Monday, June 08, 2009

I see Jimi in the mirror!

A sticky wicket has been picked up.

A thorny problem has popped up out of the weeds.

Pandora's box has been breached.

A sh*tstorm will fly.

Got the idea? Then you know how I felt when I read a recent study in the May edition of the Journal of the American College of Surgeons. Entitled Trauma Surgeon Mortality Rates Correlate with Surgeon Time at Institution, this is a retrospective review of outcomes in trauma patients with a comparison of seasoned versus less experienced trauma surgeons. This is a very provocative paper, and comes from the University of Miami Miller School of Medicine in Miami. In essence, the authors took a close look at their own data to see if trauma surgeon experience played a role in how major trauma victims fared in their institution. From the abstract:

Using our prospectively collected database, we compared our results with mean mortality for high-volume American College of Surgeon–certified trauma centers reporting to the National Trauma Data Bank. Mortality rates for our 11 trauma surgeons were correlated with years of experience as faculty surgeons at our institution during a 2-year period.
That's a pretty honest look in the mirror. What did they see? Overall, their trauma center mortality rates were excellent, and were significantly better than the mean rates of the National Trauma Data Bank for patients with all levels of injury. However, despite such good numbers,
...there was a significant correlation between years of experience as a surgeon at our institution and improved outcomes for patients with an Injury Severity Score ≥ 35 (weighted linear regression, p < style="font-weight: bold;">It took, on average, 7.9 years of experience at our trauma center to reach benchmark mortality rates.
Wait a minute. That means.....yes! Experience makes a difference! Us old guys do have something to offer after all. Of course, common sense would tell us this any way, but it is nice to be "validated" every once in a while.

Oh. Wait a minute. This means......experience makes a difference. In other words, despite the best instruction in residencies and fellowships, it takes a while before even the most well trained trauma surgeon has enough accumulated knowledge and experience to reach the level of his more seasoned colleagues. And that means, well, I'm not so sure --- but I'm pretty sure that some folks would demand to only be cared for by the most experienced trauma surgeons.

One of my favorite expressions comes courtesy of my program director:
Good judgment comes from bad experience.
Bad experience comes from bad judgment.
I know that today, after 15 years in practice, that I have better judgment and better experience than I did after 1 or 2 years in practice. That has come from a whole bunch of nights on call, time spent with patients, time spent with colleagues, time spent reading......and just a whole lot of time period. The same can be said for any occupation, it's just that physicians are held under the microscope a bit more closely than most.

I suspect that if this type of study were applied across all aspects of medical care, similar results would be found. Surgery just tends to lend itself to more spectacular problems when there are errors in judgment compared to, say, dermatology. But there simply are not ever going to be enough fully experienced surgeons on call at every institution in the country every single night. I think we have to expect that there will be an ongoing learning curve for new surgeons, but we need to encourage newly minted surgeons to put themselves into positions that allow close interaction with older colleagues who can provide much needed help as well as mentoring.

However, with the average age of practicing general surgeons in this country being ~ 56, I'm not sure the mentors will stick around to pass along the wisdom they have gathered if plans for major upheavals in health care in this country actually come to pass. And that would be a huge loss in institutional experience.

We'll see.

Wednesday, June 03, 2009

Trite but True

I must admit that the idea of sifting through reams of data makes me slightly nauseated. And antsy. And irritated. Let's just say it ain't my thing. But, as they say, somebody has to do it, and I'm all for that. Because sometimes sifting yields a little golden nugget --- the trick is to figure out if it is real gold or only pyrite.

A lotta data has been generated by the American College of Surgeons' NSQIP program --- the National Surgical Quality Improvement Program. While I have quibbles with some aspects of NSQIP, particularly about patient risk stratification, it is a laudable attempt to gather enough clinical information to steer patient care in the right direction. With the data that has been rounded up to date, the data analyzers have been able to start identifying hospitals that are outliers in certain areas, basically those with higher than or lower than expected complication and mortality rates. Figuring out what makes those facilities tick in a positive or negative direction is the whole goal of the program, so that every facility gets information to improve patient care delivery.

At the most recent Academic Surgical Congress, NSQIP data analysis of Medicare patients undergoing colectomy from 2005-2006 was presented. A total of 12,688 patients in 123 hospitals undergoing colectomy were included (article in ACS Surgery News). The reviewers looked at not only specific complication rates and risk-adjusted mortality rates, but also at the mortality rates following those complications -- what they termed as "failure to rescue."

High-mortality hospitals were found to have a 1.5-fold greater risk of postsurgical complications --- that stands to reason. However, there was not a linear association with increasing rates of complications and increasing mortality ---- the higher mortality facilities had a rate of mortality associated with postsurgical complications that was more than twice that of low mortality facilities (26% versus 11%).

