Evidence Base
The published research behind this product’s design decisions. Every claim is one click from its source.
Everything below comes from peer-reviewed studies, patient registries, and survey research conducted by other people. We have not interviewed any of these participants.
That distinction is not pedantry. This evidence tells us what is generally true of populations like our users. It cannot tell us whether this screener is confusing, whether this report gets read, or whether anyone will upload a lab report to this site. Those need primary interviews, and we have conducted zero.
each traced to a source
peer-reviewed and registry
changed a decision
still outstanding
Clinician barriers (8)
F-CLIN-0186% of Lyme-treating clinicians cite inadequate physician education about tick-borne disease as a cause of diagnostic delay.
What it meant for the product: The report's audience often does not know the material. It must teach, briefly, without condescending.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-02Critical61% identify false-negative laboratory tests as a cause of diagnostic delay.
What it meant for the product: A negative two-tier result is the single most common reason a correct suspicion gets dropped. The report must state the sensitivity limitation next to any test recommendation.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-0352% cite prior misdiagnosis as a cause of delay; 37% cite absence of an erythema migrans rash.
What it meant for the product: Prior misdiagnosis is both a consequence and a cause. Capturing it gives a clinician the thread to pull.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-0463% of Lyme-treating clinicians cannot book an initial consultation within one month; 19% have 2-4 month waits and 8% wait more than four months.
What it meant for the product: Referring a patient to a specialist is referring them to a queue. Urgent-tier guidance must route to urgent care, not to a specialist waitlist.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-0574% of these clinicians do not participate in insurance networks and 77% do not participate in Medicare, Medicaid, or government-supported plans; 75% identify patient inability to pay as a barrier.
What it meant for the product: A provider directory that ignores payment is a directory of care the patient cannot afford. Insurance status is not a filter, it is the filter.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-06Critical57% of Lyme-treating clinicians cite patients' cognitive impairment as a barrier to care; 79% cite illness complexity.
What it meant for the product: Brain fog is not a symptom to record, it is an accessibility constraint on every screen we ship. Long forms and dense reports actively exclude the target user.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-07Critical39% of Lyme-treating clinicians have been reported to a medical board, insurer, or subjected to a hospital quality inquiry; 75% report stigmatisation or disrespect from professional colleagues.
What it meant for the product: The clinicians best equipped to help are professionally exposed. Anything we publish that could be read as directing treatment increases their risk. Signpost, never prescribe.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
F-CLIN-0877% experience prior-authorisation denials and 71% face insurance claim denials.
What it meant for the product: Documentation that supports prior authorisation is a concrete, unglamorous thing the report could do.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
Diagnostic delay (7)
F-DELAY-0151% of Lyme patients reported it took more than three years to be diagnosed.
What it meant for the product: The screener is used by people mid-odyssey, not at symptom onset. Designing for day-one presentation is designing for the minority.
LymeDisease.org. MyLymeData Chart Book: 2019 Highlights. MyLymeData Patient Registry. source
F-DELAY-0254% saw five or more clinicians before diagnosis.
What it meant for the product: Clinician count is a strong, cheap signal of odyssey length and nobody asks for it. Capturing it costs one field.
LymeDisease.org. MyLymeData Chart Book: 2019 Highlights. MyLymeData Patient Registry. source
F-DELAY-0370% of registry patients were not diagnosed until they had late Lyme disease, i.e. six months or later after symptom onset.
What it meant for the product: The early-treatment window is missed in the large majority of the population this product serves.
LymeDisease.org. MyLymeData Chart Book: 2019 Highlights. MyLymeData Patient Registry. source
F-DELAY-04Appraisal delay was driven by symptom misattribution, intermittent symptoms, and the misperception that a bull's-eye rash is required.
What it meant for the product: The rash misconception is correctable in one sentence and directly shortens the first phase of delay.
Hirsch AG, Herman RJ, Rebman A, Moon KA, Aucott J, Heaney C, Schwartz BS. Obstacles to diagnosis and treatment of Lyme disease in the USA: a qualitative study. BMJ Open. 2018;8(6):e021367. source
F-DELAY-05Critical10 of 26 participants bypassed scheduling delay entirely by using urgent care or emergency departments, but misdiagnosis was more common in those settings, causing treatment delay.
What it meant for the product: Speed and accuracy trade off. Sending someone to urgent care buys time and costs accuracy, so what they carry in with them matters more there, not less.
