One-on-One With Memorial Healthcare CEO Patty Page LaPenn, Part II
California-based Memorial Healthcare is a four-IPA organization that includes a 35,000-visit-a-year urgent care clinic and a disease management center. The four IPAs represent 220 primary care doctors and 400 specialists – across North Orange County, Anaheim and Long Beach – centered around four hospitals. Primarily handling at-risk contracting with health plans, the IPA has begun working to bring EHRs to its providers. Recently, HCI Editor-in-Chief Anthony Guerra talked with CEO Patty Page LaPenn to learn just want IPAs want, and don’t want, from hospitals.
GUERRA: Whose responsibility is it for patients to get better? If a doctor is misled by a patient about the medications they are taking, so that the physician’s actions result in harm, who is at fault?
LaPENN: First of all, let’s wake up America, we’ve got no PCPs. While everybody is spending all the time focusing on other aspects of healthcare, two to three years from now there’s going to be no doctors to see. I think what they’re thinking is that it should be proportional. So the thing is, if a doctor sees 100 diabetics and everybody has about 7.5 percent who are noncompliant or 30 percent who are noncompliant, whatever it is, you can then find the outliers. So then they say that the doctors who have more noncompliant patients are bad doctors. I think that’s what they’re saying, but correlation does not imply causation. In my own doctor panel that just means the physician is a compassionate doctor. The doctors that have the worst rates in my panel are softies, and the noncompliant patients know other noncompliant patients, and if they find a doctor that’s willing to work with them, that doesn’t berate them or get short with them, then that doctor will build a panel of those noncompliant patients.
First of all, there’s not random assignment of patients to doctors. Doctors attract certain patients. I have doctors who are athletes. If they have a patient that has a single blood glucose of 105, which is way early, they refer them right to our diabetes education program to scare them and get them to the point where they don’t become diabetics. That’s because that particular doctor, as an athlete, is sensitive to getting patients quickly back in line.
When you get into an older population, you can risk adjust, but the biggest danger of judging doctors based on their noncompliant patients is doctors will shun noncompliant patients.
GUERRA: So they’ll arrange their practice to do well in P4P programs by not taking the difficult patients.
LaPENN: Yes. Doctors already have the right to dis-enroll a patient from their practice based on noncompliance. And so, with this scenario, if they’ve got a patient, his A1C is elevated, well, I have to tell you that the wrong kind of doctors are going to have the best scores. The doctors that will do stuff like that for money will have good scores. I have a lot of doctors, they’re family practice doctors, they wouldn’t be in family practice if they were after money, and so the best of them will be penalized and demoralized.
If you randomly assign patients to doctors, then I think this method works, but applying this research-based methodology to the practice of medicine when you don’t have random assignment doesn’t work. I have a public health degree, so I came out of public health research, and it’s obvious to me that people are taking metrics from the business or research world and applying them to the practice of medicine. If you really want to have an impact on patient health, tax them or charge them more for their insurance when they are noncompliant.
GUERRA: If you want to change behavior, hit them in the pocketbook.
LaPENN: Yes. Instead, by hitting the doctor with this, there will be some unintended consequences. We’re working through the PQRI right now on diabetes, and it’s so ridiculous creating all those transactions. They’re taking a very expensive, long way around it if the desire is to improve healthcare.
Let’s talk about predictive modeling. It says when you have a three-day length of stay for a chronic condition, 64 percent of all those people will be dead in a year, or dead in 18 months. So there’s a number of those items like that, and the thing is that if they want all this data because they’re looking for that, then the question is why. How are you going to determine if you’re in the 64 percent or in the 36 percent who’s going to live? The data that they’re asking for on all the quality improvement makes me wonder what they are going to do with it.
They’re going to hold doctors accountable for the outcomes of their patients when the patient has no skin in the game. There’s lots of research that says when you start doing an intervention with a patient, if you ask the question, “How likely are you to change anything you’re doing,” the patients that are not going to change anything they’re doing are going to tell you they’re not going to change anything they’re doing, and the fact is they’re not going to change what they’re doing.
GUERRA: {laughing} So they’re not even going to lie to you.
LaPENN: No, and it’s people who make their living sitting in cubicles on computers, who probably belong to a gym already, they’re the people who show up most frequently when we do an open call for cholesterol and heart disease screening and all that stuff, we don’t get the sick people. We get the worried and well.
GUERRA: You mentioned the PQRI program. Would you say the administrative burdens of that program are too large and the financial incentive too small?
