I thought it might be useful to write down my observations and thoughts about the financial side of the practice of medicine, based upon 25 years of private and academic practice of internal medicine, 15 years with a partner, and the last 10 years solo.
First, I have been solo for the past 10 years, because no recent medical school graduate wants to practice (and I have a big-city practice and attend at a medical school) pure internal medicine without an enormous salary guarantee that only a large group can afford to offer (which includes fronting their malpractice insurance). The reason is that HMO's and MCR both pay much more for doing than for thinking and diagnosing, since the powers that pay can measure doing, but not thinking. It is ridiculous that MCR pays me more for a 5-minute rigid sigmoidoscopy than for a 15-minute intake interview. This is why dermatology is the most popular residency. Getting paid $500 for a Botox injection every 3 months is an annuity! And since no insurance pays for it, the physicians can charge whatever they wish. Man is an economic animal, and often will respond to economic stimuli. So when Massachusetts established state-wide health care, there were not enough primary care doctors, and the average wait for a new doctor was 60 days. Just imagine the shortage when the 45 million currently uninsured patients look for a new primary care doctor. We will need at least 10,000 new primary care doctors, and where will they come from? Actually, every doctor should do Botox and Restalen injections and skin biopsies one day a week, so he/she can practice the medicine they like the other four days. And since MCR pays psychiatrists less than they pay me for the same amount of office time, why would any psychiatrist ever want to see a new MCR patient?
Second, we have socialized medicine now, and it's called Medicare. Few MCR patients want to give it up. There are many advantages to the patient, and the main advantage to the physician is that we spend less time on paperwork, and time is our least fungible resource. For a MCR patient, unlike an HMO patient, I don't have to call anyone for permission to get an MRI,and I don't have to worry that the best shoulder surgeon I know is a Cigna MD, and my patient is an Oxford patient, so my patient can't get to see the physician I prefer. I don't have to ask the patient to fax me a list of the HMO ophthalmologists so I can see if I know anyone on the list.
Also, MCR patients get back 5 to 10 times the dollar amount in medical services of the MCR premiums they paid.The tobacco companies showed (but dropped the argument because it made for poor PR) that the government saves money on every patient who dies before reaching MCR age. I also (I hope) will not have to spend time asking someone for permission to prescribe a brand name rather than a generic drug, or to prescribe a brand name drug that is not on their formulary.
Third, if everyone has a private MD, they will "crash" in the ER less often with diabetes out of control, unstable angina, end stage renal disease, etc. If you look at ER visits in Canada vs. here, there are fewer (percentage-wise) visits for acute medical conditions. Having a private physician who you can visit regularly puts a basic floor under your medical condition, even if nothing more is done than blood pressure check, PAP smear, and stool for blood. Overall, since ER visits are extremely expensive because ER doctors do every test they can think of, since they don't want to miss anything (viz. the recent article in the July, 2009 issue of "Archives of Internal Medicine" on the ER workup of syncope in the elderly) we will save a lot of money.
Fourth, technology is expensive, but it works. No patient walks up to an orthopedic surgeon and says "I want a new hip". Instead, the doctor is told "I can't move without agonizing hip pain". No patient says "I need cardiac bypass surgery", or "I need cataract surgery".
Fifth point has to do with malpractice. The most common suit against a family MD is for failure to diagnose a condition. As far as I am aware, no physician ever got sued for doing a test, but only for not doing one. When I started practice,I used to spend 30 minutes to explain to a 40 year old man with no risk factors and atypical chest pain and a normal EKG why he did not need a stress test, and the problems that can result from false positive tests. Now I suggest a stress-echo, stress-thallium and consultation with a cardiologist to the same patient. Why risk being sued even if I know I will win the case? Why spend the time and stress?
As long as we have contingency fees, we will have malpractice cases, and doctors will do extra tests to minimize their exposure, as well as make any referrals any family member of the patient suggests.
Sixth, unless you have a boutique practice, the average MD has to see one patient every 10 minutes, which involves putting 4 patients into 4 rooms, and having the NP or PA take the interval history, do the vital signs, etc. No one can get proper medical care under these conditions. I was trained to spend 15-45 minutes in the office with my patient to take a history and discuss any family stresses, then examine the patient in the exam room, and then bring the patient back into my office to discuss the results. I don't know of any doctor under 60 who practices medicine in this way, and most patients don't know what they are missing.
Seventh, there is something very wrong with the practice of medicine when the majority of doctors advise their children not to go into medicine.
Eighth, every medical system in every country rations by money, time, or availability, since the demand for medical care is almost infinite. England NHS does not transplant kidneys over a certain age, so those who can afford it fly to India to buy a kidney. Some drugs are not allowed in England because the NHS pays for all drugs, and this is how the country caps pharmacy expenses. Canada rations hospitals as to how many open-heart surgeries they can do in a given month. Germany reduces the physicians state-paid salary if their patients' prescriptions cost too much. MCR pays for only so many physical therapy visits or hospital days a year.
Ninth, I still love the practice of medicine, and intend to see my patients as long as I am able.
P.S.: About 10 years ago, there was an article published in NEJM that showed that New Haven had 5 times as many cholecystectomies per capita as did Boston, while Boston had 5 times as many CABG's per capita as did New Haven. Did one city have too many operations, or did the other have too few? No one could determine the answer.
P.P.S. The best review of the current state of internal medicine was published by David. D. Norenberg, M.D., in the Annals of Internal Medicine(Ann. Int. Med. 2009; 150:725-726) accessible at www.annals.org: "The Demise or Primary Care". Please read it and share it with everyone you know, physicians, patients, and politicians.
Saturday, July 25, 2009
Saturday, July 11, 2009
Vaccines and Immunizations for Adults
This blog is a general (search the archives of Morbidity and Mortality Weekly Report, or the CDC travel page, or www.immunizationed.org for more detailed info) review of the vaccinations/immunizations available for adults. I will be reviewing them in the order of the fatality rate for the infected but unvaccinated immunocompetent adults. WARNING: if you are pregnant, or immunosuppressed--e.g. on steroids, on cancer chemotherapy, have AIDS,etc., do NOT take any live virus vaccine without checking with your doctor, and avoid anyone who has had a live virus vaccine for 8 weeks after they have been vaccinated.
RABIES: Except for one fortunate female in Wisconsin, this disease has been 100% fatal once symptoms are evinced. The virus is carried in the salivary glands of infected meat-eating animals: bats, raccoons, skunks, foxes, dogs, cats,---.The vaccine is available from your state board of health, or vetinarians. You should seek vaccination for any bite or scratch from ANY unknown animal, as well as if you have been in the same room with a bat, since their needle-like teeth often can bite without your feeling it. It is also recommended for pre-exposure prophylaxis if you are traveling to a region where there is a high incidence of rabies in stray dogs, such as certain parts of Asia, including Katmandu, and India, Mexico, and other countries (check with CDC). Chloroquine, and possibly mefloquine, both used for malaria prophylaxis, may blunt your immune response to the vaccine, so be sure to notify the treating physician if you are taking either of these drugs, and ask for deep IM rather than SQ vaccination. Raccoons are nocturnal animals, so any raccoon seen in the daytime should be presumed to be rabid.
