CENTER ON BEHAVIORAL MEDICINE

MIND-BODY CONNECTION

PSYCHOIMMUNOLOGY

Related Papers

Psychoimmunology Menu

The Role Positive Emotions Play in Health
   
Permission graciously given by the author to reproduce this paper  

The Positive Role Emotions Play in Health

Robert W. Stewart

Abstract

The role of positive emotions is explored across a variety of stressful health-related events (e.g., AIDS, cancer, arthritis). Studies have shown that positive emotions (i.e., positive words, positive thoughts, humor) can increase healthy responses to that stress. The question is posed regarding whether timing of the emotion is important; specifically, can the development of positive emotion during a stressful health-related event influence the body's response in to a similar magnitude found when positive emotion exists at the time of an illness?

Introduction

Many of our daily activities seem to be viewed as being hazardous to life by public health officials. Although these activities are part of the risks presented by life and choices made by us, none are more likely to create a public health risk than sex. Transmission of disease, including the most deadly virus yet known to man, male violence directed at other men and at women, family discord provoked by jealousy, and sexual transgressions are likely at the top of a long list of physical health problems created by people seeking to satisfy their sex drive.

If positive emotions do not distract most of us from negative feelings, than many public health problems that result from our pursuit of happiness are compounded. So, why have positive emotions generally been ignored? 

Perhaps, many researchers are inclined to disregard the study of positive emotions all together and, perhaps, categorize them as part of humanity at best—even viewing positive emotions as a cause of much pain and vulnerability. By generally ignoring the role of positive emotions, researchers direct us to refocus on recognizing health risks, including the risks involved in pursing happiness.  To them, I suspect, the pursuit of happiness is seen as the opposite of reason and as the creator of much folly. 

Soon after the New Millennium celebrations ended in the year 2000, the New York Times reported an increase in the death rate (Hershey, 2000). That reporter cited the health department in New York City as documenting a 50% increase in deaths during the first week of the year 2000, compared with the first week of 1999. 

Maybe people celebrated too much or waited until the New Year to die. If that is true, how did they do this? Could positive emotions play a role in maintaining health?

Brown (1959) described a force that keeps us alive, and positive emotions probably come as close as to that force as something can to propel us to living. Herzberg (1966) discussed satisfaction in life as the center of human motivation. These positive strivings are not diminished with attention to those needs, as are our basic human needs; although the drive toward self-fulfillment strengthens as successes in our strivings to achieve what we value are experienced. These successes certainly are essential to quality of life and positive emotions are likely part of that foundation. Csikszentmihalyi in 1975 addressed the role positive emotions play in our quality of life, by investigating how people react when deprived of daily pleasurable experiences.

A group of 20 undergraduate students at the University of Chicago were recruited and instructed to act in a normal way by going about their daily routines. Specific instruction not to do anything non-instrumental to their day was given: All pleasure was to be avoided. Many reported strong adverse reactions such headaches, fatigue and tension. Feeling less creative, more anger, and increased disappointment were also reported.
As a result of that study, it was suggested that a steady diet of pleasurable events are needed to maintain us. Although pleasurable events are a set of needs rarely acknowledged by science, Csikszentmihalyi demonstrated that pleasure helps sustain us in even our most mundane daily activities—even if pleasure does not to solve the problems in our lives. Unfortunately, he did little to answer the questions regarding whether positive emotions are good for us. 

One area of research in which to look for the benefits of positive emotions is in cardiovascular health. There, mounting evidence is found demonstrating the role that positive emotions play helping us maintain good cardiovascular health. 

Affeck, Tennen, and Croog (1987), for example, found these benefits. They studied 287 men who suffered initial heart attacks. The men were interviewed seven weeks after their heart attacks and asked to identify something positive in their experience. Nearly half of the participants find some benefit stemming from their heart problems. Many reported that the heart attack increased their awareness of what they individually valued while others reported that they learned the importance of a healthy lifestyle. Nearly 25% reported that they modified their lifestyle in some way to improve quality of life.

