Textbook of Digital Health

Introduction

Patient engagement represents a significant shift in healthcare from a model where patients passively receive services to one where they actively participate in health decisions, self-management, and care processes (Topol, 2015). Digital tools are changing how patients access information, communicate with providers, manage chronic conditions, and participate in treatment decisions. From patient portals providing direct access to medical records to mobile apps supporting shared decision-making, technology is enabling new models of patient-centred care that improve outcomes while respecting patient autonomy.

This chapter explores patient engagement concepts, digital tools supporting engagement, health literacy considerations, and the digital divide that threatens to turn technological progress into new health disparities. A recurring tension runs through the chapter: the same tools that empower digitally confident patients risk marginalising those who are not.

Patient Engagement Concepts

Patient engagement encompasses how individuals participate in their own health and healthcare, from self-management and shared decision-making through to involvement in quality improvement and policy. The concept of patient activation - the knowledge, skill, and confidence individuals have for managing their health - provides a useful framework, and research consistently shows that more activated patients achieve better outcomes, experience fewer hospitalisations, and report higher satisfaction. The Patient Activation Measure (PAM) (Hibbard et al., 2004) is a validated instrument assessing these three elements and locating respondents at one of four developmental stages of activation - from passively believing the patient role is not important through to maintaining active self-management under stress.

Several forces have driven the shift toward greater patient engagement. Consumer expectations shaped by other industries (retail, banking, travel) create demand for healthcare experiences that match the convenience and transparency people encounter elsewhere. Growing evidence links engagement to improved clinical outcomes, and value-based payment models increasingly incentivise the outcomes that engagement supports. Yet engagement is not simply a matter of providing tools and expecting patients to use them. Health literacy limitations, cultural factors affecting comfort with active participation, power dynamics in clinical relationships, and systemic barriers all influence whether engagement opportunities are genuinely accessible. A portal that technically offers transparency achieves little if the patient cannot interpret what they see, feels unable to question their clinician, or lacks internet access - themes this chapter returns to in later sections.

Patient Portals

Before patient portals, accessing one's own medical records often meant navigating bureaucratic processes. A patient wanting copies of laboratory results might submit written requests, wait weeks, and pay per-page copying fees, all for information about their own body. The medical record felt like something that belonged to the institution, not the individual. Patient portals have substantially changed this relationship, providing electronic access to health information and services that places patients at the centre of their own care.

Consider what this shift means in practice for a hypothetical patient, Sarah. When her GP orders blood tests to monitor her thyroid medication, she no longer waits anxiously for a phone call that may never come. Within hours of the laboratory processing her sample, results appear in her portal with reference ranges clearly marked. She can see that her TSH level has normalised, reassuring her before she even speaks to her doctor. This transparency helps patients become informed partners who can track their own health trajectories over time.

Yet portals offer far more than record access. The ability to schedule appointments online, request prescription refills, and message clinical teams asynchronously reduces reliance on automated telephone menus and removes business-hours-only limitations. A working parent can request a medication refill at 10pm after the children are asleep rather than taking time off work for a phone call. Someone with social anxiety can compose a careful message about a sensitive symptom rather than struggling through an uncomfortable phone conversation.

The secure messaging function particularly illustrates how portals reshape the patient-provider relationship. Rather than hoarding questions for rushed appointment visits or wondering whether a concern warrants an office call, patients can communicate asynchronously about non-urgent matters. "I've developed a new rash since starting this medication. Is this something to worry about?" The care team can respond thoughtfully, perhaps with a request for a photograph, guidance for monitoring, or a recommendation to come in. This dialogue happens on both parties' timelines, respecting that neither patients nor clinicians are always immediately available.

One of the more complex portal functions involves test results, particularly those carrying significant implications. When a pathology report becomes available, should it appear instantly in the patient's portal? The question reveals an underlying tension. Immediate release treats patients as adults with rights to their own information, but it also means some people learn of cancer diagnoses while alone at home, without clinical support or context. Different organisations have landed on different approaches: some delay sensitive results to allow provider review and contact, while others release everything immediately, trusting patients to reach out when needed.

