Available via license: CC BY 4.0
Content may be subject to copyright.
1
Type of Paper: Original Paper
Title: Online positive affect journaling improves mental distress and well-being in general
medical patients with elevated anxiety symptoms: Evidence from a preliminary randomized
controlled trial
Authors: Joshua M. Smyth1,2*, PhD, Jillian A. Johnson1, PhD, Brandon J. Auer2, PhD, Erik
Lehman3, MS, Giampaolo Talamo2, MD, & Christopher N. Sciamanna2*, MD, MPH
Affiliations:
1Department of Biobehavioral Health, The Pennsylvania State University
2Department of Medicine, The Hershey Medical Center and Pennsylvania State University
3Department of Public Health Sciences, The Hershey Medical Center and Pennsylvania State
University College of Medicine
*Corresponding Authors
Corresponding Author Contact Information:
Joshua M. Smyth, PhD
231 Biobehavioral Health Building
The Pennsylvania State University
University Park, PA, 16802
814-863-8402
(fax) 814863-7525
jms1187@psu.edu
Christopher N. Sciamanna, MD, MPH
Milton S. Hershey Medical Center
West Campus Health Clinic, Campus Drive
500 University Drive
Hershey, PA, 17033
717-531-4221
(fax) 717-531-0151
cns10@psu.edu
2
ABSTRACT
Background: Positive affect journaling (PAJ), an emotion-focused self-regulation intervention,
has been associated with positive outcomes among medical populations. It may be adapted for
online dissemination to address a need for scalable, evidence-based psychosocial interventions
among distressed patients with medical conditions.
Objective: This study examined the impact of a 12-week online PAJ intervention on
psychological distress and quality of life in general medical patients.
Methods: Seventy adults with various medical conditions and elevated anxiety symptoms were
recruited from local clinics and randomly assigned to an online PAJ intervention (n=35) or usual
care (n=35). The intervention group completed 15-minute online PAJ sessions on three days each
week for 12 weeks. At baseline and the end of months 1 through 3, surveys of psychological,
interpersonal, and physical well-being were completed.
Results: Patients evidenced moderate sustained adherence to online intervention. PAJ was
associated with decreased mental distress (p’s≤.045) and increased well-being (p’s ≤.046)
relative to baseline. PAJ was also associated with less depressive symptoms (p=.047) and anxiety
(p=.01) after one month, and greater resilience after the first (p=.044) and second month (p=.01),
relative to usual care.
Conclusions: Online PAJ may serve as an effective intervention for mitigating mental distress,
increasing well-being, and enhancing physical functioning among medical populations. PAJ may
be integrated into routine medical care to improve quality of life.
Trial Registration: ClinicalTrials.gov NCT01873599
3
Keywords: positive affect journaling, online intervention, depression, anxiety, expressive
writing
4
INTRODUCTION
At present, 60% of all people living in the United States have at least one chronic health
condition, and 42% have multiple chronic conditions [1]. As advances in the treatment of disease
continue to prolong life and the overall population continues to age, these numbers are likely to
increase. The significant costs associated with managing medical conditions are well known. The
majority of diseases hold the potential to worsen the overall health of patients by limiting their
functional capacity, productivity, and health-related quality of life, and are a major contributor to
health care expenditures [2-4]. Patients with medical conditions face several challenges and often
need to modify life aspirations, daily routines, and employment. Although some patients
experience periods of grieving and adjustment after a diagnosis, many others experience
sustained distress that can further impact physical and mental health, and quality of life [4].
Given the link between severe or chronic medical conditions and psychological distress, it is not
surprising that comorbidity between medical and mental health conditions is the rule rather than
the exception [5]. The 2001–2003 National Comorbidity Survey Replication, for example, found
that more than 68% of adults with a mental health disorder reported having at least one general
medical illness, and that 29% of people with a medical condition also had a comorbid mental
health problem [6,7]. Stressful life events often precede anxiety and mood disorders [8] and the
accompanying psychological strain associated with the diagnosis of, and living with, a major
medical illness places these patients at risk for comorbidity and worse health outcomes overall.
Stress and dysphoric mood more generally may worsen the prognosis and progression of
disparate diseases [9,10] and is a major contributor to many of the leading causes of death in the
United States such as cancer, coronary heart disease, respiratory disorders, and suicide, among
5
others [11]. In light of evidence that stress and dysphoria might be a modifiable risk factor for
the development and progression of medical illnesses, finding ways to reduce distress in patients
with one or more existing medical conditions is a major public health concern.
Psychological interventions (e.g., Cognitive Behavioral Therapy, CBT) have been shown
to reduce psychological distress in chronic disease populations [12-14]. Although psychological
interventions are increasingly desirable among patients [15], there are several barriers to
accessing face-to-face psychological care among people with chronic health conditions (e.g.,
cost/insurance coverage, access, stigma) [15-18]. The internet has emerged as an effective tool
for disseminating efficacious mental health interventions [19] and may serve to overcome some
of these barriers to accessing mental health services. For example, a meta-analysis of internet-
based CBT interventions observed that they are effective for reducing depression and anxiety
[20], and Farrer and colleagues [21] observed a 44% reduction in depressive symptoms over 6
months among those randomized to internet-based CBT versus only 11% among controls. To
date, however, these evidence-based internet interventions are either not readily accessible or
widely disseminated among the general population, and therefore do not address the problem of
access to psychological services.
Relative to internet-based therapeutic or counseling interventions, Positive Affect
Journaling (PAJ), a simple intervention that is cost-efficient and easily disseminated to patients,
is becoming increasingly popular. PAJ is a modified version of the traditional expressive writing
paradigm [22,23] wherein participants write about a traumatic experience for approximately 15
to 20-minute intervals, often across a period of 3 to 5 days. Reviews of expressive writing
suggested that it was modestly effective in improving a number of physical and mental health
outcomes [24,25], although large heterogeneities in efficacy have been documented.
6
For example, several studies have found clinical benefits tied to expressive writing in
patients with autoimmune and inflammatory conditions such as arthritic conditions, lupus, and
asthma [25-29], fibromyalgia [30,31], irritable bowel syndrome [32], and HIV/AIDS [33,34]. In
addition, expressive writing has been found to have beneficial effects on blood pressure [35] and
on several health-relevant outcomes following the experience of a heart attack, such as reduced
numbers of medical appointments and prescription medications, increased self-care behaviors,
improved cardiac symptoms [36] and improved health-related quality of life [37]. Expressive
writing has also been associated with small, but consistent, improvements to well-being among
diverse cancer groups—especially breast, renal, and prostate cancer patients [38]. Finally, a
relatively small study of 40 people diagnosed with major depressive disorder found that those
writing about their deepest thoughts and feelings related to emotional events had significant
reductions in depression immediately after writing and over one month thereafter [39].
A number of efforts have been made to modify the original expressive writing approach
to be better suited for use across several contexts and populations. One stream of this process is
reflected in the integration of positive psychology, a large and growing area of research that has
linked positive psychological and emotional dispositions and states of being (e.g., optimism,
happiness, subjective well-being, positive affect) to various beneficial outcomes. Some of the
reported benefits of these positive dispositions include: fewer physical symptoms [40], faster
wound healing [41], healthier functioning biological processes (e.g., neuroendocrine,
inflammatory, and cardiovascular activity) [42], better interpersonal relationships [43], higher
quality of life [44], increased longevity [45], and decreased morbidity [46,47]. As such, the
expressive writing paradigm has been adapted to have participants write about positive aspects of
their lives and themselves (e.g., making meaning out of or finding benefit in past experiences;
7
[48,49]; focusing on positive aspects of one’s self [50]) under the notion that this would yield
similar benefits to those observed in the positive psychology literature. As a whole, we refer to
this array of positive-focused writing approaches as PAJ.
Positive affect interventions among both patients and healthy individuals have led to
improvements in a number of health outcomes. In two studies comparing an education control
(i.e., educational workbook and behavioral contract) to a positive affect intervention (i.e., self-
affirmation inducement over bi-monthly telephone sessions with staff and unexpected gifts prior
to calls), positive affect improved medication adherence in hypertensive African American
patients [51], and physical activity in patients following a percutaneous coronary procedure [52].
