Life Expectancy and Health Care Expenditures: Fun with Statistics

Recent data from the University of California Santa Cruz show that although the United States spends far more on health per capita than any other nation, the average life expectancy of its citizens ranks only 28^th in the world.

To support this, the website displays the following graph. It depicts life expectancy as bars and the purple line denotes health care expenditures per capita.

Can you see any problems with the data or the way the data are displayed?

I can. Let’s look at the countries ranked ahead of the US. Of the 27 countries ranking higher than the US, 16 have populations of under 11M with 6 of those having populations under 513K [Luxembourg 512K, Iceland 318K, Malta 418K, Andorra 85K, Monaco 35K, San Marino 32K]. At 312M, the US is by far the largest country in the top 30. Japan is second with 128M people.

The US has a much more heterogeneous population than any other country on the list. The per capita expenditures do not take social factors into account. For example, when anyone twists a knee in the US, an expensive MRI is likely to be ordered, not because it’s necessary, but because it’s expected by the patient. There are issues of non-compliance with medications and self-abuse with drugs, alcohol and tobacco. We also expend a lot of money on futile end-of-life interventions.

The bar graph itself is deceptive in that the scale range is from 74 to 82 years old. This makes the difference between the US and Japan seem much larger than the four years it actually is. This is a chart showing the actual difference on a scale of 1 to 85.

Now scroll down the page on the UCSC website and find this:

Although Cuba has limited resources and many economic problems, it has made health care a priority. It is not alone. Sri Lanka, China and the Indian State of Kerala are considered "low-income, high well-being" countries, which have adopted policies that not only reduce inequality but also increase overall health and well-being. The results of these policy priorities are significant, and can be measured in survival indicators, such as average life expectancy.

Now go back and look at the figure. You will not find Sri Lanka, China or the Indian State of Kerala in the list of the top 30 countries in life expectancy.

I am not saying we should not try to do better with our health care dollars. But I think the discussion should not be distorted by those with agendas. The behavior and expectations of our citizens will have to undergo a real sea change before anything meaningful will happen.

How many medications can a patient take?

Here is a list of medications that a 75-year-old patient is taking at home. All are orally taken except where noted.

Acidophilus 2 twice a day

Colace 100 mg twice a day

Coumadin 5 mg once a day

Detrol 4 mg once a day

Diflucan 50 mg once a day

Duragesic 12 ug transdermal patch every 3 days

Klor-Con 10 meq once a day

Lactulose 30 mL twice a day

Lantus 15 units subcutaneously once a day

Lasix 40 mg alternating with 20 mg every other day

Lisinopril 2.5 mg once a day

Metoprolol 25 mg twice a day

Multivitamin 1 tab once a day

Novolog Sliding Scale subcutaneously with meals [dose varies according to blood sugar]

Oxycontin 20 mg every 12 hours

Prilosec 20 mg once a day

Reglan 5 mg before each meal

Vitamin A 25,000 units once a day

Zoloft 50 mg once a day

Docusate 100 mg twice a day as needed

Hyoscyamine 0.25 mg sublingual every 4 hours as needed

Oxycodone 5 mg every 8 hours as needed

That’s 22 different drugs with a minimum of 24 pills per day and a maximum of 35.

With the Lasix and the sliding scale Novolog, do you think she ever misses a dose or takes a wrong dose?

The above patient is on three narcotics Duragesic, Oxycontin and as needed oxycodone. Add to those meds Zoloft and Reglan. I wonder if she’s drowsy or confused? I bet it will be a big mystery when she falls and breaks her hip.

What effect do you suppose 10 meq of oral potassium per day is having on her total body potassium [total body contains some 3000 meq of potassium] or even serum potassium?

This is not close to the record. I once was consulted on a patient who was on 31 different medications at home. I’m sure that someone has seen a patient taking more than 31 medications. Have you?

“I could teach a monkey how to operate”

“I could teach a monkey how to operate” or so some people think. This cliché has been around for years. Recently, I saw it on a Sermo [MD gripe site] post and someone said it in a comment on one of my blogs. The other day I used it in a tweet trying to entice people to read my blog on the pitfalls of dictating. And then I thought about it.

I will admit that many procedures we do are simple. I have taught physician assistants to perform such tasks as closing a wound, excising small skin lesions, draining abscesses and the like. PAs who work in cardiac surgery can harvest veins and perform other more advanced procedures. When I was teaching residents, I taught them just about all types of operations, from the most basic to the most complex.