TETANUS: Last year there were over 50 fatalities from tetanus in the U.S., and there should be none.It is 100% preventable by an antitoxin shot (no longer made from horse serum, so there is little likelihood of a strong reaction). It is a paralysis caused by a toxin generated by an anaerobic bacterium that inhabits the soil, especially where there is horse manure. You should be vaccinated every 10 years, but there is no harm done if the ER vaccinates you whenever you come in with a laceration. More women than men die each year (often from a non-remembered puncture wound from a garden rose thorn, etc.): Since men get injured more often than women, they are more likely to have been in the ER and been re-vaccinated. If you don't remember when you received your last tetanus shot, please ask your doctor for one at your next visit. I should also mention that the only vaccine currently available in the USA is dT, so you are also re-vaccinated against diphtheria at the same time. It would take up too much space here to explain why, but the reason is not medical.
YELLOW FEVER: This is a LIVE vaccine, and the vaccination is good for 10 years. It is only available from physicians or centers certified by the US gov't. Yellow fever is endemic in many areas of countries in South America, Central America, and Africa. It is spread by mosquitoes both in jungles and cities and one infection confers lifetime immnunity. The case fatality rate can be as high as 50%, and there is no treatment, since we have no drugs that reliably attack flaviviruses. Therefore, when in doubt, take the vaccine. (As a historical point, at one time it was endemic in the United States, from New Orleans to Philadelphia.)
POLIOMYELITIS: The oral vaccine (OPV) is live, and the injected vaccine (IPV) contains inactivated virus. One dose if IPV is recommended for travelers to underdeveloped countries or those with poor sanitation (check with the CDC).
HEPATITIS B: This vaccine is recommended not only because Hepatitis B can progress to cirrhosis of the liver even with treatment, but also because infection with Hepatitis B (or C) is a proven risk factor for cancer of the liver. It can be spread by unprotected sex, exposure to blood, including transfusions, or sharing of needles by drug users. Because it is considered a STD, vaccination for all children has been recommended before they reach the age of sexual activity. As an adult, vaccination is recommended for those exposed to blood (ER workers, dental hygienists and dentists, etc.) and those with compromised immune systems, including those on renal dialysis. You should specifically ask your physician if you should have the vaccine. If you are immuno-incompetent, then it is recommended that you be tested for protective antibodies one month after vaccination. Unless you are tested annually for antibodies, it is unknown how long your protection will last.
N.B.: If you want to help your country, then please donate a unit of blood. The blood bank will test your blood, at no charge to you, for Hepatitis B and C, syphilis, AIDS, and a few other diseases. You will not be permitted to donate blood if you have been in a country where malaria is endemic (including Mexico, even if in a non-malarious area) within the previous 12 months.
SMALLPOX: A LIVE vaccine. No longer given routinely, except to U.S. military personnel. Cidofovir has shown some success in animal models of this disease. Case fatality rate at least 20%.
HPV: Vaccine against Human Papilloma Virus. This is considered to be permissive, if not an inducer for cancer of the cervix. There are many strains of HPV, and the vaccine does not protect against all of them, so women should still have pap smears, even if vaccinated. Recommended for all females before first menses, and any woman who has not received the vaccination.
INFLUENZA: Everyone should have annual vaccination with the regular flu vaccine.
Since the antibody levels decline after 4 to 6 months, I recommend biannual vaccinations to my all my patients. In the past, the vaccine was recommended only for those over 65, as well as health care workers and teachers, etc., because there was not enough to give everyone in the U.S. It is especially important to get this vaccine because Tamiflu is no longer effective against it.
AVIAN FLU: The government hopes to have a vaccine against this available by October, 2009. Fortunately, this flu virus is generally susceptible to Tamiflu. We are all sincerely worried about this flu strain developing resistance to Tamiflu so I(we) are not prescribing it to patients to have "just in case". Patients born after 1956 have no natural antibodies to this H1N1 strain, and so are at risk for death. Patients born before this date were probably infected with a strain somewhat homologous to avian flu, and therefore become less ill.
PNEUMOCOCCAL PNEUMONIA: Patients over 65, those without a spleen (either at birth or after surgery), patients with TB,diabetes and a host of other chronically ill patients are at risk of death from pneumococcal pneumonia. However, it was found that patients younger than 65 develop a stronger immune response than older patients, Most physicians now will give the vaccine to their (normal) patients when they reach age 60, and a booster is recommended 5 years later.
MENINGOCOCCAL MENINGITIS: Recommended for those traveling to areas where this is endemic. The antibody levels decline after 2 years. Saudi Arabia requires this vaccination for those traveling to Mecca.
JAPANESE B ENCEPHALITIS: May to September in Asia. Low risk of infection, and high rate of side effects and reactions. Discuss with your doctor.
PLAGUE: Recommended only for travelers in regions where it is endemic (but this does include the U.S. Southwest). Booster every 1 to 2 years.
TYPHOID: Oral vaccine. Spread by contaminated food and water. Suggested if you are going to be in underdeveloped country for 3 or more months, but you should still be careful what you eat and drink.
HEPATITIS A: Virtually never fatal, but it can ruin your vacation. Spread by contaminated food and water. can be combined with Hepatitis B vaccine.
SHINGLES/HERPES ZOSTER:LIVE virus. This is reactivation of the virus that gave you chicken pox. It lives in your body forever, and can be reactivated along pain nerves as you get older and your immune surveillance competence drops. Classically, it causes redness and blisters along the path of a pain nerve, but you can get non-eruptive zoster, wherein the virus re-activates WITHOUT the rash. The problem is extreme residual pain along the path of the inflamed nerve, and it seems that older patients get more severe pain. We recommend it for all patients, because the pain can be disabling. However, vaccination is not 100% protective. Also, a prior zoster attack does not prevent a second attack.
RABIES: Except for one fortunate female in Wisconsin, this disease has been 100% fatal once symptoms are evinced. The virus is carried in the salivary glands of infected meat-eating animals: bats, raccoons, skunks, foxes, dogs, cats,---.The vaccine is available from your state board of health, or vetinarians. You should seek vaccination for any bite or scratch from ANY unknown animal, as well as if you have been in the same room with a bat, since their needle-like teeth often can bite without your feeling it. It is also recommended for pre-exposure prophylaxis if you are traveling to a region where there is a high incidence of rabies in stray dogs, such as certain parts of Asia, including Katmandu, and India, Mexico, and other countries (check with CDC). Chloroquine, and possibly mefloquine, both used for malaria prophylaxis, may blunt your immune response to the vaccine, so be sure to notify the treating physician if you are taking either of these drugs, and ask for deep IM rather than SQ vaccination. Raccoons are nocturnal animals, so any raccoon seen in the daytime should be presumed to be rabid.
