Please do not remove this page

Effectiveness of Mindfulness Intervention in

Reducing Stress and Burnout for Mental Health
Professionals in Singapore
Suyi, Yang; Meredith, Pamela; Khan, Asaduzzaman
https://research.usc.edu.au/esploro/outputs/99568107302621/filesAndLinks?institution=61USC_INST&index=null

Suyi, Meredith, P., & Khan, A. (2017). Effectiveness of Mindfulness Intervention in Reducing Stress and
Burnout for Mental Health Professionals in Singapore. Explore, 13(5), 319–326.
https://doi.org/10.1016/j.explore.2017.06.001
Document Type: Accepted Version

Link to Published Version: https://doi.org/10.1016/j.explore.2017.06.001

USC Research Bank: https://research.usc.edu.au

[email protected]
CC BY-NC-ND V4.0
Copyright © 2017. This manuscript version is made available under the CC-BY-NC-ND 4.0 license
http://creativecommons.org/licenses/by-nc-nd/4.0/
Downloaded On 2023/02/16 01:06:01 +1000

Please do not remove this page

Author’s Accepted Manuscript

EFFECTIVENESS OF MINDFULNESS
INTERVENTION IN REDUCING STRESS AND
BURNOUT FOR MENTAL HEALTH
PROFESSIONALS IN SINGAPORE

Yang Suyi, Pamela Meredith, Asaduzzaman Khan

www.elsevier.com/locate/jsch

PII: S1550-8307(17)30235-5
DOI: http://dx.doi.org/10.1016/j.explore.2017.06.001
Reference: JSCH2211
To appear in: Explore: The Journal of Science and Healing
Cite this article as: Yang Suyi, Pamela Meredith and Asaduzzaman Khan,
EFFECTIVENESS OF MINDFULNESS INTERVENTION IN REDUCING
STRESS AND BURNOUT FOR MENTAL HEALTH PROFESSIONALS IN
SINGAPORE, Explore: The Journal of Science and Healing,
http://dx.doi.org/10.1016/j.explore.2017.06.001
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 Effectiveness of mindfulness training in reducing stress and burnout for mental health

professionals in Singapore

Dr Yang Suyi (Corresponding author)

Senior Occupational Therapist, Institute of Mental Health
10 Buangkok Green Medical Park
Singapore 539747
Phone: +65 6389 2912
Fax: +65 63892907
Email: [email protected]

Dr Pamela Meredith
Senior Lecturer
School of Health and Rehabilitation Sciences, The University of Queensland
St Lucia, Queensland, 4072, Australia

Dr Asaduzzaman Khan
Senior Lecturer
School of Health and Rehabilitation Sciences, The University of Queensland
St Lucia, Queensland, 4072, Australia

1
Effectiveness of mindfulness intervention in reducing stress and burnout for mental health
professionals in Singapore

Abstract

Stress and burnout have been shown to be a concern among mental health professionals in
several countries including Singapore, and can affect quality of care and staff turnover. The aim
of this study was to examine the effectiveness of a mindfulness program in increasing
mindfulness and compassion, and reducing stress and burnout, among mental health
professionals in Singapore. The study utilised data from a prospective pre-post study design with
follow-up. A total of 37 mental health professionals participated in the program, which was
conducted in three cohorts over nine months. The program consisted of six two-hour sessions
offered once a week over six weeks, and used a range of mindfulness techniques to teach
participants to cultivate compassionate and non-judgemental attitudes towards their inner
experiences. Data were collected at three stages: pre- and post-intervention, and three months
follow-up. Assessments considered mindfulness (Five Facets Mindfulness Questionnaire),
compassion (Self-Compassion Scale-SF, Compassion Scale), stress (Perceived Stress Scale-10),
and burnout (Oldenburg Burnout Inventory). Participants demonstrated significant improvement
in four of the five mindfulness facets (observe, describe, non-judge, non-react) and in
compassion levels, and a significant reduction in stress, following intervention. The gains in
mindfulness and self-compassion scores were maintained at three months follow-up. No change
was observed for burnout variables. Results suggest that mindfulness training was effective in
reducing stress and improving mindfulness and compassion, but not decreasing burnout, for this
group of mental health professionals in Singapore. Future experimental research with larger
samples is warranted to validate the findings of the present study.

Keywords: Burnout, mental health professionals, mindfulness, Singapore, stress

2
 Introduction

There are reports of relatively high levels of burnout among mental health professionals in the
United States (Acker, 2012), Australia (Lloyd & King, 2004), and Singapore (Yang, Meredith, &
Khan, 2015). This is of particular concern as burnout has been associated with high staff
turnover (Chiang & Chang, 2012) and decreased patient care (Shanafelt, Bradley, Wipf, & Back,
2002) in samples of healthcare professionals. Although a number of studies have explored the
prevalence of burnout in mental health professionals, there has been limited research into the
outcomes of programs designed to reduce stress and burnout among mental health professionals.