What, exactly, does that mean? It means that a certain percentage of patients are going to have complications, and that complications are more frequent in higher-mortality hospitals. But it also means that if a patient has a complication in a higher-mortality hospital, their likelihood of mortality is greater than if they had a complication at a lower-mortality hospital. An unwelcome double whammy, to say the least

Why might that be the case? Here is where the trite but true saying comes into play --- it takes teamwork to get patients successfully through a hospitalization. Avoidance of postoperative complications starts well before surgery, with appropriate preoperative evaluation and testing; this includes the assistance of other physicians (cardiologists, pulmonologists, etc.) and staff (following protocols for preop lab and EKGs, initiation of DVT prophylaxis, etc.). In the OR, having a team approach is critical to minimize the risks for excess blood loss, prolonged OR time, avoidance of temperature loss, etc., ad infinitum. Postoperative care is crucial, with nurses, physical and respiratory therapists, and physicians being attentive to mobilization, pulmonary toilet, glucose get the picture.

"Failure to rescue" then may occur with any person or department involved in a patient's care --- the nurse who doesn't recognize that a patient's low blood pressure may indicate bleeding; the respiratory therapist who thinks a patient with worsening respiratory function will do OK through the night; the physician who doesn't see a patient who is doing a bit more poorly than expected in a timely fashion; the blood bank that doesn't get needed products to the patient's bedside quickly enough; failure to implement protocols to deal with DVT prophylaxis, antibiotic prophylaxis, ventilator management, etc.

To draw on the current phraseology of hospital management-types, it boils down to culture. Hospitals with lower mortality rates, I suspect, aggressively engender a culture of high expectations, where everyone down the line understands how important of a role they play on the team. High expectations come with accountability, and the squishiness of some administrators when it comes to meting out that accountability can lead to poor outcomes. That accountability must also apply to the physicians, and a physician culture that demands the best for our patients in our hospitals --- from the physicians and staff alike --- pays big dividends.

Tuesday, June 02, 2009

CMS : Flying Against the Headwind of Reality

When good science, good medical care, and common sense sit athwart government bureaucrats, who wins? I think all of you know the answer, but it bears repeating.

One of the hidden dangers that lurk for patients -- particularly those who have undergone surgery or who have had trauma -- is the risk for developing a venous thromboembilism. I have written about this in the past, so I won't bore you with the details. Simply put, we try to aggressively treat patients with prophylactic measures to try to avoid the development of VTE, using medications (Heparin or Lovenox), early ambulation, and sequential compression devices. There are a few problems with this, however:

  • Some patients cannot be mobilized, due to injury, ventilator-dependence, etc.
  • Some patients cannot be given chemoprophylaxis, with injuries to the central nervous system, spleen, or liver which could bleed when they are given medications that interfere with clotting.
  • Some patients will develop VTE, regardless of whether or not they are treated with appropriate prophylaxis.
That's right, Kemo Sabe, some patients will develop a deep vein thrombosis or pulmonary embolism no matter what we do. While we may have prior knowledge of a hypercoaguable state in some patients, more often than not it becomes apparent only after the fact. Sometimes the hypercoaguable state is temporary, associated only with the episode of trauma, and sometimes it is genetically predetermined. But a really smart trauma surgeon at the University of Colorado has developed a test that appears to be able to detect patients who are at a significantly higher risk for VTE.

In the May issue of Surgery News (link is to a pdf file), Dr. Jeffry Kashuk describes the test, known as rapid thromboelastography (r-TEG), using a device manufactured by Haemoscope. For those of you who are interested in the chemical processes involved, read the article for the details that were presented at the Central Surgical Association's annual meeting (which I suspect will be published formally in the not to distant future). The bottom line? -->
  • 19% of the hypercoaguable patients experienced a thromboembolic event despite chemoprophylaxis, compared with none of the patients who had normal coaguability.
  • Evidence of a hypercoaguable state predicted thromboembolic events with a 100% sensitivity and 45% specificity in patients who received chemoprophylaxis.
Cool. If this pans out in larger studies, it will provide us with another tool to treat patients in a more tailored fashion. For example, we may be quicker to place a temporary vena cava filter in some patients, or give them greater than standard doses of Heparin until the r-TEG results normalize. Alternately, we may be able to avoid placement of some IVC filters in trauma patients who cannot be given chemoprophylaxis if their r-TEG tests do not demonstrate a hypercoaguable process. Once again, this has the potential to be a very useful tool if it pans out.

Whoa, Nellie. Stop right there. According to the Baghdad Bob the Centers for Medicare & Medicaid Services, venous thromboembolism should never happen! It is, in their parlance, a "never event." That's sort of like saying that flat tires, frozen pipes, or computer crashes should never happen. It flies in the face of reality, an intentional offense to those caring for patients in this country. I say intentional, because the goal is not improving care, but denial of payment. (More on "never events" can be found here and here.)

So, we know that some patients are at an increased risk for VTE, and some are going to get VTE even with currently appropriate prophylactic measures. This test may help us identify some of those patients, and start trials on treating them differently. CMS, ignoring the science and accumulated weight of decades of clinical evidence, by declaring this to be a "never event" has rendered this type of investigation moot, as they simply will not fund care for "never events."

Let this be a little introduction to government-run health care.