TETANUS: Last year there were over 50 fatalities from tetanus in the U.S., and there should be none.It is 100% preventable by an antitoxin shot (no longer made from horse serum, so there is little likelihood of a strong reaction). It is a paralysis caused by a toxin generated by an anaerobic bacterium that inhabits the soil, especially where there is horse manure. You should be vaccinated every 10 years, but there is no harm done if the ER vaccinates you whenever you come in with a laceration. More women than men die each year (often from a non-remembered puncture wound from a garden rose thorn, etc.): Since men get injured more often than women, they are more likely to have been in the ER and been re-vaccinated. If you don't remember when you received your last tetanus shot, please ask your doctor for one at your next visit. I should also mention that the only vaccine currently available in the USA is dT, so you are also re-vaccinated against diphtheria at the same time. It would take up too much space here to explain why, but the reason is not medical.
YELLOW FEVER: This is a LIVE vaccine, and the vaccination is good for 10 years. It is only available from physicians or centers certified by the US gov't. Yellow fever is endemic in many areas of countries in South America, Central America, and Africa. It is spread by mosquitoes both in jungles and cities and one infection confers lifetime immnunity. The case fatality rate can be as high as 50%, and there is no treatment, since we have no drugs that reliably attack flaviviruses. Therefore, when in doubt, take the vaccine. (As a historical point, at one time it was endemic in the United States, from New Orleans to Philadelphia.)
POLIOMYELITIS: The oral vaccine (OPV) is live, and the injected vaccine (IPV) contains inactivated virus. One dose if IPV is recommended for travelers to underdeveloped countries or those with poor sanitation (check with the CDC).
HEPATITIS B: This vaccine is recommended not only because Hepatitis B can progress to cirrhosis of the liver even with treatment, but also because infection with Hepatitis B (or C) is a proven risk factor for cancer of the liver. It can be spread by unprotected sex, exposure to blood, including transfusions, or sharing of needles by drug users. Because it is considered a STD, vaccination for all children has been recommended before they reach the age of sexual activity. As an adult, vaccination is recommended for those exposed to blood (ER workers, dental hygienists and dentists, etc.) and those with compromised immune systems, including those on renal dialysis. You should specifically ask your physician if you should have the vaccine. If you are immuno-incompetent, then it is recommended that you be tested for protective antibodies one month after vaccination. Unless you are tested annually for antibodies, it is unknown how long your protection will last.
N.B.: If you want to help your country, then please donate a unit of blood. The blood bank will test your blood, at no charge to you, for Hepatitis B and C, syphilis, AIDS, and a few other diseases. You will not be permitted to donate blood if you have been in a country where malaria is endemic (including Mexico, even if in a non-malarious area) within the previous 12 months.
SMALLPOX: A LIVE vaccine. No longer given routinely, except to U.S. military personnel. Cidofovir has shown some success in animal models of this disease. Case fatality rate at least 20%.
HPV: Vaccine against Human Papilloma Virus. This is considered to be permissive, if not an inducer for cancer of the cervix. There are many strains of HPV, and the vaccine does not protect against all of them, so women should still have pap smears, even if vaccinated. Recommended for all females before first menses, and any woman who has not received the vaccination.
INFLUENZA: Everyone should have annual vaccination with the regular flu vaccine.
Since the antibody levels decline after 4 to 6 months, I recommend biannual vaccinations to my all my patients. In the past, the vaccine was recommended only for those over 65, as well as health care workers and teachers, etc., because there was not enough to give everyone in the U.S. It is especially important to get this vaccine because Tamiflu is no longer effective against it.
AVIAN FLU: The government hopes to have a vaccine against this available by October, 2009. Fortunately, this flu virus is generally susceptible to Tamiflu. We are all sincerely worried about this flu strain developing resistance to Tamiflu so I(we) are not prescribing it to patients to have "just in case". Patients born after 1956 have no natural antibodies to this H1N1 strain, and so are at risk for death. Patients born before this date were probably infected with a strain somewhat homologous to avian flu, and therefore become less ill.
PNEUMOCOCCAL PNEUMONIA: Patients over 65, those without a spleen (either at birth or after surgery), patients with TB,diabetes and a host of other chronically ill patients are at risk of death from pneumococcal pneumonia. However, it was found that patients younger than 65 develop a stronger immune response than older patients, Most physicians now will give the vaccine to their (normal) patients when they reach age 60, and a booster is recommended 5 years later.
MENINGOCOCCAL MENINGITIS: Recommended for those traveling to areas where this is endemic. The antibody levels decline after 2 years. Saudi Arabia requires this vaccination for those traveling to Mecca.
JAPANESE B ENCEPHALITIS: May to September in Asia. Low risk of infection, and high rate of side effects and reactions. Discuss with your doctor.
PLAGUE: Recommended only for travelers in regions where it is endemic (but this does include the U.S. Southwest). Booster every 1 to 2 years.
TYPHOID: Oral vaccine. Spread by contaminated food and water. Suggested if you are going to be in underdeveloped country for 3 or more months, but you should still be careful what you eat and drink.
HEPATITIS A: Virtually never fatal, but it can ruin your vacation. Spread by contaminated food and water. can be combined with Hepatitis B vaccine.
SHINGLES/HERPES ZOSTER:LIVE virus. This is reactivation of the virus that gave you chicken pox. It lives in your body forever, and can be reactivated along pain nerves as you get older and your immune surveillance competence drops. Classically, it causes redness and blisters along the path of a pain nerve, but you can get non-eruptive zoster, wherein the virus re-activates WITHOUT the rash. The problem is extreme residual pain along the path of the inflamed nerve, and it seems that older patients get more severe pain. We recommend it for all patients, because the pain can be disabling. However, vaccination is not 100% protective. Also, a prior zoster attack does not prevent a second attack.
Wednesday, July 1, 2009
Aspirin, Lipoprotein A, CRP, Anti-oxidants, Heart disease
There have been a number of articles published recently in Lancet, Journal of the AMA, and newspapers about primary prevention of heart disease. By primary prevention, we mean the reduction of risk of a heart attack in patients who have not had a heart attack. Since the greatest risk for having a heart attack is having had one, the number of patients in a study of primary prevention has to be larger and last for a longer time than a study of secondary prevention, where aspirin, beta-blockers, statins, etc., have been shown to be of help.
I should also mention that the over-riding problem in prevention of any disease is not lack of communication between doctors and patients,or lack of information on the part of patients, but the unwillingness of patients to change their behavior (and no, I am not blaming the victim). Most patients know that they should exercise, lose weight, and stop smoking to reduce their primary heart attack risk, but how many do, even if, as in NYC, the amount of calories per dish is published in the menu of every chain restaurant?
1) C-reactive protein, or CRP. Numerous studies have shown a correlation between elevated CRP and heart attacks. This correlation is about as strong as that between elevated homocysteine and heart attacks. However, just as pulling down on the metal elevator arrow in the lobby of a skyscraper does not bring the elevator down, lowering the homocysteine has not been shown to lower heart attack risk. Whether the CRP as a marker for inflammation means that inflammation is a primary risk factor remains to be seen. A recent published genetic analysis (similar to the Lp(A) analysis) seem to show that lowering CRP does NOT lower the risk for heart disease.
