There is a body of work that provides empirical support for a relationship between talking or writing about traumatic events and health outcomes. Several studies, both correlational and experimental, have shown that the inhibition of traumatic experiences is detrimental to health and that talking or writing about a trauma has positive effects on psychological and physical health.
Questionnaire studies have shown that individuals who have had a childhood trauma and not talked about it have more physician visits than those who have disclosed their trauma (Pennebaker, 1992). Similarly, in a sample of 200 corporation employees, the reporting of objective health problems, including cancer and hypertension was significantly predicted by the experience of an early trauma; whether or not they were confided added to the predictive power. These effects held when SES, gender, type of trauma and social support indicators (size of network) were controlled for (Pennebaker & Susman, 1988).
Experimental studies of disclosure and health have involved subjects randomly assigned to write or talk about traumatic or superficial topics, usually for brief periods over several consecutive days, followed by the measurement of physiological reactivity and long or short term health outcomes. In one of the earliest of the writing experiments, Pennebaker and Beall (1986) found that students writing about traumatic experiences made significantly fewer health centre visits in the five months after the study, than those who wrote about non-traumatic topics. Similar results were also found by Pennebaker, Colder and Sharp (1990) and Pennebaker, Kiecolt-Glaser & Glaser (1988). The latter also demonstrated enhanced immune functioning as well as drops in physician visits for the trauma writing group. Francis and Pennebaker (1992) also found positive health effects when using the work absence records and blood assays of university employees as outcome measures. Esterling, Antoni, Kumar, and Schneiderman (1990) found that higher emotional disclosure in students who wrote about traumatic experiences was associated with better immune function.
This last study introduces an emphasis on one important aspect of the disclosure findings: emotion as an essential aspect of describing trauma. Pennebaker and Beall (1986) reported that people who wrote only about the facts of their trauma showed no more health benefits than the non-trauma writers. People who wrote about the emotions and thoughts surrounding traumas showed the most health benefits. Further studies examining the emotionality of speakers disclosing trauma, including Holocaust survivors (Pennebaker, Barger & Tiebout, 1989; Pennebaker, Hughes & O'Heeron, 1987) found immediate physical effects and long-term health benefits for those who conveyed the greatest emotion. Pennebaker (1992) believes that emotionally charged expression activates traumatic memory so that the traumatic experience may be cognitively processed. This cognitive processing must incorporate both emotional and objective features of the event.
Pennebaker, Mayne and Francis (1997) have described the central role of language in their model developed to explain and predict such findings. Using language to label an emotion and an experience creates a structure which facilitates the assimilation and understanding of the event and thus the reduction of emotional arousal which is detrimental to physical and psychological health. Pennebaker and his colleagues have recognised the importance of language in the `processing' of trauma and subsequent health outcomes. To conclude their study of disclosure and immune responses, Petrie, Booth, Pennebaker, Davison and Thomas (1995) wrote that it is important to investigate the nature of emotional expression itself and to distinguish the types of emotional expression that influence the immune system and investigations have been conducted to explore the nature of written expression in particular. Hughes, Uhlmann & Pennebaker (1994) measured the skin conductance levels (SCL) associated with writing on a word by word basis. One of the most interesting of their mixed findings was that people showed increases in SCL as they wrote each negative emotion word, whereas in previous studies most people who expressed negative emotions over their whole essay, showed decreases in SCL. Two studies (Pennebaker & Francis, 1996;
Pennebaker, Mayne & Francis, 1997) assessed the use of language, in essays disclosing traumatic experiences, by the degree to which the subjects used either negative or positive emotion words and the degree to which their writing used words suggestive of cognitive change. These ratings were compared to physical and mental health outcomes. Once again the results were mixed and sometimes contradicted previous findings. For example, neither the increased use of negative nor positive emotion words was related to health benefits in one study while, contrary to previous results, the increased use of positive emotion words was found to be related to increased physical health in another (Pennebaker et al., 1997). Increased use of insight and causation words was related to better physical health but not to self-reported distress nor to psychological health outcomes in a different sample.
