Journal of Pain and Symptom Management
Volume 35, Issue 6 , Pages 623-631, June 2008

Emotional Disclosure Through Patient Narrative May Improve Pain and Well-Being: Results of a Randomized Controlled Trial in Patients with Cancer Pain

  • M. Soledad Cepeda, MD, PhD

      Affiliations

    • Department of Anesthesia, Tufts-New England Medical Center, Boston, Massachusetts
    • Corresponding Author InformationAddress correspondence to: M. Soledad Cepeda, MD, PhD, Department of Anesthesia, Tufts-New England Medical Center, 750 Washington Street, Box 298, Boston, MA 02111, USA.
    • ,
  • C. Richard Chapman, PhD

      Affiliations

    • Pain Research Center, University of Utah, Salt Lake City, Utah, USA
    • ,
  • Nelcy Miranda, RN

      Affiliations

    • Department of Anesthesia, San Ignacio Hospital, Bogota
    • ,
  • Ricardo Sanchez, MD, MSc

      Affiliations

    • National University School of Medicine, Bogota
    • ,
  • Carlos H. Rodriguez, MD

      Affiliations

    • Pain and Palliative Care Service, Colombian National Institute of Cancer, Bogota, Colombia
    • ,
  • Andres E. Restrepo, MD

      Affiliations

    • Department of Anesthesia, San Ignacio Hospital, Bogota
    • ,
  • Lina M. Ferrer, MD

      Affiliations

    • Department of Anesthesia, San Ignacio Hospital, Bogota
    • ,
  • Rene A. Linares, MD

      Affiliations

    • Pain and Palliative Care Service, Colombian National Institute of Cancer, Bogota, Colombia
    • ,
  • Daniel B. Carr, MD

      Affiliations

    • Department of Anesthesia, Tufts-New England Medical Center, Boston, Massachusetts

Accepted 4 August 2007. published online 25 March 2008.

Article Outline

Abstract 

Narrative medicine is based upon physicians’ awareness of patients’ narration of their suffering, their hopes, and how illness has affected them. It offers a model for improving health outcomes. To determine whether incorporating a narrative approach in patients with cancer decreases pain intensity and improves their global sense of well-being, we performed a randomized, single-blind controlled trial in adult patients with cancer and average pain intensity levels of at least 5/10. Two hundred thirty-four patients were randomized into three groups: (1) narrative (n=79), in which patients wrote a story about how cancer affected their lives for at least 20 minutes once a week for three weeks; (2) questionnaire (n=77), in which patients filled out the McGill Pain Questionnaire; and (3) control (n=78), in which patients came weekly to medical visits during which they received usual customary care. Patients rated their pain on a 0–10 scale and their well-being on a seven-point Likert scale weekly for eight weeks. Two raters independently evaluated the emotional content of the narratives. Pain intensity and sense of well-being were similar in all groups before and after treatment. Subgroup analyses showed that patients whose narratives had high emotional disclosure had significantly less pain and reported higher well-being scores than patients whose narratives were less emotional. Further study is needed to demonstrate whether the implementation of narrative medicine is associated with health benefits in this and other contexts.

Key Words: Narrative medicine, emotional disclosure, RCT, pain treatment, symptom control

 

Back to Article Outline

Introduction 

Narrative medicine offers a model for improving communication between patients and physicians,1 quality of care, and health outcomes. It is based on a clinician's hearing the patient's story of what that patient is going through; his or her fears, suffering, and hopes; and a description of how illness has affected him or her.1, 2

Narrative medicine aims to empower patients, giving them a sense of control during a time when patients often have lost control.3, 4 It balances the current practice of medicine, making it more humane, and less technologically focused. Narrative medicine facilitates the creation of therapeutic alliances with patients and the selection of treatments that make sense within the story lived by patients and their families.5 Evidence suggests that the patient–doctor interaction plays an important role in the use of health resources,6 and patient participation in decisions reduces health care costs.

