Medical Science

Evidence-based Medicine: What You Need to Know

In this post I will write about how the evidence for how well medicines work, and the risk of side effects are not always what they seem. I’m going to show that relative risk reduction (or relative risk increase) always looks a lot bigger than absolute risk reduction or absolute risk increase. Journals and advertisements always report relative risk reduction for medicines or other treatments because they look more impressive. On the other hand, side effects are almost always reported as absolute risk increase because that looks a lot smaller. I will show you how to calculate both kinds of risk reduction and increase. I will argue that absolute risk reduction or increase is what you really want to know before deciding to take a medicine or other treatment.

Any study of a medicine or treatment always has a group that gets the treatment and a group that gets a placebo. That is the only way to know if the medicine or treatment really works. There is always a placebo effect for any medicine. That is, a certain portion of the people who get a placebo get better. If the number of people who get better from the actual treatment is higher than the number of people who get better from the placebo, then the treatment works.

People who participate in a study know that they might get the treatment or a placebo, but they don’t know which one they got until the end of the study. In medical terminology, they are blinded from knowing whether they got treatment or placebo. If the study is double blind (the most reliable kind) then neither the investigators who administer the treatment nor the participants know which study participants got the study treatment or the placebo until the end of the study.

If the treatment works better than the placebo, that result can be reported in several different ways

Relative Risk Reduction or Increase

Relative risk is the proportion of people who have the disease or condition being studied in the treatment group divided by the proportion of people who have the condition in the placebo group.

For example, let’s say we are testing a treatment to prevent diabetes. The control group and the treatment group each have 100 people. 30 people get diabetes in the control group and only 10 people get diabetes in the treatment group. The risk in the treatment group is 10/100 = 0.1 The risk in the placebo group is 30/100 = 0.3 To compare those risks we divide the risk in the treatment group by the risk in the control group. 0.1/0.3= .33. That means that the relative risk of diabetes in the treatment group is 1/3 of the risk in the placebo group. To change that to the relative risk reduction percentage we use the formula 100(1- Relative Risk). Plugging our values into that formula gives 100(1-.33)= 67%. In other words, the treatment reduces the risk of diabetes by 67% relative to the risk of diabetes in the placebo group. That sounds like a big treatment effect!

All medicines or treatments have some side effects. Placebos can have side effects too, especially if people are told (as they must be) what are the possible side effects of the treatment. Placebo side effects are called “nocebo” effects.

In the hypothetical diabetes study described above, let’s say that a side effect of the treatment is bladder infection. Let’s suppose that 5 out of the 100 people in the treatment group get a bladder infection while only 1 out of 100 in the control group get a bladder infection. The relative risk of a bladder infection in the placebo group is 1/100=0.01. The risk of bladder infection in the treatment group is 5/100=.05. The relative risk of getting a bladder infection in the placebo group compared to the treatment group is .01/.05=0.2. To change that to a relative percentage increase, we use our formula again. 100(1-.2)=80%, This means that the risk of the side effect of bladder infection is 80% more likely in the treatment group compared to the risk of bladder infection in the placebo group.

Absolute Risk Reduction or Increase

Relative risk reduction or increase does not take into account the baseline risk of getting the disease or condition. Absolute risk reduction does take into account the baseline risk.

In our example above the absolute risk reduction is 30% (risk of diabetes in the control group) minus 10% (risk of getting diabetes in the treatment group) = 20%. That means that treatment reduces risk of diabetes by 20%. Notice that this is a much smaller number than the 67% relative risk reduction, but it more accurately reflects how much the treatment would reduce your risk of diabetes.

In our example the absolute risk increase of getting a bladder infection with treatment is 5% (the risk of bladder infection in the treatment group) minus 1% (the risk of getting bladder infection in the control group) = 4%. That means that the risk of getting a bladder infection from the treatment is 4% more than no treatment. Again, a much smaller number than relative risk increase of 80%.

Number Needed to Treat (or Harm)

Another way to look at how well a medicine or treatment works compared to placebo is the number of people that need to be treated in order to help one person. This is called Number Needed to Treat, abbreviated as NNT. The NNT = 1 divided by the absolute risk reduction. In our example the absolute risk reduction of getting diabetes in the treatment group was 20%. 1 divided by 0.20 = 5. That means you would need to treat 5 people to prevent 1 case of diabetes with this hypothetical treatment.

