Search This Blog

Follow by Email

Wednesday, July 18, 2012

Sigmoidoscopy--might it be better than colonoscopy?

Sigmoidoscopy means visualization of the sigmoid colon, usually by a flexible fiberoptic scope. Colonoscopy is visualization of the whole colon by means of a longer fiberoptic scope. Both have been recommended as methods of detecting colon cancer early or preventing it by identifying abnormalities such as polyps that can lead to colon cancer.

Sigmoidoscopy looks at most of the left portion of the colon and can be done without anesthesia, in a doctor's office after an enema prep the morning of the procedure. After the procedure most people can go have breakfast and go back to work. A colonoscopy, when successful, looks at the entire colon after a patient is sedated by various intravenous medications, and follows a bowel prep that usual involves drinking 1/2-1 gallon of a polyethylene glycol solution that tastes a little like lemonade with baking soda in it and having many many bowel movements starting the afternoon before the procedure. The anesthesia, which is usually quite pleasant, leaves a hangover for a significant portion of the day and most people are unable to return to work. Because a person having a colonoscopy has a longer tube wending its way through their lower intestines and that person is pleasantly snoozing during the time this is happening, there is a higher risk that the operator, a gastroenterologist usually, will inadvertently push the long tube through the wall of the intestine, which can be a real mess and potentially fatal. This rarely happens, but most of us docs have heard of a few cases, often requiring surgery.

On June 21 the New England Journal of Medicine reported a study that looked at whether sigmoidoscopy really could reduce the risk of developing or dying from colorectal cancer. 154,900 patients were enrolled and randomized to either being offered sigmoidoscopy twice in 3-5 years or just seeing their primary care physicians and following or not following their recommendations for screening. 84% of the intervention patients had sigmoidoscopy once and a little over 50% had it done twice. Both groups were followed for 12 years. The incidence of colorectal cancer was 21% lower in the sigmoidoscopically screened group and mortality was 26% lower. If something was found on sigmoidoscopy the patient would get colonoscopy and have a biopsy or polyp removal done. In the control group ("usual care") screening colonoscopies happened, at about the rate that is reported nationwide, of a little below 50%.

So what happened? Most patients in the usual care group were probably offered colonoscopy, and the patients in the intervention group were offered colonoscopy's poor cousin, the sigmoidoscopy. Why did the poor cousin patients do better? It sounds like they did better not because the procedure (sigmoidoscopy) was necessarily better than full colonoscopy, but because it sounded more feasible to the patients so they actually had it done.

A couple of years ago at a table in a board meeting of my medical group a perfectly intelligent doctor colleague of mine said "does anyone here really recommend sigmoidoscopies anymore?" To which I said "yes." I offered them and performed them in my office during the many years I was in private practice and had no complications and discovered the occasional significant abnormality. None of the patients I screened in this way ever had a colon cancer that was clearly missed by the more limited procedure. As the standard of care in my community moved more and more to colonoscopy, I did fewer flexible sigmoidoscopies and referred more patients to the gastroenterologist for colonoscopies. There were occasional complications related to the fact that the colonoscopy was a more complex procedure.

So--is sigmoidoscopy better than colonoscopy? Maybe yes, maybe no. It is definitely cheaper to the uninsured, though now, with the passage of the health care reform bill, all insured patients have full coverage without copay for either sigmoidoscopy or colonoscopy. It is definitely less risky in terms of complications. It also can miss significant abnormalities in the part of the colon not screened, including precancerous polyps and cancers. For patients at average risk of colon cancer, looking for a minimally disruptive and adequate way to undergo screening, sigmoidoscopy should definitely be strongly considered instead of routine colonoscopy.

New recommendation about Pap smear screening for cervical cancer--we REALLY don't need this test yearly!

In the 1920s a couple of pathologists, Aurel Babes of Romania and Georgios Papanikolaou from Greece discovered that in women who had uterine cancer, malignant cells could be found in samples of vaginal secretions. The details of their original reports and recommendation are shrouded in internet obscurity (that is to say I can't find any of the original articles, which would probably be really interesting) but it appears that the idea of sampling the contents of the vagina or the cervix therein did not immediately engender enthusiasm.

In 1941 Dr. Papanikolaou and his colleague, a gynecologist Dr. Herbert Traut, published a paper in the American Journal of Obstetrics and Gynecology describing how to make the diagnosis, and at some point screening for uterine cancer by sampling cells shed from the lining of the uterine cervix became a common practice. Dr. P. himself appears to have been a major force in the practical aspects of making screening an expected part of being female, at least in the US.

