Public report cards for PICU performance?

September 13, 2010  |  General  |  No Comments

The short answer is that there aren’t any. I’ve written before about how hard it is for parents to know if the PICU there child is in is a good one or not. All they can really do is see if the components of a good PICU are in place: is the facility state-of-the-art, is it staffed by board-certified pediatric intensivists and pediatric surgical specialists, and do the intensivists have the assistance of the full array of subspecialists if your child needs them. But what parents really want to know is how the results of the PICU, the patient outcomes, stack up against similar units. Even we intensivists have little such information, and what we have isn’t available to parents. The biggest problem standing in the way of even collecting such information is the hugely varied mix of children who need the PICU, so in comparing PICUs we would need to be sure that we aren’t comparing apples to oranges.

One aspect of PICU practice that is fairly standard from place to place is heart surgery; the surgery is similar from child to child, and all children who have heart surgery need the PICU afterwards. The skill of the surgeon, of course, is the biggest variable in how the children do, but the competence of the PICU team is also a key factor.

I’m not aware of any data that compares outcomes among PICUs following heart surgery (or any other condition), but now we do have such information for adult heart surgery. Here is a very good recent editorial from The New England Journal of Medicine discussing what the implications are of such information. One important potential negative effect would be cherry-picking — doctors or hospitals avoiding caring for very sick patients out of fear of worsening their report card. From the article:

“There is no question about the need for accountability on the part of health care providers or the central role of measurement in the improvement of health care. Nonetheless, questions remain about the role of public reporting in improving health care. Performance measurements audited by regulators are one alternative, especially in situations in which the information is too complex for patients to use in discriminating among care sites. Insofar as public reporting drives improvement of all outcomes, it benefits everyone; insofar as risk aversion leads to changes in the population receiving an indicated service, the net effect can be nil or even negative.”

I think that this sort of reporting is something all of us in critical care, PICUs included, will ultimately be seeing.

What’s the electronic medical record (EMR) really for, anyway?

September 7, 2010  |  General  |  No Comments

This is a cross-post of an item I wrote for Maggie Mahar’s excellent heathcare blog, Healthbeat

The electronic medical record, the EMR, is upon us. For those of us who learned medicine entirely with paper charts, some have enthusiastically embraced the EMR and some have refused, to the extent they can, to deal with it at all. But most of us have plowed ahead into learning how to use it as best we can. It seems to me that the degree of enthusiasm physicians show for the EMR relates less to the particular version of it we have chosen (or, more commonly, was chosen for us) than it does to the kind of medicine we practice. The old paper records worked reasonably well for all of us; in contrast, the several versions of the EMR I’ve used work very well for some kinds of doctors, but less well for other kinds. I think a good part of this disparity is that the basic purpose of the medical record has changed over the past half-century or so, and some of these new roles can conflict with the old ones.

The oldest repository of continuous patient medical records is at the Mayo Clinic in Rochester, Minnesota. In a very real sense the modern medical record was invented there, in the first decade of the twentieth century, by Dr. Henry Plummer. As the first multi-specialty clinic, it made more sense for each patient to have a single record that traveled from doctor to doctor with the patient, rather than having each doctor keep his or her own record for each patient. The idea was to have a single packet of paper that contained everything that had happened to the patient. From that it’s only a short step to the notion that the record should travel with patients wherever they go, even if it is to physicians not associated with each other. This is a key promise of the EMR.

Medical records began as the possession of the doctor. This paradigm is changing. Very soon, although medical facilities will have copies, the records will essentially belong to the patient, with doctors only using them from time to time as need requires. Of course this could, in theory, happen with paper records, but it would be cumbersome. One of the things that first attracted me to pediatrics was the sheer size of the pile of paper that the medical records clerk would plop in front of the hapless medical student admitting an octogenarian to the hospital; in contrast, a toddler’s chart fits neatly in a small packet. The EMR allows these massive piles to be reduced to disks or microchips. It also allows the record to be organized into searchable form, so important things don’t get missed because they are buried in the disorganized mess of sequential folders.

