measuring mindfulness mses_b_cayoun_nz2012
TRANSCRIPT
The Validity, Reliability and
Factor Structure of the
Mindfulness-based Self-
Efficacy Scale
Bruno A. Cayoun (UTAS)
Sarah E. Francis (Swinburne U)
Natasha Kasselis (UTAS
Clive Skilbeck (UTAS)
Practice & Operationalisation:
2 Main Aspects1. Concentration on breathing (attention regulation)
– Training in sustained attention, meta-cognitive awareness, inhibitory
control and attention switching
– Benefits include decreased distractibility from irrelevant thoughts
and sensory stimuli and cognitive flexibility
2. Systematic body scanning to develop equanimity (emotion
regulation)
– Generalised interoceptive exposure and response prevention
– Requires inhibitory control to prevent learned responses and foster
acceptance
• General Aims
– Develop greater self-awareness, objectivity, acceptance and
detachment with each experiences
The Need for Measuring
Progress with Therapy
• Helps measuring treatment efficacy
• Assists in understanding of the mechanisms
• Evidence that clients whose therapist has access to
outcome and alliance information are
– less likely to deteriorate
– Less likely to drop out of therapy
– twice more likely to achieve a clinically significant change
(Lambert et al., 2001; Miller, Duncan, & Hubble, 2004)
Main Current Measures
• Freiburg Mindfulness Inventory (FMI; Buchheld,
Grossman, & Walach, 2001)
• Mindfulness Attention Awareness Scale (MAAS;
Brown & Ryan, 2003)
• Cognitive Affective Mindfulness Scale (CAMS;
Feldman, Hayes, Kumar, & Greeson, 2004)
• Kentucky Inventory of Mindfulness Skills (KIMS; Baer
Smith, & Allen, 2004)
• Five Factor Mindfulness Questionnaire (FFMQ; Baer,
et al., 2006)
– FFMQ was derived from all of the above
Limitation of Current Measures
1. Problems with construct validity– KIMS 20/FFMQ: “I make judgments about whether my thoughts
are good or bad”
– KIMS 34/FFMQ: “My natural tendency is to put my experiences
into words”
– CAMS 6/FFMQ: “I am easily distracted”
– MAAS 12: “I drive places on “automatic pilot” and then wonder
why I went there
– MASS 15: “I snack without being aware that I’m eating”
2. How do we know what we don’t know?
Mindfulness scores (FMI); means (whiskers, standard deviation) of long-terms experienced insight (Vipassana)
meditators healthy non-meditating students, and binge-drinking non-meditating students. Significant difference
between meditators and binge drinkers (p < 0.0004)differences between meditators and normal students
approached significance (p < 0.06). From Grossman, P. & van Dam, N. (2011). Mindfulness, by any other name . . . :
Trials and tribulations of sati in Western psychology and science. Contemporary Buddhism, 12, 219-239.
Limitations with
Symptoms Measurements
• Many questionnaires used by clinicians measure
change in symptoms and omit change in self-efficacy
– Decrease in symptom severity does not always represent
insight, coping abilities and resilience to future stressors
– Clients can be highly symptomatic while having good coping
skills
– Client’s avoidance can leads to poor reporting of symptoms
– Client’s exposure to stressors can lead to increased
symptoms but also more manageability of symptoms =
increased self-efficacy
Alternative Measurement
• Self-Efficacy: The perception of or belief in one’s
ability to reach a goal, or in one’s own power to
achieve
• Self-efficacy scales are most useful when constructed
for a particular context (Bandura, 1997)
– Typical consequences of mindfulness
1.Regulation of attention and emotion
2.Ability to prevent avoidance
3.Ability to relate and connect socially
4.Ability to empathise and prevent harmful action
Development of the
Mindfulness-based Self-Efficacy Scale
• Item pool collected empirically (not assumed)– Collected from several hundred statements from
clients completing 8-week mindfulness-based therapy in psychiatric clinics, various private and public hospitals and community health services (2001-2004)
• Original aim– To examine the kind of recurrent changes in self-
efficacy reported by patients during and after mindfulness training irrespective of their diagnosis.
Development of the
Mindfulness-based Self-Efficacy Scale
• Clusters formation– Items were divided into 7 generic clusters (Behavioural,
Cognitive, Affective, Somatic, Social, Avoidance, Mindfulness)
– The 5 most recurring items in each cluster were retained, totaling a 35-item scale
– Each item scored a 5-point Likert scale from 0 (not at all) to 4 (completely).