All of this gets just a touch confusing to those not familiar with the layout of the female genital tract and the words used to describe its cancers. Just south of the belly button lies a triumvirate responsible for producing babies. The uterus receives fertilized eggs from either of two ovaries by way of the fallopian tubes and the cervix (latin for neck) of the uterus is a tiny canal which permits passage of sperm for fertilization and eventually passage of a baby in the other direction. The most common kinds of uterine cancer are cancers of the skin type cells lining the cervix and cancers of the glandular cells lining the body of the uterus. We call the first cervical cancer and the second endometrial cancer, but Dr. P. originally called them both "uterine cancer." The Pap smear is only good for detection of cervical cancer.

The technique used in obtaining the Pap smear when I first learned how to do it 25 years ago was to take a small rounded wooden spatula and rotate it in the opening of the cervix, smearing the goop obtained on a slide and then applying a fixative which was sometimes just plain hairspray. Eventually we started using little bitty bottle brush thingies instead of or in addition to the little spatula, which got better samples but caused more bleeding and discomfort, and a few years ago we were told that it was even better if we took the little brush (which had undergone another transformation to something that looked like a tiny broom) and swished it around in a fixative and had the lab apply it to a slide. This sample could also be tested genetically for viruses and sexually transmitted pathogens, which before had required a separate sample. It was also more expensive and we were reimbursed more. Funny thing.

In 1984 when I performed my very first pap smear we knew that cervical cancer was more common in women with multiple sexual partners and almost unheard of in women who had never been sexually active. In the next couple of years the increasing abilities of scientists to manipulate viruses in the lab led to the discovery that human papillomavirus , a sexually transmitted wart virus, could cause cervical cancer, and over the next several years it became widely accepted that this virus was necessary for the development of cervical cancer. There are many different types of human papillomavirus and only some of them can cause cancer. There are other factors that can increase a woman's risk of cervical cancer including smoking, immune suppression such as in HIV and genetic predisposition. A vaccine against human papillomavirus was released in 2009 and a 3 dose course of this significantly reduces a woman's risk of developing invasive cervical cancer.

Recommendations about Pap testing have changed significantly over the years, but it seems to me that doctors are particularly slow to adjust to guidelines. When we found out that human papillomavirus was the culprit agent and that it was sexually transmitted, women were told that they needed to get yearly pap smears starting with or before they became sexually active. Several years ago the recommendations changed to recommend 3  yearly pap smears starting at the first sexual activity and then repeating the test every 2-3 years until age 65 at which time testing could cease. Women who had had hysterectomies with removal of the cervix could forego pap smears entirely. These recommendations were especially unpopular with gynecologists in my community who continued to do yearly Paps and Paps after hysterectomy for many years (I think some of them are still doing this.) In 2009 women were told that they didn't need to start testing until age 21 or until 3 years after "sexual debut" and could go 3 years between testing after age 30 if their results were normal.

Just last month the US Preventive Services Task Force released evidence based guidelines on Pap testing that  are even more relaxed that 2009, which makes me, as a doctor and as a woman who gets Pap tests, very happy. Cervical cancer screening should not begin until age 21, even if a woman becomes sexually active significantly before this age. After that, she should be screened only every 3 years. After reaching the age of 30 she can be tested with a standard pap smear and for evidence of HPV and if those are both negative she can have pap smears only every 5 years. She can also choose to have every 3 year testing with just the pap smear. The more expensive pap testing with the cells collected in liquid and sent to the pathologist is not more accurate and so we don't need to be doing that routinely, though it does allow for the DNA testing with only one specimen. Women who have been regularly tested can stop screening at age 65 and women who have had hysterectomies for non-malignant disease don't need paps. All of these recommendations do not apply to women who have had abnormal testing and to women who are infected with HIV. Why no HPV testing before age 30? Because apparently HPV is very common in young sexually active women and comes and goes, not requiring treatment most of the time.

But some people (especially doctors) argue that Pap smears are easy, so why reduce the frequency of testing and risk missing some disease? Pap smears are easy, but not harmless. Women are pretty good sports about Pap testing, but it is actually kind of awful to go into an impersonal doctors office, get naked, put on a skimpy gown, spread your legs (just about as vulnerable a position as I can imagine) and have a near stranger poke something (a speculum, usually made of plastic and prone to producing disconcerting clicks and pinching tender flesh) into one's private parts, followed by scraping the soft distal protrusion of a delicate internal organ, usually resulting in bleeding and cramping. But that's just the beginning. If the test is abnormal, repeat testing often occurs more often than yearly, and may lead to destruction of tissue with risks of infection, significant bleeding, work loss and future difficulty getting or remaining pregnant.