2) Aspirin. First we have to decide if we mean 81mg/day, 325 mg/day, or 325 mg twice a day. Since aspirin blocks platelet clumping for 7 days, it is not clear why the studies involved daily aspirin, rather than once or twice a week. In addition, none of the studies to date controlled for aspirin resistance, where aspirin at the suggested dose does not prevent platelet clumping. It has been a "consensus" that absent any other risk factors (smoking, diabetes,etc.) that aspirin prevention should start at age 45 for men and 55 for women. However, a recent article Lancet cast severe doubt on this recommendation, so your guess is as good as mine.
3) Lp(A), or lipoprotein A. A recent article used genetic analysis to suggest that elevated Lp(A) is an independent risk factor for heart attacks. However, no one has shown that lowering Lp(A), prevents heart attacks, nor do we have a good drug to lower Lp(A). The same could be said in reverse about low HDL, the "good" cholesterol.
4) Anti-oxidants. The prophylactic use of anti-oxidants should be approached with caution. Prospective studies have shown that extra daily Vitamin E increases the primary risk for heart attacks, and prophylactic beta carotene increases the risk for lung cancer in cigarette smokers. A recent study at Memorial Sloan-Kettering showed in vitro that vitamin C inhibited the killing effect of chemicals on cancer cells. Selenium and vitamin E have been shown to have no protective effect against prostate cancer.
The conceptual problem is, of course, that correlation does not imply causation. Just recall how coffee drinking was "shown" to be a risk factor for heart attacks until it was realized that more coffee drinkers than non-drinkers smoke cigarettes. We always have to beware of confounding factors, as well as a common cause that elevates both the risk of disease and the marker. Does anyone really understand why female Pima Indians of the American Southwest have such a high incidence of cholecystitis, or the female Parsees of India (Zoroastrians transplanted from the Mideast) have such a high incidence of breast cancer?
I should also mention that the over-riding problem in prevention of any disease is not lack of communication between doctors and patients,or lack of information on the part of patients, but the unwillingness of patients to change their behavior (and no, I am not blaming the victim). Most patients know that they should exercise, lose weight, and stop smoking to reduce their primary heart attack risk, but how many do, even if, as in NYC, the amount of calories per dish is published in the menu of every chain restaurant?
1) C-reactive protein, or CRP. Numerous studies have shown a correlation between elevated CRP and heart attacks. This correlation is about as strong as that between elevated homocysteine and heart attacks. However, just as pulling down on the metal elevator arrow in the lobby of a skyscraper does not bring the elevator down, lowering the homocysteine has not been shown to lower heart attack risk. Whether the CRP as a marker for inflammation means that inflammation is a primary risk factor remains to be seen. A recent published genetic analysis (similar to the Lp(A) analysis) seem to show that lowering CRP does NOT lower the risk for heart disease.
2) Aspirin. First we have to decide if we mean 81mg/day, 325 mg/day, or 325 mg twice a day. Since aspirin blocks platelet clumping for 7 days, it is not clear why the studies involved daily aspirin, rather than once or twice a week. In addition, none of the studies to date controlled for aspirin resistance, where aspirin at the suggested dose does not prevent platelet clumping. It has been a "consensus" that absent any other risk factors (smoking, diabetes,etc.) that aspirin prevention should start at age 45 for men and 55 for women. However, a recent article Lancet cast severe doubt on this recommendation, so your guess is as good as mine.
3) Lp(A), or lipoprotein A. A recent article used genetic analysis to suggest that elevated Lp(A) is an independent risk factor for heart attacks. However, no one has shown that lowering Lp(A), prevents heart attacks, nor do we have a good drug to lower Lp(A). The same could be said in reverse about low HDL, the "good" cholesterol.
4) Anti-oxidants. The prophylactic use of anti-oxidants should be approached with caution. Prospective studies have shown that extra daily Vitamin E increases the primary risk for heart attacks, and prophylactic beta carotene increases the risk for lung cancer in cigarette smokers. A recent study at Memorial Sloan-Kettering showed in vitro that vitamin C inhibited the killing effect of chemicals on cancer cells. Selenium and vitamin E have been shown to have no protective effect against prostate cancer.
The conceptual problem is, of course, that correlation does not imply causation. Just recall how coffee drinking was "shown" to be a risk factor for heart attacks until it was realized that more coffee drinkers than non-drinkers smoke cigarettes. We always have to beware of confounding factors, as well as a common cause that elevates both the risk of disease and the marker. Does anyone really understand why female Pima Indians of the American Southwest have such a high incidence of cholecystitis, or the female Parsees of India (Zoroastrians transplanted from the Mideast) have such a high incidence of breast cancer?
Thursday, June 25, 2009
Nutrition
I expect I will receive more e-mails from this blog than from all the others together. Nutrition is fundamental, not only to life, but, as per Freud, the first stage of mental development. Our first sensation of love, warmth, and security was at our mother's breast. Because it has such a fundamental connection with our psyche, this is probably why we have more and different personal habits and religious taboos over food than any other aspect of our lives. We also have magical thinking about "good" foods and "bad" foods.
Mother's milk is heavy in sugar and fat, which is probably why we all love ice cream. (Separate thought:I prefer chocolate to vanilla, and I often wonder if people who prefer vanilla taste the chocolate in the same way that I do.) Fat is appreciated because it adds flavor and substance to food It also markedly slows stomach emptying, so we feel fuller for a longer period of time if the meal contains fat. Some of my patients equate "low fat" to "low taste" especially low-fat mayonnaise.
There are eight essential amino acids, which are called "essential" because every animal cell needs all eight to make proteins,and cannot synthesize them. Vegetables have at most seven of these amino acids, which is why a complete vegetable dish contains rice + beans, or corn + peas (aka the American Indians' succotash); that is, at least two vegetables.
There are also two or three essential fatty acids (depending on your definition), and again any animal cell contains all of them, which again is why you need at least two vegetables if no meat is eaten. This is probably why no one looks truly happy in a vegetarian restaurant, because they sense they are missing something (or subconsciously are aware of the fact that tofu lowers men's sperm count because soy is a phytoestrogen). On the other hand, everyone is all smiles in a steak restaurant.
Everyone should be aware that Vitamin B-12 is necessary for an animal cell to synthesize animal DNA, but not vegetable DNA. Vegetables contain no vitamin B-12 whatsoever. Therefore, if you eat neither meat nor fish nor eggs, you will eventually die from pernicious anemia. It as also a curious fact that no one is literally addicted to eating meat, or can achieve an altered mental state by eating meat (licking Colorado toads being a notable exception). All the addicting or mind-bending foods are vegetables, with their totally different chemistry: nicotine, alcohol, caffeine, opium, marijuana ,peyote, "magic" mushrooms, etc. In fact, if you kill an animal or fish at random and eat it, the odds are that it will agree with you. But if you eat a random vegetable or berry, there are good odds that it will make you ill or kill you (mushrooms, choke cherries, oleander, foxglove, mistletoe, rhubarb, yew, manicheel), not to mention the severe contact allergies from poison ivy, oak and sumac.
