A 28-year-old woman came to our clinic complaining that she had felt depressed for two years and was certain that nothing could or would change for the better. She was living with her parents, unable to get out of the house for work or school, discouraged that she had no other relationships that she felt were meaningful, smoked three packages of cigarettes a day, and felt that suicide might well be the best solution for her situation. Vegetative symptoms of depression were present but not severe; there was no evidence of thought disorder and no abuse of alcohol or other drugs. She scored at the 91st percentile on the depression cluster of the Symptom Checklist-90 and satisfied NIMH Research Diagnostic Criteria for minor depression.
This rather common case of minor (neurotic-reactive) depression presents the psychiatrist with several possible psycho therapeutic options, including supportive psycho therapy, classic and short-term psychoanalytic psychotherapy, cognitive psychotherapy, group psychotherapy, and couples and family psychotherapy. Antidepressant medications are another possible treatment choice, either alone or in combination with psychotherapy. Despite claims by proponents of any particular school of therapy, objective studies indicate that most treatments are about equally and marginally effective,1 partially confirming Osier's aphorism: "Ii many drugs are used for a disease, all are insufficient."2
While tricyclic antidepressants and electroconvulsive therapy have been shown to be effective in the treatment of major (psychotic-endogenous) depressions, clearly superior treatments for the moderate (neuroticreactive) depressions have not emerged. When we consider the successes of somatic treatments for major depressions, there is at least a certain logic in seeking an effective somatic treatment for minor depressions.
The patient described above participated in a study of different treatments for depression and, instead of receiving psychotherapy, was randomly assigned to running as a treatment for her depression. She recovered within two weeks and has remained in remission for more than two years. Early in the course of treatment she experienced a three-week recurrence of depression when she was unable to run after injuring her ankle by exceeding her treatment prescription (Figure 1).
Figure 1 . A patient's response to running treatment. During week five the patient injured her ankle while exceeding the treatment prescription. When she stopped running until the ankle healed her depression returned. D = depression. A = Addition items (including vegetative symptoms of depression.)
What follows is, first, a review of the evidence supporting the antidepressant effects of physical activity in general and running in particular and, second, an exploratory statement of what we now feel are critical factors in the treatment technique that can make running either antidepressant or depressing.
How one arrives at the point of testing an unconventional (some would say bizarre) treatment for a common psychiatric disorder should be described. As continuing athletes ("Anyone trained or fit to contend in exercises requiring physical agility, stamina, etc."3), the authors have noticed subjective improvement of mood with some but not all kinds of exercise.
Some of us are addicted to once- or twicedaily running not because it is likely to lengthen our Uves but because these runs predictably improve the way we feel and seem to enhance our functioning in other areas. By contrast, none of us would make the claim that "all-out" competitive running has these properties. A properly run marathon, where one runs comfortably through the early miles and then engages happily with physiologic and psychological fatigue in the closing stages, may be an exception to the caveat about all-out racing - the limits are approached gradually and are the individual's personal limits rather than ones defined in competition with another. Running made us feel good, and some kinds of racing (more akin to our daily runs than many people's stereotyped image of racing) also produced feelings of deep personal awareness and satisfaction.
All of these positive experiences are transient, and to renew them one must run again. Nevertheless, the power and predictability of these effects made us wonder whether they might also operate in a beneficial manner on patients diagnosed as having depression who seemed to respond so slowly and incompletely to our psychotherapeutic interventions. We came to realize that exercise as a treatment for depression is less radical than we had been taught to believe. After all, depression manifests itself not only in the expressed and unconscious thoughts and feelings of patients (the verbal medium of communication and psychotherapy) but also in vegetative and behavioral symptoms that are often more troublesome for patients than their cognitive and affective complaints. "We realized that simply because people complain with words about their depression, it might be no more logical to treat such depressions with words than it would be to treat the verbal complaints associated with appendicitis or pneumonia by psychotherapy.
We reviewed the literature on the relationship between exercise and mood and found several studies4"9 describing work in populations of subjects with analogue depressions but none that dealt with the problem of depressed patients.
