Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 23

Thread: Seven Medical Myths Even Doctors Believe

  1. #11
    Quote Originally Posted by RehabRhino
    Wise

    Is there seriously no medical evidence for passive smoking causing cancer? That really surprises me. The whole UK smoking ban in public places is based on protecting employees from passive smoking.

    Is this a fallacy?
    Rehab,

    As Steven Edwards pointed out, there is a widely held opinion by researchers that passive smoking is harmful to health but the evidence that it can cause cancer is lacking.

    Wise.

  2. #12
    Quote Originally Posted by Steven Edwards
    One suggested article:

    39 out of 106 review articles concluded passive smoking wasn't harmful. 29 of those 39 studies had an author with ties to the tobacco industry -- only ten did not.

    Taking those 29 out: 67 out of 77 (87%) reviews agree that passive smoking is harmful.

    They don't all say it causes cancer, but they agree that it is detrimental to one's health.
    The study that raised the greatest ruckus is by Enstrom and Kabat in 2003 (Source). Published in the British Medical Journal, the study examined non-smoking spouses who were married to smokers from 1960-1998. They examined lung disease and cancer amongst never-smokers married to smokers compared to never-smokers married to never-smokers, finding no significant difference. This study used data from the American Cancer Society (ACS), who severely criticized the study: (Source) .

    The most serious design flaw, the society argues, is that there's an inability to distinguish people who were exposed to secondhand smoke from those who weren't. This is because:

    * Participants were enrolled in 1959, when exposure to secondhand smoke was so pervasive that virtually everyone came into contact with it, whether they were married to a smoker or not.
    * No information was collected on the sources of secondhand smoke other than spousal smoking.
    * No information on smoking habits after 1972 was included in the analysis, even though the observation period continued for another 26 years.
    * On average, participants were 52 years old when enrolled on the study. Many spouses who reported smoking in 1959 would have died, quit smoking or ended the marriage during the 38-year follow up, yet their surviving partners are still classified being passive smokers in the analysis.
    * Much of the follow up relates to older age groups where the effects of many environmental risk factors become less apparent.

    More than 50 studies on the health impacts of passive smoking have been carried out over the past 25 years, including a number of landmark studies providing significant evidence of passive smoking risks. Such work includes research by the International Agency for Research on Cancer (IARC), the Scientific Committee on Tobacco and Health and the ACS.

    Notable research includes a study published in the BMJ in 1997, conducted by Hackshaw and colleagues, which analysed 37 passive smoking studies and found a 24 per cent increase in lung cancer among people living with smokers. In fact, said the charity Action on Smoking and Health (ASH), "Tobacco specific carcinogens found in the blood of non-smokers provided clear evidence of the effect of passive smoking."

    Additionally, far more reliable data was obtained in the ACS Cancer Prevention Study II (CPS-II) study, which was about 10 times larger than Dr. Enstrom's work. They enrolled patients in the 1980s, when fewer exposures to tobacco smoke outside the home existed, and therefore far less "background noise", and follow-up has been much better (over 99 per cent). The results unquestionably show an increased risk of lung cancer and heart disease.
    I must say that if one were to look carefully at the so-called 50 studies that purport to show cancer risk, one finds that many of those studies have significant weaknesses as well. For all the criticism directed at the BMJ study, the data is actually not as bad as people argue. If the BMJ study had shown that passive smoking was harmful, I don't think that anybody would have complained about the study. The study assessed over 30,000 never-smokers who were exposed to second-hand smoke from spouses. While one can argue that the exposure was variable, the study did not verify that the smokers continued to smoke after 1972, and it is true that passive smoking was prevalent in society from 1958 to the 1980's, a reasonable argument can be made that never-smokers married to smokers are getting a much higher dose than never-smokers married to never-smokers.

    The problem with all these studies is that associations are being mistakenly proclaimed to be cause-effect relationships. There is no question that heavy smoking is associated with a greater risk of cancer but this does not necessarily mean that lower dose exposures have a linearly related risk, does not rule out the possibility of other causes, and does not take into account the likelihood that the risk depends on many factors including genetic susceptibility. Cause-effect relationships are also tenuous when the associated risks are low. The evidence for lower dose exposures, i.e. that a person who smokes less than one pack a day, causing cancer is less than 1%, requiring population studies of hundreds of thousands to detect sucgh differences with confidence. So, most of the risk was inferred by extrapolation from higher dose data ([url=http://www.mskcc.org/mskcc/html/12876.cfm]).

