Dr. Daniel Stein: What the Data Says about Your Doctor

Harlan Krumholz: Welcome to Health & Veritas. I’m Harlan Krumholz.

Howard Forman: And I’m Howie Forman. We are physicians and professors at Yale University. We’re trying to get closer to the truth about health and healthcare. This week, we will be speaking with Dr. Daniel Stein, the CEO of Embold Health.

But first, we like to check in on current health news. Harlan, there’s a paper in JAMA Network Open on heart disease and alcohol, I believe. And I think you have thoughts about that. And I’m really curious to learn more about this.

Harlan Krumholz: Yeah, this is a really important paper, came out this week. And there are a couple things about it. First of all, some people may not know, JAMA Network Open is one of the family of JAMA journals. JAMA’s one of the top medical journals. JAMA Network Open is a journal that publishes articles and makes them open access to the public. Anyone can read them. And I’ve been hearing about this study and others who are thinking of doing this kind of study for a while, and when it came out, I was just so pleased.

It comes from a top-notch research group, a group in Boston at the Broad Institute, and they used the data from the UK Biobank. Now, this is the study in the UK in which more than 500,000 citizens in the UK decided, or consented, to be willingly giving of their data and their biospecimens, and many of them got imaged, and it’s just turned out to be a wonderful resource for the research community, and that data’s very high-quality. And it includes genomic data.

So what these folks did was, they did a study using a methodology that is essentially taking advantage of your genes. And knowing that people with a certain genetic profile, in this case, were more or less likely to drink alcohol. And that’s because some people inherit genes that make them less likely to tolerate or to take alcohol.

So they’re using this to say that who got which genes shouldn’t have anything to do with your risk of, say, heart disease. And so we can’t randomize people to, “You group of people drink heavily for 20 years, and you group of people drink moderately, and you group of people become teetotalers.” That becomes an impossible experiment to run, but when it turns out that, when we’re born, we have different predispositions to certain things like drinking alcohol, we can take advantage of that. And we say that that would enable us to look, based on the genetic profile, to be able to correlate that with various outcomes, and what they did show was that the genetic score was able to characterize people in that way.

And we’ve thought for a long time that, gosh, maybe this moderate drinking is really a healthy thing to do. People talk about that French paradox where, why is it that heart attack rates are lower in France? But this study shows, I think pretty clearly, that actually the more you drink, the higher your risk of lots of things. Hypertension, atrial fibrillation, and heart disease.

And then what they also found was that there was evidence that people who drink moderately—this is the group that we thought might have benefited from alcohol—also had a lot of other healthy lifestyle habits, and the implication here is that we’ve been fooled in this, because the people who drank socially a little bit tended to be people who had healthier habits. Of course, people who drank a lot tended to have a lot of other bad habits. And people who abstained, maybe there was a reason they abstained, and maybe they also had habits that weren’t quite as healthy.

So the sweet spot appeared to be in this moderate drinking. But this article would suggest that, no, it’s not the alcohol. It’s the lifestyle that tends to be associated with people who do drink that way. But any amount of alcohol seems to be causing a problem.

I’ll say one last thing. Because I know lots of my friends who really enjoy alcohol are going to be mourning this study. The increase in risk really came after about seven drinks a week. So people who enjoy about a drink a day, lots of people are like that. What I can say is it wasn’t causing much harm. But it’s not benefiting, it appears from this study, but it wasn’t really harming you, but as you get up more than a drink a day, that’s when the risk started to escalate, and it started to escalate rapidly.

So I find this fascinating. It’s a new methodology. It’s a robust design. Good researchers. Great data. I kind of think it might be a game-changer. I don’t know. I’m interested to see what the scientific community says about it, but so far I’m pretty enthusiastic about it.

Howard Forman: Does it answer the question that also exists out there, where some people say red wine is healthy because of tannins or other chemicals that are in red wine? Can it help us answer that question, or do we still not know the answer to that?

Harlan Krumholz: Well, I think that question came up because people were assuming that moderate amount of alcohol consumption was protective. And then they went to the next question, to say, “Well, what’s in it that might be protective?” If this is saying that really, there’s not any evidence of protection in general, it sort of weakens that argument.

Now, you’re raising really good point though, because it could be that when you look at all alcohol consumption, it’s not protective at all, but it could be that there are certain types of alcohol that could contain certain substances. This study can’t really get to that.