These patients were tracked for eight years after their first heart attacks. The number of these men experiencing other heart attacks was recorded, noting how many men described physical impairments limiting their quality of live. Those men finding some benefit from their heart attacks reported less disability and experienced fewer heart attacks over that eight year period than their less positive brethren. 

What is not clear is whether those men who reported benefits actually modified their lifestyles--nor is it known whether there physiological changes could explain the health benefits associated with perceiving positive benefits. The findings are, however, suggestive of a direct correlation between positive benefit finding and cardiovascular health.

Helgeson and Fritz (1999) found direct evidence linking positive beliefs and recovery from coronary heart disease. In their study of involving a group of 199 men and 99 women who had just undergone angioplasty, cognitive adaptation to their cardiovascular problems was the only variable found predictive of continued recovery from the surgical procedure.

Helgeson and Fritz recruited their participants upon discharge from the hospital. Only about 10% refused to participate; thus, the study was likely to be representative of the general population who undergo angioplasty. 

Patients were given questionnaires to complete and then were followed over the next six months to see who experienced recurring of heart problems. Recurring heart problems was defined to include death due to a heart attack, additional angioplasty procedures, bypass surgery, and/or any other cardiac problems revealed by laboratory study. 

The results indicated that 80% of these patients did not experience any recurring or additional problems.  The question the researchers raised was whether it could be predicted whose hearts would stay healthy, based on responses to the questionnaires. Three psychosocial measures were involved: (1) measures of negative emotional states such as depression, hostility, and anxiety, measures of healthy behaviors such as regular exercise, low-fat diets, and time for relaxation, and a cluster of positive mental health measures the investigators referred to as cognitive adaptation, which included self-esteem, optimism, and sense of mastery.  After controlling for any effects by income, education, and other sociodemographic indicators, those three variables were tested for their predictive quality. 

Only the measures of cognitive adaptation reliably predicted who would sustain recovery following that surgery. All other variables were weakly predictive at best. When the investigators examined and cross-tested the measures within the adaptation index, Helgeson and Fritz found self esteem and optimism, qualities associated with positive emotion, to be responsible for the effects. When high, medium, and low scores on cognitive adaptation were classified, those patients classified as low in these attributes were found to be three times more likely to have a coronary problem than those scoring high on the indicators of positive mental health. Unfortunately, Helgeson and Fritz did not question what might be responsible for these effects of self-esteem and positive perception.

It could be expected that those who were more upbeat led healthier lifestyles. In the Helgeson and Fritz study, however, no link between the health behavior measures and coronary problems were found; hence, it is unlikely that health behavior served to limit further coronary problems. Those investigators suggested that self-esteem and optimism may serve to lower blood pressure and heart rate directly through neuroendocrine mechanisms, thus controlling the physiological reaction to stress. Other researchers have also documented the link between positive emotions and health.

By studying the neurophysiologic effects of exercise, our knowledge of how positive emotions and health have been increased. Many health care practitioners promote moderate exercise as good for our health; indeed, we are aware of the effects of exercise on endorphin levels. McCubbin's 1993 study investigated the physiological processes considered to be responsible for positive feeling states such as the "exercise high" associated with increase in endorphins. 

McCubbin focused on the role beta-endorphins play in creating that "high." Since researchers cannot directly observe the ebb and flow of these peptides within brain tissue, at the present, indirect means to measure the effects of those neurochemicals must be used. Typically, indirect measurement includes the administration a drug that is known to affect the chemical being studied, such as an agonist or antagonist. 

 In the McCubbin study, beta-endorphins were measured indirectly through the administration of an antagonist. In his study, aerobic training was provided to a group of randomly chosen volunteers. Fifty percent of those who received training and half of the controls without training were given the opioid antagonist, naltrexone. 

Naltrexone was administered just prior to evaluating the effects of the cardiovascular training on blood pressure and heart rate reactions to stress through laboratory testing. The author employed a series of complex mental arithmetic problems to be solved within significant time restrictions, a procedure known to the investigators to elevate blood pressure and heart rate, as the stress agent.

