Compression of morbidity is achieved through the reduction of behavioral risks.

The compression of morbidity hypothesis was introduced in 1980 as a challenge to the then current theory of aging, which suggested that the human life span was indefinitely extensible but that human morbidity (disability) had a fixed age at onset.

This theory was termed “the failures of success” and held that science would achieve ever greater longevity while the onset of disability remained constant, thus increasing chronic ill-health at the end of life.

In this article, I introduce and elaborate on the compression of morbidity and how best to achieve it. I then suggest ways the pharmacist may contribute to the delay of aging and thus to compression of morbidity.

The negative and pessimistic view of human aging was, in turn, challenged by behavioral theories that aging processes were surprisingly modifiable and that disability could be delayed by interventions that improved skill sets and thus individual performance levels. The common secret for increasing skills was practice, practice, practice.

Importantly, the new positive health terms (compressed aging, active aging, wellness, successful aging, delayed aging) were strongly influenced by the World Health Organization definition of health as “a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity.” These positive health concepts focus on improving health, not disease.

Compression of morbidity, however, is not the only possible health future. Individuals do not benefit equally from scientific advances, and some are less healthy than their parents. Some nations and some historical periods show worsening health. There is war, pestilence, famine, and epidemic infections. There are new drugs to use and abuse, and environments to clean up or to contaminate. Compression of morbidity is a desirable but not inevitable health outcome.

The first online figure depicts 3 scenarios of lifetime disability. The upper scenario represents disability in a typical life span, with most of this disability toward the end of life. Disability often becomes detectable at 50 to 60 years of age, increases over time, and ends with death at 75 to 80 years of age.

The middle scenario also assumes that disability is first seen at 50 to 60 years of age, but that the age of death is postponed to perhaps 95 to 100 years. There is a longer morbid period and greater lifetime disability. This is “expansion of morbidity,” or “the failures of success.”

The third scenario postulates that the age of disability onset is delayed more than is the age at death. Disability then is squeezed between a delayed age at onset and a less postponed age at death. The morbid period is shorter, and lifetime morbidity is decreased even though life expectancy has increased. This is the compression of morbidity. It usually requires postponement of the onset of disability, represented as the first arrow in each scenario.

We and others sought proofs of the compression of morbidity. Study designs that guided our research include the longitudinal Framingham Study, the National Long-Term Care Study, and the National Health Interview Study. Since we needed longitudinal data over many years, we needed the study designs which required the greatest effort, discipline, and time. We needed to select controls with high educational attainment, access to high-quality medical care, and sufficient affluence to avoid many of the adverse outcomes resulting from poverty rather than aging.

For the runners study, we also sought persons who had made a commitment to lifetime vigorous exercise and thus were likely to exercise vigorously over a long time period. These runners ran an average of 25 miles (about 4 hours) each week.

If running and other vigorous aerobic exercise did not yield detectable health benefits in these persons, then any benefits must have been small. If heavy running ultimately was deleterious to health (and some had suggested that excessive wear destroyed joints), then such effects also should be easily detected.

The Runners Study: Long-Distance Runners versus Controls

In 1984 we enrolled nearly 1000 subjects for the runners and controls study, which continues in 2015. Nearly all had college or graduate degrees, lived in the same affluent communities, held professional jobs, had stable families, did not smoke cigarettes, and averaged 58 years of age. The runners were members of a runners club and had logged an average of over 12,000 running miles at study entry, over 10 times the exercise history of the controls.

Controls were selected randomly from the same community and matched for age and gender. Subjects were assessed yearly, and over 85% of living subjects remained enrolled after 23 years of observation. Deaths were ascertained by the National Death Index and are believed to be completely accurate.

We now have observed the development of disability and the occurrence of death in runners compared with controls over more than 23 years, to an average age of 83 years. Both groups were overwhelmingly nonsmokers and had similar and normal body mass index values. The groups differed materially only with the exercise variables.

Disability development in runners and controls is shown in Online Figure 2. Differences favoring the runners group increased steadily over time, continuing to increase after more than 20 years. Note that postponement of death or disability is estimated by comparing the lines as postponement on the horizontal axis, not the vertical; this is longitudinal, not a cross-sectional, study.

Figure 2: Disability Index Scores by Age and Calendar Year, Runners Club (N=538) and Community Controls (N=423) 1984-2005

Disability was postponed 16 years in the runners compared with the controls. These differences persisted after statistical adjustment, and were seen in both women and men and in “ever-runners” compared with “never-runners.”

