Reading

Letter to the Editor

I recently had the opportunity to travel a bit and catch up with a few medical school classmates. Some interesting observations emerged about reading medical journals, which might be of interest to you and your readers.

Although we have gone our different ways, a ubiquitous finding surfaced: on all of our office desks (including mine) stood a proud tower of neatly stacked, pristine medical journals with unbroken bindings. Most observed that our patients considered clinician intelligence in direct proportion to tower height, which reached a maximum height of 4.5 feet by one estimate. Several of us devised clever ways to support large towers, using a nearby wall or another journal stack as a counterbalance.

We shared an annual New Year's resolution of self-delusional efforts to someday read our journals. Yet none of us felt lacking relative to our well-read academic colleagues, who are paid generous salaries to digest and regurgitate the tower's knowledge. The limited value of most articles you publish is apparent, considering how well my classmates and I practice our craft without them. Moreover, your new instructions to authors (January 2007) mandate more detailed reporting and disclosure, further ensuring that bindings remain unbroken.

Somewhere inside your journal lurk tidbits of useful information, and I could find them more easily if you published abstracts only, thereby sparing readers the visual cacophony of skipping over complete manuscripts. Better yet, include only the “conclusion” and “significance” sections of the abstract to avoid needless methodologic jargon. At the very least your production costs would plummet, and my colleagues and I could fit 20 times as many journal issues in our respective desktop towers. Your kind consideration is appreciated.

Respectfully yours,

Sheila Shortcut, Stressed Out Surgeon

Stressville, USA

Editor's Response

Nice as it might be to live in a world of sound-bite research, the realities of medical data are couched in probability and uncertainty. The shrewd clinician blends research, judgment, and experience to provide quality care consistent with patient values. Peer-reviewed journals are the gold standard for original research, without which patient care is possible, but rarely exceeds mediocrity. “It is astonishing with how little reading a doctor can practice medicine,” noted Osler, “but it is not astonishing how badly he may do it.” 1

Reading journals reduces uncertainty by vicarious experience. Medical faculty read more than other professionals, including chemists, physicists, engineers, and astronomers. 2 Surgeons, however, read less than internists, with average weekly times of 3.3 and 4.4 hours, respectively. 3,4 University-based medical faculty read more than physicians in private practice, and report using the literature for writing, teaching, consulting, staying current, advising others, and supporting their own research efforts. 2

Fortunately for Dr. Shortcut and her colleagues, many medical problems are self-limited, improving irrespective of what was done. Even then, research defines natural history, crafts realistic expectations, and avoids harm from ineffective or marginally effective therapies. Similarly, deciding whether to use a therapy that is “proven” effective because of “statistical significance” mandates careful consideration of how much benefit vs. harm is provided to what patients under what specific circumstances and at what cost. Abstracts are starting points for teasing this information out of a research article, not a final destination for all-or-none recommendations.

Why can't busy clinicians simply rely on abstracts for evidence-based patient care? Because drawing conclusions from an abstract is like judging a five-course dinner after tasting the appetizer. Nonetheless, in one survey internists reported reading only abstracts for more than 60 percent of articles. 4 Scanning abstracts can efficiently identify articles worth reading, which then merit further analysis to judge validity and applicability. Taking time to savor, interpret, and digest original research can be just as fulfilling as a gourmet meal. In just a few moments readers can learn and benefit from months or years of effort by the study authors.

To put the limited role of an abstract in perspective, consider the anatomy of an ideal manuscript. A concise introduction arouses interest and states clearly the study rationale. Next, methods are described with enough detail for reproducibility. Results logically blend numbers and narrative, with supporting tables and figures for key findings. A discussion puts main results in context with a candid assessment of study strengths, weaknesses, and generalizability. Lastly, references demonstrate that work of others has been duly considered. Condensing all sections into an abstract introduces potential biases, distortions, and omissions, which can only be overcome by viewing the manuscript as a unified whole.

Reading journal articles often involves detective work, because conclusions in the abstract and discussion sections may or may not be reasonable. For example, almost 90 percent of recommendations for surgery in major otolaryngology journals are based on case series without control groups. 5 Editorial peer review provides some quality control, but the process is largely untested and its effects are uncertain. 6 Only about 1 in every 14 articles published in peer-reviewed, primary healthcare journals have both high-quality methods and clinically relevant material. 7 The bottom line? Caveat lector: beware of what you read even in excellent journals.

What is the best way to read a journal article? If an article has the potential to influence clinical care, or requires indepth analysis for a journal club or other academic venue, then critical evaluation is facilitated by answering five questions: 8

Reducing manuscripts to sound-bites ignores that most research facilitates communication from scientist to scientist, not from scientist to clinician. 9 Published studies are often non-definitive tests of hypotheses and innovations, only a very small percentage of which may warrant routine clinical application. Whereas the science may be sound, the idea has not progressed beyond the laboratory, a case series, feasibility study, or preliminary report. Definitive studies constituting true scientist-to-clinician communication are rare in medical journals, and must be identified by critical appraisal of full-length articles.

The pitfalls of sound-bite reporting are illustrated daily by the media and lay press. In this fantasy world of feigned certainty, interventions either work or they don't, are effective or ineffective, and can be recommended or not. Concepts of validity, confidence, probability, effect size, and harm benefit analysis do not adapt readily to the industry norm of 7-second sound bites. Yes, there are always exceptions, but the pattern of an eye-catching title followed by strings of sound bites, generalizations, and bottom-line conclusions is disturbingly common. The errors and distortions in some reports, however, may only be obvious when the reader is already well informed about the subject matter.

The art of medicine lies in adapting knowledge gained by reading and experience to fit the unique needs of individual patients. Reading alone is insufficient, because research yields probabilities based on group outcomes. Seasoned experience adjusts probabilities to the values, preferences, and specific circumstances of a patient's illness or disorder. Abstracts provide a smattering of sound-bite probabilities, but complete articles add the needed perspective for merging probabilities with experience to achieve wisdom-based medicine.

As journal editor I strive to identify manuscripts worth reading that can be used to improve patient care. No amount of due diligence by me or the editorial staff can substitute for critical evaluation of what we publish by our readers. Yes, Dr. Shortcut's ideal journal with abstracts only would make delightful bathroom reading, but would not facilitate quality care. Regrettably, cognitive effort remains essential to reading journals, and will, as always, be rewarded with intellectual clarity and a humble respect for the perennial uncertainty that stimulates lifelong learning.

The words of Francis Bacon offer a fitting conclusion: “Reading maketh a full manhellip;and if he read little, he had need have much cunning, to seem to know that he doth not.” 10