Content uploaded by Sonja Lyubomirsky
Author content
All content in this area was uploaded by Sonja Lyubomirsky on Dec 14, 2017
Content may be subject to copyright.
Feasibility and utility of positive psychology exercises for suicidal inpatients☆
Jeff C. Huffman, M.D.
a,b,
⁎, Christina M. DuBois, B.A.
a
, Brian C. Healy, Ph.D.
b,c
, Julia K. Boehm, Ph.D.
d
,
Todd B. Kashdan, Ph.D.
e
, Christopher M. Celano, M.D.
a,b
, John W. Denninger, M.D., Ph.D.
a,b,f
,
Sonja Lyubomirsky, Ph.D.
g
a
Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA
b
Harvard Medical School, Boston, MA, USA
c
Biostatistics Center, Massachusetts General Hospital, Boston, MA, USA
d
Department of Psychology, Chapman University, Orange, CA, USA
e
Department of Psychology, George Mason University, Fairfax, VA, USA
f
Benson Henry Institute for Mind Body Medicine, BostonMA, USA
g
Department of Psychology, University of California—Riverside, Riverside, CA, USA
abstractarticle info
Article history:
Received 12 July 2013
Revised 3 October 2013
Accepted 5 October 2013
Keywords:
Gratitude
Hopelessness
Optimism
Positive psychology
Suicide
Objective: The objective was to assess the feasibility and acceptability of nine positive psychology exercises
delivered to patients hospitalized for suicidal thoughts or behaviors, and to secondarily explore the relative
impact of the exercises.
Method: Participants admitted to a psychiatric unit for suicidal ideation or behavior completed daily positive
psychology exercises while hospitalized. Likert-scale ratings of efficacy (optimism, hopelessness, perceived
utility) and ease of completion were consolidated and compared across exercises using mixed models
accounting for age, missing data and exercise order. Overall effects of exercise on efficacy and ease were also
examined using mixed models.
Results: Fifty-two (85.3%) of 61 participants completed at least one exercise, and 189/213 (88.7%) assigned
exercises were completed. There were overall effects of exercise on efficacy (χ
2
=19.39; P= .013) but not ease
of completion (χ
2
=11.64; P=.17), accounting for age, order and skipped exercises. Effect (Cohen’sd)of
exercise on both optimism and hopelessness was moderate for the majority of exercises. Exercises related to
gratitude and personal strengths ranked highest. Both gratitude exercises had efficacy scores that were
significantly (P=.001) greater than the lowest-ranked exercise (forgiveness).
Conclusion: In this exploratory project, positive psychology exercises delivered to suicidal inpatients were
feasible and associated with short-term gains in clinically relevant outcomes.
© 2014 Elsevier Inc. All rights reserved.
1. Introduction
Patients who are psychiatrically hospitalized for a suicide attempt
or suicidal ideation (SI) are at very high risk for suicide following
discharge [1–3], with rates of suicide shortly postdischarge more than
100 times greater than the rate in the general population [4]. Despite
this high risk of suicide, there has been relatively limited study of
formal interventions to reduce suicidality during or shortly after
psychiatric admission.
Existing interventions for suicidal patients (e.g., cognitive behav-
ioral therapy [5,6] and dialectical behavior therapy [7,8]) typically
target negative emotions and cognitions. However, positive cognitions
and emotions also appear to be important in reducing suicide risk, and
reducing negative emotions may not automatically increase positive
psychological states. For example, optimism and depression are only
moderately correlated (median r=−.43 in 10 studies [9]), signaling
that these constructs are not simply two sides of the same coin.
Optimism, gratitude and other positive states have been associated
with reductions in hopelessness, suicidal ideation and suicide
attempts, often independent of depression [10–16]. Positive emotions
in suicidal patients have been linked to improved problem-solving
[17], and a study of patients hospitalized for self-harm found that low
positive future orientation more strongly predicted recurrent self-
harm after discharge than did global hopelessness [18]. Therefore,
explicit targeting of positive cognitive and emotional states like
gratitude and positive future orientation may have effects on suicide
risk above and beyond standard approaches.
Positive psychology (PP) interventions could represent an inno-
vative and effective adjunctive tool for patients at high suicide risk.
PP-based exercises focus on cultivating or amplifying a specific
positive cognition or emotion. Representative interventions focus on
General Hospital Psychiatry 36 (2014) 88–94
☆Funding: This work was supported in part by grant R01-DP00336 from the US
Centers for Disease Control and Prevention to Dr. Herbert Benson. There was no
other funding source.
⁎Corresponding author. Massachusetts General Hospital, 55 Fruit Street, Blake 11
Boston, MA 02114. Tel.: +1 617 724 2910; fax: + 1 617 724 9155.
E-mail address: jhuffman@partners.org (J.C. Huffman).
0163-8343/$ –see front matter © 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.genhosppsych.2013.10.006
Contents lists available at ScienceDirect
General Hospital Psychiatry
journal homepage: http://www.ghpjournal.com
expressing optimism, generating gratitude and reestablishing life
purpose [19–22]. In contrast to more intensive interventions, PP
exercises require minimal provider training, are often straightforward
and enjoyable, and are accessible to a variety of patients, including
those who cannot attend in-person appointments.
