Originally, the task of collecting bibliographic data was left to private enterprise. It is only recently that the U.S. and other governments began providing bibliographic indices at no or nominal cost. These government databases focus upon the specific interests of the agency. Each of these databases has an annual budget ranging from about $3 million to tens of millions of dollars and employs tens to hundreds of taxpayer-supported personnel. These U.S. federal databases can contain many articles relevant to safety but the database search systems make it very difficult to find them – even for experienced users. However, it has been demonstrated that if a safety-related article is in any one of these US government-supported systems, it is also (or soon will be) included in SafetyLit – and in a way that it may be found by searchers and incorporated into their work. Indeed, SafetyLit uses these databases as a resource for identifying items for inclusion in its own database system. These government databases contribute about 35% of SafetyLit contents. Thus, their existence is instrumental to SafetyLit's success at identifying suitable material. In contrast, it must be noted that SafetyLit regularly provides its records to two of these government-supported bibliographic systems.
There is evidence that information-seekers come to SafetyLit when they suspect that their search results from one of the free U.S. government databases missed important information. A recent Alexa Clickstream report found that, for 8 percent of SafetyLit searchers, the website they visited immediately before reaching SafetyLit was one of the government databases mentioned above.
SafetyLit also draws its contents from many sources that are not included in any of the US government-supported databases. SafetyLit provides, in one place, information about interpersonal violence (gangs, partner violence, child abuse, terrorism), suicide and self-harm, and unintentional injuries from incidents involving traffic crashes, burns at home or at work, poisoning, falls, natural disasters, and recreation. Also included are reports about the risk factors for injuries (such things as alcohol and drug use, engineering and design failures, behavioral problems, etc.); the individual, familial, and societal costs and consequences resulting from injuries to individuals or to communities from natural or man-made disasters; individual and enforcer response to laws and regulations; willingness to respond to calls for evacuation or other action; and other relevant topics.
The SafetyLit database contains records from journals (including non-English ones) that are not in any other database – at least for the short term. As noted above, once the articles are added to SafetyLit they are offered to appropriate government database systems, just as SafetyLit identifies new items from those government systems. For example, in early 2014, SafetyLit obtained copies of the complete back-files of several nineteenth century and early- to mid-twentieth century German-language journals concerning the skills and abilities necessary to operate various machines and vehicles. The article titles and summaries were transcribed, translated into English, and both language versions added to the SafetyLit database. These records were offered to and accepted by the U.S. Transportation Research Board for their use in the TRID database.
All SafetyLit services are provided at no cost to the user and are presented without advertising. Keeping SafetyLit without cost and free from influences that could limit its scope or bias its focus is important. Since SafetyLit is a free service, many publishers have provided their material to us without cost. If SafetyLit were to begin charging subscription fees, some of these publisher agreements could be lost and the cost for SafetyLit to access this data would be prohibitive.
Google Scholar (GS) is a massive, free database of research literature and contains records relevant to injury research and prevention. However, finding a comprehensive listing of reports on a specific topic can be a challenge. SafetyLit has advantages over GS. While the SafetyLit search system allows a query to be done using any one term that labels a concept and optionally receive all records on the topic, users of GS would need to search on each and every term that signifies a topic to receive what a single simple search of SafetyLit will return. For example, the consumer product “baby walker” (now banned in several Western nations, but not all) is known by more than 40 other terms used by authors to describe the devices. A search of GS would require a query knowing and using each of these terms while a search using the SafetyLit optional synonym search would provide all literature containing any of these terms. A search of GS for the term “football” (see below) by someone wanting articles on soccer would provide many unwanted articles about other sports known as football. A search for “soccer” would omit hundreds of articles where the author used the common “football” term for the sport but never mentioned “soccer”. SafetyLit evaluates each article containing ambiguous terms and adds explanatory words to the abstract disambiguate any potential confusion. Further, there are few articles in GS that are not also in SafetyLit. Every day SafetyLit volunteers conduct several very sensitive GS searches, evaluate the results, discard irrelevant records, and add the pertinent ones to SafetyLit. The “few” hedge about the inclusion of GS-identified articles in SafetyLit is because if the GS-identified literature is also indexed in PubMed the article is delayed a few days until it is included in PubMed. This is because PubMed identifiers (PMID and PMCID) are required to be included among the reference lists of grant applications to several U.S. government agencies. Although GS allows users to download article metadata for its records, GS does not typically include the PMID or PMCID that are included in SafetyLit. While GS will lock-out a user who makes too many metadata requests in a 24 hour period (a limit of as few as 50 to 100 requests), the SafetyLit limit is 25 thousand within a week.
Commercial databases charge high fees to subscribers but provide added value for searchers by optionally including information about all of the references in each article or all of the later articles that cite the original article itself. This allows their academic users to track back and follow forward to gather more information about the contents of the article. Thus, the strengths of private enterprise and profit-driven innovation cannot be contested. While these extras are indeed useful – particularly for academic report-writing – there remains a need for access to information by those outside of universities and well-financed think-tanks. SafetyLit is an index of the literature and provides a path to finding the full text of articles of interest online; to identifying nearby libraries with collections that hold the wanted item; or to allow a local librarian to obtain the articles at no or very low cost (via inter-library loan programs – SafetyLit provides a quick way to print out a form containing all of the information needed by a librarian to make an inter-library loan request).
Although SafetyLit does not itself contain the full text of materials it indexes, most records contain an abstract or summary of the content. For many articles, SafetyLit will add a summary that is omitted by a publisher (and not included in other databases) so that SafetyLit users can understand what the article is about. SafetyLit allows a user to know if information about a topic exists. Published products of research transcend professional disciplines, geography, and language. SafetyLit not only brings together the research, evaluation, and policy papers of more than 30 distinct professional disciplines but provides a service not available with most commercial databases — item summaries include a brief "interpretation" of highly technical language or confusing terms used by the authors so that the information can be understood by any reasonably educated English-speaking person anywhere in the world. A very simple example of this would be distinguishing articles about football. Authors (forgetting that their article may be read by people across the world) may use the word "football" without any further explanation. SafetyLit will modify the report's summary to clarify if the article is about American football, Association football (soccer), Australian football, Canadian football, or Gaelic football—different games with different rules and equipment and very different risks for injuries. Entering the word “football” (or some other ambiguous term, e.g. tailgating) in the SafetyLit search system will display a query-clarification page that provides a listing of links to the various query topics so that the searcher may avoid paging through hundreds of items that are irrelevant to their needs.