychological and emotional impact of living with a chronic condition was another prominent concern. Participants emphasized the importance of routinely addressing these aspects during care, highlighting their impact on disease progression and quality of life. Physical activity, diet, work, social life, and travel were among the areas most affected.

Communication and coordination between care levels emerged as critical. Although some patients reported occasional access to specialists via phone or email, these options were generally considered insufficient for their needs. In-person consultations were unanimously preferred.

Patient education was consistently viewed as a cornerstone of chronic disease management, contributing to better understanding, self-monitoring, and early symptom recognition. Nursing professionals were especially valued for their empathy and closeness; however, patients also identified a need for more time and enhanced training in specific areas, such as DM management. Patient associations were recognized as important sources of education and peer support.

Specific care programs, pulmonary and cardiac rehabilitation, or nutritional and lifestyle counseling were also positively valued. Psychological support for both patients and caregivers was identified across all groups as a major unmet need.

Eighty-six percent of participants expressed a desire for more specific, targeted information during critical phases of the disease to support better self-management. Among the most highly rated proposals were the implementation of PROMs, PREMs, and CROMs (with an average score of over 8.7), as well as broader access to educational materials for patients and caregivers (with an average score of over 8.5).

Finally, all patient input collected from the focus group sessions enabled the development of an emotional map (Supplementary Fig. 5) illustrating how perceived emotional well-being evolves throughout the disease journey. The map shows how the emotional status of patients scores the highest before the onset of symptoms but drops sharply upon the appearance of severe symptoms and diagnosis. Emotional well-being improves with treatment and acceptance, although it worsens again during exacerbations. In the chronic phase, scores stabilize at lower levels, indicating a process of emotional adaptation. The need for personalized psychological support was consistently highlighted across all stages of disease.

The prioritization of proposed cross-sectional solutions differed notably between patients and professionals. Patients favor solutions that support autonomy and practical disease management, assigning higher impact to initiatives such as collecting PREMs, PROMs, and CROMs, personalized self-care plans, and the development of gamified educational tools. In contrast, professionals prioritized actions aimed at improving coordination, standardization, clinical workflows, and decision support, such as specific training for nurses in PC, and the implementation of clinical alerts in shared EHR (Supplementary Fig. 6).

Despite these differences, both patients and healthcare providers converged in their positive assessment of gamified educational materials, highlighting this solution as a potential entry point for shared strategies. Conversely, the hybrid model of care (combining in-person, virtual, and home visits) was rated as low-impact by both groups, suggesting that this strategy was perceived to have limited relevance in enhancing either care experience or system efficiency (Supplementary Fig. 6).

Discussion

The CARABELA PC-HC initiative underscores the relevance of continuity of care as a key dimension in chronic disease management, based on this cross-sectional, mixed-methods assessment of five CARABELA initiatives. The findings described here point to concrete gaps and priorities across care levels and thematic domains, supporting a transversal approach to improve PC–HC coordination. Based on prioritization and assessment analyses, these results reinforce the relevance of integrated care models in addressing the coexistence of chronic diseases, where professionals, processes, and technologies operate in coordination across levels of care.^11,29–31 This alignment fosters synergies while keeping the patient at the center of care.^32,33

In a decentralized system, such as Spain, with marked rural–urban disparities in access to healthcare and availability of professionals, fragmentation between HC and PC levels adds further challenges to care continuity in chronicity. This fragmentation impairs efficiency, equity, and patient outcomes.^13,34,35 CARABELA addresses these challenges by identifying real-world barriers and prioritized solutions perceived as feasible within local settings. It maps improvement areas and supports the implementation of context-specific interventions. However, given the transversal design and the heterogeneity of participating centers, these results should be interpreted as a system-level description of perceived needs and priorities, rather than causal evidence of effectiveness.