Most cervical cancer deaths occur in women who have never had Pap testing, especially women in developing countries and disadvantaged women in the US. Reduced frequency of testing may make it feasible to offer testing to more women and to have it be more acceptable to them. HPV vaccination certainly has great promise in reducing cervical cancer especially for women who don't have access to regular health care, but at this point does not influence recommendations for frequency of Pap testing. Newer recommendations for less frequent testing do offer those of us who dreaded our regular physicals relief from the burden of excessive testing.

I, for one, am happy to forever retire the expression "annual pap smear."

Monday, July 16, 2012

Passing the American Board of Internal Medicine MOC

I took the American Board of Internal Medicine maintenance of certification test on April 25, 2012. On June 26, 2012 the board notified me by e-mail that I passed the written test, the final step.

I started the whole process June 30, 2010, by paying for the $1570 fee to the ABIM and began to do the self study and practice evaluation modules. I was able to do the self guided modules on the website pretty easily while working full time, but I'm not sure they were that valuable, since completing them just required a multiple choice test that I did while looking for specific answers. I guess that is pretty similar to how I solve academic problems for my practice, but I don't think the information sunk in too well. The best self study module was a totally self guided one in which I asked questions as they arose with my patients and later went back and found resources to answer them and described the process.

The biggest hurdle for me was the Practice Improvement Module which required that I survey a number of my patients regarding what they noticed about my practice. I chose preventive care, so they responded to questions about that plus general questions about how they were treated. This was time consuming and a little bit disruptive, but actually in a good way, and I ended up reviewing quite a bit of material about recommended screening procedures and looking at what kind of data they were based on, and really paid attention to the controversies that arose during that time about changes in recommendations. This piece took maybe 3 months.

When I had completed the required self study and practice improvement modules the board notified me that I could take the test. I scheduled that, at a testing center not too far away, and took an update CME (which I described in a previous blog in gory detail.) I also signed up for the MKSAP (the medical knowledge self assessment program) which was excellent and really sunk in well, but was very time consuming to do right. This is as it should be, since it does take time to enter a whole mongo ton of information into my brain. I was glad I was doing hospitalist work, though, so I could have large blocks of time off to study.

The test was long, but not horrendously arduous, and seemed fair. I also wrote about this in a separate blog so I won't belabor it.

The final scores were reported differently than when I took the test 20 years ago, but I'm thinking I did about as well. I missed 20 out of a total of 180 questions which put me in the top 10 percent of people taking the test. That's good, in a way, but also a little horrifying since all of those questions were things I probably should have known, and 90% or more of the people qualifying to take the test get more wrong. About 70% of test takers pass, and the standardized passing score was 366. Mine was 651. I'm not sure how they calculate the standardized scores, but it kind of looks like you could get almost half of them wrong and still pass. Some of the questions I got wrong I might have argued that a reasonable physician would disagree with the chosen answer, but some I probably just didn't know.

So I passed. Yay! I think I will probably make a point of doing the MKSAP as a kind of regular thing so I will keep up to date on the wide and wonderful body of knowledge that is internal medicine.

Thursday, July 12, 2012

More on ultrasound use at the bedside: this short week in review

This week I was supposed to work 7 days at a busy hospital as a hospitalist. I am, however, done today, after only 3 days because the company that hired me found someone cheaper to work for 4 of the 7 days I was scheduled. I found out about this too late to refuse, but it is really not a good thing and I am moderately annoyed. It is, of course, nice to not work 7 straight very difficult days and to be headed home on a beautiful Thursday to spend some unexpected free time with my family, but the disruptions involved in working in shifts, as hospitalists do, are bad enough when we change every 7 days. The reason I'm more expensive and therefore at a disadvantage in my present job is that I am represented by a locum tenens company that is particularly nice to me and therefore costs the client hospital more money. I'm just learning this.

Nevertheless, it was a very good 3 days for learning. There were patients with mysterious illnesses that I got to discover and almost figure out, which was truly delightful. And I got to do many many bedside ultrasounds.

The best thing about working at a hospital where the patients are really sick and I am seeing lots of them is that there are so many good stories and so much interesting pathology and I get to meet so many people, patients, families and consulting doctors. The drawbacks involve my increasing wrinkles and gray hair.