    One often encounters claims such as "87% of lung cancers are associated with direct or second hand smoking". Based on such claims, the average person might think that their risk of getting lung cancer, if they smoke, is 87%. But, it is far from that. For example, one screening study assessed the actual risk having lung cancer from heavy smoking in 27,000 subjects who are smokers and found that only 400 or 1.5% had suspicious lung lesions( Source).

    In the CARET study which investigated effects of beta carotene and vitamin A supplements in 18,314 subjects who are or were heavy smokers (>1 pack per day for 20 years), they found that the a 51 year-old woman who smoked one pack per year for 29 years and stopped smoking 9 years earlier had a 0.8% risk (less than 1 in 100) risk of getting lung cancer over the following 10 years. On the other hand, a 68 year old man who smoked two packs a day for the past 50 years and continued to smoke had only a 15% (1 out of 7) chance of developing lung cancer over the following 10 years.

    If smoking is the main cause of lung cancer, lung cancer should have declined linearly as the incidence of smoking declined. Indeed, there was evidence, during 1990-1996, that the male lung cancer incidence rates fell an average 2.6 percent per year. During this same period, lung cancer rates increased in women by 0.1% per year and 1.4% per year for mortality. However, in the past decade, the rates of lung cancer have more or less stabilized over the past decade even though the prevalence and history of smoking amongst adults have continued to drop dramatically (Source), suggesting that there are other causes of lung cancer that are now maintaining cancer rates.

    It is useful to look at cancer incidences closely because one expects a dramatic decline of cancer rates associated with cessation of smoking (Source). Here is a chart of age-adjusted cancer death rates in males from 1930 to 2002. The decline in lung cancer has not been that dramatic


    In contrast, the incidence of lung cancer in women plateaued, even though the number of women smoking is less than before


    http://seniorjournal.com/NEWS/Health...nueDecline.htm

    In 2000, Pelo and Doll (Source) reported that quitting smoking even late in life significantly reduced the risk of getting lung cancer. They found that those who had quit for 10 years had a third of the risk. Those that had quit for 30 years had only 10% of the risk. If this statistic were applied to the lung cancer rate and given a >50% decline in smoking rates over the past two decades, we should see a similar decline in lung cancer.
    In the 1950's, 82% of 35-59 year old men smoked in the United States. In 2006, only 24% of all men and 18% of all women smoked, for an overall percentage of about 21%. The people with the highest smoking tendency are men with high school GED diploma (>50%) and the people with the lowest smoking tendency are Asian women (4.6%). More educated people are less likely to smoke. More than half of the people who are alive and have smoked in the past have stopped (Source). In theory, therefore, we would expect lung cancer rates to fall by at least a half to two-thirds. It just has not.
    http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5644a2.htm

    Wise.
    Last edited by Wise Young; 01-02-2008 at 07:13 PM.

  3. #13
    If passive smoking actually does cause cancer why don't all of us born in the late forties and fifties have lung cancer? I mean ALL of us. Everybody smoked.
    And they smoked everywhere. Even your doctor smoked in front of you.
    And why don't the pets of smokers have lung cancer?
    Life isn't about getting thru the storm but learning to dance in the rain.

  4. #14
    It seems quite clear that smoking greatly increases the probability of getting lung cancer. But not to 100% (answering Lindox).
    Quite apart from scientific studies, a common sense extrapolation of that statement says that any exposure to smoke should also increase the probability of getting lung cancer, dependent on the frequency and intensity of exposure. But again, not to 100%.
    - Richard

  5. #15
    Quote Originally Posted by rfbdorf
    It seems quite clear that smoking greatly increases the probability of getting lung cancer. But not to 100% (answering Lindox).
    Quite apart from scientific studies, a common sense extrapolation of that statement says that any exposure to smoke should also increase the probability of getting lung cancer, dependent on the frequency and intensity of exposure. But again, not to 100%.
    - Richard
    Richard,

    Your "common sense extrapolation" is precisely the assumption that I think should not be made, in my opinion. Most biological relationships are not linearly but tend to have threshold effects. Therefore, extrapolation from high dose to low dose is questionable and has little biological basis. Let me try to convince you of this.

