Mediums for personal health communication

In the evolving landscape of healthcare delivery and drug development, patient-centredness is increasingly acknowledged as a cornerstone of quality care. Today's patients are seeking greater autonomy over their health, driven by a desire to understand the complex factors influencing their well-being. This empowerment goes beyond traditional medical protocols, encouraging patients to adopt holistic strategies encompassing lifestyle changes and multidisciplinary approaches.

Amidst this shift, there has been a notable surge in the adoption of mindful and health-data collecting strategies, including writing practices, biometric wearables, and mobile health applications (mHealth apps). However, a critical challenge emerges: the abundance of data collected by users often fails to coalesce into comprehensible pieces of information that accurately reflect the multifaceted nature of health and well-being. For this data to be clinically valuable, especially in the context of medical consultations, digital platforms must be adept at synthesising information from diverse sources into models that are not only informative for clinicians but also informative to patients.

This essay seeks to delve into and propose design strategies aiming to bridge the communication gap between patients and healthcare providers through agnostic mediums for personal health communication.

State of Health Communication

The establishment of effective communication between patient and doctor is a complex interactional system, integral for facilitating the exchange of essential information and elicitation of patient preferences.

Human interaction attributes are a determinant factor in strengthening patients’ trust and satisfaction with healthcare entities and clinicians (Berry et al., 2021), and as patients are increasingly looking to become part of the decision-making process (Naseem, 2018), mutual agreement, empathy, and providing responses that empower patients to participate in making informed choices, and not just remain passive elements, are important assets in establishing a solid foundation for successful clinical journeys rooted in patient-centred care (PCC) (Backman et al., 2019; Langberg et al., 2019; Mikesell, 2013).

Because clinical interactions are widely diverse and individual, visualising such PCC attributes in practice might be a bit abstract. Perhaps they become more tangible though, if we reflect for a moment on an episode discussing for the first time a group of physical or mental symptoms with a doctor. How prepared were you to describe your symptoms? Did you manage to articulate them clearly, or did you struggle to convey them accurately on the spot? As the doctor inquired, did you feel your responses accurately captured the full extent of your health concerns? Was there a sense of mutual understanding achieved, or were there moments where the nuances of your experience got lost in verbal communication? These few questions highlight some of the potential barriers in patient-doctor dialogues, especially when confronting new or less linear health issues. Such interactions often leave patients wondering if they could express themselves clearly without missing important facts and if their doctors truly grasped the entirety of their concerns.

These communication challenges are inherent to any interaction model that relies heavily on verbal communication. As such, they are well-documented and somewhat anticipated. By adhering to key patient interaction guidelines, behavior-dependent limitations can be mitigated with a reasonable degree of confidence (Barsky, 2002; Laurence K., 2021). Despite these efforts, research in cognitive psychology highlights that human memory is often subject to inconsistencies, particularly in recalling events telescoped in time, leaving cognitive-dependent variables largely beyond control (Coughlin, 1990; Solga, 2001; Barsky, 2002; Jaspers et al., 2009). It is commonly observed that patients’ recollections of past symptoms, illnesses, and medical care are often inconsistent across inquiries (Barsky, 2002; Zaremba et al., 2018), leading to challenges in accurately articulating variations in symptom intensity and frequency. Additionally, patients frequently forget or underreport previous symptoms and events due to simple memory gaps, or merge separate, similar occurrences into a single, generalised memory. This process, known as memory consolidation, can result in the false recall of events and symptoms that did not occur, potentially leading to misleading accounts. Furthermore, the accuracy of recalling past medical events is influenced by various factors, including personality traits and the patient’s current emotional and physical state at the time of recall (current state bias). For example, patients experiencing anxiety, depression, pain, or bodily distress may emphasise events that might not have been recalled in a more comfortable and familiar environment (Barsky, 2002).”

Recognising the inherent limitations of memory recall, the medical community has long relied on patient diaries as a practical tool for tracking health metrics and symptoms. These diaries serve as a tangible method for capturing quantifiable data—such as blood pressure, weight, and emotional states—while also providing space for patients to record symptoms as they occur. By grounding observations in real-time documentation, patient diaries help mitigate the inaccuracies of retrospective recall, offering healthcare providers a clearer picture of patterns and potential triggers. However, as health-tracking moves into the digital realm, mHealth apps have emerged as modern alternatives, promising convenience and advanced functionality—but not without their own set of challenges.

Why Do mHealth Apps Fail?

Despite the promise of mHealth apps and a growing willingness among users to adopt them, significant challenges persist. One of the most pressing issues is the inadequate user comprehension of health information generated in consumer apps, which undermines their effectiveness (Rowland et al., 2020). This is evident in the low adherence rates, with many users abandoning health-tracking devices shortly after initial engagement (Meyerowitz-Katz et al., 2020). Furthermore, factors like user motivation, pre-existing health conditions, and human-computer interaction (HCI) attributes—such as information visualisation literacy, perceived usefulness, ease of use, and satisfaction with the information systems—play a critical role in sustained engagement (Vaghefi, 2019).