So yeah, you’re right. Saving grace for those people who want to make the argument that maybe there’s a substance in the drink I like that’s different from the overall average effect that we’re seeing with alcohol consumption in general.

I think what this definitely could put away is the idea that the alcohol itself, moderate alcohol consumption, is helpful. Whether or not there might be something you’re drinking that has alcohol in it, that also has other components that might be helpful.... I guess with that suggestion you’re going to keep the research and hope alive for those who want to continue to seek whether there’s some fountain of youth within a little bit of alcohol, just whatever your favorite is.

Howard Forman: It is fascinating to me though, that this question has remained unanswered, effectively, for decades. We’re not talking about 10 years, 20 years. We have asked and not had this question firmly answered for at least 40 years now, and it was great to see this. And I did not understand the methodology. You explained it very well. It makes it much clearer to me. But it’s nice to see people advancing the science.

Harlan Krumholz: It’ll need to be validated. Of course, this is UK population. Another question you could ask was, well, is that true for every population? Does it vary? So it would be, are there certain types of alcohol that are different than the general effect, and are there certain groups of people who might have a different experience than this group?

So there’s still lots to learn, but this is very interesting. And I think it’s going to cause people to reflect on any public health guidance that would suggest that a little bit of alcohol might be a good thing. I think people should probably... If people can enjoy alcohol, that’s fine, but don’t think that you’re taking it for your health.

Howard Forman: Good point. I’m excited to introduce Dr. Daniel Stein. Dr. Daniel Stein is the founder and CEO of Embold Health, a healthcare analytics company that is setting a new gold standard for the way we measure healthcare quality. Using real clinical data around the country, Stein’s company identifies the nation’s top physicians and healthcare systems that have consistently delivered high-quality care.

Throughout his career, Dr. Stein has sought to improve healthcare from across all angles, including work as a primary care physician, health policy staff for the U.S. Senate Committee on Finance, chief medical officer for Walmart’s care clinics, and now an entrepreneur for a growing health analytics company.

I first met Daniel when he was a rising sophomore at Yale and have been excited to watch his career and be a friend observing and advising him over these years and now learning from him. He eventually did his medical school at Hopkins, his residency at the Brigham [Brigham and Women's Hospital in Boston], and a simultaneous MBA at Harvard.

So Daniel, if we just start off for our listeners who may have no experience with this in the past, can you start off by telling them what types of data you actually have and how you might use it to improve care?

Daniel Stein: Absolutely, Howie. But before I jump into that, you missed the most important job I had in my career, which was your research assistant.

Howard Forman: Teaching assistant. Yes.

Daniel Stein: Twenty-plus years ago. The data that we have is health plan data. So we get big multi-health-plan datasets that are longitudinal, so the full set of medical records and pharmacy records for, really, hundreds of millions of patients across the country. And what we do with that data is we analyze it to understand how individual physicians perform. So we follow the members on a de-identified basis, longitudinally, to see if they’re getting care that’s based on the latest evidence, based on the latest standards. And then we take that, and we synthesize that information to look at the patterns in how the docs are practicing, we bring that data together to understand what those patterns mean, and then we make the information available to employees, consumers, individual people, to help them understand who’s going to be a high-performing doctor when they have a care need.

Harlan Krumholz: So it’s really great to have you here, Daniel. And wow, to have been a research assistant with Howie, I can’t even imagine how great that must have been. I mean, I don’t even know where you would be today, if that hadn’t occurred.

Daniel Stein: I can assure you, I would not be where I am right now.

Harlan Krumholz: If only I had had that opportunity, I don’t know what I could have done. So you went into this to measure docs. A lot of people will be wondering, how do you measure docs? I mean, how do you do that? Because the data you’re getting is largely claims. It’s billing data. And what doctors do with patients is highly complex and individualized. So how do you get down to the point where you can make determinations about the performance of individual doctors, based on that data? Could you give us a little insight about that?

Daniel Stein: Yeah. Let me give you an example, Harlan, because that might help make it a little more real. So we’re not trying to measure an individual physician-doctor interaction to say this was the right care or the wrong care for any individual patient. What we’re doing is trying to look at the patterns of care that the doctors are delivering, to see how that compares to their other peers in the market as well as the latest science.