Having bought my little pocket ultrasound machine, I now ultrasound everybody at least once and many of them I do repeated ultrasounds to followup abnormal findings. Other than the actual radiologists in this hospital and the occasional ER doctor, the only people who use ultrasound in patient care are me and the intensive care physicians who work in the Intensive Care Unit. Those of us who use ultrasound at the bedside are at a distinct advantage. That is putting it very mildly. With bedside ultrasound we can see, in minutes, what everyone else is simply guessing about. Some examples with details changed so as to be discrete:

I met a woman who was transferred from her regular doctor's office to the hospital because she had gained 30 pounds of fluid weight in a month and was starting to exude it from blisters on her legs. Very disconcerting! She was diabetic and had not had heart problems before. The possible causes could have been liver failure, kidney failure, failure of one or both sides of her heart or a host of other rare conditions. We met, talked, and I looked at her heart, lungs and abdominal cavity with my little machine and was able to see and show her what was wrong. Her heart had ceased to pump well, which had put liters of fluid into her lung and abdominal cavities and had backed up into her legs. After starting several medicines for heart failure she felt a tad better and I could show her that her heart, though still very tired, was getting a little more perky. This was very useful because, though she had a "real" echocardiogram done after I made the diagnosis, it would have been expensive and impractical to repeat that daily, and with my little machine I could show the patient exactly what I was looking at.

An elderly man who had a hip fracture did fine after surgery but then was unable to maintain a normal blood pressure an hour later and was not waking up. IV fluids helped only a little. A bedside echo at 11 at night showed his heart to be thickened, noncompliant and not a very strong pump (couldn't fill, couldn't empty). Ultrasound of his lungs showed that they were not yet wet, which was good. He could get more fluids, and probably what had happened was that his small firm heart had lost pressure due to dilated blood vessels with anesthesia and was temporarily out of service. No heart attack, not occult blood loss, not severe infection.

The very most interesting thing for me, though, was in the Intensive Care Unit where I sometimes hang out after my shift to learn things from the truly awe inspiring docs who work there. When I walked in, the patient, a 60 year old woman with severe lung disease, morbid obesity and a really bad pneumonia was getting a central venous catheter put in to her subclavian vein for infusion of the various potions that she would need for surviving, were she to do that. Ultrasound was, of course, used to put the line in, which was great, because there was a lot of tissue covering the landmarks usually used to put in such lines and the ultrasound made sure that the needle went into a vein rather than an artery and avoided the lung. As we sat talking, the patient's blood pressure started to fall. The ultrasound said that there was no pneumothorax, but it wasn't quite as good a picture as would be ideal, so we looked at the x-ray too, which took forever to be processed (it seemed like forever, but was probably 15 minutes, which is forever when someone is dying.) So the blood pressure problem was not a complication of the line placement. She then had no blood pressure, requiring cardiopulmonary resuscitation, chest compressions, the whole thing. Definitely bad news. There was a rhythm on the monitor but no blood pressure, electromechanical dissociation, a very bad prognosis. What was going on? Bedside echo shows that the heart is hardly pumping. Blood gas shows severe acidemia. Appropriate treatment is to correct the acid base balance and support the heart's pumping function artificially with epinephrine. Repeat echo in 5 minutes: heart is working fine. This only lasted about 20 minutes, then her pressure dropped again, CPR started again, echo shows heart is pumping fine, so the patient now needs more fluid due to the effects of septicemia from the pneumonia. The resuscitation went on intermittently for hours, and as such things go, did not end well, though the patient was stabilized as far as blood pressure and heart function went.

Quite the week. I think I'll go outside now and see what the sun looks like.

Monday, July 2, 2012

Mysteries revealed: why do medical supplies cost so much and how do the poor and uninsured pay for their hospital care?

A couple of weeks ago I wrote a post about the cost of various medical supplies at our hospital and about the cost of similar supplies from a veterinary source. Central venous catheters of various types (also called "central lines" or "PICC lines") when billed out by our hospital cost about 30 times what similar products cost vets. I ordered a central venous catheter from a veterinary supply company to see what 29 bucks could buy. I got it in the mail a few days ago.

The veterinary central line looked fine. It was almost exactly what we used maybe 25 years ago, which is only slightly different from what we use now. There have been some design changes that involve making sure that unintentional needle sticks to the provider placing the line are avoided, due to the risk of blood borne diseases like AIDS and Hepatitis C. There was no drape or sterile gown, but those are pretty inexpensive, especially from the vet source. I also bought a regular peripheral IV, the kind that we bill out for $60 or so, because it was only $1, and I was curious. The nurse in the ICU said that it looked like the kind we used a couple of years ago, which worked fine, but we use a different kind now, also designed differently because of the risk of blood borne diseases. I'm not quite sure why I bought the surgeon's sterile gown, but it was cheap too, and is exactly what we use for humans. It was about $3 and will work well as a kitchen apron if I do any deep fat frying.