Although to our knowledge there are no randomized controlled trials that evaluate the benefits of narrative medicine, there are reports on the effects upon health outcomes of writing about traumatic events.7, 8, 9, 10 Meta-analyses that examined the effect of writing on health found that doing so improved physical symptoms, but the magnitude of effect was small.11, 12 Although in many of the studies patients wrote about stressful experiences, in other studies patients wrote about their hopes, suffering, and how their disease affected their lives. The latter type of written narrative is termed “emotional writing” or “emotional disclosure.”

Composing emotional narratives may permit patients to understand their needs and emotions, and such insight could lead to better health decisions and health outcomes. Thus, the health provider's awareness of the patient's story may add minimal benefit beyond the patient's “writing the story.”

Narrative medicine could be a powerful therapeutic tool for patients with cancer, who often endure a great deal of suffering.13, 14, 15, 16 Emotional disclosure in patients with cancer has been evaluated in randomized controlled trials with mixed results. Some trials have found it ineffective,17, 18 whereas others have shown it to decrease physical symptoms.19, 20, 21 However, the burden of disease in such studies has not been severe with the exception of one study that evaluated patients with metastatic cancer. This latter study had a small sample size (42 patients) and the benefits were seen only in few of the outcomes evaluated, such as sleep quality.21

We therefore sought to determine whether incorporating a structured narrative approach (writing about how the illness has affected the dimensions of life) in patients with advanced cancer decreases pain intensity and improves patients’ global assessment of their well-being.

Back to Article Outline

Methods 

Study Design 

We performed a randomized, single-blind (evaluator), controlled trial. The trial was approved by the Institutional Review Boards of the Javeriana University School of Medicine and the Colombian National Institute of Cancer. Individuals who agreed to participate provided written informed consent. The study was registered in the U.S. clinicaltrials.gov registry.

Study Population 

We recruited patients from the oncology, radiotherapy, and pain management services of the Javeriana University Oncology Center and the Colombian National Institute of Cancer. To facilitate the generalizability of the study results, the sample included adult patients with any type of cancer who reported average pain intensity levels of at least 5/10 on a 0–10 scale, were able to read and write, and scored 50% or higher on the Karnofsky scale (lower scores on this 0–100 scale represent lower functional capacity that might indicate patients’ inability to write or visit the clinic during control sessions).22

Randomization. To ensure that all groups had similar numbers of patients with roughly the same disease burden, we performed a stratified randomization with a cutoff of 80% on the Karnofsky scale as a measure of disease severity. A score of 80% or higher represents being able to work. For the randomization, we used a computer-generated random number program (STATA, version 9SE). We concealed the randomization schedule using serially numbered, sealed, opaque envelopes.

Patients were randomized into three groups: (1) Narrative: patients were asked to write, while at home, for at least 20 minutes once a week for three weeks, a story about how cancer affected their lives; (2) Questionnaire: patients were asked to complete, while at home, the McGill Pain Questionnaire; or (3) Controls: these patients were asked simply to attend weekly medical follow-up visits (i.e., the same clinic schedule as the other two groups) to receive usual customary care.

The questionnaire group was included as an “attention control.” Some studies have incorporated a “neutral writing” group that writes about routine day-to-day activities instead of more meaningful thoughts and feelings.21, 23 Because writing per se could have a therapeutic effect, we decided that completing the McGill Pain Questionnaire (which includes a bodily pain map to identify painful areas, and offers clusters of words to indicate the sensory and affective dimensions of pain24) was a better experimental option.

Patients in the narrative group received assurance that their narratives would be confidential and not included in their medical records. We asked them to write about how cancer has affected their lives, calling upon their deepest thoughts, feelings and fears if present, and to do so about once a week for at least 20 minutes. We asked them not to disclose any form of identification (signature, name, etc.) within the narrative, nor to concern themselves with polishing their grammar, spelling, or style.

The “exposure dose”—at least 20 minutes once a week for three weeks—followed from the results of meta-analyses on the effect of writing about stressful events. In the original studies, the length of each writing session ranged from 15 to 30 minutes and the number of sessions ranged from one to five.11

Three weeks after initial randomization, patients were telephoned by research personnel two or three days in advance of their scheduled visits to remind the patients about writing their stories, filling out the questionnaires, and coming to the office.

