Important Note

Tufts ended funding for its Open Courseware initiative in 2014. We are now planning to retire this site on June 30, 2018. Content will be available for Tufts contributors after that date. If you have any questions about this please write to

Tufts OpenCourseware
Author: Richard Glickman-Simon, M.D.
  • Provide a definition of health, illness and disease and describe various ways to quantify each
  • Identify the multiple genetic and environmental determinants of health and describe how virtually all outcomes result from their combined effects
  • Explain how an individual’s adaptation to stress, or lack thereof, influences the full range of health outcomes
  • Describe the ways in which physicians can and cannot affect the major determinants of health
Color Key
Important key words or phrases.
Important concepts or main ideas.

1. Introduction

What makes people sick? What ever it is be glad for it. Medical school would otherwise be a big mistake.

When it comes to illness, physicians deal far more with the consequences than with the causes.We don’t generally probe deep enough to find out why our patients are sick. Say, for example, a patient suddenly develops slurred speech and weakness in her right arm and leg. We would routinely diagnose a stroke and begin to treat to her. We usually explore a little further and find out that the stroke most likely resulted from long-standing hypertension and atherosclerosis. Sometimes our investigation ends there, though often we go on to inquire about the patient’s lifestyle and family history to see how they may have contributed to her high blood pressure and serum cholesterol levels. Then we stop. We generally don’t ask, for example, what accounted for her high fat diet, cigarette smoking, lack of exercise or genetic predisposition to cerebrovascular disease. In other words, we focus on the proximal associations rather than underlying etiologies of disease. We rarely try to get to the bottom of the problem.

This is not out of laziness or apathy. There are good reasons why we can often only scratch the surface. What are they? Why not dig deeper?

2. Are you healthy?

Before we can discuss the determinants of health, we need to consider what health is and how to measure it.

Assume for a moment you are healthy (which your probably are.) Why? What does it take to be healthy? This is no idle philosophical exercise. If you don’t know what it looks like, how are you going to recognize health when you see it in your patients? If health were simply the absence of disease, life would be easy. Identifying relatively disease is easy, defining health is not.

A patient comes in complaining of fatigue. You find nothing wrong. At what point would tell him he’s healthy? The next patient comes in for a routine physical. She feels fine. What information would you need before you declared her healthy?

For the most part, physicians are trained to identify and treat disease. That’s when we are most comfortable. Sometimes, however, the exclusive pursuit of disease can become a negative determinant of health. A “healthy” patient with no obvious disease, for example, may undergo multiple screening tests to look for one. Preventive medicine is based largely on the principle that disease often masquerades as health, and physicians ought to inform their patients that they are not as healthy as they may think. While identifying a life threatening disease before it becomes symptomatic can be of great benefit to a patient, care must be taken not to pursue an asymptomatic disease when there’s no reasonable chance of finding it.Looking for a disease in a “healthy person” can sometimes harm them.

2.1. Measuring Health Outcomes

Our definition of health influences how we measure it. Being able to reliably measure health is important for at least three reasons.

  • First, it makes it possible to follow your patients’ progress over time and determine whether they still need your care.
  • Second, in order to pay for medical products and services, it is necessary to assign some value to a quantifiable unit of health.
  • And third, well-defined health measures are necessary to conduct epidemiologic and clinical research, without which we could not practice scientifically-based medicine (the subject of a subsequent lecture).

Death, of course, is the ultimate measure of poor health. Mortality rates are particularly useful when it comes to measuring the health of a population, but clinicians understandably prefer to rely on other (more reversible) health indicators. Morbidity, a term often used by epidemiologists to describe the burden of disease in a population, is expressed as the incidence or prevalence of a diagnosable condition of interest. In a clinical context, morbidity generally refers to disease states having serious medical implications. While this is a useful measure of a disease’s existence, and to some degree its severity, it says little about the impact it has on a patient’s overall health. The most useful health measures capture a patient’s level of disability or the quality of their life. Since they are more subjective, however, disability and quality of life are more difficult to quantify and compare, compromising their usefulness, particular in research.

Another way to think about measures of health outcomes is to divide them into two types: disease-oriented and patient-oriented. As the name implies, disease-oriented outcomes reflect the status of a disease process, which may or may not be of immediate concern to the patient. Let’s return to our stoke patient. If a researcher wanted to know whether or not a drug is more effective than a placebo at treating hypertension, the most expedient outcome to measure would be blood pressure. Such a study can be carried out in a matter of months. What ultimately matters, however, is not the patient’s blood pressure but whether or not the medication prevents the major complication of hypertension, namely stroke.

