Bereavement Outcomes for Atheist Individuals: The Role of Spirituality,Discrimination,and Meaning

Abstract

The present study applied several concepts typically included in thanatology research to an atheist sample. Atheists are a growing segment of the population in the United States, though little is known about this group. A sample of 355 adults who self-identify as atheist completed an online survey assessing forms of spirituality, anti-atheist discrimination, and meaning reconstruction in order to examine associations between these variables and bereavement outcomes of complicated grief and psychological distress. Results of a multiple regression analysis suggested that spirituality was not related to bereavement outcomes, anti-atheist discrimination was related to poorer bereavement outcomes, while the relationship between meaning reconstruction and bereavement outcomes was mixed. These results provide a foundation for additional exploration of bereavement processes in atheist individuals, and implications for future research and practice are discussed.

Keywords

atheism bereavement spirituality discrimination meaning

The number of individuals in the United States (U.S.) who identify as religiously unaffiliated is growing rapidly (Pew Research Center, 2019). In addition to the growing number of those who are religiously unaffiliated, recent surveys suggest that around 26% (or around 80 million) individuals in the U.S. are atheists, which has been defined as a lack of belief in God or gods (Gervais & Najle, 2018; Hunsberger and Altemeyer, 2006). Despite this growth, empirical research on atheist and other nonbelieving individuals is lacking. An analysis of published research in social science journals from 2001 to 2012 found that only 100 included a focus on atheist individuals (Brewster et al., 2014). Of these, only one addressed aspects related to death, loss, and bereavement. Research focused on atheists and other nonbelievers has persisted in the years since the content analysis conducted by Brewster et al. (2014), though there is still a dearth of empirical literature focused on atheist bereavement. While one recent study suggested that belief or nonbelief in god(s) is a significant predictor of bereavement outcomes such as posttraumatic growth, complicated grief, and psychological distress (Sawyer & Brewster, 2019), little is known about the specific factors that influence these reactions. Therefore, this study sought to examine the relationships between concepts frequently found in thanatology research (i.e., spirituality, disenfranchised grief, and meaning reconstruction) and psychological outcomes during bereavement (i.e., complicated grief and psychological distress).

Associations Between Spirituality and Bereavement Outcomes

Although researchers often suggest that spirituality is a universal human phenomenon essential to mental well-being (e.g., Miller, 2015), a recent survey suggests that about 18% of the U.S. population consider themselves neither religious nor spiritual (Lipka & Gecewicz, 2017). The experiences of this group are frequently neglected in research focused on spiritual coping. Furthermore, an agreed upon definition of spirituality does not exist (see Oman, 2013 for a review). This has resulted in a wide variety of experiences falling under the umbrella of spiritual coping methods in prior research that include both religious and secular elements.

Researchers have noted also that there has been little research assessing the assumption that spirituality is helpful during bereavement (Wortmann & Park, 2008). In Wortmann and Park’s (2008) review of 73 empirical articles focused on religion and spirituality and their relationship with bereavement outcomes, they found seven articles that assessed spirituality, and six of these articles indicated that spirituality is positively associated with adjustment. Wortmann and Park also conclude that “relations between religion and adjustment to bereavement are generally positive but inconsistent and vary depending in part on how religion/spirituality is measured” (p. 703).

Even less is known about methods of coping during bereavement for atheist people in ways that are typically described as spiritual. One study on end-of-life preferences in an atheist sample found that meaning, connectedness to family and friends, and enjoyment of the natural world were the top three themes (Smith-Stoner, 2007). Thus, it is likely that atheist individuals cope with death and bereavement in ways that might be described as spiritual, even without a belief in god(s) or self-identification as spiritual due to the religious connotations of the term.

Associations Between Discrimination and Bereavement Outcomes

As many grief rituals are based in religious tradition, atheist individuals may experience their grief reactions to be outside of what is socially supported in their home or community environment. For example, atheist people would likely not believe they will see their loved ones again in heaven and they may be compelled to participate in religious events that are inconsistent with their preferred methods of coping of the loss. This may result in an increased likelihood that atheist individuals experience disenfranchised grief, which has been described as “the grief experienced by those who incur a loss that is not, or cannot be, openly acknowledged, publicly mourned, or socially supported” (Doka, 1999, p. 37). Researchers have examined a variety of characteristics that are related to disenfranchised grief in the past several decades. This includes bereavement following specific types of death such as substance use, cancer, HIV/AIDS, and suicide (Houck, 2007; Valentine et al., 2016), bereavement responses in LGBTQ+ individuals (Fenge, 2014; Wheat & Thacker, 2019), and pet loss (Cordaro, 2012). These examples offer a snapshot of the types of factors related to grief and bereavement that fall under the umbrella of disenfranchised grief but are not exhaustive.