Overall, PP exercises have been successfully used by more than
4000 participants in over 50 clinical trials [20,21], though typically in
nonclinical populations. However, PP exercises, including those
focusing on gratitude, mindfulness and using strengths, have led to
(often persistent) improvements in mood symptoms in patients with
depressive symptoms [19,23,24], and a prior intervention on
meaningful goals in older adults with suicidal thoughts led to greater
self-efficacy and psychological well-being [25].
Such interventions have the potential to be useful for patients in
crisis. These exercises could be easily administered to hospitalized
patients, with the goal of increasing positive psychological states by
discharge, when suicide risk is high. In addition, these interventions
could be used postdischarge, and an efficient exploration of feasibility in
suicidal patients could also inform that future implementation. Still, it
has been unclear whether PP exercises would be feasible or acceptable
in a broad population of hospitalized suicidal patients, or whether they
would impact clinical outcomes in this seriously ill cohort, with valid
concerns that such exercises may be a “bridge too far.”
Accordingly, in this study, we examined as a primary aim the
feasibility and acceptability of nine PP exercises in patients admitted
to a psychiatric unit for SI or a suicide attempt. Secondarily, we sought
to explore the self-reported efficacy of the exercises as a whole and to
compare such efficacy ratings among individual exercises.
2. Methods
2.1. Study overview
This study assessed the feasibility of PP exercises among suicidal
inpatients (clinicaltrials.gov identifier=NCT01398891). While hos-
pitalized, participants completed one randomly assigned PP exercise
daily under the guidance of a study trainer, self-assessed their
optimism and hopelessness immediately before and after exercise
completion, and rated the ease and utility of each exercise.
Institutional Review Board approval was obtained prior to all
study procedures.
2.2. Participants
Eligible participants were adults admitted to the inpatient
psychiatric unit at an urban academic medical center, for SI or
following a suicide attempt, between May 2011 and February 2013.
The 24-bed inpatient psychiatric unit is housed within the general
hospital and cares for a broad population of adults 18 years and older.
Mood disorders are the most common primary diagnosis, although
the unit cares for patients across the spectrum of psychiatric illness.
The mean admission duration is approximately 9 days.
2.3. Inclusion and exclusion criteria
Inclusion criteria for the study were admission to the unit for SI
(passive or active) or a suicide attempt confirmed by medical record.
Potential participants were identified by review of the unit census,
followed by a discussion with the patient’s attending psychiatrist to
determine eligibility. Patients who expressed interest in an optional
study were then assessed by a physician or social work investigator
for exclusion criteria.
Patients were excluded if they were to be discharged within 24 h
or were unable to meaningfully participate due to inability to
communicate in English or cognitive deficits identified by a six-item
screen [26]. Patients were also excluded if they had current psychotic
symptoms (identified via the Mini International Neuropsychiatric
Interview) [27]. If patients met all study criteria, they were provided a
verbal description of the study and written consent form, and
informed consent was obtained from a licensed study investigator.
Of note, although primary admission diagnosis was recorded,
patients were not excluded from the study based on their primary
psychiatric diagnosis. While studying a more homogeneous sample of
patients can increase internal validity, the goal of the project was to
identify the acceptability and utility of PP exercises in a broad
population of suicidal patients given that clinical implementation of
the exercises in this setting would likely occur across diagnoses rather
than in a single diagnostic cohort.
2.4. Study design
Prior to the recruitment phase, nine exercises were identified via
published literature [19,28–39] or directly from researchers, and
modified appropriately for this population. Additional text outlining
the rationale and instructions for each exercise was added to create
written packets for each exercise to be provided to patients. Staff
members involved in intervention delivery (a licensed clinical social
worker and a Bachelor’s-level research coordinator with relevant
inpatient clinical experience) were trained in several stages. They
reviewed a provider training manual created specifically for the
project (exercise packets and training manual available from the
authors); this manual described the rationale and procedures for each
exercise, provided guidance for maintaining the focus of the
interaction solely on the PP exercise and its review, and contained
specific advice to convey to participants to facilitate completion of the
given exercise (e.g., methods of brainstorming). Staff also completed
relevant background reading about PP, observed sessions and
completed all exercises in pairs to gain experience performing and
reviewing each exercise. Once enrollment was ongoing, sessions were
reviewed with the principal investigator, who provided feedback.
2.5. Exercises
The nine PP exercises were selected after literature review and
consultation with senior team members. The exercises were as
follows (with related citations):
•Gratitude letter [19,32,33]: Participants recalled another’s kind
act and wrote a letter to the person that described feelings of
gratitude associated with this event. Participants could send the
letter if desired.
•Personal strengths [19,40]: Participants completed a brief survey
of personal strengths, selected a strength (e.g., perseverance,
humility) to be used deliberately in the next 24 hours and then
wrote about how they used the strength and the outcome.
•Acts of kindness [34,35]: Participants performed three kind
acts for others within a single day and wrote about the acts and
the outcome.
•Important, enjoyable and meaningful activities [36]: Participants
intentionally completed three acts in a single day —a pleasurable
act done alone (e.g., reading), a pleasurable act done with others
(e.g., playing cards), and a meaningful or important act (e.g.,
creating a blood sugar log) —and wrote about them.
•Counting blessings [19,28,37,38]: Participants recalled three
events in the past week for which they were grateful and
recorded them in detail.
•Best possible self (social relationships) [29,32,33]: Participants
imagined and wrote about their best possible future interper-
sonal relationships and considered how to take steps toward
these relationships.