A key finding was the adoption of a transversal approach, looking at the full clinical picture rather than addressing isolated pathologies. For instance, a patient with the triad of HF, CKD, and COPD benefits from coordinated strategies that optimize treatment across conditions, acknowledging shared risks and interaction. The initiative identified and prioritized 82 improvement areas and 282 proposed solutions, of which 54 were consistently prioritized across pathologies and care levels and were therefore classified as transversal and high-impact. Importantly, patterns of prioritization differed by care level: PC professionals more frequently emphasized operational needs (e.g., digital tools, health education, system overload), whereas HC professionals prioritized continuity of care, protocol harmonization, and training.

Within the Resources domain, the analysis of improvement areas revealed critical gaps in diagnostic capabilities between care levels. PC professionals called for enhanced diagnostic tools and data recording systems, while hospital professionals prioritized rehabilitation resources. This underscores the importance of strategic resource allocation that promotes complementarity rather than duplication of services across care settings. Across domains, organizational and technological components (e.g., structured pathways, inter-level coordination protocols, and decision-support tools) repeatedly emerged as areas with room for improvement, consistent with the cross-pathology characterization.

Across the evaluated initiatives, protocol standardization was among the most consistently prioritized solution types, emerging as a cornerstone of care continuity. Multidisciplinary referral and coordination protocols were among the most highly prioritized solutions. These protocols are the foundation of transversal care, offering clear guidance for managing multimorbidity while allowing flexibility for individual patient needs. In line with the results, protocol-related solutions were consistently prioritized across care levels, supporting their relevance as actionable levers for improving PC–HC coordination.

Strengthened coordination between PC and HC brings direct benefits, such as reduced hospital admissions, better adherence, and improved outcomes. It also mitigates territorial inequities by promoting more equitable access to quality care.^36,37 The CARABELA model suggests that such improvements may be pursued without major structural reforms, relying instead on more efficient use of existing resources, guided by shared objectives and enhanced cross-level collaboration. These broader system implications should be interpreted cautiously in this context and require prospective evaluation.

Technology was consistently highlighted as a key enabler of PC–HC coordination. Shared EHRs, clinical alerts, and digital decision-support tools were among the most prioritized solutions among professionals. These are essential to ensure continuity, monitor care processes, and enable decision-making and timely interventions.^38–41 However, as noted, access to technology remains uneven, limiting its current impact.⁴² Beyond infrastructure, organizational change requires professional engagement, particularly from PC and nursing teams. Their active role extends beyond clinical tasks to include participation in multidisciplinary training, standardized protocols, and knowledge-sharing.

A distinctive feature of CARABELA is the incorporation of a robust evaluation component, with 46 transversal QIs covering structure, process, and outcome domains. These QIs reflect both organizational capabilities (e.g., shared EHRs, training programs) and patient outcomes (e.g., avoidable hospitalizations, disease control, satisfaction) and allow for objective monitoring of implementation and impact. This focus on measurable outcomes strengthens the credibility of proposals and supports their adoption by healthcare managers and policymakers. Their performance over time will need to be assessed in prospective implementation phases.

Patient participation is another defining feature of the initiative. Rather than being passive recipients of care, patients were involved as active agents in the evaluation of solutions, providing insights that emphasized the value of gamified education, self-care plans, and the use of PROMs and PREMs. This reflects a growing interest in empowerment and participatory care models, aligned with principles of person-centered medicine.^43,44

The initiative also revealed key divergences in perspectives among the parties involved. For instance, the prioritization of solutions differs between care levels and between patients and health care professionals. Patients tended to prioritize autonomy- and self-management–oriented solutions (e.g., PROMs/PREMs/CROMs, personalized self-care plans, gamified education), whereas professionals more frequently prioritized workflow and coordination levers (e.g., nursing training and EHR alerts). These differences highlight the importance of stakeholder-specific implementation strategies that acknowledge varying priorities while maintaining focus on the overarching goals of care continuity and transversal management.