LaPENN: The carrot is not enough to change behavior. One of my practices has 10 physicians and four nurse practitioners, and they’re only going to get $3,600 a year because they’re a family practice. The thing is, most of this work is done by primary care doctors who make the least amount of money in Medicare. So all of the money that’s being spent is on the specialty side. If there was PQRI in orthopedics or neurosurgery or ophthalmology, the financial rewards would be significant, but for a primary care doctor who’s getting $100 a visit, you’re talking about $2 a visit.
If you order a test but you don’t know the result, you have to send in, “I ordered this test,” but you have to put an AT modifier in there which says you’re not going to do it for an undisclosed reason. Why are you ever sending something that says something like that? And you’re supposed to document that you have reviewed the results and you have to document the day that you document it and send that transaction. Who does this help?
And you have to change the way doctors do things. So right now, when our diabetics go to see their doctor, the doctor talks to them, looks at their last lab test, does an exam, gives them the referral to go see an ophthalmologist, and then when the ophthalmologist report comes back, they sign the report and they date it. If there’s anything wrong, they say, “patient notified,” “referred to” and then, “referred to ophthalmologist.” And then if they are on an electronic record, they scan that in or if they’re on a paper record, they put it in the chart. Well, now they have to create a transaction out of that. So you have to change the whole way that doctors do things in order to transactionalize it, which means you’ve got doctors spending their energies thinking about transactions rather than caring for their patients, or they have to invest in a lot of expensive technology that costs more than the incentives they’re going to get.
GUERRA: I want to shift gears and talk about the integration between the ambulatory and acute environments. You’re CEO of this large IPA. It represents 220 primary physicians, 400 specialists, and many, many practices, none more than 10 doctors. Talk about the integration among those practices or between those practice and acute care facilities.
LaPENN: We scan. We have a patient record and we capture important events, not every event because when a patient is admitted to the hospital and the hospital turns on their auto fax system (if they have one), we get the H&P and discharge summary. We get those two pieces of information and scan them in so that when a doctor needs them they can easily get to them.
A patient who’s in the hospital for four days could easily have 50 pieces of paper that come through, and we don’t need all of it.
We get that for all the patients. My doctors are one-third HMO and two-thirds fee for service, and we manage all of the at-risk stuff, and we’re owned by those doctors. We are a doctor-owned organization. We’re like a little co-op. We only do patients that are paneled to us.
GUERRA: So you’re getting some faxes from the hospitals at this point and scanning those into the EMR?
LaPENN: We used to until they got an electronic medical record. Now we have to go in and print them one at a time.
GUERRA: Until the hospitals got an EMR?
LaPENN: Yes.
GUERRA: So when the hospitals went on to an EMR, you got less integrated?
LaPENN: Yes.
GUERRA: Explain that to me.
LaPENN: We get a census – I’m not going to say who this hospital is – but we get a census so we know which of our patients are in the hospital, and we have a hospitalist group. And so we know who’s there, and then we go in and find the information we need and print it out, then we scan it into our system.
GUERRA: So you can go into the hospital EMR, but there’s no way to download that into your system? You have to print it out, scan it in, and upload it.
LaPENN: For the hospitals to push out to every medical group that could be on all kinds of systems, it’s very expensive, and they’ve got their own problems.
GUERRA: So this is where we get into the interoperability difficulties.
LaPENN: Right, right. Can I say something? Give us a national patient identifier. All of this stuff is being developed and we still do not have the patient identifier we can use that transcends everything else. We can’t use social security. So we’re trying to do all this interoperability, and we don’t even have a patient identifier. What application in any other business, in real estate, in banking, can you imagine not having an account number that you need for customers? I don’t know why there’s all this delay. I mean, I do know why, but it’s ridiculous. Give us a 13-digit patient identifier that stays with the patient. They now have it for doctors but we need it for patients, and that will make interoperability much easier.
GUERRA: You’ve got all these practices that probably need integration to each of the major inpatient systems. It gets to be a mess.
LaPENN: Yes. And of course there’s all kinds of companies that want to do that for you at an additional cost, but there really is not enough profit in primary care to pay for all those additional costs.
GUERRA: For all those interfaces?
LaPENN: Yes. The average primary care doctor is making about $150,000 a year. They’re grossing about $300,000 a year. Their overhead, if they’re lucky, is 50 percent, giving them an extra $3,600 a year. I mean it isn’t even that much per-doctor on the Medicare program, but even if it were $3,600 a year, that’s no real incentive when they have to buy $6,000 worth of technology.