89J.C. Huffman et al. / General Hospital Psychiatry 36 (2014) 88–94
•Best possible self (accomplishments) [29,32,33,39]: Participants
imagined and wrote about their best possible future accom-
plishments and considered how to actualize this future.
•Forgiveness letter [30,41,42]: Participants selected a hurtful
event that occurred at least 2 years prior and wrote a letter to a
transgressor. The goal of the letter was to forgive the transgressor
for at least some aspect of the event. The letter was not to be sent.
•Behavioral commitment to values-based activities [31]: Partic-
ipants selected from an extensive list a principle that guided the
way they hoped to live (e.g., being physically fit, creating beauty)
and then made (and wrote about) a small step toward
reconnecting with this principle.
2.6. Procedures
To help characterize this population, enrolled participants com-
pleted baseline measures of depression (Quick Inventory of Depres-
sive Symptomatology—Self-Report [QIDS-SR] [43]), hopelessness
(Beck Hopelessness Scale [BHS] [44]) and optimism (Life Orientation
Test Revised [LOT-R] [45]). Baseline demographic and psychiatric
characteristics were obtained through chart review.
After completion of baseline measures, a study trainer met with
each participant for approximately 10 min to provide and review a
brief introductory packet that described the rationale for using PP
exercises in this setting. The trainer provided guidance on exercise
completion and described the structure of the study (daily exercise
completion, exercise ratings before and after completion, and daily
meetings to review exercises).
The participant was then assigned a PP exercise to be completed in
the next 24 h. The order of exercise assignment for each participant
was determined via random number generator. The trainer provided a
written exercise packet, and participants were told to complete the
activity and write about it in the packet.
At the front of each exercise packet was a preexercise rating sheet
with a pair of five-point Likert scales on which participants rated their
current levels of optimism and hopelessness (1 = not at all optimistic,
5=completely optimistic;1=not at all hopeless,5=very hopeless);
this was completed immediately prior to undertaking the exercise.
After exercise completion and before review with the study trainer,
participants completed a postexercise rating sheet at the end of the
packet. This sheet included the same scales for optimism and
hopelessness, along with scales regarding ease of completion (1 =
not at all easy,5=very easy) and overall utility (1=not at all helpful,
5=very helpful). Finally, the sheet contained open-ended questions
about how the exercise was useful and how it could be improved.
We chose hopelessness and optimism as key outcome measures
due to their established links to suicidality [10–12]. Hopelessness has
been associated with completed suicide, independent of severity of
mood symptoms [46,47], and therefore we chose hopelessness rather
than depression as our primary negative psychological measure,
especially given the transdiagnostic nature of our study cohort. We
chose to utilize this guided self-help approach (rather than creating a
psychotherapeutic intervention) to provide participants some sup-
port and guidance around exercise completion without developing an
intervention that would require extensive training, schooling or cost.
These factors are aligned with suggestions to create interventions
packaged for wide dissemination [48].
Each weekday, following exercise completion, the study trainer
met with the participant to discuss the prior exercise and to assign a
new exercise. These meetings typically lasted 15 min. On weekends,
participants were assigned two exercises, and both were reviewed the
following Monday. Participants’involvement was complete at
discharge or after completion of all nine exercises, whichever came
first. If an exercise was not completed, it was skipped. If participants
had clear clinical worsening or expressed suicidal thoughts, study staff
personally notified clinical teams and asked participants if they
wished to suspend participation.
2.7. Statistical analyses
Descriptive statistics (e.g., proportions, means) were calculated for
baseline characteristics and all individual exercise ratings. Indepen-
dent-samples ttests and χ
2
analyses were used to compare baseline
characteristics between participants who completed at least one
exercise and those who did not.
The primary goal of the study was to assess the feasibility and
acceptability of the exercises as a whole. To assess these metrics, we
calculated the proportion of enrolled participants who completed at
least one exercise, rates of overall exercise completion and mean
number of exercises completed per participant.
The secondary goal of our study was to rank the exercises to identify
the most promising interventions for future study. To compare metrics
among individual exercises, we recorded rates of exercise completion,
pre–postchanges in optimism and hopelessness, and ratingsof ease and
utility for each exercise. Given that there were multiple ratings of
exercise efficacy, we created, a priori, an efficacy score that combined
three ratings (pre-/postexercise change in hopelessness + pre–post
change in optimism + overall utility= efficacy score, with higherscores
indicative of greater efficacy) for completed exercises.
Given that not all subjects completed each exercise and to account
for the order of exercise administration and age (given data about
greater impact of PP exercises in older adults [20]), we compared
scores across the exercises on all outcome variables (change in
hopelessness, change in optimism, utility, ease, efficacy score) using
mixed-effects regression models with a patient-specific random effect
and fixed effects for age and exercise order. Using this model, adjusted
means for each outcome and exercise were calculated, and a global
test comparing all exercises was used to assess whether there was a
significant overall effect of exercise on outcomes independent of
exercise order or age.
Since participants occasionally did not complete an assigned
exercise, two approaches were used for analysis in the presence of
missing data. The first approach analyzed all available data assuming
that noncompletion of an exercise was uninformative about the
missing value and therefore used the efficacy score and ease as
described above. Because subjects with missing data may have failed
to complete an exercise because the exercise was difficult or had poor
perceived utility for the subject, a second approach created a modified
efficacy score that assumed no pre–post change and assigned the
lowest utility score when an exercise was not completed. After the
value was imputed, the same mixed model from above was used for
analysis. We similarly created a modified ease score using for
uncompleted exercises.
