Who Benefits More? Optimising Mindfulness Based Interventions for Improved Psychological Outcomes

Health and well-being are increasingly being emphasized in modern society. In response,
mindfulness-based training (MBT) (e.g. Mindfulness Based Stress Reduction (MBSR)
(Kabat-Zinn,1990)) is slowly gaining popularity as a method to alleviate the stress of modern
day living. MBT involves bringing one's attention to the present moment, stopping one's
thoughts from drifting into concerns about the past or the future It has significant effects
on improving mental health via reducing symptoms of stress, anxiety and depression (Fjorback,
Arendt, Ørnbøl, Fink & Walach, 2019). Local studies have similarly demonstrated the efficacy
of mindful training in stress reduction for mental health professionals in Singapore (Suyi,
Meredith & Khan, 2017).

Reducing stress and improving metacognitive processes also contributes to better emotional
regulation, thus facilitating better sleep quality (Chiesa et al., 2011; Zeidan et al.,
2010).. By breaking up cycles of rumination and worry, mindfulness is hypothesized to reduce
"verbal overregulation" and facilitate the disengagement necessary to fall asleep (Gross et
al., 2011). Mindfulness programs have been found to improve sleep quality in both healthy
individuals (Lazar et al., 2000) and patients with medical or psychiatric illness, including
insomnia and depression (Carlson & Garland, 2005; Heidenreich et al., 2006).

While there is much documentation supporting the efficacy of mindfulness training on stress
and sleep, there is little research that attempts to evaluate the interpersonal differences
or factors that predict the efficacy of mindfulness training on sleep and stress. Isolating
these predictive variables consequently allows a better way to evaluate suitability of the
mindfulness intervention on an individual basis; maximizing the results for the individual.
With this new information, individuals suffering from stress and bad sleep would be able to
better be able to select the most effective intervention.

While the concepts above have often been explored in relation to mindfulness, no studies to
date have used machine learning to assess how these variables predispose an individual to
benefit from mindfulness training. Generating this algorithm will allow for the prediction of
treatment response in future trainees. Further, a better understanding of the effect of
individual differences on efficacy can open the door for future research to examine how and
why these differing results exist and incorporate this information to increase the
effectiveness of mindfulness training.

Participants signing up for the 4 or 8 weeks mindfulness training courses at the Brahm Centre
and run by Potential Project (two external organization providing Mindfulness based courses)
will be automatically sent a link to the questionnaire packet several days after course
enrolment. Participants will be provided with a survey link in the enrolment email that is
sent to participants to confirm their enrolment, separate recruitment e-mails will not be
sent out.

Participants will be informed that the data will form part of a research study, that any
information provided is anonymous, and that they can opt out of completing the questionnaires
without any penalty to their class enrolment. Recruitment emails will not be sent out. On
enrollment in any of the courses, participants will be informed that there is an option to
take part in the study.

Staff sending out emails will be separate from staff carrying out the mindfulness training
programs. This will ensure that there will be no scenario where a dependent relationship will
occur.

Participants can decide if they prefer to sign up for the 4 weeks or 8 weeks Mindfulness
courses, depending on availability at the Mindfulness centres.

Questionnaires will be administered through the platform SurveyMonkey ©. SSL encryption will
be used to protect sensitive data as it is relayed between the respondent's computer and
SurveyMonkey© servers.

Questionnaires will be administered twice, with the predictor surveys administered at
pre-intervention only, and outcome variables collected both pre- and post-intervention (PSQI
and PSAS).

Predictive factors 10 predictive factors were chosen to evaluate their predictive qualities
on the effectiveness of mindfulness training in reducing stress and improving sleep:
Personality: Mindfulness training has been shown to be more effective in students with higher
scores in conscientiousness and neuroticism (Winning & Boag, 2015). As such, the NEO-FFI-3
will be used to measure the dimensions of personality.

Trait Mindfulness: Baseline trait mindfulness has been shown to be a significant moderator of
MBSR intervention effects (Shapiro, Brown, Thoresen, & Plante, 2010). The investigators have
included trait mindfulness to corroborate previous studies that have shown this relationship
as well as to explore its effect within a local population.

Mood and depression: While depression and mood has been often measured in conjunction with
Mindfulness training, there has been little to no research on how it relates to mindfulness
training efficacy State-Trait Anxiety: Mindfulness has been shown to alleviate state trait
anxiety in individuals (Bergen-Cico & Cheon, 2013). The investigators are thus interested in
whether a higher or lower baseline score in the STAI will affect the efficacy of mindfulness
training Empathy: There are mixed results in regards to the relationship between empathy and
mindfulness; with some studies finding no significant connection (Bergen-Cico & Cheon, 2013),
and others showing differing effectiveness dependent on personality traits (Winning & Boag,
2015; Ridderinkhof, de Bruin, Brummelman, & Bögels, 2017).

Self-compassion: Self compassion is believed to be developed and enhanced through mindfulness
training. Birnie, Speca & Carlson (2010) found strong associations between self-compassion
and mindfulness. It was found that changes in self-compassion were predictive of changes in
mindfulness.

Learning styles: This was included to investigate whether the difference in learning styles
in individuals would affect the ways in which the mindfulness training are internalized; in
turn influencing the efficacy of mindfulness training.

Expectancy/Credibility: Expectancy has been shown to be predictive in evaluating the outcome
on some measures (Devilly & Borkovec, 2000). It would be beneficial to understand its effect
on the effectiveness of mindfulness training.

Emotional Regulation: Emotion regulation has been shown to improve with mindfulness-based
interventions (Guendelman, Medeiros & Rampes, 2017). The investigators are interested in
understanding the effect of emotional regulation on mindfulness training.

Coping Methods: Mindfulness has been shown to affect the coping strategies used in
individuals. The effect of coping methods on mindfulness training efficacy is thus of
interest to our study.

Questionnaires Pittsburgh Sleep Quality Index (PSQI; Buysse et al., 1989)

Perceived Stress Scale (PSS; Cohen, Kamarck & Mermelstein,1983)

Objectives

This research has four objectives:

To corroborate findings on the effect of mindfulness training on perceived stress.

Hypothesis: based on previous literature, the investigators predict that mindfulness
interventions will decrease perceived stress, with lower PSS score post intervention.

To test whether individual differences might be predictive of the effectiveness of
mindfulness training on stress Hypothesis: through the use of machine learning, the
investigators aim to create a model that can predict changes in perceived Stress (PSS) post
intervention based on individual predictors as defined above.

To test whether individual differences might be predictive of the effectiveness of
mindfulness training on sleep Hypothesis: through the use of machine learning, the
investigators aim to create a model that can predict changes in Sleep quality (PSQI) post
intervention based on individual predictors as defined above.

To test the effects of group mindfulness interventions on stress and sleep quality during a
period of global uncertainty, a partial lockdown due to the COVID-19 pandemic, compared to
Mindfulness intervention in a non-emergency situation.

Hypothesis (i):the investigator predict that participants would report higher levels of
perceived stress and poorer sleep quality at baseline during the period of the pandemic
(DORSCON Orange and lockdown) relative to an earlier control period; Hypothesis (ii): the
investigators predict that online mindfulness training (during lockdown) would have be
non-inferior/equivalent in reducing perceived stress and improving sleep quality compared
with DORSCON Orange and the control period.