Number Needed to Harm, abbreviated as NNH is 1 divided by absolute risk increase. In our hypothetical example NNH = 1 divided by .04 = 25. This means that you would need to treat 25 people for one person to get a bladder infection

A Real World Example: Fosamax to prevent hip fracture in women with osteoporosis

In the real world, we rarely see a treatment effect as big as in our hypothetical example. Let’s look at a real study on a real medicine. Here are some numbers from a big four year study on using Fosamax (generic name alendronate) to prevent hip fracture in women with osteoporosis (thinning of bones). This study was reported in the Journal of the American Medical Association (JAMA) in 1988. It was the first large study to show that alendronate reduced hip fractures in women with osteoporosis. The study included over 4000 women with osteoporosis (shown by a type of bone scan called a DEXA scan). There were 2,218 women in the placebo group and 2,214 women in the treatment group. In the placebo group 812 women (36.6%) had severe bone thinning in the hips by Dexa scan. In the treatment group 819 (37%) women had severe bone thinning in the hips by Dexa scan. Over the four years there were 18 hip fractures in the placebo group (2.2%) and there were 8 hip fractures in the treatment group (1%)

Now lets do our calculations:

Relative Risk = .01 divided by .022 = .45. Relative Risk Reduction = 100(1-.45)= 55%. . This large relative risk reduction is what the article describing the study reported.

Absolute Risk Reduction = 2.2% – 1% = 1.2%. As you can see, the absolute risk reduction is tiny. Taking alendronate for 4 years reduced hip fracture by only about 1%.

Number Needed to Treat = 1/1.2% = 83.3. That means you would need to treat 83 women for four years to prevent one hip fracture.

Side Effects

Muscle or bone pain

In other studies, muscle or bone pain, sometimes severe was reported by 4% in the treatment group and 2.5% in the placebo group.

Relative Risk = .025 divided by .04 = .625. Relative Risk Increase = 100(1-.625)=37.5%. That means that bone and muscle pain are 37.5% more likely in the treatment group relative to the risk in the placebo group.

Absolute Risk Increase = 4% – 2.5% = 1.5%. This means that there is only a 1.5% increase in risk of muscle or bone pain when taking alendronate.

Number Needed to Harm = 1 divided by 1.5% = 67. That means you would need to treat 67 people for one person to have muscle or joint pain. Notice that the number needed to harm is less than the number needed to treat to prevent 1 hip fracture!

Osteonecrosis of the jaw

This is a rare but very serious side effect of alendronate. Assuming treatment group effect of .01% versus placebo of essentially zero let’s do our calculations.

Relative Risk Increase: since this complication is so rare, there are no trials comparing it with placebo.

Absolute Risk Increase: .01 %.

Number Needed to Harm: 1 divided by .01 % = 1000. This means that 1000 people would be treated before you would see one person with this serious complication.

Bottom Line

Journal articles and advertisements almost always report relative risk reduction for medicines or treatments because it makes the effect of the medicine or treatment look bigger. Side effects are almost always reported as absolute risk increases because it makes them look smaller.

Absolute Risk Reduction or Increase is what you really want to know when you are considering taking a medicine or treatment. Most of the time you will not be able to calculate absolute risk reduction, because you won’t have the actual percentages from the placebo and the treatment groups. Don’t despair though. Dr. Google is there to help. Using the following Google search will usually give you the absolute risk reduction. Type in “absolute risk reduction for (name of medicine or treatment).

Once you have the Absolute Risk Reduction (ARR) or Absolute Risk Increase (ARI)you can calculate for yourself the Number Needed to Treat or the Number Needed to Harm. Just divide 1 by either ARR or ARI.

Chronic Stress Response: It Can Make You Sick or Kill You

All mammals, including humans have an innate response to perceived threat or stress. The more common name for it is the “flight or fight” response. Our remote ancestors faced many real threats. Let’s say for example one encountered a saber tooth tiger. As soon as he (or she) saw the tiger, several things happened. Epinephrine and norepinephrine were released, speeding up the heart rate in preparation for running away. A surge of cortisol was also released, which increased glucose in the bloodstream for fuel for muscles and the brain. Cortisol also increases mental alertness. Inflammatory molecules were released to promote wound healing should that be needed.

This kind of acute stress response is a good thing. People or animals with this kind of response were more likely to survive and reproduce. Once the acute threat was over, all the hormones and neurotransmitters quickly returned to their baseline levels.

In today’s world, threats from predators are not a problem for the vast majority of people. The threats we perceive are things like poor work conditions; experiencing discrimination, hate, or abuse; poverty; homelessness; divorce or other family discord; having little control over outcomes; feeling overwhelmed.

These are all things that produce the stress response, but unlike our remote ancestors, these threats are chronic. They are either lifelong or at least last a long time. Instead of returning to normal, the stress hormones and neurotransmitters stay elevated for long periods of time. A chronic stress response is definitely not a good thing!

