Sunday, April 20, 2014

I, Robot has nothing on the truth

Here's the IMDb summary of the 2004 movie I, Robot:

In the year 2035 a techno-phobic cop investigates a crime that may have been perpetrated by a robot, which leads to a larger threat to humanity.

But we don't have to wait until 2035.  We've seen a different kind of "crime" perpetuated by a robot during the last decade or so.

If we could start all over again, would we want the medical community to spend billions of dollars in capital and operating costs on a medical device that had no proven relative efficacy compared to conventional, lower cost techniques?  I think most of us would say, "No."

But the medical community has been complicit in carrying out exactly this scenario with regard to the daVinci surgical robot.  The federal government has likewise stood by and watched, as have insurers.  At best, the establishment media has let the story go on for years, and, at worst, has been quick to jump on the bandwagon.  All to benefit the managers and shareholders of a private company.

What kind of money are we talking about?  How about 2,585 installations at a purchase price of about $1.5 million, or $3.9 billion?  And a required annual maintenance contract of about $150,000, for $387 million per year.  And disposable supplies used per surgery of about $2000, with about 100,000 cases per year, for a cost of $200 million.  (Or calculating another way, Intuitive Surgical, Inc. reported revenue of $2.3 billion in 2013, $2.2 billion in 2012, $1.8 billion in 2011.  Net income for those three years, by the way, totaled $1.8 billion.)

This has been a breathtaking saga.  We could hope that it would present lessons for the future, but there is no indication that it has done so, whether for the expanded use of this technology to other areas, or the introduction of other devices and procedures that comprise the medical arms race.

The most recent bit of evidence of the degree to which the country has been taken in is summarized in this article from Medpage Today:

Robot-assisted radical prostatectomy (RARP) led to complication rates, readmission rates, and rates of additional cancer therapy similar to those of conventional surgical prostatectomy, a review of almost 6,000 cases showed.

Patients who underwent RARP had significantly higher complications rates at 30 and 90 days, but blood loss and transfusion rates were lower, as was the risk of a prolonged hospital stay. After adjustment, the overall complication rates did not differ.

"RARP and open radical prostatectomy have comparable rates of complications and additional cancer therapies, even in the post-dissemination era," Quoc-Dien Trinh, MD, of Dana-Farber Cancer Institute in Boston, and co-authors concluded. "Although RARP was associated with lower risk of blood transfusions and a slightly shorter length of stay, these benefits do not translate to a decrease in expenditures."

But how about those supposed advantages with regard to urinary function and sexual function, i.e., quality of life issues?  Here was the result in a 2013 study in Urology:

Introduction of RALP (robot-assisted radical prostatectomy) did not result in improvement of functional outcome. There was no difference regarding urologic function/bother score or sexual function/bother score at 36-month follow-up in patients treated with LRP or RALP.

But this is not a new story. Look at this article from Urology in 2008:

Radical prostatectomy is an effective form of therapy for patients with clinically significant prostate cancer; however, outcomes are highly sensitive to variations in surgical technique. Because of the risks of perioperative complications and urinary and sexual dysfunction, which appear to be as great with robotic-assisted prostatectomy as with any other technique, patients with low-risk cancer, especially those >60 years, might be attracted to more conservative alternatives, including active surveillance, radiotherapy, and focal ablation.

Another article in Urology in 2011 talked about the learning curve associated with this "intuitive" system.  It suggests that a man might want to ask just how many procedures his surgeon has, er, under his belt:

Despite the intuitive nature of the daVinci system a definite learning curve with RALP (robotic-assisted laparoscopic prostatectomy) exists in relation to console time and LOS. Over 50 cases need to be completed to effect an appreciable improvement in learning curve. Positive surgical margin rates did not improve, suggesting that this may have a longer learning curve.