Hirsch AG, Herman RJ, Rebman A, Moon KA, Aucott J, Heaney C, Schwartz BS. Obstacles to diagnosis and treatment of Lyme disease in the USA: a qualitative study. BMJ Open. 2018;8(6):e021367. source
F-DELAY-06Insurance status was a driver of both illness delay and behavioural delay.
What it meant for the product: Cost is upstream of care-seeking, not downstream of it.
Hirsch AG, Herman RJ, Rebman A, Moon KA, Aucott J, Heaney C, Schwartz BS. Obstacles to diagnosis and treatment of Lyme disease in the USA: a qualitative study. BMJ Open. 2018;8(6):e021367. source
F-DELAY-07CriticalOnly 5% of clinicians attribute diagnostic delay to patients delaying seeking care.
What it meant for the product: Delay is a system property, not a patient failing. Any nudge implying the patient waited too long is both wrong and insulting.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
Geographic dismissal (3)
F-GEO-01Critical47% reported testing was delayed or denied because Lyme disease was not considered common in their state.
What it meant for the product: THE finding for county-level data. Nearly half of patients are turned away on a geographic argument. County incidence is the direct counter-evidence and it is federal, citable, and already in the product.
LymeDisease.org. MyLymeData Chart Book: 2019 Highlights. MyLymeData Patient Registry. source
F-GEO-0260% reported diagnosis was delayed on the basis of perceived low state incidence; 69% of those reported the delay exceeded a year.
What it meant for the product: The geographic argument does not just delay testing, it costs a year. County resolution is not a nice-to-have.
LymeDisease.org. MyLymeData Chart Book: 2019 Highlights. MyLymeData Patient Registry. source
F-GEO-0325% of Lyme-treating clinicians identify patient residence in a perceived low-incidence state as a cause of diagnostic delay.
What it meant for the product: Clinician-side corroboration of the patient-reported geographic dismissal. Both sides name the same barrier.
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. source
Life impact (3)
F-WORK-01Participants spontaneously reported the detrimental impact of Lyme disease on ability to work and on caregiving responsibilities, without being asked about it.
What it meant for the product: Unprompted disclosure is the strongest qualitative signal there is. Work and caregiving impact belongs in the report even though no clinician asks for it.
Hirsch AG, Herman RJ, Rebman A, Moon KA, Aucott J, Heaney C, Schwartz BS. Obstacles to diagnosis and treatment of Lyme disease in the USA: a qualitative study. BMJ Open. 2018;8(6):e021367. source
F-IDENTITY-01Loss of work and social withdrawal were central to the illness experience.
What it meant for the product: Energy spent on the product is energy not spent elsewhere. Every additional screen has a real cost to this user.
“I was in excellent health before ...and that's what really upsets me the most is the Lyme has taken away my job that I love, making me not be able to work.”
“I don't feel like getting myself all riled up and dressed up to go out on a social occasion.”
Verbatim participant quotes as published in the source.
Ali A, Vitulano L, Lee R, Weiss TR, Colson ER. Experiences of patients identifying with chronic Lyme disease in the healthcare system: a qualitative study. BMC Fam Pract. 2014;15:79. source
F-IDENTITY-02Patients resisted identification with more severely affected patients as a way of preserving hope.
What it meant for the product: Worst-case framing repels. Severity language should describe risk, never destiny.
“I went to the first Lyme support group meeting; there were people there with canes, people in wheelchairs and I said, 'No way...not me.'”
Verbatim participant quotes as published in the source.
Ali A, Vitulano L, Lee R, Weiss TR, Colson ER. Experiences of patients identifying with chronic Lyme disease in the healthcare system: a qualitative study. BMC Fam Pract. 2014;15:79. source
Lab result access (3)
F-LAB-01Critical96% of 8,139 respondents preferred receiving immediately released test results online even if their clinician had not yet reviewed them.
What it meant for the product: Directly supports showing extracted lab values to the patient immediately rather than gating them behind clinician review.
Steitz BD, Turer RW, Lin C-T, et al. Perspectives of Patients About Immediate Access to Test Results Through an Online Patient Portal. JAMA Netw Open. 2023;6(3):e233572. source
F-LAB-02CriticalA subset of respondents experienced increased worry after receiving abnormal results; pre-counselling by the care team before the test was ordered was associated with reduced worry.
