Mindfulness, Experiential Avoidance, and Affective Experience in Older Adults

Abstract

Subjective well-being is comprised of two components: affective experience and life satisfaction. Older adulthood is marked by changes and challenges that have the potential to diminish well-being. As the proportion of older adults in the population rises, it is becoming increasingly important to identify factors that may be protective against these potentially detrimental events. Mindfulness and acceptance constructs have been shown to be associated with affective experience across the lifespan. However, little work has examined which facets of mindfulness drive these associations. The current study aimed to explore the associations between acceptance and mindfulness processes and positive and negative affect in older adults. Eighty-five healthy older adults completed self-report measures of mindfulness, experiential avoidance, and positive and negative affect. Results from linear regression analyses indicated that the act with awareness and nonreactivity facets of mindfulness were particularly important in their contribution to positive affect. Additionally, higher levels of experiential avoidance accounted for significant variance in negative affect. These findings help to elucidate how mindfulness and acceptance processes play a role in affective experience in older adults. Future studies should explore these facets in clinical populations to help create more targeted clinical goals.

Older adulthood is a time when individuals are faced with many new and evolving challenges, and these encounters can impact one’s psychological health. For example, declines in physical health, loss of social support, and cognitive deterioration could potentially contribute to diminished well-being (Blazer & Hybels, 2005). With a rapidly aging population, it is important to acknowledge the impact these changes may have on overall well-being in older adults, and to better understand factors that may protect and enhance well-being in this population (Bruce, 2002). The construct of subjective well-being is typically defined in the literature by two components: affective experience and life satisfaction (Diener, Oishi, & Lucas, 2003). A developing idea in the literature is that greater mindfulness and lower experiential avoidance may play an essential role in maintaining and enhancing both of these components of well-being across young adult and geriatric populations (Andrew & Dulin, 2007; Fiocco & Mallya, 2015; Machell, Goodman, & Kashdan, 2015; Mahoney, Segal, & Coolidge, 2015). The aim of this preliminary study was to expand upon previous work by exploring how specific facets of mindfulness and experiential avoidance may influence affective experience in a geriatric population.

Mindfulness is commonly defined as being fully in touch with, and accepting of, one’s internal experiences (Brown & Ryan, 2003). Mindfulness is often distinguished into five interrelated facets including: observe, describe, act with awareness, nonjudgment, and nonreactivity (Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006). In one meta-analysis, mindfulness measures were associated with affective experience in adult populations (r = .34 with positive affectivity; r = −.39 with negative affectivity; Giluk, 2009). More recently, Fiocco and Mallya’s work with older adults further emphasized the link between mindfulness and both affective and cognitive well-being (2015). They demonstrated that higher trait mindfulness was associated with fewer depressive symptoms and greater self-reported quality of life (Fiocco & Mallya, 2015). Others have found that mindfulness mediates the relationship between age and affect (Raes, Bruyneel, Loeys, Moerkerke, & De Raedt, 2013; Shook, Ford, Strough, Delaney, & Barker, 2017). Additionally, mindfulness-based interventions have been successful in promoting positive outcomes for older adults (Splevins, Smith, & Simpson, 2009; Young & Baime, 2010). Increases in emotional well-being and decreases in psychological distress and levels of depression have all been observed following mindfulness-based interventions (Splevins et al., 2009; Young & Baime, 2010). Furthermore, one intervention study found that changes in facets of mindfulness play an important role in older adult outcomes, with increased levels of act with awareness and nonjudgment facets leading to lower levels of depression (Splevins et al., 2009).

