Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nurs Res. Author manuscript; available in PMC 2013 January 1.
Published in final edited form as:
PMCID: PMC3237795

Barriers to Meditation by Gender and Age Among Cancer Family Caregivers

Anna-leila Williams, PhD, PA, MPH, Peter Van Ness, PhD, MPH, Jane Dixon, PhD, and Ruth McCorkle, PhD, FAAN



Despite solid basic science research supporting meditation’s physiologic benefits, meditation remains a marginalized practice for many Westerners; observational and descriptive studies indicate a spectrum of barriers to meditation practice.


To determine differences in barriers to meditation by gender and age.


A cross-sectional survey study was conducted of 150 family caregivers to adults with cancer visiting an outpatient chemotherapy center in Connecticut, United States. The primary outcome was the Determinants of Meditation Practice Inventory. Explanatory variables included demographic characteristics, Center for Epidemiologic Studies Depression scale, Big Five Inventory, and Caregiver Reactions Assessment.


Participants included 98 women and 52 men. Age range was 18–84 years (M = 52.3 years). The highest frequency of barriers for both genders related to misconceptions about meditation. The total number of barriers to meditation did not significantly vary by gender (p = .10) nor age (p = .27). After adjusting for personality trait, reactions to caregiving, and emotional distress, gender (adjusted β = 0.81, SE = 1.70, p = .63) and age (adjusted β = 0.02, SE = 0.05, p = .67) still did not predict number of barriers to meditation. Backward elimination model building showed personality trait and reactions to caregiving account for 32% of the variability in barriers.


Total number of barriers to meditation were examined and a difference was not found by age or gender. It is possible differences by age and gender exist at the item level of evaluation, but were not evident when evaluating total scores. Further study is needed with samples large enough to have statistical power for item-level analysis.

Keywords: caregivers, meditation

National surveys indicate only a small segment of the United States population practices meditation (Barnes, Bloom, & Nahin, 2008). The most recent National Health Interview Survey (n = 23,393) found prevalence of use for meditation in the general adult population is approximately 9%. A large gender imbalance was noted in the Survey among users of mind-body therapies, including meditation, with 23.8% of women and 14.4% of men reporting use in the preceding year (Barnes et al., 2008). An assessment of mind-body therapy use among a national sample (n = 2,055) found similar practice patterns with approximately 7% of respondents saying they meditated (Wolsko, Eisenberg, Davis, & Phillips, 2004).

Physiologic responses to meditation have been documented for decades, with much of the evidence providing support for meditation as an effective tool in decreasing symptoms and biomarkers of stress (Delmonte, 1985; Fan, Tang, Ma, Posner, 2010; Fang et al., 2010; Infante el al., 2001; Jung et al., 2010). Differences in EEGs have been found, with meditators showing alpha and theta rhythm increases. In more recent studies, meditators showed higher gamma-band activity than nonmeditators (Lutz, Greischar, Rawlings, Ricard, & Davidson, 2004). Functional brain mapping of meditators showed both global and specific changes, with significant increases in areas that control the autonomic nervous system (Lazar et al., 2000). A follow-up study demonstrated a greater amount of grey matter among meditators (Lazar et al., 2005).

Meditation research in the clinical setting has shown ambiguous results. While there seems to be a general trend toward efficacy in the treatment of mood disturbance, insomnia, and impaired immune function, much of the research is marred by threats to validity that compromise confidence in the results (Ospina et al., 2007).

Despite solid basic science research supporting physiologic benefits, meditation remains a marginalized practice for most Americans. Anecdotal and observational reports describe challenges to meditation; a recent descriptive study identified a spectrum of barriers to meditation (Williams, Dixon, Van Ness, & McCorkle, 2011).

