|Home | About | Journals | Submit | Contact Us | Français|
To determine the accuracy of information on infant sleep safety on the internet using Google. We hypothesized that the majority of websites would accurately reflect the AAP recommendations for infant sleep safety.
We searched for advice using 13 key phrases and analyzed the first 100 websites for each. Websites were categorized by type and assessed for accuracy of information provided, based on AAP recommendations. The accuracy of information was classified as “accurate,” “inaccurate,” or “not relevant.”
Overall, 43.5% of the 1300 websites provided accurate information, 28.1% provided inaccurate information, and 28.4% were not relevant. The search terms “infant cigarette smoking,” “infant sleep position,” and “infant sleep surface” resulted in the highest percentage of websites with accurate information. “Pacifier infant,” “infant home monitors,” and “infant co-sleeping” resulted in the lowest percentage of websites with accurate information. Government websites had the highest rate of accuracy, and blogs had the lowest rate of accurate information.
The internet frequently contains information about infant sleep safety that is inconsistent with AAP recommendations. Health care providers should realize the extent to which parents may turn to the internet for information about infant sleep safety.
The internet has increasingly become a commonly utilized source for health-related information and medical questions.1 In 2010, 59% of the U.S. population used the internet to search for health information,2, 3 and parents searching for health information regarding their children were among the top users.3 Furthermore, almost 70% of adults say that the information that they have found online has impacted upon their health decisions or actions.4 Access to health information on the internet thus has the potential to empower patients and revolutionize health care. Indeed, 61% of adults who use the internet for health information believe that it has improved how they care for themselves or someone else.4 One study found that increased access to health information, including internet information, was associated with a decrease in pediatric utilization.5
In addition, those who use the internet to search for health information generally trust the information found. In a national survey, 72% of adults said that one can believe most or all of the health information on the internet.4 However, depending on the topic and the website, the reliability and accuracy of health information available on the internet ranges from poor to excellent.6, 7 This is true for websites addressing pediatric concerns as well.7–9
The American Academy of Pediatrics (AAP) has published recommendations for infant sleep safety to reduce the risk of sudden infant death syndrome (SIDS), suffocation, strangulation, entrapment, and other accidental sleep-related infant deaths.10, 11 However, parents and health care professionals frequently have questions and concerns about specific recommendations,12–15 and the internet is likely a primary source of additional information. No study has documented the accuracy and reliability of information available on the internet regarding infant sleep safety.
Google has become one of the most popular internet search engines and is frequently used by both patients16 and physicians17 to access health-related and medical information. The typical adult seeking health information on the internet will begin, not at a medical website, but at a search website, and spend at least 30 minutes on a search, during which s/he will visit 2–5 websites.4 In order to determine the accuracy of information regarding infant sleep safety available to a typical user conducting an internet search, we attempted to simulate this experience by conducting internet searches on the Google website. We hypothesized that the majority of websites uncovered in our Google searches would accurately reflect the AAP recommendations for infant sleep safety.
We searched for advice with 13 key phrases relating to infant sleep safety (Table I). These key phrases were chosen to reflect specific AAP recommendations for infant sleep safety.10, 11 For each recommendation, up to two key phrases were used to perform a search on www.google.com. Websites were categorized by type (Table II; available at www.jpeds.com) and assessed for accuracy of the information provided, based solely on consistency with AAP recommendations10, 11 (Table I). Each appearance of a website was noted, as some websites appeared more than once in a single search. Type of website was determined by analysis of the website’s uniform resource locator (URL) and any section entitled “About Us” or something similar (Table II). The accuracy of the information found on each website was classified as “accurate” (consistent with current recommendations), “inaccurate” (inconsistent with current recommendations), or “not relevant” (did not address the topic, did not give advice, was a non-working website, or was unrelated to the key phrase).
Because we believed it unlikely that a typical search would extend further, the first 100 websites (8–10 pages of results) for each of the key phrases were analyzed. The first page (10–12 websites) of search results was also separately analyzed to determine accuracy of the websites that a typical searcher would most likely read. Four researchers independently performed the searches over a 2-month period. To minimize observer bias, strict definitions for acceptable advice were used (Table I) to determine the accuracy of a website’s content. If there was uncertainty or disagreement about the accuracy of a website’s content, this was resolved through discussion and ultimate consensus amongst the researchers. This study was granted exemption by the Children’s National Medical Center institutional review board.
