Skip navigation

The end of the beginning: the evidence cart

The quotation 'I've seen the future and it works' was made by someone after a visit to the Soviet Union in 1919. You may think it apposite, though, after reading about how rapid availability of evidence makes its use more likely [1].

Evidence cart

In order to find out whether it was feasible to bring evidence to the bedside of a general medicine in-patient service, an 'evidence-cart' was constructed that would both contain the evidence thought to be helpful, and provide the means for projecting and printing it. The cart had a computer and projector, had available computerised information on previously assessed evidence of relevance to a busy medical team, as well as Best Evidence , Cochrane Library, MEDLINE, and pile of other useful information.

It was taken on rounds for a month, during which a log was kept of the ways it was used, and each team member (medical students to consultants) was asked to complete unannounced questionnaires about their use of evidence sources during and after that month.


It was used 98 times during the care of 166 inpatients and 30 more patients who were not admitted. Some evidence sources could be called up quickly enough (10-25 seconds) to be practical on the service. Sixteen clinical questions could be answered using the cart on the ward in the time taken for a visit to the library to answer only one of them.

Most searches were for evidence that could affect diagnostic and/or treatment decisions; the remainder concerned demonstrations of specific auscultatory findings (using a device that allowed several team members to listen to the same stethoscope simultaneously) or concerned issues in biology or prognosis that would affect management decisions.

Over 90% of searches were successful. Those that were not formed the basis for 'educational prescriptions' to search and appraise the evidence, and to add it to the previously-assessed in-house evidence resource. When assessed from the perspective of the most junior team member responsible for each patient's evaluation and management, 37 (52%) of the successful searches confirmed their current or tentative diagnostic or treatment plans, 18 (25%) led to a new diagnostic skill, an additional test or a new management decision, and 16 (23%) led to a change in a previous clinical skill, diagnostic test or treatment decision.

When the cart was removed, the perceived need for searching rose sharply (from 10 to 41 occasions in the 12 team members who remained on the service at the end of the month), but were carried out only in five (12%).


A criticism of evidence-based medicine is that of time. We are busy, we don't have time to mess about with searching, or appraising, or calculating NNTs or odds ratios. All true. So make it available NOW! That's what the team at the Centre for Evidence-based Medicine in Oxford have done, made evidence instantly available - or at least within about 15 seconds.

If it's available, it's used. If not, it's not. From the ward where the evidence cart is used to one of the finest medical libraries in the UK is but a short step. Along the corridor, down one flight of steps, turn right and you are at the library door. Yet even that proved too much of a barrier when the evidence was not available on the cart.

No surprise there. What is a revelation, though, is that judiciously combining electronic evidence from a variety of sources with summaries, reviewed and appraised, of commonly occurring problems (see CATmaker on ) can be done and can be so useful, especially to younger doctors still collecting their experiences to add to their education. What a gift! Youth, enthusiasm, plus experience. Someone should patent it.

We may have to wait before we can buy a 'Docman' - a miniaturised supercomputer that fits in your pocket and puts the evidence in your ear or projects it onto your glasses before you even knew you needed it, but we've seen the future, and it works.


  1. DL Sackett, SE Strauss, for firm A of the Nuffield Department of Medicine. Finding and applying evidence during clinical rounds: The 'evidence' cart. JAMA 1998 280: 1336-8.

previous or next story in this issue