But a Medline search has failed to produce even one paper showing that a monkey has successfully been taught to operate on even another monkey. Even if a monkey could be taught to operate, would she know when and on whom to operate?

Let me give you an example. Thanks to the ever increasing use of CT scans, I have been asked to see a number of patients who have gallstones and abdominal pain. This may surprise you, but many of these patients do not need surgery because their pain is not due to the stones. A recent study with an almost 30-year follow-up confirms this.

Sometimes we get the question, “Is that tumor resectable?” In other words, Can you take it out?” My answer is usually, “Anything is resectable [see hemicorporectomy], but will the patient survive?” That is the real question.

I don’t think a monkey is who you want making that decision.

Dictations can be tricky

I was a surgical residency program director for many years. Not long before I left medical education, I had to stop allowing residents to dictate cases they did. It began to take me longer to proofread and correct their dictation than the time it would have taken for me to dictate them myself.

Here are two examples. The first is from the dictation of an open cholecystectomy.

The next one is the actual dictation of a tracheostomy in its entirety. Parts that are inaccurate and/or incomprehensible are underlined in red. 

I think I was correct in my decision to take over the dictating duties, don't you?

Readmission Rates, Colorectal Surgery and Reporting

The website Outpatient Surgery featured a story about a paper from Johns Hopkins that showed almost 25% of privately insured patients undergoing colorectal surgery are readmitted within 90 days. The paper appeared in the December, 2011 issue of Diseases of the Colon and Rectum.

The story highlighted several elements of the paper including the cost of readmission, averaging about $8800 and the role of surgical site infection as a major factor.

The report about the paper fails to mention the important limitations of the study. The research was based on commercial insurance claims, not hospital charts, and some clinical information could not be obtained. Also, the number of planned readmissions, such as those for elective closure of a colostomy after emergency surgery, could not be ascertained from the database.

The Outpatient Surgery article ends with the following:

Here's how you can reduce readmissions: 

Have nurses review patients' discharge plans before they leave, make follow-up appointments and review medication lists, tasks shown to prevent some return visits to the emergency department for minor concerns.

Have nurses follow up with patients by phone in the days after discharge. Those deemed at high risk for readmission should receive home visits from a nurse.

The only problem is that nowhere in the study do the authors state that the above interventions “can reduce readmissions.” What they do say is that such actions may be of use. But their research was retrospective and simply documented the problem of readmissions. The study was not designed to test whether nurses helping to plan discharges or telephoning patients can prevent readmissions.

Bottom line: Reporters who write about research studies should be careful about what conclusions they attribute to the authors.

Uninsured patients released from hospital sooner than insured: Significant? You tell me.

There’s a bit of excitement on Twitter today with a number of tweets and retweets about a paper just published in the Annals of Family Medicine which shows that uninsured patients are being released from  hospitals significantly sooner than insured patients. The numbers don’t lie.

From the abstract: “Across all hospital types, the mean length of stay … was significantly shorter for individuals without insurance (2.77 days) than for those with either private insurance (2.89 days, P = .04) or Medicaid (3.19, P <.01).” These are statistically significant differences.

The authors conclude, “Future research should examine whether patients without insurance are being discharged prematurely.”

Let’s look a little closer at these numbers. The difference between the uninsured length of stay (2.77 days) and those with private insurance (2.89 days) is 0.12 days or to put it another way, 2.9 hours.

Do you really think that a difference in hospital length of stay of less than 3 hours is really clinically significant? I don’t.

Here’s another problem with the paper. Length of stay is what is called a “soft” endpoint. Having practiced surgery for 40 years, I can assure you that length of stay is very often not determined by the type of illness, treatment rendered, skill of the physician or any other parameter you can think of.

Here is what I mean. Just yesterday, a patient told me he could not go home on the day he had his laparoscopic cholecystectomy because his sister, whom he lives with, gets upset whenever he comes home from the hospital. He felt she needed another day to adjust. Patients have told me, “No one can come and pick me up today.” The care manager says, “The bed at the nursing home isn’t available today.” Three weeks ago we couldn’t send some patients home because there was a massive power outage in our area. This list of excuses goes on and on.

I have written before about the problem of things being statistically significant but not clinically significant.

The paper is another example of statistical significance not corresponding to clinical significance.