At a practical level, the prioritization patterns observed across domains and care levels suggest that implementation strategies may need to be tailored by setting (PC or HC) and anchored in the domains most consistently highlighted (protocols, organization, training, and technology), while preserving patient-centered components such as structured education and the use of PROMs/PREMs. These implementation considerations should be tested through prospective evaluations using the proposed QIs and within defined organizational contexts

This work has several strengths. It integrates data from five CARABELA initiatives, covering high-prevalence chronic conditions, and involves a broad range of professionals and patients across all Spanish regions. The shared methodology, supported by a robust digital platform and consensus-building processes, ensures consistency and representativeness. The integration of patient perspectives and the cross-sectional comparative approach enhance the reach and applicability of findings.

Limitations must also be acknowledged. Self-reported data and structured questionnaires may introduce subjectivity or variability in interpretation. In addition, the cross-sectional design precludes temporal assessment and limits causal inference. External validity may also be influenced by heterogeneity across centers (organization, resources, and participation profiles) and by potential selection and reporting biases inherent to voluntary self-assessment. Although the proposed solutions and QIs were validated by expert consensus, their implementation and impact on clinical and system-level outcomes require prospective evaluation. Accordingly, these results should be interpreted as a prioritization map and an evaluation framework to guide implementation and future evaluation, rather than as evidence of outcome improvement.

Conclusions

CARABELA PC-HC offers a realistic, scalable, and actionable framework to improve care continuity for chronic patients in Spain. By integrating PC and HC and adopting a transversal approach to multimorbidity, this initiative lays the foundation for transforming chronic care from a series of fragmented, disease-specific interventions to a coordinated, patient-centered system of care delivery. Within this integrated, cross-level approach and its transversal perspective across chronic conditions, digital tools, including artificial intelligence–based solutions, may act as enabling elements to support future implementation and scalability. The CARABELA model illustrates how collaborative work across care levels and specialties may support more effective, efficient, and equitable care for patients with complex chronic conditions, such as those with HF, CKD, and COPD, rather than focusing on isolated diseases.

Funding

AstraZeneca designed and funded the present transversal analysis and performed the data analysis.

Conflicts of interest

IM was the president of the Spanish Society for Healthcare Quality (SECA) and has received payment or honoraria for lectures, presentations, speaker's bureaus, manuscript writing or educational events from AstraZeneca, MSD, Novartis, Pfizer, Bristol Myers Squibb, Bayer, and Boehringer Ingelheim; and support for attending meetings and/or travel from Novartis and Pfizer. AP, NG, JD, EF, SJ, ML, GM, AM, GPR, CM and APD are employees at the Medical Department of AstraZeneca Farmacéutica Spain. JFS declares no conflict of interest.

Acknowledgments

The CARABELA initiatives were developed through co-organization agreements between AstraZeneca Spain and several Spanish medical societies. The authors thank the following scientific societies for their support of the CARABELA initiatives: the Spanish Society of Thoracic Surgery (SEPAR), the Spanish Society of Allergology and Clinical Immunology (SEAIC), the National Institute of Severe Asthma (INAG), the Spanish Society for Healthcare Quality (SECA), the Spanish Society of Primary Care Physicians (SEMERGEN), the Spanish Society of Internal Medicine (SEMI), the Spanish Society of General and Family Physicians (SEMG), the Spanish Society of Nephrology (S.E.N.), the Spanish Society of Endocrinology and Nutrition (SEEN), the Spanish Society of Cardiology (SEC), and the Spanish Society of Healthcare Executives (SEDISA).

The authors also acknowledge the contribution of the following patient advocacy organizations: ANADI, ADIMA, AD Madrid Sur, FENAER, ALCER, and Cardioalianza. Special gratitude is extended to the patients who generously shared their time, perspectives, and lived experiences, which were essential to shaping patient-centered care models.

We further thank the healthcare professionals involved across the different phases of the XXXX0 projects, from early design to implementation, including regional workshops and pilot experiences.The authors thank Susana Cañón-Sánchez, PhD, for medical writing support, provided in accordance with Good Publication Practice guidelines (http://www.ismpp.org/gpp3). This support was funded by AstraZeneca Farmacéutica Spain.