The purpose of this study was to examine whether barriers to meditation differ by gender and age among a sample of cancer family caregivers. Cancer family caregivers were chosen because they represent a highly stressed segment of the general population who would be likely to benefit from meditation practice. National data indicate family caregivers in the US are largely White (72%), married (58%), female (66%), ages 50–64 years (35%), and educated (29% with high school education or less; National Alliance for Caregiving, 2009). Of the estimated 65.7 million people who assume the role of family caregiver, approximately 4.6 million (7%) care for someone with cancer (National Alliance for Caregiving, 2009). Since more women than men currently practice meditation, it was hypothesized that women would identify fewer barriers to meditation than men. Because of increased popularization of meditation in recent years, it was hypothesized that younger people would identify fewer barriers to meditation than older people.


Study Design

A cross-sectional survey study was conducted at the Yale Comprehensive Cancer Center, Long Wharf Chemotherapy Center from May 2008 to March 2009. The Theory of Planned Behavior (TPB; Ajzen, 1991) provided the foundation for the study design and methods, including development of the study aims and selection of the variables of interest and outcome variables. The TPB explains individual behavior through serial stages: determinant, motivational, and volitional. Determinant variables include demographics and personality, which influence motivational variables: behavioral beliefs and attitude toward the behavior; normative beliefs and subjective norms; and control beliefs and perceived control or self-efficacy. The determinant and motivational variables influence the volitional variables: intention and behavior.

Inclusion and Exclusion Criteria

Men and women over the age of 18 years identified by an adult with cancer as a family caregiver were eligible to participate. Subjects had to speak English and be cognitively intact to the extent that they could give informed consent and understand, consider, and respond to survey questions. Language skills and cognition were evaluated grossly during recruitment. Individuals were excluded if they had prior experience with meditation (other than prayer).

Sample Size and Power Calculation

Power analysis for planned multiple regression analysis (Cohen, 1988) shows a sample size of 150 achieves 85% power to detect an R-square of 0.04 attributed to one independent variable (gender) using an F-test with a significance level of .05. Adjustment was made for eight additional independent variables (age, emotional distress, reactions to caregiving, and five personality traits--Agreeableness, Openness to New Experiences, Conscientiousness, Neuroticism, and Extraversion). The R-squared value is a conservative estimate based on results from a study assessing barriers to physical activity (Fleury, Lee, Matteson, & Belyea, 2004). While the behaviors of meditation and physical activity are not analogous, there are enough commonalities (such as influence of demographics, personality, behavioral beliefs, normative beliefs, subjective norms, and perceived self-efficacy) to allow the physical activity literature to provide a reference range for power analysis.


Patients with cancer presenting for chemotherapy were referred by the clinic nurse to research staff, who asked the patient to identify a family caregiver and provide contact information. The research staff contacted the family caregiver via telephone or at the chemotherapy center, explained the study in detail, and invited those who were eligible to participate. Written informed consent was attained prior to administration of the surveys.


Demographic information was collected on each participant. Personality traits were assessed using the Big Five Inventory (BFI), a 44-item scale measuring Agreeableness, Openness to New Experiences, Conscientiousness, Neuroticism, and Extraversion (John, Naumann, & Soto, 2008). The BFI has a 5-point Likert response format. It has shown ample internal consistency, temporal stability, and convergent and divergent validity (DeYoung, 2006; John et al., 2008). The BFI has been used previously in at least one study of cancer caregivers; Cronbach’s alphas range from 0.63 to 0.84 (Kim, Duberstein, Sorensen, & Larson, 2005).

The Caregiver Reactions Assessment (CRA; Given et al., 1992) a 24-item Likert scale, was used to assess the impact of caregiving responsibilities on daily activities and relationships. The CRA has been used extensively with cancer family caregiver populations with good internal consistency and content and construct validity testing (Nijboer, Triemstra, Tempelaar, Sanderman, & van den Bos, 1999; Stommel, Wang, Given, & Given, 1992). The range of Cronbach’s alphas for each of the 5 CRA subscales are: self-esteem (0.83–0.90), family support (0.74–0.85), impact on schedule (0.81–0.82), finances (0.81–0.83), and physical health (0.68–0.80).

