Everybody hates curbside consults – the informal, “Hey, Joe, how would you treat asymptomatic pyuria in my 80-year-old nursing home patient?”-type questions that dominate those Doctor’s Lounge conversations that aren’t about sports, Wall Street, or ObamaCare.

Consultants hate being asked clinical questions out of context; they know that they may give incorrect advice if the underlying facts and assumptions aren’t right (the old garbage in, garbage out phenomenon). They also don’t enjoy giving away their time and intellectual capital for free. Risk managers hate curbside consults because they sometimes figure into the pathogenesis of a lawsuit, such as when a hospitalist or ER doctor acts after receiving (non-documented) curbside guidance and things go sideways.

There is some evidence to support this antipathy. A recent study published in the Journal of Hospital Medicine examined 47 curbside consultations by hospitalists, in which formal consults by different hospitalists (unaware of the details of the curbside encounter) were performed soon thereafter. Conducted by a team of researchers from the University of Colorado, the study found that the information given to the curbside consultant was incomplete or inaccurate roughly half the time, and that management advice offered via the two forms of consultation differed 60 percent of the time. (In those cases in which the consultant was given inaccurate or incomplete information, the advice differed more than 90 percent of the time!) This is not the first warning about the dangers of such consults (see also here and here), and it won’t be the last.

I recall several cases at my own institution in which curbside consults contributed to mistakes with tragic consequences. In a memorable one from nearly a decade ago, a cardiology fellow was curbsided to look at an ECG in a young patient with non-cardiac symptoms. The primary team attributed her symptoms to a pulmonary illness and asked the specialist whether the patient’s potentially alarming ECG findings could be seen in that syndrome. His “yes” answer – which may have been correct in theory – led the team to stick with its original diagnosis, an error that contributed to the patient’s death. There seems little question that had the cardiologist actually seen the patient, reviewed the history in detail, and looked at the electrocardiogram in that context, his recommendation would have been different, and the outcome might have been as well.

Cases like this, and studies like the JHM paper, inevitably cause some to lobby to ban curbside consults, and I’ve heard of a few organizations and subspecialty services that have done just that. While such a move seems logical on the surface, my own belief is that it would be an extremely dangerous thing to do.

Imagine a law firm or a business in which none of the partners were comfortable asking for the advice of a colleague without a formal, written request. Let’s also imagine that the advice could only be given after the colleague had reviewed the case file for half an hour and spoken to the client himself. This would be a disaster – collegial, informal exchange of information and ideas is what lubricates the gears of every effective organization; it’s what a “learning organization” looks like. Stripped of this lubrication, the machinery freezes up; before you know it you have a hidebound, bureaucratic monstrosity. One of the key lessons of the past decade is that healthcare organizations are so-called “complex adaptive systems,” in which formulaic approaches tend to fail. In such organizations, it’s crucial to nurture the informal connections that allow for the diffusion of wisdom: from senior leaders to front-line managers, from teachers to students, and yes, from specialists to generalists.

What would actually happen if we did ban curbside consultations? Picture a resident caring for a patient with a tough case of C. difficile colitis. The resident spies an overworked, underfed GI or ID fellow looking harried. The resident would realize that if he asked the fellow the question, “Cynthia, I got this guy with recurrent C. diff. What’s the best treatment?” the response would be: “Are you asking me for a formal consult?”(Accompanied by a Please-God-No facial expression and toe tapping that would put Savion Glover to shame). Faced with that scenario, it’s likely that the resident wouldn’t ask the question of the specialist, instead choosing to wing it with the help of UpToDate or perhaps another resident’s advice. In other words, while we know that curbside consults can be dangerous, what we don’t know is how much useful information is transmitted via such consultations, and whether the advantage of better formal consults would trump the loss of shared wisdom through this fractal information market. I suspect it would not.

As with so many complex issues in medicine, the right answer will require a nuanced approach. For complex clinical questions whose answers truly hinge on the consultant having a deep understanding of the patient’s history, physical examination and clinical situation, a full-bore consultation is appropriate and should be required. I’m guessing that that description covers the minority of day-to-day clinical situations. To deal with the others, we need to get creative. Rather than banning curbside consults, we should develop new “consult-lite” models: ones in which the consultant feels comfortable opining without being obligated to see the patient and the complete dataset. For example, a pulmonologist might be comfortable rendering a recommendation after hearing a thumbnail history and seeing a chest CT; a dermatologist might need little more than a photo of a rash; a neurologist might be able to observe a hospitalist examining a patient through a video link and make a recommendation with confidence. (These recommendations and the information on which they were based should be briefly documented in the medical record.) Of course, there is a chance that their judgments might have changed had they spent 30 minutes talking to and examining the patient (and some will undoubtedly complain about the further dehumanization of medicine), so we need to weigh these concerns against the efficiency of this consult-lite approach.

To make all of this feasible, it will be important to take the matter of professional fee billing off the table. One hopes that the movement toward a value-not-volume payment system will give healthcare organizations the freedom to organize themselves in ways that promote appropriate types of consultation and information exchange. In places like Kaiser Permanente, which offer us a head start on envisioning the future, they try to maximize these informal interactions (for example, through co-locating specialists and primary care doctors in ambulatory practices). Their hope is that this structure allows primary care doctors and hospitalists to deliver appropriate care, less expensively than requiring that specialists be involved every step of the way.

A complete prohibition of curbside consultation would create only two options when it came to generalist-specialist interactions: purely educational forums (aka CME) and formal consultation. There’s a lot of good care and information exchange that lives in-between these poles. In a 2005 Annals of Internal Medicine article, Ferrer and colleagues addressed the need for flexibility when it comes to generalist-specialist interactions:

Generalists should work with specialists to address the following questions: What are the volume–outcome relationships for a specific condition in both primary and specialty care and how can they be optimized? When is referral too early, and when is it too late? They should create communication patterns that support the proper selection of steps along the referral continuum of advice, formal consultation, co-management, or referral.

To thrive in the next decade, healthcare delivery systems will need new models of generalist-specialist information exchange that produce the best outcomes at the lowest cost. To be successful, these models must leave all of the involved clinicians feeling professionally satisfied, protected from undue malpractice risk, and fairly compensated. I’m pretty sure that they will not include an outright ban on curbside consults.

In 2003, cutting residency duty hours seemed like a straightforward solution to a tangible problem: resident fatigue. Yet we now know that this change did not result in improved safety (or education, for that matter), because we failed to address the many collateral issues, ranging from resident scut work to the dangers of handoffs, nor fully appreciate how care is actually delivered in the trenches. Banning curbside consults is similarly seductive: it seems like a straightforward solution to a palpable problem. But I’m afraid it would likely have the same unfortunate outcome. Let’s be smarter – and more imaginative – this time around.

Bob Wachter is chair, American Board of Internal Medicine and professor of medicine, University of California, San Francisco. He coined the term “hospitalist” and is one of the nation’s leading experts in health care quality and patient safety. He is author of Understanding Patient Safety, Second Edition, and blogs at Wachter’s World, where this post originally appeared.