Despite their potential, portals have not reached everyone equally. Adoption data reveal a concerning pattern: portal use is consistently lower among older adults unfamiliar with digital technology, people with limited educational backgrounds, non-English speakers navigating interfaces designed for native speakers, lower-income individuals who may lack reliable internet access, and people with certain disabilities that make standard interfaces inaccessible. When portals become the primary pathway to scheduling, communication, and results, these populations risk being relegated to a second-class care experience. The chapter returns to these disparities, and the organisational responses they demand, in the section on the digital divide.

Personal Health Records

Patient portals, for all their benefits, share a practical limitation: each one provides a window into only one organisation's slice of a patient's care. The person managing diabetes through their GP, seeing a cardiologist at the regional hospital, and receiving physiotherapy at a private clinic might have three separate portals, each offering a partial view that never connects with the others. Personal health records (PHRs) emerged from the recognition that patients themselves are the common thread across these fragmented systems, and perhaps the solution lies in giving them tools to weave their own comprehensive health narratives.

The vision is compelling. Imagine a single record containing clinical information from every provider encounter, patient-entered observations about symptoms and lifestyle, data streaming from fitness trackers and glucose monitors, and scanned copies of records from long-ago providers or international care. Such a longitudinal view could reveal patterns invisible to any single clinician: the medication started at one practice that coincided with symptoms documented at another, or the gradual trajectory of a condition across decades of care.

This vision has attracted significant investment. Apple Health Records, which uses FHIR-based integration to pull clinical data directly from participating health systems into the iPhone Health app, represents one of the most prominent efforts. By the mid-2020s, Epic MyChart had become one of the most widely used patient portal and PHR platforms, aggregating records across Epic-using organisations. Numerous startups have also targeted specific populations or use cases. Technical standards like FHIR have matured to the point where data aggregation is technically feasible, and regulations increasingly require healthcare organisations to make data available for patient-directed access.

Yet despite these enabling conditions, PHR adoption remains limited. The gap between vision and reality reflects several persistent barriers. Many patients simply do not perceive sufficient value relative to the effort of setting up aggregation, particularly when they are healthy and their care is uncomplicated. Technical hurdles persist, as connecting to multiple organisations requires navigating different authentication systems and data formats. Privacy concerns loom large when sensitive health information moves outside clinical systems with established security infrastructure. Most importantly, providers rarely integrate with patient-held records in ways that make the PHR genuinely useful for care coordination rather than merely a personal archive.

The promise of patient-controlled comprehensive health records remains largely unfulfilled, though the enabling technologies continue to improve.

Shared Decision-Making

A woman in her fifties receives unexpected news: a mammogram has revealed a small, early-stage breast cancer. Her oncologist explains that for her particular tumour characteristics, surgery followed by radiation produces roughly equivalent survival outcomes to mastectomy alone. Both are medically reasonable options. "So which would you recommend, doctor?" she asks. But this is precisely the kind of decision where clinical evidence alone cannot determine the "right" answer; her choice depends on values only she can articulate. Does she prioritise preserving her breast or minimising treatment duration? How does she weigh the daily radiation appointments against a more extensive surgery? What matters most to her sense of self and wellbeing? Shared decision-making recognises that good medical care requires combining clinical expertise about options with patient expertise about their own lives (Elwyn et al., 2017).

This philosophy has given rise to an ecosystem of digital tools designed to support patients as genuine partners in consequential health decisions. Decision aids represent the most thoroughly studied category: interactive presentations that walk patients through available options, what research reveals about benefits and risks, and what remains uncertain. A decision aid for prostate cancer treatment might explain that active surveillance, surgery, and radiation produce similar survival outcomes for low-risk disease, while differing substantially in their effects on urinary, sexual, and bowel function. Research demonstrates that patients who use quality decision aids have better knowledge, more realistic expectations, and ultimately make choices more consistent with their own values (Stacey et al., 2024).

Yet knowing the options is only part of informed decision-making. Patients also need to understand their personal risk rather than population averages. Risk calculators transform individual characteristics into personalised probability estimates. A woman considering preventive mastectomy because her mother had breast cancer faces a very different calculation if her own risk is 15% versus 65%. Cardiovascular risk calculators help patients understand whether their particular combination of factors warrants medication, lifestyle changes, or both. These tools move beyond generic pamphlets toward genuinely individualised information that enables meaningful participation in decisions.