Follow-Up. Patients participated for eight weeks following randomization. All patients were seen weekly for the first three weeks in the pain clinic or palliative care service, then called weekly for four more weeks, and then asked to come to the last session of this intervention at the eighth week. All office visits included a regular checkup. To make sure that all patients received similar care, physicians were given a list of symptoms to be assessed: pain, nausea, anorexia, constipation, and sleep disruption.

Maintaining the Assessor's Blinding. At each visit, patients handed their written narratives to one of the research nurses who then brought the patient to his or her physician follow-up visit. A second research nurse who was unaware of patients’ group assignments then collected information for the study. Weekly follow-up phone calls were made by research personnel unaware of patient allocation.

Primary Outcome. Average pain intensity was our primary outcome. Patients rated their average pain intensity during the prior week using a verbal numerical rating scale in which 0 is “no pain” and 10 is “the worst pain imaginable.” We evaluated pain intensity at baseline (when patients agreed to participate) and then weekly for eight weeks.

Secondary Outcome. As a secondary outcome, simultaneously with pain intensity, we evaluated each patient's sense of general well-being on a seven-point Likert scale that ranged from “awful” to “excellent.”

The major advantage of using a single question with a Likert scale to measure a health-related quality-of-life outcome is that it is minimally burdensome. The simplicity and practicality do not affect the validity of the responses25, 26 and, in fact, appeal to patients.25, 26 We used a six-point scale because it is more sensitive to detect change than five-point Likert scales.27

Emotionality of the Narrative. Emotional disclosure and increasing insight has been linked to improved health status.28, 29, 30, 31 Therefore, we assessed the emotional disclosure of the patients’ narratives using a six-point Likert scale that ranged from “none” to “very much.” Similar assessments of the magnitude of emotional disclosure have been performed in previous studies.20, 32 Nonetheless, because of the subjectivity in rendering this assessment, two raters graded the degree of emotional disclosure in the patients’ narratives independently.

Sample Size. For patients with acute33 or chronic34 pain of moderate or severe intensity, a pain intensity difference of two units on a 0–10 scale is clinically meaningful. To detect such a difference between the groups with 80% power and an alpha error of 0.05, assuming that the baseline pain intensity in our patients was on average 6±4 (mean±standard deviation) on a 0–10 scale, we estimated a need for 63 patients per group. We increased this figure by 20% to 77 patients per group, in anticipation of dropouts.

Data Analysis 

We used an intent-to-treat analysis. To analyze the effect of the treatment on pain intensity, we used an analysis of repeated measures using generalized estimating equations (GEE). The GEE method adjusts standard errors because the multiple observations on the same subject are not independent.

To analyze the effect of the treatment on patients’ sensation of well-being, we used a similar approach as described for pain intensity, despite assessing well-being with a Likert scale. We took this approach to avoid having to select a specific time at which to calculate the proportions of subjects with a specific degree of well-being in each of the three groups. The GEE approach permitted us to assess potential changes across time.

To evaluate the effect of emotional disclosure on pain intensity and well-being, we included in the GEE model the emotional disclosure of the narrative as a dichotomous variable (“very much” vs. anything less than “very much”). In addition, we tried other categorizations to contrast the degree of emotional disclosure (i.e., “very much” and “much” vs. anything less than “much”). We calculated agreement between the two evaluators using the Kappa statistic for ordinal variables.

All values presented are mean±standard deviation.

Back to Article Outline

Results 

Two hundred thirty-four people were randomized. Figure 1 depicts patient flow within the study. Demographics and clinical characteristics of the groups were similar at randomization (Table 1).