    Most biological substances fall into six categories.
    1. Irritant at very low to low doses and toxic at moderately high doses. These include dust, smoke, allergens and other substances that cause adverse responses in the body. However, the body accomodates.
    2. Therapeutic at very low doses, toxic in low doses. Many drugs belong in this category. Nicotine, aspirin, ouabain, and others "natural" drugs belong in this category.
    3. Beneficial at moderate doses, deadly at high doses. Alcohol belongs in this category. Many studies have suggested that ethanol when imbibed in low doses can extend life but when overused can be quite toxic.
    4. Essential for life at low doses, toxic at high doses. Vitamin A or carotene is quite toxic at high doses. But, at low and appropriate doses, it is essential for life. This is true of many vitamins.
    5. Nutrient at high doses, toxic at very high doses. Most food and liquids that we ingest belongs in this category. Even water, if drunk to an excess, can be quite toxic.
    6. Harmless at very low doses but toxic if allowed to accumulate. Lead, mercury, and other metals, for example, fall into this category. We cannot get rid of these efficient and they kill cells when they accumulate.


    So, most substances do not have a linear extrapolatable dose-toxicity curve. Most have thresholds for toxicity and may even be beneficial at lower doses. Almost none have linear toxicity or beneficial dose-response curves. The greatest threat of material that we ingest or inhale is when it is not efficiently eliminated by the body and thereby accumulates to toxic levels. Thus, heavy metals such as lead and mercury belongs in the sixth category of toxic materials that accumulate. Asbestos also belongs in this category. We just don't have good mechanisms for eliminating them and subsequent accumulation is very bad.

    Tobacco smoking has two components, smoke which falls into the first category and nicotine which is in the second. There may be some people who are very sensitive or may even be allergic to tobacco smoke and they probably should not be exposed. However, for a majority of people, exposure to tobacco smoke probably does not have a lasting effect. For people who smoke, the body accomodates to the smoke and the nicotine, to the point that there may be withdrawal symptoms. However, withdrawal is not the cause of addiction. The driving force of addiction in smoking is not from phuysiological withdrawal but the psychological need for stress reduction.

    It is remarkable how many people are 50-100 pack-year smokers and survive in relatively good health into their 80's. A pack-year is to smoke a pack of cigarette daily for a year. If you have smoked two packs a day for 25 years, you have incurred a 50 pack-year dose. I tried to give some examples of the risks of lung cancer from smoking in the post that I made earlier. I am sorry that I kept on editing it to try to make it more clear. For example, a 51-year old woman who had smoked 1 pack per day for 29 years (from age 13-42) and stopped at age 42 has only a 0.8% risk of developing lung cancer in the subsequent 10 years. Even the 100 pack-year man at age 68 has only a 15% chance of getting lung cancer over the following 10 years.

    Wise.
    Last edited by Wise Young; 01-03-2008 at 01:01 AM.

  6. #16
    Wise -
    I certainly take your point about nonlinearities and thresholds - if the body can dispose of something faster than it accumulates, then it may well be expected to have little, if any effect. And I am aware that there are precious few linear relationships in biological systems (well, maybe cardiac output = stroke volume * heart rate!)
    However, I did not mean common sense extrapolation in the sense of a linear extrapolation, but in the sense that I prefer to avoid exposure to smoke (to which I was referring in particular), trichloroethane, carbon tet, MEK, etc. (I imagine the latter 3 would fall into your category 6), because I have only a vague idea where my integrated exposure level lies with relation to any threshold that may exist. That's where I consider that common sense enters.
    Additionally, if a substance (especially type 6) is considered harmless at a low dosage, that may be only because its toxic effect is small enough that it is lost in the noise of the many other insults to which our bodies are daily exposed.
    - Richard

  7. #17
    Quote Originally Posted by rfbdorf
    Wise -
    I certainly take your point about nonlinearities and thresholds - if the body can dispose of something faster than it accumulates, then it may well be expected to have little, if any effect. And I am aware that there are precious few linear relationships in biological systems (well, maybe cardiac output = stroke volume * heart rate!)
    However, I did not mean common sense extrapolation in the sense of a linear extrapolation, but in the sense that I prefer to avoid exposure to smoke (to which I was referring in particular), trichloroethane, carbon tet, MEK, etc. (I imagine the latter 3 would fall into your category 6), because I have only a vague idea where my integrated exposure level lies with relation to any threshold that may exist. That's where I consider that common sense enters.
    Additionally, if a substance (especially type 6) is considered harmless at a low dosage, that may be only because its toxic effect is small enough that it is lost in the noise of the many other insults to which our bodies are daily exposed.
    - Richard
    Richard,

    I thought that it was interesting that you used the example of cardiac output being equal to volume and heart rate as a linear relationship. To me, that is not a cause effect relationship so much as it is simply a definition. The variables on each side of the equation are related. I think of relationships as how two independent variables change relative to each other and tried ot think of an example of linear relationships. It is an interesting exercise because I start out with the thought that something "must be" linear and then realize on second thought that it might not be.

