Patients presenting with multiple, interacting chronic conditions represent a growing challenge in modern medicine [1]. Effective management often requires integrating expertise from various specialties and adapting treatment plans based on the patient’s real-world experiences and responses, which can be difficult to coordinate in traditional, fragmented health care settings [2]. The need for improved approaches to navigate diagnostic and therapeutic uncertainty in complex cases is paramount [3].

The evolution of the internet, particularly the transition from Web 1.0 (a passive library) to Web 2.0 (an interactive forum allowing user participation) and beyond, has paved the way for new models of health care interaction. Concepts such as user-driven health care (UDHC) [4,5], where multiple health care stakeholders, including patients, interact online to understand and make decisions, and “medical cognition” [6] encompassing various human and artificial cognitive tools to resolve clinical complexity, are emerging. These approaches aim to leverage collective intelligence and asynchronous communication to improve patient care.

From a relational perspective, records of patient journeys are to be understood not as static carriers of information, but as a dynamic flow of information that shapes and is influenced by users’ actions, technology availability, institutional settings, and cultural norms [7]. This also opens up the possibility for learning from dialogic interactions, where patient journey records, health care providers, and clinical learners co-construct meaning over time across various contexts.

In this article, we demonstrate how our case-based blended learning ecosystem (CBBLE) combines online and offline components for case-based learning and integrates patient journey records (PaJRs) to facilitate participatory medical cognition [8]. The PaJR focuses on patient-reported outcomes, allowing us to learn from the patient’s journey beyond the prescribed treatment and the evolving decisions made during the phases of illness along the course of his life [9]. These platforms not only provide material resources but also actively structure collaborative interactions, adapting to sociocultural contexts and supporting learning in cases of complex multimorbidity. Drawing upon the concept of relational affordances [9,10], the PaJR-CBBLE system serves not merely as a digital repository but as a communication scaffold that enables training in human medical cognition. We illustrate this participatory approach through the management of a 44-year-old female patient with diverse and fluctuating chronic conditions, drawing upon the detailed record of discussions, interventions, and outcomes captured within the online platform. Similar to challenges highlighted in other contexts, such as the ethical and clinical dilemmas of overused medicine and the lack of shared decision-making in interventions like percutaneous coronary intervention (PCI) [11], this case presented its own complexities requiring careful, evidence-informed management. A previous study by Podder et al [8] demonstrated the potential of CBBLE in reducing overdiagnosis and overtreatment by fostering evidence-based, patient-centered inputs through coordinated team communication. This aligns with the patient’s expressed gratitude for achieving health improvements “without much medication and without getting expensive tests” that were previously common but yielded no improvement. The aim of our study was to understand and explain collaboration around PaJR workflow not just as information sharing, but as a dialogic and socio-material process [12,13], where we contribute toward a more nuanced understanding of how PaJRs can serve as a communication device in CBBLE for training of medical cognition [14,15] for managing complex multimorbidity. Unlike conventional EMR-driven systems that primarily support clinician documentation, the PaJR-CBBLE model brings patients and artificial intelligence (AI) tools directly into the collaborative reasoning process.

The case presented above demonstrates the application and advantages of a collaborative online platform grounded in the principles of medical cognition and UDHC. This participatory setting, acting as an online e-log book and a component of a CBBLE, enabled remote, multidisciplinary management of a patient with multiple comorbidities (diabetes, corneal dystrophy, recurrent infections, and pain syndromes). In the instrumental details section, we discuss various affordances around PaJR that became relevant as the patient’s care unfolded as a dynamic process involving ongoing interpretations and decision-making shaped by material tools and dialogic exchanges.

The platform facilitated input from various specialists at one place (physiatry, multiple ophthalmologists, general physicians, and PPMs), which is often challenging in traditional clinical set-ups, overburdening the patient to keep up with multiple follow-ups from different departments and extra-logistical work. This was crucial in navigating diagnostic nuances, such as distinguishing between potential episcleritis and hordeolum, or understanding the implications of Meesmann corneal epithelial dystrophy in a patient status post-PTK. For example, the patient’s historical use of heart rate, modulating drugs for perceived arrhythmia was reinterpreted as somatic anxiety rather than arrhythmia, prompting appropriate discontinuation.

The platform fostered collective evidence review and synthesis. Experts actively discussed and critiqued medical literature, reflecting a core aspect of evidence-based medicine. The detailed debate surrounding the efficacy of antibiotics for hordeolum, referencing published literature and discussing the hierarchy of evidence, is a prime example. PPM 5’s comment regarding the need to accept different levels of evidence in the context of Web 3.0 medical cognition highlights the practical challenges and adaptive thinking required when high-quality randomized controlled trial evidence is lacking. Also, the clinical learners (health care providers) participating (observing or engaging) in this process are encouraged to embody the system 2 thinking inherent in medical cognition, facilitating slower, more analytical processing of information.

While EMR systems also enable storage and sharing of patient data among physicians, they are typically oriented toward documentation rather than dialogic interaction. By contrast, the PaJR platform extends beyond a clinician-EMR dyad by integrating patients, PAs, and multidisciplinary experts into a single asynchronous workspace. The addition of AI tools further differentiates PaJR, not as a passive repository, but as an active cognitive partner that supports evidence retrieval, synthesis, and translation within ongoing case discussions. This combination, PaJR+ doctor+patient+AI, creates a participatory and iterative process that conventional EMR-based collaboration does not provide.

