Correlation vs. Causation In Health And Disease

correlation-causation-comicWhen we discuss various topics in health, we typically interchange the words correlation and causation. We look for common patterns and tend to conclude causation. This approach simplifies the process for discovery, but inadvertently reaches unfounded and inaccurate conclusions in many cases. When seeking “today’s truth,” we must pursue realities that reach beyond the boundaries of “correlation.”

Just because something “appears” to correlate doesn’t prove causation. Until recently, the health care field claimed a correlation existed between cholesterol in food and cholesterol levels in the blood. Without validating this correlation to be causal, the industry placed dietary guidelines recommending no more than 300mg of cholesterol be consumed on a daily basis in the 1960’s. For more than 50 years this “apparent” correlation was considered causal for high cholesterol. Today, this correlation has been completely invalidated.

Until recently the health care field claimed a correlation existed between moderate to high levels of fat consumption and cardiovascular disease. Poorly designed scientific studies concluded these dietary patterns were CAUSAL for strokes and heart attacks. Today, Ketogenic diets, Paleo diets, Atkin diets, etc… all profess to be solutions to risk factors associated with cardiovascular disease even though each is higher in fat consumption. Once again we see examples where correlation has not conclusively been the factor determining direct causation.

The germ theory is probably one of the best examples showing the misrepresentation between correlation and causation. It was determined in the 19th century that a correlation existed between “germ exposure”  (bacteria, virus, fungi, etc…) and disease development. We saw outbreaks and epidemics that resulted in disease and death. It seemed clear that “germs” were the CAUSATIVE factor since a CORRELATION was seen among those exposed to the germ. Once again our conclusions were wrong. They were based on fear, not reality. Today we recognize that germs alone are not causal for disease. There is certainly a correlation between germs and the potential for health maladies, but the CAUSATIVE factor associated with disease is a compromised immune system incapable of optimal performance. If germs were the “causative factor” everyone exposed to the germ would succumb to disease. How often has an entire family living together been exposed to the same germ, yet experienced different health outcomes? How many times have groups of people eaten foods contaminated with bacteria, yet only some experience the symptoms of “food poisoning.” The “germ” plays a factor, but is not directly causal for the outcome.

These three examples (out of countless numbers) aim to show how “experts” setting health care policy use CORRELATION to prove CAUSATION. Causation only exists if it occurs EACH AND EVERY TIME. Correlation, on the other hand, has a significantly lower standard that establishes a commonality to a condition. We can conclude, therefore, even a high degree of CORRELATION will NOT necessarily be the CAUSATIVE factor in an outcome. (Ex. A high correlation may be exist among poor diets and elevated blood sugar, but we can’t conclude that Diabetes is CAUSED by poor dieting.) This is an important point, because it demonstrates the complexity in understanding CAUSATION. CORRELATION may play a role, but we can’t scientifically conclude it is always CAUSATIVE.

Our new science is changing traditional thinking and expanding our outdated views of health. Today, we are moving in a direction that understands the uniqueness of each individual regardless of common anatomy and physiology. We are turning away from the concept of correlation and seeking INDIVIDUAL CAUSATION by expanding our diagnostic skills and searching for COMPREHENSIVE ANSWERS as it applies to each and every person. This new approach will seek deficiencies rather than symptoms; its treatment plans will be designed to re-balance physical, mental and emotional weaknesses rather than relying on answers solely found in life long prescriptions. It will convert our current DISEASE CARE MODEL into a true HEALTH CARE MODEL. Understanding the differences between CORRELATION and CAUSATION as it applies to health and disease is an important first step as we transition our health care system and its approach to providing quality care.

correlation_causation1

Advertisements

21 comments

  1. I understand why medicine should be shiftting to this approach. How do we as individuals apply this thinking in our lives?

