Paying 4 Performance

Steffie Woolhandler, Dan Ariely, and David Himmelstein have a real cool post on Paying 4 Performance (P4P) in healthcare, wherein docs are giving financial incentives to be better doctors. Medicare and private insurers are moving toward P4P pay structures for providers, but these behavioral economists argue that it’s an ineffective tool for improving care. “Quality breeches in medical care” are most often attributed to “fatigue; poorly designed workflow and care systems; undue commercial influence; knowledge gaps; memory lapses; reliance on inappropriate heuristics; poor interpersonal skills and insufficient teamwork, to name just a few. But ‘not trying’ is rarely cited. Yet P4P implicitly blames lack of motivation for poor quality care.” So that’s interesting. Yo, doc. You have too much to do, not enough help, you haven’t been trained on this procedure and you’ve been working for 16 hours but I’ll give you a thousand bucks to get this done right. jaskldfjakd nope.

There’s also some evidence that P4P makes people perform WORSE—because of the pressure: “Huge incentives offered to rural villagers in India—equivalent to about half of their annual money income—worsened performance on complex memory and puzzle-solving tasks. High stakes incentives may be distracting, interfering with cognitive focus and creativity.” MAKES SENSE 2 ME.

---
---
---
---
---
---

23 Comments / Post A Comment

deepomega (#22)

Yeah but there are KINDS of “not trying.” For instance, if you are overworked, you might be willing to hire staff to help if you knew that the overwork was keeping you from getting P4P bonuses. If it was costing you money, maybe you’d work on your interpersonal skills instead of just saying “fuck it, I’m the doctor.”

deepomega (#22)

@deepomega And I know I’ve brought this up before, but have you noticed how EVERYONE thinks their field is the ONE career that can’t be improved by performance-based pay? “Oh no, you can never possibly tell whether doctor/teacher/police officer/stevedor is doing better or worse than his coworkers!”

wondajules (#80)

@deepomega in addition, the current system does little to assure care of patients and much more to assure unnecessary testing. Procedure based billing simply encourages running MRIs “just to make sure” and then being able to bill for them. Regardless of whether or not it was necessary.
ALSO “Knowledge gaps” are IMPROVED with P4P because if the doctor has to perform on something, she is more likely to know about it. The P4P programs outline necessary procedures for both chronic care and preventative care and are updated yearly to account for latest standards of care.
That’s not to say they’re perfect, but they shouldn’t be demonized.

@wondajules Yeah I was going to say that paying for outcomes rather than procedures would make sense. Having a fixed payment for each set of treatment protocols, which you get for following them, sounds good for me (and also encourages docs to keep up with the latest protocols rather than just whatever they happen to remember from med school 20 years ago and the literature drug companies gave them).

Imagine if McDonalds charged people for each time they flipped a burger or turned the fryer on and off, rather than saying “OK you give me your $4.00 and I give you a completed cheeseburger.”

deepomega (#22)

@stuffisthings Hamburgers are a public good, man, McDonalds should just make them because of their social value.

anecdata (#2,683)

@stuffisthings Paying for outcomes is great in theory, but difficult in practice, since outcomes are often pretty distant from treatment (for example, statin treatment and coronary heart disease), and people move in and out of hospitals, regions, and practices. I think most propoponents of P4P agree that rewarding outcomes would probably be preferable.

The other problem, of course, is that rewarding outcomes can encourage gaming the system, i.e., doctors or practices selecting healthier patients.

I’m so embarrassed that I’ve been lurking around the Billfold for months, and it takes a health policy post to make me comment.

wondajules (#80)

@anecdata I know! Me too. But this is what I DOOOOO. Good ol healthcare policy. I hate that it’s so screwed up, but I do appreciate that it means GREAT job security for me…

But yes, in general no one thinks that P4P is perfect, but it does have great potential to lower costs by changing the culture of over billing. It’s a start, anyway.

deepomega (#22)

@wondajules Are there any alternatives being suggested, really? Only thing I’ve heard is using single payer shit to force it down, which scares me.

@anecdata Well I started writing “outcomes” because that is how we talk in the international development world, but then I shifted to “following the protocols” because I realized the outcome would be distant and hard to measure (and a lot of times even if you did the best treatment, you get the worst outcome — patient dies). Maybe a payment based on the diagnosis (and recommended treatment for that diagnosis) and a bigger reimbursement to the docs for preventative care activities. I dunno. Just saying that it’s possible to design such a system in a way that “gaming” is not a concern (for example, in the UK doctors get a bonus for successfully convincing their patients to quit smoking.) Unfortunately the US does not have a good track record of designing such systems so that alone might be reason enough to be skeptical.

pocket-witch (#1,576)

@deepomega “ALSO “Knowledge gaps” are IMPROVED with P4P because if the doctor has to perform on something, she is more likely to know about it.”

