Web-use for symptom appraisal for physical health conditions: A systematic review

Research output: Contribution to journalArticle

  • External authors:
  • Alan Davies
  • Julio Vega Hernandez


Background: The Web has become an important information source for appraising symptoms. We need to understand the role it currently plays in help-seeking and symptom evaluation, to leverage its potential to support healthcare delivery.
Objective: To systematically review the literature currently available on Web-use for symptom appraisal.
Methods: We searched Pubmed, Embase, PsycInfo, ACM digital library, SCOPUS, and Web of Science for any empirical studies that address the use of the Web by lay people to evaluate symptoms for physical conditions. Papers were excluded if they did not meet minimum quality criteria. Study findings were synthesised using a thematic approach.
Results: Thirty-two studies were included. Study designs included cross-sectional surveys, qualitative studies, experimental studies, and studies involving website/search engine usage data. Approximately 35% of adults engage in Web-use for symptom appraisal, but this proportion varies between 23-75% depending on sociodemographic and disease-related factors. Most searches are symptom-based rather than condition-based. Users view only the top search results and interact more with results that mention serious conditions. Web-use for symptom appraisal appears to impact on the decision to present to health services, communication with health professionals, and anxiety.
Conclusions: Web-use for symptom appraisal has the potential to influence the timing of help-seeking for symptoms, and the communication between patients and healthcare professionals during consultations. However, studies lack suitable comparison groups as well as follow-up of participants over time to determine whether Web-use results in healthcare utilisation and diagnosis. Future research should involve longitudinal follow-up so that we can weigh the benefits of Web-use for symptom appraisal (e.g. reductions in delays to diagnosis) against the disadvantages (unnecessary anxiety and healthcare use) and relate these to health care costs.

Bibliographical metadata

Original languageEnglish
JournalJournal of Medical Internet Research
Issue number6
Early online date13 Jun 2017
StatePublished - 2017