Expert Advice and Patient Expectations: Title and subTitle BreakLaboratory Testing and Antibiotics for Lyme Disease

The reported incidence of Lyme disease, an arthropod-borne infection, is about the same as another spirochetal infection, syphilis.¹ But Lyme disease receives far more public and media attention than the sexually transmitted infection does.² However, this attention is not hard to understand: Lyme disease is primarily a disorder of suburban, educated middle- and upper-class people. Lyme disease can be as disabling as syphilis, but there usually is not a stigma to having Borrelia burgdorferi infection. The demographics of Lyme disease also make it likely that those at risk of infection would be comparatively knowledgeable about the disease and may even have opinions about how to manage it. Does this level of interest—a set of patient expectations and perhaps even demands—influence care itself? In this issue of THE JOURNAL, the article by Fix and colleagues,³ while not directly answering that question, does reveal a possible consequence of patient expectations. The article also demonstrates that laboratory tests and antibiotic therapy—whatever the driving force behind them—often are not consonant with expert advice about Lyme disease care.

Fix et al studied the use patterns of serologic testing and antibiotic therapy for tick bites and suspected Lyme disease in an endemic area for B burgdorferi infection near the Maryland shore. In their study, as in studies from California and the Netherlands, tick bites were a common reason for people to seek medical care.^{4 - 5} Fix et al found that about half of the patients who presented primarily for tick bite had a serologic test performed at the initial visit but did not have the follow-up "convalescent" assay. Most patients with tick bites received antibiotics regardless of whether a test was performed or regardless of the result of the test. In most cases of suspected or confirmed Lyme disease, the recommended Western blot was not obtained after a positive or equivocal result of enzyme-linked immunosorbent assay or other first-tier test.⁶

The issue of antibiotic prophylaxis after tick bites in an endemic area remains unsettled. Some physicians treat empirically, whereas others withhold antibiotic treatment.^{3 ,7} The sum of published studies is equivocal about the value of prophylaxis in most instances.^{8 - 10} But if there are any benefits from prophylactic antibiotics compared with their cost in dollars and adverse effects, study populations including thousands of patients may be needed to detect a significant difference, because Lyme disease is so infrequent after a tick bite.^{8 ,10} Less controversial among the experts is the issue of a single serologic test in the workup for a tick bite: the only justification for drawing blood at the first patient visit is if antibiotics will be withheld and a repeat serologic study will be performed a few to several weeks later.⁶ There is little or no chance that a patient would have detectable antibodies to B burgdorferi from a new infection at the time of the tick bite. A "positive" serologic test at that time would likely represent either a false-positive result or evidence of a past exposure to the agent.

The findings of the study by Fix et al are important for clinicians who manage patients with Lyme disease and for those who pay for that care. Clearly, money was wasted in many instances. But the findings also raise questions of broader significance; for example, why were the physicians' decisions so often at odds with guidelines of the experts? As the authors point out, the results may simply reflect that physicians did not know what the recommendations were. But in an area with such a high incidence of Lyme disease that explanation is hard to imagine. Alternatively, the authors point out that many patients are "intensely concerned about Lyme disease and strongly desire chemoprophylaxis as well as serologic testing for tick bites" and wonder whether physicians were simply responding to patients' concerns. In a study of the use of serologic tests for Lyme disease in California, almost half the tests were initiated by the patient rather than the physician.⁴ This may be the equivalent of the patient's expectation for antibiotics for a viral upper respiratory tract infection, notwithstanding the lack of proven effect of such treatment.¹¹ In different societies, patients or their families often expect to receive medicine when they see their physician.^{12 - 14}

What may be different about Lyme disease is that patients may not only wish in an ill-defined way for some sort of prescription but also may expect a particular blood test and a specific antibiotic,^{3 - 4} the effect of a better-informed, if not necessarily well-informed, population. Currently, there are many sources of information about Lyme disease, much of which is in disagreement with the experts' advice.² These sources include the Internet, books on Lyme disease written by laypersons, and pamphlets, newsletters, and call-in help lines of patient advocacy groups. A person calling or seeing a physician about Lyme disease frequently has already obtained information about the disorder from at least one of these other sources.

Of course, some issues, such as the details of antibiotic prophylaxis, remain unresolved, and a thoughtful discussion among all concerned can be useful for finding new approaches to the problem. But most expert recommendations appearing in peer-reviewed journals and professional books are reasonable and based on well-controlled trials or studies. When the experts are of one mind about a recommendation, physicians who choose to act contrarily do so out of ignorance, out of a belief in their own experience, or in response to patient expectations. We can applaud the greater participation of patients and their families in decisions about their medical care, but physicians may be too quick to accede to patients' demands in some instances.

Studies of this phenomenon in other situations, such as the expectation for antibiotics for upper respiratory tract infections, have shown that patients are more willing to accept a wait-and-see strategy than physicans are willing to give them credit for.^{12 - 15} But for patients to be satisfied with the no-test, no-medicine option means that additional time must be taken to explain the rationale behind the decision.¹⁵ Although patients may have many facts about Lyme disease—perhaps more than their physician—many of those "facts" may be inaccurate or need to be placed in the right context. Doing so requires straight talking from the physician, rather than reaching for the lab order sheet or prescription pad.