So take for example knee scopes. So arthroscopy is one of the most common orthopedic procedure that’s done, in this country at least. We’ve known for, gosh, 19 years now that patients with degenerative joint disease—so, arthritis—generally don’t benefit from knee scopes. Through randomized controlled trials done—starting at Baylor, they’ve been replicated in Europe—there’ve been a number of trials that have shown that, for that population of patients, this is a procedure that generally doesn’t offer much, if any, benefit. And so what we do, Harlan, is we go, and using that medical claims data, we identify all the patients with newly diagnosed degenerative joint disease. So not patients who’ve had it for years, but based on their medical bills, who’s got a new onset of it? And then we look for the different orthopedic surgeons that they see. Are they getting a knee scope or not?

Harlan Krumholz: But if it’s not an effective treatment, why don’t the insurers just stop paying for it? I mean, why [not] look at the doctors and say, “This person’s doing a lot of it and this one’s not doing a lot of it”? I mean, why not just go directly and say that this tends to not be a useful... and they have, of course, all of these processes in place, what people call pre-authorization—

Daniel Stein: Pre-auth.

Harlan Krumholz: ... where doctors have to convince payers that this is someone who needs something. So why don’t they just do that straight away, as opposed to trying to rank docs on whether they’re doing something that they don’t think has value?

Daniel Stein: I mean, I think that’s a multi-billion-dollar question, Harlan, that I don’t know the answer to. So I don’t know why, but what I can tell you, when we look at the data... and these are fully paid-for claims. So they’re adjudicated. It’s gotten through whatever processes, prior authorization, care management, you name it.

In a typical market, when we look at how often that procedure is being done, in the first year of patients with newly diagnosed degenerative joint disease, you see an average rate between 10 and 15% of patients who are getting a knee scope. And if you break it down then to the individual doctor, you see some orthopedic surgeons in that market are doing it 1, 2, 3% of the time. So, infrequent. And some were orthopedic surgeons in that same market, with that same payer mix, so presumably the same programs, prior auth and others, are doing it 25 to 30% of the time within the first year of a newly diagnosed patient. So it’s a big question why this stuff slips through and why it isn’t caught.

Harlan Krumholz: But are you also looking at, for the people who do it, who’s the best person to do that procedure? Are you able to do that kind of thing too?

Daniel Stein: Yeah. So, I mean, this builds on a lot of the work that you all have done at Yale, on thinking about not just appropriateness of care—you know, is that care warranted and based on the latest evidence?—but also outcomes. Effectiveness of care.

So one of the things we look at, with orthopedic surgeons in particular, is risk-adjusted complication rates, risk-adjusted readmission rates, your likelihood of getting an infection, your likelihood of having to have a redo within that first year, et cetera. Because at the end of the day, when I think about who I want to be referring patients to, it’s to a doc that’s going to do care when it’s necessary but not do a bunch of unnecessary stuff when it’s not. And when they do it, they’re going to get good outcomes and they’re going to get good results.

And in the commercial world—it’s a little different, obviously, in Medicare—but in the commercial world the other thing that we look at is cost. But for us, we’re not looking at cost for just the unit price differences. We’re looking at total cost of care are caring for a similar population of patients, with the idea being, some physicians at high-cost medical centers, including the academics, who are really efficient and are only doing things when they’re necessary and getting good outcomes, they actually look great from a total cost of care, even though the unit prices are more expensive.

Harlan Krumholz: And just one more quick thing, and then I’ll yield to Howie. I know I’m taking up all the airtime. Can anyone see your ratings, or is it only for the customers of the certain plans? That sounds very interesting, and it would be great to take a look at. But I mean, I know it’s a business, but I was just wondering, how do people get a chance to see the rankings?

Daniel Stein: So the primary customer that we have for the business, for Embold, is we work with large self-insured employers for their employees, to help them get their employees to top-performing docs who are going to deliver that high-quality care. But part of our commitment at Embold is that we’ll be fully transparent with the measures, methodology, and the results.

So for example, doctors in any of these markets where we’ve evaluated them, and we’re working with employers, we’ve got physician scorecard reports that we make available to them. And it’s not just how they rank, and it’s not just the measures and the methodology, but we try to take it a step further.