Those are a couple of the brave promises of the EMR, but we all know we are a long way from realizing them. One huge barrier is that, as of yet, there is no standard platform for the EMR. Like many physicians, I’ve had to learn several because different facilities choose different vendors. In our pluralistic medical system (if one can indeed call it a system), it’s a free-for-all. And each of them has its own maddening quirks.

I think there is a broader problem here: over the ensuing years from Dr. Plummer’s era the medical record has taken on roles unheard of back then. For one thing, now the record is a legal document, a buttress against anyone who accuses us later of bad care. This process began long before the EMR, or course, which is one reason the charts I had to grapple with as a medical student ballooned so much. As a graduate student in history of medicine I had the chance to review many of the Mayo charts from earlier, simpler times. I recall one chart, from the nineteen-forties, describing the course of a very critically ill child. Overnight the child’s condition had markedly deteriorated; it was easy to see this from the recorded blood pressures and heart rates. The physician’s note for the following day analyzed these developments with only four words: “mustard plaster didn’t work.” Now the EMR offers the possibility of recording all we do easily and without getting writer’s cramp.

The medical record has also become something else it wasn’t back then: it is also now a commercial document, proof of what we did and why, used by payers to check up on us to make sure we should be paid for what we bill. Today’s payers want to know what the doctor did and why. They want to know, quite precisely, why that mustard plaster didn’t work and all that we did to make it work.

I think some of the problems with the promise of the EMR are that these legal and commercial roles can clash with the original purpose of the chart, which is taking care of the patient. The computer whizzes who design the software don’t always seem to me to have quite the same goals as we doctors who use it. The old paper charts were easy to adapt to new things, new procedures. All we needed was a different sheet to add to them and stuff in the folder. Upgrades and tweaks to the EMR are much more formidable things.

In spite of all these things I find the EMR to be a powerful addition to my practice. In fact, I think I’m a better doctor for using it. I think a key reason for that is because of what I practice – critical care medicine. In the ICU we love to measure and count things. We want minute-to-minute monitoring of variables, which in the old days resulted in huge paper flow sheets covered with dots and numbers. Rummaging through them to identify key moments in a patient’s care was often difficult. In the ICU, each patient gets a large number of tests each day, results which used to get stuck on clipboards with all the other paper. Important things got missed. Now I can sit at a computer screen and find it all with a mouse click, and the EMR makes it very hard not to notice anything important.

In contrast, I have friends who hate the EMR. It causes them hours of pain in training time, pain for which they aren’t compensated, and is slower for them to use than paper records were. In their minds, it gives them little or no advantage over paper in caring for their patients. I’ve noticed that they practice specialties that are less concerned with number-crunching than mine. They also tend to be office-based, rather then hospital-based, and don’t have to deal with as many other physicians as I do each day in the ICU. Thus many of their notes are written for themselves, not for other members of a large clinical team. Yet now they are asked to conform to how others want their charts to be.

I don’t know how all of this will work out. The EMR is here to stay. On balance, I think this will ultimately be good for doctors and their patients. But we don’t really know yet just what it is and what it should look like. I worry it will end up like one of those military boondoggles – it gets loaded with so many bells and whistles because it is supposed to serve so many purposes that it ends up being an expensive monstrosity that doesn’t perform any of its missions well.

Still, I’m an optimist. I prefer to be excited by the possibilities, rather than discouraged by the obstacles. I think the EMR will be good for patients, and will make us better physicians. For a while though, things will continue to be more than a little messy.

Under-insurance among American children

August 30, 2010  |  General  |  No Comments

There’s been a lot of attention, appropriate attention, focused on children without health insurance. These children tend to be in families with incomes too high to qualify for Medicaid, but too low to allow purchase of decent insurance. Medicaid covers around a quarter of America’s children.

There’s another insurance problem facing America’s children — under-insurance. For families who can’t get insurance through an employer, buying health insurance on the private market is expensive — prohibitively so for many families. So they buy a bare-bones policy that can easily turn out to be a total waste of money because it doesn’t give the coverage needed when a child actually gets sick. That’s called under-insurance. How common is it among America’s children? A recent study in the New England Journal of Medicine gives us some idea of the answer.