In addition, Stanton and colleagues [53] found that four sessions of written expressive disclosure
or benefit finding resulted in lower physical symptom reports and medical appointments among
breast cancer patients at 3-month follow-up. In healthy samples, Armitage and colleagues [54]
found beneficial effects of completing a self-affirmation questionnaire or self-affirming
implementation intention on alcohol intake at one-month follow-up, while Burton and King [55]
observed that participants randomized to write only two minutes for two consecutive days in a
lab about a recent positive event showed moderate reductions in physical symptoms (Cohen’s
d=.65) at 4 to 6-week follow-up.
The goal of this randomized controlled trial was to examine whether a 12-week internet-
based PAJ intervention could reduce mental distress (primary outcome) and positively influence
psychological, interpersonal, and physical well-being (secondary outcomes), relative to usual
care, in a heterogeneous sample of patients with elevated anxiety symptoms. It was hypothesized
that participants randomized to the intervention would experience decreases in mental distress
(i.e., Hospital Anxiety and Depression Scale score; HADS) and improvements in psychological
8
well-being (e.g., perceived stress, resilience), interpersonal well-being (i.e., social support), and
physical well-being (e.g., days during which pain inhibited usual activities) over the 12-week
intervention period. It was also hypothesized that participants randomized to receive the
intervention would report less mental distress and greater levels of psychological, interpersonal,
and physical well-being than those in the control condition at each assessment period.
METHODS
Sample and Recruitment
All study procedures were approved by The Pennsylvania State Hershey Medical
Center’s (PSHMC) Institutional Review Board and all participants provided written informed
consent prior to engaging in any research related activity. This study was registered on
ClinicalTrials.gov (reference number NCT01873599), with recruitment and active intervention
occurring from June 2013 - February 2014.
Potential participants were recruited through flyers placed around the PSHMC campus
and advertisements placed in PSHMC media and local community newspapers in central
Pennsylvania. Additionally, oncology patients at The Pennsylvania State University Hershey
Cancer Institute with an Eastern Cooperative Oncology Group (ECOG) Performance Status
score of 0-3 (not completely disabled) were identified through registry review and sent a letter
describing the study. Participants were provided with a toll-free number to call if they were
interested in participating, as well as an opt-out card that could be mailed back by those who
were uninterested. Individuals who did not respond were contacted via phone by a research staff
member within two weeks to determine their interest in participating.
Eligibility for inclusion was based on: (1) English fluency, (2) between 21 and 80 years
of age, (3) internet access, (4) self-report of moderate to significant stress during the past month,
9
(5) not currently pregnant and no plans to become pregnant within the next 3 months, (6) no
plans to move within the next 6 months, (7) no hospitalization for a psychiatric condition in the
past year, (8) not a high risk for suicidality as assessed by selected questions from the Structured
Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders [56]. Although not
an explicit requirement, it was assumed that potential participants be familiar with using a
computer and accessing web-sites.
Individuals interested in participation and who met the initial inclusion criteria were
invited for a laboratory visit, and further assessed for eligibility. Eligible participants: (1)
reported a score of 8 to 15 on the anxiety subscale of the HADS [57]; and (2) had an ECOG
Performance Status of 0 (fully active) through 3 (limited self-care) [58]. Participants that met all
inclusion criteria were invited to participate.
Random Assignment
Randomization (1:1) was done via sealed envelopes prepared by someone other than the
research staff conducting the study visits and opened by participants during the baseline visit
after completing informed consent. See Figure 1 for flow diagram of recruitment procedure.
10
Figure 1. Participant flowchart
Procedure
Phone Calls Received (n=142)
Non-Starters (n=18)
Cancelled (n=13)
No show (n=3)
On hold (n=2)
Intervention Group (n=35)
Lost to follow-up (n=1)Lost to follow-up (n=2)
Consented and Randomized (n=70)
Eligible (n=88)
Not Screened (n=43)
Not interested (n=28)
Called after study closed (n=9)
Unable to contact (n=7)
Assessed for Eligibility (n=99)
Ineligible (n=11)
Reported “almost none” or “relatively
little” stress (n=7)
No access to internet (n=4)
Control Group (n=35)
Completed Study (n=33) Completed Study (n=34)
11
Eligible participants met with research staff during a scheduled baseline visit to discuss
study procedures and provide written informed consent. During the baseline visit, all participants
completed baseline surveys and were randomized (via computer generated sequences provided in
sealed envelopes) to one of the two conditions. Participants assigned to the intervention
condition received an introduction and training session to orient them to the intervention website
where they would complete the writing sessions. All participants completed self-report survey
assessments online at the end of months 1, 2, and 3 using a secure data capture system
(REDCap). Participants received gift cards following the completion of each survey (i.e., $40
compensation for completing all three assessments).
Intervention
Participants in the PAJ intervention condition were asked to complete online writing
sessions for 15 minutes on three days each week for the duration of the 12-week study. The
amount of time spent writing at each session is similar to prior expressive writing studies,
although the duration of the intervention in this study was longer than many other prior studies
[23,25,59] in order to ensure the potential for adequate ‘dose’ of intervention. This was the first
version tested of this intervention; the intervention content was ‘frozen’ during the trial and not
adjusted. During each online writing session, participants logged onto the study website and
wrote a journal entry on one of seven commonly used positive affect prompts (e.g., What are you
thankful for? What did someone else do for you?; full details available upon request) [60]; all
entries were saved on a secure server. During the study, journal entries of participants in the
intervention condition were screened by research staff to monitor content. Participants that did
not complete a journal entry within any given 7-day period were sent an email reminder (this
email reminder also included reminders of how study staff could help them resolve any technical
12
difficulties in case any existed). As there is no clinical standard of care treatment for medical
patients with mild to moderate anxiety symptoms, participants randomized to the wait-list
control group received their usual care for the duration of the study. After they had completed all
study procedures, participants in the control condition were given access to the PAJ intervention.
Measures
Sociodemographics and health behaviors
Participants’ age, gender, race, ethnicity, marital status, and education level were obtained
at baseline. During this time, participants self-reported basic information related to their specific
disease and health behaviors (i.e., smoking, physical activity, alcohol use) using standard self-
report items from the Behavioral Risk Factor Surveillance System [61].
Primary Outcome
The Hospital Anxiety and Depression Scale (HADS) [57] consists of two scales, anxiety
and depression, with each consisting of 7 items rated on a scale from 0 through 3. Items are
aggregated for each subscale (range = 0-21), with higher scores indicating greater anxiety or
depressive symptom severity, and for a total HADS score (range = 0-42), with higher scores
indicating greater mental distress. Various cut-off scores are available for the HADS. A score of 8
or greater on the anxiety subscale (HADS-A) has a specificity of 0.78 and sensitivity of 0.9 for
clinically significant anxiety, while scores below 8 indicate non-cases [62]; the inclusion
criterion of a HADS-A score of 8 to 15 was intended to include participants with mild to
moderate symptoms, while excluding those with non-significant or severe symptoms (HADS-A
scores of 15 to 21), as the PAJ intervention was expected to have limited benefit for those
individuals. In this study, Cronbach’s alpha at baseline for anxiety = 0.65, depression = 0.86, and
HADS total score = 0.85.
13
Secondary Outcomes
The Brief Resilience Scale (BRS) [63] is a 6-item measure of perceived resilience,
including items such as “It does not take me long to recover from a stressful event”. Each item is
rated on a scale from 1 (strongly disagree) to 5 (strongly agree). All items are aggregated for a
total score (range = 6-30); higher scores indicate greater resilience. The BRS has high levels of
internal consistency, with Cronbach’s alpha ranging from .80-.91 [63]. In this study, Cronbach’s
alpha = 0.90 at baseline.
The Healthy Days Measure [61] assesses self-reported physical health and functioning.
Respondents answered four items to indicate: (1) general health (i.e., “Would you say that in
general your health is “poor” – “excellent”), (2) days during which pain inhibited their usual
activities (i.e., “During the past 30 days, for about how many days did pain make it hard for you
to do your usual activities…?"), (3) sleep quality (i.e., “During the past 30 days, for about how
many days have you felt you did not get enough rest or sleep?”), and (4) number of days they felt
healthy and full of energy (i.e., “During the past 30 days, for about how many days have you felt
very healthy and full of energy?”).