So if I just go back to that orthopedic arthroscopy example, we’ll show them the primary literature. If there’s clinical practice guidelines that apply, we’ll show them the clinical practice guidelines and try to give them some tips, some things that they can do. Because it’s my belief—and this is as a physician who’s spent my life in this world—doctors got into this field because they want to do the right thing, and they want to help people. And so when we see a lot of this variation, even an inappropriate, unwarranted variation, it’s not generally because you got bad actors. It’s just practice patterns have existed however they were going to be, for however long. And so I’m a big believer in the power of feedback, but to do that, you have to be open and you have to be transparent.

Howard Forman: It’s been well over 50 years since the Dartmouth Atlas group started looking at the variations that you’re talking about, at least some of them, in a gross way. And it’s been about, I don’t know, probably five years since Berkshire Hathaway, Amazon, and JPMorgan Chase recruited Atul Gawande to lead an effort to do, in a huge way, what you’re describing here.

I have asked the question for at least 26 years: Why aren’t employers doing more to improve care delivery and hold down cost? And you’re stepping into that space right now. Exactly the space that I’ve been asking about. Why is this so hard?

Daniel Stein: Yeah. It’s funny, Howie. I like to tell people we’re not doing anything new. And yet what we’re doing feels very new. In many ways we’re on the exact same approach and methodology of that Dartmouth work from 50 years ago. It’s just now the data is more available. We can apply it to a bigger sample.

I think the answer for employers, Howie, is they haven’t had access to tools before. Historically—Harlan, to your question—they very much use the tools that they get from the carriers, the data, the results, the networks. And so what we’re trying to do is give them a new way, a new path forward. And it’s not to say, “You have to do this instead of what you’re getting from the carrier,” but you’ve got access to a different set of levers now that historically you just haven’t had access to, which our hope is, empowers them to be able to be a more active purchaser.

Howard Forman: And you talked about doing this for employers and particularly self-insured employers. But there are a host of innovators out there. Many of them you know. Places like Aledade, Cityblock, CareMore, Oak Street Health. Is that another avenue that you might be able to use your data and help inform them? They’re using their own data as well, but is there a way to work with them, to also help influence the practice?

Daniel Stein: Yeah. Howie, those are groups that we’re spending a lot of time with now, as partners and as an additional place where we’re going to take the analytics and take the business over time.

It’s funny, that revelation that you described around physician variation, it feels like that world, the risk-bearing primary care, whether it’s government or commercial or both, it feels like there’s been a sea change and an eye-opening the last year, where all of a sudden people have realized so much of the spend happens through specialists.

And yes, you can deliver really exceptional primary care, and doing everything you can within the four walls, and care coordination, and care management and pop health, and a lot of the spend is actually being driven by what specialists you go and see. And so for us, as we think about, how do we want to have impact and drive higher quality and lower cost, it’s helping empower those risk-bearing primary care groups, and those documents, to help get their patients to those high-performing specialists. So that’s absolutely where we’re going to next.

Harlan Krumholz: So I wanted to dig in just a second, about the thing I opened up before about the life of a startup CEO. And I just wonder if you could just give the listeners a little sense of what your days are like and the things you’re juggling. Just because I don’t think many people really get that much of a glimpse into that. And I can only imagine the day-to-day. You’ve got to wear so many hats and do so many things in order to get this thing going. You’ve been so successful so far, but you still have a ways to go. How is it for you so far?

Daniel Stein: Each day is a new adventure, Harlan. Seriously, I wake up in the morning and I like to write out my schedule, either the night before or early in the morning, and write out what I want to do, what I’ve got scheduled, and then invariably, you come into the office or get a phone call, and there’s some new fire, some new crisis du jour, some new problem that we didn’t even anticipate.

A couple Sundays ago I was doing some work, and I guess I had one of those “woe is me” moments that you sometimes get, where you feel bad for yourself. And my nine-year-old daughter came in and looked at me, and I said, “I can’t. I’ve got to do work right now.” And I think she must have seen a little bit of a grimace on my face. And she said, “Well, Dad, you’re the one who decided to do this. No one told you had to be the boss.”

Harlan Krumholz: They’re so smart, you know?

Howard Forman: Without giving away corporate strategy, and maybe big plans, but I’m just wondering, you must in your mind right now think about what the company ideally could look like in five years and maybe what its impact on the country could be in five years. Can you give us a blurb in Wikipedia about you in five years, or about the company in five years, that might tell us what you hope for?