In addition, unadjusted (i.e., raw) pre–post changes in optimism
and hopelessness for each individual exercise were also investigated
using paired ttests. From this analysis, we calculated Cohen’sd(effect
size [49]) and the proportion of participants experiencing a pre–post
improvement of one or more points in hopelessness and optimism (if
not at the ceiling value on the preexercise rating). All quantitative
analyses were performed using Stata version 11.0 (StataCorp, College
Station, TX, USA); all tests were two-tailed.
Finally, open-ended qualitative responses were transcribed and
independently reviewed for themes by two staff members with
experience in qualitative research. A list of themes was generated and
combined, with disagreements resolved via discussion and review of
the primary material.
3. Results
Fig. 1 displays the recruitment and enrollment flow for the study.
Fifty-two (85.3%) of 61 consenting participants completed baseline
90 J.C. Huffman et al. / General Hospital Psychiatry 36 (2014) 88–94
measures and at least one exercise. Participants who completed at
least one exercise did not differ significantly on any baseline
characteristic from those who dropped out (all PN.10).
Table 1 displays the baseline characteristics of the 52 participants
who completed at least oneexercise. Major depressive disorder was the
primary diagnosis in 82.7% (n=43) patients. Participants had moderate
to severe depression at enrollment (QIDS-SR=16.49 [standard devia-
tion {S.D.}6.08]; cutoff for severe depression=16 [43]) and moderate to
severe hopelessness (BHS=11.37 [S.D. 5.80]; cutoff for high suicide
risk=9 [50]). In total, 189 (88.7%) of 213 assigned exercises were
completed, with participants completing a mean of 3.63 (S.D. 1.93) out
of 4.10 (S.D. 2.05) assigned exercises. The most common reported
reason for exercise noncompletion was insufficient time (n=9).
Table 2 displays participant ratings using adjusted means
calculated via mixed models and the proportion of participants who
completed each exercise. Individual exercise completion ranged from
77% (best possible self social relationships) to 100% (counting blessings).
Pre–post optimism and hopelessness improved across exercises, with
adjusted mean change between 0.24 and 1.12 for optimism and
between 0.30 and 1.20 for hopelessness.
With respect to individual exercise efficacy, there was an global
effect of exercise on modified efficacy scores, independent of age and
exercise order (χ
2
=19.39; df=8, P=.013), suggesting that exercise
content had a significant effect on efficacy. Regarding specific
exercises, the gratitude letter,counting blessings and personal strengths
exercises had the highest efficacy and modified efficacy scores,
whereas the forgiveness letter, best possible self (social relationships)
and values-based activities exercises had the lowest efficacy and
modified efficacy scores. Though pairwise comparisons were not
specified beforehand, exploratory analyses demonstrated that the
forgiveness letter had a significantly lower adjusted mean modified
efficacy score than the gratitude letter (adjusted mean difference=1.9
[95% confidence interval {CI} 0.8–3.0]; P=.001) and counting blessings
(adjusted mean difference=1.8 [95% CI 0.7–2.9]; P=.001) indepen-
dent of age and exercise order; these effects remain significant after
Bonferroni correction.
Modified ease scores (Table 2) suggested that four exercises (acts
of kindness; important, enjoyable and meaningful activities; gratitude
letter and personal strengths) ranked highest (adjusted mean modified
ease scores=3.50–3.55 among these four exercises), while forgiveness
letter (adjusted score=2.70) ranked lowest. There was no significant
effect of exercise on modified ease scores (χ
2
=11.64; df=8; P=.17).
Paired ttest analyses (Table 3) demonstrated that unadjusted pre–
post changes in optimism and hopelessness were statistically
significant (Pb.01 in most cases) for all exercises except the
forgiveness letter, which had a nonsignificant change in optimism
(P=.38). At least half of participants experienced a one-point
improvement in optimism and hopelessness for each individual
exercise except the forgiveness letter. Cohen’sdfor each exercise other
than the forgiveness letter ranged from 0.48 to 1.19 for both optimism
and hopelessness.
The majority of qualitative comments were positive (138 of 190
comments [72.3%]), with most comments focusing on the high ease of
Fig. 1. Diagram of recruitment and enrollment.
Table 1
Baseline demographic, diagnostic and self-report data
Characteristic Participants completing at
least 1 exercise (n=52)
Age (mean [S.D.]) 41.63 (14.68)
Women (n[%]) 33 (63.5%)
Caucasian (n[%]) 49 (94.2%)
Primary diagnosis (n[%])
Major depressive disorder 43 (82.7%)
Bipolar affective disorder 6 (11.5%)
Generalized anxiety disorder 1 (1.9%)
Eating disorder 1 (1.9%)
Posttraumatic stress disorder 1 (1.9%)
Insurance (n[%])
Uninsured/free care/connector health plan 7 (13.5%)
Medicare 12 (23.1%)
Medicaid 8 (15.4%)
Commercial 25 (48.1%)
Baseline scores (mean [S.D.])
LOT-R—Total 15.00 (6.24)
a
LOT-R—Optimism 7.55 (3.61)
a
LOT-R—Pessimism 7.45 (3.35)
a
BHS 11.37 (5.80)
b
QIDS-SR 16.49 (6.08)
c
Note. All scores based on n=52 unless otherwise noted.
a
n=51.
b
n=49.
c
n=45.