The Center for Epidemiological Studies-Depression scale (CES-D) consists of 20 items across six major symptom areas. Each item is scored from 0 to 3, indicating frequency of occurrence of the symptom during the past week. Total scale score may range from 0 to 60, with a score of 16 or more indicating clinically significant depressive symptomatology (Radloff, 1977). Use of the CES-D with cancer family caregiver populations has shown internal consistency, typically around 0.90 (Kurtz, Kurtz, Given, & Given, 2004, 2005).

The Determinants of Meditation Practice Inventory (DMPI) is a 17-item Likert scale. The items fall within three domains: Perceptions and Misconceptions, Pragmatic Concerns, and Sociocultural Beliefs. The Perceptions and Misconceptions domain addresses the respondent’s understanding of the practice of meditation. Included are items pertaining to perceived physical, mental, and emotional constraints necessary to practice, as well as presumed outcomes of the practice. The Pragmatic Concerns domain addresses the respondent’s practical and technical barriers to practicing meditation. Included are items pertaining to the environment, time, priorities, and intrinsic and extrinsic motivation. The Sociocultural Beliefs domain addresses the respondent’s social and cultural barriers to practicing meditation. Included are items pertaining to religious beliefs, family and friend support, beliefs about appropriate interpersonal behaviors, and the supernatural.

The DMPI was developed and tested using a five-step, mixed-methods, standardized process: (a) delineate content domains; (b) develop operational definitions of the domains; (c) generate items; (d) use an expert panel to content validate the items; and (e) use a community-based sample to content validate and pilot test the items. The details of this process are reported elsewhere. Initial psychometric testing showed response variability for each item. The Cronbach’s coefficient alpha is 0.87. Item-total correlations range from 0.42–0.66 (Williams et al., 2011).


All analyses were conducted using the statistical program SAS version 9.1. Descriptive statistics regarding barriers to meditation were addressed via univariate analysis and tests for Gaussian distributions for the Determinants of Meditation Practice Inventory. T-tests were conducted to determine differences in number of barriers among the sample by age and gender. Age was dichotomized at the median for the purpose of conducting the t-test. Age data was continuous for all other analyses.

Number of barriers was calculated by summing positive responses obtained on the DMPI. Associations between the outcome variable (number of barriers) and explanatory variables (demographic characteristics, emotional distress, personality traits, and reactions to caregiving) were determined using an adjusted linear regression model. Multivariate statistical modeling was done and model fit was assessed with residual and diagnostic analyses to assure the assumptions of the model were met. Backward elimination of independent variables was conducted to find the most parsimonious model. Two-sided significance tests with p-values less than .05 were interpreted as statistically significant.

The Yale University School of Nursing Human Subjects Research Review Committee provided ethical approval for the study.


Sample Description

Among the sample of 150 caregivers, there were almost twice as many female as male participants (65.3% women, 34.7% men). Age range for the sample was 18–84 years (M = 52.3, SD = 16.2, median = 53.0). Participants predominately self-identified as White (83.3%). Almost two-thirds of participants reported at least some college education. Approximately 57% of participants were employed outside the home either full-time or part-time. Further details of the sample’s demographic characteristics can be seen in Table 1.

Table 1
Study Sample Demographic Characteristics (n = 150)

Univariate and Bivariate Statistics

No statistically significant difference was found in DMPI total score means by gender (p = .12) or by age (p = .23). Item-level differences were not evaluated for statistical significance secondary to lack of statistical power. Figure 1 is a graphic comparison of the percent of men and women who agreed or strongly agreed with each individual DMPI item to show similarities and differences by gender. For both men and women, the most frequently identified barrier to meditation was the use of prayer (item #8) and the least frequently identified barriers were conflict with religion (item #13) and concern about the harmful effects of meditation (item #17). Among items in the domain Perceptions and Misconceptions (items #1–8), a greater percentage of women than men agreed with the items with the exception of item #5 (“It might be boring.”) and item #6 (“It’s a waste of time to sit and do nothing.”), which had a higher percentage of men in agreement. In the domain Pragmatic Concerns (items #9–12), a greater percentage of women than men agreed with all items. This is in contrast to the domain Sociocultural Beliefs (items #13–17), in which a greater percentage of men than women agreed with all items.