Allostatic Load

The medical term for the acute stress response is called allostasis. Here is the definition of allostasis from Wikipedia: “Allostasis is the efficient regulation required to prepare the body to satisfy its needs before they arise by budgeting those needed resources such as oxygen, insulin etc., as opposed to homeostasis, in which the goal is a steady state.” Allostasis is an adaptive response to acute stress. Allostatic load on the other hand is the long-term result of failed allostasis, resulting in dysregulation (abnormal function) of multiple systems including the neuroendocrine, cardiovascular, immune, and metabolic systems.

Allostatic load is measured traditionally by 10 indicators of chronic stress. Primary indicators are the hormones and neurotransmitters released by stress. Secondary outcomes are measurements of the systemic effects of the primary indicators. All of these indicators are associated with the perception of stress. Below is a table showing the 10 indicators, how they are measured, and which body systems are affected. Here is a link to the full article from which this table comes: Allostatic Load: Importance, Markers, and Score Determination in Minority and Disparity Populations

CategoryMarkerFunctional purpose
Primary mediatorsDehydroepiandrosterone sulfate (DHEA), serumSecreted by the adrenal glands. When high with stress it tends to lower cortisol and be protective in the stress response.
Cortisol, urinaryIntegrated measure of 12-hour hypothalamic–pituitary–adrenal axis activity. Secreted by the adrenal glands. Has multiple effects in stress response.
Epinephrine, urinaryIntegrated indices of 12-hour sympathetic nervous system activity. Sympathetic nervous system activation increases heart rate and blood pressure.
Norepinephrine, urinary
Secondary outcomesSystolic blood pressureIndices of cardiovascular activity and major risk factor for vascular disease
Diastolic blood pressure
Waist–hip ratioIndex of long-term levels of metabolism and adipose (fat) tissue deposition. High value means fat around internal organs which increases inflammation and increases LDL (bad cholesterol) and triglycerides.
High-density lipoprotein cholesterolIndex of atherosclerotic risk protection. Low value increases risk of heart disease.
Total cholesterolIndex of long-term atherosclerotic risk
Hemoglobin A1CIntegrated measure of high blood sugar over 2–3 months

Each indicator that is a certain distance out of the normal range counts as one point. The score can range from zero to ten. The higher the score, the greater the risk of illness or death.

Other Indicators

Although the ten indicators were the ones described in the original papers about allostatic load, other indicators have been used as well.

  • Heart rate variability is the normal beat to beat variability in the heart rate. In a healthy heart there is slight variation in the timing of one heartbeat to the next. Chronic stress reduces or even eliminates this beat to beat variation.
  • High sensitivity C-reactive protein (CRP). This is a measure of systemic inflammation that can result from chronic stress.

How is the stress reaction triggered?

The stress reaction begins in the brain. Something in the environment is perceived in a part of the front of the brain called the prefrontal cortex. This is the executive decision maker in the brain. If the prefrontal cortex perceives something in the environment as a threat, then it sends messages to the limbic system (the part of the brain that is involved with emotions). It also sends messages to centers lower in the brain, especially the hypothalamus. The hypothalamus sends messages to the adrenal glands which secrete cortisone, norepinephrine and epinephrine. The hypothalamus secretes DHEA. Messages from the hypothalamus are also sent to the white blood cells which secrete inflammatory chemicals called cytokines. All of this prepares the body to deal with the perceived threat. Different people may perceive different things as a threat. It is the reaction to perceived threats that causes allostatic load. If another person experiences the same thing in the environment as not a threat, then there is no stress reaction.

Diseases associated with high allostatic load (high chronic stress)

A high allostatic load score is not disease in itself, but if chronic stress continues then disease in the cardiac, metabolic, neuroendocrine and immune system can occur. Here is a list of diseases associated with persistent high allostatic load.

  1. Heart disease, primarily progressive blockage of the coronary arteries. This can lead to angina and/or heart attack. Congestive heart failure and arrhythmia like atrial fibrillation can also occur
  2. Peripheral arterial disease. That is blockage in arteries in the legs and sometime fingers.
  3. High blood pressure
  4. Stroke
  5. Autoimmune diseases like rheumatoid arthritis or lupus
  6. Diabetes
  7. Fibromyalgia
  8. Chronic Fatigue Syndrome
  9. Dementia or decreased cognitive function
  10. Depression
  11. PTSD
  12. Cancer, particularly breast and ovarian cancer. The increase in cancer is probably related to decreased immune system function

Allostatic Load and Mortality

Many studies have shown that people with persistently hight allostatic load have about a 25% higher premature death rate than people with low allostatic load.

Disparities in Health Outcomes

The response to chronic stress (allostatic load) may explain some of the disparities we see in health outcomes. We know, for example that Adverse Childhood Events (ACE), which include things like abandonment and abuse, increase the risk of many chronic diseases in adulthood. Studies have shown that adults with a history of ACE have high allostatic load scores.

