This big (8.5 x 11 inches) book contains more than 6400 definitions of more than 4300 terms that have to do with clinical trials and the design, conduct, analysis, interpretation, and discussion of them. As its title forecasts, this book is more than just a dictionary. It is a guide not only to clarity but to good manners when communicating about clinical trials. Its author should know about both; Curt Meinert was a founder of the Society for Clinical Trials and the journal titled Controlled Clinical Trials, and he was at the center of the UGDP trial, the results of which condemned oral hypoglycemic agents and provoked ill-tempered polemics from its opponents.

The book opens with a fascinating set of guides and admonitions on how to inform, illuminate, and fight fair when discussing and debating clinical trials and their results. If limited to just this section, it would already be worth its price for many. After advising readers about redundancies, repetitions, and not saying what we mean, the author moves on to address problems that arise from humanizing the inanimate and from using pre-, post-, and small troublesome words. But the real meat in this section lies in the discussion of the language of criticism, where readers can sharpen their skills in recognizing empty rhetoric (because every study is imperfect, critics are not giving us the news when they describe one they don't like as flawed); recognize the languages of intimidation, implication, position, equivocation, weaseling, and jargon; and reinforce their resistance to the practice of trying to establish truth through declaration and repetition.

Telling quotes open each section of the dictionary (for example, the section for the letter "R" begins with: "For the actual number of people to be enrolled into a trial, divide the number promised by 4 and multiply the time stated for getting them by 2"). I used the main body of the Dictionary for a month and found most of its entries to be comprehensive, concise, and crisp and its recommendations about their use (often not recommended) thought-provoking.

This dictionary let me down only at the interface between clinical trials and clinical practice, and my disappointments were due more to omission than commission. Absent were those measures of treatment effects that we commonly use at the bedside, such as relative and absolute risk reduction and the number of patients that need to be treated to prevent one event. I also missed an important guide to the critical appraisal of a trial report (avoiding co-intervention, the use of additional diagnostic or therapeutic interventions on experimental but not control patients).

The clinical-methodologic interface also identifies differences in emphasis. Dr. Meinert finds the distinction between explanatory and pragmatic trials generally unhelpful, whereas clinical collaborators of mine find this distinction vital when hammering out the questions posed in trials and the means used to answer them. Similarly, clinicians are far more interested in the particularization (term absent) of a trial's result to their individual patient than in generalization to all patients (perhaps neither term quite fits and we should think about using extrapolation for both purposes).

Finally, the Dictionary has not quite captured the international, quickly evolving field of systematic reviews (term absent) of multiple clinical trials. Preferring the general term "meta-analysis," which others reserve for the statistical analyses that illuminate some but by no means all overviews (a discouraged term), Dr. Meinert includes "heterogeneous" but not the bane of meta-analysts, "heterogeneity."

But the author admits that such a work is never finished, and declares only that it is done for now. He has accomplished a brilliant first edition, and it will reward all who use it.