Daniel Stein: Yeah. It’s fun to get to do that, Howie, Harlan. Spend so much in the day-to-day, you name it, issues. But to be able to take a step back and think about, “Well, why am I doing this? What do we want to accomplish?” And for me, it’s raising the bar and raising the tide. I just think there’s such an immense opportunity in this country to improve quality by surfacing some of this variation and empowering people to do it, and that’s something that is what motivates me. And so for me, Howie, if you look back five years from now, when I say we’re successful, it’s because a lot of people, both employees and members and doctors, are using our data to improve quality and reduce unnecessary spend. That, to me, is what I want us to be in five years.

Harlan Krumholz: So I know we’re coming to end of the time, and we really appreciate all this time you’ve spent with us, especially given everything you’re juggling. But I do have one other question, which is a little outside of this company, Embold Health, but based on your prior experience. So you were very involved with leading Walmart’s efforts around their clinics. And I wonder if you could just share your thoughts about what you think the future of primary care is, and the role of these companies. Because it’s not just Walmart, of course. A lot of companies are now cropping up to disrupt the market in primary care and providing services. What’s your thought of what this is going to look like in maybe five, 10 years?

Daniel Stein: I’m excited, Harlan, and I’m excited because primary care, I feel like, is finally getting the attention that it deserves. So I’m a PCP.

Harlan Krumholz: Primary care provider, for people who are listening.

Daniel Stein: Thank you, Harlan, for helping me not medical-speak here. It’s my belief that primary care is, in many ways, the heart and soul of medicine. And in this country we’ve traditionally underinvested in it. There aren’t enough primary care providers, and they haven’t gotten enough attention. And a lot of them are being asked to just do more without the resources to help them do it.

So whether it’s Walmart or a CVS coming in, trying to be the front door of health, or whether it’s some of the companies, Howie, you mentioned, the Oak Streets, the Aledades, who are saying, “We’re going to deliver comprehensive primary care and we’re going to take risk on it, around both quality and total costs,” to me, there’s this huge energy, and renaissance, and capital, that’s flowing into primary care, which I think is awesome.

I think it’s great for patients. And back to my goal of helping raise the quality bar and take out unnecessary spend, having really high-quality primary care docs that are armed with smart information and that have aligned financial incentives to help get their employees to the best community specialists—that, to me, is an exciting spot for the industry, moving forward. And which of these models “win”? The retail, or the virtual, or the full at-risk bricks and mortars? I don’t know, but my belief is there’s enough space for a lot of these folks to be successful in this new world.

Howard Forman: One last question before we wrap it up. You have in common with a lot of the people we talked about earlier, in that you have government experience. You spent time working on and oversaw the passage of the Medicare Modernization Act of 2003. You worked in the Senate during that time. How has that informed your ability to do this type of work, the fact that you actually have legislative experience?

Daniel Stein: I mean, Howie, I think healthcare is so complex. And one of the challenges that I’ve seen with some of these companies that have come in—maybe not the ones you mentioned, Harlan, some of the tech companies—is, “There’s an ethos, so we’re just going to disrupt things. We’re going to come in, it’s broken, we’re going to change it, we’re going to blow it up, we’re going to do things differently.”

And my belief is, this industry is too complex to intertwine. There’s so many incumbents that I just don’t think that that’s very realistic, to take that approach. And so for me, my experience on the Senate Finance Committee was this incredible grounding experience, to realize just how many people are affected by policy. Individuals, stakeholders, companies, advocates. And to realize that if you’re going to come up with solutions that work, that actually have the chance to move this thing forward, you got to be open to hearing their perspectives and making sure that there’s a space for them at the table. And I think, at the Senate, as you know from your experience there, that’s just a requirement to get anything done. And so early on in my career, it kind of got beat into my ethos that this is a complex ecosystem, and there are a lot of good-intentioned people. But if you’re going to come up with solutions that help move this forward, you got to take a big-tent approach rather than try to go it alone.

Harlan Krumholz: Yeah. And a lot of vested interest too. I mean, I think that’s one thing, which is, if you want to move left, someone’s pushing you right. I mean, there’s lots of folks that they’re doing well in the current system. So maybe that experience also gave you an appreciation for all the different stakeholders and how they’re trying to influence the future.

Daniel Stein: And people are very good at protecting the status quo.