An often underappreciated element of shared decision-making involves values clarification: structured exercises that help patients explore and articulate what actually matters most to them. When treatments offer different tradeoffs, the "best" choice depends entirely on how a particular person weighs competing outcomes. One patient facing knee replacement might prioritise returning to golf as quickly as possible; another might be more concerned about minimising anaesthesia risks given her age. One person values aggressive cancer treatment even with significant side effects; another prioritises quality of remaining time over maximum duration. Without surfacing these preferences explicitly, clinicians may assume patients share their own values, or patients may defer to medical authority for decisions that are fundamentally personal.

Making shared decision-making work in practice requires more than excellent tools. Busy clinical schedules leave little time for extended conversations. Patients accustomed to paternalistic models may be uncomfortable when asked what they want rather than told what to do. Clinicians need training in both decision aids and the communication skills that enable genuine partnership. The tools should supplement rather than replace human conversation. A decision aid reviewed before an appointment prepares a patient to engage meaningfully during the consultation, but the technology cannot replace the rapport and clinical judgment that direct conversation provides.

Health Literacy

"Take this medication twice daily with food, and we'll check your INR next week to see if we need to adjust the dose." To many patients, this instruction seems straightforward. But consider what understanding it actually requires: knowing what INR means and why it matters, recognising that "twice daily" means roughly twelve hours apart rather than any two points in a day, understanding that certain foods interact with the medication, and grasping that laboratory monitoring is needed because the therapeutic window is narrow. Health literacy (the constellation of skills needed to obtain, process, understand, and apply health information) underpins every aspect of patient engagement. And limited health literacy is far more common than many clinicians assume: estimates suggest more than one-third of adults struggle to understand typical health materials; in the United States, the 2003 National Assessment of Adult Literacy found that 36% of adults had basic or below basic health literacy (Kutner et al., 2006).

The challenge operates on multiple levels. At the most basic level, functional health literacy involves reading ability, numerical comprehension, and document navigation applied to health contexts. Medication labels, appointment letters, consent forms, and discharge instructions all assume reading proficiency that many patients lack. Numerical concepts pose particular difficulties. Understanding that a medication reduces heart attack risk from 4% to 3% requires comfort with percentages and the ability to recognise that "25% relative reduction" and "1% absolute reduction" describe the same effect very differently. Even well-educated patients often struggle with probabilistic information, which is precisely what informed medical decision-making requires.

Beyond these foundational skills, communicative health literacy enables people to actively seek information, ask questions, and adapt to changing circumstances. The patient who can read discharge instructions still needs communicative literacy to recognise when symptoms deviate from expected recovery, formulate questions about concerning changes, and navigate the healthcare system to get appropriate follow-up. These interactive capabilities prove essential for managing anything beyond the simplest acute conditions.

A third dimension, critical health literacy, has become increasingly important in an era of abundant health information of highly variable quality. Distinguishing reliable sources from unreliable ones, recognising when personal testimonials and dramatic anecdotes should not outweigh systematic evidence, and understanding how commercial interests shape health information all require analytical skills that formal education often fails to develop. The patient researching treatment options online encounters peer-reviewed research summaries alongside supplement advertisements, celebrity wellness advice, and conspiracy theories about pharmaceutical companies. Critical health literacy means sorting through this information without being manipulated.

Table 10.1: Dimensions of Health Literacy

Digital health tools add yet another layer of complexity. Someone must now possess technological skills to benefit from portals, apps, and devices: creating accounts, navigating interfaces, troubleshooting problems, and recognising security risks. For populations already disadvantaged by limited traditional health literacy, these digital requirements raise additional barriers. The patient who struggled to understand paper medication instructions does not find the task easier when the same information appears on a smartphone app.

Designing for these realities requires deliberate effort. Plain language that avoids jargon, clear visual organisation, limited information volume, strategic use of images and multimedia, and teach-back methods that confirm understanding through patient explanation all help bridge literacy gaps (AHRQ, 2023). Yet inclusive design requires humility: the recognition that most health communicators overestimate how clearly their materials communicate to diverse audiences.

