Dr. H. Gilbert Welch has written a new book Over-diagnosed: Making People Sick in the Pursuit of Health, with co-authors Lisa Schwartz and Steven Woloshin.  It identifies a serious problem, debunks medical misconceptions and contains words of wisdom.

We are healthier, but we are increasingly being told we are sick. We are labeled with diagnoses that may not mean anything to our health. People used to go to the doctor when they were sick, and diagnoses were based on symptoms. Today diagnoses are increasingly made on the basis of detected abnormalities in people who have no symptoms and might never have developed them. Overdiagnosis constitutes one of the biggest problems in modern medicine. Welch explains why and calls for a new paradigm to correct the problem.

Where to draw the line? FDR had a BP of 200/100 at the time of his re-election in 1944 and subsequently died of a stroke with a BP of 300/190. At that time, elevated BP was not commonly recognized as a problem requiring treatment. Then studies showed that the higher the BP, the greater the risk, and now everyone diagnosed with HBP is treated. That has undoubtedly saved many lives; but for someone with only a mild elevation, the risk of heart attacks and strokes is smaller and the risk of complications from treatment becomes less acceptable. So where do you draw the line and start treatment?  When the limit of 160 systolic was dropped to 140, the new definition instantly turned 13 million people with “normal” BP into patients with hypertension. Not all of those new patients were better off with treatment. Welch gives the example of an 82 year old man who was treated for mild HBP at a level where the number needed to treat for one person to benefit (NNT) was 20; he passed out from medication side effects and declined further treatment.

Changing the Rules: We’ve changed the diagnostic thresholds for many diseases, so that people who were previously classified as normal are now diagnosed with diabetes, high cholesterol and osteoporosis. Dropping the threshold of fasting blood sugar from 140 to 126 instantly created 1.6 million new diabetics, diabetics who were less likely to develop symptoms and complications and were less likely to benefit from treatment. He tells about one of his patients who was put on blood sugar-lowering medication because of the new rules and passed out while driving and broke his neck because the medication brought his blood sugar too low.

Osteoporosis: here are the numbers for treatment of decreased bone density:

  • Winners (treatment saved them from a fracture): 5%
  • Treated for naught (had a fracture anyway, despite treatment): 44%
  • Losers (treated but never would have had a fracture without treatment): 51%

Seeing too much: New technology allows us to detect abnormalities that would never have caused harm. In people without back pain, over 50% have bulging discs on MRI; 10% of asymptomatic people have gallstones on ultrasound. In patients without symptoms, what’s the value of knowing about these findings? In people with symptoms, such findings may lead to a false diagnosis.

10% of the general population and 7% of people under the age of 50 have findings of stroke on MRI. Whole body CT scanning finds abnormalities in 86% of asymptomatic people. The higher the resolution of your testing method, the more anomalies you will detect; but how many of them are important to know about? How will finding them affect health outcomes?

Prostate cancer: the harder you look, the more you find, and the smaller the cancers you detect, most of which would never have hurt the patient. Welch estimates that for every prostate cancer death avoided by screening, between 30 and 100 patients are harmed by unnecessary treatment.

In breast cancer, for every death prevented by mammography, 2 to 10 women are overdiagnosed and treated unnecessarily, 5 to 15 are diagnosed earlier without any effect on final outcome, 250-500 will have a false alarm and half of these will be biopsied. 999 out of 1000 women do not benefit from mammography. A study in Norway showed that screening resulted in 22% more diagnoses of invasive cancer; apparently some invasive breast cancers in the unscreened group had spontaneously regressed.

Other cancers: In an autopsy study, researchers determined that almost everyone has small thyroid cancers; so many that they could be considered “normal” findings. The US Preventive Services Task Force (USPSTF) recommended against screening for thyroid cancer, since it increases the diagnosis rate without affecting the death rate, and increases morbidity from unnecessary surgery and other treatments.

There is overdiagnosis of melanoma and lung cancer. For colon cancer and cervical cancer there is overdiagnosis of precancerous abnormalities.

The good news: We are learning that many, perhaps most, small cancers either regress or never progress. Spontaneous remissions may be far more common that we ever imagined. In one study, 14% of kidney cancers got smaller without any treatment. So we don’t really need to know if any cancer is present: we need to know if a cancer is present that is likely to progress and harm the patient. And so far we have no way of distinguishing which these are.

Incidentalomas are nodules or other unexpected findings noticed on imaging studies, often in body parts adjacent to the area being studied. About half of virtual colonosopies detect abnormalities outside the colon. More than 99% of the time, these are not cancers and not important to know about; but they lead to anxiety, further studies, surgeries, and complications. Protocols are being developed to follow incidentalomas suggestive of kidney and lung cancers over time rather than immediately pursuing diagnosis.

Routine electronic fetal monitoring has minuscule benefits and results in many more C-sections.

Vascular screenings: The Lifeline company and other commercial ventures offer tests direct to the public, tests that the USPSTF doesn’t recommend and that have not been shown to benefit those screened.

Genetic screening. These tests are not done for symptoms, and do not even detect signs of early disease, but just estimate future risks using inadequate data. Welch reminds us that genetics is not destiny and abnormal genes do not equal disease. The predictive value of these tests is small, and we seldom know what to do about the risk after we identify it. Low risk for a condition doesn’t mean you can’t get it, and everyone is at high risk of something.

A paradigm shift is needed, but it will be difficult to achieve for many reasons:

  • It is hard to ignore information.
  • Most people believe the more information, the better.
  • Accepted wisdom and common sense are hard to overturn.
  • Most people are convinced that it is always in people’s interest to detect health problems early, even though the data say otherwise.
  • There is a common belief that early detection is cost-effective, even though the data show it actually ends up costing more.
  • We find it hard to tolerate uncertainty.
  • Commercial interests benefit from screening and overdiagnosis.
  • Doctors fear being sued if they omit tests.
  • Anecdotes about lives saved are emotionally persuasive.

We are easily impressed by anecdotes from people who believe their lives were saved by early detection; but we don’t hear anecdotes from people who were harmed by a diagnosis of a condition that would never have hurt them, mainly because we have no way of knowing which ones they were. I am a case in point: I had a suspicious mammogram and an excisional biopsy that removed a lobular carcinoma in situ. That is not really a cancer, but more like a risk factor for cancer. Did my surgery remove a part of my breast that would have eventually developed invasive cancer and killed me, or did it uselessly remove a harmless chunk of tissue? Did it save my life or just mutilate me? I will never know.

What’s the solution? Maintaining a healthy skepticism about early diagnosis. Informed consent for screening tests, based on accurate information. Resisting over-simplified hype about the benefits of screening. Putting our efforts into prevention (exercise, smoking cessation, healthy diet, etc.) rather than pursuing early detection. Pursuing health without paying too much attention to it and without developing anxieties about it. Welch argues for not even mentioning incidentalomas on imaging reports, but I think radiologists and lawyers would object to that strategy. He says

Severe abnormalities warrant action because net benefit is likely. But the best strategy for mild ones may be to leave well enough alone, otherwise net harm is likely. In fact, it may be better not to look for them in the first place…An overdiagnosed patient cannot benefit from treatment… [but] can only be harmed.

He doesn’t offer prescriptions. He recognizes that different individuals will assess the risk/benefit ratio differently; based on the same data, some will choose to be screened and some won’t. But they deserve accurate information to base their decisions on, and this book offers a lot of good data and thought-provoking analysis.

I couldn’t help but like this book, since it says many of the same things I have been saying about screening tests , colonoscopy, osteoporosis treatment, PSA tests , not always treating , ultrasound testing , overuse of CT angiograms, genetic testing in general and in specific situations, and the pitfalls of diagnostic tests. It explains complicated concepts like lead-time bias in simple terms and spices the story with patient anecdotes. I found it a bit repetitive but that is probably an asset for driving the message home to a general audience. Both patients and doctors would benefit from reading this book and thinking about the issues it raises.

Posted in: Book & movie reviews, Cancer, Diagnostic tests & procedures

Leave a Comment (56) ↓

56 thoughts on “Overdiagnosis

  1. Stephen Wood says:

    Interesting stuff. Can you give the reference for the spontaneous remission studies, particularly the kidney cancer one?