• Past studies– E.g., Husinger (2006) used 1250 UMASS students (α=0.86),
showed good concurrent validity when compared with the FMI (r= 0.62, p < .001), the KIMS (r = 0.49, p < .001), and the MAAS (r= 0.57, p < .001).
Study 1
• Aim– Evaluate concurrent validity MSES in the general community
• Participants: – 197 from general population and Swinburne University
– Incomplete data (n=34) were excluded, 163 were retained
– 69% females, 31% males, 76% Australians, 60% married/de facto
– Age: 18-25 years = 5%, 26-35 years = 23%, 36-45 years = 21%, 46-55 years
= 29%, 56-65 years = 18%, over 66 years = 4%
– Education: postgrad = 54%, undergrad = 58%, high school = 10%
– Income: more than $100K = 16%
• Measures– Mindfulness (MSES, FMI, KIMS, MAAS)
• Internal consistency – Good (Chronbach’s Alpha = .88)
Participants who scored higher on the MSES were Australian, older, married or in a de facto
relationship
Correlations Between Measures
Study 2• Aim
– Evaluate further concurrent validity, convergent validity, test-retest reliability
and factor structure of the MSES in the general community
– Compare scores in mental illness, current distress and no distress
• Participants: – 521 from general population and University of Tasmania
– Incomplete data (n=10) were excluded, 511 were retained
– 75% females, 25% males, 18% Mental illness, 40% married/de facto
– Age: 18-25 years = 46%, 26-35 years = 17%, 36-45 years = 15%, 46-55
years = 10%, 56-65 years = 9%, over 66 years = 3%
– Education: postgrad = 14%, undergrad = 47%, Dip/high school = 36%
– Income: more than $80K = 8%
• Measures– Mindfulness (MSES, FFMQ, DASS)
• Internal consistency – Good: (Chronbach’s Alpha = .86)
Sample CharacteristicsMental illness, psychological distress, exposure to mindfulness
Exploratory Factor Analysis
• A 6-factor solution was retained, accounting for 57.5% of the
variance was found using Kaiser’s Criterion as the cut-off
• Other extraction methods and the exclusion of potentially
problematic demographical groups from the sample were
tried but did not improve parsimony
• To address co-linearity and on theoretical grounds, loadings
below .5 were excluded and items that were theoretically
uninterpretable were removed. Factors with 2 or less items
on them were removed. In total 13 items were removed.
Factor Structure of proposed 22-item Mindfulness-based Self Efficacy Scale Containing 22 Items (N = 511)
Factor Interpretation
1. Emotion regulation
– Relates to an involuntary or subconscious emotional response that is well
modulated and falls within the expected normal range of responses
2. Equanimity
– Relates to the ability to normalise difficulties and prevent reactivity
3. Taking Responsibility
– Relates to clarity of interpersonal boundaries and locus of control
4. Distress Tolerance
– Relates to voluntary effort which inhibits avoidance of experiential intolerance
or discomfort
5. Social skills
– Relates to social abilities in the broader sphere of interaction
6. Interpersonal effectiveness
– Relates to the ability to connect with others within the intimate sphere of
relationships
Test-Retest Reliability
• Positive correlation between initial test scores and retest
scores (N = 100). Two weeks interval.
• Original 35 item MSES: r = .82, p < .01.
– Shared variance was 68%.
• Proposed 22-item MSES: r = .88, p < .01
– Shared variance was 78%.
Notes: MSES = Mindfulness-based Self Efficacy Scale. N= 521.
** Correlation is significant at the 0.01 level.
Correlations between the factor scores were modest but significant. This was expected because whilst
they are distinct from each other they are also somewhat correlated, as they are designed to measure
the same underlying construct of mindfulness-based self efficacy.
Correlations between MSES factor scores
Discriminant Validity: Correlations with the DASS21
Notes: *p < .01
Discriminant Validity: Correlations with the DASS21
Notes: ** p< .01. Non-judging = Non-judging of Inner Experience. Non-reactivity = Non-reactivity to Inner Experience.
Correlates of MSES Factor Scores and FFMQ Factors (N = 511)
Convergent Validity
• Independent samples t-tests compared factor scores of participants with current mental illness with those without.
• There was significant differences between these two groups scores.
• Confirms unpublished studies that the MSES is excellent at differentiating between clinical and non-clinical populations.
• The proposed 22 item scale differentiated the groups slightly better than the original 35-item scale.