Across group comparisons found significant differences between the groups: Solving complex math problems raised the blood pressure less in those with the exercise training. In the group whose beta-endorphin effects were blocked, the beneficial effects of exercise disappeared, suggesting that the body's opioid system is essentially involved in limiting the blood pressure response during times of stress. 

McCubbin's research team in 1996 also found that relaxation training to have beneficial effects in limiting the physiological reaction to stress; hence, physical conditioning can be ruled out as being the critical factor. Instead, it seems positive emotions, regardless how they are induced whether appear to be the keys that open the physiological controls regarding stress. The question is then raised whether positive emotions affect health problems other than problems related to the cardiovascular system and stress.

Cohen et al. (2002) investigated the relationship between emotions and illness. In this study, the link between the common cold and emotion was explored; specifically, whether people with many positive and/or negative emotions have increased susceptibility to the common cold was examined. Adult men and women were recruited though advertisements in the newspaper. The extent of their everyday emotions was rated. To obtain this information, data was collected telephonically, three days a week for two consecutive weeks. 

The raters inquired whether each participant felt negative emotions (e.g., depression, unhappiness, nervousness, tension, hostility, anger). They also asked participants about positive emotions. A separate index was constructed using such positive words as lively, happy, cheerful, calm, and relaxed. 

At the end of the two weeks, each gave participant was administered nose drops with one of two types of rhinovirus and then quarantined for several days. Objective measurement of whether a participant contracted a cold was obtained by measuring the antibody response to the virus and the amount of nasal mucous secreted during the five days following exposure. In order to maintain a subjective estimate of symptoms, each participant reported any cold symptoms to the investigators.

The objective criteria showed more than one-third of the participants became symptomatic following exposure to the virus. Interestingly, those participants who reported high levels of positive emotions prior to exposure were found less likely than those reporting low positive emotion to become ill. Such within group variations were not found across those participants reporting negative emotions. These findings held for symptoms both objectively and subjectively reported: Tense and angry people were found no more vulnerable to the flu than participants who were not tense or angry; thus, those who were least vulnerable to the flu reported high levels of positive emotion. The vulnerable people were those who were not cheerful and energetic.

Recovery from Traumatic Health Events: A Role for Positive Emotions?

Thus far, the role of positive emotions in the ebb and flow of daily life has been discussed. It may be instructive to extend this discussion to major, even traumatic, events, such as an untimely death of a close family member, to see what has been uncovered regarding health effects. Traumatic events are likely among the most studied and the least understood experiences.

There is little doubt that these experiences can have profound effects on our psychological well-being. Janoff-Bulman (1992) has studied traumatic events extensively. Although traumas differ from one another in many ways, she has identified one unifying experience among those coping with these unexpected calamities: a shattering of fundamental assumptions. 

She notes that certain beliefs sustain us in our everyday lives. Prominent among those beliefs is that the world is a benevolent place, a place in which things happen for a reason. Traumas challenge those assumptions and at times even raise doubts about our self-worth.

The assumptions we have about the world seem to allow us to maintain an orderly view of our relationship to events and reduce uncertainty about the future. The arbitrariness of traumatic experiences, however, can lift the curtain on our lives to reveal chaos where there had been order and uncertainty where there had been faith and trust. The stress is often severe, and the need for recovery of a coherent view of one's place in the world is great. 

 The healing process itself may take many forms, but there is one form that is the most popular: talking about one's emotional reactions to the event with someone who cares. This form is commonly referred to as self-disclosure, and it has an extensive following among those who teach, perform, and engage in psychotherapy.

There are many theories concerning why self-disclosure may be helpful in recovery from major traumas. One popular model is that some people inhibit their expressions of negative feelings following severe crises.  

This inhibition is thought to bottle up those emotions, and such constraints on emotional expression are considered harmful not only to psychological adjustment but also to physiological adaptation. People may inhibit their self-expression because they are afraid of the powerful emotions they harbor, but, according to some theorists, failure to get it out in the open may leave an open wound, delaying the cognitive processing needed to come to terms with the experience and to reach closure. 

