Thursday, June 30, 2005

Database Highlight: PIER

PIER (Physicians' Information and Education Resource), produced by the American College of Physicians - American Society of Internal Medicine, is a web-based information support tool specifically designed to deliver rapidly accessible, comprehensive, up-to-date, and evidence-based guidance to clinicians. The information, written by experts and peer-reviewed, is presented in a standard structure that mirrors questions and decisions that physicians routinely address when caring for patients.

Content is organized around five modules: diseases; screening and prevention; complementary and alternative medicine; ethical and legal issues; and procedures. Each section begins with a few succinct evidence-based guidance statements - and clicking on a guidance statement brings up a specific recommendation and its supporting rationale, evidence, and comments. All recommendation statements have letter grades that alert the user to the strength of supporting evidence.

The Medical Library subscribes to PIER through the Stat!Ref service.

Give it a try! It's really easy to use.

FYI
Available from the All Major Resources A-Z list. Click on Stat!Ref.

Tuesday, June 28, 2005

MEDLINE Database on Tap

MEDLINE Database on Tap (aka MD on Tap or MDoT) is a Web application which targets mobile health care professionals by facilitating access to medical information at the point of care. MDoT provides a way to search the National Library of Medicine's MEDLINE, read abstracts from a set of 120 core clinical journals, and access the ClinicalTrials.gov database. To use it, you need a PDA with Palm 4.0 OS or higher and a wireless Internet connection. An equivalent Pocket PC client is underdevelopment.

The client is available for PDAs and smartphones with Palm operating systems and a wireless connection to the Internet. You can download the newest version from http://mdot.nlm.nih.gov/proj/mdot/downloadPalm.php.

With this newest version, there are two cool new features:

  • When the Auto Spell Check option is selected, ambiguous search terms are replaced with those terms suggested by an internal NLM e-spell utility for the actual search. The actual terms used are shown at the bottom of the Search tab.
  • As an alternative to the default PubMed search engine, you can select Essie, an experimental probabilistic search engine developed at the Lister Hill National Center for Biomedical Communications. Essie ranks results by relevance, then by date. These options can be combined with any (or all!) of the other search options. Also, a big, friendly “Go” button has replaced the magnifying glass as the icon to tap to execute your search.

Check out the screenshots.
Download the program and give it a try!

Monday, June 27, 2005

PubMed RSS feeds

RSS feeds are available from PubMed. To add a PubMed search to your RSS feeds, run a search in PubMed and choose RSS Feed from the Send To pull-down menu.

For detailed instructions see the NLM Technical Bulletin .

Friday, June 17, 2005

Entrez: The Life Sciences Search Engine

In my previous post, I discussed the rich linking environment in Entrez. What makes this resource even more impressive is the search gateway page that NCBI has created to search across all of Entrez's inter-connected nodes with one query.


The gateway is a wonderful place to start a search of the Entrez system. For example, I saw a recent Yale Medicine article on zebrafish and I’d like a summary of related information on that topic.

Try typing in zebrafish on the query line and the system searches across all of Entrez's nodes. Sometimes there are hundreds or thousands of results from a particular database and sometimes there are no relevant results returned. This makes sense, since not all nodes may contain data on a particular topic.

As always, NCBI provides excellent Help for this resource, including mouse-over synopses of the nodes and their scope. Entrez even searches for relevant NCBI and other outside websites and resources, including the National Library of Medicine's on-line catalog. Another nice feature of the gateway is that it uses MeSH (Medical Subject Headings) to search PubMed.

The Entrez gateway is a great place to start researching a topic, but it has its limitations. To perform a thorough search of a specific aspect of your topic, such as genomic information or nucleotide sequences, it's always best to use the search screens for the individual nodes.

Take the Entrez gateway out for a spin at www.ncbi.nlm.nih.gov/gquery/gquery.fcgi
You'll be amazed at the power of this search engine.

Tip: The Entrez gateway is called different things, depending on which Entrez node you are searching. If you are in PubMed, the link to the gateway is called All Databases. In other nodes, it's called Entrez.

Posted by Judy Spak, guest blogger.

Thursday, June 16, 2005

ISI Essential Science Indicators

Essential Science Indicators (ESI) provides internet access to a compilation of essential science performance statistics and science trends data derived from ISI's databases. The chief indicators of output, or productivity, are journal article publication counts. For influence and impact measures, ESI employs both total citation counts and cites per paper scores. The former reveals gross influence while the latter shows weighted influence, also called impact.

Types of data featured in ESI include most cited author rankings, institutional (university, corporate, government research lab) rankings, national rankings, and journal rankings. Another unique feature is the listing of research areas called Research Fronts, algorithmically derived topics reflecting research intensive and breakthrough areas of current science. An editorial feature called Special Topics gives special attention to selected areas of research.

Highly cited papers and hot papers have been selected for ESI based on percentile rankings specific to fields of science and specific time periods. While highly cited papers are chosen from the most recent 10 years of data, hot papers focus on very recent papers (from the past two years) that show an unusual rate of citation in the current period. Both highly cited and hot papers are searchable by a variety of attributes.