Mindfulness training has been recognised to reduce stress and burnout, and promote positive
attitudes among healthcare professionals more generally (Escuriex & Labbé, 2011; Irving,
Dobkin, & Park, 2009). During mindfulness training, a range of mindfulness practices are used
to teach participants to cultivate an observant, accepting, and compassionate attitude towards
their own internal experiences including body sensations, emotions, and thoughts. The training
sessions are usually conducted in groups and include teachings, group discussions, and practical
exercises. One of the most widely used mindfulness programs is the Mindfulness-Based Stress
Reduction (MBSR) course developed by Jon Kabat-Zinn (Cullen, 2011; Kabat-Zinn, 1991). This
is an eight-week program with weekly group sessions of two and a half hours and daily 45-
minute homework sessions.

In their review of twenty studies that examined the effectiveness of mindfulness-based

interventions in improving the psychosocial health of healthcare providers, Escuriex and Labbé
(2011) concluded that the efficacy of mindfulness-based interventions was tentatively supported.
More studies examining the effect of mindfulness-based intervention with healthcare
professionals have been conducted in recent years with promising results (Amutio et al., 2014;
Asuero et al., 2014; Gauthier et al., 2015; Goodman & Schorling, 2012). However, several
limitations were reported in the reviews and papers, including small sample sizes, lack of
consideration to therapist adherence to treatment protocol, lack of adjunctive physiological
measures, and lack of follow-up data to determine whether gains made in intervention were
sustained. For example, Pipe et al. (2009) conducted a randomised study in which 33 nurses were
assigned to either a four week mindfulness course or a leadership course (control). They reported

3
that participants in the mindfulness course showed significantly more improvement on the
Positive Symptom Distress Index and Global Severity Index, which assess symptom intensity
and overall psychological distress respectively. Despite the use of a gold standard study design,
the study had several limitations including small sample size and lack of follow-up measure. In
another larger study by Goodman and Schorling (2012), 93 healthcare providers in the United
States participated in an eight-week MBSR intervention utilizing a pre-post study design. Their
results indicated that the improvement in mental health, emotional exhaustion, depersonalization,
and personal accomplishment scores among the healthcare providers after the intervention were
all statistically significant. Despite a large sample size, no follow-up data was gathered; thus, it is
not known if these gains were maintained.

Several studies with health professionals have considered follow-up after a mindfulness training
program (Asuero et al., 2014; Cohen-Katz et al., 2005). For example, Asuero et al. (2014)
examined the effectiveness of MBSR with 29 healthcare professionals, and found that the
significant reduction in self-reported psychological distress post-intervention was maintained
three months later. Other studies have reported inconsistent results; for example, Cohen-Katz et
al. (2005) investigated the effectiveness of mindfulness training in reducing burnout in 27 nurses.
Using the Maslach Burnout Inventory, they reported significant improvement in emotional
exhaustion post-intervention, which was maintained after three months; however, no significant
change was noted for personal accomplishment and depersonalization either post-intervention or
three months later. No research regarding the outcomes of mindfulness-based programs designed
to reduce stress and burnout specific to mental health professionals was identified.

In addition to the limitations discussed above, all studies described earlier were conducted with
western populations and the authors cautioned against the generalisation of the results to
different population groups. With improvements in technology and communication, cultural
“levelling” is occurring, and the values of the Western world are being assimilated in Singapore
(Henslin, 1999; Zhang, 2002). Nevertheless, Singapore’s healthcare professionals are
predominantly of Asian origins, and might respond differently to mindfulness training compared
to their Western counterparts. As the MBSR program was developed in the West, the course
content and structure might be more appealing to a western population than an Asian population.
Despite a rigorous search, no published research on the efficacy of a mindfulness program for

4
healthcare professionals in Singapore was located. In the only published study pertaining to
mindfulness in Singapore, Feng, Cao, Zhang, Wee, and Kua (2011) examined the outcomes of a
mindfulness program for 11 individuals with mild anxiety. They reported improvements in
overall mental well-being and anxiety symptoms after the eight-week program. As only one
mindfulness study conducted in Singapore had been found, there is a lack of evidence to suggest
that mindfulness intervention will be effective in reducing stress and burnout among mental
health professionals in Singapore.

The Present Study

Although evidence from empirical studies in Western cultures tentatively supports the efficacy of
mindfulness programs in promoting positive attitudes and reducing stress and burnout in
healthcare professionals (Asuero et al., 2014; Goodman & Schorling, 2012), there was no
literature examining its effectiveness with mental health professionals generally, or with
health/mental health professionals in Singapore. Further, findings regarding the sustainability of
gains from mindfulness training have been somewhat inconsistent. The aim of this study was to
examine the effectiveness of a mindfulness program in promoting mindfulness and compassion,
and reducing stress and burnout, among mental health professionals in Singapore. Study
hypotheses were that, for a sample of mental health professionals in Singapore:

1. Levels of mindfulness, self-compassion, and compassion for others would improve after a six-
week mindfulness program, and this improvement would be maintained three months after the
mindfulness program.

2. Levels of stress and burnout would decrease after a six-week mindfulness program, and this
decrease would be maintained three months after the mindfulness program.

Methods

Ethical approval for this study was granted by the National Healthcare Group Domain Specific
Review Board in Singapore, and The University of Queensland Behavioural and Social Science
Ethical Review Committee in Australia.

