Gold Standard Search
[This link has been removed because it is no longer valid] 1998 NLM Pricing Algorithm Chart
Gold Standard Search
Thank you to everyone who mailed in search strategies in response to the "Gold Standard Search" article in the May-June 1997 issue of the Technical Bulletin. This feature offers searchers a chance to match wits with NLM staff on a search strategy formulation. The Gold Standard search strategies are developed here at NLM. The search request published in the May-June 1997 issue was:
A physician is trying to decide whether or not to recommend yearly mammographies to her 40-49 year old patients and wants to get some information to help her make this decision.
There was no winner for this search. However, there was one honorable mention, Katherine Alexander of the Northwest Medical Center in Springdale, Arkansas. Ms. Alexander received the honorable mention because of the extra efforts she made to find very relevant information from additional sources as discussed below in the Analysis of the Search Strategies.
Choosing a File
The first step in this Gold Standard Search was to choose the best database for the subject. For this search, CANCERLIT, would seem to be the obvious choice. However, since, MEDLINE and PREMEDLINE have a considerable overlap with CANCERLIT and are searchable free of charge on the Web, if cost savings were a concern, command language searchers might think of going to PubMed first to run their strategy and then going into CANCERLIT and running the same strategy, eliminating the MEDLINE overlap and printing out only the unique CANCERLIT citations there. The Gold Standard search was run in CANCERLIT and PREMEDLINE. However, the Gold Standard Search did not retrieve any unique citations that would not have been found in MEDLINE. Searchers should routinely consider searching by Text Word in PREMEDLINE, in addition to searching the primary file of interest, to get the latest information on their subject.
Introduction to the Search
The inspiration for this search was a recent controversy in the United States surrounding the adoption of new mammography guidelines by the National Cancer Institute. Two key citations describing this event are shown below:
UI - 97260518 AU - Eastman P TI - NCI adopts new mammography screening guidelines for women [news] SO - J Natl Cancer Inst 1997 Apr 16;89(8):538-9 2 UI - 97261753 AU - Taubes G TI - NCI reverses one expert panel, sides with another [news] SO - Science 1997 Apr 4;276(5309):27-8
While knowledge of this event was not completely necessary in structuring the terms required for a search, it does help limit the timeframe of the search. It is clearly not necessary to go back very far to find the current information. In fact, a search could be limited to 1997 for the purposes of clinical decision making.
Analysis of the Search Strategies
The selection of the appropriate MeSH headings was important and fairly straightforward. In this search, BREAST NEOPLASMS, MAMMOGRAPHY and MASS SCREENING were the key headings. Almost all entries contained these terms. Further refinement or narrowing, can be obtained by combining the above with PRACTICE GUIDELINES, GUIDELINES, PATIENT SELECTION and/or some of the age terms. Again, the choice about how many terms to add and how to limit the search will depend on your interpretation of how much information is enough for your client. The Gold Standard Search Strategy is shown in Figure 1.
In addition to searching the NLM databases, Ms. Alexander also accessed CancerFax at 301-402-5874. This service provides access to many types of cancer information, for the cost of a phone call to the CancerFax computer (you must have a fax machine to utilize this service). The CancerFax Contents list allows you to select from the listings and have any of the items faxed to you. Ms. Alexander obtained copies of the following publications and found the first one to be "the most on-target information" she found.
400181: Mammography screening recommendations, National Cancer Advisory Board Issues. NCAB: mammography recommendations for women ages 40-49. March 1997. 6 pages.
600528: Mammography screening, Questions and Answers
400183: Mammography, statement from NCI on NCAB recommendations.
With the electronic resources now available, it will become increasingly necessary to look beyond the boundaries of standard database searches in order to obtain the most current information available on a subject.
Search Strategy Suggestions
The most common error was the use of subheadings in a search that had a relatively small retrieval. If you find fewer than 10 citations in a file, it is generally unnecessary to restrict the retrieval further by including subheadings. Other strategy errors that appeared in this search included:
1. Not searching in CANCERLIT and PREMEDLINE. Several entries searched only in MEDLINE. CANCERLIT should be used for a cancer-related search; and for the most current information, a PREMEDLINE text word search is necessary.
Using ADULT (PX) AND NOT AGED to eliminate unwanted citations. Articles are frequently indexed for more than one age group, and eliminating AGED could remove relevant articles that discuss both the desired age group and the elderly. In this search it is better to use ADULT or MIDDLE AGE and not try to narrow it further.
Eliminating LETTER (PT). Eliminating letters on such a recent topic could miss relevant information because good discussions, which have not yet appeared in articles, may be contained in letters.