Scholars have noted that disenfranchisement is likely broader and deeper than what has been typically thought, and includes reactions “not confined merely to the structural elements of bereavement or to grief understood in a global way; it can extend to every aspect or dimension of experience of bereavement and be applied to all of the dynamics of grief and mourning” (Corr, 1999, p. 17). There are currently no known studies that focus on the associations between experiences of discrimination and bereavement outcomes in atheist individuals, though previous studies have generally found negative mental health outcomes related to discrimination and other forms of marginalization towards atheist people in ways that are similar to other marginalized groups in the U.S. (e.g., Brewster et al., 2016; Hammer et al., 2012). Given these findings, identification as an atheist may be another demographic variable associated with more complicated bereavement outcomes if one’s grief responses are invalidated.

Associations Between Meaning and Bereavement Outcomes

One important task for bereaved individuals is meaning making after the loss of a loved one. While there are a number of ways that the bereaved go about this, a recently developed scale assessing grief and meaning reconstruction found five distinct factors that indicate whether one has made meaning after the loss (Gillies et al., 2015). Gillies et al. (2015) state:
… experiencing a sense of peace with the loss, reaffirming the bond with the deceased,appreciating life, and/or perceiving oneself as having grown through adversity are all indications that one has made meaning in the context of bereavement. In contrast, a sense of emptiness or meaninglessness serves as an indication that meaning has not been made or found in the grief experience (p. 69).
These five factors are further grounded in research examining the role of continued bonds and posttraumatic growth in bereavement outcomes, which have received increased attention in recent years by thanatology researchers. In terms of continued bonds, which has been defined as an ongoing attachment to the deceased that can take on a number of forms (Field et al., 2003), research findings have been inconclusive. For instance, one study examining adjustment to conjugal loss found that using possessions of the deceased for comfort was related to higher levels of distress, while thinking about fond memories of the deceased was related to less distress (Field et al., 1999). Another study found that continued bonds in the form of recovered memories of the deceased was a strong predictor of grief, but not depression, while use of the deceased’s possessions was not a strong predictor of grief or depression (Boelen et al., 2006). Posttraumatic growth has been defined as positive change that results from difficult or traumatic events (Tedeschi & Calhoun, 2004). Importantly, posttraumatic growth often occurs as a result of significant distress and higher levels of grief can be related to corresponding levels of growth (Bellet et al., 2018).

While past research has found that meaning making is an important concept for atheist individuals (Smith-Stoner, 2007), very few studies have examined this process. One recent study suggested that presence of meaning is associated with lower levels of complicated grief and psychological distress in bereaved atheist individuals (Sawyer & Brewster, 2019), however, the specific methods in which atheist individuals make meaning during bereavement is unknown at this time. The current study utilized the aforementioned measure of meaning reconstruction developed by Gillies et al. (2015) in order to gain additional insights into this process for atheist individuals.

The Present Study

This study was exploratory in nature with an attempt to better understand how variables from previous studies might apply to a bereaved atheist sample. Despite an exploratory approach, several hypotheses were made based on findings from prior studies. The first was that the three elements of spirituality that would likely be more applicable to an atheist sample (i.e., self spirituality, nature spirituality, and human spirituality) were expected to be related to lower levels of complicated grief and psychological distress (Hypothesis A). Based on the theory of disenfranchised grief, experiences of anti-atheist discrimination were expected to be related to higher levels of complicated grief and psychological distress (Hypothesis B). Finally, based on research related to meaning making and meaning reconstruction, a greater sense of peace and valuing life were expected to be related to lower levels of complicated grief and psychological distress, while higher levels of emptiness/meaninglessness and personal growth were expected to be related to higher levels of complicated grief and psychological distress (Hypothesis C). Due to the difficulty of predicting the impact of continued bonds (Field, 2006; Stroebe & Schut, 2005), no prediction was made for this variable.