Thirteen hundred websites (100 each for the 13 key phrases) were analyzed in July and August 2011 and re-analyzed for accuracy after the most recent AAP guidelines were published11, specifically with regards to the new recommendation against all bumper pad use. The proportion of websites with accurate information varied, depending on the key phrase that was used. Overall, 564 (43.4%) of the 1300 websites searched provided accurate information, 365 (28.2%) provided inaccurate information, and 370 (28.5%) were not relevant to the key phrase (Table III). When the websites that were not relevant were excluded, 60.8% of the websites provided accurate information.
The key phrases that resulted in the highest percentage of websites with accurate information were “infant cigarette smoking”, “infant sleep position”, and “infant sleep surface, “with 82%, 74%, and 73% of the respective websites providing accurate information. In each of the other 10 key phrase searches, less than 58% (range, 14–58%) of websites contained accurate information. The key phrases that resulted in the highest percentage of websites with inaccurate information were “pacifier infant,” “infant home monitors,” and “infant co-sleeping,” with 14%, 18%, and 20% of the respective websites providing accurate information. When the search results were limited to the first page of websites (Table IV), excluding the 24 websites that were not relevant, 67.3% of the websites contained accurate information.
When websites were categorized by type (Table V), the most common type of website encountered in our searches were company/interest groups (250), closely followed by retail product review (246) and educational websites (241). Analysis of the accuracy of information provided by the type of website demonstrated that government websites had the highest rate of accuracy, with accurate information being presented 80.9% of the time, followed by organization websites (72.5% accurate). Company/interest group, news, sponsored links, and educational websites had similar rates of accuracy, ranging from 50.6% to 52.4%. Thirty percent and 25.7%, respectively, of websites from individuals and blogs provided accurate information. Retail product review websites provided accurate information 8.5% of the time. When sites that were not relevant were excluded, the websites providing the highest and lowest rates of accurate information remained the same. Government and organization websites had the highest rates of accurate information (86.7% and 83.1%, respectively), and blogs, retail product review websites, and websites from individuals had the lowest rates of accurate information (30.9%, 36.2%, and 45.5%, respectively).
National surveys have demonstrated that a high proportion of adults both believe most of the health information that is available on the internet4 and make decisions for themselves and others based on that information.4 Other studies have found variability in the reliability and accuracy of websites that contain pediatric health information.7, 9 Furthermore, one study found that half of those who used the internet to search for health information did not verify the source or currency of the information, relying instead on “common sense;” the authors speculate that this is perhaps because of the great deal of trust placed on the reliability of the internet.4 However, it is clear that when parents hear mixed messages about a specific health recommendation, they are more likely to discount that recommendation as unimportant.18 It is therefore disturbing that less than one-half (43.5%) of the websites found in our Google searches for key phrases related to infant sleep safety provided accurate information. Even when websites that were not relevant were excluded, only 60.8% of websites were accurate in the advice provided.
In general, government and organization websites had the highest rates of accurate information (80.1% and 72.5%, respectively). Government websites with either inaccurate or not relevant information generally contained outdated information. There was no specific topic that predominated in the inaccurate category for government websites. Organization websites, on the contrary, were generally consistent in containing updated information about infant sleep safety. The organizations represented in these websites included health awareness, parenting, healthy infant advocacy, and SIDS awareness organizations.
Educational websites inclued those with .edu in the URL, books, scientific/medical institution pages, and peer-reviewed articles; only 52.4% of these websites contained accurate information. Because peer-reviewed articles were assessed for accuracy based on the information available, those that only provided a title and abstract and required a subscription to access the entire article may have had additional information in the full text that could have affected the assessment for accuracy. The majority of books found online were either not relevant to the key phrase about infant sleep or provided outdated information.