As for cholesterol, it is an essential component of the membrane of every animal cell, through which all nutrients pass. (Plant cells have a cell wall outside the cell membrane, which contains no cholesterol whatsoever.) So when we chemically lower cholesterol, we lower a constituent of the membrane of every cell of the body. This in turn alters trans-membrane potentials, ion fluxes, molecular transport, and sensitivity to neural stimulation. This may explain why some patients develop leg cramps (muscle cell irritability and sensitivity), or altered liver tests (? inflammation of the cell membrane and enzyme leakage), and sometimes just don't "feel right"(altered electric potentials in brain cells). If cholesterol is lowered by weight loss, the body responds as a whole, and therefore there are no such side effects. When Wise's Potato Chips tried to advertise their chips as "having absolutely no cholesterol", which is a literal scientific fact, the FTC pulled their add, claiming that patients would confuse "no cholesterol" with "no fat".
We don't even know how many meals we should eat daily: one, two, three,or six. We don't know the proper time of day to eat the largest meal, even though we know that all our minerals and hormones have a diurnal cycle (cortisol is highest in the AM, serum iron in the PM, etc.). We don't know if the fact semi-starved rats exceed their natural life span applies also to humans. We don't understand why, if a normal BMI is optimal for health, how is it that people with an elevated BMI survive heart attacks better.
As for vitamins, a vitamin is a chemical which, if you do not ingest, you get ill and may die (scurvy, beri-beri, pernicious anemia, etc.), because your body needs it as a co-factor for essential enzymes, and cannot synthesize it.A,D,E, and K are fat-soluble vitamins, but C and all the B's will be excreted in your urine, if ingested in excess. Oddly enough, an excess of daily vitamin B-6 causes neurological symptoms, just as a deficiency of the same vitamin does. (cf article in NEJM, more than 10 years ago; research done at Albert Einstein Hospital)Since it is a pure chemical, your body cannot tell the difference between "natural" and synthesized vitamins. Since the FDA does not regulate or supervise vitamin manufacture, there is also absolutely no guarantee as to if a capsule labeled "400 U of Vitamin E" contains 0, 400, or 1200. And the idea that too much of a vitamin cannot hurt you never applies to fat-soluble vitamins, which are stored in your body. We have already seen studies of the coronary artery dangers of too much daily Vitamin E.
Finally, all the suggestions about "healthy" diets are based on retrospective and/or epidemiological studies, which all fail to remember that correlation does not mean causation. The only food intervention shown to lengthen life, mostly by preventing heart attacks, is alcohol. I have read studies with beer, scotch slivovitz, red wine, etc. It is the C2H5OH molecule that saves you. There was an article as far back in 1974 in JAMA, in which comparative autopsies showed that moderate male drinkers had larger diameter (i.e. wider and less clogged) coronary arteries than age and weight matched controls (non-smokers only). Every time a forward study (low fat, beta-carotene, selenium) has been done, the results have been nil, or even negative (especially with vitamin E).
The overall problem is that no prospective food study has ever been done with adult humans, so all are beliefs are just that; beliefs without a sound foundation in experimental fact.
Mother's milk is heavy in sugar and fat, which is probably why we all love ice cream. (Separate thought:I prefer chocolate to vanilla, and I often wonder if people who prefer vanilla taste the chocolate in the same way that I do.) Fat is appreciated because it adds flavor and substance to food It also markedly slows stomach emptying, so we feel fuller for a longer period of time if the meal contains fat. Some of my patients equate "low fat" to "low taste" especially low-fat mayonnaise.
There are eight essential amino acids, which are called "essential" because every animal cell needs all eight to make proteins,and cannot synthesize them. Vegetables have at most seven of these amino acids, which is why a complete vegetable dish contains rice + beans, or corn + peas (aka the American Indians' succotash); that is, at least two vegetables.
There are also two or three essential fatty acids (depending on your definition), and again any animal cell contains all of them, which again is why you need at least two vegetables if no meat is eaten. This is probably why no one looks truly happy in a vegetarian restaurant, because they sense they are missing something (or subconsciously are aware of the fact that tofu lowers men's sperm count because soy is a phytoestrogen). On the other hand, everyone is all smiles in a steak restaurant.
Everyone should be aware that Vitamin B-12 is necessary for an animal cell to synthesize animal DNA, but not vegetable DNA. Vegetables contain no vitamin B-12 whatsoever. Therefore, if you eat neither meat nor fish nor eggs, you will eventually die from pernicious anemia. It as also a curious fact that no one is literally addicted to eating meat, or can achieve an altered mental state by eating meat (licking Colorado toads being a notable exception). All the addicting or mind-bending foods are vegetables, with their totally different chemistry: nicotine, alcohol, caffeine, opium, marijuana ,peyote, "magic" mushrooms, etc. In fact, if you kill an animal or fish at random and eat it, the odds are that it will agree with you. But if you eat a random vegetable or berry, there are good odds that it will make you ill or kill you (mushrooms, choke cherries, oleander, foxglove, mistletoe, rhubarb, yew, manicheel), not to mention the severe contact allergies from poison ivy, oak and sumac.
As for cholesterol, it is an essential component of the membrane of every animal cell, through which all nutrients pass. (Plant cells have a cell wall outside the cell membrane, which contains no cholesterol whatsoever.) So when we chemically lower cholesterol, we lower a constituent of the membrane of every cell of the body. This in turn alters trans-membrane potentials, ion fluxes, molecular transport, and sensitivity to neural stimulation. This may explain why some patients develop leg cramps (muscle cell irritability and sensitivity), or altered liver tests (? inflammation of the cell membrane and enzyme leakage), and sometimes just don't "feel right"(altered electric potentials in brain cells). If cholesterol is lowered by weight loss, the body responds as a whole, and therefore there are no such side effects. When Wise's Potato Chips tried to advertise their chips as "having absolutely no cholesterol", which is a literal scientific fact, the FTC pulled their add, claiming that patients would confuse "no cholesterol" with "no fat".
We don't even know how many meals we should eat daily: one, two, three,or six. We don't know the proper time of day to eat the largest meal, even though we know that all our minerals and hormones have a diurnal cycle (cortisol is highest in the AM, serum iron in the PM, etc.). We don't know if the fact semi-starved rats exceed their natural life span applies also to humans. We don't understand why, if a normal BMI is optimal for health, how is it that people with an elevated BMI survive heart attacks better.
As for vitamins, a vitamin is a chemical which, if you do not ingest, you get ill and may die (scurvy, beri-beri, pernicious anemia, etc.), because your body needs it as a co-factor for essential enzymes, and cannot synthesize it.A,D,E, and K are fat-soluble vitamins, but C and all the B's will be excreted in your urine, if ingested in excess. Oddly enough, an excess of daily vitamin B-6 causes neurological symptoms, just as a deficiency of the same vitamin does. (cf article in NEJM, more than 10 years ago; research done at Albert Einstein Hospital)Since it is a pure chemical, your body cannot tell the difference between "natural" and synthesized vitamins. Since the FDA does not regulate or supervise vitamin manufacture, there is also absolutely no guarantee as to if a capsule labeled "400 U of Vitamin E" contains 0, 400, or 1200. And the idea that too much of a vitamin cannot hurt you never applies to fat-soluble vitamins, which are stored in your body. We have already seen studies of the coronary artery dangers of too much daily Vitamin E.