The study takes the form of a survey done in 2007. It found that 19% of all children had inadequate health insurance. This was far more children than had no insurance at all during the year or who spent part of the year without insurance. What this means in practice is that a major illness in a child, such we often see in the PICU, can bankrupt a family even if they have some sort of coverage. Not surprisingly, “under-insured children were significantly more likely to have delayed or forgone care, and to have difficulty in obtaining needed specialist care.”

It’s too soon to see if the new healthcare bill will have any impact on this grim statistic. There are lots of interesting statistics in the article, analyzing many subgroups of children by income level and geographic location. If you’re interested at all in health policy, it’s well worth a look.

Reminders: another simple, low-tech way to reduce hospital infections

August 20, 2010  |  General  |  2 Comments

There’s a best-selling book out about how simple checklists can prevent complications of medical treatments. It’s simple, the lowest of low tech, requires no expensive equipment, and it works. It seems a bit sad that we need research to confirm common sense. Now a new study describes how yet another very simple thing can reduced complications — simple, automated reminders.

Medical care in the PICU is quite complicated at times and the pace is often hectic. When we are going from patient to patient we focus on the major stuff — the mechanical ventilator, for example. It’s easy to overlook smaller items; smaller items which, in the longer term, can become much bigger things. Urinary catheters are an example of this.

We use these tubes to drain urine out of the bladder of a patient who is unable to urinate themselves. They’re useful and handy. But urinary catheters do carry some risk of infection, especially if they are left in for a long time.

A recent study showed that simply asking the doctor — prompting them with a small reminder each day — if the catheter was still needed led to a 50% drop in catheter-associated infections. It’s simple, easy, and cheap. It’s something I now have on my daily checklist for each patient.

Calling nurses and doctors by their first names: why the one but not the other?

August 15, 2010  |  General  |  1 Comment

I’ve worked in hospitals since I was 16 years old — 42 years ago now. I was first an orderly, then a nurse’s aid, then a practical nurse, and a finally a surgical technician before I became a physician.

When I started, female nurses wore caps, the details of which identified which nursing school they had graduated from, as well as a pin that gave the same information. They wore starched, white dresses, white shoes, and white hose. They were never called by their first name except by those who knew them personally. Male nurses were few and far between. We had none in the medium-sized community hospital where I worked.

Of course things have changed a lot from those times. Female nurses no longer have to wear those awful hats and uncomfortable starched dresses. Having everybody in scrubs does improve comfort, although it can make it hard to tell the PICU nurse from the housekeeping person cleaning the PICU.

The most significant change to me is that nurses now expect both patients and doctors to address them by their first names. In fact, they have to: my name badge has my full name on it, but the PICU nurses only have their first names and the first initial of their last name. I’m told that at hospitals which still have a nurse’s last name on the badge, the nurses themselves put tape over it to obscure it.

I’m told the reason for this change is personal safety and security. Nurses have close, intimate contact with patients and families, and they fear stalkers. Yet I’ve also been told by security people that, if somebody really wanted to find out the last name of a particular nurse, it wouldn’t be that difficult. I’d love to see some actual data about this issue.

Medicine has long been rigidly hierarchical. Nurses, whose relationship to physicians for many decades was more or less a master-servant one, have struggled for recognition and respect. The fact that physicians were once overwhelmingly men and nearly all nurses were women compounded this effect. (In pediatrics, at least, this profile is changing — now over half of pediatricians in training are women.)

I have wondered now and then about calling nurses by their first names, but continuing to call physicians “doctor.” Is that fair? Somehow it seems to me it should be all one or all the other — both sides using the first name or neither.

This question crops up from time to time on nursing discussion boards, such as here, and always seems to lead to pro and con debates. Nursing leaders also ponder the implications of this new familiarity, such as here. I’m curious what anybody else thinks about it.

Comparative effectiveness research in medicine: we need it, but how to get it?

