Clinical excellence that withstands scrutiny

All health technology companies advocate their best-in-class solutions. 

Objective research, peer reviewed and published by the scientific community, helps people distinguish between claims and clinical evidence to judge for themselves. 

Industry benchmarking allows health companies to compete. Scientific scrutiny allows them to lead.

Peer-reviewed studies and publications in collaboration with Ada

  • Patients’ utilization and perception of an artificial intelligence based symptom assessment and advice technology in a British primary care waiting room: Exploratory pilot study.
    JMIR Hum Factors (2020) doi: 10.2196/19713.

    Miller, S., et al.

    Publication typeExploratory pilot study
    Findings
    • Ada performed well in a study of 523 patients in a primary care setting.
    • Ada was rated very or quite easy to use by 97.8 percent of participants.
    • With Ada’s advice, 12.8 percent of participants would have used lower-intensity care or delayed their appointment.
  • Rare diseases 2030: How augmented AI will support diagnosis and treatment of rare diseases in the future.
    Ann Rheum Dis (2020) doi: 10.1136/annrheumdis-2020-217125.

    Hirsch, M. C., et al.

    Publication typeFuture view
    Findings
    • Within the next decade, specialists may work closely with tools like Ada to simplify the rare disease care journey.
    • Ada could support detection of genetic autoinflammatory disorders such as familial Mediterranean fever.
    • Chatbots ask for red flag symptoms to exclude emergencies.
  • Can a decision support system accelerate rare disease diagnosis? Evaluating the potential impact of Ada DX in a retrospective study.
    Orphanet J Rare Dis (2019) doi: 10.1186/s13023-019-1040-6.

    Ronicke, S., et al.

    Publication typeRetrospective real-patient study
    Findings
    • Ada’s top suggestion matched the confirmed diagnosis in 89.25% of cases (83 of 93 cases).
    • In more than 56% of cases, Ada provided correct disease suggestions earlier than the time of clinical diagnosis.
    • More than 33% of patients could have been identified as having a rare disease in the first documented clinical visit.

Conference posters and preprints of studies, by or in collaboration with Ada, submitted for peer review

  • How accurate are digital symptom assessment apps for suggesting conditions and urgency advice?: A clinical vignettes comparison to GPs.
    medRxiv (2020) doi: 10.1101/2020.05.07.20093872.

    Gilbert, S., et al.

    Publication typeComparative vignette study
    Findings
    • Ada outperformed competitors, providing safe advice 97 percent of the time with 99 percent condition coverage.
    • Ada achieved 70 percent accuracy for the top three suggestion fit. The closest competitor achieved 43 percent accuracy, and the overall average for competitors was 38 percent.
  • Syndromic surveillance insights from a symptom assessment app before and during COVID-19 measures in Germany and the United Kingdom: Results from repeated cross-sectional analyses.
    medRxiv (2020) doi: 10.1101/2020.06.16.20126466.

    Mehl, A., et al.

    Publication typeComparative descriptive analysis
    Findings
    • Ada can act as a real-time syndromic surveillance system that generates novel health insights.
    • Ada could facilitate understanding of the implications of public health policies like the COVID-19 lockdown measures. 
    • Ada can help complement and cross-validate hypotheses based on data collected through more traditional channels.
  • The potential for digital patient symptom recording through symptom assessment applications to optimize patient flow and reduce waiting times in Urgent Care Centers: A simulation study.
    medRxiv (2020) doi: 10.1101/2020.06.19.20135590.

    Montazeri, M., et al.

    Publication typeSimulation study
    Findings
    • Ada could decrease the average patient waiting time to see a triage nurse by 54 percent.
    • Reducing the time to see the triage nurse can positively impact patient and staff anxiety and improve quality of care.
  • Just how rare are rare diseases on Ada, a medical symptom checker?
    Findings
    • Rare diseases were Ada’s top fit suggestion in 4 percent of 15 million user assessments, in keeping with estimated rates in the general population. 
    • Rare diseases appeared in Ada’s top five suggestions in 17 percent of assessments.
    • Ada could help shorten the diagnostic odyssey of people living with rare diseases.