Table 1. Baseline Characteristics of Patients Included in the Study by Treatment Group
VariableNarrative Group, n=79Questionnaire Group, n=77Control Group, n=78
Women555144
Age (years±SD)50.2±10.549.1±14.246.2±12.6
Weight (kg±SD)61.3±1361.3±1462.9±13.2

Educational level
Elementary education232722
High school education272729
College education292327
Karnofsky≥80%605756

Pain etiology
Somatic pain424850
Visceral pain171714
Neuropathic pain1378
Mixed etiology756

Pain duration
Less than six months626654
Six months to a year9513
More than a year8611

Baseline pain intensity7.5±1.87.2±1.87.6±1.7

Level of well-being (self-report)
Excellent100
Very good536
Good191919
Fair373233
Bad798
Very587
Awful565

Type of cancer
Breast cancer Stage I–III15610
Breast cancer Stage IV9105
Cancer of unknown origin254
Carcinoma of the uterine cervix Stage I–II978
Carcinoma of the uterine cervix Stage III–IV513
Chondrosarcoma Stage II–III302
Colon cancer Stage II–III134
Colon cancer Stage IV201
Gastric cancer with distant metastasis365
Gastric cancer with no distant metastasis020
Hodgkin's lymphoma Stage II–III012
Kidney cancer Stage III–IV201
Lung cancer Stage II731
Lung cancer Stage IV103
Melanoma Stage II–III011
Melanoma Stage IV031
Multiple myeloma Stage II–III031
NonHodgkin's lymphoma Stage II–III103
NonHodgkin's lymphoma Stage IV012
Prostate cancer with no distant metastasis234
Prostate cancer with distant metastasis220
Rectal cancer Stage II–III354
Rectal cancer Stage IV012
Retroperitoneal liposarcoma230
Other types478

Pain Intensity 

Mean pain intensity was similar for all groups (Fig. 2). Throughout the study, the narrative group had 0.1 units higher pain intensity than the questionnaire group (95% confidence interval[CI] −0.7, 0.5) and 0.02 units higher pain intensity than the control group (95% CI −0.6, 0.5).

  • View full-size image.
  • Fig. 2 

    Pain intensity was similar for all groups throughout the study. Narrative group had 0.1 units higher pain intensity than the questionnaire group (95% CI −0.7, 0.5) and 0.02 units higher pain intensity than the control group (95% CI −0.6, 0.5).

Well-Being 

The sense of well-being was similar in all groups. Throughout the study, the narrative group reported 0.08 units greater sense of well-being (on a 0–6 scale) than the questionnaire group (95% CI −0.3, 0.2) and 0.02 units lesser sense of well-being than the control group (95% CI −0.2, 0.3).

Agreement Between Raters 

The agreement between the two raters was good (ordinal kappa 0.8, P=0.0001).

Adherence 

Fifty-three percent of patients (42/79) in the narrative group wrote the three essays as instructed, 86% (66/77) of patients in the questionnaire group completed questionnaires as instructed, and 90% (70/78) of patients in the control group came to their medical visits as scheduled (P=0.0001).

Emotional Disclosure of Narrative and Effect on Outcomes 

Twenty patients had no emotional disclosure in one or more of the three written essays; in five patients, the emotional disclosure in at least one of the narratives was rated “very much.” These latter patients had significantly lower pain intensity than patients whose emotional disclosure in their narratives was never rated as great as “very much” (Fig. 3). The average difference in pain intensity between these two latter groups was 2.5 units (95% CI −4.2, −0.9). At subsequent follow up, patients whose narrative emotional disclosure rated highest also had higher well-being scores than patients whose narratives had lower emotional disclosure (1.37 difference, 95% CI 1, 1.7).

  • View full-size image.
  • Fig. 3 

    Patients with very high degree of emotional disclosure had lower levels of pain (average difference 2.5 units [95% CI −4.2, −0.9]) than patients with lower levels of emotional disclosure.

These improved outcomes in patients with a high degree of emotional disclosure cannot be explained by lower levels of pain intensity or higher senses of well-being at the start of the study. Baseline pain intensity scores in the patients whose narratives had the highest emotional disclosure were 8.7±2.5, vs. 7.5±1.8 in patients with lesser emotional disclosure. In terms of well-being, baseline scores were 3.8±11 in the higher emotionally disclosing group and 3.75±1.5 in individuals with less emotionally disclosing narratives.