To determine this, researchers would need to follow subjects far longer and in greater numbers to see if the medication could effectively lower the risk of stroke. While blood pressure reduction is disease-oriented, stroke risk reduction is patient-oriented. Even though, out of necessity, most clinical research is disease-oriented, physicians often extrapolate data from large epidemiologic studies, which suggest that the average stroke rate in populations with uncontrolled hypertension is higher than in normotensive populations. Due to the well-known flaws inherent in epidemiologic studies, of course, these assumptions may prove to be incorrect.

Always remember, it is the patient-oriented outcome that matters most. Physicians spend their days diagnosing, treating and preventing diseases. Our patients, on the other hand, suffer illnesses, which represents their perception of a disease process – if one exists. While the medical view of their disease often overlaps with the patients’ experience of it, in many cases there is little correlation between the two. By focusing our attention exclusively on disease, we may miss our patients’ illnesses.

Can you think of a situation where you might identify a disease where there is no illness? How about the reverse, an illness with no disease?

3. What Determines Health?

This is an extraordinarily complex question, one that does not lend itself to our customary way of linear thinking. Sometimes event A does lead directly to health outcome B. In most cases, however, this simple relationship does not exist. When a basketball player, for example, twists his ankle after landing on another player’s foot following a rebound, it is clear that the forceful inversion of his foot tore the fibers in one or more ligaments leading to an ankle sprain. This simple cause and effect relationship, however, would never apply to a patient arriving in the emergency department with crushing chest pain from a myocardial infarction. While there may be a straightforward linear relationship between the sudden thrombosis (blood clot) of a coronary artery and death of myocardial tissue, the aggregation of factors leading up to that event is complicated, obscure and impossible to know for certain. We do know, however, that on that day numerous conditions in the patient’s past combined in such a way as to increase his probably of a having myocardial infarction to 100%. These conditions could have occurred as long as ago as 100,000 years or as recent as breakfast that morning. As the time separating two events increases, it becomes more difficult to formulate linear associations between them, let alone do anything to change their interaction. This is why in medicine, the more recent the predisposing condition or risk factor, the more attention we pay to it.

It is possible to obtain some insights into these complex relationships by viewing various health determinants along a timeline. As a species we are prone to experience certain health outcomes under a given set of conditions due to our evolutionary history. Some have argued, in fact, that the existence of virtually all human disease ultimately has an evolutionary explanation (all of which, by the way, are difficult if not impossible to test). On an individual level, the genes we inherit transmit this evolutionary vulnerability from one generation to the next. While we will never be able to change out evolutionary history, some believe we are rapidly approaching a time when will be able to change our genetic inheritance.

Let’s assume, for the time being, that we come with a fixed genome. This package of genes predisposes each of us to specific health outcomes – good and bad. Over a lifetime, these genes encounter a range of environmental conditions (some quite extreme) along with opportunities for adaptation (some quite limited). While some genes are relatively stable, others show considerable plasticity, switching on and off in response to constantly changing environments. The vast majority of our health outcomes are a manifestation of our genetically-influenced response to the nature and timing of these environmental exposures.

Depending on our genome, one environmental hit at a vulnerable time can have a greater impact on long-term health than multiple hits over decades. Witnessing a tragic event in early childhood, for example, may lead to decades of ill health, while smoking a pack of cigarettes a day for sixty years may produce no discernable health consequences. As you can see, in virtually all cases it is neither nature nor nurture. It’s both.

4. Environmental Determinants

Listed below are the major environmental determinants of health:

  • Lifestyle choices or health behaviors
  • Pathogenic or toxic exposures
  • Health care
  • Socioeconomic status
  • Stress

While you will often see them listed separately, it is impossible to isolate the influence of one determinant from all the others. Take health behaviors as an example. Individuals who practice destructive behaviors like smoking, overeating or excessive drinking are often assumed to do so out of choice. They continue these behaviors despite their risk either because they don’t know better, or they know the risks but choose to ignore them. Or so the story goes. This kind of oversimplification, based on proximal notions of cause and effect, is not only naïve but potentially harmful. Physicians may wrongly decide that in order for patients to adopt healthful lifestyles, they simply need to be educated. Or even worse, they may blame these individuals for their unhealthful habits and the illnesses they cause. The truth is, health behaviors are the end result of complex interactions between multiple factors, many of which will never be identified or addressed in the clinical setting.