African Americans have higher incidence of many cancers, as well as poorer outcomes from those cancers. They also have worse outcomes from heart disease, high blood pressure and diabetes. While a good portion of these poorer outcomes are related to lack of access to health care, these disparities persist to some degree even in middle class and upper middle class African Americans. Almost all African Americans have experienced or still experience racism on a chronic basis. African Americans of all social classes have higher allostatic load scores than caucasians. Chronic stress and response to it may be the common denominator for these disparities as well as for health outcome disparities in other marginalized populations.

How to reduce allostatic load

There is typically a long time between the presence of indicators of allostatic load and illness and death caused by diseases associated with these indicators. That presents an opportunity to reduce allostatic load before the chronic stress response leads to illness and death. So how do we reduce allostatic load?

Some of the things that cause allostatic load can only be reduced by societal changes. Things like poverty, structural racism and homelessness cannot be decreased by individual effort. Even these causes, though, can respond to the mind body methods discussed below. On the other hand, if you don’t have enough to eat, have no home, or have a job that gives you no control of your life, it is not likely that you will have the energy or the will, or the financial means to do many of the mind body methods discussed below. We should not be distracted from working to decrease the inequities that are responsible for societal causes of chronic stress.

Mind-Body Medicine

Remember that an external threat is first received by the peripheral nervous system and transmitted to the pre-frontal cortex. In order to reduce allostatic load we can either reduce the threat perception in the prefrontal cortex (top down) or reduce the transmission of threat in the peripheral nerves (bottom up).

Top Down Treatments

Top down treatments start with intentional activity in the prefrontal cortex. The idea is to decrease activation of the limbic system and the hypothalamus. This can be accomplished by mindfulness meditation, hypnosis (including self hypnosis), mental imagery and progressive muscle relaxation. All of these techniques when done regularly have been found to decrease allostatic load indicators and to reduce the risk of stress related illnesses.

Bottom Up Treatments

Bottom up treatments decrease the threat transmission to the prefrontal cortex. They include yoga, Tai Chi, massage and biofeedback. These treatments have also been shown to decrease allostatic load and to reduce stress related illness.

Bottom up and top down are somewhat of an oversimplification. All of these treatments have some aspects of both top down and bottom up. Yoga, for example includes aspects of meditation. The same goes for Tai Chi. Biofeedback involves some attention from the prefrontal cortex. Massage also includes progressive muscle relaxation.

Bottom Line

The body’s reaction to a perceived threat includes a complex cascade of messages from the executive center in the prefrontal cortex to multiple body systems including the nervous system, the endocrine system, the cardiovascular system and the immune system. All of these things prepare the body to deal with the threat. As long as the threat is short term the stress response is very useful to the organism.

Perception of chronic stress leads to continuous secretion of all the stress hormones and inflammatory cytokines and this leads to dysfunction of multiple body systems and eventually to illness and death.

Mind body treatments, both top down and bottom up can reduce the allostatic load (chronic stress response) and reduce the risk of stress induced illness and death.

Many causes of chronic stress have to do with the structure of our society, such as poverty, homelessness and structural racism. Individual effort is not likely to ameliorate the effect of these causes of chronic stress. All of us should be working toward societal change to reduce chronic stress response in marginalized populations.

Why are US Health Costs so High?

As I pointed out in a previous post: Cost and Quality of Healthcare in the US, we have the highest health care costs in the world, and yet have much worse outcomes than other industrialized countries.  In this post I will write about why our costs are so high and what we might do to fix that.

Cost Drivers

Use of imaging

The US uses more sophisticated imaging technologies such as CT scans, MRI scans and PET scans per capita than any other industrialized country.  The cost per scan is also much higher. For example an MRI in the US averages $1145, while the average cost in Switzerland is $138.  Note that more imaging per person does not translate into better health.

Use of Prescription Drugs

The US is “addicted” to prescription drugs. The average person over 18 in the US is on 2.2 regular prescription drugs. That is the highest among industrialized countries. In the Netherlands, for example that number is 1.2; half as much.  Prescription drugs in the US also cost more than twice as much for the same drugs from the same manufacturers as in other industrialized countries.

Costs of procedures

Routine surgical procedures cost much more in the US than in other industrialized countries.  The average cost of an appendectomy in the US is $13,910. Contrast that with $4995, which is the average price in the Netherlands.  The average cost of coronary bypass surgery in the US is $75,345.  The average price in the Netherlands is $15,742 – 80% less!

Administrative costs

Administrative costs represent 30% of the cost of US health care.  That is much greater than other western countries. Canada, for example has health care administrative costs of about 16%.  Medicare administrative costs are about 2%.  That means 98 cents of every dollar Medicare spends actually pays for medical care.  Pretty good for an “inefficient” government program.