  2. BillyJoe says:

    “Putting our efforts into prevention (exercise, smoking cessation, healthy diet, etc.) rather than pursuing early detection. Pursuing health without paying too much attention to it and without developing anxieties about it. ”

    For me, it’s mostly the sentiment expressed in these two sentences, except that the words “efforts” and “pursuing” seem too strong. The health benefits of exercise are merely a bonus side issue and, if a balanced diet (a bit of everything) is a healthy diet, then I have that. But, I have chosen never to have any medical examinations, blood tests or screening tests done looking for disease or disease risk.

    So far so good. :)

  3. marcus welby says:

    A contributing factor is common in my experience, and alluded to in Dr. Hall’s comments: the imaging studies are so sensitive that an uninformed or unthinking physician or physician extender may glance at the imaging report and direct the patient with a minor finding of no significance on for specialty consultation. As mentioned by Dr. Hall, lumbar spine MRI studies will nearly always show disc abnormalities in adults, and MRI of the knee likewise. Some busy primary care practices have triage by physician assistants or nurse practitioners, or, for that matter, physicians who are not using their diagnostic thinking caps. A patient with a pain complaint gets an MRI, then the report comes back indicating an abnormality….the primary care folks refer the patient for specialty care without consideration of the significance or lack of it…of the imaging finding. A variation on this scenario is the personal injury victim who is referred to a chiropractor (or colluding MD) following a fall or auto accident, when Personal Injury Protection funds run out, the chiropractor or MD cites the MRI as evidence of permanent injury suitable for compensation.

  4. Jan Willem Nienhuys says:

    Does the book say anything about overdiagnosis in psychiatry? I think that ADHD is seriously overdiagnosed, autism has been stretched to autism spectrum to the extent that children that are able to concentrate for a long time on something and children that are naturally very active both are considered deviant. In the review above the emphasis seems to be on advanced physical methods (scan, ultrasound, biochemical tests and prodding with ‘scopes’) which seems to preclude attention for ‘tests’ in psychology and psychiatry based on vague questionaires and symptom lists.

  5. A couple questions – Firstly – “The US Preventive Services Task Force (USPSTF) recommended against screening for thyroid cancer, since it increases the diagnosis rate without affecting the death rate, and increases morbidity from unnecessary surgery and other treatments.”

    As a laymen, I think I’m a little unclear about the word screening (I looked it up, but still not sure). Are we talking about screening of a general population for thyroid cancer or testing a person with thyroid disease and nodules of a certain size?

    I guess I have the same question for genetic testing. While I can see that screening for genes linked to cancer or heart disease may be difficult to know what to do with, currently many children get genetic testing due to congenital malformations that are suggestive of genetic or hereditary conditions or syndromes. If a known genetic condition is found, that can (not always) be useful in planning medical care as well as future family planning for the parents and the effected child when they reach adulthood.

    Interesting to think about these things, though. We do often make the assumption that early detection of all disease or having a diagnoses for a symptom is good. It’s helpful to aware of, and thoughtful about the ways that it is counterproductive.

  6. Rick says:

    Reminds me of the book Overtreated by Shannon Brownlee. Great book that is a quick read. One of the predomint people featured is Dr. David Eddy (MD, PhD) who helped develop Archimedes (a mathematical model of human physiology, diseases, interventions and healthcare systems). I highly recommed his thoughts on healthcare.

  7. What’s the solution?

    In addition to those mentioned, how about a national database of “incidental findings” that could provide data for ongoing epidemiological studies? Let’s try to find out which ones may be worth pursuing. This would pose problems of informed consent, privacy, and follow-up data, but it would be too bad for such information to be wasted.

  8. cervantes says:

    In response to MicheleinMichigan —

    It is indeed important to understand the difference between screening tests, and diagnostic tests. Screening means giving a test to everybody within a very widely defined group, such as all women over 40 or all men over 50 or even all people. A diagnostic test is given only when there is reason for suspicion of a specific problem, as you suggest.

    A very important idea, which can often be counterintuitive, is called Bayes theorem, named after its discoverer. Even if a test is highly “specific,” for example if 9 times out of 10 a person who doesn’t have the problem will test negative, if the problem is rare most positive tests will be false positives. I could explain it further but I’ll leave that to a front page poster to do it justice.

    The point is, if you give even a very accurate test to people who are at low risk, most of the positive tests will end up causing harm — anxiety, further tests, treatment with its risks and pain and side effects, monetary cost and wasted time — without doing any good. The unnecessary treatment happens when we can’t really tell which positive results are really meaningful and which aren’t, as in the case of prostate cancer and some breast cancers.

    But, if the prior probability of a real problem is high — you have symptoms of some kind that are highly suggestive of a problem for which there is a test — a positive result is much more reliably an indicator that the problem really exists. That’s a diagnostic test.

  9. BillyJoe “But, I have chosen never to have any medical examinations, blood tests or screening tests done looking for disease or disease risk.

    So far so good. :)

    Not that it’s any of my business, but since you put it out there. My husband recently went in for a eye examine for new glasses. While there, they did a standard glaucoma screening, which got suspicious results. They followed up with a couple tests that indicated a possibility of glaucoma. He was referred to a ophthalmologist who did an examine and tests and indicated that while he does not have glaucoma, he does have some results that could indicate a pre-glaucoma phase. He recommended periodic testing…watch and wait.

    Considering that glaucoma is one of those disease that can progress without obvious symptoms and the damage is reasonably irreversible, but has a good prognosis when treated early, I would think screening of the at risk population (by age, family history, etc) is a responsible approach.

    I would think that there are some other screening tests that fit that profile and other tests that don’t. To me it seems best to decide on a case by case basis, evaluating risk/benefit, etc. Rather than just broadly deciding ‘no screening, exams, etc’.

    Which is not to suggest that you have decided broadly, perhaps you did evaluate things on a case by case basis and no tests was your answer.

  10. cervantes – Screening test vs diagnostic test.

    Ahhh, That helps. Thanks!

  11. David Gorski says:

    A contributing factor is common in my experience, and alluded to in Dr. Hall’s comments: the imaging studies are so sensitive that an uninformed or unthinking physician or physician extender may glance at the imaging report and direct the patient with a minor finding of no significance on for specialty consultation.

    Imaging sensitivity also contributes to overtreatment, through a phenomenon known as stage migration or, colloquially, the Will Rogers effect. I discussed stage migration in the context of discussing H. Gilbert Welch’s earlier work, namely a classic paper in the NEJM published in 1993 that is just as relevant today as it was 18 years ago:

    The early detection of cancer and improved survival: More complicated than most people think

    Early detection of cancer, part 2: Breast cancer and MRI

    And the problem of overdiagnosis and overtreatment:

    Do over one in five breast cancers detected by mammography alone really spontaneously regress?

    Are one in three breast cancers really overdiagnosed and overtreated?

    The mammography wars heat up again

    Then of course, all these issues come to a head when it comes to making science-based guidelines for screening asymptomatic populations by, for example, mammography:

    The USPSTF recommendations for breast cancer screening: Not the final word

  12. Jan Willem Nienhuys says:

    @ cervantes on 01 Feb 2011 at 9:35 am

    The philosophy about false positives is too important to leave it to the occasional poster. It is also very simple if you calculate in odds, i.e. ratios of chances yes:no. If the odds are small, then they are about equal to the chance, if they are large then 1/odds is the chance of ‘no’.

    Suppose that the odds of having a disease are 1:100 (i.e. in 101 randomly chosen people the disease is present in 1 and absent in 100). These are called prior odds.

    Do a test. How to compute the posterior odds if the test is positive? Suppose the percentage c of correct positive is known and also the percentage f of false positives. Multiply the ratio 1:100 by c/f. That’s all. So if c=80% and f=8%, the prior odds 1:100 change into posterior odds 10:100.

    In practice (I hope) c is somewhere between 50% and 100%. So only if the false positive percentage is very small, the posterior odds are much larger than the prior odds.

    If one starts with very small prior odds (i.e. rare diseases) then the posterior odds will be also small, unless one has a test with an extremely low false positive rate (example: hiv-test with Western blot plus ELISA).