Does Oral Contraceptive Use Increase Rates of Prostate Cancer? I Doubt It.

A recent paper linking the use of oral contraceptives by women to higher rates of prostate cancer in men has received some media attention. Two urologic oncologists from the University of Toronto in Canada published this hypothesis in an open access journal, BMJ Open. The full text is available here.

The paper’s plausibility is based on some previous work suggesting the exposure to estrogen might increase the risk of prostate cancer.

The study was done using retrospective data from various sources. Information on contraceptive use and prostate cancer was obtained for countries with all levels of economic development on all continents. The main finding was that oral contraceptive use was associated with both the incidence of and mortality from prostate cancer, while all other types of contraception showed no such association.

A fair amount of statistical manipulation was used to generate the results. When one first reads the paper, it is almost believable until one reaches the section postulating the mechanism by which this association could possibly be real. The means by which oral contraceptive use could affect prostate cancer rates is posited as increased levels of so-called endocrine disruptive compounds (in this instance, estrogens) in drinking water secondary to excretion by women.

To their credit, the authors themselves admit that the paper has some serious limitations.

Obviously, more developed countries would have higher rates of cancer screening and detection. Rates of prostate specific antigen (PSA) testing were not available. The authors felt that their method of controlling for gross domestic product obviated the need to have actual PSA screening data.

The authors further concede that there are no available data on the levels of endocrine disruptive compounds in the drinking water of a single country.

As a mere general surgeon with no ecologic or public health credibility but having seen the vast reservoirs of drinking water in my area, I find it difficult to believe that there could be significant levels of endocrine disruptive compounds in the public water supply. And what about the fact that in many highly developed countries, people drink bottled water presumably free of endocrine disruptive compounds?

An [assumed] editorial summary of the paper states “This study is an ecological study and thus has significant limitations with respect to causal inference.” Translation: There is absolutely no proof that oral contraceptive use in women causes higher rates of prostate cancer in men.

Bottom line: I’m not going to stop drinking water. Are you?

Online etiquette. Stop telling me what to do.

Don't blog/tweet anonymously. Tweet before you blog. Bloggers should meditate [of course]. Don't use emoticons. Don't “dis” patients. Mind your HIPAA (often misspelled "HIPPA"). You have to have a social media presence. I could go on.

I blog anonymously because I want to. I started blogging and tweeting at the same time and I'm doing fine. I use emoticons because sarcasm and satire are difficult to convey in a tweet or email. I find they help me express myself and avoid misunderstandings.

Occasionally, patients who do silly things should be dissed. I rarely do it myself but if others do, it's their business. Also, this can be funny.

Digression: HIPAA is a major nuisance. It's one of the most annoying and misunderstood federal laws ever enacted. Really, are there large numbers of people trying to look at my patients' x-rays so I need to go through logins to get into the PACS? It's not a HIPAA violation to exchange patient information in the process of caring for said patient. But many times doctors and hospital medical records people stubbornly refuse to tell you vital information needed in emergency situations.

I don't want to have a social media presence for my practice. I don't need it. I don't have the type of practice that would benefit by it.

So do me a favor; don’t tell me what I should or should not do online.

Surgical Residents Surveyed; Concerns about Experience & Confidence

A recent survey of general surgery residents was published online by the Journal of the American College of Surgeons yesterday. More than 4200 residents who took the 2008 American Board of Surgery In-ServiceTraining Exam (ABSITE) responded. This accounts for 82% of all categorical (five-year) general surgery trainees. The authors of the paper represented the American Board of Surgery, Yale University and Memorial Sloan Kettering Cancer Center.

Compared to those in university programs, residents in community hospital programs had more positive responses to questions about satisfaction with their operative experience, didactic teaching and support from their programs.

But a significant number (27%) of all residents surveyed worried that they would not feel confident to perform surgery by themselves when they finished training. A similar number were not satisfied with their operative experience. Almost half of all residents were not satisfied with the level of didactic teaching being offered.

Something is wrong if over 25% of surgical residents are uncertain that they will be able to operate independently when they finish training. And just how does one identify those surgeons? Don’t say, “Check to see if they’re board-certified.” The boards don’t test operative skill.

Not long ago, I blogged about the coming shortage of general surgeons. The paper discussed above would indicate that expanding existing general surgery residency programs may not be the answer. If a quarter of all residents feel they are not getting enough operative experience and are not confident in their skills, how can programs be expanded?