Specialist to Generalist ratio

Generalist physicians are considered to be in family medicine, general pediatrics or general internal medicine. In the US, the ratio of specialists to these generalist physicians is 2:1 (two specialists for every generalist). In the rest of the developed world, the ratio is exactly the opposite: 1:2 (two generalists for every specialist). Specialists in the US make an average annual salary of $350,300 while generalists make an average salary of $242,400. Those are the highest physician incomes in the world except for the tiny country of Luxembourg. In the Netherlands, average annual specialist income is $200,300 and average generalist income is $137,500. As you can see from these figures, the high specialist to generalist ratio in the US is a major driver of US health care costs. More specialists does not mean better care. In fact, the higher the specialist to generalist ratio, the worse the health outcomes. Here is a link to an article by Barbara Starfield describing how more specialists = worse health outcomes: The Effect of Specialist Supply on Population Health.

Unnecessary Care

The Institute of Health estimates that unnecessary procedures and tests add 210 billion dollars to US health care costs, making it the single largest contributor to waste. Too many specialists who do procedures means more procedures get done. Here is a link to an opinion piece in Time magazine that gives some examples of egregious unnecessary care: One Patient, Too Many Doctors: The Terrible Expense of Overspecialization.

Consequences of the High Costs of Medical Care

The extraordinarily high cost of US healthcare means that the cost of health insurance is also high. Companies who provide health care insurance for their employees have moved toward high deductible plans. Many workers, especially blue collar workers cannot afford to pay several thousand dollars for health care before their insurance starts to pay. As a result, they defer needed care.

Low and middle income people do not have enough access to medical care. Rich people, on the other hand get too much medical care and are much more likely to get unnecessary care. See my recent post: Social Determinants of Health.

Can anything be done to reduce US healthcare costs?

The answer is yes, but it will require the political will to resist the giant money machine of the medical-industrial complex. In 2020 the healthcare sector spent more than 623 million dollars on lobbying. There are a few things that would make a significant difference.

If congress would allow Medicare to negotiate the price of medicines for Medicare recipients, US drug prices would be much closer to the prices paid in other western countries.

The specialist-generalist imbalance could be addressed by forgiving student loans for medical graduates who choose generalist residencies, such as family medicine and general pediatrics. Unfortunately, there are almost no general internal medicine residencies left in the US.

Most difficult of all, but also most important would be universal health insurance not tied to one’s employer. The US is the only western country that does not have universal healthcare. The Affordable Healthcare Act (sometimes called Obamacare) made some progress in this direction, but not nearly enough. Other western countries have found many different ways to provide universal health insurance. They range from a government run system like the UK to a completely private insurance system with subsidies such as Germany or Singapore.

Bottom Line

The US spends twice as much on healthcare as Switzerland, the next highest western country. US health outcomes are worse than any other rich western country. There are multiple factors responsible for our astronomical healthcare costs. Getting our healthcare costs in line with other western countries will require political solutions. The medical-industrial complex spends millions of dollars lobbying to keep things the way they are.

The Placebo Effect: Not Just Psychological

I’m going to take a break for a while from writing posts on COVID-19. I will do a post later on the Omicron variant, but we don’t yet know enough about it to make that useful right now. This post is about a powerful effect of any treatment that is independent of its direct effect on a system in the body.

Placebos in double blind trials

We normally think of placebos (medicines that have no direct effect) in the context of research. In order to tell if any treatment really works it has to be compared to a placebo. In this kind of research the placebo is an inactive pill that looks just like the active pill being tested. Why is it necessary to use placebos in clinical trials? Because a substantial portion, often up to twenty to thirty per cent of people will get better with the inactive medicine. These are not people just convincing themselves that they are better. One can measure changes in endorphins (the body’s natural pain killer) and other physiological changes. This is powerful medicine!

The placebo effect in everyday medical care

Placebo effects happen all the time in any visit to a health professional and with any medicine that is prescribed. I prefer to call these context effects. The word placebo should be reserved for clinical trials. It would be unethical for a clinician to give a patient something that he or she knew was inactive. Nonetheless there are context effects that enhance the effect of any active medicine or treatment. Master clinicians know this and know how to maximize the context effects that enhance the effect of any treatment. Below are a few of the things that clinicians can do that research has shown produce context effects that enhance medical treatment.

Believing in the treatment

Clinicians who really believe that a treatment will work explain it confidently and maximize the patient’s hope and expectation that he/she will get better.

Invite patients to be partners

Patients who feel that they have some control over treatment decisions and/or timing have less pain and side effects from medicines. The clinician must be still be seen as the confident leader of the treatment team in order to maximize this context effect.

