At the annual Harvard Healthcare Conference a few years ago, I had the pleasure of listening to David Kirchhoff, the CEO of Weight Watchers, speak about the obsession in America around weight loss and the obesity epidemic. And, although the years have somewhat impeded my exact memory of the presentation, I distinctly remember starting to think about the people who went to these Weight Watchers clinics.

I started, in fact, to think about Weight Watchers themselves as a singular organisation. And as Kirchhoff continued to speak about the revenue and the research and development objectives and the number of employees, I began to wonder if Weight Watchers was a pharmaceutical company disguised as a lifestyle management outfit or a series of obesity clinics disguised as weight loss clinics. Or something in between. Or none of the above.

And then I began to wonder about the people (patients?) who attended these Weight Watchers venues. If, in fact, these are patients with real illnesses, co-morbid conditions and psychosocial issues, then don't we need to completely change our model of how we view them if we're going to ensure the provision of services to this patient population?

If we expect that obese people are going to be sitting in our physicians' waiting rooms with their hands neatly folded on their laps waiting for their names to be called so that we can refer them for bariatric surgery, gastric bypass surgery, sleeve gastrectomy or even a less (supposedly) invasive approach like The Lap-Band System, then maybe we're wrong. Maybe, we're going to keep on waiting and the reality is that these patients are not going to show up in our physicians' offices any more. Maybe we need to ensure that the delivery of care to these patients takes into account where they actually are and not where we expect them to be.

What about mental health patients? Another classic example of 'assuming where the patients will be' instead of actually 'knowing where they are'. If I asked you to name the largest mental health institution in North America (aka the largest mental hospital), what would you guess? McLean Hospital? Johns Hopkins? Maybe New York Presbyterian? You'd be wrong on all three counts. The largest mental hospital in North America is Cook County Jail. A jail.

If we're going to change the slope of the mental health curve and bring meaningful interventions to this group of patients, then we ought to know where a significant majority of them are, shouldn't we? If your argument is something along the lines of 'are we really going to change the way we deliver mental health services based on a cohort of people who are behind bars?', then you know better than that. You know that the incarcerated population is simply a surrogate marker for any disenfranchised population segment that has unequal access to mental health services.

The goal should be to work towards a model of geolocation of care

There are, of course, other examples. Plenty of other examples of disease states where patients are not lined up neatly waiting for the provision of services. Disease states where the lines are blurred between where we think patients are and where they actually are. And the point of these examples is that we need to inject, within our health policy thinking, significant brainpower aimed at better understanding where patients are and, more importantly, how to deliver to them the services they need where they are.

Instead of continuously trying to devise ways of driving patients to the provider, the goal should be to work towards a model of geolocation of care. We have become adept at throwing around buzzwords like 'point of care' and 'point of diagnosis' without stopping to think that this 'point' is not a fixed point (as the fixed point theorems of mathematics might suggest). In fact this 'point' is not a point at all. It is a series of points. And all of these individual points that make up this larger 'point' of care represent patients.

We've spent much time debating and discussing the cost of therapies and how we can ensure price transparency in order to control costs. We've gutted ourselves with the angst of trying to prevent the cost shifting happening in the private payer market with higher co-pays, annual caps and lifetime caps. Everywhere you turn, someone wants to talk about cost. Every think tank has a slant on the topic. Here's a sobering thought: can you imagine how much we'd be spending on healthcare if we actually knew where all the patients were?