FYI

  • It is important to recognize that the data in ESI are limited to ISI-indexed journal articles only. No books, book chapters, or articles published in journals not indexed by ISI are taken into account here, either in terms of publication or citation counts.
  • Data is updated every two to four months.
  • Check out the HELP (and I'd put a link in here but for the fact that the page consists of frames! UGH!)
  • ESI is a part of the ISI Web of Knowledge.

Wednesday, June 15, 2005

Journal Citation Reports

If you've ever wondered about how to choose the 'best' journal in a discipline to submit your article to, Journal Citation Reports is the resource for you.

This comprehensive and unique resource tool allows you to evaluate and compare journals using citation data drawn from over 7,000 scholarly and technical journals from more than 3,300 publishers in over 60 countries. It is the only source of citation data on journals, and includes virtually all specialties in the areas of science, technology, and social sciences. Journal Citation Reports can show you the:
  • Most frequently cited journals in a field
  • Highest impact journals in a field
  • Hottest journals in a field
  • Leading journals in a field
  • Most published articles in a field
FYI
JCR is available from the Medical Library's home page - look for Journal Citation Reports under Major Resources.
If you're new to this resource, check out the tutorial and if you really need detailed information, check out the HELP
JCR is part of the ISI Web of Knowledge.

Tuesday, June 14, 2005

Introducing Entrez

Most people are familiar with PubMed, but I wonder how many know about the extremely rich databases, that along with PubMed, make up the Entrez system.

Created by the National Center for Biotechnology Information (NCBI) in 1991, Entrez began as a system that contained three "nodes", one for published articles (MEDLINE, later called PubMed), a nucleotide database and a protein database. Today, Entrez is made up of more than 20 nodes or databases. Each node represents specific data objects of the same type, e.g., protein sequences.

What makes Entrez such a robust search system is that it integrates the biomedical literature, DNA and protein sequence databases, 3D protein structure and protein domain data, population study datasets, expression data, assemblies of complete genomes, and taxonomic information into a tightly interlinked system. THIS IS VERY POWERFUL STUFF! These links allow you to search in one of the nodes or databases and link to related information in the other databases.
 

The image above is a thumbnail of an interactive graphic that demonstrates the rich linking between the Entrez nodes. Check it out on the Model of Entrez Databases page . Place your mouse on one of the nodes and it will show the various links to other databases, including how many records are linked from each resource. Very cool.

For a more in-depth description of the Entrez search and retrieval system, visit the Entrez information pages from NCBI.

Posted by Judy Spak, guest blogger. Stay tuned for more Entrez posts coming soon.

Monday, June 13, 2005

Searching for Guidelines

Searching for practice guidelines is much easier when you use the National Guideline Clearinghouse.

This comprehensive database of evidence-based clinical practice guidelines and related documents is an initiative of the Agency for Health Care Research and Quality (AHRQ), U.S. Department of Health and Human Services.

The mission of NGC is to "provide physicians, nurses, and other health professionals, health care providers, health plans, integrated delivery systems, purchasers and others an accessible mechanism for obtaining objective, detailed information on clinical practice guidelines and to further their dissemination, implementation and use."

Key Features

Search
There are three ways to search NGC. The first is the basic search box that appears on every page; the second is the detailed search allows you to create very specific search queries based on various attributes such as target population; and the third, frequently requested searches, is designed to assist NGC users in finding guidelines related to frequently requested search topics.
Browse
This handy feature allows users to scan for guidelines available on the NGC site by disease/condition, treatment/intervention, or developing organization.
Compare
This unique feature provides you with the ability to generate side-by-side comparisons for any combination of two or more guidelines. You can also look at syntheses of selected guidelines that cover similar topic areas.
Resources
NGC has compiled a lengthy list of resources including annotated bibliographies, discussion list, FAQs, patient resources, PDA downloads, guideline archives, etc.

Keep this resource in mind when you are looking for clinical guidelines...it will save you time.

Friday, June 10, 2005

Field Guide to MEDLINE

Here's a handy book that I recommend to anyone who wants to build their searching skills. It covers an overview of content, basic tools and search strategies in the second chapter and then in the third and fourth chapter applies the techniques to PubMed and Ovid. So chapter 3 covers PubMed and chapter 4 covers Ovid. There are ample screen shots to illustrate each technique.

And even though both of these interfaces are changing and might not look exactly like they did when this book was produced in 2003, the theory and techniques will not change.

Borrow it from your local library...I'm sure you'll find it useful.


Stave, Christopher D. Field Guide to MEDLINE: Making Searching Simple. Philadelphia: Lippincott Williams & Wilkins, 2003.

Thursday, June 09, 2005

Database Highlight: images.MD

images.MD compiles over 50,000 high-quality images spanning all of internal medicine, all derived from Current Medicine’s series of illustrated atlases. Each image is accompanied by detailed text written by over 2,000 contributing experts.