5
Participants

The study sample was drawn from healthcare professionals employed at the Institute of Mental
Health (IMH), Singapore. Emails were sent to all mental health professionals working in one
mental health setting in Singapore to invite them to participate in the study. Inclusion criteria for
the study was that participants must be: (1) psychiatrists, doctors, allied health professionals,
case managers, pharmacists, researchers, or nurses working at the Institute of Mental Health,
Singapore, (2) above the age of 18 years, and (3) able to attend at least four out of the six
sessions of mindfulness training. Participants were excluded if they reported a previous diagnosis
of psychosis, severe depression, borderline personality disorder, high suicide risk, and/or
alcohol/ drug misuse or dependence at any point in their life. The Principal Investigator (PI)
performed the eligibility check after participants indicated their interest in joining the study.

A total of 44 participants were initially recruited; however, seven withdrew from the study. Four
of these seven participants reported a change in work schedule which prevented them from
attending the training, while three took annual or medical leave and were unable to attend the
minimum four training sessions. A total of 37 participants completed the mindfulness training.
The majority were female, Chinese, and working full-time (see Table 1). Most of the participants
reported that they had no regular meditation practice, except for one participant who reported a
15 minutes breath meditation daily. All 37 participants also completed the 3 month follow-up
assessment.

This study utilised a pre-post study design with follow-up. Participants attended six sessions of
mindfulness training, offered once a week. The participants were evaluated at three points using
the same assessments: (Time 1) pre-intervention, (Time 2) post-intervention, and (Time 3) three
months after the intervention. Participants were given a small monetary reimbursement after
each study visit.

<Insert Table 1 about here>

6
Measures

Demographics

Demographics included gender, age, race, education, and years of work experience and were
collected using multiple choice options. Participants were also asked whether they had engaged
in regular meditation and/or yoga practice in the last three months. This information was
collected during the first study visit conducted by the PI of this study.

Five Facets Mindfulness Questionnaire, FFMQ (Baer et al., 2008). The FFMQ is a 39-item self-
report measure of mindfulness. It consists of five subscales: observe, describe, act with
awareness, non-judge, and non-react. An example item is: “I notice how foods and drinks affect
my thoughts, bodily sensations, and emotions” (observe). In the present study, confirmatory
factor analysis indicated that the data demonstrated good fit with a second-order hierarchical
model in which the five factors are subsets of an overall mindfulness factor (χ2 (85) = 132.99, χ2 /
df = 1.56, CFI = .97, TLI = .96, RMSEA = 0.05, SRMR = 0.07). The Cronbach’s alphas for the
FFMQ in this present study, measured at pre-intervention, post-intervention and three months
later, ranged between 0.91-0.92.

Self-Compassion Scale-Short Form, SCS-SF (Neff, 2003; Raes, Pommier, Neff, & Van Gucht,
2011). The SCM-SF is a 12-item self-report questionnaire, measuring self-compassion. This is
defined as being open and kind to one-self and taking a non-judgmental attitude towards difficult
feelings or thoughts. A sample item is: “I try to be understanding and patient towards those
aspects of my personality I don’t like”. The SCS-SF total score also demonstrated adequate
internal consistency of 0.87 (Neff & Germer, 2013) and a near-perfect correlation (r ≥ .97) with
the long form SCS (Raes et al., 2011). The Cronbach’s alphas of SCS-SF, measured at pre-
intervention, post-intervention and three months later, ranged between 0.81-0.86.

Compassion Scale, CS (Pommier, 2010). The CS is a 24-item self-report questionnaire,

measuring compassion for others. It consists of both positively and negatively worded items. A
sample items is: “I like to be there for others in times of difficulty”. The scale demonstrates good
convergent validity with moderate, significant associations with similar constructs such as

7
compassionate love, social connectedness, and empathy (Pommier, 2010). The Cronbach’s alpha
of CS for the present study, measured at pre-intervention, post-intervention and three months
later, ranged from 0.89-0.91.

Perceived Stress Scale, PSS-10 (Cohen, Kamarck, & Mermelstein, 1983; Lee, 2012). The PSS is
a 10-item questionnaire measuring perceived stress as experienced by an individual in daily life.
An example item is: “In the last month, how often have you felt difficulties were piling up so
high that you could not overcome them?” Mitchell, Crane, and Kim (2008) reported that the
PSS-10 demonstrated good psychometric properties with adequate convergent and concurrent
validity. The Cronbach’s alpha of PSS-10 for the present study, measured at pre-intervention,
post-intervention and three months later, ranged from 0.81-0.91.

Oldenburg Burnout Inventory, OLBI (Demerouti & Halbesleben, 2005; Vardakou, Bakker,
Demerouti, & Kantas, 2003). The OLBI is a 16-item self-report questionnaire measuring burnout
and consisting of two subscales: exhaustion and disengagement from work. An example items
include “After working, I have enough energy for my leisure activities” (exhaustion). Good
psychometric properties have been demonstrated for the OLBI in a sample of working adults
(Halbesleben & Demerouti, 2005). The Cronbach’s alphas for the exhaustion and disengagement
subscales in the present study, measured at pre-intervention, post-intervention and three months
later, ranged from 0.63-0.89 and 0.79-0.88 respectively.