Finally, all the suggestions about "healthy" diets are based on retrospective and/or epidemiological studies, which all fail to remember that correlation does not mean causation. The only food intervention shown to lengthen life, mostly by preventing heart attacks, is alcohol. I have read studies with beer, scotch slivovitz, red wine, etc. It is the C2H5OH molecule that saves you. There was an article as far back in 1974 in JAMA, in which comparative autopsies showed that moderate male drinkers had larger diameter (i.e. wider and less clogged) coronary arteries than age and weight matched controls (non-smokers only). Every time a forward study (low fat, beta-carotene, selenium) has been done, the results have been nil, or even negative (especially with vitamin E).
The overall problem is that no prospective food study has ever been done with adult humans, so all are beliefs are just that; beliefs without a sound foundation in experimental fact.
Friday, June 19, 2009
Fatigue, Insomnia, Sleep Apnea, and Hypnotics
There are many beliefs about sleep, but not much verified knowledge. For instance, it is not true that one hour of sleep before midnight is worth two after.
It is true that a lot of fatigue is due to mental processes, rather than physical fatigue.
We are probably all underslept, since most adults have trouble getting "catch-up" sleep. When you were in high school or college, you could probably make up for the sleep deficit you incurred during the week by sleeping till noon or later on weekends. As we get older, we seem to lose this ability, while at the same time we awaken at night more frequently.
As a physician, I observe that my chronically underslept patients are more irritable, with degraded ego functions as well as more muscular pains and aches, and have some degree of depression. They also seem to function closer to baseline when doing well-known and repetitive tasks, but have trouble concentrating on crossword puzzles and the like. (Or as we used to say as interns, we could run a cardiac arrest with no sleep, but it was difficult to do an intake interview.)
Another danger of being underslept is falling asleep behind the wheel of your car, or making other errors. The government recognizes this by putting a limit on the number of consecutive hours a truck driver can drive, or a pilot can fly. The number of hours a medical intern can be on call at a hospital is also limited, but strangely enough a 60-year-old medical attending has no limits placed on his hours. In fact, if a patient called me at 10 at night and I told him/her that instead of getting my advice, because I had been awake for 16 hours ,they should go to the nearest emergency room where they could be seen by a well-rested doctor, they would be very upset.
Since lack of sleep is so corrosive and destructive, both mentally and physically, I see nothing wrong with the use of sleeping pills. I would prefer the use of sleeping pills to a patient falling asleep at the wheel of his/her car. Some patients are concerned about becoming dependent on sleeping pills, but I believe that many adults have "irritable sleep syndrome" and need some chemical help to get sufficient sleep.
Sleep apnea studies are in a curious position. They are reviewed by psychiatrists, neurologists, pulmonologists, and ENT doctors. The definition of sleep apnea is somewhat imprecise. (If you are not an overweight male who snores, the chance of your having sleep apnea is greatly lessened.) We all have apneas, generally defined as cessation of breathing for 10 or more seconds. The diagnosis is firmer if the oxygen saturation drops below 85% at the same time. There are also hypopneas, which are episodes of underbreathing without complete cessation. We have also known for a long time that everyone drops oxygen saturation during sleep, which is why we put nasal oxygen on all heart attack patients. Oxygen desaturation is a potent pulmonary artery vasoconstrictor, which overloads the right ventricle, which can tolerate volume overload better than pressure overload.
Since the number of apneas increases with age, the definition of sleep apnea depends in part on a patient's age. My two additional criticisms of all sleep apnea studies are:(1) no one asks the patient (usually a male) if he would be willing to wear a positive pressure sleep mask for the rest of his life, and (2) it would be more efficient to put a patient suspected of sleep apnea on a CPAP and then a BIPAP mask to see if they felt less tired, etc. the next day, but insurance and Medicare only pay for the mask if you first diagnose sleep apnea. In other words, unlike putting a patient on an antidepressant, CPAP cannot be used as a diagnostic/therapeutic treatment and test.
Before I send a chronically fatigued patient for a sleep apnea test, I try one month of Prozac or other SSRI, then one month on Effexor XR or Wellbutrin SR, and finally one month on Provigil, which is FDA approved for CNS stimulation of patients with sleep-shift disorder or chronic insomnia/fatigue, and seems to have no ischemic cardiac effects.
It is true that a lot of fatigue is due to mental processes, rather than physical fatigue.
We are probably all underslept, since most adults have trouble getting "catch-up" sleep. When you were in high school or college, you could probably make up for the sleep deficit you incurred during the week by sleeping till noon or later on weekends. As we get older, we seem to lose this ability, while at the same time we awaken at night more frequently.
As a physician, I observe that my chronically underslept patients are more irritable, with degraded ego functions as well as more muscular pains and aches, and have some degree of depression. They also seem to function closer to baseline when doing well-known and repetitive tasks, but have trouble concentrating on crossword puzzles and the like. (Or as we used to say as interns, we could run a cardiac arrest with no sleep, but it was difficult to do an intake interview.)
Another danger of being underslept is falling asleep behind the wheel of your car, or making other errors. The government recognizes this by putting a limit on the number of consecutive hours a truck driver can drive, or a pilot can fly. The number of hours a medical intern can be on call at a hospital is also limited, but strangely enough a 60-year-old medical attending has no limits placed on his hours. In fact, if a patient called me at 10 at night and I told him/her that instead of getting my advice, because I had been awake for 16 hours ,they should go to the nearest emergency room where they could be seen by a well-rested doctor, they would be very upset.
Since lack of sleep is so corrosive and destructive, both mentally and physically, I see nothing wrong with the use of sleeping pills. I would prefer the use of sleeping pills to a patient falling asleep at the wheel of his/her car. Some patients are concerned about becoming dependent on sleeping pills, but I believe that many adults have "irritable sleep syndrome" and need some chemical help to get sufficient sleep.
Sleep apnea studies are in a curious position. They are reviewed by psychiatrists, neurologists, pulmonologists, and ENT doctors. The definition of sleep apnea is somewhat imprecise. (If you are not an overweight male who snores, the chance of your having sleep apnea is greatly lessened.) We all have apneas, generally defined as cessation of breathing for 10 or more seconds. The diagnosis is firmer if the oxygen saturation drops below 85% at the same time. There are also hypopneas, which are episodes of underbreathing without complete cessation. We have also known for a long time that everyone drops oxygen saturation during sleep, which is why we put nasal oxygen on all heart attack patients. Oxygen desaturation is a potent pulmonary artery vasoconstrictor, which overloads the right ventricle, which can tolerate volume overload better than pressure overload.
Since the number of apneas increases with age, the definition of sleep apnea depends in part on a patient's age. My two additional criticisms of all sleep apnea studies are:(1) no one asks the patient (usually a male) if he would be willing to wear a positive pressure sleep mask for the rest of his life, and (2) it would be more efficient to put a patient suspected of sleep apnea on a CPAP and then a BIPAP mask to see if they felt less tired, etc. the next day, but insurance and Medicare only pay for the mask if you first diagnose sleep apnea. In other words, unlike putting a patient on an antidepressant, CPAP cannot be used as a diagnostic/therapeutic treatment and test.
