Save patients from media hype

My family was happily scaling sand dunes on a vacation in Nevada in 2002 when our idyllic bliss was interrupted by a panicked call from one of my nurses.

"What do we tell all these women calling about their hormones?" Miles from civilization, I had no idea what the fuss was all about. My nurse explained that news reports about the Women's Health Initiative results had prompted several of my patients, suffering from hot flashes but terrified of instant death, to abruptly stop hormone replacement during a stifling heat wave.

After returning home, I told these women that they had only a very small individual risk of dangerous side effects. Further, I explained that the benefits of hormone replacement in small doses, for a short time in low-risk women with severe symptoms, far outweighed the risks.

How do we explain "hot" health news and data in terms our patients can understand? First, we have to comprehend these statistics well enough ourselves to correctly interpret them when patients arrive, printout in hand.

Taking the Greek out of study results

Evidence-based medicine has entered the media's and thus our patients' vernacular. EBM integrates a doctor's clinical expertise, the best current research evidence, and a patient's values and expectations, to make decisions about the care of that individual patient. It's a fine principle, but you can drown in the numbers. Just to discuss the reliability of diagnostic tests, we're awash in a sea of figures: sensitivity, specificity, accuracy, number of false positives and negatives, and likelihood ratios. To determine if research findings apply to our patient, we first look at the studied population and measured outcomes before we confront the next wave of figures: statistical and practical significance, statistical power, absolute and relative risk reduction, and numbers needed to treat or harm.

For help in understanding these terms and figures, I turn to the second edition of late journalist Victor Cohn's book, co-authored with Lewis Cope, News & Numbers: A Guide to Reporting Statistical Claims and Controversies in Health and Other Fields (Blackwell Publishing, 2001). This volume makes clear the sense behind medical and scientific stats, in terms you can use with patients. (Online excerpts from the first edition are available at http://www.facsnet.org/tools/ref_tutor/newsandnumbers.) For more resources to help you understand, access, and interpret statistics, see the box.

How patients read medical studies

Besides understanding the data, we should also understand our patients. Their values and experiences are an integral, and often overlooked, part of the decision-making process. Patients have a laundry list of issues that color their feelings about treatment-fear, family matters, religion, cost-and it's important to discuss these. But let's look at what impacts a patient's view of medical research. Here are some misconceptions I frequently encounter.

Risk is imminent. Patients fear problems that are random, rare, devastating, or beyond their control. Reports from the media on environmental toxins like mercury, or infectious diseases such as bird flu or HIV, only fuel these worries. Our job is to put the patient's true risk in context.

Association equals causation. Many patients don't know the difference between association and causation; just because two things happen at the same time, doesn't mean that one causes the other. After all, if you touch a toad and later develop a wart, you can't necessarily blame the wart on that toad.

If it's published, it must be true. People tend to believe everything they read. Yet, authors may be biased, and even editors of medical journals, not just talk-show hosts, can be fooled by authors' fabrications.