The Perceived Stress Scale (PSS) [64] consists of 10 items that assess perceived stress,
rated on a scale from 0 (never) to 4 (very often). A sample item includes: “In the last month, how
often have you felt nervous and stressed?” Items are combined for a total score with higher
scores indicating greater stress. The measure demonstrates strong internal consistency. In this
study, Cronbach’s alpha = 0.91 at baseline.
The Positive and Negative Affect Schedule (PANAS) [65] consists of two subscales, each
including 10 items. Respondents indicate the extent to which they felt specific positive emotions
(e.g., excited, proud) and negative emotions (e.g., upset, afraid) over the past month on a scale
14
from 1 (not at all) to 5 (extremely). Subscales are scored separately (range = 10-50); higher
scores indicate greater positive affect and greater negative affect. Internal consistencies are high
(Cronbach’s alpha = 0.85-0.88) [65]. In this study, Cronbach’s alpha for the positive affect
subscale = 0.90 and negative affect subscale = 0.90 at baseline.
The Satisfaction with Life Scale (SWLS) [66] is a five-item scale that assesses overall
life satisfaction with items such as: “In most ways my life is close to my ideal” and “I am
satisfied with my life.” Each item is rated on a scale from 1 (strongly disagree) to 7 (strongly
agree), and a total score is calculated using all items (range = 5-35); higher scores indicate
greater satisfaction with one’s life. The SWLS has a test-retest reliability of 0.82 [66]. In this
study, Cronbach’s alpha = 0.93 at baseline.
The Social Provisions Scale (SPS) [67] assesses various dimensions of social support.
The scale consists of 24 items that make up six components, each consisting of four items
pertaining to: attachment, social integration, reassurance of worth, reliable alliance, guidance,
opportunity for nurturance. The respondent rates the extent to which each statement describes
their current social network on a scale from 1 (strongly disagree) to 4 (strongly agree). Items are
aggregated separately for each component (range = 4-16) and for a total perceived support score
(range = 24-96); higher scores indicate a greater degree of perceived support provisions. In this
study, Cronbach’s alpha for attachment = 0.76, social integration = 0.88, reassurance of worth =
0.72, reliable alliance = 0.84, guidance = 0.83, opportunity for nurturance = 0.79, and total
perceived support = 0.93 at baseline.
Adherence
Adherence generally describes the extent to which individuals are exposed to the content
of the intervention. For the current study, participants were asked to complete online PAJ
15
sessions an average of three 15-minute sessions per week, over 12 weeks, for a total of 36
journaling sessions throughout the course of the study. Overall PAJ adherence rate was calculated
using two methods: 1) weekly journaling counts for each participant—derived from online user
login counts—were recoded into a binary variable (i.e., yes, no) based on journaling > 1 times
per week. The journaling counts for all weeks were then summed, divided by 12, and multiplied
by 100 to calculate the overall 12-week adherence rate; 2) total journaling counts for all
participants were summed for all weeks of the study, divided by 36, and multiplied by 100.
Although it would be desirable to count actual time (minutes/session) spent engaged in the PAJ
intervention, the website was not capable of accurately tracking this information (e.g., if a person
left the computer to complete another task).
Sample Size
The sample size was calculated based on an anticipated baseline mean of 11.0 (SD=3) on
the HADS-A. This anticipated value was derived from a study of 273 medical patients
participating in a web-based education program. Using G*Power, we assumed treatment
condition standard deviations similar to those reported by Yun and colleagues [68] and a 5% type
I error rate for a two-sided hypothesis test, concluding that 31 subjects per group would provide
80% power to detect a difference in the HADS-A at 3 months (10.0 v. 8.0). This effect size is
based on a clinical trial of CBT for distressed medical patients, wherein the CBT arm decreased
their HADS-A score by 3.1 points more than controls (7), and a web-based CBT intervention by
Farrer and colleagues [21], that observed a 44% reduction in depression scores over 6 months.
The current study estimated a reduction of 2.0 in the HADS-A measure (an 18% reduction).
Anticipating a dropout rate of <10%, we planned to recruit 70 subjects at baseline.
Analytic Plan
16
All analyses were carried out using SAS Software version 9.4 (SAS Institute, Cary, NC).
First, descriptive statistics were calculated for all variables at baseline and each of the three
follow-up assessments, and response rates were calculated using the online user login tracking
logs. Categorical variables were summarized with frequencies and percentages, and continuous
variables were summarized with means, standard deviations, medians, and quartiles. The
distribution of continuous variables was checked using box plots, histograms, and normal
probability plots. For demographic variables and other characteristics measured at baseline,
comparison tests were conducted between the intervention and control groups using a two-
sample t-test or Wilcoxon Rank Sum test with means for continuous variables and using a Chi-
square test with percentages for categorical variables. A Fisher’s Exact test was used as needed
when cell counts were too small for the Chi-square test to be valid.
Second, in making comparisons of the differences from baseline to each of the 3 months
within and between groups, we used two approaches depending on the type of outcome variable.
For continuous outcome variables, we first found the change from baseline at each subsequent
month. A linear mixed effects model was then employed that included factors for group
(intervention versus control), month, the interaction between the intervention group and month,
and the baseline measurement for adjustment, and the differences between groups were
quantified with means. For binary outcome variables, a Generalized Estimating Equations model
was utilized that included factors for group, month, and the interaction between the group and
month, and differences between groups were quantified with percentages and odds ratios. All
comparisons were adjusted for age, sex, income, and preexisting journaling—reflecting self-
reported frequency (i.e., “Never,” “Less than once per month,” “1-3 times per month,” and “At
least once per week”) of writing in a diary or journal in the year leading up to the study—by
17
including these factors as additional covariates in the models. Missing data was not a significant
problem for the primary outcome variable (at less than 5%) or for the secondary outcome
variables (at less than 10% at most) and was not an issue for any independent variables.
RESULTS
Participants
A total of 99 people were assessed for eligibility of which 88 patients were interested in
participating. After further screening, 70 people were eligible, consented, and randomized to the
intervention (n=35) or usual care (n=35) condition (Fig 1). Three participants were lost to
follow-up during the 12-week assessment period and all participants were included in analyses.
No unanticipated harms were reported, and there were no privacy breaches or major technical
problems during the trial. Participants in this study had a broad range of chronic health
conditions, including: arthritic conditions (e.g., rheumatoid arthritis, gout, lupus, fibromyalgia;
27.1%), diabetes (type I or type II; 17.4%), asthma (17.1%), cancer (11.4 – 18.5%, all cancer
types combined), pre-diabetes (5.7%), kidney disease (not including kidney stones, bladder
infection, or incontinence; 4.3%), chronic obstructive pulmonary disease (1.4%), heart disease
(1.4%), and stroke (1.4%). Demographic and other baseline characteristics are shown in Table 1.
There were no significant differences between the intervention and control groups on any
baseline characteristics (demographics, primary or secondary outcomes).