Experiential avoidance (EA), also referred to as psychological inflexibility, can be characterized as the opposite of mindfulness and acceptance. This construct refers to attempts to alter the duration, frequency, or intensity of internal experiences (e.g., thoughts, emotions), even when doing so results in detrimental outcomes (Hayes, Wilson, Gifford, Follette, & Strosahl, 1996). Through attempts to suppress internal experiences, increased frequency of these experiences often paradoxically ensues (Hayes et al., 1996). Additionally, while continually engaging in attempts to control or avoid, individuals may lose contact with values-based positive reinforcers in their lives. Work by several research groups has demonstrated EA to be important in predicting symptoms of depression and anxiety in older adults (Andrew & Dulin, 2007; Mahoney et al., 2015). Other work in college-aged students found inverse links between well-being and EA, with higher levels of daily EA predicting lower daily positive affect (PA), meaning in life, and enjoyment of activities, and higher levels of daily negative affect (NA; Machell et al., 2015). The current study expands upon this previous work by exploring how EA may be associated with varying levels of PA and NA, an important component of well-being, in an older adult population.

Previous research examining the associations between mindfulness, EA, and components of well-being has largely examined these constructs in younger, rather than older, adults. Importantly, previous studies that have investigated the associations between mindfulness and subjective well-being in older adults have not examined whether specific mindfulness facets are particularly important in these associations (e.g. Fiocco & Mallya, 2015; Giluk, 2009; Raes et al., 2013). A more comprehensive understanding of which facets of mindfulness are associated with affective experience may prove useful in promoting well-being outcomes. This exploratory and preliminary study sought to expand upon the limited research in this area to examine the associations between mindfulness facets, EA, and affect in a sample of 85 older adults.

Material and Methods

Participants

One-hundred and fourteen healthy, community-dwelling older adults were recruited through local senior citizen organizations, churches, and synagogues, as part of a larger ongoing study. The broader study examined cognitive functioning in healthy older adults, thus, participants aged 60 and older were screened for a series of neurological events, major surgeries, and medications that could affect cognitive functioning, and for a history of significant primary psychiatric disease (after Tranel, Benton, & Olson, 1997). Of the initial 114 participants contacted, nine were excluded during the screening process and 20 did not complete all necessary procedures, resulting in a total sample of 85 participants (Mean = 74.74 years; SD = 5.44 years; Median = 75.0 years; range = 60-88 years; 52.9% female). Sociodemographic and sample characteristics are displayed in Table 1.

Table 1.

Descriptive statistics, N = 85.

Sex
Male, N (%)	40 (47.1%)
Female, N (%)	45 (52.9%)
Age, M (SD)	74.74 (5.44)
Median	75.00
Years of Education, M (SD)	16.32 (3.00)
Race
White, N (%)	85 (100%)
Non-White (%)	0 (0%)
Ethnicity
Not Hispanic or Latino, N (%)	75 (88.2%)
Hispanic or Latino, N (%)	1 (1.2%)
Ethnicity not reported, N (%)	9 (10.6%)
Current marital status
Married, N (%)	57 (67.1%)
Widowed, N (%)	13 (15.3%)
Divorced, N (%)	7 (8.2%)
Other (Single, Partnership), N (%)	5 (5.9%)
Status not reported, N (%)	3 (3.5%)
AAQ-II Total, M (SD)	13.08 (6.32)
FFMQ-Observe, M (SD)	26.34 (5.46)
FFMQ-Describe, M (SD)	29.47 (6.26)
FFMQ-Act with Awareness, M (SD)	29.74 (5.09)
FFMQ-Nonjudgment, M (SD)	30.63 (5.94)
FFMQ-Nonreactivity, M (SD)	24.30 (4.33)
PANAS-Positive Affect, M (SD)	33.09 (7.20)
PANAS-Negative Affect, M (SD)	11.80 (2.65)

Note: for FFMQ-Nonreactivity, N = 83 due to missing data points. PANAS-PA/NA = Positive and Negative Affective Schedule-Expanded Form. AAQ-II = Acceptance and Action Questionnaire, 2^nd edition. FFMQ = Five Facet Mindfulness Questionnaire.