Patient Factors

Expectation that a treatment will work produces the strongest context effects. Context effects are also strengthened by the amount of confidence a person has in his/her physician.

How big are context effects?

The best way to evaluate the power of context effects is to look at data from treatments that were thought initially to be effective and were later found out to be ineffective in clinical trials. In a study of five treatments for angina (heart pain related to partial blockage of blood flow to the heart) that were later shown to be ineffective it was found that over 80% of the patients who received these treatments had substantial improvement in their heart pain that in some cases lasted for a year or more! That is a powerful effect!

How do context effects work?

Expectation that a treatment will work produces physiological changes. I already talked about how patients produce opioid substances called endorphins that can decrease pain.

Another example is Parkinson’s disease. People with Parkinson’s disease have trouble starting movements. A placebo that a Parkinson’s patient expects will help their movement causes measurable releases of dopamine in the brain that does in fact help them move better.

Context effects may also affect the immune system. Reseach is ongoing here, but it is possible that a positive attitude may speed recovery from or even prevent certain infections.

Nocebo effects

There are also things that clinicians or other health professionals can do that can make a patient worse, despite being given the correct medicine. As opposed to the placebo effect, things that make people worse are called nocebo effects. If a researcher carefully explains all the possible side effects of a medicine in a placebo controlled clinical trial, a substantial portion of the people who get the placebo (inactive medicine) will experience some of the side effects that they were told about. This also happens in clinical practice with effective medicines. Some clinicians, worried about malpractice law suits, will detail all the possible side effects of a medicine or treatment and remain somewhat neutral about the positive effects. In this case the patient is much more likely to have side effects from the medicine than if the clinician emphasizes the potential benefit of the medicine.

Patients who expect to have lots of pain after a surgical procedure will experience much more pain than someone who expects to be able to manage their pain after the same surgical procedure.

What can you do?

Many years ago, Norman Vincent Peale wrote a book called ”The Power of Positive Thinking.” At the time, there was absolutely no evidence for his claims. It turns out though that now there is a lot of evidence supporting health benefits of a positive attitude. Many studies have shown that optimists are healthier than pessimists, have up to 50% less heart disease than pessimists and have longer life expectancy then pessimists. Pessimists are more realistic about their expectations than optimists, but that realism comes with less good health. Clearly optimists experience positive context effects. What you can do to maximize these effects on your health is to cultivate a positive attitude in general. Easier said than done, of course, but the evidence is clear.

Bottom Line

The context in which medicines and other treatments are administered can have powerful physiologic effects independent of the direct effect of the medicine or treatment. These effects can be positive or negative. Clinicians can enhance the effect of medicines or treatments by projecting confidence that patients will improve and by giving patients appropriate control over decisions about treatment and timing of treatment. Patients who have confidence in their physician and who have high expectations of treatment also are more likely to have positive context effects. A positive attitude in general leads to less illness, less chronic disease and a longer life expectancy.

Lies, Anecdotes and Evidence

Lies

Lies are deliberate falsehoods. The person telling a lie knows the truth, but chooses not to tell it. There are several main reasons for lying. The first, and most common is to avoid responsibility or punishment for something you have done. We are all familiar with the young child who lies for this reason. “I did not break the lamp, Mommy. It just fell off the table.” An alternative strategy to avoid blame or punishment involves blaming someone else. “I did not break the lamp, Mommy. Jaimie did it.” The third reason is to manipulate someone else’s behavior to get them to do something that you want. The classic form of this kind of lie is the telephone scam. The caller asks for money and promises to do something he/she has no intention of doing.

Lying is always destructive, but when lying becomes common in the public sphere, and when lies are told by people who would normally be trusted sources of information, then trust is eroded for everyone, even people who are telling the truth. A substantial portion of the public begins to believe that everyone is lying. This clearly happened through most of 2020 with regard to the pandemic.

Sometimes things are called lies that are not. If someone says something they truly believe to be the truth and are later proved incorrect by subsequent evidence, they are not lying; that is they are not telling a deliberate falsehood. An example of this is that early in the pandemic, Dr. Fauci and other experts at the CDC said that universal mask wearing was not necessary. As evidence accumulated, it became clear that this was incorrect; that universal masking markedly reduced coronavirus transmission. Their initial recommendations were incorrect as proved by later evidence, but they were not lies.

Anecdotes

We have all seen articles such as these online: “Physician dies after receiving COVID vaccine”; “Patient gets infected with COVID-19 two weeks after second COVID vaccination”; “Study shows hyrdoxychloroquine prevents COVID-19.”

These are all examples of anecdotes. Anecdotes are stories, based on personal experience or reports of results based on small non-representative samples. Often these stories are dramatic as shown in the examples above.