Table 1. Participant and baseline characteriscs
Characteristic
Total
(N=70)
Control
(n=35)
Intervention
(n=35) p
Age 46.9 ± 12.8 47.2 ± 12.3 46.5 ± 13.5 .82
Female, % 87.0 85.7 88.2 .99
White, % 95.5 94.3 97.0 .99
Hispanic, % 1.5 2.9 0.0 .99
18
Married, % 63.8 71.4 55.9 .18
Education: College 4+ years, % 59.4 51.4 67.7 .17
Employed for Wages, % 76.8 74.3 79.4 .61
Income: <$50,000, % 32.8 30.0 35.3 .65
Current Smoker, % 4.3 2.9 5.7 .99
General health: Excellent/very good, % 49.3 41.2 57.1 .19
Hospital Anxiety and Depression Scale
- total
14.3 ± 6.6 14.3 ± 7.1 14.3 ± 6.1 .74
Hospital Anxiety and Depression Scale
- anxiety
9.8 ± 3.4 9.5 ± 3.4 10.1 ± 3.4 .44
Hospital Anxiety and Depression Scale
- depression
4.6 ± 4.0 4.9 ± 4.2 4.3 ± 3.7 .68
Perceived Stress Scale 19.9 ± 7.2 20.4 ± 6.7 19.4 ± 7.7 .55
Brief Resilience Scale 20.1 ± 5.3 20.8 ± 4.3 19.5 ± 6.1 .37
Satisfaction with Life Scale 19.0 ± 8.0 19.7 ± 7.3 18.3 ± 8.6 .48
Social Provisions Scale - total 79.0 ± 12.0 78.6 ± 10.9 79.3 ± 13.1 .55
Social Provisions Scale - attachment 12.5 ± 2.6 12.6 ± 2.4 12.5 ± 2.8 .94
Social Provisions Scale - social
integration
13.2 ± 2.7 13.3 ± 2.4 13.1 ± 2.9 .94
Social Provisions Scale - reassurance of
worth
12.9 ± 2.2 12.8 ± 1.9 12.9 ± 2.4 .41
Social Provisions Scale - reliable
alliance
13.6 ± 2.6 13.4 ± 2.6 13.8 ± 2.6 .47
Social Provisions Scale - guidance 13.3 ± 2.7 13.1 ± 2.6 13.5 ± 2.9 .35
Social Provisions Scale - opportunity
for nurturance
13.6 ± 2.5 13.7 ± 2.2 13.5 ± 2.7 .96
Positive and Negative Affect Scale -
positive affect
31.9 ± 8.2 33.4 ± 7.8 30.4 ± 8.4 .15
Positive and Negative Affect Scale -
negative effect
16.0 ± 6.6 16.7 ± 6.9 15.3 ± 6.2 .20
Days pain inhibited usual activities 3.5 ± 6.9 3.7 ± 6.1 3.4 ± 7.6 .08
Days not getting enough sleep 12.3 ± 9.7 13.1 ± 10.3 11.5 ± 9.2 .68
Days felt healthy and full of energy 14.0 ± 10.0 12.8 ± 9.4 15.3 ± 10.5 .30
Better mental health than 1 month ago 24.6 17.7 31.4 .18
Note.
p-values for variables based on means are from a two-sample t-test or Wilcoxon Rank Sum test;
p-values for variables based on %’s from a Chi-square test; exact tests were used as needed.
Within-Group Differences
Our initial analyses examined changes over time within each group. All results for the
within-group differences across the 3-month study period are shown in Table 2. Results indicated
that the PAJ intervention reduced mental distress and improved well-being. Specifically, the
19
intervention group reported lower HADS-A at all three assessments (at the end of months 1
through 3), more resilience at the end of month 2, less perceived stress at all three assessment
points, and a greater percentage (i.e., 56.3% vs. 31.3%) of participants reported better mental
health at the end of the first month, relative to baseline. No other within-group differences were
observed in the intervention group. Compared to baseline, the control group reported less social
integration at the end of month 3, more days in pain inhibiting usual activities at the end of
month 2, and a greater percentage (i.e., 41.4% vs. 20%) of participants reported better mental
health at the end of month 3 relative to the previous month. No other within-group differences
were observed in the control group.
Between-Group Differences
We next examined differences between the groups, over time. Results for the between-
group differences across the 3-month study period are also indicated in Table 2. Compared to the
control group, the intervention group exhibited lower anxiety at the end of month 1, lower
mental distress at the end of months 1 and 2, greater resilience and lower perceived stress at the
end of month 1, greater self-reported social integration at the end of month 2, and at the end of
month 2 they reported fewer days in which pain prohibited usual activities.
Table 2. Outcome variables over me
Control (n=30) Intervention (n=32) Interven.
vs.
Control
Outcome Time Mean [95% CI]
or %
d
or
h
pMean [95% CI]
or %
d
or
h
p p
Hospital Anxiety and Depression Scale - total
0 14.4 [11.7, 17.1] 13.8 [11.9, 15.7]
1 14.8 [12.2, 17.3] .06 .44 11.4 [9.5, 13.2] .48 .06 .01
2 14.3 [11.5, 17.1] .01 .72 11.5 [9.6, 13.5] .43 .06 .03
20
3 14.3 [11.4, 17.1] .02 .92 12.2 [9.7, 14.6] .27 .25 .27
Hospital Anxiety and Depression Scale - anxiety
0 9.6 [8.3, 10.9] 10.0 [8.8, 11.2]
1 9.7 [8.2, 11.1] .01 .60 7.9 [6.8, 9.0] .64 <0.001 .01
2 9.1 [7.5, 10.6] .14 .18 8.2 [7.0, 9.3] .55 0.002 .09
3 9.0 [7.5, 10.5] .16 .14 8.3 [7.0, 9.6] .49 0.003 .19
Hospital Anxiety and Depression Scale - depression
0 4.9 [3.3, 6.4] 3.8 [2.9, 4.8]
1 4.9 [3.3, 6.4] .02 .22 3.4 [2.4, 4.4] .14 .97 .19
2 5.1 [3.7, 6.5] .06 .14 3.4 [2.3, 4.4] .16 .89 .11
3 5.0 [3.5, 6.6] .04 .32 3.9 [2.5, 5.2] .02 .58 .69
Brief Resilience Scale
0 20.3 [18.7, 21.9] 19.7 [17.6, 21.8]
1 19.2 [17.2, 21.1] .24 .82 20.7 [18.8, 22.6] .17 .09 .049
2 20.1 [18.3, 21.9] .05 .50 21.5 [19.5, 23.6] .31 .01 .06
3 20.1 [17.9, 22.3] .04 .43 20.9 [18.9, 22.8] .20 .08 .35
Perceived Stress Scale
0 19.9 [17.4, 22.4] 19.0 [16.3, 21.7]
1 20.3 [17.7, 23.0] .06 .93 16.9 [14.5, 19.3] .30 .03 .047
2 18.9 [16.6, 21.3] .15 .27 16.6 [14.0, 19.3] .32 .02 .24
3 19.9 [17.1, 22.7] .00 .48 17.0 [14.1, 19.9] .26 .04 .17
Satisfaction With Life Scale
0 20.2 [17.4, 22.9] 18.8 [15.9, 21.7]
1 20.3 [17.4, 23.3] .02 .78 19.8 [17.0, 22.5] .12 .50 .69
2 20.2 [17.1, 23.2] .00 .88 20.2 [17.3, 23.0] .18 .32 .41
3 20.3 [17.1, 23.5] .02 .73 19.6 [16.7, 22.5] .10 .59 .86
Social Provisions Scale - total
0 79.0 [74.6, 83.3] 79.5 [74.8, 84.1]
1 78.1 [73.3, 82.8] .08 .31 79.5 [75.2, 83.8] .00 .73 .49
2 78.2 [73.9, 82.5] .07 .41 81.8 [77.1, 86.5] .18 .48 .11
3 78.1 [73.2, 82.9] .08 .33 80.3 [75.8, 84.9] .07 .98 .31
Social Provisions Scale - attachment
0 12.7 [11.7, 13.6] 12.5 [11.5, 13.5]
1 12.2 [11.1, 13.4] .15 .47 12.7 [11.7, 13.8] .08 .66 .29
2 12.4 [11.5, 13.4] .10 .68 12.9 [11.8, 13.9] .13 .42 .22
3 12.6 [11.6, 13.6] .02 .89 13.1 [12.1, 14.1] .20 .23 .18
Social Provisions Scale - social integration
0 13.3 [12.4, 14.2] 13.2 [12.1, 14.2]
1 12.7 [11.8, 13.6] .26 .11 12.9 [12.0, 13.7] .11 .40 .43
2 12.9 [12.0, 13.7] .20 .25 13.5 [12.6, 14.5] .15 .37 .04
3 12.4 [11.5, 13.2] .42 .03 13.2 [12.3, 14.1] .01 .93 .05
Social Provisions Scale - reassurance of worth
0 12.6 [11.9, 13.3] 13.2 [12.4, 13.9]
1 12.5 [11.6, 13.4] .07 .15 13.2 [12.3, 14.0] .00 .61 .35
2 12.2 [11.4, 13.1] .19 .09 13.6 [12.8, 14.4] .20 .84 .05
21
3 12.1 [11.2, 13.0] .27 .07 13.2 [12.4, 14.0] .01 .58 .21
Social Provisions Scale - reliable alliance
0 13.3 [12.3, 14.3] 13.6 [12.6, 14.5]
1 12.9 [11.7, 14.0] .15 .20 13.9 [13.1, 14.7] .13 .63 .08
2 13.3 [12.4, 14.3] .01 .79 14.3 [13.5, 15.2] .30 .16 .08
3 13.0 [11.9, 14.2] .10 .48 13.6 [12.6, 14.5] .00 .85 .61
Social Provisions Scale - guidance
0 13.1 [12.1, 14.0] 13.5 [12.4, 14.5]
1 13.2 [12.3, 14.1] .07 .86 13.5 [12.5, 14.4] .00 .97 .89