91J.C. Huffman et al. / General Hospital Psychiatry 36 (2014) 88–94
completion and generation of positive emotions related to the
exercises; negative comments focused on feeling too overwhelmed
to complete the writing portion or to carry out an activity requiring
interpersonal engagement. A substantial (n=9) number of negative
comments were related to the forgiveness letter, with themes related
to the aversive nature of recalling a past slight and having negative
feelings emerge when writing about the event.
4. Discussion
In a population of suicidal patients with high levels of hopelessness
and depression, on our primary aim, we found that administration of
PP exercises was feasible and well accepted. Nearly 90% of assigned
exercises were completed, a substantial finding considering that
participants were in crisis and receiving additional intensive treat-
ment. However, a small proportion of participants were unable to
participate in exercise completion, suggesting that a subset of this
population may be too severely ill or distressed to engage even in this
simple intervention.
Furthermore, regarding our secondary aim, PP exercises were
associated with improvement of hopelessness and optimism, central
therapeutic targets in this clinical cohort because of their independent
links to suicidal thoughts and behaviors [50,51], with moderate effect
sizes on these key outcomes. This suggests that PP interventions may
impact meaningful clinical outcomes, at least in the short term, in this
important high-risk cohort.
The number of previous exercises completed had the potential to
impact outcome assessments. We addressed this in two ways. First,
Table 2
Pre-/postexercise ratings (exercises listed in order of adjusted modified efficacy score)
Exercise
Completed/assigned
Change in
hopelessness
Change in
optimism
Utility Efficacy Modified
efficacy
Ease Modified
ease
Gratitude letter
21/23 (91.3%)
1.00 (0.20) 1.13 (0.22) 3.84 (0.22) 5.90 (0.44) 5.59 (0.47) 3.69 (0.24) 3.50 (0.28)
Counting blessings
25/25 (100%)
0.69 (0.19) 0.78 (0.20) 4.01 (0.20) 5.48 (0.40) 5.52 (0.46) 2.97 (0.23) 2.97 (0.27)
Personal strengths
16/18 (88.9%)
0.81 (0.23) 0.79 (0.25) 4.25 (0.25) 5.79 (0.50) 5.29 (0.53) 3.91 (0.28) 3.53 (0.32)
Acts of kindness
18/21 (85.7%)
1.20 (0.22) 0.83 (0.23) 3.59 (0.23) 5.65 (0.47) 5.02 (0.49) 3.91 (0.26) 3.55 (0.29)
Important, enjoyable and meaningful activities
25/26 (96.2%)
0.59 (0.19) 0.78 (0.20) 3.87 (0.20) 5.22 (0.41) 4.86 (0.45) 3.75 (0.23) 3.51 (0.27)
Best self (accomplishments)
22/25 (88.0%)
0.76 (0.20) 0.76 (0.21) 3.31 (0.21) 4.82 (0.42) 4.48 (0.45) 3.20 (0.24) 2.98 (0.27)
Behavioral commitment to values-based activities
18/19 (94.7%)
0.50 (0.23) 0.58 (0.25) 3.49 (0.24) 4.60 (0.49) 4.32 (0.54) 3.38 (0.28) 3.17 (0.32)
Best self (social relationships)
20/26 (76.9%)
0.84 (0.21) 0.53 (0.23) 3.79 (0.22) 5.17 (0.45) 4.12 (0.46) 3.93 (0.25) 3.19 (0.27)
Forgiveness letter
24/30 (80.0%)
0.30 (0.19) 0.24 (0.20) 3.77 (0.20) 4.34 (0.41) 3.69 (0.41) 3.14 (0.23) 2.70 (0.25)
Overall
189/213 (88.7%)
0.73 (0.09) 0.71 (0.10) 3.77 (0.10) 5.20 (0.21) 4.72 (0.22) 3.51 (0.12) 3.21 (0.12)
Note. Ratings are adjusted means and standard errors from random-effects model accounting for age and exercise order. All ratings based on five-point Likert scales. Modified
efficacy index and ease scores accounted for participants who were assigned but did not complete an exercise.
Table 3
Paired ttests assessing pre- and postexercise changes in hopelessness and optimism, separated by exercise
Exercise Mean improvement
a
95% CI tPvalue Cohen’sd(mean/S.D.) Improvement
Pre/post change in hopelessness
Counting blessings 0.64 0.27–1.01 3.53 0.002 0.71 12/21 (57.1%)
Best self
(accomplishments)
0.77 0.30–1.25 3.40 0.003 0.72 11/19 (57.9%)
Best self
(social relationships)
0.80 0.33–1.27 3.56 0.002 0.80 10/17 (58.8%)
Forgiveness letter 0.29 0.06–0.52 2.60 0.016 0.53 6/22 (27.3%)
Gratitude letter 1.05 0.60–1.49 4.93 b0.001 1.08 14/16 (87.5%)
Important, enjoyable and meaningful activities 0.68 0.35–1.01 4.24 b0.001 0.85 8/16 (50.0%)
Acts of kindness 1.17 0.65–1.69 4.74 b0.001 1.12 12/13 (92.3%)
Personal strengths 1.00 0.35–1.65 3.30 0.005 0.83 8/15 (53.3%)
Values-based activities 0.50 0.41–0.96 2.30 0.035 0.54 8/16 (50.0%)
Across all: 95/157 (60.5%)
Pre/post change in optimism
Counting blessings 0.80 0.28–1.32 3.18 0.004 0.64 11/21 (52.4%)
Best self (accomplishments) 0.73 0.29–1.16 3.46 0.002 0.74 13/19 (68.4%)
Best self (social relationships) 0.50 0.01–0.99 2.13 0.047 0.48 9/17 (52.9%)
Forgiveness letter 0.13 −0.16–0.41 0.90 0.38 0.18 5/22 (22.7%)
Gratitude letter 1.14 0.70–1.58 5.43 b0.001 1.19 15/16 (93.8%)
Important, enjoyable and meaningful activities 0.76 0.44–1.08 4.88 b0.001 0.98 15/18 (83.3%)
Acts of kindness 0.78 0.12–1.43 2.52 0.022 0.59 9/13 (69.2%)
Personal strengths 0.81 0.25–1.37 3.10 0.007 0.78 9/15 (60.0%)
Values-based activities 0.44 0.09–0.79 2.68 0.016 0.63 8/16 (50.0%)
Across all: 95/173 (54.9%)
a
On 1–5 Likert scale. Improvement scores based on participants with greater than or equal to one-point (positive) changes in hopelessness and optimism. Denominator based on
participants who completed exercises but did not have ceiling scores for prehopelessness or preoptimism.
92 J.C. Huffman et al. / General Hospital Psychiatry 36 (2014) 88–94
we accounted for potential improvements in preexercise baseline
ratings over time by measuring pre/post changes in optimism and
hopelessness (rather than just postexercise scores). In addition, the
random-effects model (estimating the effect of each exercise)
accounted for the ordering of the exercise, suggesting that differences
in outcomes for different PP exercises were not caused solely by
differences in the number of previous exercises completed.
Though the primary goal of this pilot project was to assess overall
feasibility and utility, we found meaningful variability in the degree to
which exercises were experienced as easy to complete and clinically
useful. Though the cumulative effect of the package of exercises may
have influenced outcomes, the global test of exercise efficacy found
that specific exercise type, independent of exercise order, was
significantly associated with self-reported efficacy.
Straightforward exercises that did not require substantial
introspection appeared to perform best. For example, exercises on
gratitude—the gratitude letter and counting blessings—had high
utility scores and were associated with substantial improvements
in optimism. This finding is consistent with prior work that has
linked gratitude to lower levels of depression, hopelessness and
suicidal thoughts/attempts [13,52–56], though is in some contrast
to a study that found a gratitude letter to reduce depressive
symptoms in mildly, but not more severely, depressed participants
[57]. The exercise focusing on a personal strength was also
associated with substantial improvements in this population and
was perceived as easy to complete, consistent with prior work
finding this exercise to be associated with prolonged improvement
in depressive symptoms [19].
In contrast, the forgiveness letter exercise performed most poorly.
Qualitatively, many patients experienced a resurgence of anger or
sadness when recalling a past slight and found it difficult to move past
these feelings in the midst of crisis [58]. In addition, exercises focused
on life purpose and optimism had lower efficacy ratings than most
other exercises (though with Cohen’sdN.4 on optimism and
hopelessness in all cases). Asking patients to take on ‘big’issues
such as life purpose or imagining an optimal future might have been
more difficult for some patients at this stage. Alterations in delivery or
timing (e.g., after some symptom recovery) might render these
exercises more useful in this population.
This preliminary study had several limitations. Our sample was a
largely Caucasian population on a single psychiatric unit. There were
no sham or control exercises. Participants concurrently received
psychological and psychopharmacological interventions during ad-
mission, although ratings were completed in the narrow window
between exercise initiation and completion, minimizing potential
“noise”from other interventions. However, this meant that we were
not able to assess the overall/cumulative effect of the exercises as a
package. Additionally, patients may have felt compelled to rate higher
improvements in symptoms after exercise completion than before,
either to justify their participation or to please the researchers,
although as noted there were substantial differences in improvement
among various exercises. Finally, in this initial study that was
primarily aimed at feasibility (and short-term effects on optimism
and hopelessness), we did not serially measure suicidality via a
validated scale or obtain postdischarge outcomes, and thus, we cannot
make claims about the ultimate impact of these exercises on suicide
risk postdischarge.
In conclusion, PP exercises administered to suicidal inpatients
were well accepted, were completed at high rates and appeared to
be associated with self-rated improvements in clinically relevant
short-term outcomes. Specific exercises, especially those related to
gratitude and using personal strengths, appeared to be most
effective in this context. Additional work is needed to confirm
this pilot work, refine suboptimal exercises, test the exercises in a
larger cohort with a control group and assess their impact on
longer-term outcomes.
References
[1] Appleby L, Dennehy JA, Thomas CS, Faragher EB, Lewis G. Aftercare and clinical
characteristics of people with mental illness who commit suicide: a case–control
study. Lancet 1999;353:1397–400.
[2] Appleby L, Shaw J, Amos T, McDonnell R, Harris C, McCann K, et al. Suicide within
12 months of contact with mental health services: national clinical survey. BMJ
1999;318:1235–9.
[3] Goldacre M, Seagroatt V, Hawton K. Suicide after discharge from psychiatric
inpatient care. Lancet 1993;342:283–6.
[4] Qin P, Nordentoft M. Suicide risk in relation to psychiatric hospitalization:
evidence based on longitudinal registers. Arch Gen Psychiatry 2005;62:427–32.