Method

Participants

Participants included 355 adults living in the U.S. who ranged in age from 18 to 80 years old (M = 43.77, SD = 13.45). Approximately 68% of the sample identified as women, 30% as men, and 1% as another gender (e.g., androgynous, gender nonconforming, genderqueer). In terms of race, approximately 90% of the sample identified as White, 4% as multiracial, 2% as Hispanic/Latino/a, 1% as Native American/Indigenous American, 1% Asian American, 1% African American/Black, and 1% another race not listed. In terms of highest level of educational attainment, the sample was highly educated, as approximately 30% completed a four-year degree, 26% completed some college, 18% completed a master’s degree, 7% completed a doctorate, 14% completed a two-year degree, 6% completed high school degree, and 1% completed less than a high school degree. In addition to self-identifying as an atheist, participants were presented with several other categories to indicate self-identification and were instructed to select as many as were appropriate. Approximately 57% also self-identified as a humanist, 45% as secular, 35% as a freethinker, 23% as agnostic, and 12% as another identity representing their nonbelief (e.g., bright, anti-theist, Satanist, skeptic). About 88% identified as neither religious nor spiritual and 12% as spiritual but not religious. The amount of time that participants self-identified as an atheist ranged from one year to 66 years (M = 22.27, SD = 14.90). Furthermore, only about 1% of participants in this sample self-identified as an atheist for less than two years. Since participants were required to have experienced the death of a close friend or family member in the past two years to proceed with the survey items, very few participants in this sample identified as an atheist only after this most recent death. A summary of demographic information can also be found in Table 1.

Table 1.
Demographic Information.

Variable

Age M = 43.77 SD = 13.45

Race

White 90%

Multiracial 4%

Latina/o or Hispanic 2%

Asian American 1%

Black/African American 1%

Native/Indigenous American 1%

Another race not listed 1%

Gender

Woman 68%

Man 30%

Another gender not listed (e.g., “genderqueer”) 1%

Highest level of education

Four-year college degree 30%

Some college 26%

Master’s degree 18%

Two-year college degree 14%

Doctorate 7%

High school degree 6%

Less than high school degree 1%

Additional self-identified labels

Humanist 57%

Secular 45%

Freethinker 35%

Agnostic 23%

Another identity not listed (e.g., bright, anti- theist, Satanist, skeptic) 12%

Spiritual identification

Neither religious nor spiritual 88%

Spiritual but not religious 12%

Time self-identifying as atheist in years M = 22.27 SD = 14.90

Procedure

In order to participate, individuals had to confirm that they (a) were 18 years of age or older, (b) currently live in the U.S., (c) have experienced the death of a close friend or family member in the past 2 years, and (d) self-identify as an atheist. Participants were asked to provide informed consent and were able to continue with the survey after confirming they met all aspects of the inclusion criteria. Participants were informed that their participation will help to better understand how people perceive aspects of death and bereavement in nonbelievers and no other renumeration was provided. The survey was advertised as taking between 15–25 minutes to complete, was hosted by Qualtrics, and was accessed by computer or smartphone. The survey was approved by the Pennsylvania State University Institutional Review Board. Participants were recruited online via atheist listservs, social media sites (e.g., Twitter and Facebook pages with a large atheist following, e.g., “Friendly Atheist”), and snowball sampling techniques. A total of 376 participants responded to the initial survey. From there, 10 were removed after indicating that the death of a close friend or family member occurred more than two years ago, six were removed after missing more than one attention check item asking participants to mark a particular response (e.g., “Please select ‘Agree’”), four were removed after indicating they did not live in the U.S., and one was removed for omitting more than 20% of the survey items (Parent, 2013). This resulted in a final sample of 355 participants. Analyses were conducted with SPSS version 24.

Measures

Spirituality

Spirituality was assessed with three subscales of the 18-item Sources of Spirituality Scale (SOS Scale; Davis et al., 2015). The SOS Scale assesses “profiles of religious and secular spiritualities” (Davis et al., 2015, p. 512) using a 5-point Likert-type scale (1 = strongly disagree to 5 = strongly agree). The subscales used in this study included: Self spirituality (e.g., “I felt entirely authentic”), Nature spirituality (e.g., “I felt connected to nature), and Human spirituality (e.g., “I felt connected to all of humanity”). Two other subscales (i.e., Transcendent spirituality (e.g., “I had a sense of something infinite”) and Theistic spirituality (e.g., “I felt near to God”), were excluded from the analysis due to a primary focus on items that are likely more applicable to an atheist sample. Items were averaged to create three subscale scores, with higher scores indicating higher levels of endorsement for each category. Prior research has established the SOS Scale as a reliable and valid measure, with Cronbach’s α ranging from .90 to .99 across all subscales. Cronbach’s α for the Self, Nature, and Human subscales in the current sample were .89, .96, and .95, respectively.