August 7, 2010  |  General  |  No Comments

Every patient wants the best care — what is known to work. Certainly nobody wants care that doesn’t work, especially if what doesn’t work carries some risk of its own. But what if we don’t know what treatment is best for a particular condition? Shouldn’t we find out?

It is well known that there are wide variations in medical practice across the country, even for the same conditions. Even in the same region, doctors choose different therapies for the same condition. This is often because the circumstances of patients differ. But what if doctors choose different therapies for no particular reason except personal preference? And what if those therapies, although producing the same outcomes, differ in cost by a large amount?

Enter comparative effective research (CER). The idea is simple: compare two treatments and see which works better. And, if one works only a little better and the cost difference is huge, is the tiny improvement worth the cost. It seems odd to nonphysicians, but this kind research is hard to do, and is in fact rarely done. We often compare some new, experimental treatment with the standard; but it is often hard to compare two standard treatments with each other to see which works better. Why?

A recent editorial in the New England Journal of Medicine examines why CER can be challenging to do. The reason often boils down to money. In the example given in the article, a very expensive drug was being compared with a much cheaper one. Both therapies were covered under the patients’ insurance (in this case Medicare). But one group of patients would have an enormously high out-of-pocket copay fee because the copay is often calculated as a percent of the total bill. How do we deal with that? The guiding principle of a randomized, controlled trial is that the patient groups are similar. It would affect the trial if one group of patients could afford (or was willing to pay) a higher copay than the other.

Another key principle of such trials is that neither the patient nor the evaluating physician knows which group the patient is in — which drug they are receiving. This is called blinding. For experimental drugs blinding is easy; the patient gets an unidentified drug, marked only with a code that will be broken later. But if the drugs being compared are both covered by insurance, the insurance statement shows to the patient what it is.

These problems are solvable. But it is important to realize that comparative effectiveness research, which we very much need if we are to control our exploding medical costs, will not be easy to do in many situations.

Patient safety: when doctors are the problem

July 31, 2010  |  General  |  No Comments

We’ve always know that hospitals can be dangerous places for patients. In a landmark study some years ago, the Institute of Medicine, a part of the National Academy of Sciences, demonstrated just how dangerous they can be; anywhere from 50,000 to 100,000 people die annually from preventable errors. How are we doing at reducing that grim statistic? The answer is that we are making some progress, but there remain serious roadblocks.

The deaths studied by the Institute of Medicine came from a whole host of causes, and many of these causes are complex and difficult to address. But it turns out that one cause — serious infections from central venous catheters — can be easily improved. We can’t prevent all of these infections, but we can dramatically reduce them. The way to do this is absurdly simple and the lowest of low-tech: use a checklist that ensures basic procedural steps are followed in the correct order. Hospital safety guru Peter Pronovost demonstrated this some years ago. Checklists for all sorts of procedures are useful. Well-known medical author and surgeon Atul Gawande had even written a best-selling book about them. So what’s the problem? The answer is that the problem is often doctors and our medical culture. A recent editorial by Dr. Pronovost helps explain why. (The editorial is from the Journal of the American Medical Association, which requires a subscription. If anybody wants a copy, let me know.) Here’s the crux of the problem, as described by Dr. Pronovost:

“Although most physicians and hospital leaders genuinely want to prevent harming patients, and many physicians practice good teamwork, this view of not questioning physicians is pervasive. Physicians are often rushed, sleep deprived, and overworked and are offered limited training about teamwork and conflict resolution. The practice setting is not always conducive to completing recommended practice and anything that takes extra time for one patient (eg, searching for supplies) detracts from the care of others. Physicians also may not receive feedback on individual performance or hospital infection rates. Social, cultural, educational, and financial differences between physicians and nurses also may inhibit some nurses from speaking up, even when physicians may welcome such feedback.

Moreover, many physicians have not accepted that fallibilities are part of the human condition. Thus, when a nurse questions them, it causes embarrassment or shame. Clinicians are sometimes arrogant, believing they have all the answers, dismissing team input, responding aggressively when questioned. The line between autonomy and arrogance is fine and nuanced. Society has benefited tremendously from physician autonomy and innovation, producing new drugs, devices, therapies, operations, and anesthetics. Therefore, autonomy and innovation must be continued. However, autonomy becomes arrogance when actions are mindless and not mindful, when something is done simply because a physician demands it, when a clinician does not learn from mistakes, and when experimentation occurs without a clear rationale or testable hypothesis. Too often autonomy is mindless and driven by arrogance. When placing a catheter, reliability not autonomy is needed.