Intervention

The MBSR program developed by Jon Kabat-Zinn (Kabat-Zinn, 1991) was used as a guide in the
development of the mindfulness program employed in this study. However, the length of the
program was shortened as a longer program is associated with a higher attrition rate (Irving et al.,
2009). This was of particular concern for mental health professionals in Singapore with a high
caseloads and tight schedules. To better suit the needs of busy staff members, one author (YS)
designed a shorter program. As participants were all mental health professionals, it was assumed
that topics such as stress physiology and emotional reactivity had been taught during
participants’ undergraduate or post-graduate training, and also in their on-going training as
mental health professionals. Thus, discussion of these topics was abbreviated. Prior to this study,
the author (YS) piloted two mindfulness programs, which lasted four-weeks and six-weeks

8
respectively. The participants in the four-week program suggested that the program was too short
while the participants in the six-week program indicated that the duration was just right. Thus,
for this study, a six-week mindfulness program was used.

The mindfulness training was provided once a week, for two hours per session, over six weeks.
The topics were: 1) Welcome and introduction to practice, 2) Perception and engaging with
practice, 3) Awareness of being stuck in one’s life and how to get unstuck, 4) Reacting and
responding to stress, exploring perceptions and thoughts, 5) Mindful communication in stressful
situations, and 6) Cultivating kindness towards self and others. During the training, meditation
practices such as body scan, breath meditation, kindness meditation, mindfulness of
feelings/thoughts meditation, and mindful movement/yoga were taught. There was also a
discussion period where participants were invited to share stories about their mindfulness
practice, or thoughts on the topics presented during the training. In addition to attending the
group, participants were requested to practice 30 minutes of formal meditation daily. Recordings
of guided meditation were given to participants for use during this homework practice.
Participants were also encouraged to apply mindfulness in their daily life.

Three mindfulness training programs were conducted one after the other, each with participants.
Each group had two instructors: the main instructor (YS) and a co-facilitator. The main instructor
is a certified MBSR instructor who started meditation practice 16 years ago. She has three years
of experience in teaching mindfulness to both mental health professionals and people with a
mental illness. The co-facilitator role was filled by two people, both certified Mindfulness-Based
Cognitive Therapy instructors. The first co-facilitator helped in the first mindfulness training
program, while the second co-facilitator helped in the second and third programs.

Data analysis

The data were analysed using SPSS (Version 17.0). Bivariate analyses were used to check for
any between group differences at baseline and after intervention. No significant differences were
noted. Repeated measures ANOVA was used to determine whether scores of mindfulness
(FFMQ), self-compassion (SCS-S), compassion for others (CS), stress (PSS-10), and burnout
(OLBI) changed post-intervention and whether these changes were maintained three months
after the intervention. Shapiro-Wilk tests were performed to assess the normality of residuals.

9
The residuals of SCS-S and PSS-10 were normally distributed but residuals of FFMQ, CS and
OLBI were not. As the normality assumption was violated in a small sample of n=37, non-
parametric tests (Friedman’s ANOVA) were used to examine the changes in scores in FFMQ,
CS and OLBI. Statistical significance for all analysis was set at p<.05, unless otherwise stated.

Repeated measures within groups ANOVA, including Mauchly’s sphericity tests and Bonferroni
post-hoc tests, were used to examine changes in PSS-10 and SCS-S. Mauchly’s tests for both
measures were non-significant, indicating that the assumption of sphericity was not violated.
Effect sizes for pre- and post-intervention were calculated using Cohen’s d. A value of 0.2 was
considered small, 0.5 was considered medium, and 0.8 was considered large.

Friedman’s ANOVA was used to examine changes in FFMQ, CS, and OLBI, with Wilcoxon
signed-rank tests as follow-up post-hoc procedures. A Bonferroni adjustment with a significant
level of p<.02 was adopted for the post-hoc procedures. Effect sizes for pre- and post-
intervention were calculated using the formula, ES (Effect size) = Z/√N (Z=Z-score from
Wilcoxon signed-rank test). A value of 0.1 was considered small, 0.3 was considered medium,
and 0.5 was considered large (Field, 2005).

Results

Attendance

A total of 37 (84%) participants completed the program: 20 (45%) attended all six sessions, 14
(32%) attended five sessions, and 3 (6%) attended four sessions. All 37 participants also attended
the 3 months follow-up assessment.

Changes in mindfulness

Friedman’s ANOVA revealed that there was a significant change across the three time points in
four of the five mindfulness facets: observe (χ2(2)=21.5, p<.001), describe (χ2(2)=12.01, p=002),
non-judge (χ2(2)=11.51, p=.003), and non-react (χ2(2)=32.08, p<.001). There was no significant
change with act with awareness (χ2(2)=.10, p=.95) score. Results of post-hoc analyses, using
Wilcoxon tests, for the five facets are reported below:

10
1) Observe: there was a significant increase in observe scores from pre-intervention (Median=23)
to post-intervention (Median=28 p<.001) with a medium effect size (ES=.49). There was also
significant change from pre-intervention to three months follow-up (Median=28, p<.001),
indicating that the improvement in observe scores was maintained at three months follow-up.

