In my last post I spoke about the difference between pay for performance (P4P) as opposed to pay for outcomes. There is a difference, so it's worth spending some time looking at them.
P4P tries to raise all boats to the same level. In British Columbia, family physicians are paid for ensuring that their patients are followed according to certain guidelines. So diabetic patients for example would need to have certain tests like a Haemoglobin A1c done every three months.
Of course, most of us, if we are smart should be following guidelines, so in essence the doctors are being paid for showing their homework. That takes time which is one reason why many physicians resent having to log into a government site to file their work. It is also clerical work, which is something that most doctors are not tempermentally inclined towards. This could be overcome by having electronic medical records automatically do the reporting.
From a management point of view, this does provide some indication of how many doctors are showing their work, but it's a fair bet that many others are doing this but not reporting. There are other problems. There is no direct link to patient outcomes for one, and for another, guidelines tend to ossify - and slavish adherence to them discourages innovation. Finally, these guidelines can drive costs upwards, especially when we look at diagnostic services like labs and imaging.
To use an analogy used by Michael Porter and Elizabeth Olmstead Teisberg in their book Redefining Health Care, most of us don't care how our cars are repaired or how long it took to repair them. We just want the things to work well when they come back from the shop. Most don't care whether our cars were filled with synthetic or mineral oil. And frankly most patients only care about feeling better - and they'd like to get that way with the least possible effort.
Enter pay for outcomes. Physicians could be paid according to the risk adjusted outcomes they acheive. The paying agency wouldn't care how those results were acheived, nor would it put roadblocks in the way. Things like restrictive formularies would be dispensed with, guidelines would be published but not enforced and innovation would be encouraged. We wouldn't care if the work was done by a physician, a nurse, or a mechanic, as long as the outcomes were good.
So how would we do this? Exactly what do we mean by an outcome? Those details are important. We can easily reward the wrong behaviour if we don't get it right. Fortunately in BC we have some time. The next Physician Master Agreementbetween the BCMA and the Province of British Columbia won't be signed until 2012.
Let's consider who we should pay. We could pay individual doctors. That would have the advantage of not requiring any significant changes in infrastructure, but comes with some disadvantages as well. It ignores the fact that many individuals contribute to an outcome. If we were to reward groups of doctors, we would be giving them an incentive to work and play well together.
Newsflash: that's not how we've been trained. We've been trained to excel individually and to distrust the conclusions raised by others. That training has been re-inforced by the court system. Unfortunately, none of this is in the patient's best interest. Patients complain about having to tell their story over and over again.
So let's assume that we encourage physicians to form into incorporated networks. These networks could be at the community level, the hospital level or both. At the hospital level, the network could consist of a tertiary care group that would include the surgeons, internists, radiologists and pathologists who contribute to an outcome. In the community it might include family physicians and the consultants who support them.
Ideally, we would want to report outcomes publicly. Network A gets these results for a particular condition, as opposed to network B. I'd take this a step further and have the health authorities (all six of them in BC) report their results. And the Medical Services Plan should benchmark it's results against other paying agencies in Canada and the USA.
Right now, I'm not sure what the outcome indicators should be. In my own specialty, we might want to look at days missed from school because of asthma as being a good indicator of asthma care. Cardiac networks might want to look at the percentage of their patients who return to work.
We also need to consider what should trigger the payment. Would we reward only the top 80% of physicians and leave the 20% at the bottom with a goose egg, or should we consider rewarding only those who acheive a certain target?
I look forward to your thoughts on this idea. Let's six hat it- what is good about this, what is bad, how can we make it better, what data do we need, and what is your gut reaction?