Figure 1
Percentage of participants who “Agree” or “Strongly Agree” with an item by gender

Linear Regression Models

Tests of unadjusted associations between gender and barriers to meditation (p = .10) and age and barriers to meditation (p = .27) do not provide evidence of statistically significant differences. When adjustments are made for confounders (personality type: Extraversion, Agreeableness, Conscientiousness, Neuroticism, Openness to New Experiences; reactions to caregiving; and emotional distress), age and gender still do not predict barriers to meditation (Table 2). The R-Square for the adjusted model was 0.33, indicating the combination of the nine explanatory variables account for approximately 33% of variability in the DMPI total score. Residual and diagnostic analyses provided assurance that the assumptions of the model were met.

Table 2
Full Model Multiple Linear Regression Predicting Barriers to Meditation (n = 150)

Using a backward elimination approach to model building, all key variables were entered initially into the model and then least significant variables were eliminated one at a time. Listed in order of elimination, the variables that were removed were Extraversion, Agreeableness, age, and CES-D. The most parsimonious model included gender (the primary explanatory variable), Conscientiousness, Neuroticism, Openness to New Experiences, and reactions to caregiving (Table 3). The R-Square for the most parsimonious model was 0.32, indicating the combination of the five retained explanatory variables account for approximately 32% of the variability in the DMPI total score.

Table 3
Parsimonious Multiple Linear Regression Model Predicting Barriers to Meditation (n = 150)


The study’s sample population of 150 cancer family caregivers is similar to U.S. national samples on several criteria (National Alliance for Caregiving, 2009). Namely, the study sample is predominantly comprised of married women employed outside the home.

Since national data indicate that women use mind-body practices, including meditation, at a higher rate than men (Barnes et al., 2008), it was anticipated that women would present fewer barriers. Comparison of the DMPI total score means by gender shows no statistically significant differences. However, a visual comparison of individual items on the DMPI by gender shows considerable differences. These differences may be due solely to chance; however, the gender differences may have been obscured when only considering a total score. Items in the Pragmatic Concerns domain appear, in their entirety, as greater barriers to women than men. These results must be interpreted with caution until they are replicated with a larger sample.

It should be noted, overall, the domain Perceptions and Misconceptions has the highest frequency of barriers for both genders. This may indicate a prevailing lack of knowledge about meditation among the study sample. It appears that for meditation teachers or intervention researchers to overcome widespread fallacies and attract sizeable numbers of participants, recruitment procedures must describe the experience of meditation fully and clearly, in a way that is understandable to the target population.

The most parsimonious multiple linear regression model shows gender and Conscientiousness, Neuroticism, Openness to New Experiences (inversely), and reactions to caregiving account for 32% of the variability in the DMPI score. It seems reasonable that the negative emotions associated with Neuroticism trait would contribute to an individual’s perception of many barriers to a new behavior such as meditation. Individuals with high Conscientiousness trait are motivated to adhere to rules and conventions (John et al., 2008). Subsequently, without knowledge of meditation, a person with high Conscientiousness might perceive meditation as so unconventional as to present major barriers. It is possible that someone who is conscientious about his or her responsibilities as a caregiver will not want to deviate from an established pattern of activities by the addition of a new behavior. That said, some of the characteristics associated with the Conscientiousness trait, such as self-discipline and orderliness, would serve someone well who is embarking on a meditation practice. An individual who reports a high caregiver burden is likely to be overwhelmed at the prospect of adding something new to his or her daily activities, and therefore would perceive many barriers to meditation. It also seems likely that an individual high in the Openness to New Experiences trait characteristics would perceive few barriers to a new behavior and would likely welcome the opportunity to learn meditation.