Sensitivity to Psychopathology
Study 3
• Aim– Evaluate the sensitivity to change of the MSES in the clinical population
during and after mindfulness-based treatment.
• Participants: – Random sample of 99 adults (age 18-78) who completed treatment of
moderate to severe psychological disorders at the MiCBT Institute, Hobart
– All clients were referred by GP or self-referred
– Disorders: GAD, PD, Dys, MDD, PSTD, OCD, BiPD, BPD, APD, DPD, OCPD
– Comorbidity: 30% with 3+ conditions; 80% with 2 conditions, 12% D/Alcohol
– Incomplete data in mid-treatment (n=32) and post-treatment (n=81) were
excluded
• Measures reported – MSES, DASS21
Sensitivity to Treatment Effects
15.00
25.00
35.00
45.00
55.00
65.00
Pre Mid Post
DASS Total
MSES Total
Sensitivity to Treatment Effects
Sensitivity to Treatment Effects
Case Example
• Panic Disorder with Agrophobia,
• Major Depressive Disorder (Chronic)
Summary• The MSES-Revised comprises 6 subscales of self-efficacy
– Test-retest reliability is very good (r = .88, N = 100, p < .01. Shared
variance for the 22-item scale was 78%).
• Internal consistency is high (Chronbach alpha = .86)
• Good discriminant validity
– Inverse relationship with the Depression Anxiety and Stress-Short
Form (DASS21)
– Clear discrimination of scorers who report having a mental illness
from those who do not.
• Good convergent validity
– Consistently in the good range with the KIMS, FFMQ, FMI and MAAS.
• Good sensitivity to change
– Consistently showing that scores improve as a function of successful
treatment or decreased psychological distress over time
Appendix: Additional Information
Scale item Mean SD I-T
1
I am able to think about what I am about to do before I act
3.93
.69
.35
2 When an unpleasant thought enters my mind I can cope with it 3.82 .68 .56
3 When I relax, I can feel sensations in my body 3.56 .99 .39
4 I get easily overwhelmed by my emotions 1.83 .89 .42
5 I find it difficult to make new friends 2.10 .93 .34
6 I try to avoid uncomfortable situations even when they are really important 1.92 1.01 .42
7 I am aware when I am about to do something that could hurt me or someone else 3.89 .86 .28
8 Stopping myself from engaging in unwanted or hurtful behaviours is very difficult 2.26 .81 .45
9 I know that my thoughts don’t have the power to hurt me 3.49 1.29 .46
10 When I am stressed, I am aware of unpleasant body sensations 3.41 1.06 .24
11 When I feel very emotional, it takes a long time for it to pass 1.84 .82 .47
12 I feel comfortable saying sorry when I feel I am in the wrong 3.9 1.11 .32
13 It is OK for me to feel strong emotions 4.11 .99 .55
14 It is often too late when I realize I overreacted in a stressful situation 1.83 .92 .35
15 If something needs to be done, I am able to complete it within a reasonable time 3.94 .88 .45
16 I get so caught up in my thoughts that I end up feeling very sad or anxious 2.04 .85 .48
17 When I have unpleasant feelings in my body, I prefer to push them away 1.75 1.01 .54
18 I believe that I can make my life peaceful 3.98 .84 .46
19 I can resolve problems easily with my partner (or my best friend if single) 3.75 .89 .48
20 I can face my thoughts, even if they are unpleasant 4.03 .87 .55
21 I am tolerant with myself when I am repeating old habits that are no longer helpful 3.18 1.02 .46
22 My actions are often controlled by other people or circumstances 1.95 .83 .44
23 I get caught up in unpleasant memories or anxious thoughts about the future 1.91 .95 .34
24 I can deal with physical discomfort 3.45 .97 .31
25 I feel I cannot love anyone 2.72 .69 .50
26 I am often in conflict with one (or more) family member 2.42 .78 .33
27 I avoid feeling my body when there is pain or other discomfort 2.35 .72 .43
28 I find it difficult to accept unpleasant experiences 1.89 .85 .54
29 I do things that make me feel good straight away even if I will feel bad later 1.96 .85 .35
30 When I have a problem, I tend to believe if will ruin my whole life 2.63 .71 .49
31 When I feel physical discomfort, I relax because I know it will pass 3.34 .97 .48
32 Even when things are difficult I can feel happy 3.13 .86 .56
33 I can feel comfortable around people 3.84 .87 .57
34 Seeing or hearing someone with strong emotions is unbearable to me 2.24 .85 .36
35 If I get angry or anxious it is generally because of others 1.80 .97 .33