A failure to talk about what has happened may lead to rumination over the events, perhaps even with recurring mental images of the trauma. When people talk about traumatic experiences, they are reopening channels of communication between their emotions and processes of cognitive adaptation, allowing for better understanding of the event and a recovery of beliefs of fairness and other fundamental assumptions that had been challenged by the traumatic experience.

Pennebaker (1990) designed studies to put these theories to empirical test. He asked college students to write about a stressful experience that was very meaningful to them, and he instructed other students to write about something else unrelated to events in their past. Those students who wrote about their traumas had fewer subsequent visits to the student health center and were happier, when assessed a month or two later, than those given the task of writing about something inconsequential. This result has been replicated numerous times with different groups and different measures of outcome, ranging from better work attendance to greater immune cell activity in the presence of a pathogen. 

Beneficial health effects also have arisen from talking, as well as writing, about the events. The key appears to be the expression of thoughts and feelings about the stressful experience. How are we to understand the mechanisms of adaptation involved in recovery from these traumas? 

After some 10 years of research and considerable success in reproducing these beneficial effects on health due to self-disclosure, Pennebaker, Mayne, and Francis (1997) examined some of the potential mechanisms that they believed were responsible for these effects. They analyzed the words people use to express themselves. In this case, the language people use is thought to be a window into their thought processes. 

Pennebaker and her group reasoned that the use of language containing strong negative feelings in the self-disclosing essays would be an important indicator that the person was allowing the expression of hidden emotions. Such expressions of negative emotional states would then lead to more accurate cognitive processing of the trauma and its impact and to better integration of the experience and a more coherent structure of meaning. In contrast, the expression of positive emotions would distract from the primary task of self-disclosure and recovery: Therefore, these researchers predicted that those people with many negative and few positive affect-laden words in their disclosures would benefit the most from telling their stories.

The mechanisms of healing, however, may be quite different than those identified by the Pennebaker group. From a two-dimensional view of emotional health, one of the potentially damaging effects of traumatic events is a loss of complexity in our emotional lives. Traumatic events not only raise negative affects but also reduce positive affects, and they may do so dramatically. These events could disturb the normal wide berth between negative feelings and the capacity to experience positive emotions. 

Such is the subtext of grief following the death of a loved one and of being paralyzed by fear following a severe threat to life or livelihood. Positive emotions then disappear, and the capacity to experience and express these feelings maybe recovered only slowly for some after a period of time. 

Usually when we think of the impact of trauma on health, we think of the impact of negative affective states over an extended period of time. If we dissect the impact of these traumatic events, we see that two processes are occurring: the increase in negative emotions and the loss in the ability to experience positive feelings due to the collapse of affective dimensions following stressful events. 

After the terrorist attack on September 11th, 2001, many of us went through the succeeding days grim-faced, humorless, stone-cold in affect. It is this loss of positive capacities that could be the most disruptive to adaptation and health in the long-term: not the inhibition of expression of negative feelings but the inhibition of positive feeling states. Returning to the research on expression of emotions about past traumatic events to see how this model predicts different results from Pennebaker's research from 1990 and 1997. 

Essentially, I argue that a return to health would be marked by a return to the relative independence of positive and negative affects. Those people who are able to reestablish a connection to positive emotion when reviewing the traumatic experiences should be most likely to show health benefit's. In short, the more a person expresses positive emotions while retelling the story, the more he or she should find health benefits. The benefit should extend, however, only to those who reexamine the events fully enough to make the transformation real. 

Those who fail to reveal any negative emotions in the stories of their stressful experiences may still be in an emotional straitjacket, unable to allow co-expression of positive and negative feelings. Without this emotional richness, self-disclosure may not be beneficial.

These predictions, derived from the dynamic model of stress and emotions, were substantiated. In 1997, when Pennebaker and her colleagues counted the number of times the study participants expressed various emotions in their narratives, it was the number of positive affect expressions, not the number of negative affective words, that predicted better health.