None of the other categorizations used to dichotomize the degree of emotional disclosure was associated with lower pain intensity levels or higher senses of well-being.

Back to Article Outline

Discussion 

Overall pain intensity and sense of well-being were similar in all three groups, suggesting that writing a narrative does not alter those outcomes. Other clinical trials have failed to show that writing has therapeutic benefit35, 36 and an update of a systematic review on the effect of emotional disclosure concluded that writing did not produce any clear improvement in physical health or other heath outcomes.37 This systematic review included the studies that evaluated patients with cancer.

We found three randomized controlled trials concerning pain of noncancer origin, not included in the systematic review described above, which also assessed pain intensity as an outcome. Two evaluated patients with fibromyalgia,23, 38 and the other examined those with chronic pelvic pain.32 In all three trials, patients wrote about stressful events, yet their results are contradictory. Of the studies that evaluated fibromyalgia, one found a small benefit23 and the other found no benefit at one month, but a small benefit at three months.38 The study that evaluated chronic pelvic pain found a small benefit upon pain intensity, but no effect in other dimensions of pain such as sensory, affective pain or interference.32

The lack of effect of emotional disclosure on health outcomes could be explained by inclusion of individuals who may not feel any need to disclose, are reluctant to disclose fully, or have difficulty with the cognitive or affective tasks needed to process and resolve emotional experience and to articulate these experiences and conclusions.32 The absence of a psychotherapist could also be a factor promoting a negative response, as subjects may revisit painful emotions without forming new insights.37 In our study, patients were encouraged to write about how cancer affected them, but there was no interaction with, or acknowledgment of, the content of the narrative on the part of health care providers. Further, with respect to limits upon generalization of our results, only 6% of the present sample had narratives with the highest degree of emotional disclosure.

We found that patients who were able to disclose a very high degree of emotion in their narratives reported lower pain intensity and had higher well-being scores. Hence, the disclosure of emotion contributes to the effect of narrative medicine upon health outcomes in patients with cancer pain. Although evaluating the impact of high emotionally disclosing narratives on pain and well-being was approached by means of ad hoc subgroup analyses, other ad hoc dichotomizations of emotional disclosure, which also included substantial emotional disclosure, showed no association between emotional disclosure and pain levels or well-being scores. In addition, we cannot rule out the possibility that our findings are subject to the additional confounding effect of our not having randomized patients to disclose narrative material of a neutral or emotional nature but instead relied upon patients to self-allocate within these categories. Nonetheless, confirmation of the present results would be worthwhile because, if our results are generalizable, the benefit of writing emotionally disclosing narrative may be clinically important. The 2.5-unit decrease in pain intensity observed in our study is typically considered by patients to be clinically meaningful.33

Despite efforts to assure adherence, only 50% of patients had the desired “exposure”—writing three narratives—and 25% of patients provided no emotional disclosure in one or more of the narratives. Patients may have not invested the time necessary to develop emotional themes in their stories; we did not measure how much time patients dedicated to writing these. This lack of “adherence” could explain the absence of a clear effect of composing versus not composing narratives upon pain and well-being. These findings may also be interpreted as showing that patients with advanced cancer and significant disease burden find it difficult to compose emotional narratives. Although previous studies that included patients with cancer17, 20 have found that such patients were able to construct emotionally disclosing stories, these trials included patients with early stage disease.

Our study's assessment of the degree of emotional disclosure in the narratives was subjective. Formal linguistic analysis could have provided data on the number of positive and emotional words to confirm or refute the subjective appraisals.23 However, the interrater agreement was good and other studies have used similar rating scales to evaluate the magnitude of emotional disclosure.17, 20

Cultural differences, including beliefs and values, can shape how patients experience pain, how they form and express their narratives, their willingness to disclose narrative information, and the impact of the narrative on quality of life. Therefore, such issues should be taken into consideration. For example, ethnic differences in expressiveness of symptoms have been described: the Latino population in aggregate exhibits more expressiveness than Europeans, and Chinese Americans as a group exhibit less expressiveness than European Americans.39, 40, 41 These prior studies, together with the present results, suggest the potential practical value of studying both expressiveness and ethnicity in the context of palliative care and narrative medicine.

