
Oxford, UK — In a groundbreaking development for healthcare communication, Oxford-based company Aide Health has launched Mirror, an artificial intelligence-powered application designed to tackle one of medicine’s most pervasive yet overlooked challenges: patients’ inability to remember vital information from medical appointments.
The statistics surrounding patient recall of medical information are sobering and represent a significant risk factor in healthcare delivery. Research consistently demonstrates that patients struggle to retain critical information shared during consultations:
“These aren’t just statistics—they represent real risks to patient safety and treatment outcomes,” explains Dr. Christopher Sharp, Chief Medical Information Officer at Stanford Health Care. “When patients can’t remember medication dosages, follow-up appointment dates, or warning signs to watch for, it can have serious health consequences.”
| Recall Type | Percentage | Timeframe |
|---|---|---|
| Accurate recall without prompting | 49% | 1 week post-visit |
| Accurate recall with prompting | 36% | 1 week post-visit |
| Incorrect or no recall | 15% | 1 week post-visit |
| Overall information forgotten | 40-80% | Post-discharge |
| Information remembered incorrectly | ~50% | Varies |
Ian Wharton, creator of Mirror and founder of Aide Health, was driven to develop the platform after witnessing his father’s struggles with early-stage Alzheimer’s disease. “It was the vast majority of information that was given to him that he just couldn’t recall, and my biggest fear was not being by his bedside when he was told something important,” Wharton explained.
The need for such technology becomes even more critical when considering vulnerable populations. Over 7.2 million Americans aged 65 and older are living with Alzheimer’s disease in 2025, with episodic memory—the type used to recall medical appointments—being particularly affected by cognitive decline.
“They’re moments of quite often intense information, which are very brief for you to digest often high-stakes news,” Wharton said, highlighting the perfect storm of stress, information overload, and time pressure that characterizes many medical consultations.
| Factor | Impact on Recall |
|---|---|
| Number of instructions (3-4 items) | 26-56 times worse recall |
| Age (older adults) | Moderate negative correlation |
| Education level (college vs. less than high school) | 60-70% improvement in recall |
| Stress and anxiety during consultation | Significant reduction via attentional narrowing |
| Provider talk time (higher ratio) | Worse recall (β = -0.04, p = 0.01) |
| Structured vs. unstructured information | 42% improvement with structure (8.1 vs 5.7 items) |
The Mirror app employs sophisticated artificial intelligence to listen during medical appointments and automatically generate detailed, structured summaries of consultations. The technology represents what Wharton describes as “a really modern application of AI to solve a challenge that is this hidden risk in every healthcare consultation that we have.”
Janette Alfrey, one of Mirror’s first users, described the experience as transformative. “It had given me a heading as to why I was there, and then it had talked about what was about to happen and why they were doing it, and what was going to happen when they did the next operation,” she said.
The practical benefits extended beyond her own understanding. “When friends and colleagues phoned up over the next couple of days asking what was going to happen, I could just send them a screenshot of what I was told. It saved me having to go through everything again.”
This aligns with research showing that consultation recordings benefit patients by reducing anxiety, enhancing information retention, enabling better-informed decision-making, and improving communication with family members.
In an era of heightened concern about healthcare data privacy, Wharton emphasized Mirror’s commitment to patient data ownership. “Patients’ data recorded by the app is theirs, we don’t do anything with it—we don’t sell it, we don’t give it to third parties,” he stated categorically.
The current version of Mirror operates during in-person consultations only, though expansion plans are in development.
Wharton envisions Mirror evolving far beyond its current passive recording and summarization capabilities. “Right now our app is passive—it listens and summarizes for you—but in the future it will be your advocate, it will chirp, it will speak out if it thinks there’s something you should ask,” he explained.
This proactive approach could address another critical gap in healthcare communication: patients often don’t know what questions to ask or when important topics are being glossed over. “That’s where this technology is going, and this is absolutely where we’re going in how we manage our health,” Wharton added.
Mirror joins a growing ecosystem of applications designed to improve patient-provider communication through recording and AI technology. Here’s how it compares:
| App Name | Developer | Primary Features | Target Users | Key Differentiator |
|---|---|---|---|---|
| Mirror | Aide Health (UK) | AI summaries, in-person recording, secure sharing | Patients, especially those with memory issues | Advanced AI summarization with future advocacy features |
| Medcorder | Medcorder Inc. (US) | Recording, transcription, photo sharing, family sharing | General patients, caregivers | Free app with focus on family communication |
| The Medical Memory | The Medical Memory (US) | Video recording, provider and patient apps, educational videos | Patients and healthcare providers | Dual-sided platform with institutional endorsement |
| My Care Conversations | Cancer Care Alberta (Canada) | Recording, note-taking, organizationally endorsed | Cancer patients, general hospital patients | First organizationally endorsed app in North America |
| Rev Voice Recorder | Rev.com (US) | Recording with AI transcription and note-taking | General patients | Multi-purpose with medical use case; HIPAA compliant |
| App Name | Focus | Key Statistics | Integration |
|---|---|---|---|
| DAX Copilot | Microsoft/Nuance | 96% easy to use, 78% expedites note-taking | Stanford, Atrium Health |
| Abridge | Enterprise healthcare software | Mayo Clinic: 2,000+ physicians | Epic EHR integration |
| DeepScribe | Ambient AI documentation | Used at Ochsner Health | Value-based care focus |
| Freed AI | Independent clinicians | Clinicians report leaving work on time | 14+ languages |
| Sunoh.ai | Medical scribe | 80,000+ healthcare providers, saves 2 hrs/day | Multi-specialty, EHR integration |
| Heidi AI | Global healthcare | 100% customer retention (KLAS Research) | Offline capability |
| Lyrebird Health | Australian market | <20 second note generation | Web-based, no installation |
Key Distinction: While clinician-focused tools help doctors reduce administrative burden, patient-focused apps like Mirror prioritize patient understanding and information retention—addressing different but complementary aspects of the healthcare communication challenge.