Procedure

As a part of this larger ongoing study, participants completed two separate visits. During Visit #1, participants provided informed consent and were screened using a semi-structured interview to ensure they met the aforementioned inclusion criteria. Next, they completed a neuropsychological evaluation that lasted approximately two hours. During Visit #2, the PANAS-X (Watson & Clark, 1994) and several other self-report personality and affect measures were completed. Following both visits, packets containing the self-report mindfulness and experiential avoidance assessments were mailed to the participants’ homes. These assessments were completed at home and were sent back via mail in pre-paid envelopes. Individuals were provided compensation for their participation, and all study procedures were approved by the University of Iowa’s Institutional Review Board.

Mindfulness

The Five Facet Mindfulness Questionnaire (FFMQ; Baer et al., 2006) was used to measure five mindfulness processes including: observe (attending to internal experiences), describe (labeling internal experiences), act with awareness (deliberately choosing one’s actions in the moment), nonjudgment (not judging one’s thoughts or emotions), and nonreactivity (allowing thoughts to come and go freely without reacting). This 39-item inventory uses a 5-point Likert scale with higher scores on each subscale indicating greater levels of mindfulness. The validation study reported good convergent and discriminant validities, and adequate internal consistencies on all subscales (α = .75-.91; Baer et al., 2006). Reliabilities for this sample were acceptable to excellent (observe: α = .797; describe: α = .918; act with awareness: α = .872; nonjudgment: α = .877; nonreactivity: α = .737).

Experiential Avoidance

The Acceptance and Action Questionnaire II (AAQ-II; Bond et al., 2011) was used to measure experiential avoidance. The AAQ-II is a 7-item measure that utilizes a 7-point Likert scale to create a total possible score (7-49), with higher scores indicating greater avoidance. The measure has demonstrated good test-retest reliability at three and twelve months (Bond et al., 2011). Adequate evidence was found for both convergent and discriminant validity, as well as excellent internal reliability across various samples (mean alpha coefficient = .84; Bond et al., 2011). Cronbach’s alpha in this sample was excellent (α = .922).

Positive and Negative Affect

The Positive and Negative Affective Schedule-Expanded Form (PANAS-X; Watson & Clark, 1994) is a 60-item instrument used to measure positive and negative affect (PA/NA), which comprises the affective experience component of subjective well-being. Participants are asked to rate the extent to which they have experienced various emotions in the past few weeks. Items are rated on a 5-point Likert scale with higher scores indicating higher levels of their respective scales. The PANAS-X demonstrates convergent and discriminant validity, moderate test-retest reliability, and good to excellent internal consistency, with PA and NA alpha coefficients reported at .83-.90 and .79-.93, respectively (Watson & Clark, 1994). Alphas for this sample were .875 for PA, and .775 for NA (Cronbach’s alpha is based on 84/85 participants as item level data was not available for one subject).

Statistical Analyses

Missing data on mindfulness measures were handled using person-mean imputation when less than 20% of data was missing on a given subscale. When greater than 20% of items on a given subscale were missing (2+ items), data were excluded from any analyses that included that subscale. Two participants’ data were excluded due to missing data.

All analyses were performed using IBM SPSS Statistics for Windows, version 25 (IBM Corp., Armonk, N.Y., USA). First, we examined potential confounders of the associations of interest. Bivariate correlations between age and PA and between age and NA were examined. A Mann-Whitney test was used to examine if PA or NA varied by sex. Next, bivariate correlations between all mindfulness facets, experiential avoidance, and PA and NA were examined. Then, individual linear regressions with each variable of interest were used to predict PA and NA. The facets that accounted for significant variance in the respective affective outcome variables were then used to construct models using both stepwise and backward linear regressions to explore the relative contributions of experiential avoidance and facets of mindfulness in predicting PA and NA.

Results

Preliminary analyses

Preliminary analyses indicated that age was not significantly associated with positive affect (PA) or negative affect (NA; ps > .05). A Mann-Whitney test indicated that PA and NA were not significantly different for males and females (ps > .05). Thus, age and sex were not entered as covariates in subsequent analyses. A correlation matrix displaying correlations between all predictor (experiential avoidance and mindfulness facets) and outcome (PA and NA) variables is displayed in Table 2. To avoid the problem of multicollinearity, we examined variance inflation factors (VIFs). All VIFs were under five, which was deemed appropriate for this study. In regard to power to detect effects, based on statistical estimates, the analyses would be able to detect small to moderate effects.

