Value, in medicine as in other industries, is a simple ratio of quality to cost:

 Value = Quality/Cost

As quality goes up or cost goes down, value increases. In the clinical setting, though, consider the tension that this produces: on the one hand, patients value “facetime” with the doctor. There is some validity to their desire. Studies consistently show that outcomes for patients who like and trust their doctor are modestly improved compared to patients who do not have a good relationship with their doctor.

On the other hand, more powerful studies show that it’s not necessarily the doctor-patient relationship that produces the best outcomes on a population basis (like the population of employees your company insures). Instead, it’s the careful application of care protocols.

This is hard to swallow. We’re all the stars of our own sitcoms, and we’d like to think that our doctor makes a personalized, artisanal, set of decisions for our care at each visit based on solid training and artful instinct. Unfortunately, that philosophy is not born out in the data. This is because most clinical interventions have a small effect on the individual patient. Prescribing a common older class of osteoporosis medications, for example, reduces the three-year risk of a hip fracture from three percent to two percent (a 35% relative risk reduction, but only a 1% absolute risk reduction). But that effect, which is small to the individual patient, is extremely powerful when amplified up to the population level: preventing 1,000 hip fractures in a population of 100,000 patients over the course of a few years is an amazing outcome. Each hip fracture costs more than $50,000, and hip fractures cost American healthcare purchasers almost $6 billion a year, a number big enough to really move the value equation above. The contribution of a single encounter with a patient to that outcome is almost trivial, though.

As sexy as House-style exotic diagnoses of individual patients can be, we’re better off thinking in terms of treating populations, not patients, with protocolized strategies. We’ve blogged before about examples of this, like Kaiser Northern California’s hypertension treatment algorithm that is largely executed by their medical assistants, or MAs. The MAs are overseen by the doctors, but the algorithm itself makes most of the decisions. With this approach, Kaiser has such good blood pressure control system-wide that their patients are more likely to die from cancer than from a heart attack. This is astonishing.

To the older attending physicians I worked with during my medical training, there was hardly a more biting insult than to say that someone was practicing “cookbook medicine.” They considered the decision-making process in any individual patient encounter sacrosanct. And Kaiser’s protocol is proudly “cookbook,” at least for the majority of patients. The job of the modern physician, then, is to tease out which patients may not be well-served by the “cookbook.” A small but non-trivial number of patients have underlying medical conditions, drug allergies, or other factors that may make them poor candidates for certain algorithms. These are the patients from whom the doctors should be making their money.

So the next time you’re thinking of who is going to care for your employees, especially if you’re thinking of direct contracting with physicians (which we’re very much in favor of at KBGH), ask the practitioners what they think of this philosophy. Is every patient an individual for whom a personalized solution must be hand-crafted for every problem? Or are they more interested in good systems, especially for common illnesses, screenings, transitions of care, and continuity of care? If their answer is the former, dig deeper before you commit.

As the Medical Director of the Kansas Business Group on Health I’m sometimes asked to weigh in on hot topics that might affect employers or employees. This is a reprint of a blog post from KBGH.