Those who conduct analogue studies are sometimes offended by clinicians' reluctance to generalize from analogue studies to work with actual patients. As one analogue researcher complained, "You don't require a model airplane to fly across the Atlantic," to which clinical researchers would answer, "a model airplane cannot fly across the Atlantic." Nevertheless, the analogue literature was generally positive and suggested that there might be objective changes in mood associated with exercise that would extend our subjective impressions to depressed patients.
Encouraged by the results of controlled analogue studies of the effect of exercise on mood and a single report from Cavanaugh about his work with patients experiencing depression after myocardial infarction,10 we decided to conduct a controlled pilot study of running as 3 treatment for depression in psychiatric outpatients.11 Patients were randomly assigned to time-limited, ??-session psychotherapy, time- unlimited psychotherapy, or individual running with a running leader who had no training in psychotherapy. Patients initially met with the running leader three times per week for 30 to 45 minutes of comfortable movement, which at the outset was often long walks linked by short, slow runs. As ten weeks of active running treatment passed, patients were encouraged to run four and then five days per week, and the running leader met with them less frequently.
Figure 2. Depression scores of patients treated with running only, and with two kinds of psychotherapy. Time-limited psychotherapy consisted of 1 0 behaviorally focused sessions. Time-unlimited psychotherapy was dynamic, insightoriented psychotherapy.
Discussion of depression was not encouraged during the runs or afterward, when a brief stretching session was introduced to avoid the tightening and shortening of posterior leg muscles that often accompanies regular running. We were pleasantly surprised with the rapidity and degree of improvement shown by depressed patients assigned to running treatment as compared with changes observed with comparable patients in psychotherapy (Figure 2). Improvement has persisted in the two years since the study was completed.
The results of this pilot study suggest that running may be an effective symptomatic treatment for at least some people with moderate depression. Patients in this study were not psychotic, actively suicidal, or receiving antidepressant medications. Those randomly assigned to running did at least as well as patients treated with either of two kinds of psychotherapy, and the cost for providing this treatment (including treadmill tests) is approximately 25 per cent of the cost of 10 sessions of individual psychotherapy with a psychiatrist.
KostrubaJa reported anecdotedly that running was effective in the treatment of depression of some patients he was caring for. Several other therapists have described single cases in which depressed individuáis have improved when they began running.
Still, we feel there are a number of difficulties with the present advocacy of running as a treatment for depression. All of the anecdotal reports are fraught with limitations that Kostrubala candidly acknowledges: "The trouble is that my sample is too small to be of statistical significance, and I am unable to separate out the factor of my own enthusiasm which often affects the outcome of a treatment modality."12 As Hans Zinnser said, "two mice are no mice at all." As far as we have been able to discover, our pilot study is the only reported work that attempts to control some of the variables that make research on selfremitting illnesses so difficult. Our pilot study was flawed in several ways,13 and so we are now conducting -another study that attempts to remedy those deficiencies.
The difficulty with any treatment lies in striking a sensible balance between blind acceptance of the first uncontrolled reports and waiting too long for incontrovertible proof, which may be very difficult to obtain. While controversy still surrounds the role of cholesterol in the etiology of atherosclerosis and continuing study suggests that carrier Iipoproteins may play an important part in the healthy and pathologic metabolism of cholesterol and, thus, in the development and prevention of atherosclerosis, few responsible physicians would advocate that their patients consume a diet unusually high in cholesterolcontaining foods or saturated fats. When evaluating new treatments, one must also consider possible deleterious side effects of the treatment should the main effect prove to be weak or nonbeneficial. As Chalmers pointed out, "If a new therapy has harmful as well as beneficial effects, the physician owes the patient a 50-50 chance of not receiving it."14
Unlike the major medical and surgical procedures Chalmers was discussing, running seems to have a potential for producing a number of beneficial side effects that would justify its advocacy even in the absence of clear proof of an antidepressant effect. While it is far from clear that exercise will extend a subject's life span beyond what he might naturally expect, given general good health, it appears that the weight reduction, lower blood pressure, increased cardiopulmonary reserve, decreased cigarette smoking, and increased relaxation that usually accompany regular exercise are likely, on a statistical basis, to reduce a person's risk of dying from a myocardial infarction "before his time."