Some expression of negative feelings was also useful. In fact, the people with the best health outcomes fit the profile identified: Their disclosures of traumas included many positive emotion-laden words and a few negative affect words.

Certainly, this set of findings will not be the last word on how people benefit from talking about difficult life experiences. This area of research is rich in complexity. Nevertheless, we are again alerted to how positive emotions emerge; when we might least expect them to, as a key benefit to adaptation and health.

A striking illustration of the life-sustaining properties of positive emotion comes from the "nun study." Danner, Snowdon, & Friesen (2001), completed an intensive examination of the health and longevity of 180 Sisters of Notre Dame conducted in the 1990s. These Catholic nuns were born prior to 1917 and ranged in age from 75 to 95. Over the course of the nine year follow-up, 42% of these women died, and the researchers sought variables that would predict survival.

When the nuns took their vows for the order, they had been asked to write autobiographies. The researchers found these hand-written documents. Danner and his colleagues wondered whether the emotions these nuns expressed in their autobiographies would predict who was still alive at the end of the study. Each word was coded for the presence of negative emotion, positive emotion, and the absence of emotion; each sentence was then scored in a similar way. Multiple raters were used to ensure reliability in these judgments, and the researchers were indeed able to confirm high levels of agreement among raters on the levels of emotion expressed in the autobiographies.

The number words expressing positive and negative emotion as well as the number of positive and negative emotion sentences were summed. Each nun was then ranked according to her emotional expressiveness. The researchers then tested whether nuns who ranked high in emotional expression lived longer. 

The number of negative emotions that were expressed did not predict who would stay alive; however, the presence of positive emotions in the writings was strongly associated was longevity. Chance of survival across the nine year study was 2.5 times greater for the 25% who expressed the greatest number of positive emotions. The researchers estimated that nuns in the top quartile in their use of positive emotions in their storytelling lived an average of 10 years longer. The question is now raised regarding how positive emotions operate to preserve our health and make us more resilient in the face of difficult, even traumatic, experiences? 

Let us return to the physiological model examined earlier for positive emotions generated from exercise: the opioids. Among their many actions, opioid systems are known to play a key role in regulating neuro-endocrine responses to stress, and in turn, to stress-related down-regulations of the immune system. Could gains in positive emotion lead to longevity through promoting a healthier immune response?

Even speculations about such a relationship probably would have been scoffed over the past 10 years ago by cancer researchers. Now there are some data, not in themselves conclusive but nonetheless suggestive, that this may well be the case. Researchers followed the lives of 40 gay men who had tested positive for HIV infection (Bower, Kemeny, Taylor, & Fahey, 1998). These men were part of Folkman et al. (1996) study regarding post-bereavement and pre-bereavement predictors in HIV+ and HIV- men. Each of these men had experienced a death of a close friend or partner due to AIDS. 

Bower's group scored the transcripts of their interviews with these men to judge whether or not the men had found some meaning or purpose in the deaths of their friends and loved ones. Sixteen or 40% of the group did discuss finding meaning. Some reported important lessons, such as spending more time with people who they cared about and living life in the most satisfying way possible.

One marker of disease progression for those who are HIV+ is the rate of decline in CD4 cells, which are key cells in immune function lost over time to the AIDS virus (Garcia, 1997). Those partners who found meaning in the deaths of their loved ones had a slower rate of decline in circulating CD4 cells. In addition, more of those partners who found meaning were alive three years later than those who had not found some benefit. 

Many researchers are skeptical of findings that come from observational studies such as these. From a methodological point of view, only those investigations that manipulate the independent variable are "true" experiments. In Folkman, finding meaning was the key measure, and it defined individual-differences in how the HIV patients coped with bereavement. No matter how carefully any study is done, there always seems to be some question that can be raised about whether other, unmeasured differences between patients that were hidden from the investigators could account for the findings. Only by experimentally inducing "meaning" in some way can researchers eliminate these uncertainties.