To explore these limitations and objectively trace information-supportive dynamics, I conducted an evaluation of selected leading health apps, complemented by user feedback. This assessment revealed recurring themes and highlighted interface challenges from both theoretical and practical perspectives. From these insights, I inferred speculative design experiments aimed at answering a broad yet defining question: How might raw data be transformed into a vivid, pictorial narrative of the self?

For the things of the world are their stories, identified by their paths of movement in an unfolding field of relations. Things occur where things meet, occurrences intertwine, as each becomes bound up in the other's story. It is in such binding that knowledge is generated.

(Ingold, 2011, 159)

Occurrences interwining, Ingold, T., 2011. Being alive: essays on movement, knowledge and description. Routledge, London ; New York.

The Fragmentation of Health Data

A common limitation of popular health-tracking apps is their inability to present information within a broader context. While many excel at data collection, they often fail to integrate diverse datasets effectively, preventing users from obtaining a broad and comprehensive overview of their health information (Yao et al., 2019).

Although intuitive infographics can effectively communicate specific domains, the inability to integrate multivariate data visualisation within a single graph limits the user’s ability to uncover meaningful correlations between overlapping events. For instance, a monthly view of rosacea symptoms may provide insights into their frequency and severity. However, layering or juxtaposing a second variable, such as emotional or stress levels, onto the same timeline allows disparate health metrics to converge, shaping a new context with potentially valuable correlations that enhance the user’s comprehension.

Building such contextual views supports the creation of coherent mental models, enabling users to better understand how different events and experiences are interconnected over time. By situating these variables within a shared timeline, users can interpret how the sequencing and relationships between factors influence their health. This contextualised visualisation not only allows them to explore potential correlations and patterns that might otherwise remain hidden but also facilitates more effective personal insights and enhances communication with healthcare providers, fostering a more comprehensive understanding of their health experiences.

— Data abundance doesn’t ensure better understanding—excess can result in confusion, anxiety, and eventually, resource inoperability.

This challenge is compounded when users rely on multiple apps to manage their health, each specialising in a narrow domain. For instance, one app might track dietary intake while another logs symptoms, but without interoperability, users are left to manually correlate these data points. This fragmented experience detracts from the effectiveness of these tools, making it harder for users to uncover patterns and manage conditions efficiently.

The Apple Health app displays symptoms in isolated views, requiring multiple steps of navigation to access each symptom individually. This fragmented design prevents users from viewing symptoms in context with one another, making it difficult to identify correlations or patterns across symptoms. The lack of an integrated view limits a comprehensive overview into health trends and adds unnecessary complexity to the navigation flow.

Visualising time

Preserving context across timeline views is a common challenge in interface design. Timelines represent sequences of events in chronological order and can vary in complexity. A basic timeline might display event types, counts, and order, while more complex timelines include details such as timing, duration, and overlap of events. Transitions between these views often risk compromising information continuity, visual stability, and variable correlations. For example, moving from a broad overview of health events to a more detailed view can disorient users. Without careful design, such transitions may lead to a loss of context, making it harder for users to understand how individual events relate to their overall health journey.

Given a timeline story comprised of a sequence of narrative points, there may not exist a single timeline design that adequately communicates all of these points; as a consequence, a storyteller might incorporate multiple timeline designs into a single story.

(Brehmer et al., 2017).

Balancing

A reader-driven approach provides users with a high level of interactivity, enabling them to independently navigate and analyse their health data. This method encourages active exploration, appealing to users inclined toward self-led discovery and detailed analysis. However, it can also be overwhelming for those less comfortable with data interpretation, potentially leading to confusion or misinterpretation. Conversely, an author-driven design presents health data in a predetermined, linear format with limited opportunities for user interaction.

While a pre-structured interface is essential for establishing the user experience, the ideal solution would strike a balance between structure and adaptability (Chen et al., 2020). This balance would allow users to explore beyond predefined paths by interacting with familiar patterns, such as filters, enabling them to uncover insights that programmers may not have anticipated.

This raises some key questions: What kinds of inquiries would most effectively enhance patient-doctor interactions? How can we design interfaces that strike the right balance between exploratory freedom and intuitive ease of use? And how might systems intelligently resurface meaningful patterns from months of collected data, revealing connections users didn’t even know to look for?

Design Goals

There is not one singular timeline view to ideally present the information necessary to answer these questions, thus, adopting different representations might be necessary. Drawing from invaluable principles of data storytelling, I am exploring various dynamics rooted in fundamental spheres of prior work encompassing temporal logic, and temporal visualisation.

This framework outlines the theoretical underpinnings of the visual design process, providing a structured approach to bridge the gap between visual analytics and narrative frameworks rooted in storytelling principles. The goal is to create meaningful, context-oriented visual dialogues that enhance the understanding and interpretation of health data.

The correspondence between visual analytics and storytelling has been conceptualised to address the following key objectives:

• Transform large, complex data entries into dynamic and easily interpretable information visualisations.

• Break data silos by integrating disparate sources into a unified timeline, enabling the contextualisation of events as they unfold.

• Facilitate comprehension of both single-point events and interval-based events within their temporal and relational sequence.