If we are to advocate running for our patients, what kind of running treatment should we prescribe? Many regimens have been suggested, and each appears to have been successful in helping some people become regular runners. Cooper's aerobics program has been tried by several million people, and some people feel that the detailed aerobics points system for different physical activities proves initially helpful both by providing a quantitative measure of progress and by avoiding excessive activity that might lead to injury.
We have no quarrel with any approach that helps people begin exercising. Still, the rather compulsive nature of the aerobics system palls for many and quickly becomes inappropriate for long-distance runners (for example, for the large number of people who run 50 miles per week at a pace faster than eight minutes per mile, 300 points accumulate each week, fully 10 times the 30 points Cooper recommends for fitness). Drop-out rates between 30 per cent and 70 per cent are commonly reported for jogging groups, and most attrition occurs in the first six weeks. We have had a 10 per cent drop-out rate with both depressed and nondepressed beginning runners by utilizing an approach that emphasizes the pleasures and satisfactions in each day's run rather than point tallies or other, more distant, goals, such as completing a marathon.
We answer the tyro's question, "When will I be a runner?" by saying, "The moment you get both feet off the ground at some point of the movement cycle you are running." When asked, "How long will it be before I can run a mile?" we respond, "You can probably run a mile right now if you're willing to experience hurt, pain, and agony. And you'll probabJy never want to run again, and you would be right not to want to run again if you mistreated yourself in this way." We are convinced by our experience (both successes and failures) in helping several hundred people begin and continue running over the past 10 years that concern for particular goals or achievements early in a running career interferes with development of the habit of regular, comfortable running. Once a habit of enjoyable running is established the runner can begin, if he wishes, to point toward objectives outside the pleasures of the daily run.
The first time we run with someone, we inquire how many times he has tried to run before. The usual answer varies between one and six times, and when we ask the person to describe what happened, the scenario too often runs as follows:
I didn't like the way Ï looked or felt and I admired the way my friend who runs looked and seemed to be on top of her life. I expected starting to run would be tough, and I steeled myself for a lot of nausea, gut pain, and sore muscles. I realized how quickly I had quit before, so 1 found a friend who was in the same boat, and we resolved to run together every day.
The first day, we started out and were amazed how good it felt to run. After I had gone a short way, I noticed my breathing was very rapid. I had that familiar sweet-sick taste in my mouth, and my legs were beginning to feel tight. Becaue my friend was there, I pressed on and managed to hang on until her discomfort became so severe that she stopped. We both bent over, hands on our knees, panting, and were nearly sick. I remember thinking, "God, this is tough, I don't know how Jane a regular runner) does it every day."
After a while, we ran home, having to stop once more before we got there, feeling totally wiped out. I was exhausted and fell asleep quickly that night but had a nightmare about running in place and not getting anywhere. I was so sore the next day that if 1 hadn't agreed to meet my friend., I wouldn't have gone at ali.
We started out looking like walking wounded and hobbled along until, rather surprisingly, we limbered up a bit and the running began to feel pretty good. Almost immediately, though, we noticed a return of the unpleasant feelings from the day before. We knew how far we had gone the first day and managed to go a few steps farther the second day we ran.
The third day, my friend failed to show up and I was relieved to be able to return home without running to soak in a hot tub.
We tell the patients that we are responsible for the distance and pace we run together and that they are to stay with us and follow our lead. We ask them to concentrate on deepening rheir breathing as it quickens naturally with running. We also ask that they hold their spine in an erect position. (In later sessions, we introduce other kinesiotactile instructions such as, "feel yourself moving lightly over the ground . . . put energy in every step ... let your arms relax and drop down as much as feels comfortable.")*
In the first session, it's important to keep the instructions and the experience simple. We run with the beginners and proceed at a pace slower than most of them have run in their other, abortive, beginnings. When we sense that their breathing is beginning to labor (and this can be in as little as 50 feet), we begin to walk, thus modeling for them the principle of remaining comfortable and well within our physiologic adaptive limits while running. Most people are quite relieved to find that we're not going to "run them into the ground," an unexpressed fear engendered in part by previous experiences with competitive "friends" who have done just that.