What it meant for the product: Anticipatory guidance BEFORE the result appears is the mitigation the evidence supports. Warn before extraction, not after.
Steitz BD, Turer RW, Lin C-T, et al. Perspectives of Patients About Immediate Access to Test Results Through an Online Patient Portal. JAMA Netw Open. 2023;6(3):e233572. source
F-LAB-03Clinicians' pre-implementation concerns that patients would be upset or confused largely did not materialise after rollout.
What it meant for the product: Paternalistic gating is not evidence-based. Do not withhold a value from the person whose body produced it.
Steitz BD, Turer RW, Lin C-T, et al. Perspectives of Patients About Immediate Access to Test Results Through an Online Patient Portal. JAMA Netw Open. 2023;6(3):e233572. source
Bringing information to a visit (3)
F-INFO-01CriticalMost patients who bring internet-sourced information to a consultation do so because they value the clinician's opinion on it, not to confront or make demands.
What it meant for the product: The report should be framed as a question to the clinician, not an argument. Tone is the difference between being read and being dismissed.
Ahluwalia S, Murray E, Stevenson F, Kerr C, Burns J. 'A heartbeat moment': qualitative study of GP views of patients bringing health information from the internet to a consultation. Br J Gen Pract. 2010;60(571):88-94. source
F-INFO-02CriticalInternet-informed patients can induce anxiety in GPs, who fear losing control of the consultation, being seen as ignorant, or being devalued.
What it meant for the product: A report that looks like a verdict threatens the clinician and gets discarded. A report that looks like structured history gets used.
Ahluwalia S, Murray E, Stevenson F, Kerr C, Burns J. 'A heartbeat moment': qualitative study of GP views of patients bringing health information from the internet to a consultation. Br J Gen Pract. 2010;60(571):88-94. source
F-INFO-03Patients reported a better relationship when the clinician was receptive to discussing the information they brought.
What it meant for the product: The upside of getting the tone right is not neutral, it is a better relationship. Worth designing for.
Ahluwalia S, Murray E, Stevenson F, Kerr C, Burns J. 'A heartbeat moment': qualitative study of GP views of patients bringing health information from the internet to a consultation. Br J Gen Pract. 2010;60(571):88-94. source
Data sharing (3)
F-DATA-01CriticalPooled willingness to share health data is 77%, but individual studies range from 24% to 100% depending on use and recipient. Privacy, consent, and transparency are the determining factors.
What it meant for the product: The A4 assumption is plausible but the variance is enormous and driven by exactly the things F3 was built around. This is supporting evidence, not proof.
Worldwide willingness to share health data high but privacy, consent and transparency paramount: a meta-analysis. npj Digit Med. 2025. source
F-DATA-02Willingness is high for sharing with one's own healthcare providers and much lower for secondary and third-party uses.
What it meant for the product: Framing matters: 'so you can show your doctor' is a supported use case, 'to improve our models' is not. Research consent must stay separate and off by default.
Worldwide willingness to share health data high but privacy, consent and transparency paramount: a meta-analysis. npj Digit Med. 2025. source
F-DATA-03Approximately 75% of US patients surveyed by the AMA in 2022 were concerned about the privacy of their personal health information, and most were unaware which companies had access to their data.
What it meant for the product: Stating plainly where the data goes is not boilerplate, it is the deciding factor for three quarters of users.
Worldwide willingness to share health data high but privacy, consent and transparency paramount: a meta-analysis. npj Digit Med. 2025. source
Trust in AI (2)
F-AI-01CriticalAround 89-90% of patients say a clear escalate-to-human option is essential for trusting AI support, and 91% say patients should be able to opt out of AI-driven clinical recommendations entirely.
What it meant for the product: An AI-only path with no human alternative fails for nine in ten users. A manual-entry route is not a nice-to-have.
Worldwide willingness to share health data high but privacy, consent and transparency paramount: a meta-analysis. npj Digit Med. 2025. source
F-AI-02CriticalPatients are roughly three times more likely to trust an AI agent embedded in a clinical system than one offered as a public chatbot.
What it meant for the product: As a standalone site we start from the lower-trust position. Provenance and verifiability have to do the work institutional context would otherwise do.
