C
IDENTIFYING THE LITERATURE

This appendix is a brief overview of the approach used by the Institute of Medicine committee and staff in identifying the scientific and medical literature that would form the basis of the committee’s review. It is not meant to be a comprehensive guide to searching the insecticide and solvent literature.

LITERATURE SEARCHES

Online Databases

Online databases were used extensively to identify the relevant peer-reviewed literature. Most of the literature searches were conducted in medical and scientific bibliographic databases available through Dialog, a commercial vendor of over 470 databases that cover a wide array of disciplines. From Dialog’s catalog of databases, staff chose the ones that contained peer-reviewed scientific literature in the biomedical sciences and in environmental and occupational health, public health, toxicology, biology, and chemistry (Table C.1). There was subject and content overlap, but each database has a unique subject emphasis and indexes literature not available elsewhere. Given the study’s focus on long-term human health effects of insecticides and solvents, MEDLINE and EMBASE were particularly useful because of their extensive coverage of US and international scientific and biomedical journals. MEDLINE is produced by the US National Library of Medicine and covers over 4300 journals published throughout the world. The database provides coverage from 1966 to the present and contains over 11 million citations. EMBASE, produced by Elsevier Science, indexes over 3300 primary journals with a focus on the international literature. EMBASE indexes journal articles from 1974 to the present and contains over 8 million citations. Both databases add about 40,000 records each year. To provide essential toxicologic and chemical information on each of the insecticides and solvents, factual databases (Table C.2) were searched. Together, the bibliographic and factual databases offer the most effective means of identifying the international peer-reviewed scientific literature published since the middle 1960s. Other strategies were used to identify literature published before then (see “Other Sources” below).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 565
Gulf War and Health: Insecticides and Solvents, Volume 2 C IDENTIFYING THE LITERATURE This appendix is a brief overview of the approach used by the Institute of Medicine committee and staff in identifying the scientific and medical literature that would form the basis of the committee’s review. It is not meant to be a comprehensive guide to searching the insecticide and solvent literature. LITERATURE SEARCHES Online Databases Online databases were used extensively to identify the relevant peer-reviewed literature. Most of the literature searches were conducted in medical and scientific bibliographic databases available through Dialog, a commercial vendor of over 470 databases that cover a wide array of disciplines. From Dialog’s catalog of databases, staff chose the ones that contained peer-reviewed scientific literature in the biomedical sciences and in environmental and occupational health, public health, toxicology, biology, and chemistry (Table C.1). There was subject and content overlap, but each database has a unique subject emphasis and indexes literature not available elsewhere. Given the study’s focus on long-term human health effects of insecticides and solvents, MEDLINE and EMBASE were particularly useful because of their extensive coverage of US and international scientific and biomedical journals. MEDLINE is produced by the US National Library of Medicine and covers over 4300 journals published throughout the world. The database provides coverage from 1966 to the present and contains over 11 million citations. EMBASE, produced by Elsevier Science, indexes over 3300 primary journals with a focus on the international literature. EMBASE indexes journal articles from 1974 to the present and contains over 8 million citations. Both databases add about 40,000 records each year. To provide essential toxicologic and chemical information on each of the insecticides and solvents, factual databases (Table C.2) were searched. Together, the bibliographic and factual databases offer the most effective means of identifying the international peer-reviewed scientific literature published since the middle 1960s. Other strategies were used to identify literature published before then (see “Other Sources” below).

OCR for page 565
Gulf War and Health: Insecticides and Solvents, Volume 2 TABLE C.1 Bibliographic Databases Name Producer Coverage Size BIOSIS Previews BIOSIS 1969 to present Over 12,257,000 records (as of May 2000) CAB HEALTH CAB International 1973 to present 616,000 records (as of Dec 1997) CANCERLIT US National Cancer Institute 1975 to present Over 1,693,000 records (as of Aug. 2001) EMBASE Elsevier Science BV 1974 to present Over 8,052,000 records (as of Apr. 2000) Environmental Bibliography Environmental Studies Institute 1973 to present Over 590,000 records (as of March 1998) Life Sciences Collection Cambridge Scientific Abstracts 1982 to present Over 1,600,000 records (as of Dec. 1997) MEDLINE US National Library of Medicine 1966 to present Over 11,149,899 records (as of Sept. 2001) National Technical Information Service US Department of Commerce 1964 to present Over 2,112,000 records (as of May 2000) Occupational Safety and Health National Institute for Occupational Safety and Health 1973–1998 210,155 records (file closed) PsycINFO American Psychological Association 1887 to present Over 1,688,000 records (as of Dec. 2000) Science Citation Index Institute for Scientific Information—Thomson Scientific 1975 to present Over 10,048,000 records (as of Oct. 2001) TOXLINE US National Library of Medicine 1965 to present 2,400,00 records (as of Feb. 1999 WorldCat OCLC Online Computer Library Center, Inc Unlimited Over 46,000,000 cataloging records TABLE C.2 Factual Databases ChemID Plus Extension Toxicology Network (EXTOXNET) Hazardous Substances Data Bank (HSDB) Integrated Risk Information System (IRIS) Dialog’s robust search engine allows the use of complex search strategies in multiple databases simultaneously. However, databases often contain unique fields of data or use different indexing systems. For example, the records of BIOSIS and CAB HEALTH include Chemical Abstract Service (CAS) registry numbers; other databases do not include CAS registry numbers as a data element but instead rely on chemical names and synonyms. All search strategies were designed to capture relevant CAS registry numbers wherever available but with the recognition that databases lacking CAS registry numbers would not be as well represented in the retrieval set and that additional searching was needed. Another challenge in conducting relevant and comprehensive literature searches is the diversity of indexing systems used by database producers. For example, MEDLINE, EMBASE, and PsycINFO are professionally indexed with a controlled set of vocabulary terms or subject codes; using MEDLINE’s index term carbaryl automatically retrieves all studies that use the term carbaryl and ones that use the alternative spelling carbaril or the trade name Sevin. By using the controlled thesaurus terms, staff could be certain that all synonyms, alternative spellings, trade names, and equivalent conceptual terms were automatically retrieved in the search results. Other databases, however, do not use a controlled vocabulary but instead rely on author-assigned index terms. To address that problem, search strategies were expanded to include specifically not only the standard