As Pogo said many years ago: “We have met the problem, and he is us.”

What inflammation looks like up close and personal, part I

July 29, 2010  |  General  |  No Comments

Here’s a snippet from the first chapter of my new book, How Your Child Heals. It picks up at the point where you, the reader, have begun a microscopic voyage to see what an infected splinter looks like from the perspective of inside your child’s body.

Now that you have had the full-sized, outside view of what happened to your son’s finger, it is time for you to go inside to places where the ancient physicians could not go. It is time to take a seat in the audience of the microscopic drama. You are about to make the first of several trips you will make throughout this book in a tiny, imaginary, high-tech vessel. Think of it as a cross between a submarine and an all-terrain vehicle; it can swim in the blood stream or leave the circulation to crawl around between the cells of the body. It is well-equipped with spotlights and spacious windows, allowing you to see what is happening all around you. The dramatic setting for your first foray is the time just before you called the doctor’s office to ask what to do about it.

The blood vessels in the body form an immense, self-contained system that is divided into two halves. We need oxygen to live, and one half of the circulation, the arteries, carries oxygen-rich blood out to all the parts of the body, down to the tiniest places. The other half, the veins, brings oxygen-depleted and carbon dioxide-laden blood back to the lungs to get more oxygen, which we breathe in, and dump the carbon dioxide waste, which we breathe out. The two halves of the circulation join in a microscopic meshwork of vessels called the capillaries. This is where the true business of circulation happens, where oxygen and other important nutrients get delivered to the body’s cells.

The capillary bed of your son’s throbbing finger is the key place to visit as you investigate what is causing all the problems, but to get there you must first get inside his circulation. You need a location where the tiniest of blood vessels are accessible, close to the surface. The lining of the eye is such a place.

Imagine you begin by poising your craft at the base of one of his lower eyelashes. You look over the edge into the wet, shiny world below. Your son momentarily pulls down his lower lid, revealing the pink inner lining of his lower lid, called the conjunctiva. You seize your chance, zip over the edge, and find yourself motoring about in the clear liquid of his tears, nature’s way of keeping our eyeballs moist. Here there are blood vessels close at hand, just below the surface. You slide your craft into the nearest one and then drift along with the stream, ever faster, as it takes you toward the heart.

You do not stay in his heart long, though, because nothing does. The blood rockets out of the heart like a fire hose because the heart pumps an enormous amount of blood very quickly. A typical adult heart, for example, sends out about a gallon and a half of blood every minute, proportionately less in a child. The effect on your vessel is the equivalent of taking a trip over Niagara Falls. You get bounced around, but soon find yourself in the aorta, the large vessel exiting from his heart.

The aorta is wide and fast, but it soon divides, then subdivides, into multiple rounds of ever smaller vessels. As this branching happens, the velocity of the stream in each of them slows down dramatically. Within seconds after leaving his heart you are scooting down one of these tributaries, headed for his painful index finger.

Things were moving so fast in the aorta and the first couple of branchings that you could not see any details in the surrounding walls of the blood vessels. Although you are going slower now, your pace is still a brisk one, and the flow still pulses along–now faster, now slower–in rhythm with your son’s heartbeat. Soon the stream slows down enough for you to get some idea of just what kind of pipe you are traveling through. The first thing you see when you shine a light at the walls is that the surface is covered by a bumpy layer of cobblestone-appearing cells. The junctions between these cells make a completely watertight barrier; no blood can leave this sealed pipe, and thus you cannot see what is going on in the tissues outside of it.

You soon find you are slowing down even further as you come closer to the sore on his finger, and you notice a dramatic change in the walls of the blood-filled passage you are passing down. For one thing, the wall of the tube is now translucent; you can shine your light right through and get a hazy view of what lies beyond. There are now some small gaps between the pavement of flat cells that makes up the walls, but the cells still mostly touch one another along their edges.