[5] Stanley B, Brown G, Brent DA, Wells K, Poling K, Curry J, et al. Cognitive–behavioral
therapy for suicide preven tion (CBT-SP): treatment model, feasibility, and
acceptability. J Am Acad Child Adolesc Psychiatry 2009;48:1005–13.
[6] Tarrier N, Taylor K, Gooding P. Cognitive–behavioral interventions to reduce
suicide behavior: a systematic review and meta-analysis. Behav Modif 2008;32:
77–108.
[7] Stanley B, Brodsky B, Nelson JD, Dulit R. Brief dialectical behavior therapy (DBT-B)
for suicidal behavior and non-suicidal self injury. Arch Suicide Res 2007;11:
337–41.
[8] Bloom JM, Woodward EN, Susmaras T, Pantalone DW. Use of dialectical behavior
therapy in inpatient treatment of borderline personality disorder: a systematic
review. Psychiatr Serv 2012;63:881–8.
[9] DuBois CM, Beach SR, Kashdan TB, Nyer MB, Park ER, Celano CM, et al. Positive
psychological attributes and cardiac outcomes: associations, mechanisms, and
interventions. Psychosomatics 2012;53:303–18.
[10] Hirsch JK, Conner KR, Duberstein PR. Optimism and suicide ideation among young
adult college students. Arch Suicide Res 2007;11:177–85.
[11] Hirsch JK, Duberstein PR, Conner KR, Heisel MJ, Beckman A, Franus N, et al. Future
orientation and suicide ideation and attempts in depressed adults ages 50 and
over. Am J Geriatr Psychiatry 2006;14:752–7.
[12] Hirsch JK, Duberstein PR, Conner KR, Heisel MJ, Beckman A, Franus N, et al. Future
orientation moderates the relationship between functional status and suicide
ideation in depressed adults. Depress Anxiety 2007;24:196–201.
[13] Li D, Zhang W, Li X, Li N, Ye B. Gratitude and suicidal ideation and suicide attempts
among Chinese adolescents: direct, mediated, and moderated effects. J Adolesc
2012;35:55–66.
[14] Yamokoski CA, Scheel KR, Rogers JR. The role of affect in suicidal thoughts and
behaviors. Suicide Life Threat Behav 2011;41:160–70.
[15] Heisel MJ, Flett GL. Purpose in life, satisfation with life, and suicide ideation in a
clinical sample. J Psychopathol Behav Assess 2004;26:127–35.
[16] Seidlitz L, Conwell Y, Duberstein P, Cox C, Denning D. Emotion traits in older
suicide attempters and non-attempters. J Affect Disord 2001;66:123–31.
[17] Joiner TE, Pettit JW, Perez M, Burns AB, Gencoz T, Gencoz F. Can positive emotion
influence problem-solving attitudes among suicidal adults? Prof Psychol Res Pract
2001;32:507–12.
[18] O'Connor RC, Fraser L, Whyte MC, Machale S, Masterton G. A comparison of
specific positive future expectancies and global hopelessness as predictors of
suicidal ideation in a prospective study of repeat self-harmers. J Affect Disord
2008;110:207–14.
[19] Seligman ME, Steen TA, Park N, Peterson C. Positive psychology progress:
empirical validation of interventions. Am Psychol 2005;60:410–21.
[20] Sin NL, Lyubomirsky S. Enhancing well-being and alleviating depressive
symptoms with positive psychology interventions: a practice-friendly meta-
analysis. J Clin Psychol 2009;65:467–87.
[21] Bolier L, Haverman M, Westerhof GJ, Riper H, Smit F, Bohlmeijer E. Positive
psychology interventions: a meta-analysis of randomized controlled studies. BMC
Public Health 2013;13:119.
[22] Lyubomirsky S, Layous K. How do simple positive activities increase well-being?
Curr Dir Psychol Sci 2013;22:57–62.
[23] Grossman P, Tiefenthaler-Gilmer U, Raysz A, Kesper U. Mindfulness training as an
intervention for fibromyalgia: evidence of postintervention and 3-year follow-up
benefits in well-being. Psychother Psychosom 2007;76:226–33.
[24] Seligman ME, Rashid T, Parks AC. Positive psychotherapy. Am Psychol 2006;61:
774–88.
[25] Lapierre S, Dube M, Bouffard L, Alain M. Addressing suicidal ideations through the
realization of meaningful personal goals. Crisis 2007;28:16–25.
[26] Callahan CM, Unverzagt FW, Hui SL, Perkins AJ, Hendrie HC. Six-item screener to
identify cognitive impairment among potential subjects for clinical research. Med
Care 2002;40:771–81.
[27] Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The
Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and
validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
J Clin Psychiatry 1998;59(Suppl 20):22–33.
[28] Emmons RA, McCullough ME. Counting blessings versus burdens: an experimental
investigation of gratitude and subjective well-being in daily life. J Pers Soc Psychol
2003;84:377–89.
[29] Sheldon KM, Lyubomirsky S. How to increase and sustain positive emotion: the
effects of expressing gratitude and visualizing best possible selves. J Positive
Psychology 2006;1:73–82.
[30] McCullough ME, Root LM, Cohen AD. Writing about the benefits of an
interpersonal transgression facilitates forgiveness. J Consult Clin Psychol
2006;74:887–97.
[31] Kashdan TB, McKnight PE. Origins of purpose in life: refining our understanding of
a life well lived. Psychological Topics 2009;18:303–16.