Cruess et al. (2000), for example, encouraged a group of women with breast cancer to “find meaning" with interesting results being found. They enlisted 34 women who had just had surgery and randomly assigned them to one of two conditions: a 10 week cognitive-behavioral stress management program or a waiting-list control group that would receive the same treatment but only after the experiment was completed. 

Those investigators monitored two key outcomes from women in treatment and control conditions: degree of psychological distress and levels of cortisol in circulation in the bloodstream. It was expected that treatment effects on psychological and physiological health would be found. Their reasoning was that cortisol is a good marker for stress responses that suppress immune function; thus, high cortisol levels would be particularly dangerous to breast cancer patients, who must rely on a responsive immune system to arrest further spread of malignant cells.

When Cruess and his colleagues examined changes pre-to-post intervention, no difference between groups in level of psychological distress was found. The intervention apparently did not influence the patient's experience of negative emotions; although they did find differences in cortisol: The treatment group had lower levels of this immune-suppressing hormone in their bloodstreams than did the control group.

The cognitive treatment had emphasized patients finding new and more positive interpretations of their illness as ways of transforming the stresses that this illness had brought about. When the investigators compared groups, the treatment group was, in fact, much more likely to report finding benefits from their experiences with breast cancer. Some remarked, for example, that their illness brought the family closer together. 

Others reported a new appreciation of themselves and their lives with comment such as, "my illness made me more accepting of things." The treatment lowered cortisol levels and increased benefit finding. The next and final step was to see if benefit finding lowered cortisol levels.

Those participants in the treatment group who showed lower levels of this stress hormone after the treatment were those who were able to find benefits from their experiences with cancer. The engagement of positive emotions through benefit finding appeared to be the key ingredient in improvements for these women, not a reduction in distress. These findings are dramatic illustrations of the potential links between immune system functioning and positive emotions. As it turns out, this finding is becoming increasingly commonplace. No one expected it, least of all those researchers investigating the role of stress hormones on immune processes. 

The focus on negative affective states has left many feeling blindsided by these findings; thus, more than a little reluctant to delve into the realm of positive emotions. It has been a place where angels fear to tread.
Perhaps scientists in such novel mind-body fields as psychoneuroimmunology have been particularly reticent, fearing the huckster label a designation that has been applied so aptly in the past to many who have promoted various snake oil preparations that promise to cure all that ails.

All the pieces of the puzzle are certainly not in place: A biological mechanism by which positive emotions can turn on the immune system and thereby foster healthier and longer lives has not yet been identified. There are some candidates among the many chemical messengers, such as neuropeptide Y (Zukowska-Grojec, 1995); however, the absence of attention to this area leaves us with few careful experiments that are grounded in sound theoretical work. 

Nevertheless, empirical demonstrations of the effects of positive states on immune function are dotting the landscape with increasing frequency, as predicted by pioneers in this field (Melnuchuk, 1988; Panksepp, 1993).

Considerable attention has focused on those immune processes involved in natural killer (NK) cell activity. These cells apparently have the ability to proliferate in response to the presence of cancer cells is frequently used as a measure of immune system health because such activity directly parallels a key immune function: protection against cancer cell growth (Janeway, 2005). Valdimarsdottir and Bovbjerg (1997), for example, conducted a straightforward study of the relationship between positive emotion and NK cell activity, with intriguing results. Forty-eight healthy middle-aged women were recruited and asked questions about their positive and negative feelings at the time, and drew blood on two consecutive days. 

It was found that women who reported more positive mood had higher levels of NK cell activity than women with less positive mood, but not always. Only those women who reported negative mood got an immune system boost from positive mood. Those women who reported no negative emotions during those two days were unaffected by their positive moods. The presence of psychological distress appeared to have created a physiological context that favored the influence of positive states on immunity; thus, suggesting that stress-related processes may trigger the healing properties of positive states.  Another example comes from the work of Segerstrom, Taylor, Kemeny, and Fahey (1998). 