    Liked by 1 person

    1. I believe it begins by realizing in many cases we don’t apply this thinking to our lives. We then need to begin asking ourselves the question, “WHY” more frequently. Answering what appears to be a simple question can unveil more hidden “truths” than we currently understand. WHY do we get sick?, WHY do we choose unhealthy foods as our main source of “nutrition?,” WHY did the senate just vote on legislation to place the consumer in the dark about foods ingredients in products? (fortunately, it did not pass!,) WHY does the United States rank 18th in the quality of care and 1st in cost?, …and so on.
      I am not advocating any political agenda, but Bernie Sanders is basing his campaign on awakening the public and getting them to participate to change the legal monopolization that currently exists allowing big business to control the health and welfare of our economy. I am trying to create a similar awakening.
      The willingness to open our minds to new ideas and the willingness to ask the question, WHY can have profound health benefits for our population. The question becomes, is the average person willing to participate and become part of the solution.

      Liked by 2 people

      1. Your final question is part of the dilemma. That statistic about us being 18th yet first in costs in disturbing. Shows what is wrong in this country. The public needs to be awakened. Keep up the good work.

        Liked by 1 person

  2. I really like this. It seems that the “experts” like to make sure the theory fits in their “box” so they can wrap it up and consider the case closed. Finding an answer seems to usurp the validity of the answer. Very thoughtful post!

    Liked by 1 person

    1. This has been the traditional approach to health. People accepted their “limited role” and believed their well being was in the hands of the doctor and the medications he or she prescribed. Recognizing “health care” has been based on a model of DISEASE rather than a model of HEALTH has begun to resonate in the general population stirring up greater dissatisfaction with the traditional model. Patients are becoming tired of answers that simply increase the number of prescriptions they must take without ever RESOLVING their underlying problems. Maintaining a chronic state of reduced health is no longer an acceptable option. It’s nice to see and experience more and more people seeking a modern day approach to improving health and quality living by following the concepts I write about. Today, these options are no longer dismissed as “alternative thinking.” History is proving their LONG TERM BENEFITS!

      Like

  3. Your best yet! really good work…keep it up!

    Liked by 1 person

    1. Thank you Frank. Just working on my mission to reach out to as many people as possible to show them a healthier and happier path in life. The hardest part is getting people to cross the STARTING LINE. Once they begin the journey and experience the benefits the results help motivate the ongoing transition. It’s a beautiful thing to witness.
      Hope everyone at home is well.
      Stay healthy and happy.

      Liked by 1 person

  4. The final cartoon is the funniest antidote I have seen to confusing correlation with causation.

    On the other hand, saying that “Causation only exists if it occurs EACH AND EVERY TIME” sets the bar impossibly high. Exact replication of experiments with people is impossible: we are too complex, diverse, and ever-changing. Exposure to tobacco smoke does “cause” lung cancer in a weaker but legitimate sense. Carcinogens shove probabilities in one direction; having a healthy immune system shoves them in the other. Determining statistical (probability-shoving) causation is often a tricky business.

    Liked by 1 person

    1. Your point is well taken, however, if (using your example) exposure to tobacco smoke caused lung cancer (rather than my opinion that it correlates well with increased chances for developing lung cancer) than greater percentages of those exposed to tobacco smoke would develop lung cancer. In my opinion, greater proof of correlation vs causation would take this example up a notch and use actual smokers. Smokers only account for 30% of all cancers. It accounts for 87% of all lung cancer deaths. As a society we are more inclined to use the word CAUSATION where the word CORRELATION more accurately describes the situation. Therefore, I would argue there is a correlation between smoking (or tobacco exposure) and cancer. Those who smoke and develop lung cancer specifically, however, can make the CAUSATION argument. Without this distinction, people tend to use the CAUSAL term rather loosely.
      I appreciate your insight as well as your view on this topic. Makes for an interesting debate.

      Liked by 1 person

      1. People do speak loosely of “THE” cause of something that may or may not happen, with a probability that could be raised by some behaviors/exposures and lowered by others. When there is a good reason to believe that X makes Y more likely, I am more comfortable with saying that X is “A” cause of Y than with just saying that X and Y are correlated. Statisticians and philosophers of science have been wrestling with the methodology of probability measurement for centuries, with no end in sight.