2 13.0 [12.1, 13.9] .03 .73 13.7 [12.6, 14.9] .09 .60 .39
3 12.7 [11.7, 13.8] .13 .46 13.7 [12.8, 14.7] .10 .52 .17
Social Provisions Scale - opportunity for nurturance
0 13.8 [13.1, 14.5] 13.5 [12.5, 14.4]
1 14.0 [13.2, 14.8] .08 .13 13.1 [12.2, 14.0] .15 .76 .09
2 13.9 [13.2, 14.7] .05 .15 13.4 [12.5, 14.3] .03 .62 .35
3 14.0 [13.2, 14.8] .08 .07 13.5 [12.6, 14.5] .02 .41 .33
Positive and Negative Affect Scale - positive affect
0 32.9 [30.1, 35.8] 30.6 [27.6, 33.6]
1 30.7 [27.7, 33.7] .28 .44 27.9 [24.3, 31.5] .29 .21 .65
2 30.8 [27.5, 34.2] .25 .46 29.2 [25.4, 33.0] .14 .51 .91
3 30.9 [27.6, 34.1] .25 .52 28.0 [24.0, 32.1] .26 .22 .59
Positive and Negative Affect Scale - negative affect
0 16.9 [14.1, 19.6] 14.9 [13.0, 16.8]
1 19.6 [16.4, 22.8] .34 .14 15.8 [13.8, 17.7] .16 .89 .17
2 19.6 [16.8, 22.4] .37 .15 17.3 [14.7, 19.8] .38 .25 .75
3 18.7 [15.4, 22.0] .23 .42 16.0 [13.6, 18.5] .19 .73 .61
General health (%)
0 44.8 59.4
1 34.5 .21 .31 59.4 .00 .99 .35
2 30.0 .31 .14 53.1 .13 .14 .43
3 34.5 .21 .45 46.9 .25 .21 .81
Days pain inhibited usual activities
0 4.0 [1.6, 6.4] 3.7 [0.8, 6.5]
1 4.9 [1.5, 8.2] .11 .24 3.1 [0.3, 5.9] .08 .95 .24
2 6.2 [3.0, 9.3] .29 .02 3.3 [0.6, 6.0] .05 .93 .02
3 3.9 [1.7, 6.0] .02 .62 4.2 [1.0, 7.3] .06 .49 .86
Days not getting enough sleep
0 13.8 [9.9, 17.7] 11.7 [8.3, 15.2]
1 13.4 [9.7, 17.1] .05 .51 11.1 [7.9, 14.3] .07 .19 .49
2 14.0 [10.5, 17.4] .01 .59 11.4 [8.2, 14.5] .04 .23 .51
3 13.6 [9.7, 17.4] .03 .50 10.9 [7.6, 14.1] .10 .19 .55
Days felt healthy and full of energy
0 13.2 [9.7, 16.7] 14.8 [11.0, 18.7]
1 12.9 [9.2, 16.5] .04 .80 15.4 [11.6, 19.2] .06 .38 .53
2 13.9 [10.4, 17.3] .07 .49 15.7 [12.1, 19.4] .09 .50 .98
22
3 15.1 [11.6, 18.5] .21 .21 13.5 [10.0, 17.0] .13 .91 .18
Better mental health (%)
0 20.0 31.3
1 26.7 .16 .47 56.3 .51 .03 .37
2 30.0 .23 .36 43.8 .26 .32 .97
3 41.4 .47 .02 40.6 .19 .37 .28
Note. p-values are from a linear mixed effects model (means) model with group, month, and
interaction between month and group as factors in addition n to covariates for adjustment.
Monthly means are raw means while change from baseline means are from the model adjusted
for covariates. Significant p-values are bold and italicized.
Adherence
We also examined patient adherence to suggested journaling frequency (or “dose”).
Overall adherence to the intervention, operationalized by dividing the mean amount of
completed sessions by the maximum amount of sessions, was moderate in the current sample
(mean = 47.8% with a range of 2.8% - 172.2%; one participant journaled 62 times with the
remainder having rates at or below the expected 100%). When operationalizing adherence as
completing at least one journaling session per week, a level consistent with the broader
expressive writing literature [23], the adherence rate was 66.4% (range of 41.7% - 100%). After
the first week of journaling, participants journaled an average of 0.94 times with a peak of 2.3
times per week in week 2. Overall, the number of journaling sessions generally decreased as time
progressed (see Figure 2). Adherence to PAJ sessions was largely unrelated to outcomes (data not
shown; results available upon request).
23
Figure 2. Average number of journaling sessions completed by participants over the 12-week
study period.
DISCUSSION
The primary aim of the current study was to examine whether a 12-week online PAJ
intervention could reduce mental distress, and improve psychological, interpersonal, and
physical well-being in a heterogeneous sample of medical patients with significant anxiety
symptoms. Compared to patients receiving standard care, patients randomized to the PAJ
intervention exhibited reduced mental distress, anxiety, and perceived stress, greater perceived
personal resilience and social integration, and fewer days on which pain inhibited usual
activities. The PAJ intervention was not associated with improvements in depressive symptoms,
satisfaction with life, other indices of social support (i.e., attachment, reassurance of worth,
reliable alliance, guidance, opportunity for nurturance, and overall perceived support), or
24
positive and negative affect. Overall, the current findings suggest that PAJ has potential utility as
an intervention for managing mental distress, particularly elevated anxiety symptoms, and other
aspects of well-being among general medical patients. This is consistent with, and extends, prior
research on positive writing interventions as a way to improve aspects of health and well-being
[55,69,70,71].
Effects on Well-being
The current study demonstrates that PAJ can improve several factors associated with
psychological well-being among patients with mild to moderate anxiety, each of which may have
implications for long-term health outcomes. Some of the most notable findings from this study
were that the PAJ intervention was associated with better mental health, including lower anxiety,
mental distress, and perceived stress after only one month of the intervention, and was associated
with reduced mental distress across time (e.g., continued reduction in anxiety and perceived
stress across the 12-week intervention period). As such, PAJ may be an effective way to improve
mental well-being and potentially increase longevity via improvements in on these outcomes in a
variety of patient populations. Additionally, PAJ was associated with higher perceived resilience.
Though PAJ may be a useful intervention for improving this outcome, more work is needed to
understand why the beneficial effects of PAJ on resiliency appear to taper off over-time, and
whether the initial increase in resiliency may serve as a mechanistic pathway between PAJ and
disease outcomes across a variety of medical conditions.
Another aspect of psychological well-being with potential implications for health
outcomes are perceptions of one’s social environment, such as perceived social integration. For
example, perceived social isolation, which is conceptually the opposite of perceived social
integration, is known to contribute to increased risk for early mortality [72]. The current study
indicates that PAJ may be beneficial to this end as the results demonstrate greater self-reported
25
social integration in the intervention group relative to controls at the end of the second month.
Again, more work is needed to understand why the effects of PAJ on perceived social integration
do not appear to hold over time and whether these benefits can be prolonged.
Surprisingly, one indicator of improved well-being (i.e., percent of patients who self-
reported “somewhat” or “much better” mental health compared to previous month) was observed
in the control group. This indicates that patients receiving usual care will fluctuate in their
perceived well-being (i.e., have occasional up-swings in self-reported mental health).