Mortality in controls (higher risk) and runners (lower risk) over 25 years is shown in Online Figure 3. Analyses over 25 years comparing the higher-risk controls and the lower-risk runners show a 7-year postponement of mortality in the runners. After 25 years, runners also have had only 60% of the mortality of controls.


Figure 3: Mortality in Community Controls (Higher Risk) and Runners Club (Lower Risk) 1984-2009

Runners thus have had a 10 to 16 year postponement of disability and only a 3.3- to 7-year postponement of death, documenting compression of morbidity. It should not be surprising that runners are more fit and less disabled than more sedentary controls, nor that vigorous exercisers should outlive more sedentary controls. There are of course thousands of published studies of all types that document that exercise is good for you.

The dominant findings here, however, are that the postponement of mortality in runners is not nearly as long as the postponement of disability and that the differences between runners and controls continue to increase after a quarter century. This substantial compression of morbidity is associated with very large differences in physical exercise levels.

Thus, morbidity is compressed by exercise. To our knowledge, this is the greatest compression of morbidity yet documented, and is far greater than that reported in studies of populations or medical interventions.

The comparisons of the runners and control cohorts were conservatively evaluated. We used control subjects that had previously been healthy controls for the Lipid Clinics Study. They were a hardy and healthy bunch, and a quarter of them were also runners. Runners who were included in the control cohort were retained permanently in that cohort, ensuring a conservative evaluation design.

We searched for self-selection bias by statistical adjustment for possible confounding variables, such as injuries, gender, age, education, birth defects, leg length, ever-runners versus never-runners, comorbid diseases, spousal reports versus subject reports, family history of diseases, and study dropouts. We found no sign of selection bias in these or other variables.

In a confirmatory study, we had similar results when we studied 2000 college alumni from 69 years of age for the risk factors of inactivity, smoking, and obesity, alone and in combination. These 2 longitudinal studies (see “For Further Reading” online), performed under National Institutes of Health protocols, document that lower levels of behavioral risks compress morbidity.

These new data reported here suggest that the earlier in life a postponement intervention is begun, the greater its likely effect. Compression of morbidity is based on postponing the onset of disability (Online Figure 1) to the greatest degree possible. Results cannot precede an intervention. Thus, there now are proven preventive measures that can delay the processes associated with human aging, and the most important of these appears to be exercise.




Reducing Behavioral Health Risks

Over the past 35 years, the compression of morbidity hypothesis has become the compression of morbidity paradigm. Among the many lines of evidence are these longitudinal studies with large effects upon delaying morbidity. Morbidity compression also has been documented in population studies of US senior populations, notably the National Long Term Care Study from 1972 to 2004 and the National Health Interview Survey over similarly long periods. It receives strong support by health promotion studies such as the randomized control Bank of America study, with over 50,000 subjects, in which reducing behavioral health risks was successful in reducing the need for medical care.

Large national reductions in smoking behavior in the United States have been strongly linked to large reductions in lung cancer, and substantial reductions in heart attacks and cardiovascular deaths. Medications such as low-dose aspirin, lipid-lowering statins, and angiotensin-converting enzyme inhibitors substantially delay heart attacks and therefore must certainly compress morbidity.

Over the same period, there has been expansion of morbidity elsewhere, seemingly resulting from increases in health risks that are causally related to public health failures. The most obvious is the increasing pandemic of obesity, but decreases in vigorous exercise and increasingly sedentary lives spent in front of television and computer screens may be having similar effects. Serious exercise commitment delays morbidity by as much as 16 years, far longer than the postponement of mortality. Nonsmokers have greater longevity and less morbidity than smokers. Non-obese individuals live longer and spend less time in the hospital.

Currently we have solid proof of concept of both the compression of morbidity and of the expansion of morbidity, related to opposite trends in major health risks, the most important of which is lack of adequate exercise. Vigorous aerobic exercise is a unifying risk reducer; it is associated with less smoking and less obesity. It has abundant proven mechanisms of action. It increases reserve in cardiac, pulmonary, muscular, and other organ systems; it postpones many diseases, from heart attacks and strokes to blood clots and diabetes. It appears to reduce depression and to improve feelings of well-being. It delays loss of ability to perform activities of daily living and postpones the development of frailty.