2) Describe: there was a significant increase in describe scores from pre-intervention

(Median=23) to post-intervention (Median=28, p=.01) with a small effect size (ES=.29). There
was also significant change from pre-intervention to three months follow-up (Median=28,
p=.01), indicating that the improvement in describe scores was maintained at three months
follow-up.

3) Non-judge: there was a significant increase in non-judge scores from pre-intervention

(Median=23) to post-intervention (Median=26, p=.02) with a small effect size (ES=.27). There
was also significant change from pre-intervention to three months follow-up (Median=26,
p=.01), indicating that the improvement in non-judge scores was maintained at three months
follow-up.

4) Non-react: there was a significant increase in observe scores from pre-intervention

(Median=20) to post-intervention (Median=23, p<.001) with a medium effect size (ES=.48).
There was also significant change from pre-intervention to three months follow-up (Median=23,
p<.001), indicating that the improvement in non-react scores was maintained at three months
follow-up.

5) Act with awareness: There was no significant change in these scores (χ2(2)=.10, p=.95) from
pre-intervention (Median=26) to post-intervention (Median=25), or from pre-intervention to
three months follow-up (Median=26).

6) Total mindfulness score: there was a significant increase in total mindfulness scores from pre-
intervention (Median=121) to post-intervention (Median=131, p<.001) with a large effect size
(ES=.67); there was also significant change from pre-intervention to three months follow-up
(Median=132, p<.001), indicating that the improvement in non-react scores was maintained at
three months follow-up (see Table 2).

11
<Insert Table 2 about here>

Changes in compassion

There was a significant change in self-compassion scores (F(2,72)=15.69, p<.001) across the
three time points as determined using repeated measures within group ANOVA. Post-hoc test
using Bonferroni correction revealed that there was an increase in self-compassion scores from
pre- (M=36.57, SD=7.55) to post-intervention (M=40.0, SD=6.37), which was statistically
significant (p=.001). The difference in self-compassion scores between pre-intervention and
three months follow-up (M=41.51, SD=7.13) also reached statistical significance (p<.001),
indicating that the increase in self-compassion scores from pre- to post-intervention was
maintained at follow-up. A small effect size (d=.49) was found for the improvement from pre- to
post-intervention (see Table 3).

Similarly, Friedman’s ANOVA showed that there was a significant change in compassion for
other scores across the three time points (χ2(2)=7.73, p=.02). Post-hoc analysis revealed a
significant increase in compassion for others from pre- to post-intervention (p=.002), although
the change from pre-intervention to follow-up did not quite reach significance (p=.06),
suggesting that the gain after the intervention was not sustained at three months follow-up. A
medium effect size (ES=.37) was found from pre- to post-intervention (see Table 2).

<Insert Table 3 about here>

Changes in stress

A repeated measures, within-group ANOVA showed that there was a significant change in stress
scores across the three time points (F(2,72)=4.52, p=0.01). A post-hoc test using Bonferroni
adjustment revealed a statistically significant reduction of stress scores from pre- (M=18.64,
SD=5.53) to post-intervention (M=15.59, SD=5.80, p=.02). Although there was a slight
reduction of stress from pre-intervention to three months follow-up (M=17.03, SD=5.33), it was
not statistically significant (p=.36), suggesting that the change had not been sustained. A medium
effect size (d=.54) was found for the reduction in stress scores from pre- to post-intervention (see
Table 3).

12
Changes in burnout

Exhaustion: There was a slight change in exhaustion scores from pre-intervention

(Median=2.37) to post-intervention (Median=2.25), and no change from post-intervention to
three months follow-up (Median=2.25). Friedman’s ANOVA showed no significant change
(χ2(2)=2.9 , p=.24) over the three time points (see Table 2).

Disengagement: There was no change in disengagement scores from pre-intervention

(Median=2.25) to post-intervention (Median=2.25), and from post-intervention to follow-up
(M=2.25). Friedman’s ANOVA confirmed that there was no statistically significant change in
disengagement scores (χ2(2)=5.11, p=.078) over the three time points (see Table 2).

Discussion

The aim of the present study was to investigate the effectiveness of a mindfulness intervention in
reducing stress and burnout in mental health professionals in Singapore. In support of the first
hypothesis, participants reported significant improvement in levels of mindfulness, self-
compassion, and compassion for others after a six-week mindfulness program. Improvements in
mindfulness and self-compassion scores were maintained three months later. However, the
second hypothesis was only partially supported. Although there was a significant reduction in
stress levels after the mindfulness program, the reduction was not maintained at three months
follow-up. In addition, there was no significant change in burnout measured pre-post intervention
and three months later. These results are discussed in detail below.

The significant reduction in perceived stress levels from pre-intervention to post-intervention

was consistent with other research. For example, Martín-Asuero and García-Banda (2010)
reported a reduction in stress-related psychological distress with moderate-large effect size in 29
health professionals working in Spain after attending an eight-week mindfulness program, while
Schenström, Rönnberg, and Bodlund (2006) reported reduction in perceived stress in 52
healthcare professionals working in Sweden after attending a 50 hours mindfulness program. Our
study extends existing literature to mental health professionals in Singapore.