        Liked by 1 person

        1. I respect your position and agree that a single conclusion will unlikely ever be reached. I may personally be more sensitive to these words and concepts because of my professional relationship with patients who often blame unhealthy outcomes on CAUSAL FACTORS (ex. genetics) when the greater reality is CORRELATION (ex. obesity and environmental factors.) When people believe they are “powerless” to “causal factors” they commonly submit to poor choices justifying their actions with phrases like, “it doesn’t matter anyway.” My intent is to encourage greater participation in one’s health AND outcome. I want people to feel greater self empowerment so they are willing and able (in a larger way) to produce results of choice rather than believing they are victims of “fate”.

          Liked by 1 person

  5. Love this post too….to often we forget to question ourselves….just accept….I have always been the one in the group to say what minute….why are we doing this…thanks for another eye opener….kat

    Liked by 1 person

  6. Like you said, it only takes one additional minute to think something through BEFORE we take actions. That one minute can make all the difference in the world.

    Like

  7. What country is first in quality of care? I’m sure Canada is higher than the US, but I doubt it’s the top. I hope you write an article on how to improve immune systems one day…because mine stinks! I catch EVERYTHING. When JP is sick, I wish he’d move out for awhile.

    Like

  8. Today, France sits in the #1 position. I must also correct myself; #18 belongs to the United Kingdom, not the United States. #30 goes to Canada and #37 belongs to the United States. Nothing like paying the highest costs in the world per person and sitting in 37th place. Can you imagine if this philosophy translated into job performance. Be less productive and receive a salary increase. What is the United State’s response to this? Raise the cost of health care! Simply amazing!!

    Like

    1. Thank you.

      Liked by 1 person

  9. Great article. Thank you for sharing it. Hopefully science will catch on one day and we will once again have more actual/real research rather than ghost written articles from drug companies, whose sole purpose is to improve their financial position!

    Liked by 1 person

    1. People have a tendency to believe positions that favor their PASSIVE WANTS over their ACTIVE NEEDS. Until people are willing to assume greater self responsibility for their health, ghost writing will remain a lucrative venture.

      Like

  10. This is a very important issue close to my heart. Many thanks for bringing it up, excellent post. I come late to this post but having read the responses I can’t resist providing one myself. You are so right in pointing out the distinction between correlation and causation. However, I am a little confused by your very strict requirements for causation. Take a genotoxic carcinogen that doesn’t really have a minimum threshold concentration. Even one molecule could theoretically cause a genetic change leading to a cancer. Fortunately we have reparative mechanisms attempting to kill the faulty cell with success most but not all of the time. Where the faulty cell survives we have a cancer forming. This molecule is thus causative of cancer even if we are lucky and heal ourselves most of the time. Equally some bugs cause food poisoning. I wouldn’t trust my immune system to be able to fight them off all the time as they are clearly causative not just associated by correlation. But again thanks for the encouraging post bringing on the debate.

    Liked by 1 person

    1. I use a strict definition to minimize the comments from those claiming the science doesn’t DEFINITIVELY prove a causal outcome. Placing one’s hand in a fire will CAUSE the skin to burn.” It is a repeatable event proven across race, gender, etc… Your sentence, “Even one molecule could theoretically cause a genetic change leading to a cancer” would therefore fit more closely to “CORRELATIVE” because the cancerous outcome may only occur a “percentage” of the time. If exposure to the genotoxic carcinogen consistently resulted in a cancer diagnosis, it would be easier to universally accept this as a causal relation. Using “correlation” reinforces a significant ASSOCIATION and DANGER with cancer (in this case) and avoids confrontation from those claiming (for example) “to regularly eat heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons (PAHs)” without evidence of cancer. Just because certain chemicals CAN alter cellular genetic material resulting in cancerous outcomes doesn’t mean this factor ALONE determines a cancer diagnosis.

      The same conclusion can be made about common “food poisoning.” The bacteria certainly plays a role, but other factors determine the clinical manifestation of the disease. If this was not the case, everyone eating the “tainted” food in the same quantity would end up sick.

      If these two terms are defined more broadly, we create greater room for additional discrepancy.

      Thank you again for your perspective. Theoretically, I agree with your points. In reality (when dealing with Agribusiness and Pharmaceuticals,) they have made the “CAUSAL” argument of toxicities associated with their respective products invalid.

      Like

Your comment can positively impact the lives of others.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: