Evidence-Based On Call Acute Medicine was designed to bring “what we really know” about acute illness to hospital physicians. Motivated by the lack of use of evidence in practice, Ball and Phillips compiled this book along with the companion web site (http://www.eboncall.co.uk [not reviewed]) to be a dynamic, continuously revised data resource to inform real clinical decisions in the hospital. As such, we evaluated the book's ability to address our clinical questions on the basis of 3 important criteria: the quickness of retrieving the information, the reliability of the data presented, and the sufficiency (or completeness) of the answers.

Ball CM, Phillips R.Evidence-Based On Call Acute Medicine. Edinburgh: Churchill Livingstone, 2001.

Evidence-Based On Call Acute Medicine presents > 1000 critically appraised topics (CATs) to support the sequence of management steps that make up the care of patients admitted to hospital. The book is quite ambitious, presenting an exhaustive amount of data to support, or refute, some of the most common ideas about important clinical conditions. The authors' present summaries of the data from the CATs with references and recommendations for 37 alphabetically organised major topics. Each chapter is identified not only by a diagnosis but also by a common treatment scenario or presenting symptom (eg, cellulitis, anticoagulation, and syncope). Within each topic, chapters are organised according to the evidence for prevalence, aspects of the clinical presentation, diagnostic tests, and therapeutic interventions.

The reader can find a relevant general topic easily. However, finding an answer to a specific clinical question is a challenge. The indexing is sporadic and is not cross-referenced extensively, limiting its swiftness as a resource for answering specific questions. Despite this drawback, the overall format and presentation of data is easy to use and pleasing to the eye. We tested its quickness on call recently. Wondering whether low molecular weight heparin is safe as an anticoagulant for prosthetic heart valves, we immediately targeted and searched the comprehensive section on anticoagulation, first by using the index and then by general browsing. After 90 seconds, we were convinced that no sound evidence existed to answer our question. A Medline search showed a few reviews of case reports^1,2 but only 1 non-randomised clinical trial that addressed the issue,³ validating the results in the book.

The method of selecting references is rigorous, appraisal criteria are explicit, and grading of data is well done. The authors approached their quest for data by using clinical questions, but specific search terms were not divulged. For each query, they embarked on a search beginning with Best Evidence, proceeding to the Cochrane Library, and finally going on to PubMed only when questions were still unanswered. This method is appropriate and inspires the reader with confidence in the completeness of the information provided.

The irony of a book format to capture the dynamic nature of medical knowledge highlights the necessity of a fully functioning companion web site. The authors assure us that over time each topic will be periodically updated with recent CATs on relevant articles and posted to the web site. Expiration dates are assigned in the text to inform the reader when the search and appraisal process for each topic will be totally redone, but they do not present explicit criteria for determining those dates. Whereas we find the overall reliability of this resource impeccable, the notion of expiration dates misleads the reader, imparting an unwarranted “good-until” seal of confidence to an ever-changing sea of information.

In its current format, Evidence-Based On Call Acute Medicine is certainly not sufficient as a source of clinical advice (as the authors readily admit). For example, it would probably change little of the initial approach to patient management. Knowing the likelihood ratios for reproducible chest pain as a diagnostic test for unstable angina may not change what we do “on call” for the patient admitted from the emergency room with this sign. Furthermore, clinicians face a variety of other challenges in information seeking (eg, formulating a precise clinical question and knowing what to do with the data) that Evidence-Based On Call Acute Medicine cannot and does not pretend to meet.

However, taken as a compendium of evidence, Evidence-Based On Call Acute Medicine holds great hope for accelerating clinical wisdom by offering concise data at key moments of decision making that often occur after the on-call storm. To that end, it has great value and is a powerful educational resource for hospital physicians at all levels of learning.