Table 2.

Correlations between measures, N = 85.

	1	2	3	4	5	6	7	8
1. PANAS-PA
2. PANAS-NA	−.202
3. AAQ-II	−.235*	.460***
4. FFMQ-Observe	.133	.093	.173
5. FFMQ-Describe	.311**	−.079	−.264*	.225*
6. FFMQ-Act with awareness	.315**	−.240*	−.409***	−.098	.375***
7. FFMQ-Nonjudgment	.097	−.216*	−.598***	−.265*	.219*	.338**
8. FFMQ-Nonreactivity	.276*	−.074	−.225*	.351**	.508***	.103	.200

Note.

^*p < .05;

^**p < .01;

^***p <. 001.

PANAS-PA/NA = Positive and Negative Affective Schedule-Expanded Form. AAQ-II = Acceptance and Action Questionnaire, 2^nd edition. FFMQ = Five Facet Mindfulness Questionnaire.

Positive Affect

Individual linear regressions indicated that describe, act with awareness, nonreactivity, and experiential avoidance (EA) all accounted for significant variance in positive affect (PA). Thus, only these variables were entered into the stepwise and backward regression models. In the stepwise and backward regressions, act with awareness and nonreactivity remained significant predictors of PA in the final model, with higher levels of act with awareness and nonreactivity being associated with higher levels of PA (t = 2.794, Std. β = .288, p = .007 and t = 2.388, Std. β = .246, p= .019, respectively). For both regressions, these variables accounted for 15.8% of the total variability in PA (R² = .158). It might be noted as surprising that the describe facet was not a significant predictor in the final model of PA, even though it was more significantly associated with the outcome variable than nonreactivity (r = .311, p = .004 versus r = .276, p = .012). Attention, however, should be drawn to the significant correlation between these two variables, r = .508, p < .001. Given the strength of this association, it can be assumed that when entered into the regression analyses, the describe facet was not able to significantly account for additional variance in PA, above and beyond that which nonreactivity was able to account for. This was confirmed by running an additional regression analysis where describe and act with awareness were entered as predictor variables, in which case the final model was still significant (p = .002), but accounted for less total variance in PA than the original model (R² = .142 versus R² = .158).

Negative Affect

Individual linear regressions indicated that experiential EA, act with awareness, and nonjudgment all accounted for significant variance in negative affect (NA). Thus, only these variables were entered into further regression analyses. Stepwise and backward linear regressions indicated that EA was the only variable that remained significant in predicting NA in the final model, with higher levels of EA being associated with higher levels of NA (t = 4.723, Std. β = .460, p < .001). In total, EA accounted for 21.2% of the variability in NA (R² = .212).

Discussion

The present study explored the associations between mindfulness facets, experiential avoidance (EA), and positive and negative affect (PA/NA) in a sample of healthy older adults using linear regression analyses. Results indicated that the act with awareness and nonreactivity facets of mindfulness appeared to be particularly important in their associations with PA, while levels of EA accounted for significant variance in NA.