OCR for page 565
Gulf War and Health: Insecticides and Solvents, Volume 2 insecticide and solvent terms but also other spelling variations, synonyms, and commercial names. Thus, a series of search strategies were developed to take advantage of the classification scheme, controlled vocabulary, and distinct structure of each database. In addition to the complexities presented by the databases, the literature on the insecticides and solvents being reviewed in this study afforded its own set of challenges as committee members and staff worked to identify the pertinent literature comprehensively while minimizing the retrieval of nonrelevant citations. For instance, ethylene glycol and butyl acetate, two solvents under study, are referred to in numerous articles that describe laboratory procedures but not human health effects. Searching for the solvent ethanol was confounded by the retrieval of numerous articles on alcoholism. Similarly, many efficacy and treatment-related studies, such as those of the effectiveness of various insecticides on mosquito nets or evaluating treatment options for head lice, were retrieved by a search for malathion and permethrin. It is not always possible to design a comprehensive search strategy that retrieves the citations needed for the committee’s purposes but eliminates all citations that are not useful to the committee. The search for insecticides included the 11 insecticides listed in the congressional legislation and three additional insecticides (d-phenothrin, azamethiphos, and bendiocarb) identified by the Office of the Special Assistant for Gulf War Illnesses as having been used by military personnel during the Gulf War. The search for solvents included the 53 specific solvents that were sent to the Gulf War as determined by the Department of Defense’s Defense Logistics Agency. Although the search strategies varied somewhat with their purpose and the structure of the databases being searched, the foundation of every major search contained the same basic elements: specific insecticides and solvents, their synonyms, alternative spellings, trade names, CAS registry numbers, insecticide and solvent classes (such as organophosphates, carbamates, acetates, and glycols), and the general terms pesticides, insecticides, and solvents. All databases were searched in their entirety, and all foreign-language citations were retained; any studies that the committee deemed significant were translated into English. Because the focus of the committee’s work was to review the literature regarding the human health effects of insecticide and solvent exposure, the primary searches used the comprehensive search strategy discussed above and then limited the retrieval to human studies. For some database searches, terms like adverse effects, epidemiologic study, controlled study, and cohort were added to the search strategies. Nearly 30,000 citations were initially identified and reviewed for their relevance to the committee. However, as discussed above, because of the attempt to be as comprehensive as possible, the search strategies retrieved many references that were not relevant for the committee’s purposes. A second goal of the literature search was to provide the committee with a broad overview of the toxicology of insecticides and solvents so that it could assess biologic plausibility. Textbooks and reports, such as the Agency for Toxic Substances and Disease Registry (ATSDR) Toxicological Profiles, provided much of the background material, but online bibliographic databases were used to identify the current literature and review articles. Throughout the study, smaller targeted searches were conducted on specific topics, such as insecticide and solvent interactions, occupational exposures, and inert ingredients found in insecticide and solvent formulations. The secondary searches were especially

OCR for page 565
Gulf War and Health: Insecticides and Solvents, Volume 2 useful for double-checking the thoroughness of the primary search strategies. In addition, the primary and secondary searches combined yielded many pertinent case reports, background, and review articles that were retained for use as supporting material as needed by the committee. To ensure that the committee reviewed the complete body of evidence before forming its conclusions, searches were conducted through August 2001. Other Sources Identifying the pertinent literature was a continuing process throughout the study. In addition to the formal online searches, the committee and staff examined the reference lists of major epidemiologic studies, review articles, and textbooks for relevant citations. The bibliographies of such reports as the International Agency for Research on Cancer’s monograph series, ATSDR’s Toxicological Profiles, the World Health Organization’s Environmental Health Criteria documents, and the technical reports of the European Centre for Ecotoxicology and Toxicology of Chemicals provided many relevant citations. Online library catalogs, including those of the US National Library of Medicine and the National Institutes of Health, were searched for textbooks and other monographs that could provide pertinent overview and background materials for the study. MANAGING THE INFORMATION The results of the online searches and citations identified by other means were imported into ProCite, a software program designed to store and manage bibliographic data. When the search phase was completed, ProCite contained nearly 30,000 citations that included all the relevant and nonrelevant epidemiologic and toxicologic studies, background reports, and other articles. Staff reviewed each citation carefully and identified about 3000 citations as most relevant to the committee’s charge. The citations selected were studies of human populations exposed to the agents of interest that examined the potential for adverse health effects. The full text of those journal articles was retrieved and sent to committee members for their review.