Narrative medicine is a complex construct and we evaluated only one of its dimensions, that is, writing a story. Verbally recounting a story, whether to listeners who are present or simply as a solitary dictation, is another option for individuals with advanced cancer. Therefore, additional study is needed to evaluate whether the implementation of narrative medicine in general, and which of its specific modalities singly or in aggregate, are associated with measurable health benefits.

Back to Article Outline

Acknowledgments 

The authors thank the attending physicians in the Pain Clinic at San Ignacio Hospital and Palliative Care Unit at Colombian National Cancer Institute for their contributions to the success of this study.

Back to Article Outline

References 

  1. Charon R. The patient-physician relationship. Narrative medicine: a model for empathy, reflection, profession, and trust. JAMA. 2001;286:1897–1902
  2. Charon R. Physician writers. Lancet. 2004;363:404
  3. Charon R. Narrative and medicine. N Engl J Med. 2004;350:862–864
  4. Carr DB, Loeser JD, Morris DB. Narrative, pain, and suffering. Seattle, WA: IASP Press; 2005;
  5. Elwyn G, Gwyn R. Narrative based medicine: stories we hear and stories we tell: analysing talk in clinical practice. BMJ. 1999;318:186–188
  6. Redelmeier DA, Molin JP, Tibshirani RJ. A randomised trial of compassionate care for the homeless in an emergency department. Lancet. 1995;345:1131–1134
  7. Richards JM, Beal WE, Seagal JD, Pennebaker JW. Effects of disclosure of traumatic events on illness behavior among psychiatric prison inmates. J Abnorm Psychol. 2000;109:156–160
  8. Pennebaker JW. The effects of traumatic disclosure on physical and mental health: the values of writing and talking about upsetting events. Int J Emerg Ment Health. 1999;1:9–18
  9. Francis ME, Pennebaker JW. Putting stress into words: the impact of writing on physiological, absentee, and self-reported emotional well-being measures. Am J Health Promot. 1992;6:280–287
  10. Pennebaker JW, Barger SD, Tiebout J. Disclosure of traumas and health among Holocaust survivors. Psychosom Med. 1989;51:577–589
  11. Smyth JM. Written emotional expression: effect sizes, outcome types, and moderating variables. J Consult Clin Psychol. 1998;66:174–184
  12. Frisina PG, Borod JC, Lepore SJ. A meta-analysis of the effects of written emotional disclosure on the health outcomes of clinical populations. J Nerv Ment Dis. 2004;192:629–634
  13. Cleeland CS. The impact of pain on the patient with cancer. Cancer. 1984;54:2635–2641
  14. Carr DB, Goudas LC, Lawrence D, et al. Management of cancer symptoms: Pain, depression, and fatigue. AHRQ Publication No. 02-E032. Rockville, MD: Agency for Healthcare Research and Quality; 2002;
  15. Cepeda MS, Carr DB. Overview of pain management. Approaches to pain management. An essential guide for clinical leaders. Oakbrook Terrace, IL: Joint Commission Resources; 2003;1–20
  16. Chapman CR, Gavrin J. Suffering: the contributions of persistent pain. Lancet. 1999;353:2233–2237
  17. Zakowski SG, Ramati A, Morton C, et al. Written emotional disclosure buffers the effects of social constraints on distress among cancer patients. Health Psychol. 2004;23:555–563
  18. Walker BL, Nail LM, Croyle RT. Does emotional expression make a difference in reactions to breast cancer?. Oncol Nurs Forum. 1999;26:1025–1032
  19. Rosenberg HJ, Rosenberg SD, Ernstoff MS, et al. Expressive disclosure and health outcomes in a prostate cancer population. Int J Psychiatry Med. 2002;32:37–53
  20. Stanton AL, noff-Burg S, Sworowski LA, et al. Randomized, controlled trial of written emotional expression and benefit finding in breast cancer patients. J Clin Oncol. 2002;20:4160–4168