Additionally, PAJ was not observed to improve all indices of well-being (i.e., we did not see
beneficial effects on depression, satisfaction with life, indices of social support other than
perceived integration, or positive and negative affect). This indicates that this intervention may
be effective for improving some, but certainly not all aspects of well-being. To explore the
robustness of PAJ for enhancing additional indices of well-being, future studies would benefit
from exploring ways to modify the expressive writing methods used in the current study to
increase their effectiveness (e.g., across more well-being outcomes, for longer durations of time,
or both). For example, the writing schedule in the current study was fairly dense (three writing
sessions per week for 12 weeks). Perhaps patients would benefit from a less dense writing
schedule, or greater variability in the topics offered. It may also be possible to optimize benefit
by tailoring PAJ instructions (overall, or adaptively over time) to individual patient needs.
Furthermore, the writing task utilized in the current study was a modified version of that
developed by Pennebaker and colleagues [23], particularly in terms of the number of overall
sessions; it is possible that writing in a manner more consistent with earlier expressive writing
studies would be preferable.
Timing Effects
26
Regarding the timing effects as a whole, the first month of the PAJ intervention provides
a considerable number of improvements in quality of life that are still observed two and three
months later, albeit to a lesser degree. For example, PAJ was associated with decreased mental
distress and improved well-being in the first month, but the number of benefits and between-
group differences diminished over time. Perhaps the benefits of PAJ are largely observed within
only at the start of the intervention and do not provide sustained improvements in mental distress
and well-being over time. However, other studies have demonstrated that longer-lasting positive
psychology interventions are effective at improving subjective and psychological well-being
among cancer patients [73], and the “short-term” improvements in cancer patients’ mental well-
being observed in this study may translate into longer-term health benefits. As such, future
studies are needed to determine whether PAJ beyond 3 months would provide additional
upswings in well-being. Conducting longer investigations of expressive writing in clinical
populations may be particularly important, as previous work has found the benefits to dissipate
after several months of discontinuation [28]. It is worth noting that the sample size of our study
was modest, and, even after just 3 months, improvements were observed that favor the PAJ
group. Although a portion of the effect sizes for these significant improvements were small (i.e.,
Cohen’s d or h: .5 – .49), there were also several effect sizes of moderate size (i.e., Cohen’s d or
h: .51 – .64; see Table 2). Given the potential for cost-efficiency and reach of this online
intervention, we view these preliminary results as promising and supportive of a larger follow up
study examining the clinical utility of PAJ interventions.
Feasibility of Intervention
An important aspect of this study is the demonstrated feasibility of the web-based writing
task intervention. First, participants generally enjoyed the intervention (i.e., 39.4% reported that
27
the journaling activity made them feel “somewhat better” and 18.2% reported that it made them
feel “much better”). A total of 67 out of 70 consented and randomized participants competed the
study for an overall excellent completion rate of 95%; this compares favorably to other
randomized expressive writing interventions in chronic illness samples (e.g., 73% [74]; 81%
[53]). Overall adherence to the intervention was moderate in the current sample (mean = 47.8%).
However, when operationalizing adherence as completing at least one journaling session per
week, the adherence rate rose to 66.4% (range of 41.7% - 100%). Once-weekly sessions are
common in therapeutic practice and are frequently used in randomized trials of CBT (e.g., [75]).
While these adherence rates are acceptable, it remains unclear why adherence was not even
higher given the relative ease through which online PAJ modules could be accessed.
Generally, the number of completed journaling sessions decreased over the course of the
intervention. The reasons for this decrease are uncertain, but several plausible explanations exist.
One possibility, though unassessed in the current study, is that participants began the journaling
process with enthusiasm in the early weeks, but experienced reduced interest or increased fatigue
with the intervention over time. Another possible explanation is that participants believed there
were diminishing returns on therapeutic benefit as journaling sessions increased; a single weekly
journaling session could have been deemed therapeutically equivalent to multiple sessions, for
example. Online interventions are becoming more prevalent (e.g., [76]) as they can be
administered at lower costs and disseminated to more people. Future work should investigate
factors that drive adherence to online PAJ interventions and explore opportunities to improve the
interventions themselves, given the benefits observed in this study.
Limitations
Given the preliminary nature of investigating this novel online intervention, we included
several outcomes and conducted a large number of statistical tests, and the small sample size
28
reduced our power to detect some effects (especially when contrasting between groups) and may
contribute to spurious findings. As such, care should be taken in interpreting any specific effect
and replication of these effects is warranted before strong conclusions can be made about
potential efficacy. Additionally, the length of the study was relatively short, and it remains
unclear whether longer-term interventions would be sustainable or show similar improvements in
various indices of well-being. Evidence from some clinical populations (e.g., patients with
asthma [29]) suggests that expressive writing may offer the most benefit for those with moderate
level of disease—patients that are relatively healthy do not have much room for improvement,
and those with very severe illness may require a more powerful treatment alternative. A large
percentage (44.3%) of patients in the current study reported excellent or very good health at
baseline, possibly limiting the therapeutic benefit that could be observed from journaling; less
healthy patients may have greater gains to make in well-being from baseline relative to their
healthier counterparts, and future studies may consider testing this intervention in a sample of
patients with greater disease severity. Finally, the homogenous nature of our small study sample
(95.5% White, 87.0% female), combined with the relatively brief study timeframe, limits the
generalizability of our findings to more diverse patient samples.
Conclusions
The results of this randomized controlled trial provide preliminary evidence that PAJ is a
feasible and well-accepted intervention that can be implemented online for effectively reducing
some aspects of mental distress and improving aspects of well-being among medical patients
with mild to moderate anxiety symptoms. Moreover, PAJ is likely to be a more pleasant and
uplifting treatment for patients compared to the traditional expressive writing interventions that
focus on writing about deeply distressing and traumatic experiences from the past; this may
promote acceptability and treatment engagement relative to other treatments. Thus, this relatively
29
simple and cost-effective intervention may represent a low-risk way to improve a variety of well-
being domains, particularly among underserved patients.
30
ACKNOWLEDGEMENTS
Funding for this work was provided by the Penn State Social Science Research Institute (SSRI).
The funders have no role in the study design, data collection and analysis, decision to publish, or
manuscript preparation. Study data were collected and managed using Research Electronic Data
Capture tools (REDCap) hosted at the Penn State Milton S. Hershey Medical Center and College
of Medicine. The authors would like to thank Vanessa Juth for feedback on an earlier version of
this work.
CONFLICTS OF INTEREST
The authors have no conflicts of interest to report.
31
References
[1] Buttorff, C., Ruder, T., & Bauman, M. (2017). Multiple Chronic Conditions in the United
States. RAND Corporation: Santa Monica, CA. ISBN: 978-0-8330-9737-8
[2] Megari, K. (2013). Quality of life in chronic disease patients. Health Psychol Res, 1,
e27. PMID: 26973912
[3] Sambamoorthi, U., Tan, X., & Deb, A. (2015). Multiple chronic conditions and healthcare
costs among adults. Expert Rev Pharmacoecon Outcomes Res, 15, 823–832. PMID: 26400220
[4] Turner, J., & Kelly, B. (2000). Emotional dimensions of chronic disease. West J Med, 172,
124-128. PMID: 10693376
[5] Druss, B. G. & Walker, E. R. (2011). Mental disorders and medical comorbidity. Synth Proj
Res Synth Rep, 21, 1-26. PMID: 21675009
[6] Alegria, M., Jackson, J. S., Kessler, R. C., & Takeuchi, D. (2015). United States National
Comorbidity Survey: Replication 2001–2003. Ann Arbor, MI: Inter-university Consortium for
Political and Social Research [distributor]. doi:10.3886/ICPSR20240.v7
[7] Kessler, R. C., Berglund, P., Chiu, W. T., Demler, O., Heeringa, S., Hiripi, E., . . . Zheng, H.
(2004). The US national comorbidity survey replication (NCS-R): Design and field
procedures. Int J Methods Psychiatr Res, 13, 69-92. PMID: 15297905
[8] Schneiderman, N., Ironson, G., & Siegel, S. D. (2005). Stress and health: Psychological,
behavioral, and biological determinants. Annu Rev Clin Psychol, 1, 607–628. PMID: 17716101
[9] Hiller, A., Quinn, J., & Schmidt, P. (2017). Does psychological stress affect the progression
of Parkinson’s Disease. Neurology, 88, Supp. N5.002.