The study is limited by the modest racial or ethnic variability represented among the study sample. Effort should be made to repeat the study with racial and ethnic groups who may have different barriers to meditation than the dominant population of cancer caregivers.


The study indicates there are barriers present for a sizeable percentage of both men and women of various ages. It is commonly understood by meditation teachers that many of the barriers queried on the DMPI can be modified effectively. For example, if members of a target population identify “I can’t sit still long enough to meditate” as a barrier, teachers or researchers can design recruitment materials that assure potential participants they can walk or lie down to meditate. If “There is no quiet place where I can meditate” is identified as a barrier, a teacher can describe the experience of meditating on ambient sound. Intervention studies are needed to determine if the barriers are indeed modifiable by these techniques.

Whether in a clinical or research setting, the DMPI may provide information to recruiters that will enable them to improve enrollment and minimize attrition. The DMPI can be used by researchers to understand (a) why some individuals choose to enroll and others do not; (b) why some participants respond to an intervention while others do not; and (c) how individuals with missing data (for instance, due to low attendance or dropouts) differ from those with complete data. Use of the DMPI, as described here, can be one step in a process to promote methodological rigor in meditation research and enhance participation in meditation interventions in the clinical setting.


This study was supported by the National Institute of Health, National Center for Complementary and Alternative Medicine (F31AT003535) and the Francisco J. Varela Contemplative Science Grant Award, Mind and Life Institute. Dr. Van Ness was supported in part by funding from the Claude D. Pepper Older Americans Independence Center at the Yale University School of Medicine (2P30 AG021342-06).

Contributor Information

Anna-leila Williams, Assistant Professor of Medicine, Quinnipiac University Medical School, Hamden, Connecticut.

Peter Van Ness, Research Scientist in Medicine (Geriatrics), Yale University School of Medicine and Co-Director, Biostatistics Core, Yale Program on Aging, New Haven, Connecticut.

Jane Dixon, Professor, Yale School of Nursing, New Haven, Connecticut.

Ruth McCorkle, Florence S. Wald Professor of Nursing, Yale School of Nursing and Professor of Chronic Disease Epidemiology, Yale School of Public Health, New Haven, Connecticut.