Peer-reviewed papers and preprints in third-party publications that analyze or discuss Ada

  • Position paper of the commission on digital rheumatology of the German Society of Rheumatology: Tasks, targets, and perspectives for a modern rheumatology.
    Zeitschrift für Rheumatologie (2020) doi: 10.1007/s00393-020-00834-y.

    Knitza, J., et al.

    Publication typePosition paper
    Findings
    • Ada was validated as a multidisciplinary symptom assessment tool that creates lists of condition suggestions.
    • Ada provided advice on the degree of urgency and the appropriate care destination.
  • Accuracy of online symptom checkers and the potential impact on service utilisation.
    medRxiv (2020) doi: 10.1101/2020.07.07.20147975.

    Ceney, A., et al.

    Publication typeComparative vignette study
    Findings
    • Ada achieved 73 percent accuracy compared to an all app average of 38 percent. 
    • Ada’s top five suggestion accuracy was 84 percent compared to a 51 percent all app average.
    • Ada had 97 percent disposition safety compared to an all-app average of 83 percent.
  • Ethical and legal challenges of artificial intelligence-driven healthcare.
    Artificial Intelligence in Healthcare (2020) doi: 10.1016/B978-0-12-818438-7.00012-5.

    Gerke, S., Minssen, T., & Cohen, G.

    Publication typeBook chapter
    Findings
    • Ada is an important example of an AI health application developed in the European regulatory setting.
    • Ada was identified as a CE-marked (class I under the Medical Device Directive) medical device that complies with the EU General Data Protection Regulation 2016/679 (GDPR).
  • Can mobile health apps replace GPs? A scoping review of comparisons between mobile apps and GP tasks.
    BMC Med Inform Decis Mak (2020) doi: 10.1186/s12911-019-1016-4.

    Wattanapisit, A., et al.

    Publication typeScoping review
    Findings
    • Ada can provide primary care information and health-related suggestions to users.
    • Ada can produce a list of provisional and differential diagnoses based on the information collected by AI.
    • mHealth-related expertise should be recognised as an essential competency for GPs.
  • Accuracy of a chatbot (Ada) in the diagnosis of mental disorders: Comparative case study with lay and expert users.
    JMIR Form Res (2019) doi: 10.2196/13863.

    Jungmann, S. M., et al.

    Publication typeComparative vignette study
    Findings
    • Ada could assist diagnosticians as a screening tool to identify mental disorders in adults.
    • Ada was tested with 20 case vignettes covering a broad spectrum of mental disorders.
    • Ada asked an average of 34 questions and took seven minutes to complete mental disorder assessments.
  • From symptom to diagnosis-symptom checkers re-evaluated : Are symptom checkers finally sufficient and accurate to use? An update from the ENT perspective
    HNO (2019) doi: 10.1007/s00106-019-0666-y.

    Nateqi, J., et al.

    Publication typeComparative vignette study
    Findings
    • Ada was the second-best performing app out of 24 tested in the ENT discipline.
  • What happened when Pulse tested symptom checker apps.
    Pulse Today (2019).

    Publication typeComparative vignette assessment
    Findings
    • Experienced GPs tested four symptom assessment apps against each other: NHS, Babylon, Ada, and Your.MD.
    • Ada showed the highest accuracy for disposition advice and condition suggestion.
    • Ada is fast, easy to use, and could help identify serious conditions such as heart attack.
  • Can you really trust the medical apps on your phone?
    Wired UK (2017).

    Burgess, M.

    Publication typeComparative vignette assessment
    Findings
    • Symptom assessment apps’ performance can vary significantly when tested against each other using the same symptoms.
    • Three symptom assessment apps were tested against each other: Ada, Babylon, and Your.MD.
    • Ada was ‘by far the best’ app tested, asking clear questions about the most important symptoms and providing the best condition suggestions.

Preprints of studies and publications by or in collaboration with Ada

  • The quality of condition suggestions and urgency advice provided by the Ada symptom assessment app assessed with independently generated vignettes optimized for Australia.
    medRxiv (2020) doi: 10.1101/2020.06.16.20132845.

    Gilbert, S., et al.

    Publication typeComparative vignette study
    Findings
    • Ada’s top condition suggestion was correct in 65 percent of cases, and the correct condition was in the top three in 83 percent.
    • Ada’s urgency advice matched the gold standard in 63 percent of cases, including 67 percent of urgent cases and 57 percent of those less serious.