If you had to choose one of the five determinants listed above, which do you think has the broadest affect on health outcomes? Where would you rank the relative influence of health care?

4.1. Socioeconomic Status and Health

There is little debate that socioeconomic status (SES) has a profound affect on health. You needed look any further than developing countries to see how abject poverty leads to high infant mortality rates and low life expectancy. Until recently, it was assumed that once a country was able to meet the basic human needs of the majority of its population, the health status of most citizens would be roughly the same, particular in countries where access to medical services is fairly uniform. But numerous epidemiologic studies over the past 25 years have shown that a continuous gradient, rather than a single threshold effect, describes the association between SES and health, even in populations well-above the poverty line. In societies able to meet the basic health needs of its citizens – plentiful sources of safe food and drinking water, effective sanitations systems, a public health infrastructure and ready access to clinical services – the health of individuals occupying adjacent socioeconomic stations will differ. And, for the most part, this phenomenon is condition-independent; even slightly higher SES means lower mortality and morbidity from the vast majority of chronic diseases.

There are only a handful of conditions whose prevalence increases with affluence. Can you think of any examples?

What still remains a mystery is how exactly this SES gradient translates into health outcomes. So far, none of the various markers commonly used to measure SES – income, education, job status and neighborhood – has adequately explained the association, alone or in combination. This suggests that there is at least one other underlying factor common to both SES and health yet to be identified. One possibility has to do with concept of social divisiveness. Accumulating evidence suggests that a reliable predictor of the health status of a society is how its members treat each other.This may partly explain why we Americans are not as healthy as we are wealthy. Hierarchical societies sharply divided along class lines, like in the United States, tend to have higher infant mortality rates and lower life expectancies, then more egalitarian societies like in Sweden. According to this model, what matters is not the uneven distribution of material resources, but the harmful psychosocial consequences of a society that tolerates – even encourages – the alienation of social classes.

4.2. Stress and Health

While this model provides a intriguing and plausible explanation for the link between SES and health, it fails to identify the mechanism by which social divisiveness actually leads to poor health. Hypertension, for example, is more common in African Americans than whites, and it is certainly conceivable that the long history of slavery and racism in this country contributes to this disparity. Even if this turns out to be true, what we still don’t know is exactly how racism raises blood pressure. Adaptation to stress may be one possibility.

Imagine for a moment that you are living in the Stone Age, some 10,000 years ago. Your day largely consists of gathering food for your family to eat. Life is stressful, but you know how to survive, and you take pride in your ability to handle whatever comes your way. Now imagine that it is dusk, and you and some companions are returning home from a long, exhausting day of hunting. You suddenly encounter a small pack of startled wolves obviously threatened by your presence. A violent confrontation is imminent, and to survive, you must quickly assess the situation and prepare to take action. You and your companions know exactly what to do. You reflexively coordinate your positions, ominously raise your weapons and prepare for a fight. Unconsciously, your neuroendocrine system springs into action. A rapid surge in epinephrine and cortisol from the adrenal glands sets in motion a widespread set of physiologic changes coordinated to enhance your chances of survival.

The standoff continues for about a minute before the wolves retreat. As the danger subsides, so does the stress response, which is highly adaptable to a rapidly changing environment. As quickly as it turns on, it turns off once the threat has passed. You proceed home, a little shaken, but otherwise feeling fine as your stress hormone levels return to normal.

This scenario makes two important points.

  • First, stress is normal. For our entire evolutionary history life on earth has been stressful, and the fact that we are here to talk about it means we have successfully adapted to it. Even today there are many examples where stress produces a positive, or adaptive, effect. You already know that a certain level of stress prior to an exam, for example, tends to enhance performance. You also know that stress can have the opposite effect, leading to disastrous consequences.
  • This raises the second point, stress is adaptive only in so far as it is contained and appropriate to the situation. Our prehistoric counterpart presumably lived on to fight another day because he or she was able to mount the stress response quickly and appropriately, and then extinguish it just as fast.

Here’s the problem. While the physiology of this fight-or-flight response presumably hasn’t changed much in thousands of years, our behavioral response has changed dramatically. Often the result is a full blown fight or flight response with nowhere to go – we neither fight nor fly. And, while we may never encounter the random pack of wolves, we still confront major stressors, many of which occur frequently or even continuously over a lifetime. Repeatedly taking stressful hits without the opportunity for an adaptive behavioral response translates into a continuous barrage of stress-related hormones and neurotransmitters. Considerable experimental evidence suggest that the accumulated effects of this psycho-physiological reaction, referred to as allostatic load, leads to widespread damage in virtually every organ system in the body, and is probably involved in the pathogenesis of most chronic diseases.

