How Should Your Doctor Be Paid? Part 2: Episodes of Care

We’ve had some great discussion in the comments on Part 1, so if you haven’t seen it yet, I’d suggest going back and taking a look.

The presidential campaign of 2008 wasn’t that long ago, and one of the areas of health care that John McCain reliably talked about was paying doctors based on “outcomes,” though he was never specific on quite what he meant by that.  Barack Obama also campaigned on “aligning incentives for excellence” – again, a little hazy on what that meant or how you get there.  To patients, these seem unobjectionable because they fit in the mindset of “I got better, so my doctor should get paid.”  However, it scares the bejeesus out of the doctors I know.  If their pay entirely shifted to an analysis of whether the patient got better, they’d be at the mercy of the patient following instructions.  The nightmare looming in their mind is the patient who is overweight, with high blood pressure and warning-sign cholesterol levels, who irregularly takes his pills and makes no serious effort to change lifestyle.  Their compensation would be very poor indeed – through no fault of their own.

In the 1970s, we were wrestling with the same questions.  An alternate method to fee for service that some began using them was “capitation” payments.  The doctor or hospital would get a lump sum per patient per year.  There’d be a financial incentive to keep the patient healthy – if the patient only needed to come in once or twice per year for preventative care, you’d have money left over at the end of the year, which became your profit.  This was the model behind the HMO.  Unfortunately, the money again trumped patient care as a consideration.  Capitation directly led to the HMOs’ reputation as likely to seemingly arbitrarily deny or restrict care, as well as helped accelerate the process of “cherry picking” healthy customers who you knew you wouldn’t use the whole lump sum on each year.

That cautionary tale brings us to the new proposal – one that’s been embraced both by Tom Daschle and by Sen. Max Baucus in his white paper on health care reform.  This is payment by “episode of care.”  Rather than paying for each procedure/service used in the treating of an ailment, the doctor or hospital is compensated a lump sum for everything to do with that ailment.  So for hip replacement surgery, it’s not one fee for the pre-op X-rays, and one fee for the anesthesia, and one fee for the surgery, and one fee for the hospital stay, and a regular set of fees for the rehabilitation process – it’s one fee for the entire health care “episode.”  The other example cited by WSJ’s Health Blog is oncology.  Currently, oncologists are paid for prescribing and administering chemotherapy drugs with an incentive (conscious or not) to prescribe more expensive drugs, and not much financial incentive to do anything else.  The “episode of care” model would give a lump sum for a few months of cancer treatments – everything that goes into treatment.  The doctor is then able to focus on what’s necessary for the patient.  An “episode of care” program for oncologists is currently being tried out by United Health.

Daschle says payment by episode of care is worth trying in a pilot program through Medicare because it could lead to “better outcomes and lower cost, and far less hassle for providers.”  And indeed, there have been some positive experiments.  Medicare has already tried the lump-sum approach with hospitals – a Daschle-led pilot program would roll it out with doctors in private practice, as well as situations where the cost of care is shared by a hospital and a doctor.  Minnesota is experimenting with this on the state level, creating lump sum “basket of care” payments to treat chronic conditions like asthma and diabetes.

But although this seems like a promising experiment, the old fears of capitation come back.  Does this inevitably lead to cherry-picking the patient who’s least likely to have a complication?  When the lump sum is gone but the patient still needs tests, treatment and care for another few weeks, does the doctor or hospital provide services at a loss or does the patient get rushed out of care?  How can we structure this experiment to make sure we’re not making the situation worse?

Next:  I’ll answer Bob’s question about doctors on salary, and whether Health IT might enable a real compensation based on outcomes.

(Photo credit:  Olivier B on Flickr.)