13
Contrary to expectations, there was no significant change in exhaustion and disengagement
levels (subscales of burnout as measured using OLBI) on post-intervention or at three months
follow-up. This was in contrast to findings in the study by Goodman and Schorling (2012), in
which a significant reduction in emotional exhaustion, depersonalization, and improvement in
personal accomplishment (subscales of burnout measured using Maslach Burnout Inventory) was
reported in 93 healthcare providers after eight-week MBSR program. The eight-week MBSR
program used in the later study consisted of eight weekly, two and a half hour sessions, and
included a seven-hour meditation retreat. Participants were also asked to engage in 45 minutes of
meditation a day. In contrast, the mindfulness program in the present study consisted of six
weekly, two-hour sessions and a daily 30 minute meditation practice. Thus, it is possible that a
longer duration of mindfulness training may be required before positive changes in burnout
levels might be noticed. Another possible reason is that external (environmental) factors such as
workload and marital discord might have contributed to participants’ levels of burnout. These
issues were not measured or addressed in this study, and future studies should consider change in
burnout after such external factors are addressed.

Our results also indicated that there was a significant improvement in four of the five
mindfulness facets (observe, describe, non-judge and non-react) after the intervention, with
small-to-medium effect sizes. Significant improvement in self-compassion and compassion for
others with small-to-medium effect sizes was also noted. These changes were all sustained after
three months without additional booster sessions. This suggests that participation in the six-week
mindfulness program was associated with improvements in mindfulness and self-compassion
levels of the participants. These results are consistent with the study conducted by Neff and
Germer (2013), who reported improvements in mindfulness and compassion in their participants
after an 8-week Mindful Self-compassion program.

While there were significant changes in the scores in four of the mindfulness facets with
medium-to-large effect sizes, it was surprising to note that there was no significant change in the
act with awareness facet after the intervention. Act with awareness is the ability to pay attention
to the tasks at hand (Baer et al., 2008), and the ability to attend is considered a fundamental
practice in mindfulness practices (Kabat-Zinn, 1991; Shapiro & Carlson, 2009; Yongey Mingyur
Rinpoche, 2009). Interestingly, similar results were reported by Amutio, Martínez-Taboada,

14
Hermosilla, and Delgado (2014), in a study of 42 physicians after an eight-week MBSR
program. It is unclear why the ability to attend remained unchanged after mindfulness training.
One possible reason is the influence of the physical environment (e.g., noise level) on one’s
ability to pay attention, motivation levels, and interest in the task at hand. Thus, even if a
participant showed improvement in the ability to pay attention during a sitting meditation, their
attention in a noisy environment, or when doing tasks that they dislike, could still be short. Given
the importance in attention training in mindfulness practice, more research is warranted to further
investigate to the various factors that impact on the ability to pay attention.

Limitations

The study had several limitations. Firstly, although randomised control is considered the gold
standard when examining the efficacy of an intervention, a non-experimental one-arm study
without control was chosen for two main reasons: (1) the sample size required for a randomised
control trial was unlikely to be achievable in the setting from which the sample were drawn, and
(2) all participants for the study were drawn from the same institution. Thus, the use of a control
group would result high risk of contamination due to the sharing of knowledge and experiences
among participants in the intervention and control arms (Creswell, 2014; Heddle, 2002). Due to
the lack of a control group in the present study, it is possible that the noted improvements are a
result of non-specific effects of intervention. Thus, future experimental research with random
sampling to reduce selection bias, and a control group across different institutions, will be useful
to further investigate the effectiveness of this intervention. Secondly, one of the investigators of
this study was the main instructor for the mindfulness program. This might result in experimental
and social desirability bias (Strickland & Suben, 2012; Van de Mortel, 2008), with participants
possibly reporting a more positive outcome. Thirdly, the study was conducted with mental health
professionals and the results cannot be generalised to other healthcare professionals. Future
studies to assess whether the program might benefit healthcare professionals working in different
settings are warranted. Fourthly, the sample size for this study was under-powered to estimate
the influence of attendance and homework completion on outcome. Thus, future studies with
larger sample size are required to further investigate these influences.

15
 Conclusion

Overall, results suggest that mindfulness training was associated with improvement in
mindfulness and compassion, and reduction in stress, among this sample of mental health
professionals in one health care setting in Singapore. These results extend existing evidence from
the United States (Goodman & Schorling, 2012), Spain (Amutio et al., 2014), and Sweden
(Schenström et al., 2006). Our findings also showed that positive gains in mindfulness and self-
compassion levels made after the mindfulness training program were sustained at three months
follow-up. However, no significant change was noted in burnout levels after the mindfulness
training. Future experimental studies are warranted to validate the findings of the present study,
and to assess whether the program might benefit healthcare professionals working in different
settings in Singapore.

References

Acker, G. M. (2012). Burnout among mental health care providers. Journal of Social Work,
12(5), 475-490. doi:10.1177/1468017310392418

Amutio, A., Martínez-Taboada, C., Hermosilla, D., & Delgado, L. C. (2014). Enhancing
relaxation states and positive emotions in physicians through a mindfulness training
program: A one-year study. Psychology, Health and Medicine, 20(6), 720-731.
doi:10.1080/13548506.2014.986143

Asuero, A. M., Queraltó, J. M., Pujol-Ribera, E., Berenguera, A., Rodriguez-Blanco, T., &
Epstein, R. M. (2014). Effectiveness of a mindfulness education program in primary
health care professionals: A pragmatic controlled trial. Journal of Continuing Education
in the Health Professions, 34(1), 4-12. doi:10.1002/chp.21211

Baer, R. A., Smith, G. T., Lykins, E., Button, D., Krietemeyer, J., Sauer, S., . . . Williams, J. M.
G. (2008). Construct validity of the five facet mindfulness questionnaire in meditating
and nonmeditating samples. Assessment, 15(3), 329-342.

16
Chiang, Y., & Chang, Y. (2012). Stress, depression, and intention to leave among nurses in
different medical units: implications for healthcare management/nursing practice. Health
Policy, 108(2-3), 149-157.

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress.
Journal of health and social behavior, 24(4), 385-396.

Cohen-Katz, J., Wiley, S. D., Capuano, T., Baker, D. M., Kimmel, S., & Shapiro, S. (2005). The
effects of mindfulness-based stress reduction on nurse stress and burnout, Part II: A
quantitative and qualitative study. Holistic Nursing Practice, 19(1), 26.

Creswell, J. W. (2014). Research design: qualitative, quantitative, and mixed methods

approaches. Thousand Oaks: SAGE Publications.

Cullen, M. (2011). Mindfulness-Based Interventions: An Emerging Phenomenon. Mindfulness,

2(3), 186-193.

Demerouti, E., & Halbesleben, J. R. B. (2005). The construct validity of an alternative measure
of burnout: investigating the English translation of the Oldenburg burnout inventory.
Work and Stress, 19(3), 208-220. doi:10.1080/02678370500340728

Escuriex, B. F., & Labbé, E. E. (2011). Health Care Providers' Mindfulness and Treatment
Outcomes: A Critical Review of the Research Literature. Mindfulness, 2(4), 242-253.

Feng, L., Cao, Y., Zhang, Y., Wee, S. T., & Kua, E. H. (2011). Psychological therapy with
Chinese patients. Asia‐Pacific Psychiatry, 3(4), 167-172. doi:10.1111/j.1758-
5872.2011.00148.x

Field, A. P. (2005). Discovering statistics using SPSS: and sex, drugs and rock 'n' roll. Thousand
Oaks, Calif: Sage Publications.

Gauthier, T., Meyer, R. M. L., Grefe, D., & Gold, J. I. (2015). An On-the-Job Mindfulness-based
Intervention For Pediatric ICU Nurses: A Pilot. Journal of Pediatric Nursing, 30(2), 402-
409. doi:10.1016/j.pedn.2014.10.005

17
Goodman, M. J., & Schorling, J. B. (2012). A mindfulness course decreases burnout and
improves well-being among healthcare providers. International Journal of Psychiatry in
Medicine, 43(2), 119-128.

Halbesleben, J. R. B., & Demerouti, E. (2005). The construct validity of an alternative measure
of burnout: Investigating the English translation of the Oldenburg Burnout Inventory.
Work and Stress, 19(3), 208-220.

Heddle, N. M. (2002). Clinical research designs: quantitative studies. Vox sanguinis, 83 Suppl 1,
247-250. doi:10.1111/j.1423-0410.2002.tb05312.x

Henslin, J. M. (1999). Sociology: A Down-to-Earth Approach (4th edn.). Boston: Allyn and
Bacon.

Irving, J. A., Dobkin, P. L., & Park, J. (2009). Cultivating mindfulness in health care
professionals: A review of empirical studies of mindfulness-based stress reduction
(MBSR). Complementary Therapies in Clinical Practice, 15(2), 61-66.

Kabat-Zinn, J. (1991). Full catastrophe living: using the wisdom of your body and mind to face
stress, pain, and illness. New York: Dell Publishing.

Lee, E. (2012). Review of the Psychometric Evidence of the Perceived Stress Scale. Asian
Nursing Research, 6(4), 121-127. doi:10.1016/j.anr.2012.08.004

Lloyd, C., & King, R. (2004). A survey of burnout among Australian mental health occupational
therapists and social workers. Social psychiatry and psychiatric epidemiology, 39(9),
752-757. doi:10.1007/s00127-004-0808-7

Martín-Asuero, A., & García-Banda, G. (2010). The mindfulness-based stress reduction program
(MBSR) reduces stress-related psychological distress in healthcare professionals. Spanish
Journal of Psychology, 13(2), 897-905.