Chronic stress can contribute to allostatic load in two ways: physiologic and behavioral. In those individuals genetically predisposed to vascular disease, for example, chronic exposure to the neuroendocrine products of stress may lead directly to the development of hypertension and accelerated atherosclerosis. And maladaptive behaviors associated with psychological stress – tobacco use, excessive alcohol consumption, unhealthful diets and non-compliance with anti-hypertensive medications – can clearly exacerbate a genetic tendency toward hypertension and stroke.

It is important to keep in mind that stress is not the problem. It’s the maladaptive response to stress that leads to disease. Physiologists use the term stress to describe any environmental exposure that challenges an organism’s homeostasis. A more useful definition of stress for our purposes is a psychological response to a perceived environmental challenge, real or not, that a person feels incapable of coping with.Perception is key. While some situations are universally stressful (loss of a loved one, for example), in most cases the experience of stress has far less to do with the stressor than with an individual’s adaptability to psychological and physical challenges.One person’s stressor is another person’s opportunity.

4.2.1. Stress Management

There are three basic ways to avoid the damaging effects of stress. Eliminate or minimize stress in the first place

The first is to endeavor to eliminate, or at least minimize, stressful exposures in the first place. Even if this were possible, spending energy on trying to escape life’s inevitable challenges is wasteful and self-defeating. The problem, after all, is not the stressor; it is the inability to adapt to it. Respond to stress immediately

The second approach is to deal with the stress response once it occurs. This is most commonly accomplished with drugs. Medications like benzodiazepines effectively blunt the perception of stress, but there is no evidence that they diminish its damaging effects over time. β-blockers are a class of medications that directly interfere with sympathetic stimulation throughout the body, and unlike benzodiazepines, β-blockers have been shown to decrease the damaging effects of stress on the cardiovascular system. Finally, various mind-body interventions can permit patients to gain access to, and control over, their own autonomic nervous system. While there is considerable experimental evidence that techniques like meditation and biofeedback can affect stress physiology, there is little clinical or epidemiologic evidence that these interventions actually lead to favorable health outcomes over the long-term. Learn ot constructively adapt to stress

The third option, which lies between the first two, is to help patients constructively adapt to the inevitable stressors they are sure to encounter. This approach makes a lot of sense, since stress is ultimately a problem of adaptation, rather than simply a physiologic response to an environmental exposure. Cognitive-behavioral therapy (CBT) is probably the best known and most widely studied method for influencing stress adaptability. Recognizing that maladaptive stress responses involve harmful attitudes and behaviors, cognitive-behavioral therapists help patients think differently about their stressful encounters and act in a way to promote health. While CBT has been used extensively in treating psychiatric diagnoses, its use in the prevention and management of medical conditions is a relatively new application.

5. Affecting Health Determinants

Given the enormous complexity of health and its determinants, what can a linear-thinking, busy clinician realistically expect to achieve? Oftentimes, a simple intervention is enough. Putting an ace wrap on a sprained ankle and taping it up before the next game is all there is to do. At other times, we can only scratch the surface of what’s really going on. While we cannot be reasonably expected to recognize, let alone manage, every possible health determinant, we can always ask these three questions for every patient we see:

Why is my patient sick?
What can I realistically do about it?
And, if I can’t help, who can?

Let’s turn one last time to our stroke patient. By the time you leave medical school it will become clear what you are expected to do for patients like this. Manage the hypertension and respond to the stroke when and if it occurs. This is the minimum. How much farther you decide to go is your call.

6. Ancillary Material

6.1. Readings


6.2. References

  • McCally M. Poverty and ill health: physicians can, and should, make a difference. Ann Int Med 1998; 129:726-33
  • Adler N, et al. Socioeconomic status and health: What we know and what we don’t. Ann NY Acad Sci 1999; 896:3-15.
  • McEwen, et al. Protective and damaging effects of mediators of stress. Ann NY Acad Sci 1999; 896:30-47.
  • Baum A, et al. Socioeconomic status and chronic stress: Does stress account for SES effects on health? Ann NY Acad Sci 1999; 896:131-44.
  • Taylor S, et al. Health psychology: What is an unhealthy environment and how does it get under the skin?