Segerstrom and his colleagues studied changes in NK cell activity in 50 students in their first semesters of law school. At the beginning of the semester the researchers drew blood and asked the students to report on their degree of optimism. Segerstrom's group used two measures of optimism in this study. One measure focused on the global beliefs held by the students and expectations held by the students about the course of their lives by asking them questions such as, "In uncertain times, I usually expect the best." Another measure focused more on their immediate beliefs in themselves with questions about current situations, for example, "It is unlikely I will fail (at law school)."

Those researchers referred to this measure as situational optimism, as opposed to the other, global measures that they termed "dispositional optimism." The interesting aspect of this study was that the researchers took advantage of the students' midterm exams as a naturally occurring stressor. The research question involved whether these two forms of optimism would influence change in immune activity identified from assays of the initial blood drawn at the start of the semester compared with the immune assays on blood drawn during midterms.

A global optimistic outlook had no effect on changes in NK cell activity during the stress of exams, thus, so-called Pollyanna models of health received no support. There was, however, a strong and reliable relationship between situational optimism and immune activity among these law students: Those students who were optimistic about their success in coping with the stresses of law school also had immune systems better prepared to engulf and destroy cancer cells than those who were not as optimistic. Two factors were thus important: (1) They were under stress, and (2) the positive state had relevance to the stressful situation.

For many illnesses such as cancer, we assume that greater immune system activity is beneficial; autoimmune conditions such as rheumatoid arthritis result from overabundant immune responses that cause abnormal joint inflammation and pain (Weil, 1998). It is reasonable to expect that, for people with autoimmune disorders, the effects of positive emotions may be quite different from those observed in cancer patients. Health from the standpoint of the immune system may be defined best by using theories of adaptation such as McEwen's (1998) model of allostatic load. 

The key issue is with this model is not whether there is a sufficient immune response to stress but whether there is adequate homeostatic regulation of both the initial response and recovery. An imbalance in either direction can thus be harmful to health. Yoshino, Fujimori, and Kohda (1996) as well as Nakajima, Hirai, and Yoshino (1999) conducted studies with rheumatoid arthritis patients to examine the effects of positive emotions on autoimmune processes.

They brought groups of patients into a lecture hall after lunch and showed them an hour-long video of traditional Japanese comic stories. In the former study, the patients' arthritis pain levels before and after the video was noted. 

Blood was drawn, before and after the video, to test for changes in levels of interleukin 6 (IL-6)--a pro-inflammatory cytokine associated with autoimmune disease processes in rheumatoid arthritis. In the latter study, the researchers compared the IL-6 levels of rheumatoid patients who saw the video with a control group of rheumatoid patients who were randomly selected not to see the video.

Both studies showed significant effects of humor on disease processes for patients in pain. Not only was the pain substantially lower, but also IL6 levels were reduced significantly for those rheumatoid patients who watched the video. 

Increasing positive emotions in both studies reduced the production of immune products responsible for inflammation, pain, and damage to the joints. The question is then raised whether studies on cancer and rheumatoid arthritis provide any clues to what mechanisms we use to convert positive emotion into better health.

One aspect that has been especially overlooked is that the timing of the emotion. Positive emotion, when present at the time of stress, appears maximally effective in strengthening resistance. This raises the question whether the development of positive emotion during the stressful event can have the same effect.
 
References

Affeck, G., Tennen, H., & Croog, S. (1987). Casual attribution, perceived benefits, and morbidity after a heart attack: An eight year study. Journal of Consulting and Clinical Psychology, 55, 29-35.

Bower, J. R., Kemeny, M. E., Taylor, S. E., & Fahey, J. L. (1998). Cognitive processing, discovery of meaning, CD4 decline, and AIDS-related mortality among bereaved HIV seropositive men. Journal of Consulting and Clinical Psychology, 60, 979-986.

Brown, N. O. (1959). Life against death: The psychoanalytical meaning of history.
Middletown, CT: Wesleyan University Press.

Cohen, S., Doyle, W. J., Turner, R. B., Alper, C. M., & Skoner, D. P. (2002). Emotional style and susceptibility to the common cold. Unpublished manuscript.