Mitchell, A. M., Crane, P. A., & Kim, Y. (2008). Perceived stress in survivors of suicide:
Psychometric properties of the perceived stress scale. Research in Nursing and Health,
31(6), 576-585. doi:10.1002/nur.20284

18
Neff, K. D. (2003). The Development and Validation of a Scale to Measure Self-Compassion.
Self and Identity, 2(3), 223-250. doi:10.1080/15298860309027

Neff, K. D., & Germer, C. K. (2013). A Pilot Study and Randomized Controlled Trial of the
Mindful Self‐Compassion Program. Journal of Clinical Psychology, 69(1), 28-44.
doi:10.1002/jclp.21923

Pommier, E. A. (2010). The compassion scale. (Doctorate dissertation). Retrieved from

ProQuest, UMI Dissertations Publishing

Raes, F., Pommier, E., Neff, K. D., & Van Gucht, D. (2011). Construction and factorial
validation of a short form of the Self-Compassion Scale. Clinical Psychology &
Psychotherapy, 18(3), 250-255. doi:10.1002/cpp.702

Scanlan, J. N., & Still, M. (2013). Job satisfaction, burnout and turnover intention in
occupational therapists working in mental health. Australian occupational therapy
journal, 60(5), 310-318.

Schenström, A., Rönnberg, S., & Bodlund, O. (2006). Mindfulness-based cognitive attitude
training for primary care staff: a pilot study. Complementary Health Practice Review,
11(3), 144-152.

Shanafelt, T. D., Bradley, K. A., Wipf, J. E., & Back, A. L. (2002). Burnout and self-reported
patient care in an internal medicine residency program. Annals of internal medicine,
136(5), 358-367.

Strickland, B., & Suben, A. (2012). Experimenter Philosophy: The Problem of Experimenter
Bias in Experimental Philosophy. Review of Philosophy and Psychology, 3(3), 457-467.
doi:10.1007/s13164-012-0100-9

Van de Mortel, T. F. (2008). Faking it: Social desirability response bias in selfreport research.
Australian Journal of Advanced Nursing, 25(4), 40-48.

19
Vardakou, I., Bakker, A. B., Demerouti, E., & Kantas, A. (2003). The convergent validity of two
burnout instruments : a multitrait-multimethod analysis. European Journal of
Psychological Assessment, 19(1), 12-23. doi:10.1027/1015-5759.19.1.12

Yang, S., Meredith, P., & Khan, A. (2015). Stress and burnout among healthcare professionals
working in a mental health setting in Singapore. Asian Journal of Psychiatry, 15, 15-20.

Yongey Mingyur Rinpoche. (2009). Joyful Wisdom: Embracing change and finding freedom.
New York: Three Rivers Press.

Zhang, W. (2002). Singapore’s Modernization: Westernization and Modernizing Confucian

Manifestations. New York: Nova Science.

20
Table 1. Socio-demographics characteristics of participants, N=37.

Variable n %

Profession Nurse 8 21.1

Occupational therapist 3 8.1

Doctor/ Psychiatrist 3 8.1

Social worker 7 18.9

Case manager 1 2.7

Pharmacist 1 2.7

Psychologist/ Counsellor 11 29.7

Researchers 3 8.1

Age group <25 2 5.4

25-30 10 27.0

30-35 9 24.3

35-40 5 13.5

40-45 6 16.2

>45 5 13.5

Ethnicity Chinese 29 78.4

Malay 2 5.4

Indian 4 10.8

21
 Gender Male 7 18.9

Female 30 81.1

Education Level Diploma 2 5.4

Degree 20 54.1

Post-graduate 14 37.8

Employment Full-time 35 94.6

Part-time 2 5.4

Year of Experience <5 16 43.2

6-10 6 16.2

11-20 12 32.4

>20 3 8.1

Note: Sum of each category might not add up to 37 due to missing data

22
Table 2. Median values for FFMQ, CS and OLBI along with Friedman’s ANOVA results at pre-,
post-, and three months after intervention.
Pre Post ES Follow-up χ2(2) p

Median Median Median

FFMQ-total 121 131 .67 132 16.68 <.001

Observe 23 28 .49 28 21.5 <.001

Describe 25 26 .29 28 12.01 .002

Act with 26 25 .09 26 0.10 .95

awareness

Non-judge 23 26 .27 26 11.5 .003

Non-react 20 23 .48 23 32.08 <.001

CS 92 96 .37 97 7.73 .02

OLBI

Exhaustion 2.38 2.25 .05 2.25 2.90 .24

Disengagement 2.25 2.25 .14 2.25 5.11 .08

Note: FFMQ = Five Facets Mindfulness Questionnaire; CS = Compassion Scale; OLBI =

Oldenburg Burnout Inventory; IQR = Interquartile Range; ES = Effect size calculated for pre-
and post-intervention (Field, 2005)

23
Table 3. Mean values for SCS-S and PSS and within-groups ANOVA at pre-, post-, and three
months after intervention

Pre Post d Follow- F(2, p

up 72)

Mean SD Mean SD Mean SD

SCS-S 36.57 7.55 40.0 6.37 .49 41.5 7.13 15.69 <.001

PSS 18.65 5.53 15.59 5.80 .54 17.03 5.33 4.52 .01

Note: SCS-S = Self-compassion Scale-Short form; PSS = Perceived Stress Scale; d = Cohen’s d
value calculated for pre- and post-intervention

24