Before I send a chronically fatigued patient for a sleep apnea test, I try one month of Prozac or other SSRI, then one month on Effexor XR or Wellbutrin SR, and finally one month on Provigil, which is FDA approved for CNS stimulation of patients with sleep-shift disorder or chronic insomnia/fatigue, and seems to have no ischemic cardiac effects.
Sunday, June 7, 2009
Generic Drugs, Brand Names, and Tier I Drugs
Most people think that generic drugs always work, that all Tier I drugs of an HMO are pharmacologically equivalent,and that generic drugs are cheaper. Only the last statement is true, and the pharmacies also love generics because they have higher profit margins. (Brand name $80, charge to you $100, profit$20; generic $20, price to you $50, profit $30.) However, the generics do not always work as well as brand names, because there can be different "inert" compounds added to the parent drug to make the tablet. The FDA requires that the blood levels of the generic be the same as the brand name (+/- 10%), but only the brand name drug has been tested for clinical efficiency. There is no way of comparing tissue levels of the brand name drug to the generic, and this is especially important with drugs that act on the central nervous system (tranquilizers, sleeping pills, pain pills) after crossing the blood-brain barrier. Again, no generic is tested for clinical efficiency against the brand name drug.
The blood-brain barrier is nature's way of protecting the brain from noxious compounds that are dissolved in your blood. A drug has to cross the blood-brain barrier (through capillary walls) to reach the brain and achieve its effect. Whether a particular drug can do this is usually only determined by trial and error. For instance, penicillin readily crosses the blood-brain barrier, and is therefore extremely active in treating meningitis. Keflex, a cephalosporin, cannot cross the blood-brain barrier, and is therefore useless in treating meningitis, although it generally can treat any soft tissue infection or lung infection that penicillin does.
Within a particular class of drugs (e.g. beta-blockers) if a drug is lipophilic, this affects the ability of the brain to absorb the drug once it gets there. Marijuana is extremely lipophilic, so it is released very slowly from fatty tissue in the brain, and your urine can test positive for at least two weeks after inhaling it. There can also be subtle differences within a class of drugs: Coreg is the only beta-blocker shown to reduce microalbinuria in diabetic patients.
When it comes to other groups of drugs, there is wide human variation in responsiveness, since we are all biochemically different. Claritin and Zyrtec are over-the-counter (OTC) anti-histamines, and sometimes one will work on a patient's allergies, and sometimes the other, since they are chemically very different. Allegra is a prescription antihistamine, which works better in many patients, but unlike Medicare Part D, most HMO's ask me to certify that both OTC drugs did not work before they will pay for Allegra. The same problem arises is proton-pump inhibitors (PPI's) for acid reflux. The HMO's want me to try the patient on OTC Prilosec before paying for Prevacid, Aciphex or Nexium. Furthermore, if I want to prescribe Nexium at a double dose to suppress acid symptoms, which some patients need, I generally have to send them to a GI doctor to have the payment okayed. Remember, if you are forced to buy an OTC drug, your co-pay is 100%, and the cost to your HMO is $0.
The worst scenario is in CNS-acting drugs. In my experience, generic Ativan works less than 10% of the time, so if a patient does not respond to generic Ativan, I am not sure of the cause. I also have found that generic Prozac works less than 50% of the time, and there are treatment problems with generic Wellbutrin. If a patient responds well to the anti-depressant Lexapro, the HMO generally asks me to switch the patient to a generic form of Celexa, even though they are chemically totally different drugs. I have also found that the beta-blocker Nadolol is excellent prophylaxis against migraines, as is Inderal LA, but since only Inderal (which requires dosing 4 x a day) is available in a generic form, the LA Rx. is generally not paid for.
In your own case, I am certain that aspirin, Advil, and Alleve all work differently in you, even though they all are NSAID's (anti-inflammatories).
Viagra, Levitra, and Cialis all work to treat erectile dysfunction, but sometimes one drug will work, and sometimes another.
The main point is that drugs with the same physiological endpoint in the human body have different chemical structures, and they cannot all work equally well.
The blood-brain barrier is nature's way of protecting the brain from noxious compounds that are dissolved in your blood. A drug has to cross the blood-brain barrier (through capillary walls) to reach the brain and achieve its effect. Whether a particular drug can do this is usually only determined by trial and error. For instance, penicillin readily crosses the blood-brain barrier, and is therefore extremely active in treating meningitis. Keflex, a cephalosporin, cannot cross the blood-brain barrier, and is therefore useless in treating meningitis, although it generally can treat any soft tissue infection or lung infection that penicillin does.
Within a particular class of drugs (e.g. beta-blockers) if a drug is lipophilic, this affects the ability of the brain to absorb the drug once it gets there. Marijuana is extremely lipophilic, so it is released very slowly from fatty tissue in the brain, and your urine can test positive for at least two weeks after inhaling it. There can also be subtle differences within a class of drugs: Coreg is the only beta-blocker shown to reduce microalbinuria in diabetic patients.
When it comes to other groups of drugs, there is wide human variation in responsiveness, since we are all biochemically different. Claritin and Zyrtec are over-the-counter (OTC) anti-histamines, and sometimes one will work on a patient's allergies, and sometimes the other, since they are chemically very different. Allegra is a prescription antihistamine, which works better in many patients, but unlike Medicare Part D, most HMO's ask me to certify that both OTC drugs did not work before they will pay for Allegra. The same problem arises is proton-pump inhibitors (PPI's) for acid reflux. The HMO's want me to try the patient on OTC Prilosec before paying for Prevacid, Aciphex or Nexium. Furthermore, if I want to prescribe Nexium at a double dose to suppress acid symptoms, which some patients need, I generally have to send them to a GI doctor to have the payment okayed. Remember, if you are forced to buy an OTC drug, your co-pay is 100%, and the cost to your HMO is $0.
The worst scenario is in CNS-acting drugs. In my experience, generic Ativan works less than 10% of the time, so if a patient does not respond to generic Ativan, I am not sure of the cause. I also have found that generic Prozac works less than 50% of the time, and there are treatment problems with generic Wellbutrin. If a patient responds well to the anti-depressant Lexapro, the HMO generally asks me to switch the patient to a generic form of Celexa, even though they are chemically totally different drugs. I have also found that the beta-blocker Nadolol is excellent prophylaxis against migraines, as is Inderal LA, but since only Inderal (which requires dosing 4 x a day) is available in a generic form, the LA Rx. is generally not paid for.
In your own case, I am certain that aspirin, Advil, and Alleve all work differently in you, even though they all are NSAID's (anti-inflammatories).
Viagra, Levitra, and Cialis all work to treat erectile dysfunction, but sometimes one drug will work, and sometimes another.
The main point is that drugs with the same physiological endpoint in the human body have different chemical structures, and they cannot all work equally well.
Saturday, May 30, 2009
Analysis of Meta-analysis
There have recently been several articles published in noted medical journals which derived from meta-analysis of pooled studies. Usually the purpose of a medical study/experiment is to see if intervention (X) affects condition (Y) in a patient, which condition may be discrete (e.g. heart attack) or continuous (e.g. blood pressure). Other studies are retrospective epidemiological studies, which should be used to suggest interventional studies. Unfortunately retrospective studies (e.g. a low fat diet lessons the risk of cancer) as well as meta-analyses (e.g. beta-carotene lowers the heart attack rate) have been shown on prospective, interventional studies to be incorrect. (Of course, the fact that a 5-year low fat diet in white females failed to lower their incidence of either breast cancer or colon cancer has not prevented the American Cancer Society from recommending a low fat diet to prevent cancer, but that discussion is for another blog.)