Notably, the platform’s role is here not constrained to data logging; it facilitates participatory distributed medical cognition [31] through discussion, dissent, and iterative reinterpretation. We noted that participants were also actively involved in discussions, and all the questions and inputs were facilitated in a conversational approach, in contrast to the usual medical learning and teaching pattern, which often involves long didactic lectures. Therefore, this aspect aligns with user-driven learning and the creation of user-driven learning community ontologies, which weave contextual patient data into a tapestry for reasoning and AI-driven processing and allow participatory learning for medical students. We are calling this participatory medical cognition.

This was powerfully echoed by the patient, who expressed being “extremely happy” with reduced medication and regained confidence, stating “I’ve gotten back a much better life” and “I can now do my own work.” This positive sentiment underscores the platform’s capacity to facilitate holistic care. This reduction is particularly striking given that when she was on antidiabetic medications with metformin 850 mg after breakfast and a combination of metformin 500 mg and glimepiride 1 mg after dinner a decade ago. Through the platform’s participatory care approach, she now takes only 750 mg Metformin daily and has stopped both glimepiride and BP medicine entirely since March 2025. This was possible only because of the patient’s commitment to battling her sarcopenia and trunkal obesity through regular sharing of her muscle strengthening exercises from the gym, as seen in the images logged on April 30 in the 44-year-old female case’s PaJR [32] and her well-regulated diet, as seen in the images of food plates in the 44-year-old female case’s PaJR [32]. This demonstrates the role of continuous lifestyle-guided care, intensive monitoring, and informed stepwise withdrawal of medication. At the same time, data logging enabled the patient to critique care gaps, where the clinical collaboration either empowered self-care (particularly weight loss, diet, and exercise) or did not offer symptom relief (arm-straightening and eye complaints), showing both the empowering and limiting dimensions of participatory care. Progressive reduction of blood pressure and anti-diabetic medications can be seen in Multimedia Appendix 1.

The integration of technology and AI was evident. AI tools like Meta AI and Perplexity were used for rapid information retrieval and synthesis on specific topics, such as the link between inflammation and hyperglycemia or evidence for antibiotic use. While human expert validation remained crucial, AI served as a tool to augment the collective information-gathering process. This aligns with the use of asynchronous intelligence AI tools in human cognition, where continued reliance on Web 3.0 tools (Meta AI, Perplexity, and large language models) is integrated as a dialogic supplement for medial cognition (another participant), not a replacement, to clinical judgment.

While not explicitly framed within a formal shared decision-making (SDM) model as discussed in the context of PCI, the process inherently involves shared information and facilitates a degree of patient involvement (through the PA) in understanding options and rationale, potentially leading to more confident choices. The continuous logging of patient data by the PA allowed the team to identify patterns, understand potential triggers (diet, travel, and weather), and observe the interplay between conditions. Discrepancies noted between the patient’s self-monitored BP and sugar readings and ward measurements prompted collaborative troubleshooting and investigation into potential device issues or measurement techniques. This was instrumental in the patient’s experience and to a certain extent, the emergence of her own medical cognition (or health cognition), as she felt “monitored very closely which helps her adhere to a good lifestyle” and found the group provided “accountability especially with regard to food intake.” The platform also has emergent affordances for the PA. The PA, acting as the liaison, was empowered to share detailed patient information and receive collective expert advice relatively quickly. This participatory approach also provided the patient with “mental strength and a sense of hope that she is not alone,” enabling her to “ask and update about her symptoms in the group everyday.” She also found motivation to “make innovative healthy food for herself”’ and became “‘more conscious about outside food.”

The CR’s role in deidentification and case report maintenance was vital to this process. The patient’s direct query about the impact of publishing her case report “‘will this open up new avenues.. for students?”’ and the assurance that it would “‘spread further in the scientific community”’ to help “‘many more patients”’ clearly demonstrates the platform’s role in fostering user-driven learning and community ontologies.

Overall, we find that maintenance of a structured case report via the PaJR system also serves as documentation and a learning resource. The PaJR captures the patient’s journey, discussions, and findings, contributing to a broader “participatory medical cognition” and serving as an educational tool for participants. The timeline in a case report represents more than a linear record; it embodies ongoing negotiations of meaning where patient experiences, clinical interpretations, and treatment decisions are continuously shaped by social and technological interactions within the platform.

We looked for patients connected via PaJR for similar outcomes. Interestingly, we noted another such case where a 63-year-old male with metabolic syndrome, dyspnea and previously diagnosed hypertension was able to stop his antihypertensive therapy with regular BP recordings shared on his PaJR journey by engaging with some clinicians as part of this study. It was argued, and doctors concluded, that BP recordings were not consistently high enough to merit further antihypertensive therapy [33]. This suggests the current clinical practices observed in different parts of the world, where prescription of antihypertensives based on single clinic BP recordings could be leading to the over-diagnosis of hypertension. Thus, we can argue on the basis of another PaJR that clinical judgment and decisions when made using active participatory data sharing approaches can help ensure optimal treatment plan, preventing over-diagnosis following over-treatment. Table 2 shows the specific uses of AI in clinical cognition and communication.