Sarah Redston1, Sharon de Botte1, Carl Smith2. 1. Health Press Ltd, Elizabeth House, Queen Street, Abingdon, UK. 2. School of Biology, University of St Andrews, St Andrews, UK; Department of Ecology & Vertebrate Zoology, University of Łodz, Łodz, Poland. Electronic address: cs101@st-andrews.ac.uk.
Abstract
PURPOSE: Reliance on online health information is proliferating and the Internet has the potential to revolutionize the provision of public health information. The anonymity of online health information may be particularly appealing to people seeking advice on 'embarrassing' health problems. The purpose of this study was to investigate (1) whether data generated by the embarrassingproblems.com health information site showed any temporal patterns in problem resolution, and (2) whether successful resolution of a medical problem using online information varied with the type of medical problem. METHODS: We analyzed the responses of visitors to the embarrassingproblems.com website on the resolution of their problems. The dataset comprised 100,561 responses to information provided on 77 different embarrassing problems grouped into 9 classes of medical problem over an 82-month period. Data were analyzed with a Bernoulli Generalized Linear Model using Bayesian inference. RESULTS: We detected a statistically important interaction between embarrassing problem type and the time period in which data were collected, with an improvement in problem resolution over time for all of the classes of medical problem on the website but with a lower rate of increase in resolution for urinary health problems and medical problems associated with the mouth and face. As far as we are aware, this is the first analysis of data of this nature. CONCLUSIONS: Findings support the growing recognition that online health information can contribute to the resolution of embarrassing medical problems, but demonstrate that outcomes may vary with medical problem type. The results indicate that building data collection into online information provision can help to refine and focus health information for online users.
PURPOSE: Reliance on online health information is proliferating and the Internet has the potential to revolutionize the provision of public health information. The anonymity of online health information may be particularly appealing to people seeking advice on 'embarrassing' health problems. The purpose of this study was to investigate (1) whether data generated by the embarrassingproblems.com health information site showed any temporal patterns in problem resolution, and (2) whether successful resolution of a medical problem using online information varied with the type of medical problem. METHODS: We analyzed the responses of visitors to the embarrassingproblems.com website on the resolution of their problems. The dataset comprised 100,561 responses to information provided on 77 different embarrassing problems grouped into 9 classes of medical problem over an 82-month period. Data were analyzed with a Bernoulli Generalized Linear Model using Bayesian inference. RESULTS: We detected a statistically important interaction between embarrassing problem type and the time period in which data were collected, with an improvement in problem resolution over time for all of the classes of medical problem on the website but with a lower rate of increase in resolution for urinary health problems and medical problems associated with the mouth and face. As far as we are aware, this is the first analysis of data of this nature. CONCLUSIONS: Findings support the growing recognition that online health information can contribute to the resolution of embarrassing medical problems, but demonstrate that outcomes may vary with medical problem type. The results indicate that building data collection into online information provision can help to refine and focus health information for online users.