I realized, though, that what the items cost at the supply company would not be the same as what a hospital bills for the item, since clearly there would tend to be a markup from a wholesale purchase by the hospital to a retail sale to the patient or insurance company. What, I wondered, did the hospital pay for these central venous catheters, and what did the consumer actually end up paying for them?

Today I went in to the hospital and talked to the various billing people and to the guy in central supply who know the answers to these questions. I learned a ton of stuff that clears up questions I didn't even know I had.

First, the numbers. When a hospital buys things like central lines and PICC lines, they get a really good deal because they buy in bulk. We pay about $110 for a triple lumen central line kit, including the drapes and gowns and hat and gloves (which nobody uses because they are one size fits nobody) and suture and tiny bottle of numbing solution. We bill out almost $900.  For the single lumen PICC line we pay about $170 and bill out around $1300. We pay $10 for a peripheral IV.

I asked the billing people how much Medicare pays for these things. Do they actually pay us 800 and something dollars for the central line kit? Well, actually, no. They don't pay us for any of our supplies. They pay a daily fee to the hospital for all supplies and room and board and social workers and cardiac monitoring and medicines and nursing. They also pay a percentage of professional fees, which include all physician services, on top of the daily fee. Hospitals get different daily rates from Medicare based on all kinds of things, and ours, a critical access hospital in a rural area gets about $2300 a day for patients insured by Medicare.

What about other insurance companies? It turns out that they usually pay a flat fee based on a diagnosis, so hospitalization of a very sick person with pneumonia and delirium and dehydration would result in a certain level of reimbursement for the hospital no matter how many days they were in the hospital or how many central lines or IV's they had placed. There are, however, some insurance companies that still pay a percentage of what is billed, and it is probably for those payers that prices like the $1300 for a PICC line are created. In my experience the price that is charged is usually just a little more than the allowable payment by the highest paying insurance company. In this way, a provider, hospital or primary care or specialist, is able to make the most amount of money from those who pay the most generously.

So what about the patient who is uninsured? That person will be looking at a bill that is larger than any insurance company will ever pay. This is not because the hospital wants to penalize the uninsured. The prices the hospital charges have to be the same for everyone, but different groups can negotiate for lower reimbursement. This doesn't entirely make sense to me. I'm not sure why it is a different thing to charge a different amount versus to accept a different amount in payment. I will eventually find the right person to ask about this.

But what, again, about this hypothetically uninsured patient? Is she really expected to pay this premium bill? And does she? In not-for-profit charity hospitals (only about 18% of hospitals are for-profit) patients are charged a percentage of their bill based on their income. Some patients are eligible to have their bills completely written off. Others are billed a percentage and the hospital works on payment plans with them so they can pay off a bill over months or years. Most patients  apparently try really hard to pay their bills. One of my sources estimated that about 50% of uninsured and underinsured patients paid their bills, based on the sliding scale. (This is, of course, a gross estimate based on one very small geographical area.)

Hospitals try to achieve charitable, not-for-profit status because they are then tax exempt, which is a major financial benefit. To do this they must provide services to everyone, regardless of their ability to pay and must have emergency services available. They must also feed profits either back into the community or into the hospital in such a way that patient care is improved. New legislation is adding some other requirements ( The cost of care for those who can't afford to pay will inevitable be shifted to those who can, be it individuals or insurance companies.

Wow. That was a large amount of dense and maybe confusing information. My interpretation of the take-home points would be:

1. Basic medical supplies aren't all that expensive to hospitals. What suppliers charge for these things is probably not a big part of the problem of excessively high healthcare costs.
2. Veterinary supplies are quite a lot cheaper and pretty similar.
3. The amount that a hospital bills for a supply is hugely inflated but mostly irrelevant to the majority of patients and insurers since they pay a flat fee for to the hospital for everything, but very relevant to the uninsured patient.
4. Insurance companies (third party payers) determine the prices the hospital bills by the amount that they agree to pay. Without a third party payer system that pays fee-for-service or fee-for-item, costs would be lower.
5. Most hospitals are not-for-profit and so charge uninsured patients less than they bill, according to a sliding scale. Since hospitals cannot run at a loss, costs for these patients' care will be shifted to patients or insurers who pay more.