One explanation for the Pennebaker model is provided by theories of traumatic stress in which the cognitive processing of a disturbing event is explained in terms of organising or working through the memories of the event. These theories employ the concept of cognitive schemas: structures of organised knowledge based on generalisations from previous experience, which guide the processing of new information (Fiske & Linville, 1980). Cognitive schema theories often emphasise the importance of information and appraisal for assisting with the assimilation of inconsistent experiences by expanding the schematic base. Horowitz (1993) has proposed a stage theory which suggests that traumatic events overwhelm the information processing system of the individual. Information that does not fit with the cognitive schemas that people use to interpret everyday events, but is stored as memories, must be reviewed over time to bring existing schemas and traumatic memories into alignment. Hence the task for people with traumatic memories, i.e. those that do not fit existing schemas, is a gradual processing in which schemas are adjusted so that the new information is included. Those who cannot complete this process suffer from recurring denial and intrusive memories and the accompanying psychological and physical toll that this process takes on the individual. Horowitz and Reidbord (1992) have additionally developed a theory of person schemas, in which emotions are regarded as an integral part of meaning systems and hence interrelated with cognition. According to Horowitz, the process of working through includes talking about the trauma and its related emotions so that the experience can be assimilated.
Janoff-Bulman (1992) has described a model of adaptation to trauma which is also based on the notion of cognitive schemas. The theory holds that people have core assumptions about the self and world which guide their day to day functioning. These schemas are inherently conservative - individuals will first try to fit new information to existing schemas - and extraordinary events involve the development of revised schemas that will account for the new information, preserve self-worth and the ability to cope with emotions. This development process, as in Horowitz' model requires time. According to Janoff-Bulman, talking about and sharing trauma-related thoughts and feelings, provides opportunities for trauma victims to work through their experience.
Since Pennebaker's seminal work there have been many studies using this experimental paradigm to test the effects of the disclosure of emotions on many aspects of physical and psychological health. The results of these studies have been mixed and a number of studies have failed to find consistent effects on psychological distress. However, a meta-analysis by Smyth (1998) which included both published and unpublished manuscripts, reported that, in general, writing about emotional topics is associated with significant reductions in distress. Thus, there is empirical evidence for a link between talking or writing about traumatic events and health benefits. Explanations for this link have been couched in terms of theories of cognitive schemas and the need for processing and assimilating the radically new information that does not readily fit existing schemas. Emotions are an integral part of the cognitions that form schemas and the need to find meaning in life's events is seen as a force behind the incorporation of new information into existing schemas that guide people's behaviours and expectations of future events.
A review of the evidence by Esterling et al. (1999) has described the support that the available evidence provides for the suggestion that the disclosure of traumatic experiences can improve physical and mental health. Esterling et al. focus on the clinical relevance of this evidence to suggest that writing could be used as a powerful therapeutic technique, however, their review also points to some of the limitations of the available evidence and the need for ongoing research in this area. The two papers in this special edition provide the results of ongoing investigation area and attempts to address the limitations and answer questions about the use of the writing paradigm to prevent the deleterious health effects of trauma.
The studies reported here are both examples of home based interventions that use structured writing tasks. The first study (Smyth et al., in this issue) directly addresses one of the limitations described by Esterling et al. (1999); the use of groups with uncontrolled stressors. Esterling et al. suggested that future research could usefully focus on groups who have experienced a single stressful event and the Smyth et al. study answers this need with a well controlled study based on intervention and control groups who have experienced the same natural disaster (with a community reference group for additional control). Both the intervention and control group improved over time on psychological and physical symptoms, and both groups experienced intrusive thoughts about their experience, which points to the importance of the control group. However, the intrusive thoughts of the intervention group, members of which wrote about their experience in a structured way, were not related to the negative health outcomes following the intervention. This is an important finding that contributes to the evidence for the utility of such an intervention for groups affected by natural disaster.