• Develop varying levels of immersion that allow users to analyse and interact with information at different degrees of detail, balancing high-level overviews with deeper explorations.

Visual storytelling

Designed to support a closer and more collaborative relationship between patients and clinicians, the principles of visual storytelling provide a foundation for imagining what a personal health communication medium might look like.

Visual storytelling excels in anchoring elements over time, helping users situate and connect events within a broader timeline. Unlike static, isolated representations, multivariable visualisations weave disparate data points—such as symptoms, external influences, and lifestyle factors—into a unified narrative. This approach provides a way to manage the interplay of multiple variables in health-tracking systems (Segel, 2010) while shaping these variables into coherent stories that reflect robust models of reality.

To know someone or something is to know their story, and to be able to join that story to one’s own.’

(Ingold, 2011: 161)

A visual medium

On one hand, a visual medium can support patients in articulating their symptoms and emotions with increased clarity and confidence; on the other, they provide clinicians with a shared factual foundation, inviting them into a relatable, patient-centered narrative. This mutual understanding has the potential to enhance diagnostic accuracy, treatment adherence, and overall clinical care (Domin, 2021; Lupton, 2021; Melbye et al., 2020).

At the outset of this essay, I invited you to recall a moment when you shared a complex story for the first time—did your words capture its depth, and did you feel assured that your doctor fully grasped its meaning?

This is precisely the purpose such a tool should serve: to provide immersive snapshots of time that make personal experiences tangible. By transcending the limitations of active recall and verbal communication, these tools offer quick, intuitive overviews of layered health information, transforming fragmented data into coherent narratives that enhance patient confidence (Schröder et al., 2023). In doing so, they may cultivate deeper, more informed relationships between patients, clinicians, and caregivers (Vaughn et al., 2021).

— Narratives provide structure to what we record and share, helping us develop a subjective and contextual understanding of our state over time.

A particular challenge I aim to address is the ability to represent diverse types of information—captured from multiple sources and spanning extended time periods—in a way that is flexible, scalable, and ultimately yields easily interpretable and meaningful insights.

Design implementation

Query Presets: Glanceable patient-centered snapshots

Incorporate pre-set queries designed to answer specific and commonly posed health questions, such as:

• How have you been for the past X months?

• How would you describe your most recent flare-up?

• How impactful have anxiety and depression been recently?

• Have your sleep patterns been affected during that period?

By visually representing these queries, users can explore their health data with greater relevance and precision. This approach provides clear, glanceable snapshots that are simple to interpret and share, offering a personalised and intuitive method for retrieving meaningful insights.

This screen visualises a 6-month health timeline, integrating symptom severity, interventions, and mental state data into a cohesive view. Highlighted regions and layered graphs provide context for flare duration, symptom spikes, treatment periods, and emotional fluctuations, enabling users and clinicians to identify patterns and correlations across metrics.

Gradients between aggregation and details on demand

Facilitate shifting between context and focus states to balance the navigation between event overview and detailed exploration by providing two main degrees of immersion:

Context Level

Present a dynamic, linear timeline that sequences multivariate events, summarising data through aggregated layouts (clusters) for a broad overview of health patterns over time. This view enables users to quickly identify trends and high-level insights without overwhelming them with excessive detail.

A rosacea flare from February 15 to April 2 is visualized alongside symptom severity and treatment interventions. Medications such as Soolantra and Metrogel are aligned with the timeline, offering context for understanding the progression of symptoms and the potential effects of treatments.

The 47-day rosacea flare is contextualised with a comparative duration graph, contrasting it against two previous flares lasting 12 days and 30 days. This visualization provides users and clinicians with a clear reference for understanding the relative severity and duration of flares over time.

Focus Level

Allow users to interact with clusters to explore their contents in greater detail, such as viewing symptoms associated with a particular event or period. This approach uses a layered timeline design to break down aggregated data into its individual components, offering users a clearer view of how events and variables interrelate. By enabling seamless transitions between broad overviews and more focused perspectives, users can engage in an immersive exploration of individual health events and their connections over time.

The use of zooming animations, touch gestures, and hover states to move between these facets should be determined as to aid in preserving context continuity and spatial stability, facilitating an uninterrupted users sense of place in the timeline.

This visualisation of the rosacea flare employs a stacked timeline chart to break down symptom categories, including skin-related issues like oedema, burning, and erythema, as well as digestive symptoms. The layered bar design provides a clear representation of symptom intensity and duration, while treatment interventions with Soolantra and Metrogel are displayed as overlaid bars. This approach offers a detailed and structured view of symptom progression in relation to management strategies.

Closing notes

The process of writing the notes that ultimately shaped this essay has been a profoundly enlightening and invigorating journey. The convergence of diverse disciplines and scientific fields uncovered new perspectives, challenging initial design assumptions and encouraging me to reassess intentions and redefine priorities. This iterative process has led to a more nuanced understanding, raising questions that extend beyond the scope of these notes and the resources currently at my disposal.

Although limited, the insights gained through this process have been invaluable in framing new possibilities and guiding the exploratory design choices presented here. I am hopeful that these reflections will serve as a foundation for deeper inquiries and be instrumental in future research opportunities should they arise.