Addressing the Digital Divide

A rural GP surgery invests significantly in a patient portal, celebrating when two-thirds of patients activate their accounts. But deeper analysis reveals a troubling pattern: activation rates approach 90% in the affluent village three miles away while barely reaching 40% in the deprived housing estate closer to the surgery itself. The estate has patchy mobile signal, residents are more likely to be older or have left school early, and many households lack home broadband. For these patients, the portal might as well not exist. As the surgery increasingly routes appointments, results, and communication through digital channels, they find themselves marginalised from their own care. The digital divide, which describes disparities in technology access, connectivity, and skills, risks translating technological progress into healthcare inequity.

The divide manifests in interconnected ways. Physical access remains unevenly distributed: rural areas often lack the broadband infrastructure that urban populations take for granted, and lower-income households may have only smartphone internet access, functional for social media but limiting for the detailed reading and form completion that patient portals often require. Beyond access, skills disparities mean that even people with devices and connectivity may struggle to use digital tools effectively. Age, educational background, prior technology exposure, and cognitive factors all shape digital capability, and these characteristics are not randomly distributed across populations.

Less visible but equally consequential, design disparities exclude people whose needs the original developers never considered. A portal that works perfectly for sighted English speakers may be utterly inaccessible to someone using a screen reader or someone whose primary language is Polish or Urdu. Web Content Accessibility Guidelines exist precisely because good intentions do not automatically produce inclusive design (W3C Web Accessibility Initiative, 2023); accessibility requires deliberate attention throughout development. Similarly, multilingual interfaces are essential for equitable care in diverse communities.

Addressing these disparities requires concrete organisational action (ONC, 2023). Healthcare organisations should begin by actually assessing their patient populations' digital circumstances rather than assuming that national statistics apply to their specific community. What proportion have reliable broadband? Smartphones but not computers? Limited English proficiency? These questions have answers, and those answers should shape strategy. Alternative pathways (telephone booking, postal reminders, in-person registration) must remain genuinely functional options rather than grudging afterthoughts for patients who cannot or choose not to engage digitally. Proactive outreach can help interested patients develop digital skills, while community partnerships with libraries, senior centres, and community organisations extend support beyond healthcare settings into trusted local environments.

The key question organisations must answer is whether digital initiatives will serve all patients or primarily those already most advantaged. Technology itself is neither equitable nor inequitable; those outcomes depend entirely on the choices made in implementation.

Patient Perspective: Mr Kovač (Hypothetical)

Mr Kovač is a hypothetical patient whose experience reflects common barriers faced by patients with limited English proficiency in digital health. A 58-year-old Croatian immigrant working in construction, he has Type 2 diabetes and limited English literacy. When his GP surgery launched a patient portal, he received login details but never activated the account. The interface was only in English, and he found the medical terminology confusing even with translation apps. His diabetes control deteriorated because he missed appointment reminders and could not easily request prescription refills. The surgery identified the problem when analysing portal usage disparities by postcode. They now offer portal registration support sessions with an interpreter, have enabled text message reminders in multiple languages, and maintain telephone booking for patients who prefer it. His daughter helped him download a translation app on his phone, and he now uses the portal with her help during visits. His story illustrates how digital health tools designed without considering linguistic and literacy diversity can inadvertently exclude the patients who might benefit most.

Summary

Digital tools can meaningfully support patient engagement - portals provide transparency that was once difficult to obtain, decision aids help patients participate as genuine partners in consequential choices, and shared decision-making tools surface the personal values that should inform treatment. The evidence for these tools is encouraging, particularly for decision aids, which consistently improve patient knowledge and produce choices more aligned with individual preferences.

Yet the chapter's recurring theme is that these benefits are not distributed equally. Portal adoption is lowest among those with the greatest disease burden. Health literacy limitations affect over a third of adults, and digital tools add technological demands on top of existing literacy requirements. The digital divide means that as healthcare organisations route more services through digital channels, they risk marginalising the patients who can least afford to be excluded. Addressing this means auditing the digital circumstances of the actual patient population, funding telephone and postal routes well enough to keep them usable, and testing interfaces against accessibility guidelines and with non-English speakers before launch rather than after complaints.