The authors in this edition of the Australasian Journal of Disaster and Trauma Studies, point to the need for ongoing research in this area in regard to the safety of the application of such interventions. One particular concern raised by Smyth et al., was the unknown effect of such interventions on those at high risk, with particular needs, or with higher psychopathology. This concern had also been raised as another important limitation by Esterling et al. (1999) and these concerns are supported by the second paper. In the Sheffield et al. (in this issue) study, the participants were students who wrote about a variety of undisclosed traumas. One of the important findings here was that, as in previous studies, the group that wrote about their emotions in regard to upsetting events were more likely to have more physical symptoms and illness, three weeks after writing. As the authors suggest, the unsupported exposure to distressing thoughts in the home environment could be detrimental and raises additional concerns for those at greater risk than these relatively healthy young adults. As both studies show, most people are resilient and do recover from their traumatic experiences. However, the relatively small number of people who will develop major disorders such as PTSD are those that we must be especially concerned about when planning group interventions for disaster victims.
Such findings suggest that guidance, during disclosure, may be an important aspect of its benefits and there are already indications of the directions that such guidance could take. For example, Pennebaker, Mayne and Francis (1997) suggest that there are two independent processes that are important in the writing or talking about a traumatic event: the need to construct a story or explanation of an event, and the importance of the labelling of the emotions. Lutgendorf and Antoni (1999) reported that, whereas depth of involvement in the disclosure process predicted greater recovery of mood, the use of more words during disclosure predicted higher levels of intrusive thoughts and negative mood. This suggests that the time of writing itself should be limited. Drawing conclusions from recent empirical and theoretical work, Gidron et al.(2000) have proposed and begun testing a guided disclosure protocol.
A final concern that has not been touched upon yet in this area of application is the importance of the timing of disclosure and the social stages of disclosure. Pennebaker and Harber (1993) studied people's reactions following an earthquake and the Persian Gulf War. They found three distinct social stages in regard to the degree of willingness or usefulness of disclosure and the time of the event. Such findings may have important implications for the timing of interventions based on writing and expressing emotions.
The use of interventions based on the expression of emotions about a traumatic experience in a structured manner is based on a background of substantial empirical and theoretical support. The use of writing protocol interventions in disaster or group trauma situations is promising, particularly because such interventions are practicable for use with groups of people and may allow maximum effectiveness with fewer resources. The evidence so far is promising, and further work must be done to replicate these results with attention to issues of control, timing and monitoring of individual utilisation of and response to the writing task.
Esterling, B. A., Antoni, M. H., Kumar, M., & Schneiderman, N. (1990). Emotional repression, stress disclosure responses, and Epstein-Barr viral capsid antigen titers. Psychosomatic Medicine, 52, 397- 410.
Esterling, B.A., L'Abate, L., Murray, E.J., & Pennebaker, J.W. (1999). Empirical foundations for writing in prevention and psychotherapy: Mental and physical health outcomes. Clinical Pscyhology Review, 19, 1, 79-96.
Fiske, S. T. & Linville, P. W. (1980). What does the schema concept buy us? Personality and Social Psychology Bulletin, 6, 543-557.
Francis, M. E., & Pennebaker, J. W. (1992). Putting stress into words: the impact of writing on physiological, absentee, and self-reported emotional well-being measures. American Journal of Health Promotion, 6 (4), 280-287.
Gidron, Y., Duncan, E., Lazar, A., Biderman, M. A., Tandeter, H., Schwartzman, P. (2000). Effects of Guided Written-Disclosure of Traumas on Somatization and Clinic Visits among Frequent Israeli clinic visitors. Proceedings of BPS, vol 8, No 1, pp 15.
Harber, K. D., & Pennebaker, J. W. (1992). Overcoming traumatic memories. In S. Christianson (Ed.), The handbook of emotion and memory: Research and theory. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Horowitz, M. J. (1993). Stress-response syndromes: a review of posttraumatic stress and adjustment disorders. In J. P. Wilson & B. Raphael (Eds.), International Handbook of Traumatic Stress Syndromes. New York: Plenum Press.
Horowitz, M. J., & Reidbord, S. P. (1992). Memory, emotion and response to trauma. In S. Christianson (Ed.), The handbook of emotion and memory: Research and theory. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Hughes, C. F., Uhlmann, C., & Pennebaker, J. (1994). The body's response to processing emotional trauma: Linking verbal text with autonomic activity. Journal of Personality, 62, 565-585.