  • Ajzen I. The theory of planned behavior. Organizational Behavior and Human Decision Processes. 1991;50:179–211.
  • Barnes PM, Bloom B, Nahin RL. Complementary and alternative medicine use among adults and children: United States, 2007. National Health Statistics Reports. 2008;12:1–23. [PubMed]
  • Cohen J. Statistical power analysis for the behavioral sciences. 2. Hillsdale, NJ: Erlbaum; 1988.
  • Delmonte M. Biochemical indices associated with meditation practice: A literature review. Neuroscience and Biobehavioral Reviews. 1985;9:557–561. [PubMed]
  • DeYoung CG. Higher-order factors of the Big Five in a multi-informant sample. Journal of Personality and Social Psychology. 2006;91:1138–1151. [PubMed]
  • Fan Y, Tang YY, Ma Y, Posner M. Mucosal immunity modulated by integrative meditation in a dose-dependent fashion. Journal of Alternative and Complementary Medicine. 2010;16:151–155. [PubMed]
  • Fang CY, Reibel DK, Longacre ML, Rosenzweig S, Campbell DE, Douglas SD. Enhanced psychosocial well-being following participation in a mindfulness-based stress reduction program is associated with increased natural killer cell activity. Journal of Alternative and Complementary Medicine. 2010;16:531–538. [PMC free article] [PubMed]
  • Fleury J, Lee SM, Matteson B, Belyea M. Barriers to physical activity maintenance after cardiac rehabilitation. Journal of Cardiopulmonary Rehabilitation. 2004;24:296–305. [PubMed]
  • Given CW, Given B, Stommel M, Collins C, King S, Franklin S. The Caregiver Reaction Assessment for caregivers to persons with chronic physical and mental impairments. Research in Nursing & Health. 1992;15:271–283. [PubMed]
  • Infante J, Torres-Avisbal M, Pinel P, Vallejo JA, Peran F, Gonzalez F, Latre JM. Catecholamine levels in practitioners of the transcendental meditation technique. Physiology & Behavior. 2001;72:141–146. [PubMed]
  • John OP, Naumann LP, Soto CJ. Paradigm shift to the integrative Big Five trait taxonomy. In: John OP, Robins RW, Pervin LA, editors. Handbook of personality: Theory and research. 3. New York, NY: Guilford Press; 2008. pp. 114–158.
  • Jung YH, Kang DH, Jang JH, Park HY, Byun MS, Kwon SJ. The effect of mind-body training on stress reduction, positive affect, and plasma catecholamines. Neuroscience Letters. 2010;479:138–142. [PubMed]
  • Kim Y, Duberstein PR, Sorensen S, Larson MR. Levels of depressive symptoms in spouses of people with lung cancer: Effects of personality, social support, and caregiving burden. Psychosomatics. 2005;46:123–130. [PubMed]
  • Kurtz M, Kurtz J, Given C, Given B. Depression and physical health among family caregivers of geriatric patients with cancer--a longitudinal view. Medical Science Monitor. 2004;10:CR447–CR456. [PubMed]
  • Kurtz M, Kurtz J, Given C, Given B. A randomized, controlled trial of a patient/caregiver symptom control intervention: Effects on depressive symptomatology of caregivers of cancer patients. Journal of Pain and Symptom Management. 2005;30:112–122. [PMC free article] [PubMed]
  • Lazar S, Bush G, Gollub R, Fricchione G, Khalsa G, Benson H. Functional brain mapping of the relaxation response and meditation. Neuroreport. 2000;11:1581–1585. [PubMed]
  • Lazar SW, Kerr CE, Wasserman RH, Gray JR, Greve DN, Treadway MT, Fischl B. Meditation experience is associated with increased cortical thickness. Neuroreport. 2005;16:1893–1897. [PMC free article] [PubMed]
  • Lutz A, Greischar L, Rawlings N, Ricard M, Davidson R. Long-term meditators self-induce high-amplitude gamma synchrony during mental practice. Proceedings of the National Academy of Sciences of the United States of America. 2004;101:16369–16373. [PubMed]
  • National Alliance for Caregiving. Caregiving in the U.S. 2009 Retrieved from
  • Nijboer C, Triemstra M, Tempelaar R, Sanderman R, van den Bos GA. Measuring both negative and positive reactions to giving care to cancer patients: Psychometric qualities of the Caregiver Reaction Assessment (CRA) Social Science & Medicine. 1999;48:1259–1269. [PubMed]
  • Ospina MB, Bond TK, Karkhaneh M, Tjosvold L, Vendermeer B, Liang Y, Klassen TP. Evidence Report/Technology Assessment No 155 AHRQ Publication No 07-E010. 2007. Meditation practices for health: State of the research; pp. 1–263. [PubMed]
  • Radloff LS. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401.
  • Stommel M, Wang S, Given CW, Given B. Confirmatory factor analysis as a method to assess measurement equivalence. Research in Nursing & Health. 1992;15:399–405. [PubMed]
  • Williams AL, Dixon J, Van Ness PH, McCorkle RM. Determinants of Meditation Practice Inventory: development, content validation and initial psychometric testing. Alternative Therapies in Health and Medicine. 2011;17:16–23. [PubMed]
  • Wolsko P, Eisenberg D, Davis R, Phillips R. Use of mind-body medical therapies. Journal of General Internal Medicine. 2004;19:43–50. [PMC free article] [PubMed]