We continue moving in this way, alternating walks and runs for 30 to 45 minutes, and meet with these runners up to three times per week for similar sessions. Usually within two or three weeks, runners ask whether they can run more frequently and we tell them that they certainly may as long as they're not feeling fatigued, irritable, or sore.
We also show the runners three or more stretching exercises that tend to counteract the shortening and tightening of posterior leg muscles that may accompany regular running.
While it is impossible to adequately convey the running technique we use with beginning runners in words alone, the following principles can be emphasized:
1. It is important to make each run comfortable, so that one finishes with more energy than one had at the beginning.
2. Pace and distance are dictated by bodily feedback of breathing rate and depth, leg lightness, and energy.
3. In the beginning, it is better to stay well within one's adaptive limits. This means making each run so gentle or comfortable that one is sure to look forward to the next run.
Running appears to have antidepressant properties for some persons with moderate (neurotic-reactive) depression. Evidence to date is based on anecdotes, several controlled analogue studies, and one controlled pilot study with depressed patients. Additional controlled studies are necessary before clinicians can properly assess the role of running and other exercise in the management of this and other kinds of depression.
For those who wish to prescribe a trial of running treatment for depressed patients before these studies are completed, we feel it is important to emphasize comfortable running and to minimize concerns about distance, pace, and competition with others. Using this technique, we have been successful in helping 90 per cent of the patients who come to us for running treatment to become continuing runners.
The treatment of depression with running seems to us to be neither more nor less complicated than helping people become runners and then watching with them to see whether their depression remits. It is the same with any treatment in clinical practice: one applies the treatment in the most skillful way possible and then monitors the patient's course, modifying or changing the treatment regimen as required.
1. Luborsky, L. Singer, B. and Luborsky, L. Comparative studies of psychotherapies: Is it true that "everyone has won and alt must have prizes"?. Arch. Gen. Psychiatry 32 (1975), 995-1008.
2. Osier, W. S. In Bean, W. B, (ed.). Sir William Osier: Aphorisims from his Bedside Teaching and Writings. Number 211. New York: Henry Schuman, 1950, p. 101.
3. Websters Seventh New Collegiate Dictionary. Springfield, Mass.: G. and C. Merriam Company. 1971.
4. Morgan, W, F. Psychological Consequences of Vigorous Physical Activity and Sport. In Introduction to Sport Psychology. St. Louis: C. V. Mosby, 1976.
5. Morgan, W. P. Selected physicological and psychomotor correlates of depression in psychiatric patients. RES. Q. 39 (1968), 10371043.
6. Morgan, W. P. A pilot investigation of physical working capacity in depressed and non-depressed psychiatric malea. Res. Q. 40 (1969), 849-861.
7. Morgan, W. P. Physical working capacity in depressed and non-depressed females; A preliminary study. Am. Con-ecf. Tfier. J. 24 (1970), 14-16.
8. Morgan, W. P., Roberts, J. A., Brand, F. R., etal. Psychological effects of chronic physical activity. Med. Sd. Sports 2 (1970), 213217.
9. Brown, R. The prescription of exercise for depression. Physician Sportsmai. 6 (1978), 34-49.
10. Cavanaugh, T., Shepard, R, ]., and Tuck, J. A. Depression after myocardial infarction. Can. Med. Assoc. ]. 113 (1975), 23-27.
11. Creisi, J. H" Klein, M. H., Eischens, R. R., and Paris, J. W. Antidepressant running. Behaa. Med. June, 1978, 19-24.
12. Kostrubala, T. ioy of Running, New York: J. B. Lippincott Company, 1976., p. 129.
13. Greist, J. H., Klein, M. H., Eischens, R. R., Paris, J. W., Gurman, A. S-, and Morgan, W. P. Running as treatment for depression. Compr. Psychiatry. (In press.)
14. Chalmers, T. C. Mortality rate versus funeral rate in clinical medicine. Gastroenterology 46 (1964), 788-791.
15. Eischens, R., Greist, J. H., and Mdnvaille, T. Ran to Reality. Madison, Wis.·. Madison Running Press, 1976, p. 90.