Janoff-Bulman, R. (1989). Assumptive worlds and the stress of traumatic events: Applications of the schema construct. Social Cognition, 7, 113-136.
Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychology of trauma. New York: The Free Press
Lutgendorf, S. K., & Antoni, M. H. (1999). Emotional and cognitive processing in a trauma disclosure paradigm. Cognitive Therapy and Research, 23, 423-440.
Pennebaker, J. W. (1988). Confiding traumatic experiences and health. In S. Fisher & J. Reason (Eds.), Handbook of life stress, cognition and health. New York: John Wiley & Sons.
Pennebaker, J. W. (1992). Inhibition as the linchpin of health. In H. S. Friedman (Ed.), Hostility coping and health. Washington: American Psychological Association.
Pennebaker, J. W. (1993). Putting stress into words: health, linguistic, and therapeutic implications. Behaviour research and therapy, 31, 539-548.
Pennebaker, James W. (2000). The effects of traumatic disclosure on physical and mental health: The values of writing and talking about upsetting events. In, J.M. Violanti & D.Paton (Eds), Posttraumatic stress intervention: Challenges, issues, and perspectives. (pp. 97-114). Springfield, IL, US: Charles C Thomas.
Pennebaker, J. W., Barger, S. D., & Tiebout, J. (1989). Disclosure of traumas and health among holocaust survivors. Psychosomatic Medicine, 51, 577-589.
Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: towards and understanding of inhibition and disease. Journal of Abnormal Psychology, 95, 274-281.
Pennebaker, J. W., Colder, M., & Sharp, L. K. (1990). Accelerating the coping process. Journal of Personality and Social Psychology, 58, 528-537.
Pennebaker, J. W. & Francis, M. E. (1996). Cognitive, emotional, and language processes in disclosure. Cognition and Emtion, 10, 601-626.
Pennebaker, J.W., & Harber, K.D. (1993). A social stage model of collective coping: The Loma Prieta earthquake and the Persian Gulf War. Journal of Social Issues, 49, 125-145.
Pennebaker, J. W., Hughes, C. F., & O' Heeron, R. C. (1987). The psychophysiology of confession: linking inhibitory and psychosomatic processes. Journal of Personality and Social Psychology, 52, 781-79
Pennebaker, J. W., Kiecolt-Glaser, J. K., & Glaser, R. (1988). Disclosure of traumas and immune function: health implications for psychotherapy. Journal of Consulting and Clinical Psychology, 56, 239-245.
Pennebaker, J. W., & Mayne, T., & Francis, M. E. (1997). Linguistic predictors of adaptive bereavement. Journal of Personality and Social Psychology, 72, 863-871.
Pennebaker, J. W., & Susman, J. R. (1988). Disclosure of traumas and psychosomatic processes. Social Science and Medicine, 26, 327-332.
Petrie, K. J., Booth, R. J., Pennebaker, J. W., Davison, K. P., & Thomas, M. G. (1995). Disclosure of trauma and immune response to a hepatitis B vaccination program. Journal of Consulting and Clinical Psychology, 63, 787-792.
Sheffield, D., Duncan, E., Thomson, K., & Johal, S. S. (2002). Written emotional expression, and psychological and physical well-being: results from a home-based study. Australasian Journal of Disaster and Trauma Studies, 2002-1. Retrieved April 3, 2002 from the WWW: http://www.massey.ac.nz/~trauma/issues/2002-1/sheffield.htm
Smyth, J. (1998). Written emotional expression: effect size, outcome types, and moderating variables. Journal of Consulting and Clinical Psychology, 66, 174-184.
Smyth, J. M., Hockemeyer, K. Anderson, C. Strandberg, K., Koch, M., O'Neill, H. K., McCammon, S. (2002). Structured writing about a natural disaster buffers the effect of intrusive thoughts on negative affect and physical symptoms. Australasian Journal of Disaster and Trauma Studies, 2002-1. Retrieved April 3, 2002 from the WWW: http://www.massey.ac.nz/~trauma/issues/2002-1/smyth.htm
Massey University, New Zealand
Last changed April