    Hiv provides a good example of a typical testing problem: the prior odds for someone who doesn’t belong to any risk group and just takes the test because a life insurance company wants it, are extremely low and in the same order as the false positive rate. The posterior odds are: prior odds divided by false positive rate, which is: very small not precisely known number divided by another very small not precisely known number. The hapless positive testee still doesn’t know: is s/he the rare instance of a false positive or the rare instance of a person who is in a low risk group and still got the infection?

    The above analysis tells you that mass screening for rare diseases is dubious. However, if someone is tested when there are symptoms or if such a person belongs to a group in which the disease is much more prevalent, then the prior odds are higher for that person, and then the posterior odds can be quite large.

  13. Draal says:

    I know, lets screen for DNA markers and start treatments based on the results. What could go wrong?

    And the CDC is still stating that only 1 out of 3 Americans are receiving treatment for high blood pressure.

  14. David Gorski says:

    FDR had a BP of 200/100 at the time of his re-election in 1944 and subsequently died of a stroke with a BP of 300/190. At that time, elevated BP was not commonly recognized as a problem requiring treatment.

    Is it really true, though, that extreme hypertension wasn’t recognized as a problem? I thought that it was more a nihilistic view that there just wasn’t much that could be done about it, a view that prevailed until the 1960s, when several antihypertensives came into use. The problem in the 1940s, if I recall correctly, is that there just weren’t really any good oral drugs to lower blood pressure safely on a chronic basis, as this section in the Wikipedia entry on hypertension describes. While it is certainly true that treating “prehypertension” (DBP between 80 and 90) is a recent phenomenon, it’s been appreciated for a long time that very high blood pressures (as in DBP greater than around 115) are dangerous–at least since the 1920s. However, before at least the postwar era and the development of hexamethonium, hydralazine and reserpine, the only treatments were strict sodium restriction, sympathectomy, and pyrogen therapy. Drug therapy was limited pretty much to sodium thiocyanate, which had a lot of side effects, so many that few patients could handle them.

    I do find it of interest, though, that blood pressure was not routinely measured as a vital sign in this country until the early 1900s. This was after the invention of a convenient sphygmomanometer and its popularization by Cushing.

  15. cervantes says:

    HIV testing is less problematic than others because it is extremely specific and confirmatory testing is done routinely.

    Blood pressure screening is essentially a no-cost test since it happens on routine visits and has no consumables. Alas, it’s not very specific — white coat hypertension and all that — but you could certainly set a threshold for treatment that’s high enough you know it will be cost effective, especially given generic drugs with little to be concerned about in terms of side effects. However, setting just the right threshold for treatment is very difficult, and involves a lot of variables including values.

  16. S.C. former shruggie says:

    I just wanted to say this is an excellent post, and so are the comments. Thank you! I love discussions like this as much as I love my biology classes!

  17. Harriet Hall says:

    @ michelinmichigan,

    The USPSTF doesn’t recommend for or against screening for glaucoma with intraocular pressures tests. See

  18. weing says:

    “While it is certainly true that treating “prehypertension” (DBP between 80 and 90) is a recent phenomenon”

    I admit to treating prehypertension. My treatment is TLC. Not tender loving care but therapeutic life changes.

  19. Harriet Hall says:

    @ David Gorski,

    According to Welch, FDR’s BP was 200/100 in November 1944, and “it is unclear whether his doctors recognized it was a problem.”

    This article mentions his doctor’s statement that Roosevelt’s BP was normal for a man of his age. We can only guess whether he really believed that or said it for political reasons.

  20. Harriet Hall says:

    @ Jan Willem,

    No, the book doesn’t mention psychiatry; but the same principles would presumably apply.

  21. David Gorski says:

    One thing that should always be remembered when screening a population for any condition is that the more you look, the more you will find. Always. That’s because nearly all conditions can be preclinical or have preclinical phases. In the case of cancer, which I know best, pretty close to everyone over the age of 40 probably has some malignant cells in one organ or another, but most people will not die of cancer. The classic example is prostate cancer, where at least 80-90% of men over the age of 80 have microscopic foci of prostate cancer, but the vast majority of men do not die of prostate cancer.

  22. David Gorski says:

    This article mentions his doctor’s statement that Roosevelt’s BP was normal for a man of his age. We can only guess whether he really believed that or said it for political reasons.

    Now that’s clearly covering up, particular the part later in the article you cite where Roosevelt’s blood pressure was listed as 210/120, which was recognized as early as the 1920s to be a dangerously high blood pressure. My guess is that the doctor knew; he was simply doing what was done so many times before and downplaying the seriousness of the President’s condition. Either that, or he really was clueless. The rest of the article points out that FDR had clearly developed signs of congestive heart failure.

  23. lizditz says:

    Jan Willem Neinhuys:

    Does the book say anything about overdiagnosis in psychiatry? I think that ADHD is seriously overdiagnosed, autism has been stretched to autism spectrum to the extent that children that are able to concentrate for a long time on something and children that are naturally very active both are considered deviant.

    As far as pediatric ADHD diagnoses go, it is likely that some population segments are underdiagnosed and undertreated. I am not referring just to medication, but other methods of treatment such as behavior management and parent training.

    I do agree that school settings can penalize normal behavior.

  24. Harriet Hall – Thanks for the link – it is quite interesting and I will pass it on to my husband.

    I’m sure that you are aware, though, that as a patient, as complex as the issue is, you still have to decide to either to test or not test.

    To test, or not to test: that is the question:
    Whether ’tis safer in the mind to suffer
    The pokes and prods and spend an outrageous fortune,
    Or resist imagining a sea of possible troubles,
    And by ignoring, end them?

    or possibly succumb to them. :)

    Not always an easy decision.

  25. wales says:

    Thanks HH for a well written review. I have been an admirer of Gilbert Welch for some time and look forward to reading the new book. I also recommend two other’s he co-authored: “Should I Be Tested For Cancer? Maybe Not and Here’s Why” and “Know Your Chances: Understanding Health Statistics” as well as “Worried Sick: A Prescription For Health in an Overtreated America” by Nortin Hadler.

    My jaw dropped this morning when an acquaintance praised his healthcare HMO by saying “they give CT scans at the drop of a hat, no waiting!”

  26. cervantes says:

    Michele — If you’re lucky, you’ll have a health care provider who understands these issues and can discuss them with you without pushing tests on you that ought to be your choice. My doctor gets all huffy when I refuse prostate cancer screening but in fact the consensus guidelines say I should make my own decision.

    If you have the right doc, you can go over the considerations in your case — and if your risk factors for something are higher, it makes more sense to be screened — and make an informed, sensible decision. Unfortunately too many doctors a) don’t understand Bayes theorem (really — they have no idea and think that a 90% specific test means if you test positive, there’s a 90% chance you have the thing); b) don’t understand the down sides of screening and c) (it’s usually an unconscious influence but it’s real) they make money by doing stuff.

    This is a major problem with medical education, and the way the health care system is organized and financed.

  27. marilynmann says:

    I had one of those Lifeline (or a similar company) screenings at my health unit at work years ago. They told me I might have a leaky mitral valve and recommended I see a cardiologist. The cardiologist did an EKG and found a mild abnormality. He told me he “had to” order a stress test because of the EKG, even though it might be normal for me. I ended up have a stress test and an echo (if memory serves). Conclusion: everything normal. So one screening test can lead to more tests. Everybody orders all the tests necessary to CYA. I wasn’t really harmed by all those tests, but sometimes people *are* harmed.

    Even when people have symptoms, it doesn’t always make sense to order all possible tests “just to be sure.” There was an article by researchers at the Cleveland Clinic in Archives of Internal Medicine recently where a low risk woman had chest pain and through a series of unfortunate events ended up having to have a heart transplant.

  28. Jan Willem Nienhuys on spectrum disorders in psychiatry: “I think that ADHD is seriously overdiagnosed, autism has been stretched to autism spectrum to the extent that children that are able to concentrate for a long time on something and children that are naturally very active both are considered deviant.”

    A disorder by definition causes a problem. Someone can be on the autistic spectrum and be severely disabled or not at all disabled. If someone is not disabled — they just have really good concentration — then they have no disorder. They might still be on the spectrum, but there’s nothing to diagnose.

