Variation in Quality of Women's Health Topic Information from Systematic Internet Searches

Background/Objectives: The internet has unquestionably altered how people acquire health information. Instead of consulting with a medical professional, billions of pages of information can be accessed by anyone with a smartphone. Women's health issues have been historically and culturally taboo in many cultures globally; therefore, internet searches may be particularly useful when researching these topics. Methods: As an exercise in scientific information evaluation, we chose 12 non-cancer topics specific to women's health and developed a scoring metric based on quantifiable webpage attributes to answer: What topics generate the highest and lowest scores? Does the quality of information (mean score) vary across topics? Does the variation (score deviation) differ among topics? Data were collected following systematic searches after filtering with advanced features of Google and analyzed in a Bayesian framework. Results: The mean score per topic was significantly correlated with the number of sources cited within an article. There were significant differences in the quality scores across topics; "pregnancy" and "sleep" scored the highest and had more sources cited per page than all other topics. The greatest variation in scores were for "cortisol" and "weight". Conclusions: A purposeful, systematic internet search of 12 critical women's health topics suggests that scrutiny is necessary when this information is obtained by a typical internet user. Future work should include review by medical professionals based on their interaction with patients who self-report what they know or think about a condition they present and respect, while educating, patients' own internet searching.