Low, C.A., Stanton, A.L., Bower, J.E., & Gyllenhammer, L. (2010). A randomized controlled trial of emotionally expressive writing for women with metastatic breast cancer. Health Psychology, 29(4), 460–466.

To test the effects of emotionally expressive writing versus disease-related writing on patients with metastatic breast cancer; to determine whether the effects of expressive writing vary as a function of perceived social support or time since diagnosis of metastasis   

Investigators used three sources of recruitment: a larger study, flyers, and advertising on a website and Listserv. All contact with patients was via telephone, postal mail, or email. Participants completed baseline assessments, which gathered data about demographics and emotional support. Investigators collected saliva samples. Patients were randomized to either the emotional or control writing condition, and patients received information about the exercises. Patients were scheduled to participate in four 20-minute sessions that occurred at patients' convenience within a three-week interval. A research assistant monitored compliance. After each session, a patient mailed her writing to the research office. At study entry and three months after the final writing session, a by-mail questionnaire measured outcomes according to stated instruments.

• The sample was composed of 62 participants, 31 were in the expressive writing group and 31 were in the control group.
• Mean patient age was 53.8 years (SD = 10.3 years). The age range was 29–78 years.
• All participants were female.
• All patients had stage IV metastatic breast cancer.
• Participants were eligible for the study if they could complete the writing exercises and assessments in English and if they were undergoing any medical treatment for cancer.
• Of all participants, 74% were college-educated, 71% were married or partnered, 87% were white, and 78% were not working outside the home. On average, participants had been diagnosed with breast cancer 7.9 years ago. Most had received the diagnosis of stage IV cancer 3.3 years ago. The cancer of most participants had metastasized to the bone.
• Of all participants, 72% had had experience with support groups, 63% were working on cancer-related issues with a mental health professional, and 63% had journaled about their cancer experience.
   

• Multisite.
• Home.
• California (October 2006–April 2009); however, information was posted online and on a Listserv.

• Phases of care: multiple phases of care
• Clinical applications: supportive care (end-of-life and palliative care), late effects and survivorship
   

Randomized controlled trial

• Center for Epidemiologic Studies Depression Scale (CESD), a 20-item scale to assess the frequency of depression symptoms in the past week.
• Seven-item intrusions subscale of the Impact of Event Scale (IES), to measure the extent of distress caused by cancer-related intrusive thoughts over the past week.
• Instrument, developed by Pennebaker (1982), to measure negative somatic symptoms. The instrument, shown to be responsive to the expressive writing intervention, asks respondents to report the number of days in the past month on which they experienced each of nine somatic symptoms.
• Pittsburgh Sleep Quality Index (PSQI), to assess sleep quality over the past month and yield a total sleep disturbance score. Scores of 5 and above indicate clinically significant sleep disruptions. This scale has been used as an outcome measure in previous trials of expressive writing as practiced by patients with advanced cancer.
• Items adapted from a tool by Alferi et al. (2001), to measure perceived emotional support.
• Questionnaire, to gather demographic data.
   

• Investigators observed no significant main effects of the experimental condition, with respect to demographic or cancer-related variables or experience with cancer, support groups, psychotherapy, or journaling. 
• Investigators reported no significant group differences, at study entry, regarding symptoms of depression, intrusive thoughts, sleep disturbance, somatic symptoms, or perceived emotional support.
• A significant condition, correlated with social support interaction, emerged regarding intrusive thoughts: Expressive writing was associated with reduced intrusive thoughts in women reporting low emotional support. The significant condition correlated with time, because investigators also observed an interaction between the measure of metastatic diagnosis and the measures of somatic symptoms and sleep disturbance. Relative to the control, participants in the intervention who were more recently diagnosed had fewer somatic symptoms. Intervention participants with longer diagnosis duration exhibited increases in sleep disturbance.
• Among outcome variables, only scores reflecting depression symptoms correlated significantly with intrusive thoughts (r = 0.56, p < 0.001), sleep disturbance (r = 0.30, p = 0.019), and somatic symptoms (r = 0.53, p < 0.001). None of the correlations between the other three outcome variables was significant.

Expressive writing did not produce reduction in psychological distress or improvement in physical health. The intervention may be beneficial for a subset of patients with metastasis and contraindicated for other patients.

• The study had a small sample size, with fewer than 100 participants.
• Participants were predominantly white and well educated.
• Outcome measures related to physical health relied on self-reporting.
• The study was underpowered in regard to detection of main effects. Study design should have included alternative writing topics or a control topic unrelated to cancer.
   

Expressive writing may be of benefit to a certain subset of patients. The intervention is cost-effective and an activity that patients with low levels of social support can do. Future study should apply the expressive-writing approach to vulnerable, underserved, and understudied populations and offer broader topics for expressive writing (e.g., benefits of the cancer experience, a topic unrelated to cancer). Investigators should provide writing supplies. Future research should consider privacy protections (possibly from family members), especially in cultures that place great value on the privacy of written material.