    1. Fluid input and output. One would expect that our urine output would be linearly related to your input. It, however, may not be. First, clearly when you drink very little fluid, your urine output not only may be less than your intake (due to other outputs such as sweating, etc.) but much depends on the time at which you measure the output. If you look at urine output at 5 minutes after having drunk a liter of water, you will find relatively little urine production. However, if you measure it several hours later, you can get an amount that is probably similar to the amount that you have drunk. So, the relationship between fluid input and output depends on volume as well as time. And, of course, it also depends on absorption.

    2. Drug dose response curves. Most drug dose response curves are S-shaped. In other words, they have little effect until a certain threshold where the response increases with dose. However, when the dose increases above a certain level, you may get a plateau, i.e. no further effect with increasing doses. The entire curve may be shifted upward, downward, leftward, or rightward depending on circumstances. The slope of the curve may be different.

    3. Structure and function relationship. An obvious example is the relationship of the number of axons the spinal cord and walking function. One would expect a S-shaped curve. At the top end of the curve, one may find a plateau. At the bottom and left side of the curve, one may find another plateau. However, in reality, the curvilinear relationship between walking performance and number of axons is really asymptotic towards zero. With only 10% of the axons, animals such as rats still can walk. The curve tends to be asymptotic to the Y-axis. Above 10% of axons, there is a second linear relationship between walking and axon number.

    I am struggling to find a linear relationship. There are so few. I am not sure that I can think on one. Yet, most doctors and scientists use linear extrapolations for biological relationships. The reason why they can get away with it is because the relationship is often linear over a limited range. But, when one is considering something like passive smoking which is perhaps several orders of magnitude less than the exposure that is associated with actually smoking. If the relationship were linear, the risk would be vanishingly small.

    Wise.

  8. #18
    Quote Originally Posted by Wise Young
    Antiquity,

    Thanks for the post. There are many other myths that doctors believe. Let me try to enumerate several additional myths that many doctors believe but what evidence there is either does not support or contradict them:
    • Passive smoking causes cancer.
    • Bedrest is good for patients. The evidence suggest that bedrest is at best not effective and often deleterious for most conditions.
    • H1 blockers are effective for treating urticaria but H2 blockers are not. The evidence suggests that the combination of the two are better than H1 blockers alone.
    • Patients allergic to penicillins are also allergic to cephalosporins.
    • Prednisone must always be tapered. The evidence indicates that short course of prednisone and other glucocorticoid steroids that are 8 days in duration does need to be tapered.
    • High blood pressure (such as that resulting from autonomic dysreflexia must be emergently treated with sublingual nifedine (a calcium channel blocker). Some evidence suggests that rapid reduction of hypertension may be associted complications. Sublingual nifedipine should not be used for this purpose.
    • beta-blockers should not be used in congestive heart failure. Some evidence suggests that they may be beneficial for heart failure.

    etc. etc.

    http://www.montana.edu/wwwebm/myths.htm
    wow, I thought alot of these were true. I think just two days ago my resident told me that beta-blockers should be used in MI but not CHF.

    So prednisone doesn't need to be tapered if its given for less than 8 days regardless of the dose? Also, Dr. Young, reminds me of something else I was wondering but wasn't able to find an answer to -- if prednisone suppresses the immune system, why does it cause the white count to go up? Is it that prednisone suppresses the immune response at the cellular level, and the body reacts to this by increasing white cell production? (sorry for going off topic...)

    p.s. Happy belated birthday Dr. Young and happy new year!