Our brains are hard-wired to look for patterns. This was very useful in the evolution of our species, because not recognizing patterns was much more likely to be fatal! Unfortunately, this tendency to look for patterns causes us often to see patterns where there aren’t any. That is why people are predisposed to believe dramatic stories that are based on little or no evidence.

Here are a couple of examples not related to COVID-19:

Flu shots can give you the flu

As a family physician I saw this one every fall. Flu shots are given at the beginning of the season when respiratory viruses are becoming more common. Inevitably, some people would get sick with a viral infection shortly after receiving their flu shot. Flu shots, of course, cannot cause flu. They are made of pieces of viral proteins that cannot infect cells in the body. Such is the power of anecdote though that all my scientific explanations made little difference to people who were convinced that the flu shot caused their viral infection.

Childhood immunizations cause autism

Symptoms of autism generally begin about age two. By that age almost all children have received several immunizations. Once again, this is the power of anecdote. Many parents of children diagnosed with autism are convinced that immunizations caused the autism. This hypothesis has been studied in several very large, well designed studies and thoroughly disproved. Many of these parents remained convinced because of their anecdotal experience that immunizations caused their children’s autism.

Physicians and anecdotes

Physicians are not immune to the power of anecdotes. As a matter of fact many continue to prescribe medicines or treatments that have been shown to be worthless or even harmful. Their excuse is almost always that “in my own experience this treatment works” or “I have never seen these harmful side effects in my patients on this treatment.” I have been guilty of this myself. Years ago a study from Canada showed that a simple exam of an ankle injury could rule out a fracture and save getting x-rays on all of these. The first time I tried it, the exam showed no evidence of a fracture, but the patient ended up having a broken ankle. Despite the fact that this was a perfect example of anecdotal evidence, I got an x-ray of every ankle injury for a long time after that.

Evidence

Evidence, as opposed to anecdotes, comes from systematic studies of large groups of people. Evidence helps answer important questions related to health and disease. There are several forms of evidence that are useful. The kind of evidence depends to some extent on the question. Below are brief descriptions of studies that helps us understand and better treat illnesses such as COVID-19.

Double blind randomized controlled trials

There is only one kind of evidence that gives us confidence that A causes B, or that A does not cause B. It is called a double blind randomized controlled trial. Here is how it works. A large number of people are recruited that are representative of the population who might use the treatment. The subjects are randomized into two groups by the equivalent of flipping a coin for each person. Instead of a coin, a computer does the equivalent of a coin flip. Dividing subjects in this way means that each person has an equal chance to be included in either group. One group will get the experimental treatment and the other group, called the control group will get a placebo that looks just like the treatment medicine but does not have any biologic activity (a sugar pill for example). The subjects don’t know whether they are in the treatment group or the control group. They are “blinded” as to which group they are in. The medicines are labeled with a code, so that not even the investigators know who is getting the treatment and who is getting the placebo. That is the investigators are “blinded” to who is in the treatment group and who is in the control group. That is what double blind means. Neither subjects nor investigators know who is getting the treatment and who is getting the placebo.

What is the reason for all this “blinding”? It is called the placebo effect. In any trial there are a number of people who will get better even when they get a placebo. These are not just psychological effects. When the brain thinks it may be getting a medicine, hormones and neurotransmitters show real physical changes and that makes people feel better. People who are giving the medicines can unwittingly increase the placebo effect, which is why the investigators are “blinded” too. At the end of the trial the code is broken so investigators then know who got the treatment and who got the placebo. If the treatment group has statistically more positive results than the control group, then we know the treatment works. If the control group has about the same improvement as the treatment group, then we know that the treatment does not work.

Anecdotal evidence suggested that hydroxychloroquine might work to treat or prevent COVID-19. A double blind randomized trial though showed no effect on either prevention or treatment of COVID-19.

Anecdotal evidence suggested that high doses of vitamin D might prevent COVID-19 or make it milder. A double blind randomized trial showed no effect of vitamin D on prevention or severity of COVID-19

COVID-19 vaccines were also tested in double blind randomized controlled trials. That is why we are confident that they work very well at preventing infection, hospitalization and death.

Double blind randomized trials are expensive and take a long time to get results. It is not practical to do double blind randomized trials on every scientific question. Also, there are some questions that a double blind randomized controlled trial cannot answer. Sometimes a randomized controlled trial would be unethical. That would be the case if using a placebo group would clearly cause harm to that group.

There are other kinds of evidence that help us decide whether it is worth it to do a double blind randomized control trial or that provide useful evidence when placebo controlled trials are impossible or unethical.