The Next Gold Standard Search
A change is afoot for this ongoing feature of the Technical Bulletin. The purpose of these searches will remain the same, that is, they will be designed to illustrate search features and vocabulary in the NLM databases. However, PubMed and Internet Grateful Med (IGM) will be the highlighted search engines. The July-August 1997 issue of the Technical Bulletin has a PubMed Search Hints article on page 4 to help you get started and more information will be published in the coming months.
The search request is:
Your user asks for information on "Jewish genetic diseases."
Results will be published in the January-February 1998 Technical Bulletin. Entries must be received by December 15, 1997 to be considered.
Choose either PubMed or Internet Grateful Med. If you choose IGM, please indicate which database(s) you searched. Limit your retrieval to English language citations.
There are no exact answers or absolutely correct strategies, only best approximations, just like real life. Good luck!
If your strategy is selected for publication you could win an NLM T-shirt or mug. In cases of multiple submissions of identical winning strategies, the prize winner will be either the person who is a first time winner or the first winning strategy received.
Mail us a copy of your search strategy and include a printout of citations. Please note that in the interest of saving time and money we are no longer asking for abstracts.
IGM searchers can print their strategies by clicking on Details of Search and printing the page displayed there. With PubMed as your search engine, you may be unable to print the screen displaying your search strategy. Therefore, please type your strategy on paper for submission and mail it to:
Figure 1 - Gold Standard Search
SS 1 /C? USER: file cancer PROG: YOU ARE NOW CONNECTED TO THE CANCERLINE (CANCERLIT) FILE. SS 1 /C? USER: *mammography and *breast neoplasms/pc * In this particular situation it is not necessary to EXPLODE breast neoplasm because the narrower terms are to male breast neoplasms and to mammory neoplasms which pertain to animals. PROG: SS (1) PSTG (739) SS 2 /C? USER: mass screenings or practice guidelines or guidelines or patient selection * As the text explained, if the retrieval in SS1 proved to be low you should drop the use of the subheading. PROG: SS (2) PSTG (9861) SS 3 /C? USER: (pt) practice guideline or guideline PROG: SS (3) PSTG (675) SS 4 /C? USER: 2 or 3 PROG: SS (4) PSTG (10474) SS 5 /C? USER: 1 and 4 PROG: SS (5) PSTG (610) SS 6 /C? USER: 5 and not for (la) PROG: SS (6) PSTG (544) SS 7 /C? USER: 6 and human PROG: SS (7) PSTG (542) SS 8 /C? USER: 7 and adult or 7 and middle age PROG: SS (8) PSTG (364) SS 9 /C? USER: 8 and 97 (yr) PROG: SS (9) PSTG (34) SS 10 /C? USER: prt 10 ti PROG: 1 TI - The great mammogram debate. 2 TI - NBSS: changes were made, suspicious changes were not [letter] 3 TI - NIH Consensus Statement. Breast cancer screening for women ages 40-49. 4 TI - The effect of spatial definition on the allocation of clients to screening clinics. 5 TI - Mammography controversies: time for informed consent? [letter] 6 TI - Mammography screening for women aged 40 through 49-a guidelines saga and a clarion call for informed decision making. 7 TI - Age-related differences in breast carcinoma knowledge, beliefs, and perceived risk among women visiting an academic general medicine practice. 8 TI - National Institutes of Health Consensus Development Conference Statement: Breast Cancer Screening for Women Ages 40-49, January 21-23, 1997. National Institutes of Health Consensus Development Panel. 9 TI - Assessment of lifetime gained as a result of mammographic breast cancer screening using a computer model. 10 TI - Report of the NMA panel on mammography. SS 10 /C? USER: file premedline PROG: YOU ARE NOW CONNECTED TO THE PREMEDLINE FILE. NOTICE: THESE PRELIMINARY, UNINDEXED CITATIONS HAVE NOT GONE THROUGH NLM'S QUALITY CONTROL PROCESS. SS 1 /C? USER: (tw) all mammogra: PROG: SS (1) PSTG (82) SS 2 /C? USER: (tw) all guideline: or all screen: or all detect: PROG: SS (2) PSTG (5542) SS 3 /C? USER: 1 and 2 PROG: SS (3) PSTG (51) SS 4 /C? USER: 3 and not for (la) PROG: SS (4) PSTG (37) SS 5 /C? USER: prt 3 ti PROG: 1 TI - Breast carcinoma stage in relation to time interval since last mammography: a registry-based study. The Romagna Cancer Registry and Collaborators [In Process Citation] 2 TI - Self-reported use of mammography among women aged > or = 40 years - United States, 1989 and 1995 [In Process Citation] 3 TI - Comparison of the diagnostic performance of high-frequency ultrasound as a first- or second-line diagnostic tool in non-palpable lesions of the breast [In Process Citation]