  9. #19
    Quote Originally Posted by Wise Young
    ...
    I thought that it was interesting that you used the example of cardiac output being equal to volume and heart rate as a linear relationship. To me, that is not a cause effect relationship so much as it is simply a definition. The variables on each side of the equation are related. I think of relationships as how two independent variables change relative to each other and tried ot think of an example of linear relationships. It is an interesting exercise because I start out with the thought that something "must be" linear and then realize on second thought that it might not be.
    ...
    I am struggling to find a linear relationship. There are so few. I am not sure that I can think on one. Yet, most doctors and scientists use linear extrapolations for biological relationships. The reason why they can get away with it is because the relationship is often linear over a limited range. But, when one is considering something like passive smoking which is perhaps several orders of magnitude less than the exposure that is associated with actually smoking. If the relationship were linear, the risk would be vanishingly small.
    I generally think of cardiac output as a measure of the volume of blood flowing through the aorta per unit time - somewhat more than a simple definition by CO=SV*HR. But linearity in that falls apart if you look at a short enough time interval (e.g., less than one pulse interval, where heart rate begins to lose its meaning), or when you consider leakages, for example due to mitral regurgitation.

    For just about any purpose, one should indeed be very cautious about applying a linear extrapolation. However, linear interpolation is generally pretty safe even for biological purposes, when applied over a small enough interval, but I bet you're right that over a large range, there are no true linearities in the body. Come to think of it, maybe in the universe, considering quantum and relativistic effects!

    - Richard

  10. #20
    Quote Originally Posted by Monique
    wow, I thought alot of these were true. I think just two days ago my resident told me that beta-blockers should be used in MI but not CHF.

    So prednisone doesn't need to be tapered if its given for less than 8 days regardless of the dose? Also, Dr. Young, reminds me of something else I was wondering but wasn't able to find an answer to -- if prednisone suppresses the immune system, why does it cause the white count to go up? Is it that prednisone suppresses the immune response at the cellular level, and the body reacts to this by increasing white cell production? (sorry for going off topic...)

    p.s. Happy belated birthday Dr. Young and happy new year!
    Hi, Monique. Happy New Year to you.

    You were always the best student who asked the best questions. It is an interesting and important question that shed light on the mechanism by which glucocorticoids work as anti-inflammatory drugs. Glucocorticoids decrease inflammation by suppressing NF-kappa-B induced genes. These genes include IL-1, IL-6, and TNF-alpha, which you know all about.

    The mechanism by which glucocorticoids affect NF-kappa B activity is not well-understood. Recent work suggest that glucocorticoids do not directly interfere with DNA binding capabilities of NF-kappa-B. Nor does it seem to affect the inhibitor of NF-kappa-B (iKb). Most textbook suggest that glucocorticoids bind to cytoplasmic glucocorticoid receptors (GR) that act in combination with CREB-binding protein (CBP), p300, and SRC-1 to prevent NF-kappa-B activation. One study suggested that glucocorticoids suppress NF-kappa-B induced gene expression regardless of the levels of CBP, p300, SRC-1. It may work by interfering with p65 interaction with transcription machinery (see attached Bosscher, et al. 2000). Others (Novac, et al. 2006) have reported that the GR directly regulates the human FasL promoter. As you know, Fax ligand is a proinflammatory messenger. In any case, much data now suggests the glucocorticoids antagonize NF-kappa-B effects through multiple mechanisms (See Almawi and Melemedjian, 2002).

    So, why does glucocorticoid therapy increase white blood cell counts? I am not sure. I think that it does so by increasing the migration of white blood cells from tissues into the bloodstream, as opposed to increasing proliferation of such cells. In a normal person, glucocorticoids will increase leukocytosis and may mask the onset of an infection.

    While the mechanism of glucocorticoid induced leukocytosis is not clear, the increased in white blood cell count is understandable considering the role of glucocorticoids. This is a stress-induced hormone that prepares the body for infection and injury. If given or naturally induced for too long, i.e. more than a week, glucocorticoids can suppress immune function.

    You can usually tell a glucocorticoid-induced leukocytosis by the profile of the white cell counts. Glucocorticoids generally cause lymphocytes to undergo apoptosis and thereby reduce lymphocyte counts. However, it can increase neutrophils. The effect of steroids is also dose dependent.

    Wise.

Similar Threads

  1. Replies: 0
    Last Post: 09-13-2002, 11:11 AM
  2. Replies: 0
    Last Post: 09-11-2002, 10:30 AM
  3. Victims of medical errors may face hurdle
    By antiquity in forum Life
    Replies: 0
    Last Post: 08-11-2002, 07:45 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •