Abstract
Background. Mindfulness-based stress reduction (MBSR) has demonstrated efficacy for alleviating cancer-related distress. Although theorized to be the means by which people improve, it is yet to be determined whether outcomes are related to the development or enhancement of mindfulness among participants. This study examined the effect of participation in an MBSR program on levels of mindfulness in a heterogeneous sample of individuals with cancer, and if these changes were related to improvements in stress and mood outcomes. Methods. In all, 268 individuals with cancer completed self-report assessments of stress and mood disturbances before and after participation in an 8-week MBSR program. Of these, 177 participants completed the Mindful Attention Awareness Scale and 91 participants completed the Five Facet Mindfulness Questionnaire, at both time points. Results. Levels of mindfulness on both measures increased significantly over the course of the program. These were accompanied by significant reductions in mood disturbance (55%) and symptoms of stress (29%). Increases in mindfulness accounted for a significant percentage of the reductions in mood disturbance (21%) and symptoms of stress (14%). Being aware of the present moment and refraining from judging inner experience were the 2 most important mindfulness skills for improvements of psychological functioning among cancer patients. Conclusions. These results add to a growing literature measuring the impact of mindfulness and its relationship to improved psychological health. Moreover, specific mindfulness skills may be important in supporting these improvements.
Introduction
Mindfulness refers to the cultivation of conscious awareness in the present moment in an open and nonjudgmental manner. 1 Mindfulness is often fostered through the practice of meditation, in which individuals learn to observe both their internal state and their external environment without attachment, evaluation, or judgment. Mindfulness-based stress reduction (MBSR) is the most commonly used intervention, which incorporates and applies formal mindfulness meditation training to clinical populations in a health care setting. 2 Throughout the MBSR program, individuals learn the principles and practice the techniques of mindfulness meditation for the purpose of fostering mindfulness, enhancing emotional regulation, and decreasing maladaptive reactions to stress. 2
The efficacy of MBSR to alleviate emotional and physical difficulties associated with chronic disease including cancer has been repeatedly demonstrated (see meta-analyses3-6). Of all chronic illnesses, cancer may produce the most significant levels of distress, and the need for psychosocial interventions to improve patients’ psychological adjustment and quality of life has been well documented.7,8 The use of MBSR as a psychological intervention in oncology began at the Tom Baker Cancer Centre in Calgary, Canada in 1996. Since then, several studies have found evidence that participation in MBSR can help alleviate cancer-related distress, enhance posttraumatic growth, and improve quality of life of patients and their partners.9-14 These positive outcomes have been demonstrated in randomized controlled studies, including wait-list15,16 and usual care control groups.17,18 The durability of the psychological improvements has been demonstrated to persist for 6 to 12 months following MBSR participation.19,20 However, several literature reviews21-25 emphasize the need for further research into the active components of MBSR.
The mechanisms by which MBSR produces health benefits are yet to be determined. In particular, it is not clear how an increase in mindfulness may relate to positive psychological outcomes. This is due, in part, to the lack of conceptual consistency in defining the term “mindfulness” in the literature. Mindfulness has been referred to as a theoretical construct, a quality of awareness, and a range of meditation practices.26-28 Although the process of operationally defining mindfulness is still ongoing, several instruments attempting to measure it have been developed. These measures are conceptually, as well as structurally, quite different. For example, the Mindful Attention Awareness Scale (MAAS) 29 consists of a single factor, whereas the Five Facet Mindfulness Questionnaire (FFMQ) 30 is composed of 5 different domains thought to encompass various components of mindfulness.
Only a handful of studies have investigated the impact of MBSR on levels of mindfulness in individuals with cancer. Increases in mindfulness following MBSR participation have been reported using the FFMQ 17 and the MAAS.14,16,29 Further focused research is needed with larger samples, to strengthen these preliminary results. Even in other populations, research is yet to conclusively demonstrate that increased levels of mindfulness contribute to the positive outcomes of MBSR. 31
Across some clinical and nonclinical groups, increases in mindfulness have mediated improved psychological functioning.32-34 Among cancer patients, however, the research is less clear cut. Bränström et al 17 reported a mediating effect of increased mindfulness (assessed by the FFMQ) in reducing perceived stress, posttraumatic avoidance, and increased positive emotions, but not anxiety or depression. In contrast, Labelle et al 16 did not find a mediating role of increased mindfulness for a reduction of depressive symptoms following the MBSR program, as assessed by the MAAS. More recently, Dobkin and Zhao 35 showed an association between increased mindfulness (as assessed by the MAAS) and a reduction of self-reported depressive and physical symptoms among chronically ill patients, of which the majority (77.9%) were cancer patients, as well as an increased perception of optimism and control over their environment. Thus, a small number of studies measuring mindfulness in cancer patients have found increased mindfulness after participation in MBSR, but further evidence is required to demonstrate that development of mindfulness contributes to improvements in other measures of psychological health.
It should also be noted that these studies are cross-sectional in nature, making statements of causality impossible. Both levels of mindfulness and psychological outcomes improved simultaneously (not sequentially) over the course of the program, so statistically it is impossible to know which one accounts for the improvements in the other. Theoretically, however, it makes more sense that improvements in mindfulness skills would lead to decreases in psychological symptomatology than the reverse. However, it is entirely possible that decreases in stress and depression could be facilitated by other mechanisms inherent in the program, providing a subsequent opportunity for increased levels of mindfulness. The determination of whether early changes in mindfulness are associated with later improvements in psychological functioning can only be investigated through studies designed with multiple assessment periods over the course of the program.
Although also handicapped by a similar design with only pre- and post-assessments, this research further investigates the role of increased mindfulness as it relates to reductions in symptoms of stress and mood disturbance in a heterogeneous sample of cancer patients. It is intended to further the understanding of whether an increase in levels of mindfulness is an integral part of the MBSR program, and how this may relate to improvements in psychological well-being. Preliminary evidence has suggested that increased levels of mindfulness are associated with improvements in psychological well-being among cancer patients, but those relationships were either simply correlational at one time point 36 or inconsistent when mediation analyses were applied.16,17 This study, therefore, will examine levels of mindfulness using 2 different measures (the MAAS and FFMQ) and investigate the relationship of those scores to stress and mood outcomes in a larger group of cancer patients.
Methods
Participants
The Department of Psychosocial Resources at the Tom Baker Cancer Centre has been offering and researching the MBSR program since 1996. The data included in this study were provided by patients who participated in previous research studies between the years 2000 and 2008, resulting in a total sample of 268 heterogeneous cancer patients.
Procedures
Patients were referred to the MBSR program by hospital staff, self-referred via posted advertisements, or recruited for specific MBSR studies. Participants who provided complete pre- and post-program assessment, including measures of mindfulness, symptoms of stress, and positive and negative moods were eligible for inclusion. No restrictions were placed on gender, age, tumor location, or disease stage. Participants must have attended at least 5 of the 8 sessions to be included. This decision was made based on the clinical determination of the minimal number of session required to have an adequate understanding and meaningful experience of the intervention. The Conjoint Health Research Ethics Board at the University of Calgary gave ethical approval for data from these studies to be amalgamated and analyzed.
Measures
Mindful Attention Awareness Scale
The MAAS 29 was developed to measure the extent to which individuals attend, to and are aware of, the present moment. It consists of 15 items, and for each item the respondents are asked to indicate how often they perform behaviors without awareness (eg, I rush through activities without being really attentive to them) using a 6-point Likert-type scale. The MAAS has been used to measure mindful awareness among clinical and nonclinical populations 36 and had demonstrated good internal consistency (Cronbach’s α = .85) and test–retest reliability (Pearsons r = .81). 29
Five Facet Mindfulness Questionnaire
The FFMQ 30 is composed of 39 items generated using a factor analysis of 5 preexisting measures of mindfulness.29,37-41 The FFMQ includes 5 scales: (1) observing (eg, I notice the smell and aromas of things); (2) describing (eg, I am good at finding words to describe my feelings); (3) acting with awareness (eg, I find myself doing things without paying attention); (4) nonjudging of inner experience (eg, I think that some of my emotions are bad or inappropriate and I shouldn’t feel them); and (5) nonreactivity to inner experience (eg, I perceive my feelings and emotions without having to react to them). The α coefficients for the internal consistency of these scales ranged from .75 to .91. 30 The FFMQ has been shown to be sensitive to changes in mindfulness levels with MBSR participation, 32 and its construct validity has been established in college students and experienced meditators. 42
Calgary Symptoms of Stress Inventory
The Calgary Symptoms of Stress Inventory (C-SOSI) 43 is a 56-item measure of the cognitive, behavioral, and physiological symptoms of stress. It is a shortened version of the 95-item Symptoms of Stress Inventory (SOSI). 44 Respondents are asked to rate how often they experienced a particular symptom during the past week using a 5-point Likert-type scale. The C-SOSI includes 1 overall score and 8 subscale scores. The subscales are peripheral manifestations of stress, cardiopulmonary arousal, upper respiratory symptoms, central neurological/gastrointestinal symptoms, muscle tension, depression, emotional irritability, and cognitive disorganization. Cronbach’s α is .95 for the overall score, with subscales ranging from .80 for upper respiratory symptoms to .92 for anger. 43
Profile of Mood States
The Profile of Mood States (POMS) 45 is a 65-item measure of positive and negative moods that has been used extensively in individuals with cancer. Respondents are asked to rate mood-related adjectives on a 5-point Likert-type scale based on how accurately the word describes how they have felt during the previous week. The POMS includes a total score and six subscales. The subscales are tension–anxiety, depression–dejection, anger–hostility, fatigue–inertia, vigor–activity, and confusion–bewilderment. Cronbach’s αs range from .80, for tension–anxiety, to .91, for depression–dejection. The POMS has demonstrated sensitivity to changes in mood over the course of psychosocial interventions for cancer patients. 46
Mindfulness-Based Stress Reduction program
The MBSR program is intended to assist individuals to cope with the emotional, psychological, and physical difficulties that can be associated with cancer diagnosis and treatment. The MBSR program was adopted and modified from its original design and use at the Massachusetts Medical Centre’s Stress Reduction and Relaxation Clinic2,47 to be more applicable to the needs of an oncology population. The program consists of 8 weekly sessions, and a 6-hour silent retreat that is held on the Saturday after the sixth session. At each class, participants learn about the mind–body connection, principles of mindfulness, and practice yoga movements in a group. Participants are also encouraged to share their experience of learning mindful meditation in order to collectively address difficulties in practice and generate support from the other group members. Participants are provided with guided meditation exercises on 2 compact discs and a program manual describing the weekly teachings and practices. Homework is assigned after each class and it is recommended that participants spend 45 minutes per day practicing meditation and mindful movement. A more detailed description of the MBSR program is available elsewhere. 48
Data Analysis
Of the 268 participants included in the current study, 177 completed the MAAS, POMS, and SOSI, whereas the other 91 participants completed the FFMQ, POMS, and C-SOSI at pre- and post-MBSR. The data from participants who completed the long form of the SOSI was transformed into the shorter form, the C-SOSI, to ensure consistency. The performed questionnaire transformation only applied the scoring criteria to the items included in the short form. Pre–post effect sizes (Cohen’s d) were calculated by dividing the difference of the pre- and post-program scores by the pooled standard deviation. Change scores were calculated by subtracting the pre-assessment scores from the post-assessment scores. The regression and correlational analysis were conducted using change scores.
Correlational analyses were performed to examine the relationship between the 5 scales of the FFMQ, the C-SOSI, and the POMS. Pairwise t tests were carried out to examine differences in pre- and post-program mindfulness scores on the MAAS, FFMQ, POMS, and C-SOSI. Simple regressions were conducted to determine what percentage of variance in POMS and C-SOSI scores was accounted for by changes in MAAS scores.
Stepwise multiple regressions were then conducted to establish the respective influence of the FFMQ subscales on variation in POMS and C-SOSI scores. We used stepwise regression, rather than forward or backward selection methods because there were no predetermined theoretical reasons to order the entrance of each independent variable. For this analysis, the independent variables were the change scores for the 5 subscales of the FFMQ. The dependent variables were the change scores from the POMS and the C-SOSI. An independent variable was included in the model if it was significant at accounting for variance in the dependent variable at a P < .05 level at the point of entry, and it was retained in the model if it sustained a P < .10 level of significance.
Results
Demographics
Table 1 shows the demographic characteristics of the participants. The majority of participants were women (84%) with an average age of 53.5 years. The most common cancer diagnosis was breast cancer (58.2%) followed by blood cancers (ie, lymphoma and leukemia, 7.8%) and colorectal cancer (6.0%), with an average time since diagnosis of 1.6 years (range = 0-24.7 years). Zero years since diagnosis reflects that the assessment took place close to the time of diagnosis for some participants.
Participant Demographics
Mindfulness
Table 2 summarizes the means and standard deviations for scores on the MAAS and the FFMQ, at pre- and post-program, and the statistical results of comparisons between the 2 assessments for each measure. Level of mindfulness as measured by the MAAS increased significantly over the course of the program (t = −6.82, P < .001). This increase reflected a moderate effect size, d = 0.46. All scales of the FFMQ also increased significantly over the course of the program. The effect size corresponding with these increases ranged from d = 0.48, for the describing scale to d = 0.79, for nonreactivity to inner experience scale.
Pairwise Comparisons of Mindfulness Scores and Mood and Stress Scores
Abbreviations: MBSR, mindfulness-based stress reduction; MAAS, Mindful Attention Awareness Scale; FFMQ, Five Facet Mindfulness Questionnaire; POMS, Profile of Mood States; C-SOSI, Calgary Symptoms of Stress Inventory.
Stress and Mood Outcomes
Statistically significant improvements were found in the total scores and all the subscale scores for stress and mood outcomes. Full results for the POMS and C-SOSI are presented in Table 2. The effect size for the POMS total score was d = 0.47. The effect sizes for the POMS subscales ranged from d = 0.24 for the vigor–activity subscale to d = 0.52 for the tension–anxiety subscale. The effect size for the total score of the C-SOSI was d = 0.57. Effect sizes for the C-SOSI subscales ranged from d = 0.24 for upper respiratory symptoms to d = 0.53 for emotional irritability.
Correlations
Table 3 summarizes the Pearson product-moment correlation coefficients between the FFMQ scales and the POMS and C-SOSI total and subscale change scores. The change scores for the nonjudging of inner experience scale (P < .001), acting with awareness scale (P < .01), and observe scale (P < .05) of the FFMQ were significantly correlated with the POMS total change score. The change scores for nonjudging of inner experience scale (P < .05) and acting with awareness scale (P < .01) of the FFMQ were significantly correlated with the C-SOSI total change score. The change scores for the nonreactivity to inner experience and the describe scales of the FFMQ were not related to any subscale or total change score for either the POMS or the C-SOSI.
Correlations Between Mindfulness, Stress, and Mood Change Scores (n = 91)
Abbreviations: FFMQ, Five Facet Mindfulness Questionnaire; POMS, Profile of Mood States; C-SOSI, Calgary Symptoms of Stress Inventory.
P < .05.
P < .01.
P < .001.
Relation Between Mindfulness, Stress, and Mood Outcomes
Table 4 summarizes the results for the simple regressions investigating the amount of change in stress and mood outcomes that were accounted for by changes in mindfulness, as measured by the MAAS. The MAAS change score accounted for 20.5% of the variance in total mood disturbance. Change in mindfulness using the MAAS accounted for the greatest amount of variance in the following 3 POMS subscale change scores: tension–anxiety (18.4%), confusion–bewilderment (17.2%), and depression–dejection (14.5%). The MAAS change score accounted for 14.4% of the variance in symptoms of stress. The 3 C-SOSI subscale change scores in which the MAAS change score accounted for the greatest amount of variance were depression (13.6%), emotional irritability (13.6%), and the cognitive disorganization (10.4%).
Simple Linear Regression Using MAAS, POMS, and C-SOSI Change Scores
Abbreviations: MAAS, Mindful Attention Awareness Scale; POMS, Profile of Mood States; C-SOSI, Calgary Symptoms of Stress Inventory; ns, nonsignificant.
Tables 5 and 6 summarize the results of the stepwise regressions investigating the amount of change in stress and mood outcomes that was accounted for by changes in mindfulness using the FFMQ. Of the five scales, only nonjudging of inner experience (POMS, C-SOSI), and acting with awareness (C-SOSI) were significant in accounting for variance in the mood and stress outcomes and were included in the model. The “nonjudging of inner experience” change score accounted for 14.8% of the change in total mood disturbance. The 3 POMS subscale change scores in which the nonjudging of inner experience change score accounted for the largest amount of variance were tension–anxiety (14.5% of the variance), confusion–bewilderment (14.4% of the variance), and depression–dejection (12.2% of the variance). The nonjudging of inner experience change score accounted for 4.8% of the variance in cardiopulmonary symptom change as measured by the C-SOSI.
Stepwise Multiple Regression Using FFMQ and POMS Changes Scores
Abbreviations: FFMQ, Five Facet Mindfulness Questionnaire; POMS, Profile of Mood States; States; ns, nonsignificant.
Stepwise Multiple Regression Using FFMQ and C-SOSI Change Scores
Abbreviations: FFMQ, Five Facet Mindfulness Questionnaire; C-SOSI, Calgary Symptoms of Stress Inventory; ns, nonsignificant
The acting with awareness scale accounted for 6.6% of the variance in reductions of total symptoms of stress as measured by the C-SOSI. The subscales of the C-SOSI in which the “acting with awareness” scale of the FFMQ change score accounted for the greatest amount of variance were peripheral manifestation (6.6%) and emotional irritability (4.5%).
Discussion
This study supports previous research demonstrating a reduction of cancer-related distress, including improved mood and stress symptoms, among individuals with cancer following participation in a MBSR program.10,11,15-18 Overall, there was a 55% decrease in mood disturbance and a 29% decrease in symptoms of stress. These results are very similar to our past research on stress and mood outcomes in individuals, with a variety of cancer diagnoses, participating in a MBSR program.12,15
As hypothesized, participation in the MBSR program resulted in improved mindfulness measured by the MAAS and all 5 scales of the FFMQ. Increases observed with the MAAS represent a medium effect size, similar to previous findings with both individuals with cancer16,35 and a nonclinical population. 33 On the FFMQ, a medium effect size was demonstrated for the describing, acting with awareness, and nonjudging of inner experience scales, and a large effect size was found for the observing and nonreacting to inner experience scales in this study.
The most important finding of this study was that self-reported psychological benefits were associated with increased levels of mindfulness, as assessed by both the MAAS and the FFMQ. Increases in mindfulness as measured by the MAAS accounted for improvements in total mood disturbance and symptoms of stress. On the subscales, the MAAS showed a stronger relationship to indicators of psychological than physical functioning on both the POMS and C-SOSI. Using the FFMQ, the nonjudging of inner experience scale accounted for improvement in mood, whereas acting with awareness accounted for improvement in stress symptoms, such as muscle tension. This provides support for the link between increased mindfulness and improved psychological health following MBSR participation. This is in accordance with other findings concluding that salutary impacts of MBSR are more apparent for psychological well-being rather than for physical health domains. 49
It would be interesting to compare our results to the relative contribution of the different facets of mindfulness in the study by Bränström et al. 17 However, this is not possible as the facets were reported only as one unidimensional score, which has not been recommended by the developers. 30 Thus, by showing differential associations of the FFMQ scales (rather than the total FFMQ score), stress, and mood outcomes, our study provides important implications for understanding how different facets of mindfulness are related to alleviation of distress.
The qualities of mindfulness, acting with awareness, and nonjudging of inner experience (as determined by effect sizes and/or regression analyses), were particularly important for a reduction of stress symptoms and aversive moods. This makes sense considering the training participants undergo in the program. People are first encouraged to pay attention to their thoughts and emotions without evaluating them. This mindfulness practice has been suggested to foster alternative ways to respond to negative emotional experiences. 50 In support of this argument, Labelle et al 16 found that improved depressive symptoms following MBSR were meditated by reduced ruminative thoughts among cancer patients. Taken together, the results of this study provide further evidence that participating in a MBSR program is related to increased awareness, a reduction in judgmental thoughts, and an improvement in psychological states.
Increases in mindfulness accounted for up to 20% of the improvements in stress and mood outcomes; however, there was a small or nonexistent relationship between some facets of mindfulness and the improvements in stress and mood outcomes. For instance, the nonreacting scale of the FFMQ showed no relation to any of the outcome measures in this study, despite a large effect size. This suggests that mindfulness training helps individuals increase awareness of stress, but may not immediately prevent them from reacting to stressful situations and subsequent experience of aversive moods. This particular facet may take longer to develop with continued mindfulness practice. This possibility is supported by Bränström et al, 17 wherein nonreacting was the only component of mindfulness to show a strong association with psychological improvement when assessed at 3 months following the completion of MBSR. Therefore, it may take more practice and time to cultivate the ability of not reacting to perceived psychological threat, which may not be measurable immediately after the completion of the program.
This research needs to be understood with the following limitations in mind. Most obviously, the absence of a control group prevents any causal attributions regarding the changes observed in stress, mood, and mindfulness. Second, this study included only participants who had completed both pre- and post-program assessment, a subsample of all those who initially began the program. Thus our results may be biased, by including only those who completed the MBSR program. Although this does not allow generalization to the population at large, it does strongly reflect the actual delivery and experience of participants who choose and complete this type of intervention. Last, our participants completed either the MAAS or the FFMQ but not both. Hence, it was not possible to examine the relationship between the 2 measures of mindfulness and between mindfulness and psychological outcomes in one sample instead of 2 separate samples.
In summary, this study demonstrated that cancer patients experience psychological benefits following participation of the MBSR program. Increases in mindfulness were related to improvements in psychological functioning to a greater degree than physical functioning. Being aware of the present moment and refraining from judging thoughts and emotions are important mindful skills for reducing stress and negative moods. This knowledge may help understand which components of mindfulness contribute to response and should be emphasized in the MBSR program.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Sheila Garland is funded by a doctoral research fellowship from the CIHR funded Psychosocial Oncology Research Training (PORT) initiative. Dr. Tamagawa is jointly funded by a postdoctoral award from PORT and Alberta Innovates Health Solutions (AIHS). Dr. Carlson holds the Enbridge Research Chair in Psychosocial Oncology, co-funded by the Alberta Cancer Foundation and the Canadian Cancer Society Alberta/NWT Division. She also holds an Alberta Heritage Foundation for Medical Research Health Scholar Award.