UpToDate has 2 main goals: “to become the first place that clinicians, including students, will go when they need an answer to a specific clinical question”, and to make the information comprehensive, accurate, verifiable (well referenced), easy to access, and updated regularly. Its contents include all areas of internal medicine, but the dermatology, oncology, and neurology sections are still being developed. It is currently available on CD-ROM, with a Web version forthcoming.

UpToDate is organised in the form of “cards” (>3900). For example, see the list of cards obtained by searching the term stroke (fig 1). An example of a card is given in the next screen shot (fig 2), which describes the use of thrombolytic therapy for stroke. Each card may be 3 to 10 pages in length and includes Medline citations with abstracts for most references.

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Explicit criteria for seeking and appraising evidence are not described, but contributing authors are asked to include descriptions of the most important relevant studies and a brief review of quantitative data. They are also asked, in general terms, to note the quality of the evidence, to provide a specific clinical recommendation, and to identify areas of controversy or limited evidence. Internal and external reviewers—experts in the relevant field—subsequently review the cards. A team of in house physician editors work with the authors and section editors to update continuously the entire CD-ROM. This process includes reviewing and incorporating information from >100 clinical journals (including ACP Journal Club and Evidence-Based Medicine). New CD-ROMs are sent out at 4 month intervals.

To explore the extent to which this process leads to inclusion of valid evidence, I conducted a short review in which a colleague not familiar with UpToDate identified 10 clinical questions for which good quality evidence is available. For 9 of the 10 topics (6 related to treatment and 3 to diagnosis), UpToDate cards reviewed the data from the studies we expected to find (abstracts for most of these studies were available, but relevant Cochrane reviews were not always identified). One prognostic topic was not covered (prognosis of acute stroke), and the neurology section is still incomplete.

10 clinical trainees working on a clinical teaching unit in Hamilton, Ontario, Canada, were asked to grade the user friendliness of a variety of evidence resources. On a scale ranging from 1 (easy to use), through 3 (so-so), to 5 (difficult to use), UpToDate was assigned a “1” by all 10 trainees (Best Evidence mean score 1.8, Cochrane 3.8). UpToDate was the most commonly used information database among these trainees.

As I have previously stated, the resource is not yet entirely comprehensive because some areas are still being developed. In addition, the lack of explicit criteria by which the authors support their conclusions may occasionally (in my experience, rarely) lead to a situation where part of the evidence is omitted or an existing controversy ignored. Finally, although UpToDate can be used for general orientation to a topic, people trying to learn more about basic biology, physiology, or anatomy might want to look elsewhere.

UpToDate is practical for answering questions about management issues, and it contributes to patient based learning. Over the past year, use of UpToDate has become part of my daily clinical practice.