Case-Control Studies

This kind of study is used to answer a question about whether exposure to something causes a disease. The. “exposure” is measured in people who already have the disease (cases) and in a group of similar people who do not have the disease (controls). Unlike randomized controlled trials, case-control studies cannot tell whether the exposure causes the disease. They can only show that the exposure is associated with the disease. There is always the possibility that the difference between the cases and controls is related to some difference between cases and controls other than the exposure being measured. Years ago a case-control study showed an association between coffee drinking and pancreatic cancer. People who drink coffee were much more likely to smoke. It was the smoking that caused the increased risk of pancreatic cancer, not the coffee drinking. Systematic differences between cases and controls other than the exposure being measured are called confounders. Confounders that we know about, such as smoking, can be removed from analysis by statistical techniques. In a case-control study, however, there is always the chance that there are confounders we don’t know about. That is why we can never use a case-control study to say the exposure causes the disease or condition.

Another potential problem with case-control studies is called “recall bias.” Let’s say that the exposure we want to measure is sugar intake and the disease we want to measure is diabetes. We ask people in both the diabetic group. (cases) how much sugar they eat in a month and ask the same question for non-diabetics (controls). People are unlikely to remember accurately the amount of sugar they ate in a month. People may tend to minimize how much sugar they eat. This will obviously make the results of the study much less reliable. People tend to remember (recall) things selectively. That is recall bias.

Population Studies

These studies look at whole populations that live in a particular area. They are similar to case control studies in that They look at an exposure (say intake of saturated fats) with a disease (say heart disease). If consumption of saturated fats is different in whole populations of different countries then the incidence of heart disease in those populations will be measured as well.

The best population studies follow populations over time. Like case-control studies, they can only show association, not causation, but sometimes they can be very useful anyway.

The most famous population study, the seven country study by Ancel Keys followed populations in seven countries over many years. This study was the first to show that animal fat intake was associated with heart disease and that high blood pressure was associated with heart disease and stroke.

The Framingham Study followed the population of Framingham, Massachusetts for many years. In addition to confirming the findings of the seven country study, It showed for the first time the association between diabetes and cigarette smoking with heart disease.

Peer Review

Until the pandemic the main way that results of evidence based studies became known to the public was publication in scientific journals. In order to get a study published in a reputable scientific journal, it first has to be looked at by one of the editors of the journal. The editor decides if the study meets the criteria for that journal, and many papers get rejected at this stage. If the editor thinks the study is done well and fits the criteria for the journal, then the editor sends the paper to at least three experts in the field of the authors of the paper. They critique the paper, make suggestions to improve it, and send their comments to the authors and the editor of the journal. Many more papers get rejected at this stage. Sometimes the authors are asked to re-submit their paper after making changes suggested by the reviewers. This whole process is called peer review. This process assures that for the most part only well done studies make it to publication. The other quality control in scientific studies is replication. One study, no matter how well done, may have missed something. If other peer reviewed studies come up with the same results, then we have increased confidence that the findings are real.

Pre-print servers

The peer review process takes a lot of time. It is usually months, sometimes many months between the completion of a study and its publication in a peer reviewed journal. In the midst of the pandemic that was too long. Physicians who were treating desperately ill patients with a new disease needed to know the results of trials quickly. Scientists could publish their results online as soon as the study was completed and before submission of their studies to peer reviewed journals. The websites that allow them to do this are called pre-print servers. This process allows the results of studies to be available to doctors very rapidly, but the quality control of peer review is missing. Results published on pre-print servers should be considered preliminary. Many of them will not make it through the peer review process when they are submitted to scientific journals.

The Media

Most non-scientists do not read scientific journals or pre-print servers. They find out the results of scientific studies through the media. The media serve an essential function in translating the jargon of scientific papers into language that most people can understand. While there are a few excellent journalists who understand how to evaluate scientific papers, for the most part that is not the case. The media look for things that are splashy and they are just as likely to trumpet preliminary findings from pre-print servers as they are to discuss peer reviewed papers in journals. Even if they discuss the findings of peer-reviewed papers, they often emphasize the positive and do not report the author’s caveats about how carefully to interpret the findings.

I refer you back to my previous post about reliable media sources of medical information

Bottom Line

Lies are deliberate untruths, but when lying comes from sources that we usually can trust, that creates distrust of all sources, even ones that tell the truth.

Anecdotes are compelling stories based on personal experience or reports of small non-representative groups. Our brains are programmed to look for patterns, even when those patterns are figments of our imagination.

Reliable evidence is based on trials of large numbers of people who are representative of the population at risk of disease. Double blind randomized controlled studies are the gold standard of reliable evidence, but other kinds of studies can give good information as well.

Media reports of scientific studies can be useful, but definitely need to be taken with a grain of salt. Some media sources are much better than others.