Records of Children’s Hospital Treatment, 1883-1903

There’s a new a Web site of records that chronicle the treatment of children at the Glasgow Hospital for Sick Children from 1883 to 1903. The site is part of the Historic Hospital Admission Records Project and is available at http://hharp.org/.

The site now has about 120,000 records, of which the recently-added Glasgow Hospital records are one-tenth. At the front page you can do a first and last name search as well as a year of birth (exact or a range). I did a search for John Smith, no birth date. I got 60 results, of which 20 were available (all are available if I register/log in. Registration is free.) Search results include admission date, age, name, health issue, registration district, and the institution (the key for registration abbreviation is available at the top of the search results.) Some search results had case notes associated with them but you have to be logged in to view those.

I took a look at an 8-year-old John Smith who was admitted with rheumatism. The details page included personal details, details about the admission and length of stay, information about the health issue, admission date, and the outcome (in this case cured.) If you’re logged in and can view the case notes, they’re viewable page-by-page on the site or which can be downloaded as a single PDF file.

For more background on the records behind this site, check the About page, but also take a look at the Academic Resources page, which will give you pointers to some historical context, and the extensive and fairly well-annotated Links List.

American Medical Association Opens Up Its Archives

The American Medical Association announced yesterday that it has knocked down its access control wall for its American Medical News. The site now has ten years of full content and some older selected articles — all told, an archive of about 15,000 articles. You can access the archives at http://www.ama-assn.org/amednews/site/archives.htm. And as the AMA said in its announcement, “We invite our readers to visit the archives and link to our articles from their own sites, blogs and posts.”

The archives have full text back to 2000, with selected content back to 1995. Articles are divided up into topic/section. Poking around randomly found me articles like Fall-related injuries cost Medicare billions, Handling the toxic employee: How to avoid — or dilute — the poison , and Patient safety improving slightly, 10 years after IOM report on errors. There’s also an article search page. A search for substance abuse found 314 results sorted by date.

While you’re here, you might also want to check the new mobile version of the American Medical News at http://www.ama-assn.org/amednews/site/m.htm. Movile updates are available via e-mail or RSS. Plenty to read here.

Download PubMed Search Results Into a Spreadsheet with FLink

Hat tip to Patricia at the Dragonfly blog for her pointer to FLink, (Frequency-weighted Links), a tool for downloading PubMed results into a CSV file (suitable for opening in Excel or OpenOffice.) You can access FLink at http://www.ncbi.nlm.nih.gov/Structure/flink/flink.cgi.

You can use Patricia’s post for step-by-step directions for using FLink, but it’s pretty easy. Use the pulldown menu to specify the database you want to search (there are a few available besides PubMed.) Then from the popup window choose “Search Entrez” and enter your search keywords. FLink will think about it a minute and show your search results in the browser window. (I searched for autism and got 16502 results.)

From there, click “Download CSV.” In the case of the autism search I did, it took several minutes for FLink to assemble a CSV for me — but once it did, it was a 5MB download that listed over 16300 articles from PubMed, including PubMed ID, Authors, Title, Month and Year of publication, and Summary.

Now if direct links to the articles were included you could jimmy up a little Perl and have a nice download utility, but unfortunately links are not available. However just to be weird I took the last thousand article titles in the downloaded CSV (I wanted to do more but Wordle got stroppy) and made a tag cloud of the top 75 words.

A nice tool. I feel guilty for wishing there were more data downloaded with the CSV.

2.5 Million Images from Biomedical Literature

The National Center for Biotechnology Information (NCBI), a division of the National Library of Medicine (NLM) at the National Institutes of Health (NIH), has developed and will maintain the newly-announced Images, a database of 2.5 million images and figures from medical and life sciences journals. The new site is available at http://www.ncbi.nlm.nih.gov/images.

From the front page you can browse recently-added or most-cited literature, but you also have the option to do a simple search by keyword (an advanced search builder allows you to search through several different fields.) I did a search for autism and got 1580 results, mostly charts. (“Fig. 4. From: Automated vocal analysis of naturalistic recordings from children with autism, language delay, and typical development.”) The search results have a thumbnail, paper title, and a brief amount of context; there are also links to a citation/abstract for the article as well as full text.

This is not a fuzzy-wuzzy biology search engine; the vast majority of the results I saw were data charts. Try doing multiple keyword searches for concepts which you’re trying to relate; the first result for thyroid diabetes found me a very interesting Venn diagram.

ebrary Makes an Open Access Database on Breast Cancer

In recognition of National Breast Cancer Awareness Month, ebrary has created an open access database about breast cancer at http://site.ebrary.com/lib/breastcancer.

The front page is divided up into tabs, but I found the first tab and the search page most useful. (When I tried looking at the other tabs, ebrary suddenly decided I had to “log in” even though I was on an openly-available site. I had to open the open access database in a different browser to escape that.) The front page provides a list of “Featured Titles” though you can really explore via the topic list, which ranges from “Breast cancer treatment” to “Clinical trials.” “Disparities in Health Care” produced eight documents, most from the National Cancer Institute, while “Alternative Cancer Treatments” provided a wider variety of results, including one in Spanish.

Documents are viewable via a slightly awkward in-browser reader; terms for which you searched are highlighted. You can also search for similar documents, show the table of contents, or export bibliographical information to EndNote, Citavi, or RefWorks.

I did a simple search for genetics and got 27 results, from Family Studies Research Portfolio (National Cancer Institute again) to Cancer Clinical Trials : A Resource Guide for Outreach, Education, and Advocacy.

The resources were well-organized and easy to access, but I very much wish there were more here. As a comparison I went to the PLoS Medicine site (PLoS Medicine is a peer-reviewed medical journal published under a Creative Commons license; it’s available at http://www.plosmedicine.org/home.action) and did a keyword search for “breast cancer”. I found 129 results. Expanding that search across all of the PLoS journals found over 1800 results.

If you’re looking for information on breast cancer, ebrary’s new database is a good place to start. Just don’t stop there.