Several websites were on the topic of breastfeeding and bedsharing. There is a great deal of controversy among health professionals on this topic, as bedsharing and breastfeeding duration have been correlated,19 but bedsharing also increases the risk of SIDS, suffocation, asphyxia, entrapment, and other sudden unexpected infant deaths.20–26 For such controversial topics, it is not surprising to find a high percentage of inaccurate websites (e.g., promotion of bedsharing as a strategy to encourage breastfeeding), as accuracy was determined solely by consistency with AAP recommendations.10, 11
Websites with the highest rates of inaccurate information included retail product review websites, blogs, and individual websites. Retail product review websites with inaccurate information generally promoted products (such as infant sleep accessories) that are not recommended by the AAP. There was frequently an implication that the product was AAP-endorsed (e.g., promotes sleeping on the back, which is recommended by the AAP). As many of these products are specifically not recommended by the AAP and have no scientific data to support their claims, it is disturbing that these websites promote products that are generally unnecessary and may be dangerous to sleeping infants. Blogs and individual websites often included authors from a wide variety of backgrounds, including pediatricians and concerned parents. Thus both blogs and individual websites may present contrasting views. Included in the “individual” website category were 3 Facebook pages. It is yet unclear what impact these types of websites will have on parent behavior. However, half of US adults27 and 73% of the upcoming generation of parents (currently ages 12–17)2 currently use social network sites, so these websites may become more important sources of health information in the future. Forums were also included in the ‘Blog’ category, as there are no requirements or qualifications to become a contributor to a community conversation. Forums provide an avenue for networking in an inquisitive and supportive manner. However, these virtual conversations, by their nature, may be the source of inaccurate information28, which may in turn result in potentially dangerous practices. Parents may be in search of support from others who have, for instance, placed their infant prone for sleep without negative consequences and who currently recommend that practice to others. Indeed, in the forums that we encountered in our searches, many parents explicitly discussed how they dismissed the AAP recommendations, particularly with regards to bedsharing, sleep position, and the use of home monitors.
Sponsored link websites such as eHow.com and About.com often have hired experts, including physicians, as contributors for specific topics. It was therefore surprising that the content of these websites was frequently inaccurate. It is possible that sponsoring entities (such as infant product manufacturers) disagree with specific AAP safe sleep recommendations and/or may be promoting products that are not recommended. Video messages (primarily Youtube videos) were also categorized as sponsored link websites. Fourteen percent of U.S. adults do not have basic literacy skills;29 in addition, many others do not have the ability to read and comprehend technical reports. Video can thus be a powerful internet tool for these parents and serve as an alternative, easily accessible, source of information. We found that, with the exception of infant sleep position, for which the videos consistently provided accurate information, videos frequently provided inaccurate or not relevant information with regards to infant sleep safety.
We were surprised that news websites only provided accurate information about infant sleep safety approximately half of the time. In general, national news coverage about infant sleep care almost always correctly described AAP guidelines, but sometimes highlighted the controversy over specific guidelines rather than the importance of the guidelines themselves. Local news coverage was generally limited to stories about tragic, traumatic, and often fatal incidents, without any mention of AAP guidelines or about how other parents could prevent similar mishaps. Parents frequently gauge the importance of an issue by how frequently they hear about it in the news.18 If parents therefore have not heard about SIDS or other causes of sudden unexpected infant death in the news recently, they assume that this is no longer a problem and that they no longer need to take steps to reduce the risk of such an occurrence.
We acknowledge that this study has limitations. There may be observer bias in determining accuracy or inaccuracy of websites. We took specific measures to minimize bias. We developed strict criteria for categorizing the websites and determining the accuracy of each website. In addition, we limited each search to the first 100 websites. Expanding the searches may have altered the results. However, we felt that it was unlikely that the typical parent conducting a search about infant sleep safety would search beyond 100 websites (8–10 pages).
In conclusion, we found that the internet frequently contains inaccurate information about infant sleep safety. It is important for health care providers to realize the extent to which parents may turn to the internet for information about infant sleep safety and then act upon the advice found, regardless of the reliability of the source. Providers should consider offering URLs of specific websites that they have identified as accurately reflecting AAP guidance and educating families on how to evaluate health-related websites for trustworthiness. Websites such as Healthfinder (healthfinder.gov), Medline Plus (http://www.nlm.nih.gov/medlineplus), and Health on the Net Foundation (www.hon.ch/HONcode), may be helpful in providing such guidance. Finally, governments and other entities that host websites with infant sleep safety information should periodically review the content of the information for accuracy and currency.
M.C. was supported by the APS/SPR Student Research Program (NIH grant HD007446), which had no role in the conduct of the study or the production of the manuscript. R.M. has served as expert witness, both for the defense and the plaintiff, on a number of medical-legal cases pertaining to sudden infant death syndrome.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The other authors declare no conflicts of interest.