This study highlights that taking a more mindful, nonreactive, and accepting stance toward internal experiences may be associated with greater subjective well-being. Our findings are consistent with a previous study of young and middle-aged adults, which revealed comparable correlations between mindfulness (as measured by a total general mindfulness score on the Mindful Attention Awareness Scale) and PANAS PA and NA (r = .250 and r= −.355, respectively) (Taren, Creswell, & Gianaros, 2013). Our work extends that of Taren and colleagues in two ways. First, our study utilized a sample of older adults, which is group of individuals relatively understudied in the literature. Additionally, we explored how EA and specific facets of mindfulness are related to affective experience. Because these constructs are likely interrelated yet distinct, our work elucidates potential differential process-based targets for a clinical setting. In particular, acting with awareness is characterized by deliberately choosing and attending to one’s actions, while nonreactivity is defined as letting thoughts and emotions come and go freely without reacting. In line with these findings, it may be that when one is mindfully aware of the moment and takes a nonreactive stance toward internal experiences, chosen actions are more likely to be in line with an individual’s goals and values. Thus, it is possible that positive affectivity may be more likely to ensue. This would be comparable with previous literature suggesting that valued living may often result in greater PA (Ryan & Deci, 2001). Additionally, purposefully remaining present and aware while engaging in these values-based actions increases the likelihood of experiencing the reinforcing qualities of the actions themselves, again potentially resulting in enhanced affective experience. This further supports previous work demonstrating significant associations between increased levels of act with awareness and lower levels of depression (Splevins et al., 2009). Additionally, the finding that higher levels of EA were associated with higher levels of NA echoes previous work associating psychological inflexibility with adverse psychological outcomes (Craske, Miller, Rotunda, & Barlow, 1990; Hayes et al., 2004; Machell et al., 2015). As previous research suggests, it may be that actively trying to avoid negative emotions leads to a paradoxical increase in those same emotions (Hayes et al., 1996).

Although this preliminary study provides rich insight into the associations between mindfulness, avoidance, and affective experience, some important limitations should be acknowledged. First, the sample size was relatively small and largely homogenous, which may limit generalizability. It is possible that with our sample size we were not able to detect smaller effects for some of the other mindfulness facets that were not found to be significant predictors of PA and NA in the current analyses. Additionally, because this study was correlational in nature, the directionality and causality of the associations of interest cannot be definitively determined. Finally, self-report data introduces the potential for bias. Given that the literature on older adults and mindfulness is in its infancy, this work should be replicated in more diverse and larger sample sizes. Despite these limitations, these results are noteworthy in that they give insight into how experiential avoidance and specific facets of mindfulness are associated with PA and NA.

As a whole, this study shows that higher levels of mindfulness and acceptance may play an important role in promoting enhanced affective experience. This research helps further explain which factors of mindfulness have the most influence in these associations, which allows for more targeted clinical goals. Specifically, this work suggests that interventions should aim to increase an older individual’s ability to remain aware and nonreactive while choosing actions, in addition to limiting attempts to avoid unpleasant internal experiences. Nonetheless, this is the first study of its kind to examine these constructs in an older adult sample, so replication is necessary to validate these results. Future work should also aim to explore these constructs and related clinical interventions in specific clinical subpopulations of older adults, such as those struggling with anxiety and depression. Older adulthood can bring about a host of new challenges that have the potential to diminish subjective well-being. Taken together, this study elucidates the importance of cultivating higher levels of mindfulness and acceptance to enhance affective experiences in older adults in the midst of these challenges.

Table 3:

Stepwise regression analyses: Role of experiential avoidance and mindfulness processes in predicting positive and negative affect.

Dependent variables	Significant predictors	β (SE)	Std. β	t-value	Significance	R2	Adjusted R2	F	95% Confidence Interval for β
Positive Affect	Act with awareness	.405 (.145)	.288	2.794	.007	.158	.137	8.834**	(.146, .736)
	Nonreactivity	.406 (.170)	.246	2.388	.019				(.068, .745)
Negative Affect	Experiential avoidance	.193 (.041)	.460	4.723	<.001	.212	.202	22.302***	(.112, .274)

Note:

^*p < .05;

^**p < .01;

^***p < .001.

^[1]For the regressions, only predictor variables that were significant predictors (p<.05) of positive or negative affect on their own in individual linear regressions were utilized.

^[2]Stepwise and backward linear regressions produced the same final models.

Highlights.

• Acceptance and mindfulness processes are related to affective experience in older adults

• Acting with awareness and nonreactivity are associated with positive affect

• Experiential avoidance contributes significantly to negative affect