{"title":"设计一个搜索策略来识别和检索使用MEDLINE的循证医疗保健文章。","authors":"J Harrison","doi":"","DOIUrl":null,"url":null,"abstract":"<p><p>The practice of evidence-based health care requires that information on methodology be identified from databases such as MEDLINE. Up until this year there have been no designated medical subject headings (MeSH) for evidence-based health care. 'EVIDENCE-BASED MEDICINE' appears as a MeSH term from 1997. The absence of designated MeSH for this concept prior to 1997 provides a challenge to the searcher. This paper describes the creation of a MEDLINE search strategy to retrieve articles on the methods of evidence-based health care published prior to the introduction of the new term, where an optimal combination of free-text and MeSH terms is required to identify relevant material. The study examines both free-text and subject heading searching and attempts an optimal balance of sensitivity and specificity. It begins by examining separate free-text and subject heading searches. Sensitivity of the subject heading search was 33% and specificity 80%, while the free-text search produced a sensitivity of 50% and a specificity of 67%. The final strategy, combining both approaches, was more successful with sensitivity reaching between 82 and 90% and specificity 83%. It is therefore possible to devise a search strategy to retrieve articles on the methods of evidence-based health care with relatively successful rates of sensitivity and specificity. The limitations of MEDLINE, however, necessitate the use of additional approaches in identifying articles on the methods of evidence-based health care.</p>","PeriodicalId":73211,"journal":{"name":"Health libraries review","volume":"14 1","pages":"33-42"},"PeriodicalIF":0.0000,"publicationDate":"1997-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Designing a search strategy to identify and retrieve articles on evidence-based health care using MEDLINE.\",\"authors\":\"J Harrison\",\"doi\":\"\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>The practice of evidence-based health care requires that information on methodology be identified from databases such as MEDLINE. Up until this year there have been no designated medical subject headings (MeSH) for evidence-based health care. 'EVIDENCE-BASED MEDICINE' appears as a MeSH term from 1997. The absence of designated MeSH for this concept prior to 1997 provides a challenge to the searcher. This paper describes the creation of a MEDLINE search strategy to retrieve articles on the methods of evidence-based health care published prior to the introduction of the new term, where an optimal combination of free-text and MeSH terms is required to identify relevant material. The study examines both free-text and subject heading searching and attempts an optimal balance of sensitivity and specificity. It begins by examining separate free-text and subject heading searches. Sensitivity of the subject heading search was 33% and specificity 80%, while the free-text search produced a sensitivity of 50% and a specificity of 67%. The final strategy, combining both approaches, was more successful with sensitivity reaching between 82 and 90% and specificity 83%. It is therefore possible to devise a search strategy to retrieve articles on the methods of evidence-based health care with relatively successful rates of sensitivity and specificity. The limitations of MEDLINE, however, necessitate the use of additional approaches in identifying articles on the methods of evidence-based health care.</p>\",\"PeriodicalId\":73211,\"journal\":{\"name\":\"Health libraries review\",\"volume\":\"14 1\",\"pages\":\"33-42\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1997-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Health libraries review\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Health libraries review","FirstCategoryId":"1085","ListUrlMain":"","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Designing a search strategy to identify and retrieve articles on evidence-based health care using MEDLINE.
The practice of evidence-based health care requires that information on methodology be identified from databases such as MEDLINE. Up until this year there have been no designated medical subject headings (MeSH) for evidence-based health care. 'EVIDENCE-BASED MEDICINE' appears as a MeSH term from 1997. The absence of designated MeSH for this concept prior to 1997 provides a challenge to the searcher. This paper describes the creation of a MEDLINE search strategy to retrieve articles on the methods of evidence-based health care published prior to the introduction of the new term, where an optimal combination of free-text and MeSH terms is required to identify relevant material. The study examines both free-text and subject heading searching and attempts an optimal balance of sensitivity and specificity. It begins by examining separate free-text and subject heading searches. Sensitivity of the subject heading search was 33% and specificity 80%, while the free-text search produced a sensitivity of 50% and a specificity of 67%. The final strategy, combining both approaches, was more successful with sensitivity reaching between 82 and 90% and specificity 83%. It is therefore possible to devise a search strategy to retrieve articles on the methods of evidence-based health care with relatively successful rates of sensitivity and specificity. The limitations of MEDLINE, however, necessitate the use of additional approaches in identifying articles on the methods of evidence-based health care.