[10] Orth-Gomér, K., Wamala, S. P., Horsten, M., Schenck-Gustafsson, K., Schneiderman, N., &
Mittleman, M. A. (2000). Marital stress worsens prognosis in women with coronary heart
disease: The Stockholm female coronary risk study. JAMA, 284, 3008-3014. PMID: 11122587
[11] Salleh, M. R. (2008). Life event, stress and illness. Malays J Med Sci, 15(4), 9-18. PMID:
22589633
32
[12] Hofmann, S. G., Asnaani, A., Vonk, I. J. J., Sawyer, A. T., & Fang, A. (2012). The efficacy
of cognitive behavioral therapy: A review of meta-analyses. Cognit Ther Res, 36, 427–440.
PMID: 23459093
[13] Moorey, S., Cort, E., Kapari, M., Monroe, B., Hansford, P., Mannix, K., . . . Hotopf, M.
(2009). A cluster randomized controlled trial of cognitive behaviour therapy for common mental
disorders in patients with advanced cancer. Psychol Med, 39, 713-723. PMID: 18761755
[14] Moss-Morris, R., Dennison, L., Landau, S., Yardley, L., Silber, E., & Chalder, T. (2013). A
randomized controlled trial of cognitive behavioral therapy (CBT) for adjusting to multiple
sclerosis (the saMS trial): Does CBT work and for whom does it work? J Consult Clin Psychol,
81, 251-262. PMID: 22730954
[15] Mohr, D. C., Ho, J., Duffecy, J., Baron, K. G., Lehman, K. A., Jin, L., & Reifler, D. (2010).
Perceived barriers to psychological treatments and their relationship to depression. J Clin
Psychol, 66, 394-409. PMID: 20127795
[16] Bijl, R. V., de Graaf, R., Hiripi, E., Kessler, R. C., Kohn, R., Offord, D.R., … Wittchen, H.
U. (2003). The prevalence of treated and untreated mental disorders in five countries. Health Aff
(Millwood), 22, 122-133. PMID: 12757277
[17] Corrigan, P. (2004). How stigma interferes with mental health care. Am Psychol, 59, 614-
625. PMID: 15491256
[18] Hewitt, M., & Rowland, J. H. (2002). Mental health service use among adult cancer
survivors: Analyses of the National Health Interview Survey. J Clin Oncol, 20, 4581-4590.
PMID: 12454116
[19] Ritterband, L. M., & Tate, D. F. (2009). The science of internet interventions. Ann Behav
Med, 38, 1-3. PMID: 19816750
[20] Spek, V., Cuijpers, P., Nyklíček, I., Riper, H., Keyzer, J., & Pop, V. (2007). Internet-based
cognitive behaviour therapy for symptoms of depression and anxiety: A meta-analysis. Psychol
Med, 37, 319-328. PMID: 17112400
[21] Farrer, L., Christensen, H., Griffiths, K. M., & Mackinnon, A. (2011). Internet-based CBT
for depression with and without telephone tracking in a national helpline: Randomised controlled
trial. PLoS ONE, 6, 9. PMID: 22140514
[22] Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: Toward an
33
understanding of inhibition and disease. J Abnorm Psychol, 95, 274-281. PMID: 3745650
[23] Pennebaker, J. W., & Smyth, J. M. (2016). Opening up by writing it down: how expressive
writing improves health and eases emotional pain. New York: The Guilford Press. ISBN:
1462524923
[24] Frattaroli, J. (2006). Experimental disclosure and its moderators: A meta-analysis. Psychol
Bull, 132, 823-865. PMID: 17073523
[25] Smyth, J. M., Stone, A. A., Hurewitz, A., & Kaell, A. (1999). Effects of writing about
stressful experiences on symptom reduction in patients with asthma or rheumatoid arthritis: A
randomized trial. JAMA, 281, 1304-1309. PMID: 10208146
[26] Danoff-Burg, S., Agee, J. D., Romanoff, N. R., Kremer, J. M., & Strosberg, J. M. (2006).
Benefit finding and expressive writing in adults with lupus or rheumatoid arthritis. Psychol
Health, 21, 651-665. doi:10.1080/14768320500456996
[27] Lumley, M. A., Keefe, F. J., Mosley-Williams, A., Rice, J. R., McKee, D., Waters, S. J., . . .
Riordan, P. A. (2014). The effects of written emotional disclosure and coping skills training in
rheumatoid arthritis: A randomized clinical trial. J Consult Clin Psychol, 82, 644-658. PMID:
24865870
[28] Lumley, M. A., Leisen, J. C. C., Partridge, R. T., Meyer, T. M., Radcliffe, A. M., Macklem,
D. J., . . . Granda, J. L. (2011). Does emotional disclosure about stress improve health in
rheumatoid arthritis? Randomized, controlled trials of written and spoken disclosure. Pain, 152,
866-877. PMID: 21315515
[29] Smith, H. E., Jones, C. J., Hankins, M., Field, A., Theadom, A., Bowskill, R., . . . Frew, A. J.
(2015). The effects of expressive writing on lung function, quality of life, medication use, and
symptoms in adults with asthma: A randomized controlled trial. Psychosom Med, 77, 429-437.
PMID: 25939030
[30] Broderick, J. E., Junghaenel, D. U., & Schwartz, J. E. (2005). Written emotional expression
produces health benefits in fibromyalgia patients. Psychosom Med, 67, 326-334. PMID:
15784801
[31] Hsu, M. C., Schubiner, H., Lumley, M. A., Stracks, J. S., Clauw, D. J., & Williams, D. A.
(2010). Sustained pain reduction through affective self-awareness in fibromyalgia: A randomized
controlled trial. J Gen Intern Med, 25, 1064-1070. PMID: 20532650
34
[32] Halpert, A., Rybin, D., & Doros, G. (2010). Expressive writing is a promising therapeutic
modality for the management of IBS: A pilot study. Am J Gastroenterol, 105, 2440-2448. PMID:
20551938
[33] Mann, T. (2001). Effects of future writing and optimism on health behaviors in HIV-infected
women. Ann Behav Med, 23, 26-33. doi:10.1207/S15324796ABM2301_5
[34] Petrie, K. J., Fontanilla, I., Thomas, M. G., Booth, R. J., & Pennebaker, J. W. (2004). Effect
of written emotional expression on immune function in patients with human immunodeficiency
virus infection: A randomized trial. Psychosom Med, 66, 272-275. PMID: 15039514
[35] McGuire, K. M. B., Greenberg, M. A., & Gevirtz, R. (2005). Autonomic effects of
expressive writing in individuals with elevated blood pressure. J Health Psychol, 10, 197-209.
PMID: 15723890
[36] Willmott, L., Harris, P., Gellaitry, G., Cooper, V., & Horne, R. (2011). The effects of
expressive writing following first myocardial infarction: A randomized controlled trial. Health
Psychol, 30, 642-650. PMID: 21875209
[37] Hevey, D., Wilczkiewicz, E., & Horgan, J. H. (2012). Type D moderates the effects of
expressive writing on health-related quality of life (HRQOL) following myocardial infarction
(MI). Ir J Psychol, 33, 107-114. doi:10.1080/03033910.2012.708901
[38] Merz, E. L., Fox, R. S., & Malcarne, V. L. (2014). Expressive writing interventions in
cancer patients: A systematic review. Health Psychology Rev, 8, 339-361. PMID: 25053218
[39] Krpan, K. M., Kross, E., Berman, M. G., Deldin, P. J., Askren, M. K., & Jonides, J. (2013).