Worldwide willingness to share health data high but privacy, consent and transparency paramount: a meta-analysis. npj Digit Med. 2025. source
Patient–clinician relationship (1)
F-TRUST-01Contrasting doctor-patient relationships emerged as one of four major themes; patients distinguished sharply between clinicians who treated them as 'worthy of getting better' and those who had 'already made his mind up' before asking questions.
What it meant for the product: The report's job is to survive the second kind of clinician. It must be hard to dismiss on tone, which means citations and restraint, not persuasion.
“They treat you like, like you're someone in pain and you're someone that's worthy of getting better and getting treatment.”
“Walking in the door, he knew more than we did. Before asking us any questions at all, I felt he'd already made his mind up.”
“I couldn't continue to see someone who didn't believe I have a disease that I know I have.”
Verbatim participant quotes as published in the source.
Ali A, Vitulano L, Lee R, Weiss TR, Colson ER. Experiences of patients identifying with chronic Lyme disease in the healthcare system: a qualitative study. BMC Fam Pract. 2014;15:79. source
Johnson LB, Maloney EL. Access to Care in Lyme Disease: Clinician Barriers to Providing Care. Healthcare (Basel). 2022;10(10):1882. link
Cross-sectional survey, 30 closed items plus open text, Sep 23 - Dec 1 2021 · 155 clinicians across 30 US states; 45 provided open-text comments · Clinicians treating Lyme disease
Limitation: Recruited through LymeDisease.org's referral network, so it over-represents Lyme-focused practice. Read as 'what clinicians who treat this disease report', not as a random sample of US physicians.
LymeDisease.org. MyLymeData Chart Book: 2019 Highlights. MyLymeData Patient Registry. link
Patient-reported outcomes registry · 17,000+ enrolled patients · US patients with Lyme disease, skewing toward persistent/late-stage illness
Limitation: Self-selected registry. Over-represents patients whose illness did not resolve, which is precisely the population a screening tool needs to understand and precisely the population that resolves-quickly cohorts miss.
Hirsch AG, Herman RJ, Rebman A, Moon KA, Aucott J, Heaney C, Schwartz BS. Obstacles to diagnosis and treatment of Lyme disease in the USA: a qualitative study. BMJ Open. 2018;8(6):e021367. link
Qualitative, in-depth telephone interviews, analysed against the General Model of Total Patient Delay · 26 participants diagnosed 2014-2017 with a positive IgG western blot · Serologically confirmed Lyme patients, Pennsylvania region
Limitation: Serologic confirmation required, so the cohort excludes the seronegative early presentations where delay is often worst. Delay estimates here are conservative.
Ali A, Vitulano L, Lee R, Weiss TR, Colson ER. Experiences of patients identifying with chronic Lyme disease in the healthcare system: a qualitative study. BMC Fam Pract. 2014;15:79. link
Hermeneutic phenomenology; semi-structured face-to-face interviews, 60-90 minutes · 12 adults; mean age 41; 75% female; all White · Adults self-identifying with chronic Lyme disease
Limitation: Small and racially homogeneous. Themes are rich; prevalence claims cannot be drawn from it.
Steitz BD, Turer RW, Lin C-T, et al. Perspectives of Patients About Immediate Access to Test Results Through an Online Patient Portal. JAMA Netw Open. 2023;6(3):e233572. link
Cross-sectional survey across four US academic medical centres · 8,139 respondents · Patients receiving results via an online portal
Limitation: Academic medical centre portal users skew toward higher health literacy and engagement.
Ahluwalia S, Murray E, Stevenson F, Kerr C, Burns J. 'A heartbeat moment': qualitative study of GP views of patients bringing health information from the internet to a consultation. Br J Gen Pract. 2010;60(571):88-94. link
Qualitative semi-structured interviews with GPs · GPs in North Central London · General practitioners
Limitation: UK primary care, 2010. The dynamic it describes is durable but the setting is not US and predates the current portal era.
Worldwide willingness to share health data high but privacy, consent and transparency paramount: a meta-analysis. npj Digit Med. 2025. link
Meta-analysis of willingness-to-share studies · Pooled across studies, predominantly high-income countries · General and patient populations
The full insight → change ledger, including the assumptions still waiting on primary interviews, is in docs/user-research/traceability.md. The change worth knowing: 47% of patients report testing delayed or denied because Lyme “is not considered common” in their state (F-GEO-01). That is why the screening report now carries CDC incidence for the patient’s own county — try the screener and generate a report to see it.