You have reached the capillaries. In real life there would be millions upon millions of options for you to have chosen on your trip from the aorta as the tubes branched into ever smaller pathways, but for our purposes we assume your miniature craft has the proper instruments to sense the correct path among the myriad of choices to lead you to the sore spot on your son’s finger.

Why we doctors order far too many scans

July 23, 2010  |  General  |  6 Comments

There’s a provocative editorial in a recent New England Journal of Medicine about the explosive rise in high-tech medical imaging. Everyone knows doctors order a lot of CT scans, MRI scans, and ultrasound studies, and that the number of these has been steadily increasing. And the cost is enormous. From the article: ” . . . these costs were the fastest-growing physician-directed expenditures in the Medicare program, far outstripping general medical inflation.”

To be fair, rising use of new medical technology is expected because, well, it’s new. What is unclear is that how much of this increased use has led to improved health to justify the cost. Clearly much of it doesn’t, and unnecessary scans, particularly CT scans, lead to risk with no benefit.

The practice of “defensive medicine,” of doctors ordering tests out of a fear of being sued for missing rare conditions, is often given as a cause for overuse of scans. There is some truth to that: the article cites a Massachusetts study showing that 28% of scans are done for that reason. Lawsuits over failing to diagnose things are common; lawsuits about overuse of tests are vanishingly rare.

Physician conflict-of-interest also plays a part. Through a loophole in Medicare regulations, physicians are allowed to refer patients for scans from which the physician benefits financially. That is wrong and needs to be fixed.

But there are deeper reasons. The root cause may well be “the style and content of clinical education and their impact on medical practice.” In other words, how doctors are trained. We use scans unthinkingly, and, unthinkingly, can cause harm. Again from the editorial: “The greatest risk that patients face with unnecessary imaging is the needless exposure to downstream testing and inappropriate treatment related to misdiagnosis and the overdiagnosis of common but unimportant findings.” I’ve seen that happen more than a few times.

How many calories does your child need to grow?

July 16, 2010  |  General  |  No Comments

In the PICU we focus a lot on nutrition because critically ill children need good nutrition for them to heal from their illness or injury. We often struggle to provide those needed calories. We usually can manage it one way or another, either with high potency oral feedings, special intravenous feedings known as total parenteral nutrition (TPN), or some combination of both of these. We have formulas we use to calculate what a child’s nutritional needs are.

But what about normal children? Many mothers, and it seems most grandmothers, don’t think their children are eating enough to grow. Aside from charting progress in height and weight, how can we calculate if a normal child is getting enough calories?

Here are some good rules of thumb. It’s simplified, but it works. The first thing you must do is determine your child’s weight in kilograms, since that is how we do the calculations: 1 kilogram = 2.2 pounds. Once you know that, you can calculate the average amount of calories needed to grow this way:

  • Age newborn to 3 months:  100 calories per kilogram per day
  • Age 3 months to 3 years : 90-100 calories per kilogram per day
  • Age 3 years to 8 years: 80-90 calories per kilogram per day
  • Age 8 years to 12 years: 60-80 calories per kilogram per day
  • Age 12 years to 16 years: 45-60 calories per kilogram per day

These calculations assume a normally active child. A significantly more active child needs more. A reasonable rule of thumb for this extra need is for an active child (defined as an hour per day of sustained physical activity) to have about 1.25 times their baseline calories and very active children (defined as more than 90 minutes of sustained physical activity per day) to have 1.5 times their baseline calories.

What does this work out to be in actual numbers? A normal 6-year-old in first grade weighs about 20 kilograms (44 pounds). For him, the daily need is 1700 calories. How about a 17-year-old girl who is a serious soccer player? The average weight for a girl of that age is 55 kilograms (120 pounds), and she needs about 3700 calories; her baseline need is 2400 calories or so, multiplied by 1.5 for her intense activity.

You can find lots of growth charts listing normal heights and weights for children here, and how to use them here.