Some venous thromboembolic events can’t be prevented even with optimal care

I have written several posts about how I get things right before others see the light, but none better than one from three years ago pointing out that some of the Centers for Medicare and Medicaid Services (CMS) "never events" can't really be completely prevented and therefore should not be considered "never events."

One specific "never event" I questioned was hospital acquired venous thromboembolic (VTE) disease which encompasses deep venous thrombosis (DVT) and/or pulmonary embolism (PE). I wrote "I am unaware of any DVT study in which no patients in the experimental arm developed DVTs or PEs. Patients will develop DVT or PE even with the best evidence-based care."

Along comes a brief research letter published last month in JAMA Surgery by a group from Johns Hopkins led by surgeon Elliott R. Haut.

Of 92 patients in their institution who had VTEs in a single year, 43 (47%) had received defect-free care. That is, each of those patients received all doses of risk-appropriate pharmacological prophylaxis ordered for the entire hospitalization.

To put it another way, VTEs for those 43 patients were not preventable. There would be no way to do a quality improvement project for a group of patients who received the right prophylaxis throughout their hospital stays and still got VTEs.

The Joint Commission/CMS criterion states that a hospital is in compliance with VTE prophylaxis if a patient receives one dose of an appropriate drug within 24 hours of admission. The Hopkins study showed that of the 49 patients (53%) whose care was suboptimal, 36 (73%) missed at least one dose of prophylaxis that was correctly ordered. Other studies have shown that missing even one dose of prophylaxis at any time during a hospitalization increases the risk of VTE.

So about half of VTEs are not preventable even with perfect adherence to the prophylaxis protocol, and the standard for compliance established by the JC/CMS is inadequate to judge the quality of an institution's performance for VTE prevention.

The study shows that 1) a lot of good information can be delivered in a two-page paper, 2) JC/CMS criteria for compliance with VTE prophylaxis need to be revisited, and 3) VTE should be removed from the list of "never events.”

Defensive medicine is more of a problem than you think

You may have missed this when it first appeared.

Experts from Harvard and the University of Southern California say assumptions made by some analysts that defensive medicine is not an important facet of the high cost of healthcare may be wrong.

Those assumptions were based on data showing that malpractice reforms instituted in some states did little to reduce healthcare spending. 

According to the report from the National Center for Policy Analysis about an article in the wall Street Journal, defensive medicine ("ordering some tests or consultations simply to avoid the appearance of malpractice") is just as common in states with low as it is in those with high malpractice risk. In fact, about 2/3 of doctors in both the low and high risk states admitted to practicing defensive medicine. 

My experience is that the 2/3 figure is probably a very low estimate. Just about every physician I know has  ordered a test or consult strictly to "cover his/her ass" if something were to go wrong. I am certain it happens tens of thousands of times per day in the US.

Why the No-Pay Policy for In-Hospital Infections Failed

I told you so.

Three months ago, I blogged about the Medicare (CMS) “never events” list, diagnoses that Medicare will no longer reimburse hospitals for. In Medicare’s eyes, these diagnoses are totally preventable, should never happen and will not be reimbursed. I pointed out that several were in fact not 100% preventable despite any institution’s best efforts, and the rates of many of these occurrences would not fall to zero.

Now the esteemed New England Journal of Medicine has published a paper which confirms what I wrote back in July. Its 13 authors compared rates of central line-associated bloodstream infections (CLABSI) and catheter-associated urinary tract infections (CAUTI), two of the diagnoses on the “never events” list, with ventilator-associated pneumonia, a disease not on the list, as a control.

After reviewing data from 398 hospitals from before and after the establishment of the new Medicare rules, they found that quarterly rates of all three infections did not change and concluded that the “never events” policy was ineffective. The senior author of the study then tweeted “Our paper in NEJM - CMS non-payment policy didn't change infection rates. Do we need much stronger penalties?”

My answer to that question is “No.”

Penalizing hospitals did not work because we may have reached the lowest possible rates of infection already. Some infections will occur no matter what steps are taken. We are dealing with human patients and human care-givers. Perfection is not likely to happen.

Many people erroneously believe that all CLABSIs can be prevented with the implementation of strict sterile precautions when catheters are inserted. That has lowered infection rates but not to zero. Why not? In addition to the technique of insertion, CLABSIs can result from other factors. Solutions may become tainted. The integrity of the IV line itself may be violated during the administration of medications through the line. The dressing covering the line may loosen and allow bacteria to enter the puncture site. Patients may be immunosuppressed and unable to overcome even the slightest hint of contamination. Or maybe it’s just bad luck.

CAUTIs are also not totally preventable. Despite a major push to remove urinary catheters as soon as possible, some patients need them for days to weeks for many reasons. For example, there are patients who simply cannot urinate on their own due to old age, dementia, coma, paralysis, etc. Critically ill patients with marginal urine outputs need urinary catheters for monitoring. Patients who are incontinent of stool may contaminate their catheters despite the best nursing care.

No, much stronger penalties will not work.

How about if we simply decide what is an acceptable rate for these infections and aim for that?

The "Never Events" list should be reconsidered

You are probably familiar with the CMS “never events” initiative. CMS has decided it will not reimburse hospitals for treatment related to complications which it says should never occur. Here is the current list.

• Foreign object retained after surgery
• Air embolism
• Blood incompatibility
• Stage III and IV pressure ulcers
• Falls and trauma
• Manifestations of poor glycemic control
• Catheter-associated urinary tract infection
• Vascular catheter-associated infection
• Surgical site infection after coronary artery bypass graft, bariatric surgery for obesity and certain orthopedic procedures
• Deep vein thrombosis/pulmonary embolism after certain orthopedic procedures

According to American Medical News, two more complications have been proposed as additions. One is acquired conditions stemming from cardiac implantable electronic device surgeries and the other is iatrogenic pneumothorax associated with venous catheterization.

I have no problem with some of the items on the list. Foreign bodies like sponges or instruments should not be left in patients after surgery. Air embolism and blood incompatibility should be 100% preventable.

But I do not see how catheter-associated UTIs or vascular catheter associated infections can be completely prevented. Some sick patients with depressed immune systems are going to get infections.

I believe it is impossible to completely prevent wound infections in all clean cases. As has been shown in studies of SCIP compliance, some patients get wound infections after colon surgery despite the timely use of the right antibiotic.

DVT/PE cannot be prevented in every orthopedic procedure. I am unaware of any DVT study in which no patients in the experimental arm developed DVTs or PEs. Patients will develop DVT or PE even with the best evidence-based care.

With very few exceptions, every large published paper on central line insertions, even those using ultrasound guidance, reports some instances of post-procedure pneumothorax. There is no way it can be completely avoided. For example, this study of 937 ultrasound-guided central line insertions reported 2 (0.2%) post-procedure pneumothoraces. That’s a published study by radiologists. The real world incidence of pneumothorax is much higher, often quoted at 2-5%.

To me, these rulings are simply a way for CMS (and other payers who are sure to follow suit) to avoid paying. Where is the input from “organized medicine”? Was any evidence-based research looked at by those who decided all this?

Why are we standing around and allowing this to go unchallenged?

This post appeared on Sermo yesterday and most people who commented agreed with me.