An everyday activity as a treatment for depression: The benefits of expressive writing for people
diagnosed with major depressive disorder. J Affect Disord, 150, 1148-1151. PMID: 23790815
[40] Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and
implications of generalized outcome expectancies. Health Psychol, 4, 219-247. PMID: 4029106
[41] Robles, T. F., Brooks, K. P., & Pressman, S. D. (2009). Trait positive affect buffers the
effects of acute stress on skin barrier recovery. Health Psychol, 28, 373-378. PMID: 19450044
35
[42] Steptoe, A., Wardle, J., & Marmot, M. (2005). Positive affect and health-related
neuroendocrine, cardiovascular, and inflammatory processes. Proc Natl Acad Sci U S A, 102(18),
6508-6512. PMID: 15840727
[43] Lyubomirsky, S., King, L., & Diener, E. (2005). The benefits of frequent positive affect:
Does happiness lead to success? Psychol Bull, 131, 803-855. PMID: 16351326
[44] Conversano, C., Rotondo, A., Lensi, E., Della Vista, O., Arpone, F., & Reda, M. A. (2010).
Optimism and its impact on mental and physical well-being. Clin Pract Epidemiol Ment
Health, 6, 25-29. PMID: 20592964
[45] Diener, E., & Chan, M.Y. (2011). Happy people live longer: Subjective well-being
contributes to health and longevity. Appl Psychol Health Well-being, 3, 1-43. doi:10.1111/j.1758-
0854.2010.01045.x
[46] Cohen, S., & Pressman, S. D. (2006). Positive affect and health. Curr Dir Psychol Sci, 15,
122-125. doi:10.1111/j.0963-7214.2006.00420.x
[47] Pressman, S. D., & Cohen, S. (2005). Does positive affect influence health? Psychol Bull,
131, 925-971. PMID: 16351329
[48] Low, C. A., Stanton, A. L., & Danoff-Burg, S. (2006). Expressive disclosure and benefit
finding among breast cancer patients: Mechanisms for positive health effects. Health
Psychol, 25, 181-189. PMID: 16569109
[49] Park, C. L., & Blumberg, C. J. (2002). Disclosing trauma through writing: Testing the
meaning-making hypothesis. Cognit Ther Res, 26, 597-616. doi:10.1023/A:1020353109229
[50] King, L. A. (2001). The health benefits of writing about life goals. Pers Soc Psychol
Bull, 27, 798-807. doi:10.1177/0146167201277003
[51] Ogedegbe, G. O., Boutin-Foster, C., Wells, M. T., Allegrante, J. P., Isen, A. M., Jobe, J. B.,
& Charlson, M. E. (2012). A randomized controlled trial of positive-affect intervention and
medication adherence in hypertensive African Americans. Arch Intern Med, 172, 322–326.
PMID: 22269592
[52] Peterson, J. C., Charlson, M. E., Hoffman, Z., Wells, M. T., Wong, S.-C., Hollenberg, J. P.,
… Allegrante, J. P. (2012). Randomized controlled trial of positive affect induction to promote
36
physical activity after percutaneous coronary intervention. Arch Intern Med, 172, 329–336.
PMID: 22269589
[53] Stanton, A. L., Danoff-Burg, S., Sworowski, L. A., Collins, C. A., Branstetter, A. D.,
Rodriquez-Hanley, A., … Austenfeld, J. L. (2002). Randomized, controlled trial of written
emotional expression and benefit finding in breast cancer patients. J Clin Oncol, 20, 4160-4168.
PMID: 12377959
[54] Armitage, C. J., Harris, P. R., & Arden, M. A. (2011). Evidence that self-affirmation reduces
alcohol consumption: Randomized exploratory trial with a new, brief means of self-affirming.
Health Psychol, 30, 633-41. PMID: 21553966
[55] Burton, C. M., & King, L. A. (2008). Effects of (very) brief writing on health: The two-
minute miracle. Br J Health Psychol, 13, 9-14. PMID: 18230223
[56] Kobak, K. A., Taylor, L. v., Dottl, S. L., Greist, J. H., Jefferson, J. W., Burroughs, D., . . .
Serlin, R. C. (1997). A computer-administered telephone interview to identify mental
disorders. JAMA, 278, 905-910. PMID: 9302242
[57] Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and depression scale. Acta
Psychiatr Scand, 67, 361-370. PMID: 6880820
[58] Conill, C., Verger, E., & Salamero, M. (1990). Performance status assessment in cancer
patients. Cancer, 65, 1864-1866. PMID: 2317765
[59] Frisina, P. G., Borod, J. C., & Lepore, S. J. (2004). A meta-analysis of the effects of written
emotional disclosure on the health outcomes of clinical populations. J Nerv Ment Dis, 192, 629-
634. PMID: 15348980
[60] Aspinwall, L. G., & Tedeschi, R. G. (2010). The value of positive psychology for health
psychology: Progress and pitfalls in examining the relation of positive phenomena to health. Ann
Behav Med, 39, 4-15. PMID: 20091429
[61] Moriarty, D. G., Zack, M. M., & Kobau, R. (2003). The Centers for Disease Control and
Prevention’s Healthy Days measures – Population tracking of perceived physical and mental
health over time. Health Qual Life Outcomes, 1, 1-8. PMID: 14498988
[62] Bjelland, I., Dahl, A. A., Haug, T. T., & Neckelmann, D. (2002). The validity of the Hospital
Anxiety and Depression Scale. An updated literature review. J Psychosom Res, 52, 69–77.
PMID: 11832252
37
[63] Smith, B. W., Dalen, J., Wiggins, K., Tooley, E., Christopher, P., & Bernard, J. (2008). The
brief resilience scale: Assessing the ability to bounce back. Int J Behav Med, 15, 194-200. PMID:
18696313
[64] Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived
stress. J Health Soc Behav, 24, 385-396. PMID: 6668417
[65] Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief
measures of positive and negative affect: The PANAS scales. J Pers Soc Psychol, 54, 1063-1070.
PMID: 3397865
[66] Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life
scale. J Pers Assess, 49, 71-75. PMID: 16367493
[67] Cutrona, C., Russell, D., & Rose, J. (1986). Social support and adaptation to stress by the
elderly. Psychol Aging, 1, 47-54. PMID: 3267379
[68] Yun, Y. H., Lee, K. S., Kim, Y. W., Park, S. Y., Lee, E. S., Noh, D. Y,, … Park, S. (2012).
Web-based tailored education program for disease-free cancer survivors with cancer-related
fatigue: A randomized controlled trial. J Clin Oncol, 30, 1296-1303. PMID: 22412149
[69] Burton, C. M., & King, L. A. (2004). The health benefits of writing about intensely positive
experiences. J Res Pers, 38, 150-163. doi:10.1016/S0092-6566(03)00058-8
[70] Lambert, N. M., Gwinn, A. M., Baumeister, R. F., Strachman, A., Washburn, I. J., Gable, S.
L., & Fincham, F. D. (2013). A boost of positive affect: The perks of sharing positive
experiences. J Soc Pers Relat, 30, 24-43. doi:10.1177/0265407512449400
[71] Reiter, C., & Wilz, G. (2016). Resource diary: A positive writing intervention for promoting
well-being and preventing depression in adolescence. J Posit Psychol, 11, 99-108.
doi:10.1080/17439760.2015.1025423
[72] Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness
and social isolation as risk factors for mortality: A meta-analytic review. Perspect Psychol Sci,
10, 227-237. PMID: 25910392
[73] Bolier, L., Haverman, M., Westerhof, G. J., Riper, H., Smit, F., & Bohlmeijer, E. (2013).
Positive psychology interventions: A meta-analysis of randomized controlled studies. BMC
Public Health, 13, 119. doi:10.1186/1471-2458-13-119
38
[74] Milbury, K., Spelman, A., Wood, C., Matin, S. F., Tannir, N., Jonasch, E., … Cohen, L.
(2014). Randomized controlled trial of expressive writing for patients with renal cell carcinoma.
J Clin Oncol, 32, 663–670. PMID: 24470003
[75] Arch, J., Eifert, G. H., Davies, C., Plumb Vilardaga, J. C., Rose, R. D., & Craske, M. G.
(2012). Randomized clinical trial of cognitive behavioral therapy (CBT) versus acceptance and
commitment therapy (ACT) for mixed anxiety disorders. J Consult Clin Psychol, 80, 750-765.
PMID: 22563639
[76] Baikie, K. A., Liesbeth, G., & Wilhelm, K. (2012). Expressive writing and positive writing
for participants with mood disorders: An online randomized controlled trial. J Affect Disord, 136,
310-319. PMID: 22209127