Harlan Krumholz: That’s right. Yeah.

Howard Forman: Well, look, we appreciate that you are not protecting the status quo. You are a disruptor. And we wish you the best of luck. We wish the company the best of luck. We want to see a better healthcare system. And it’s people like you who are fighting the good fight, that hopefully will get us there. So thanks for joining us on Health & Veritas today.

Harlan Krumholz: Yeah, no, really appreciate it. I saw you studied ethics while you were at Yale too. So, I mean, I think that bringing that kind of ethical North Star to what you’re doing, and if you’re successful, then patients should be successful. So I really like that attitude that you’re taking and the approach. So good luck.

Daniel Stein: Thanks, Harlan, and thanks, Howie. It’s great to be here with you today.

Howard Forman: So that was great, hearing from Daniel. It is amazing for me like that this is over 20 years since I first met him, and he is doing great things.

This is the segment where we talk about what’s either keeping us up or what’s inspiring us. And I will be absolutely a hundred percent honest that I watched the Academy Awards live on Sunday night, and it literally kept me up at night. I could not fall asleep after it. And I was asking myself why I was so disturbed at it. And just for our listeners, who I imagine all have heard some version of this, but during the latter half of the Academy Awards, Chris Rock was about to present an award, and he made a joke at the expense of Jada Pinkett Smith, who is Will Smith’s wife and also an acclaimed actress, who at the present time has a shaved head. I don’t know whether he knew or didn’t know whether she had alopecia, but she in fact has alopecia. And so it’s been a troubling issue for her and for Will Smith. It’s been bothering them.

And he made a crude joke. We’ve certainly heard worse at the Academy Awards and at other shows, but it was at her expense. And Will Smith became enraged, went up to the stage, slapped, very hard, Chris Rock, went back to his seat and shouted expletives, but very angrily. And I think for me at least, what made it so discomforting, aside from the fact that we’re not used to seeing something like that live, is both of these are great actors.

What Chris Rock did, arguably, he does all the time. He makes jokes. They happen at the expense of individuals. And Will Smith, similarly, is absolutely not known as someone who would physically attack somebody or would curse out somebody. And it was not obvious to me who was right or who was wrong at that moment. I think I have a little more clarity now about my opinion about it. There is no way to excuse what Will Smith did, but I do understand the emotion of feeling that you are protecting your spouse who’s going through a difficult time. And I also felt bad for Chris Rock in the moment, in the sense that he’s done this a million times, and people expect it out of him. And I think he was sort of surprised by it.

They both have since apologized in different ways, and the Academy, I think, is going to issue some type of statement in the next day or so about how they’re going to rule. But it still discomforts me. There’s no obvious answer. You don’t want to get to a point where comics are having to completely filter their jokes to the point where they don’t offend anybody.

On the other hand, you also do want to have some decorum at an event show. And we’ve had some years where the decorum is worse. So I found that conflicting. I don’t know what your thoughts were. You and I were both watching it live. You and I texted at that time. At first, I think both of us thought it might have even been staged, but it clearly wasn’t.

Harlan Krumholz: Yeah. And I’ll just say, my last thought about this is that we’re all human. And I think it reflects on the human condition, which is that anyone can be triggered. What I would’ve liked for him to say was, “I cannot believe I just did that.” And to have reflected on the depth of what just occurred. But that wasn’t quite the message. And then out that night, partying as if nothing had happened.

It was a major thing that happened. Again: we’re all human. People get triggered. But that’s also, with the ubiquity of guns, why there’s so much gun violence and gun deaths. I mean, as humans, this kind of thing can happen. So anyway, it’s something to reflect on it. It was on my mind too this weekend. And thanks for bringing it up.

You’ve been listening to Health & Veritas with Harlan Krumholz and Howie Forman.

Howard Forman: So how did we do? To give us your feedback or to keep the conversation going, you can find this on Twitter.

Harlan Krumholz: I’m @hmkyale. That’s hmkyale.

Howard Forman: And I’m @thehowie. That’s @ T-H-E-H-O-W-I-E.

Harlan Krumholz: Health & Veritas is produced with the Yale School of Management. Thanks to our researcher, Sherrie Wang, and to our producer, Miranda Shafer. Talk to you soon, Howie.

Howard Forman: Thanks very much, Harlan. Talk to you soon.