    Similarly for ADHD. Some people are clearly very disabled. Others just need to keep busy all the time and perform very well. They don’t have a “disorder,” but they may still benefit from self-management techniques.

    “Screening” isn’t really the issue with autism or ADHD. Someone would “screen” for these disorders by asking if so-and-so is having problems. If they are, then by definition they are having problems. Any “treatment” is not for something that might happen in the future but for something that is definitely happening now.

    Screening for bipolar disorder is more like the physical conditions that Harriet Hall was mentioning. It may have no signs or symptoms in early childhood, some nonspecific signs and symptoms in late childhood or adolescence (young person is stormy and ambitious), become striking in young adulthood (young adult is dramatically moody and passionate; may be very high-achieving; may have interesting brief psychotic episodes) and become a clear disorder by the early thirties (person is irritable, miserable, never achieves anything again and begins a fairly deliberate course of drinking themselves to death).

    At least, this is one path. Assuming this path, what would be the benefit in screening for potentially bipolar eight-year-olds? Adolescents? College students?

    Treatment for bipolar adults relies heavily on medication that dulls the brain. You don’t want little kids taking that stuff. You don’t want to pathologize emotions or deny young people the opportunity to learn to manage them. But medical treatment appears to have some prophylactic effect such that intervening to prevent early mood episodes reduces the frequency and severity of later ones. So you don’t want to hold out until someone has completely ruined their life before offering treatment. You want to identify people who can benefit from treatment now to avoid a worse condition later. Not an easy balance, which is something like the dilemmas that Harriet Hall was discussing.

  29. SimonH says:

    I agree ADHD is grossly overdiagnosed. I see many children, especially boys, who are a bit impulsive, a bit overactive, a bit distracted, a bit oppositional at school and/or at home and I diagnose them as normal. Many of these children are on psychostimulants. When certain conditions are medicalised, it teaches children and families that if there is a problem, then the solution is to take a tablet. This can impact on them in adult years, when they come up against the normal frustrations and disappointments of life and they feel they need to take medication.

    Re: Bipolar disorder in children. I have been a paediatrician in Australia for 15 years and have never diagnosed a child with bipolar depression. I have never heard of a child diagnosed with this disorder my any of my colleagues either.

    There is a broad range of normal behaviours in children, impacted on by their development and their family situation.

    Very few children need any pharmacological intervention for behavioural issues, as many of them have normal variant behaviour.

  30. Jann Bellamy says:

    Very informative post. This must account for a huge chunk of health care spending. Any estimates on how much in the book?

    Feeding on the screening hype is the “walk-in lab,” no doctor’s order needed. We have one in Tallahassee in a shopping center, called “Any Lab Test Now.” It’s a franchise operation. You can get the “Fatigue Panel,” which includes testing for Epstein-Barre and Lyme disease, and the “Cancer Panel,” to name a couple. It would make a good SBM post.

  31. Paddy says:

    One interesting feature of overdiagnosis is how it skews cancer survival (or cancer mortality) estimates.

    “Success” in a health system’s management of cancer is primarily defined by its 1-year and 5-year from diagnosis survival rates. If you diagnose more patients early, you’ll increase these for three reasons, only one of which is of clear benefit to the patients. I’ve enumerated this with theoretical patients called Adam, Brenda and Carlos:

    A) Some patients will do better because their cancer is caught early. Adam, diagnosed through screening in January 2010 might have time to be treated for a cancer that would kill him if not spotted until July 2010.
    B) Some patients will die of cancer at the same age as they would have anyway, but have had a little longer knowing that they had it. Brenda’s survival to one-year will be a success if she’s diagnosed in January 2010 rather than July 2010 before going on to die in February 2011 anyway.
    C) Some patients will be treated for early indications of cancer that would never have amounted to cancer itself. Carlos’ survival to one year is a success if treated after being diagnosed based on screening in January 2010, but he doesn’t enter into the data if never detected as having “cancer”.

    As stated in the article, you may have very little or no idea who falls into which group. But what you do know is that this indicator will dramatically increase if you diagnose more patients early, which is partly why various levels of management in health systems from governments down to individual hospitals, wanting to perform well on this indicator, will have a considerable incentive to diagnose earlier and earlier.

    Overall, however, if you want to know whether people are truly better off being diagnosed early, you’d need to know how many patients are in Adam’s shoes, and how many in Carlos’, and, in particular, how many people in Carlos’ shoes would be severely harmed by undergoing unnecessary treatment. (Leaving the question of whether Brenda is better off or not moot).

  32. Jann Bellamy says:

    One more test of note from “Any Lab Test Now”
    ATTN: Dr. Gorski

    “Introducing the BT Test® – A Blood Test for the Early Detection of Breast Cancer
    All women over the age of 40 should get the BT Test® every year. And now you can get the BT Test directly at ANY LAB TEST NOW® across the country.”

    It costs $299.

  33. On ADHD – ADD diagnoses or over diagnoses. I have a couple of nephews that have ADHD – ADD diagnoses whom medication and therapy seems to genuinely have enabled them to lead a normal life. They both had some pretty severe dysfunction across different aspects of their life, schoolwork, home life, socializing and making friends as well as issues with impulsive, dangerous behavior. I can genuinely see the value of the diagnoses.

    On the other had, I have wondered with, some situations, if the dysfunction present is not sometimes more brought on by unrealistic expectations, classroom approaches or parenting skills.

    Example one:

    My mom was a teacher and principle for many years. While in high school and college I would occasionally help out in class. This was in a time when desks generally sat in rows facing a chalkboard, which was cleaned at the end of the day. Decorations where a bulletin board with projects and occasionally a few posters or a boarder with spelling words. Homework was seldom given.

    Now, I sometimes help out in my children’s classrooms. At my daughter’s school, desks are kept in pod of 6 or 8 with the desks next to each other and facing each other*. The walls, some windows and most of the whiteboard are completely covered with posters, flyers, spelling, math, literature, science, slogans, songs, etc. My daughter’s class is feisty and social and they are often chatting with their neighbor (whom they are shoulder to shoulder with) or the person facing them. The teacher often has to raise her voice and reprimand them to get their attention. When my daughter gets home from school, we always used to do homework (often 1/2 to 1 hour of work). This was often a complete frustration, due to how little focus she had.

    After a few months of this, and viewing the class in one of it’s more chaotic mornings, it occurred to me that my daughter was possibly over stimulated from a day in the classroom and needed time to unwind before homework. Things seemed to have improved since then.

    From my experience with this situation, I wonder how many kids that have normal attention, but are somewhat sensitive to stimulation would be pushed into a ADHD diagnoses.

    There is also the question of ADHD misdiagnoses. hyperactivity and the appearance of inattention is also a symptom for other conditions in children, anxiety, sleep apnea and hearing loss to name a few. I having seen a few forum postings from parents who have children with hearing loss complaining that teachers want their children diagnosed with ADHD, at the same time that they ignore the accommodations (preferential seating, FM sound systems use) that help the child hear adequately in class.

    This is not to slam all teachers, I’ve met a lot of smart, committed, talented teachers, only to suggest that schools can possibly have an effect on the number of diagnoses.

    *Having worked in office setting for many years, I would suggest that most adults don’t like to have open seating, facing coworkers because they find it hard to focus.

  34. SimonH says:

    The other issue for primary school-aged boys especially is that sitting in class all day is boring, so they get easily distracted, and the teachers get annoyed, and they tell the parents to take the child to a paediatrician to see if they have ADHD

  35. marilynmann says:

    Addendum to my previous comment:

    Here is the study I mentioned:

    Left Main Trunk Coronary Artery Dissection as a Consequence of Inaccurate Coronary Computed Tomographic Angiography
    Matthew C. Becker, MD; John M. Galla, MD; Steven E. Nissen, MD

    Arch Intern Med. Published online December 13, 2010. doi:10.1001/archinternmed.2010.464

    A 52-year-old woman presented to a community hospital with atypical chest pain. Her low-density lipoprotein cholesterol and high-sensitivity C-reactive protein levels were not elevated. She underwent cardiac computed tomography angiography, which showed both calcified and noncalcified coronary plaques in several locations. Her physicians subsequently performed coronary angiography, which was complicated by dissection of the left main coronary artery, requiring emergency coronary artery bypass graft surgery. Her subsequent clinical course was complicated, but eventually she required orthotropic heart transplantation for refractory heart failure. This case illustrates the hazards of the inappropriate use of cardiac computed tomography angiography in low-risk patients and emphasizes the need for restraint in applying this new technology to the evaluation of patients with atypical chest pain.