The images include: histology, pathology, radiographs, original artworks, graphs, tables, or just the state-of-the-art commentary. Detailed references accompany each image. Most importantly, because the Yale Medical Library subscribes to images.MD, you're provided with unlimited personal use of the images in your collections, plus the tools and online workspace to help you use them effectively.

This database of images provides both a basic and advanced search feature as well as an easily browsed index. The My Slides feature lets you create and manage slidesets.

Check it out! You can find images.MD on the Medical Library's homepage under the eResources scan column.

Wednesday, June 08, 2005

Selecting keywords for your journal article

The National Library of Medicine offers some general suggestions for authors as they select keywords for their articles. Use their database to help you locate appropriate vocabulary.

And don't forget that your friendly Medical Librarian can also give you assistance in choosing appropriate keywords for your journal article!

Tuesday, June 07, 2005

Using subheadings in MEDLINE

Many MEDLINE searchers utilize subheadings when they want to hone in on a specific aspect of a topic. There are 83 of these topical subheadings and when used in conjunction with a MeSH term, you can really fine-tune your results.

Here are some examples of frequently used subheadings: adverse effects, complications, diet therapy, drug therapy, diagnosis, etiology, epidemiology, therapy, psychology, nursing, etc.

I use subheadings frequently when I'm doing a clinical search. For example, if my search question has to do with how effective pharmacotherapy is in improving cognitive symptoms in patients with dementia, then linking the subheading 'drug therapy' to the subject heading 'dementia' will produce a set of citations that is specifically about the treatment of dementia with drugs. Then I would continue to search for the other concepts in my search question.


FYI
Don't forget the MeSH website. Check it out!

Monday, June 06, 2005

Indexing articles for MEDLINE

If you've ever wondered how articles are indexed and who does the indexing for MEDLINE, here's a few details.

  • Altogether, about 100 indexers perform the indexing for the National Library of Medicine (NLM) and they are all trained by NLM. Indexers have at least a Bachelor's degree in life sciences, attend a two-week formal training class at NLM, and undergo continuous review for a year. The required production rate is 4 articles per hour.
  • Indexers work in an online environment which provides extensive validation and 'reminder' programs to enhance consistency and accuracy.
  • The sequence is: understand the subject content of the article, apply MeSH terms to cover all topics substantively discussed, apply 10-12 MeSH terms per article, indicate major points of the article with an asterisk, apply the subheadings, apply check tags (organism, age, gender, etc.)
  • Authors frequently want to know how to select keywords for their articles, but NLM doesn't use author-assigned keywords in MEDLINE. Only keywords assigned from the Medical Subject Headings (MeSH) are used. NLM finds that authors frequently choose MeSH terms that are too broad - NLM policy is to select the most specific subject heading possible. A tip for authors from NLM is to use their preferred keywords in the title or abstract where they will be retrievable via textword searching.

For more information:
FAQs about Indexing

Friday, June 03, 2005

Clinical Prediction Guides Added to PubMed Clinical Queries

Check out the Clinical Queries page under PubMed Services on the left scan column.

A new clinical study category has been added - Clinical Prediction Guides.

Clinical Prediction Guides pertain to the prediction of some aspect of a disease or condition and have been advocated as a mechanism for enhancing clinical judgement. Clinical prediction studies develop or validate rules, guides, indexes, equations, scales, scores or models to predict a diagnosis, prognosis, risk (etiology), therapeutic response, therapeutic drug level or clinical outcome.

An example of a clinical prediction guide is the Modified Mid America Heart Institute Coronary Care Unit scoring system - a new comprehensive prognostic index for Coronary Care Unit patients. Med Sci Monit. 2005 Feb 25;11(3):CR95-99

Wednesday, June 01, 2005

Database Hightlight: Clinical Evidence

When you're looking for evidence on the effects of common clinical interventions, consider using Clinical Evidence.

Clinical Evidence summarizes the current state of knowledge and uncertainty about the prevention and treatment of clinical conditions, based on thorough searches and appraisal of the literature. It is neither a textbook of medicine nor a set of guidelines. It describes the best available evidence from systematic reviews, RCTs and observational studies where appropriate, and if there is no good evidence it says so.

The contents are driven by questions rather than by the availability of research evidence. Rather than start with the evidence and summarise what is there, they identify important clinical questions, and then search for and summarise the best available evidence to answer them.

The summaries in Clinical Evidence result from a rigorous process aimed at ensuring that they are both reliable and relevant to clinical practice. You can check out this process by looking at their search strategies and their critical appraisal criteria (under the 'About Us' tab on the Clinical Evidence web site)

Clinical Evidence is available from the Major Resources A-Z list on the Cushing/Whitney Medical Library's home page. And the print version is available in the Medical Library Information Room (Ref 18 RA427 C45)

The Facts

Published by BMJ Publishing Group
Updated continuously, with full literature searches in each topic every 12 months
Available in electronic and print (full-text and Concise)
Web version updated monthly; print version updated twice yearly