I find that medical statistics are still poorly understood by the medical-reading public, as well as by many doctors. For instance, the first study (of male cigarette smokers in Finland) to evaluate the effect of cholesterol-lowering by fibrates on the heart attack rate used a 2-tailed t-test, in case cholesterol lowering was harmful. When a null result was found, they redid the study, saying that lowering cholesterol could not possibly be harmful, and by using a 1-tailed t-test, the positive effect of cholesterol-lowering reached statistical significance. It is difficult to explain, without the use of mathematics, why changing the analysis in this matter is incorrect from a statistical point of view. (Just look at all the arguments about the Monte Hall problem, which is really trivial Bayesian analysis.)
Definition: meta-analysis is the combining and analysis of several studies, none of which usually reaches statistical or clinical significance because of small sample size, to determine the effect on a patient of an intervention.
In any event, it is my opinion that the result of any meta-analysis should be treated as an epidemiological event, and should only be a basis to suggest further interventionl testing. The real problem arises when a meta-analysis falsely suggests a negative effect, and then doctors are fearful of doing a prospective study because of the threat of malpractice. The basic problem with meta-analysis is its lack of mathematical rigor, and its basis on the subjective views of the lead investigator. I propose there should be an on-line journal of medical meta-analytic results, rather than formal peer-reviewed publication, and the results should be interpreted as tentative conclusions to be tested, rather than hardened "fact".
We also have read articles where two different meta-analyses of the same clinical intervention (e.g. low molecular weight heparin to prevent DVT's) reach opposite conclusions, and it is impossible to reject either study.
The following is a list of some of the problems that, in my opinion, prevent meta-analysis from reaching the statistical rigor and reproducibility that we should require of any process that affects human life and health:
Interpretation of "effect size" is totally arbitrary.
At least two authors are usually required to code the individual studies for grouping, because of the acknowledged subjective interpretation that occurs in coding. "Valid" studies require a reliability above 0.8!
The underlying level of risk in every studied group is of key importance, and often cannot be weighted properly. (e.g. the higher your cholesterol, the greater is the benefit of lowering it.)
The danger of Simpson's Paradox (or the Yule-Simpson Effect). This occurs when two smaller studies point to one result of intervention, and the pooled study points in the other direction. (Its existence is easy to demonstrate in vector space, but you can ask one of your statistical buddies to explain its basis to you.)
There are two different methods of combining results: the fixed effects model or the random effects model. The choice of approach strongly determines the results of the meta-analysis, and neither is always demonstrably "right".
Every meta-analysis requires subjective decisions, and few if any readers will redo the calculations with their own decisions.
Bias in selecting which studies to use.
There are TWO methods to measure effect: standardized mean difference, and correlation, and therefore there can be two different results (but only one is published).
There are various sources of heterogenicity and selection bias that cannot be determined.
There is often post-hoc sub-group analysis in meta-analyses, and this can easily lead to false statistical conclusions. This is similar in concept to the difference between an everywhere continuous function and a piecewise continuous function.
Results are often discarded if they do not make "medical sense", but it is always tricky and suspect to discard any result.
Comparing randomized trials may not be itself a random process.
The capacity to detect bias is small if only a small number or trials is considered.
There is no statistical solution to the problem of failure to reject the null hypothesis of homogeneous events, even if significant differences exist between the studies.
And finally: assessing the quality of a study in order to determine whether or not to include it in one's meta-analysis, is purely subjective.
I find that medical statistics are still poorly understood by the medical-reading public, as well as by many doctors. For instance, the first study (of male cigarette smokers in Finland) to evaluate the effect of cholesterol-lowering by fibrates on the heart attack rate used a 2-tailed t-test, in case cholesterol lowering was harmful. When a null result was found, they redid the study, saying that lowering cholesterol could not possibly be harmful, and by using a 1-tailed t-test, the positive effect of cholesterol-lowering reached statistical significance. It is difficult to explain, without the use of mathematics, why changing the analysis in this matter is incorrect from a statistical point of view. (Just look at all the arguments about the Monte Hall problem, which is really trivial Bayesian analysis.)
Definition: meta-analysis is the combining and analysis of several studies, none of which usually reaches statistical or clinical significance because of small sample size, to determine the effect on a patient of an intervention.
In any event, it is my opinion that the result of any meta-analysis should be treated as an epidemiological event, and should only be a basis to suggest further interventionl testing. The real problem arises when a meta-analysis falsely suggests a negative effect, and then doctors are fearful of doing a prospective study because of the threat of malpractice. The basic problem with meta-analysis is its lack of mathematical rigor, and its basis on the subjective views of the lead investigator. I propose there should be an on-line journal of medical meta-analytic results, rather than formal peer-reviewed publication, and the results should be interpreted as tentative conclusions to be tested, rather than hardened "fact".
We also have read articles where two different meta-analyses of the same clinical intervention (e.g. low molecular weight heparin to prevent DVT's) reach opposite conclusions, and it is impossible to reject either study.
The following is a list of some of the problems that, in my opinion, prevent meta-analysis from reaching the statistical rigor and reproducibility that we should require of any process that affects human life and health:
Interpretation of "effect size" is totally arbitrary.
At least two authors are usually required to code the individual studies for grouping, because of the acknowledged subjective interpretation that occurs in coding. "Valid" studies require a reliability above 0.8!
The underlying level of risk in every studied group is of key importance, and often cannot be weighted properly. (e.g. the higher your cholesterol, the greater is the benefit of lowering it.)
The danger of Simpson's Paradox (or the Yule-Simpson Effect). This occurs when two smaller studies point to one result of intervention, and the pooled study points in the other direction. (Its existence is easy to demonstrate in vector space, but you can ask one of your statistical buddies to explain its basis to you.)
There are two different methods of combining results: the fixed effects model or the random effects model. The choice of approach strongly determines the results of the meta-analysis, and neither is always demonstrably "right".
Every meta-analysis requires subjective decisions, and few if any readers will redo the calculations with their own decisions.
Bias in selecting which studies to use.
There are TWO methods to measure effect: standardized mean difference, and correlation, and therefore there can be two different results (but only one is published).
There are various sources of heterogenicity and selection bias that cannot be determined.
There is often post-hoc sub-group analysis in meta-analyses, and this can easily lead to false statistical conclusions. This is similar in concept to the difference between an everywhere continuous function and a piecewise continuous function.
Results are often discarded if they do not make "medical sense", but it is always tricky and suspect to discard any result.
Comparing randomized trials may not be itself a random process.
The capacity to detect bias is small if only a small number or trials is considered.
There is no statistical solution to the problem of failure to reject the null hypothesis of homogeneous events, even if significant differences exist between the studies.
And finally: assessing the quality of a study in order to determine whether or not to include it in one's meta-analysis, is purely subjective.
Subscribe to:
Posts (Atom)