93J.C. Huffman et al. / General Hospital Psychiatry 36 (2014) 88–94
[32] Boehm JK, Lyubomirsky S, Sheldon KM. A longitudinal experimental study
comparing the effectiveness of happiness-enhancing strategies in Anglo Ameri-
cans and Asian Americans. Cogn Emot 2011;25:1263–72.
[33] Lyubomirsky S, Dickerhoof R, Boehm JK, Sheldon KM. Becoming happier takes
both a will and a proper way: an experimental longitudinal intervention to boost
well-being. Emotion 2011;11:391–402.
[34] Dunn EW, Aknin LB, Norton MI. Spending money on others promotes happiness.
Science 2008;319:1687–8.
[35] Sheldon K, Boehm JK, Lyubomirsky S. Variety is the spice of happiness: the hedonic
adaptation prevention (HAP) model. In: Boniwell I, David S, editors. Oxford
handbook of happiness. Oxford: Oxford University Press; 2012. p. 119–33.
[36] Peterson C, Park N, Seligman ME. The full life versus the empty life. J Happiness
Stud 2005;6:25–41.
[37] Froh JJ, Sefick WJ, Emmons RA. Counting blessings in early adolescents: an
experimental study of gratitude and subjective well-being. J Sch Psychol 2008;46:
213–33.
[38] Lyubomirsky S, Sheldon KM, Schkade D. Pursuing happiness: the architecture of
sustainable change. Rev Gen Psychology 2005;9:111–31.
[39] King LA. The health benefits of writing about life goals. Pers Soc Psychol Bull
2001;27:10–7.
[40] Mitchell J, Stanimirovic R, Klein B, Vella-Brodrick DA. A randomised controlled
trial of a self-guided internet intervention promoting well-being. Comput Hum
Behav 2009;25:749–60.
[41] van Oyen Witvliet C, Ludwig TE, Vander Laan KL. Granting forgiveness or
harboring grudges: implications for emotion, physiology, and health. Psychol Sci
2001;12:117–23.
[42] Peterson C. A primer in positive psychology. USA: Oxford University Press; 2006 .
[43] Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, et al. The 16-
Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating
(QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with
chronic major depression. Biol Psychiatry 2003;54:573–83.
[44] Beck AT, Weissman A, Lester D, Trexler L. The measurement of pessimism: the
hopelessness scale. J Consult Clin Psychol 1974;42:861–5.
[45] Scheier MF, Carver CS, Bridges MW. Distinguishing optimism from neuroticism
(and trait anxiety, self-mastery, and self-esteem): a reevaluation of the Life
Orientation Test. J Pers Soc Psychol 1994;67:1063–78.
[46] Beck AT, Steer RA, Kovacs M, Garrison B. Hopelessness and eventual suicide: a 10-
year prospective study of patients hospitalized with suicidal ideation. Am J
Psychiatry 1985;142:559–63.
[47] Hill RD, Gallagher D, Thompson LW, Ishida T. Hopelessness as a measure of
suicidal intent in the depressed elderly. Psychol Aging 1988;3:230–2.
[48] Kazdin AE, Blase SL. Rebooting psychotherapy research and practice to reduce
the burden of mental illness. Perspectives on psychological science 2011;6:
21–37.
[49] Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale (NJ):
L. Erlbaum Associates; 1988.
[50] Beck AT, Brown G, Berchick RJ, Stewart BL, Steer RA. Relationship between
hopelessness and ultimate suicide: a replication with psychiatric outpatients. Am J
Psychiatry 1990;147:190–5.
[51] Rasmussen KA, Wingate LR. The role of optimism in the interpersonal-
psychological theory of suicidal behavior. Suicide Life Threat Behav 2011;41:
137–48.
[52] Rosmarin DH, Krumrei EJ, Pargament KI. Are gratitude and spirituality protective
factors against psychopathology?. Int J Existential, Psychol Psychother 2010;3.
[53] Lyubomirsky S, Tkach C, DiMatteo MR. What are the differences between
happiness and self-esteem? Soc Indic Res 2006;78:363–404.
[54] Fredrickson BL, Tugade MM, Waugh CE, Larkin GR. What good are positive
emotions in crises? A prospective study of resilience and emotions following the
terrorist attacks on the United States on September 11th, 2001. J Pers Soc Psychol
2003;84:365–76.
[55] Wood AM, Maltby J, Gillett R, Linley PA, Joseph S. The role of gratitude in the
development of social support, stress, and depression: two longitudinal studies.
J Res Pers 2008;42:854–71.
[56] Ng MY, Wong WS. The differential effects of gratitude and sleep on psychological
distress in patients with chronic pain. J Health Psychol 2013;18:263–71.
[57] Sin NL, Della Porta MD, Lyubomirsky S. Tailoring positive psychology interven-
tions to treat depressed individuals. In: Donaldson SI, Csikszentmihalyi M,
Nakamura J, editors. Applied positive psychology: improving everday life, health,
schools, work, and society. New York: Routledge; 2011. p. 79–96.
[58] Worthington EL, Kurusu TA, Collins WB, Berry JW, Ripley JS, Baier SN. Forgiving
usually takes time: a lesson learned by studying interventions to promote
forgiveness. J Psychol Theol 2000;28:3–20.
94 J.C. Huffman et al. / General Hospital Psychiatry 36 (2014) 88–94