Wouldn’t you say that “what I do could save or kill this person” is enough of an incentive to know about it?

anecdata (#2,683)

@stuffisthings There’s a really great paper by Serumaga, I think in BMJ, that looks at the effect of P4P in UK primary care practices and essentially found no association, either in processes or outcomes. Lots of possible explanations beyond “P4P doesn’t work” (the paper discusses a couple of them), but I don’t think there’s much in the way of convincing research showing the P4P benefits anyone but the payors.

Something I don’t think Ariely mentioned was the impact of negative incentives in a P4P setup.

Also- and I don’t have a citation for this- I believe the kind of gaming we’re both talking about is just less problematic in the UK, since insurance isn’t tied to employment, and medicine just flat-out isn’t as profitable. What do you think?

@anecdata I actually did my Master’s degree in England with a fair bit of focus on public governance and New Public Management theories (though largely comparative, as in: should we use these in developing countries?) and it seems like the case of P4P in public service provision in general is pretty thin. I think probably for the same reasons as it wouldn’t work well in medicine or education: the outcomes are very hard to measure, and there is no clear link between the performance of any one provider and the eventual outcome.

It’s just like if you introduced performance bonuses for train drivers, it’s unlikely that trains would run more on time — there are just too many factors involved, most of which the driver can’t control.

wondajules (#80)

@pocket-witch You would think, and in many cases it is incentive enough. But there are a variety of factors that make it far from universal:
1)rapidly changing guidelines combined with over-worked and under-staffed primary care providers making it hard to keep up with what is best in what situation
2)overlapping conditions that can cause contra-indications for medications and treatments but can be overlooked due to above mentioned reasons (and also, could be unknown to the provider because patients tend to move around a lot. Especially the really sick ones).
3) A system that prioritizes volume and de-emphasizes holistic care.
4) Plain old arrogance

Many times the provider has the patient’s best interest at heart but our system tends to make it difficult to care for the whole patient and not for the condition.

And @deepomega, as far as alternatives are concerned, Health Information Exchanges, Patient Centered Medical Homes and Accountable Care Organizations are making recent headway into ensuring that a patient is cared for across disciplines. The emphasis there is on the QUALITY of care, rather than the volume, so it’s a step in the right direction.

Look, performance-based pay should only be used in situations where the performance is earning money, not healing the sick or educating children. It’s time we stopped trying to apply a corporate, capitalist model to things that shouldn’t be tied to either.

deepomega (#22)

@Jake Reinhardt Are there any other public goods that shouldn’t get performance based pay? Putting out fires, maybe? Feeding hungry people? Defending innocents in court?

@deepomega I wasn’t aware that my inclusion of two things meant my exclusion of EVERYTHING ELSE FOR THE PUBLIC GOOD. Sheesh.

deepomega (#22)

@Jake Reinhardt Uh, you said “only be used in situations where the performance is earning money.” Putting out fires doesn’t earn money. Being a public defender doesn’t earn money. Explain why those ones, which match your criteria, DO deserve merit-based pay while teaching and doctoring don’t.

@deepomega What the hell are you talking about? Those jobs are public service, non-money oriented jobs, therefore they shouldn’t get merit-based pay. I didn’t say ONLY HEALING THE SICK AND EDUCATING CHILDREN SHOULD BE EXCLUDED. Perhaps your reading is…not comprehensive. Merit-based pay: for bankers. Is that good enough for you?

deepomega (#22)

@Jake Reinhardt Relax on the caps lock, bro. And I guess that’s fine, as long as you think that nobody ever picks jobs based on pay – otherwise you’re setting up a situation where ambitious people all become bankers. I wonder what could go wrong with that?

deepomega (#22)

@Jake Reinhardt Mostly, though, I like the idea of you determining what is and is not a public good. Pretty great!

@deepomega I see what you’re saying but you’re implying that there are only two compensation schemes in the world, pay-for-performance and low pay. Most people’s pay is not explicitly linked to performance measures — for instance, to recycle my above McDonalds metaphor, McDonalds cooks are not paid per burger. What if doctors were just paid a high, but fixed, salary regardless of how many procedures they order?

pocket-witch (#1,576)

@stuffisthings This is how the Mayo Clinic does physician salaries.

@pocket-witch I believe this how GPs are paid in England as well. Since we’re never going to get single payer in the US, I wonder if insurance companies have tried something like this? Hire a doc to work directly for them, say “OK you get $200,000 a year from us no matter what, but you have to see and treat all the patients we send to you. We handle all the paperwork and administration.” If they also did the same thing at testing companies and hospitals it could drive their costs down considerably.

Comments are closed!