“One of the things that was most remarkable to me was my ability to turn away from my keyboard, face the patient and really listen, knowing that everything shared in our conversation was being documented,” says Dr. Christopher Sharp of Stanford Health Care, describing AI scribe technology.
This physician perspective highlights how recording technology can paradoxically make consultations more personal by removing the barrier of note-taking.
Research by cognitive psychologists has identified specific phenomena that impair medical information recall:
Attentional Narrowing: When doctors deliver serious diagnoses, patients focus intensely on the central message (“you have X disease”) while peripheral information about treatment adherence or follow-up appointments fails to be processed and stored in memory.
State-Dependent Learning: Information given in a stressful clinical environment is best recalled in a similar state, yet patients need to remember it in their relaxed home environment—creating a mismatch that impairs recall.
Research identifies several communication techniques associated with better recall:
However, studies note these techniques are uncommonly used due to time pressures in clinical practice, making technological solutions like Mirror increasingly valuable.
The launch of Mirror comes at a critical time for healthcare communication. The COVID-19 pandemic accelerated the adoption of telemedicine, creating new challenges for information retention as consultations became more frequent but less face-to-face.
Additionally, the aging global population means more patients are dealing with conditions that affect memory and cognition, making tools like Mirror not just helpful but essential for safe, effective care.
The implications of poor patient recall extend beyond individual outcomes:
The legal landscape around recording medical appointments varies significantly:
Best practice recommendations emphasize informed consent and transparency, with patients explaining why they want to record and how they’ll use the recording.
Cancer Care Alberta’s implementation of their My Care Conversations app provides valuable lessons for patient recording platforms:
The approach resulted in a sustained adoption since the November 2018 launch, with ongoing improvements based on user reviews.
Research specifically examining patient recording of consultations has documented measurable benefits:
Mirror represents the vanguard of a broader transformation in how patients interact with the healthcare system. As Wharton articulates, the future involves AI not just as a passive tool but as an active participant in healthcare conversations.
Proactive Question Prompting: Future versions could alert patients in real-time if important topics aren’t being addressed or if medical jargon needs clarification.
Personalized Health Literacy: AI could automatically adjust summary complexity based on individual patient health literacy levels.
Longitudinal Health Tracking: Integration with electronic health records could provide patients with comprehensive health timelines, connecting information across multiple visits and providers.
Multi-language Support: Automatic translation and transcription could break down language barriers in healthcare.
Telemedicine Integration: Expansion beyond in-person consultations to virtual visits, addressing the rapid growth of remote healthcare.
Mirror by Aide Health addresses a critical yet often overlooked gap in healthcare: the persistent challenge of patient information recall. With nearly half of medical information forgotten or misremembered, tools that help patients retain and understand their healthcare information aren’t just convenient—they’re potentially life-saving.
As healthcare systems worldwide grapple with increasing complexity, aging populations, and resource constraints, AI-powered solutions like Mirror offer a pathway to better outcomes without requiring systemic overhaul. By empowering patients with accurate, accessible records of their medical consultations, these technologies shift the balance toward more informed, engaged, and effective healthcare participation.
For Ian Wharton, the motivation remains deeply personal: ensuring his father—and millions like him—never have to face the anxiety and risk of forgotten medical information. But the impact extends far beyond individual families to potentially transform how we all navigate our healthcare journeys.
✓ Patients forget 40-80% of medical information shared during appointments
✓ Mirror uses AI to record, transcribe, and summarize medical consultations
✓ Inspired by personal experience with early-stage Alzheimer’s disease
✓ Patient data ownership is central to the platform’s privacy approach
✓ Future versions will offer proactive advocacy, alerting patients to important questions
✓ Similar apps exist but Mirror’s AI summarization and future advocacy features offer unique value
✓ Clinical evidence supports the benefits of consultation recording for patient outcomes
✓ Growing market of both patient-focused and clinician-focused AI healthcare communication tools
Sources: Research studies from PLOS One, Journal of the Royal School of Medicine, Stanford Medicine, Cancer Care Alberta, Alzheimer’s Association, and various peer-reviewed medical journals.
Recent Posts