    * * *

    In the case report above, the woman had recently started an exercise program and had nonexertional chest pain that was aggravated when she raised her right arm. The pain was most likely of musculoskeletal origin. To reassure her, her doctor ordered a CT angiogram to reassure her. The results of the test were unclear, so she was sent to the cath lab, where she unfortunately suffered dissection of the left main coronary artery (a rare but known possible complication of cardiac catheterization), leading to an emergency bypass surgery and eventually a heart transplant. These unfortunate events could have been avoided if her doctor had not ordered the original CT angiogram.

  36. JMB says:

    When the definition of overdiagnosis becomes so broad, it is easy to make sweeping statements about the problems with overdiagnosis.

    Overdiagnosis in cancer/disease screening is clearly different from overdiagnosis in the mental health spectrum.

    Overdiagnosis in cancer/disease screening is also different with different disease processes.

    For example, consider mammography/PSA screening. In the case of breast cancer screening, few here would argue that a palpable breast lump needs to be evaluated. So the best definition of overdiagnosis due to screening mammography is the excess cases of cancer discovered by screening mammography versus usual (no screening mammography) care. I call that the best definition because the woman is faced with the decision about whether to undergo screening mammography. If she chooses not to undergo screening mammography, there will still be cases of breast cancer discovered by feeling a lump. If we use a definition of overdiagnosis which is the percentage of breast cancer cases that undergo treatment in which the treatment either did not save the patient from dying from breast cancer, or that the patient did not die from breast cancer, then overdiagnosis will occur with breast cancer discovered when the women or her doctor discovers a lump. That difference in definition is one of the reasons why percentage of overdiagnosis varies in the literature. So it is dishonest to tell a woman that if she does not undergo screening mammography, she will avoid the problems of overdiagnosis. If you want to know the figure of excess breast cancer cases from the randomized clinical trials in the group of women undergoing mammogram screening for 15 years, then 5 to 10% is the more common figure. If you want to consider the excess cases and the cases in which treatment probably didn’t make a difference, then the combination adds up to the 25 to 35% cited commonly in some other articles.

    So if you use the definition of overdiagnosis suggested in the review of this book, then the most honest statement to a woman receiving informed consent for screening mammography is that overdiagnosis occurs when a breast cancer is discovered by either finding a lump, or detecting a lesion on a screening mammogram. 90% of lesions detected by mammography that prove to be an invasive carcinoma, would have eventually been discovered after the lesion grew and became palpable. 10% of lesions detected by mammography, and shown to be invasive cancer, would not have been discovered within 15 years, and either wouldn’t grow, or spontaneously resolve.

    In the case of PSA screening, we do not have a similar scenario to women finding a lump, because most men don’t feel their prostate. Consequently, a broader definition of overdiagnosis may be useful in informed consent.


    It annoys me when people come up with new definitions, and act like they have made a new scientific discovery. The problem of overdiagnosis/overtreatment has been known for years (50+). Overdiagnosis is a factor in the calculation of benefit, overtreatment is a factor in the calculation of risk. The data and mathematics has not changed, just the verbiage/verbage.


    The solution to too much information is not burying our head in the sand and ignoring it. The solution is to improve our techniques of handling the information (increasing our channel capacity as such).


    In addition to those mentioned, how about a national database of “incidental findings” that could provide data for ongoing epidemiological studies?

    That’s a great idea. I hate incidentalomas. It makes me feel like the guy on the team that decides to go play tiddlywinks, while everyone else on the team is sweating on the field.



    It may be more than CYA medicine. The doctor also profits by ordering the tests.

    Healthcare uses loss leaders just like consumer stores. Attract the customer with a low cost (or free) screening test, and make the profit off the more expensive tests to follow. Most health fairs are full of loss leaders.


    we need to know if a cancer is present that is likely to progress and harm the patient. And so far we have no way of distinguishing which these are.

    I think we know a lot of this, it is just the there is no overall answer, just a lot of specific answers based on cell type and tumor genetics.


    999 out of 1000 women do not benefit from mammography.

    Maybe if the woman only has one mammogram and we are only talking about prevention of death, that statement is true (depending on her age). But if a woman commits to yearly screening mammography from age 40 to 74, that figure is inaccurate. The benefit of early detection is not only prevention of death, but also less invasive treatment.

    250-500 will have a false alarm and half of these will be biopsied

    This guy comes up with strange stats! Out of 1000 patients in a typical screening mammography practice, 50 to 80 will be recalled for additional views, 10 to 25 will be referred to biopsy, and 4 to 8 will be found positive for either invasive cancers or DCIS. This does vary based on the age distribution. 15 to 25% of the screen detected breast cancers can attribute a cure to early detection.

    Sound like this guy likes to cite benefit from having only one mammogram, but risks from 10 mammograms. Or maybe he’s talking about the risks and benefits in the age group 40 – 50 only.

  37. Barry2 says:

    Here’s my family’s anecdote about incidentalomas. The beginning is so convoluted I forget the exact details in this paragraph, but it goes something like this: My mother was having a urinary problem, and a scan (ultrasound?) showed no problem – except for something suspicious (in the bladder?). The newly detected problem was then scanned. It, too, turned out to be normal, except that something abnormal was now seen in the liver. A biopsy of the liver abnormality revealed that it was a carcinoid.

    Results thus far: by incredible coincidence, a cancer is discovered. Patient and family are upset and alarmed.

    We go to a local oncologist, who informs us that carcinoids are an “x-ray disease” – they grow very slowly, and normally they’re only found in autopsies of people who have died of something else. We’re told not to worry about it. I notice that the doctor sounds a bit stupid, but hey, he’s a doctor, I’m not, and this is great news.

    Results thus far: we’ve gone through a cancer scare, but fortunately it was nothing – truly an incidentaloma – at least if we don’t think about it too closely.

    However, I’m left with nagging suspicions, so a couple of months later I drag myself into a medical library. (This was before the web had really taken off). Never having taken an anatomy course, I don’t understand most of what I’m reading, but I do notice that the 5-year survival rate for carcinoids isn’t even close to 100%, and the literature is saying to always operate. I am so shocked at what I’m seeing that I start laughing.

    We manage to get to a rising star in the carcinoid field at a world-class cancer hospital. He orders more scans, which reveal an intestinal blockage and additional tumors in some lymph nodes. Apparently, my mother’s recent weight loss wasn’t really due to switching to a new type of bread, as she had assumed. (An interesting tie-in to the recent diet articles on SBM – some patients on new diets will lose weight for unrelated reasons.) And just to keep things interesting, the scans turn up another incidentaloma, this time a meningioma.

    We learn that carcinoids cause carcinoid syndrome once they reach some tipping point. The reason that carcinoid patients die of something else is that carcinoids poison their bodies.

    Results thus far: what a save!; don’t trust stupid-sounding people; me-nin-gi-o-ma?

    After major surgery, the surgeon informs us that he ended up having to remove the maximum number of lymph nodes that he could safely remove. I.e., if we had waited too much longer, the cancer might have spread to more lymph nodes. The cancer eventually starts to slowly come back a few years later, but by then, an experimental medicine has arrived (octreotide), and my mother has been doing very well on it, with no detectable cancer.

    As for the meningioma, it so far seems to be harmless, and fortunately no other incidentalomas have turned up in the many follow-up scans.

    Of course, this is just an anecdote, one that happens to have a successful conclusion. We had a top oncologist, a top medical center, and a top surgeon, which not everyone has access to. Even with these advantages, things could have gone wrong during surgery; in fact, my father noticed before surgery that my mother was being prepped with the wrong solution. But this anecdote does illustrate a couple of points:

    1) Some problems, e.g., carcinoids, often aren’t detected until it’s too late. Following up on incidentalomas is one way to find these problems in time. Perhaps routine screening for these problems would be a better way, but I wonder about the cost, practicality, and safety of submitting a large number of seemingly healthy people to a variety of scans and their associated chemical cocktails. So, maybe following up on what you do happen to find isn’t such a bad idea.