Cruess, D. G., Antoni, M. H., McGregor, B. A., Kilbourn, K. M., Boyers, A. E., Alferi, S. M., Carver, C. S., & Kumar, M. (2000). Cognitive-behavioral stress management reduces benefit finding among women being treated for early stage breast cancer. Psychosomatic Medicine, 62, 304-308.

Csikszentmihalyi, M. (1975). Beyond boredom and anxiety. San Francisco: Jossey Bass.
Danner, D. D., Snowdon, D. A., & Friesen, W. V. (2001). Positive emotions in early life and longevity: Finding from the nun study. Journal of Personality and Social Psychology, 80, 804-813.

Folkman, S., Chesney, M., Collette, L., & Boccellari, C. M. (1996). Postbereavement depressive mood and its prebereavement predictors in HIV+ and HIV- gay men. Journal of Personality and Social Psychology, 70, 336-348.

Garcia, T. (1997). Viral load in asymptomatic patients with CD4 lymphocyte counts above 500 x 106/l. AIDS (11), 53-57.

Helgeson, V. S., & Fritz, H. L. (1999). Cognitive adaptation as a predictor of new
coronary events after percutaneous transluminal coronary angioplasty. Psychosomatic
Medicine, 61, 488-495.

Hershey, R. D. (2000, January 15). Rise in death rate after New Year is tied to
the will to see 2000. New York Times, p. A1.

Herzberg, F. (1966). Work and the nature of man. New York: Crowell.
Janoff-Bulman, R. (1992). Shattered assumptions: Toward a new psychology of
trauma. New York: Free Press.

Janeway, T. (2005). Immuno-biology: The immune system in health and disease. London:
 Churchill-Livingston.

McCubbin, J. A. (1993). Stress and endogenous opioids: Behavioral and circulatory
interactions. Biological Psychiatry, 35, 91-122.

McCubbin, J. A., Wilson, J. F., Bruehl, S., Ibarra, P., Carlson, C. R., Norton, J. A.,
& Colcough, G. W. (1996). Relaxation training and opioid inhibition of blood
pressure response to stress. Journal of Consulting and Clinical Psychology, 64,
593-601.

McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New
England Journal of Medicine, 338, 171-179.

Melnuchuk, T. (1988). Emotions, brain, immunity, and health: A review. In M.
Clynes (Ed.), Emotions and psychopathology (pp. 181-247). New York: Plenum Press.

Nakajima, A., Hirai, H., & Yoshino, S. (1999). Reassessment of mirthful laughter
in rheumatoid arthritis [Letter to the editor]. Journal of Rheumatology, 26,
512-513.

Panskepp, J. (1993). Neurochemical control of moods and emotions: Amino acids to neuropeptides. In M. Lewis (Ed.), Handbook of emotions (pp. 87-107). New
York: Guilford Press.

Pennebaker, J. W. (1990). Opening up: The healing power of confiding in others. New York: Morrow.

Pennebaker, J. W., Mayne, T, & Francis, M. E. (1997). Linguistic predictors of adaptive bereavement. Journal of Personality and Social Psychology, 72, 863-871.

Segerstrom, S. C., Taylor, S. E., Kemeny, M. E., & Fahey, J. L. (1998). Optimism
is associated with mood, coping, and immune change in response to stress.
Journal of Personality and Social Psychology, 74, 1646-1655.

Valdimarsdottir, H. B., & Bovbjerg, D. H. (1997). Positive and negative mood:
Association with natural killer cell activity. Psychology and Health, 12, 319-
327.

Yoshino, S., Fujimori, J., & Kohda, M. (1996). Effects of mirthful laughter on neuroendocrine and immune systems in patients with rheumatoid arthritis. [Letter to the editor]. Journal of Rheumatology, 23, 793-794.

Weil, A. (1998). Health and healing: The philosophy of integrative medicine. New York: Houghlin Mifflin

Zukowska-Grojec, Z. (1995). Neuropeptide Y: A novel sympathetic stress hormone and more. Annals of the New York Academy of Sciences, 771, 219-233.