Compression of morbidity requires postponement of the behavioral prodromes of chronic illness, beginning well before the onset of clinical morbidity. Preventing childhood obesity, not smoking cigarettes, avoiding severe sunburns, continuing exercise after the schoolyard—these are primary and secondary prevention measures. Postponing morbidity by prolonging pre-morbidity tells us much about the necessary public and personal health actions. These approaches are less expensive than medical approaches to health improvement. But there will always be newer threats of increasing morbidity, and they may come in large part from deteriorating behavioral health risk factors.

The Pharmacist on the Health Care Team: A Focus on Postponing Aging

In this era of rapidly evolving medical care, an expanded role for the pharmacist on an expanded medical team is an obvious need, and this can come as no surprise to the Directions in Pharmacy reader. The physician is too often busy, too focused on disease, too bored by repetition, overbooked, and often has to delegate tasks.

Improving health outcomes, on the other hand, requires multiple face-to-face encounters (reinforcement), easy access, the welcoming of questions, a focus on behavior change, access to reliable information, appropriate interim advice, and health advocacy. Someone needs to fill in the gaps in care, and the pharmacist is well-positioned for this role.

To delay aging, the intervention must precede the ill effect; it must be important, preventive, timely, and relevant. Don’t sweat the small stuff. Five under-emphasized areas of pharmacist activity are particularly important to advocate and expand, and are directed at delaying aging: (1) Behavioral risk factors increase morbidity; emphasize inactivity, smoking, and obesity. (2) Injury prevention delays disability; focus on impaired or distracted driving, seat belts, car seats. (3) Preventive medicines postpone illnesses; reinforce American College of Physicians (ACP) guidelines, which are important, proven, and neglected; emphasize low-dose aspirin, beta blockers, statins, drugs that moderate blood pressure. (4) Medication noncompliance leads toward earlier morbidity; focus on making a plan and adhering to the plan. Use reminders, pill sorters, and counters. With the physician, simplify regimens toward fewer times of administration, question unneeded medicines, obtain refills before they are needed. (5) “Snake oil” salesmen detract from wiser actions to postpone aging; they are everywhere on television and other media, behind “cold” calls, and on many store counters. Pharmacists need to encourage national ACP guidelines, counsel about direct-to-consumer ads, be skeptical of snake oil advertisements, note that unproven remedies are generally ineffective, and work to save money and add value.

These themes are at once familiar and a bit strange. They point toward taking a longer view of health outcomes. But, the actions of the pharmacist toward implementing primary prevention imperatives could contribute to postponement of aging processes and a shortening of the morbid period. 


Dr. Fries is an emeritus professor of medicine at Stanford University School of Medicine. He is nationally and internationally recognized as a leader in conceptualization of strategies to promote healthy aging, in behavioral approaches to decrease morbidity, in long-term outcome assessment, in self-management strategies, in design of efficacious and effective interventions, in evaluation of long-term behavioral interventions by randomized clinical trial, and in managing large-scale patient data collection and analysis projects. He has published over 300 articles, 11 books, and numerous book chapters and invited papers, and is a frequent keynote speaker.

In 1980, he developed the compression of morbidity hypothesis, which has provided the conceptual foundation for health promotion and healthy aging programs. The compression of morbidity hypothesis holds that primary preventive factors have a greater effect upon morbidity than upon mortality and that chronic diseases with onset later in life will be present for a shorter length of time. Recent work by Dr. Fries on this hypothesis include addresses at both the Nobel Forum and The Institute of Medicine, an editorial in
The American Journal of Public Health, a major policy paper focused on Need and Demand Reduction in Health Affairs, a recent Special Article in the New England Journal of Medicine, and a progress report on proofs of the paradigm in the Annals of Internal Medicine.

He lives with his wife, 2 horses, and a dog in Woodside, California. He has run the Boston Marathon and has climbed or attempted to climb the highest mountain on each continent.


For Further Reading

Chakravarty EF, Hubert HB, Lingala VB, Fries JF. Reduced disability and mortality among aging runners: a 21-year longitudinal study. Arch Int Med. 2008;168:1638-1646.

Fries JF. Aging, natural death, and the compression of morbidity. New Eng J Med. 1980;303:130-135.

Fries JF. The theory and practice of active aging. Curr Gerontol Geriatr Res. 2012; doi:10.1155/2012/420637.

Fries JF, Bloch DA, Harrington H, Richardson N, Beck R. Two-year results of a randomized controlled trial of a health promotion program in a retiree population: the Bank of America study. Am J Med. 1993;94:455-462.

Fries JF, Koop CE, Sokolov J, Beadle CE, Wright D. Beyond health promotion: reducing need and demand for medical care. Health Aff.1998;17:70-84.