    2) Patients who seem to be healthy might actually have signs of illness that are being missed or misinterpreted. As far as I know, my mother’s primary care doctor never sent her for tests when she had an unexplained loss of weight. And my mother had been reporting digestive problems for years that are consistent with carcinoid syndrome. Do doctors really know when they have all the information they need about their patients? Why ignore information that might turn out to be useful?

    Of course, these points apply better to cancer than they do to many other problems, but I hope this anecdote helps to illuminate the subject of overdiagnosis.

  38. SimonH on bipolar disorder in children:
    “I have been a paediatrician in Australia for 15 years and have never diagnosed a child with bipolar depression. I have never heard of a child diagnosed with this disorder my any of my colleagues either.”

    Yes, exactly. Even if in retrospect an adult could look back and recognize the early stirrings of a mood disorder, how would anyone caring for the child have known that’s what it was?

    In my own case my first towering rage happened the summer I turned 11, stunning both me and the friend I was with. No special reason to think it was anything other than the stirrings of puberty. I became suicidal the first time a few months after my fourteenth birthday; the second time a few months after that. By the time I turned 17 I was used to the idea of being suicidal one or twice a year and knew it would pass if I could just sit tight and hang on.

    I may have had hypomanic episodes as well, but I didn’t recognize them as anything other than feeling the way I should feel: confident, creative, ambitious. I’ve never asked my parents. I was more interested in sex than my peers at least since I was seven (I know because my peers complained) but kids are sexual.

    How would a pediatrician have known I would go on to develop Bipolar II? Nobody even knew I was suicidal.

    By the time I was in my early twenties it was clear that I was never going to fulfill my early promise: projects were routinely compromised by depressive episodes. My grades in high school were very mixed. I never graduated from college. I put it down to lack of discipline. (I am not, nor ever have been, a drinker.)

    I was denied antidepressants as late as age 31, when I asked for them for the first time, because I was not typically depressed. (“Not depressed” is the way my GP put it.) When I was 35 my psychiatrist offered me a mood stabilizer; I didn’t accept her offer until four years later, when I accepted that the benefits of hypomania were gone forever and that I was stuck with just mixed states.

    Any screening test that would have identified me as potentially bipolar when I was seven would have had such a huge false-positive rate as to be very harmful.

    I’ve thought about it, and while I would like to have been able to succeed in school I have never been able to imagine an intervention that would have had more pros than cons and that I could have accepted – except maybe a study skills class? With a really good tutor? Dunno. This is probably the age where screening tests would move into a grey area.

    Any screening test that failed to pick me up by the time I was in my late twenties would be so lacking in sensitivity as to be utterly useless. (In my experience, depression screening tests were utterly useless for picking up my bipolar depression. What’s the point of asking someone if their appetite has changed in the last two weeks when they’ve been cycling for years and no longer have a baseline?)

    Yes, pediatricians do have bipolar patients. My point is – how would they know? And what would be the point of knowing?

  39. BillyJoe says:


    I have considered some specific issues. For example, my father died of lung cancer shortly after he was diagnosed with prostate cancer. His specialist recommended that all his sons be tested. Shortly after this I came across an article that was highly critical of prostate cancer screening. This was the start of rather long process of finding out some facts. My final conclusion was that prostate cancer screening is not a useful activity and therefore I declined to participate. The fact that my father had prostate cancer did not change my decision because about 70% of men his age have prostate cancer anyway.

    I have not gone into other areas in as much detail but, on the basis of what I have found out, I have made a decision to decline all screening tests. An important consideration for me is the number needed to test to get a positive outcome (for prostate cancer this is about 440!). Another important consideration is the number needed to treat. For example, if the number needed to treat high cholesterol to prevent one death is unacceptable to me, then there is no point in screening for high cholesterol. Odds of 1 in 1000 per year is not high enough for me to take a pill every day so I don’t get tested. I figure my hour walk/run every morning and 2-4 hour walk/hike/run through the hills every Sunday would probably swamp any benefit from the tablets. And I look forward to that.

    One small worry is that my sister had a heart attack at age 40. However, she is morbidly obese, has diabetes, smokes 25 cigs a day, and never exercises. I also have a brother would had a heart attack at age 50, but he is also a heavy smoker, obese, never exercises, and has a type A personality. I am the exact of opposite in every way to my brother and sister. And, on the other hand, my other sister and three brothers are as fit and healthy as I am, as is my mother who is in her 80s.

    But, if I ever get chest pain on one of these long walks….

  40. lillym says:

    Re: Bipolar disorder in children. I have been a paediatrician in Australia for 15 years and have never diagnosed a child with bipolar depression. I have never heard of a child diagnosed with this disorder my any of my colleagues either.

    You may not have seen it but that doesn’t mean it doesn’t exist. Or that children aren’t being misdiagonsed.

    I wasn’t diagonsed with bipolar disorder until I was 27 years old and hospitalized for a serious suicide attempt. The doctors there asked me questions no one had thought to ask before, basic stuff that if anyone had asked me in the past 17 years I could have told them what was going.

    If anyone had properly screened me they would have figured it out when I was 9 or 10 because after talkign to my current doctors I was pretty much a textbook case.

    I could have been on medication and in therapy where I could have dealt with how to control my poor impulse control, how to deal with the rages and anger, how to handle interpersonal relationships instead of spending most my life thinking (and being told by professonals) if I really wanted to I cuold change. That this was my fault and a lack of will power.

    I started seeing a therapist when I was 8 because of problems I was having – I was immature for my age, prone to day dreaming, insonmiac and having problems socializing. I started having panic attacks, uncontrollable rages, depression, etc when I was 10. School was a nightmare for me, both in terms of being unable to navigate the social structures but also the fact I would start to panic (only I didn’t know what was happening) over almost anything, I’d cry at the drop of a hat, it’s too long to explain everything.

    This didn’t happen every day, that didn’t start until I was closer to 12 and I started having panic attacks multiple times a day, I started gettign depressed, I’d fly into rages, I’d get too excited and happy about things. I’d make grand plans and then nothign would happen to them. I had horrible impulse control which led to me saying or doing things that weren’t appropriate.

    Oh and I was seeing a psychologist. I didn’t know what a panic attack was so I described it as it feeling like having a heart attack. Only to be told, “kids don’t have heart attacks.” I started thinking about sucide more frequently – daily instead of weekly. I could barely function in school. I was told, if i loved my mothe renough I wouldn’t act the way I did. If I wanted to be like all the other girls I could if only I tried hard enough.

    Some days I would be so tired and depressed I couldn’t get out of bed. I remember feeling like I had a ton of weight pressing down on me and doing anything was just too much. Taking a shower was exhausting, doing anything was overwhelming.

    And then things would be fine and I would be great.

    There were times when the thoughts would race in my head so fast I couldn’t keep up. I talked so fast people could barely understand me.

    Then there were the times when I would sit or lay in one place, unmoving, for hours, because the depression left me empty.

    It wasn’t until I was 15 and barely functioning, that my mom on wits end talked to a new doctor. He said “bring her in onw” and she did. He did a simple screening – you know rate your moods for 1-5 with 1 being worst and best and I bottomed out on every score. The dcotor told me he thought I had depression, a real illness with no magic cure, but medication and therapy that could help. I cried in relief and disbelief. Someone believed this was beyond my control and I wasn’t just “willful”. My parents always believed that but no one else.

    There was talk of admitting me to the hospital, instead I was sent to a therapist.

    I wish I’d been hospitalized. Instead I got diangosed with depression – because hey I was so depressed I wanted to die but didn’t have the energy to do anything about it.

    I cried at ever session unable to talk for the first 3 weeks (twice a week) because it was just such a relief to be believed and have a safe place.

    Anyway, after years of antidepressants not working (they made me hypo manic and then depressed) the therapst said I had “atypicall depression” and started me on lithum to handle my rages. Then on an MAOI which kept me stable for a few years.

    Now that I’ve made this long story – I wasin and out of therapy and on and off anti depressants (stopped taking MAOIs) until I was 25 when I started a long downward spiral of depression and mania until my suicide atttempt.