  21. de MC, Sterner J, Hall M, et al. A pilot study of the effects of expressive writing on psychological and behavioral adjustment in patients enrolled in a Phase II trial of vaccine therapy for metastatic renal cell carcinoma. Health Psychol. 2002;21:615–619
  22. Karnofsky scale. Available at http://www.whonamedit.com/synd.cfm/3286.html
  23. Broderick JE, Junghaenel DU, Schwartz JE. Written emotional expression produces health benefits in fibromyalgia patients. Psychosom Med. 2005;67:326–334
  24. Caraceni A, Cherny N, Fainsinger R, et al. Pain measurement tools and methods in clinical research in palliative care: recommendations of an Expert Working Group of the European Association of Palliative Care. J Pain Symptom Manage. 2002;23:239–255
  25. Davey HM, Barratt AL, Butow PN, Deeks JJ. A one-item question with a Likert or Visual Analog Scale adequately measured current anxiety. J Clin Epidemiol. 2007;60:356–360
  26. Temel JS, Pirl WF, Recklitis CJ, et al. Feasibility and validity of a one-item fatigue screen in a thoracic oncology clinic. J Thorac Oncol. 2006;1:454–459
  27. Wyrwich KW, Tardino VM. A blueprint for symptom scales and responses: measurement and reporting. Gut. 2004;53(Suppl 4):iv45–iv48
  28. Pennebaker JW. Putting stress into words: health, linguistic, and therapeutic implications. Behav Res Ther. 1993;31:539–548
  29. Campbell RS, Pennebaker JW. The secret life of pronouns: flexibility in writing style and physical health. Psychol Sci. 2003;14:60–65
  30. Pennebaker JW, Seagal JD. Forming a story: the health benefits of narrative. J Clin Psychol. 1999;55:1243–1254
  31. Suedfeld P, Pennebaker JW. Health outcomes and cognitive aspects of recalled negative life events. Psychosom Med. 1997;59:172–177
  32. Norman SA, Lumley MA, Dooley JA, Diamond MP. For whom does it work? Moderators of the effects of written emotional disclosure in a randomized trial among women with chronic pelvic pain. Psychosom Med. 2004;66:174–183
  33. Cepeda MS, Africano JM, Polo R, et al. What decline in pain intensity is meaningful to patients with acute pain?. Pain. 2003;105:151–157
  34. Farrar JT, Young JP, LaMoreaux L, et al. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain. 2001;94:149–158
  35. Klapow JC, Schmidt SM, Taylor LA, et al. Symptom management in older primary care patients: feasibility of an experimental, written self-disclosure protocol. Ann Intern Med. 2001;134:905–911
  36. Harris AH, Thoresen CE, Humphreys K, Faul J. Does writing affect asthma? A randomized trial. Psychosom Med. 2005;67:130–136
  37. Meads C, Nouwen A. Does emotional disclosure have any effects? A systematic review of the literature with meta-analyses. Int J Technol Assess Health Care. 2005;21:153–164
  38. Gillis ME, Lumley MA, Mosley-Williams A, et al. The health effects of at-home written emotional disclosure in fibromyalgia: a randomized trial. Ann Behav Med. 2006;32:135–146
  39. Lasch KE. Culture and pain. Pain Clin Updates. 2002;X:1–8
  40. Kripke DF, Jean-Louis G, Elliott JA, et al. Ethnicity, sleep, mood, and illumination in postmenopausal women. BMC Psychiatry. 2004;4:8
  41. Leong FTL, Ming E, Lee S. The relationship between family dynamics and career interests among Chinese Americans and European Americans. J Career Assess. 2004;12:65–84

 Support for this study was provided by the Saltonstall Fund for Pain Research and a small donation from Grunenthal.

PII: S0885-3924(08)00060-2

doi:10.1016/j.jpainsymman.2007.08.011

Journal of Pain and Symptom Management
Volume 35, Issue 6 , Pages 623-631, June 2008

Article Tools

* Image Usage & Resolution