Had people been alert to this 2007 article in Urology, they could have seen that, in addition to the patient perspective, there is an institutional one, too, associated with the cost of learning curve for this equipment:

The literature search involved 8 series, with a range of learning curves from 13 to 200 cases. The least expensive learning curve was $49,613 and the most expensive learning curve was $554,694. The average learning curve was 77 cases and cost $217,034.


Costs associated with operative time while learning RAP (robotic-assisted prostatectomy) are substantial, and should be considered when deciding whether to implement RAP at an individual institution.

A 2008 article presented the business case as follows:

Purchase of a robot reduces income by at least $415,000 per year, due to the cost of the device and the service contract. Donation of a robot lessens the financial impact by $300,000 per year. If an institution maintains an identical caseload when switching from LRP to RAP, then it cannot maintain equivalent profits.. This holds true even if the robot is donated.

To maintain profits, an increase in caseload is needed to cover the added costs of the robot. At all levels of baseline productivity, purchase of a robot requires a greater case volume to maintain profits, relative to donation of a robot. Centers that perform a high volume of LRP at baseline need to only make small increases in total case volume to maintain profit, while centers with low volumes at baseline need to make very large changes in operative volume to cover the additional costs of robotics.

Like an airline making money with an airplane, the robot will only pay for itself if it is used at a high volume.

Imagine what kind of financial pressure that puts on doctors and hospitals to recommend prostate surgery rather than more conservative treatment approaches. How many men have had surgery who did not need to because of those pressures? How many of those have suffered irreparable harm?


Anonymous said...

Paul, since you have made no secret of the fact that, as CEO of BIDMC you yourself reluctantly OK'ed the purchase of a robot (and blogged about the decision at the time), I think it would be highly instructive to others now in that position to hear the inside story of how those pressures come about and how they might be resisted in the current environment. Somewhere this juggernaut has to be brought to a halt. It would be nice if it came from within the healthcare profession itself rather than by external fiat.

nonlocal MD

Paul Levy said...

Here's that post in which I discussed our hospital's decision to buy this piece of equipment:

Because this company relies on direct to consumer marketing, plus signing up doctors who are willing to extoll the virtue of the technology, the juggernaut only comes to a halt if the profession itself stands up to the misleading PR. The head of the American College of Obstetrics and Gynecology has done that, and arguably that has slowed down the market spread of this product in that field.

Chris Mitchell said...

From Facebook:

As someone who worked to market and sell DaVinci and all the other "products" for years at a hospital, I can assure you your analysis and conclusions are spot on. Whether it was special procedure for back surgeries like x-Stop, DaVinci, prostate screening, heart scans, bi-plane X-Ray machines, the latest Neuro interventional procedure, or total knee and hip replacement, we sold it. We spent millions to bring in patients to fill beds and pay for equipment. We were given regular reports of margins by service line and then told where the priority was for the next campaign. Our direct mail and online campaigns were laser focused to get the right folks in the new doors.

I look back now and see it as a complete charade. It is not about health care, but about revenue. The public thinks that because you are a not-for-profit that it is all goodness and altruism. Nope. This profit still motivates the execs and raises their salaries. We build medical monuments with private rooms and beautiful furniture and wood paneling and atriums and roof gardens and LEEDS Certified Silver, all just blocks away from people who have no health coverage, no food and no money for rent and utilities. Needless to say, I no longer sell this vision of 'healthcare.'"

Thomas Pane said...

Convoluted healthcare economics in healthcare can allow a marginal technology to persist longer than it would otherwise. But even in the cash-only (cosmetic) world, there can be unproven technologies that persist for quite a while.

If the robot can allow certain types of cases to be done with superior outcomes, it may bear fruit. But those probably be liver/pancreas resections which are done in far smaller numbers than hysterectomies/prostatectomies.

Anonymous said...

"plus signing up doctors who are willing to extoll the virtue of the technology, the juggernaut only comes to a halt if the profession itself stands up to the misleading PR."

Signing up doctors for cash: This standard approach is how the HIT vendors have sold their poorly usable devices and coaxed Congress to pass HITECH.