    I lost most of my life to this disease. Most of my childhood, all of my teenage years, my twenties with my 30s being dedicated to starting over from scratching and building my life up. I’ll be honest I have very bad social skills, friendships and relationships are like a forgien language I never learned.

    Being diagnosed as a kid could have saved my parents and my siblign heartache, money, time. And stress! My god the stress my mother was under and didn’t have to be. The money thrown away with bad therapists and anti depressants taht never did any good.

    I went through mental hell and I didn’t have to.

    But what I went through did save one of my cousin’s kids that same hell. When he started having problems aroudn 10 they knew to go get him screened for bipolar disorder. And he has it and he has meds and therapy and going through a better life.

  41. megancatgirl says:

    As for ADD/ADHD, unless you’re a psychiatrist, you really have no standing to refute the diagnoses of trained practitioners, and it’s a bit presumptuous to assume that you know better than actual doctors. There almost certainly are wrong diagnoses, just as there are with every single disease. But it really isn’t a case of a kid who isn’t perfectly passive being pumped full of drugs to make them more docile. This really goes along with mental illness denial in general, where people think that these kids just need a good spanking or stricter discipline to behave. The truth is that they want to behave but have trouble doing it because of their disorder. Medication isn’t always used, and when it is, it’s generally not done exclusively. If you don’t know these kids personally and have an educated understanding of the disease, you have no right to assume their diagnoses are wrong.

  42. windriven says:

    “leading to an emergency bypass surgery and eventually a heart transplant. These unfortunate events could have been avoided if her doctor had not ordered the original CT angiogram.”

    But another patient presenting with the same symptoms and not having an angiogram might have gone on to have a fatal MI leaving an angry family suing for malpractice.

    Physicians are caught between the rock of sometimes unrealistic patient expectations and the hard place of overdiagnosis.

  43. megancatgirlon “As for ADD/ADHD, unless you’re a psychiatrist, you really have no standing to refute the diagnoses of trained practitioners, and it’s a bit presumptuous to assume that you know better than actual doctors.”

    Actually, as a layperson, I think I have every standing to question the diagnoses of a trained practitioner. That is what patients and patient’s parent’s should do. If the practitioner can not or does not want to answer the questions, that is a problem. And no, I don’t just trust all practitioners just because they have the right degree. Sometimes they make mistakes, some of them are just not very good at what they do.

    In fact, that is the whole topic…whether trained professionals may currently have a tendency to over diagnose in the general populace. If we ask that question with cancer and heart disease and it’s not “cancer and heart disease denial” Why can’t we ask that question with psychiatric disorders?

    And I don’t think that the automatic answer is yes on any of those questions, by the way.

    Rather than saying ‘don’t ask the questions’ Why not just make a good argument that ADHD or other psychiatric or developmental disorders aren’t over diagnosed or make an argument that the risks from under diagnoses is more dangerous than the risks from over diagnosing, or… something else.

  44. megancatgirl,


    When I was 44, having been on meds for about ten years at that point, my psychologist (am leery of psychologists in general but I quite like her) said something offhand like “… and then we can get you off meds.”

    I was gobsmacked. When I started seeing her we discussed goals of therapy. Her being nice to me was one. Me getting off meds was not. I had to remind her that getting off meds was not one of my goals. She was surprised: Whyever not? Doesn’t everyone want to get off meds?

    I explained carefully that I had lost the potential of youth to depression and that I had no intention of giving up my middle years to depression too. [Hey there, @lillym!] I went over my family history. I reminded her that my meds are prescribed by a psychiatrist who has known me since before I started seeing her [the psychologist]. I pointed out that I was not a child or teenager anyone needed to worry about starting on an over-medicalized approach to life. And I told her again how the shape of my life changed radically once I got access to appropriate medication — I am quite capable of doing lots of difficult things without a therapist that I couldn’t do without a psychiatrist. The issue is not insight.

    After I walked her through all the rationale for me using medication to manage a mood disorder she brightly agreed that ok, she appreciated that getting off meds was not one of my goals, that she hadn’t known all the detail of my family history, and congratulated me on my positive emphasis on getting the most out of my middle years.

    She’s a PhD psychologist and works in a hospital. Where on earth did she get the idea that she could simply deny the existence of mental illness and decide that my psychiatrist must be wrong? I had to fight hard to get where I am and having to justify myself to someone who was supposed to help me rankled. I still like her, and I understand that people make mistakes, but somehow I’ve never needed an appointment with her since. That was two years ago.

  45. RE ADHD, which seems to be the whipping boy of mental illness denialists, I’ll go back to the definition of a disorder: it’s a problem. ADHD is a disorder, but attention deficit and hyperactivity are spectrum traits. You can have really bad, disabling ADHD or you can just be very busy. There doesn’t have to be a downside of recognizing spectrum traits in anyone, or noting that the spectrum traits are pronounced enough to be a disadvantage in certain circumstances.

    What people really seem to be worried about is not really overdiagnosing ADHD. Any kid with a diagnosis is by definition someone who is already experiencing a problem of some sort. Putting the ADHD label on the problem didn’t create it.

    The real concern people seem to be having is that kids without a medical condition might be taking Ritalin. A person can have ADHD and not take Ritalin. My understanding is that standard interventions in ADHD are behavioural first, and if the behavioural interventions don’t help enough and the disability is severe enough then medications are added as well.

    A kid who just likes to keep busy would presumably do fine with an ADHD label (or identified attention deficit or hyperactivity spectrum traits) and behavioural interventions. If those interventions help, what’s the problem? Overdiagnosis did not create an issue where one did not exist before.

    In the case of cancer, one can identify something on a scan that may or may not be cancer — or that if it is cancer, may or may not progress. When scans overdiagnose early breast cancer, they are identifying something that is not causing a problem at the time.

    When a kid has trouble in school and is put through a battery of tests by a psychologist who suggests mild ADHD and behavioural supports, the psychologist has not identified something that was not a problem before. The psychologist started with “trouble in school” and put a particular label on it.

  46. “When a kid has trouble in school and is put through a battery of tests by a psychologist who suggests mild ADHD and behavioural supports, the psychologist has not identified something that was not a problem before. The psychologist started with “trouble in school” and put a particular label on it.”

    My main concern is not so much the Ritalin and it is not wanting to deny help to a child who is having a problem with some area of function. My intent is not to say “That child just needs to try harder to behave or the parents need to be stricter.”

    My issue is that I’ve heard a few educators and therapists suggest that my son (and other children, mostly boys) have “problems with attention” or are hyperactive, without regard for the setting, what kind of communication they are using with the child, or the age appropriateness of the developmental task at hand.

    I just don’t believe that one can always assume that if a child is experiencing a problem, that it is caused by the child’s biology. It can possibly be caused by the setting. It could possibly be that the adult is experiencing a problem and finds it more convenient to view it as the child’s problem rather than their own.

    For example, I had one special education teacher suggest this after a test in which my son (almost 4 at the time) was bouncing and squirming and fooling around at the table. She completely ignored that fact that: she tickled him, as praise, every time he got a correct answer, she was sitting on his deaf side and he had to twist his head to the side to hear better, we had to wait in the middle of the test for 15 minutes while she attended to another student.

    Strangely enough, a similar test preformed without interruption a month later by a low key serious teacher sitting on his hearing side was treated with more seriousness and focus on my son’s part.

    Now, I believe, the testing and interviews used to diagnose ADD – ADHD are designed to account for this kind of variation of setting and interpersonal relationships. But I’ve gotten the impression from my pediatrician that the seriousness with which the local psychiatrists apply those test and interviews varies greatly.

    I’m sure I just sound like a defensive mom, but considering my children’s birth histories, I do try to be open to the idea that their attention or learning abilities may need additional help at some point. But, if that happens, I’ll be darned if it’s going to because a teacher or therapist just wanted their job to be a bit easier and they recommend a psychiatrist who’s results they like.

  47. lillym says:

    I thought megancatgirl’s comments were directed speficially at simon and his presumptions rather than a general you.

    Alison – I’ve had my fair share of crappy therapists (post diagnosis) but some good ones. I haven’t seen one in awhile. I was seeign a great therapist, but I kinda messed things up and was embarassed to go back. but it’s so hard to find a good therapist, if I can find someone who takes my HMO, I also have to find someone who is taking new patients. The last therapist I saw was hung up on me the idea I was molested as a child. (I never indicated that I was and I haven’t been, but somehow the fact I had bladder reflux as a toddler -and bladder control issues and then night mares when my life was so bad go her on that train of thought.)

    Luckily my shrink has been a rock, his goal was to get me on as few meds as possible. So I got off the mood stabilizer and just take Seroquel.

    Oh and Strattera. I have ADD too. Which wasn’t diagonsed until a couple of years ago. I was talking to some people with ADD and realized that mixed in with all the other stuff was that. My shrink was great . We tried out Strattera it really helped and while I still have difficulties it’s more things I could change behavorial.

    I really wish I could get up with someone that does Cognative Behavioral Therapy, but I’m hoping to move cross country within the next 6 months so maybe after I move I can work on that.

  48. micheleinmichigan,

    If a kid experiencing difficulties is put through a battery of tests by an experienced, well-trained child psychologist who characterizes the child as having ADHD, then the kid probably has ADHD. There is always more to it, but that’s probably an element.

    If a kid experiencing difficulties goes through the exact same process and is characterized as having a slight memory deficit but nothing else then the kid probably doesn’t have ADHD. The kid probably has a teacher problem and might also benefit from learning to use checklists.

    In both cases the child is experiencing problems. Neither example illustrates Jan Willem Nienhuys thesis of spectrum disorders in psychiatry: “I think that ADHD is seriously overdiagnosed, autism has been stretched to autism spectrum to the extent that children that are able to concentrate for a long time on something and children that are naturally very active both are considered deviant.”

    Children who are not having problems are not considered deviant.

    Children who are having problems AND who receive a formal diagnosis of ADHD or autism by someone competent to make one need more than someone’s “I think” to dismiss the diagnosis.

    An unprepared teacher with a too-large class size, an inappropriate curriculum and inadequate infrastructure is not competent to make a formal diagnosis of a developmental disorder, though they are certainly in a position to identify children who are struggling more than others.

    (Likewise, parents may not know exactly why such-and-such a teacher is causing such difficulty for their child, but they are very well-placed to know in whose class the child thrives.)

    My parents were under a lot of pressure to let my brother be evaluated by the school psychologist so they could have him put on ritalin. Instead, they elected to pay a lot of money to an independent psychologist to evaluate both my brothers. Yes, each had various types of deficit and the psychologist made various recommendations. No, neither had ADHD. No ritalin for them. Since my parents had to disappoint the school, my father volunteered as a teacher’s aide in my brother’s class two or three days a week. His reasoning was that if the teacher knew she was being watched that she would choose some other kid to pick on.

    If Jan Willem Nienhuys has evidence that ADHD and autism diagnoses have poor validity and that testers who diagnose ADHD or autism are less accurate than those who do not, we’re all open to hearing the figures.

    We all understand how tempting it could be for schools or parents to point to the child as the problem so as to avoid having to do anything different themselves. But “I think” doesn’t cut it when asserting that most children with certain types of diagnosis do not experience difficulties.

  49. Alison – I won’t side with or against “most children with certain types of diagnosis do not experience difficulties.” Because I do not know…I would suspect, the “most” is not the case.

    I’m more interested in how over diagnoses, misdiagnoses or under diagnoses can happen, how the medical field should prevent it and how to guard against it as a patient (or patient’s parent).

    For me, that doesn’t require that most children are mistakenly diagnoses or undiagnosed, only that enough children are effected to make it a cause for concern.

    Sorry to be annoying by not getting that the central issue was the “most” statement.

  50. micheleinmichigan,

    I think this is an issue of screening vs diagnosis, and diagnosis vs misdiagnosis.

    Kids with difficulties are tested with diagnostic tests and diagnosed or mis-diagnosed with ADHD or autism. That is, we already know there’s a problem, it’s a question of deciding which one is most important. If the diagnosis is made inaccurately/ sloppily/ unethically, that doesn’t change the fact that the presentation was of a child experiencing a problem of some sort. This isn’t analagous to the “overdiagnosis” of breast cancer by screening young, low-risk women with extremely sensititive scans. It would be more like a woman finding a lump in her breast and her doctor misdiagnosing it as a boil and then managing it inappropriately.

    Kids without difficulties may be screened with screening tests — say for scoliosis. They aren’t having a problem yet, but they may in the future. Same with adults and HTN: the problem isn’t in the present, it’s a theoretical future propblem. ADHD and autism aren’t in this category. Nobody takes a child who is happy and popular, does well in school, verbal, motivated, stays out of trouble, is good company for the parents and so on, to the pediatrician to say “can you please screen this child for ADHD and autism?”

    Bipolar disorder is more like something that could be in the “screening leading to risk of overdiagnosis” category. A child in the process of developing bipolar disorder isn’t as obvious as an adult with fully-developed bipolar disorder. They may not even be experiencing problems in the present. It’s possible that someone would want to identify kids in the process of developing bipolar disorder as young as possible so that intervention would slow or stop progress of the disorder. This is where you would run into “overdiagnosis” from screening. Kids who will never have problems may be identified as possibly going to have problems later.

  51. weing says:

    This book is about type 1 errors and gives a good description of them. Some of the commenters have raised the problem of type 2 errors. To me these have always been the monsters described by Homer as Scylla and Charybdis. We are caught between them. Even the great Odysseus was unable to pass through unscathed.

  52. “Kids with difficulties are tested with diagnostic tests and diagnosed or mis-diagnosed with ADHD or autism. That is, we already know there’s a problem, it’s a question of deciding which one is most important.”

    Okay – here’s an extreme hypothetical to try to illustrate what I’m getting at. A school district decides they are going to overhaul their program for academic excellence. They decide to have an intensive all day kindergarten and eliminate recess so that they can teach more material. They decide that children will be proficient in reading after 3 months. They will have math fact memorized by the end of the year. Children will not interrupt and will keep their hands to themselves. They have an excellent staff, so 75% of the class manages to meet these milestones. The remaining 25% are not meeting milestones and %20 are disruptive in class. The 20% are having difficulties. They can then be referred to a psychologist for testing and they will have the same risks/benefits of being diagnosed or misdiagnoses as all the other kids who are referred to a psychologist for testing.

    So the idea of both autism and adhd is that the overriding behaviors are supposed to fall outside of the range of “typical” child behaviors. But if the schools or parents create a narrower definition of typical learning/behavior, you are, in effect, using a screening tool that is more sensitive.

    As an aside, children are typically screened for autism at our pediatrician’s office. It’s not that obvious, but it goes along with all the developmental milestone questions they ask at annual appointments. I might be only aware of it because my son has the speech delay, so we usually hit the second tier questions that are triggered by the speech delay question.

    Just as a stream of consciousness follow-up, kids that have some delay or deficient and receive therapeutic or educational services are given lots of tests. At age 5 my typical daughter had received maybe two, six questions tests at her preschool (can she count to ten, ABC, skip, draw a circle, write her name.) By the time my son was 5 he had received at least 6 or 7 in depth educational, language, developmental tests, plus several 2-3 page behavioral questionnaire filled out by me. This is with an obvious/known diagnoses for his speech delay. Seems to me that some of those tests could easily turn up an incidental finding. But, of course a good test giver will recognize an incidental finding, unfortunately, they’re not all good. That’s probably irrelevant, though…just sometime I’ve been thinking on.

  53. “But if the schools or parents create a narrower definition of typical learning/behavior, you are, in effect, using a screening tool that is more sensitive.”

    So even if the diagnostic tests that follow this “screening” are appropriate, there will be more false-positives because the frequency of the targeted disorder is lower.

    Ok, I buy it.

    I was just naively thinking that the nice, non-misdiagnosing psychologist would raise an eyebrow and suggest that the child was suffering from a teacher problem and suggest a school-change intervention. But you’re right, changing the (diagnostically) tested population in this way would increase the number of false positives even with a nice, non-misdiagnosing psychologist. Which would in fact be “overdiagnosis” in the sense of the original post.

    Got it, thanks!

  54. wales says:

    A nice review of Welch’s book by another physician

Comments are closed.