Anti-Atheist Discrimination

Discrimination was assessed with the 24-item Measure of Atheist Discrimination Experiences (MADE; Brewster et al., 2016). Participants rated their experiences using a 6-point Likert-type scale (1 = never to 6 = almost all of the time). Example items include, “Because I am atheist, I have been asked how I can still have a purpose in life” and “I have been told that I will ‘grow out’ of my atheism and that it is just a phase.” Items were averaged, with higher scores indicating higher levels of discrimination. Prior research has established the MADE as a valid and reliable instrument, with Cronbach’s α levels of .94 and .95 in diverse samples of atheist people (Brewster et al., 2016). Cronbach’s α for the current study was .96.

Meaning

Meaning was assed using the 29-item Grief and Meaning Reconstruction Inventory (GMRI; Gillies et al., 2015). The GMRI assesses meaning specific to bereavement using a 5-point Likert-type scale (1 = strongly disagree to 5 = strongly agree) across five subscales. The subscales include: Continuing Bonds (e.g., “Memories of my loved one bring me a sense of peace and solace”), Personal Growth (e.g., “Since the loss, I’m more self-reflective”), Sense of Peace (e.g., “I was prepared for my loved one to die), Emptiness and Meaninglessness (e.g., “I can’t understand this loss), and Valuing Life (e.g., “I value and appreciate life more). Items were averaged to create five subscale scores, with higher scores indicating higher levels of endorsement for each category. The GMRI showed evidence of validity and reliability in past research, with Cronbach’s α ranging from .76 to .85 across all subscales. Cronbach’s α for the Continuing Bonds, Personal Growth, Sense of Peace, Emptiness and Meaninglessness, and Valuing Life subscales in this study were .71, .82, .78, .73, and .60, respectively.

Complicated Grief

Complicated grief reactions were assed with the 19-item Inventory of Complicated Grief (ICG; Prigerson et al., 1995). The ICG uses a 5-point Likert-type scale (0 = never to 4 = always) and items were averaged, with higher scores indicated greater levels of complicated grief reactions (e.g., “I find myself longing for the person who died”). Items can be summed to identify if a respondent is experiencing high levels of complicated grief. Prigerson et al. (1995) suggested that a score of 25 or above indicates high levels of complicated grief, while others have used a cut-off 30 or above (Shear et al., 2005). The ICG has been found to be a valid and reliable measure, with a Cronbach’s α of .94 on prior research with an atheist sample (Sawyer & Brewster, 2019). Cronbach’s α for the current study was .90.

Psychological Distress

Psychological distress was measured with the 21-item version of the Hopkins Symptom Checklist (HSCL-21; Green et al., 1988). The HSCL-21 uses a four-point Likert-type scale (1 = not at all to 4 = extremely) to measure symptoms of distress (e.g., “Feeling inferior to others”). Items were averaged, with higher scores indicating higher levels of distress. This measure has been found to be valid and reliable, with a Cronbach’s α of .91 in two separate studies with an atheist sample (Brewster et al., 2016, 2020). Cronbach’s α for the current study was .92.

Results

An analysis of the data found no evidence of multicollinearity (Mansfield & Helms, 1982). Additionally, results of Little’s Missing Completely at Random test was nonsignificant (p = .76) which indicates there were no patterns in the missing data (Li, 2013). Correlations between variables of interest can be found in Table 2. Cohen’s (1992) guidelines for small (r = .10), medium (r = .30) and large (r = .50) effects are used to describe the strength of relationships between variables. Approximately 15% of participants had a total complicated grief score of 25 or above, and 8% of participants had a scoree of 30 or above (M = 12.78, SD = 10.14). Both cut-off scores have been used in prior research to indicate high levels of complicated grief (Prigerson et al., 1995; Shear et al., 2005). Scores for psychological distress were similar to other studies assessing bereavement responses using the HCSL-21 (e.g., Hargrave et al., 2012).

Table 2.
Descriptive Statistics, Cronbach’s Alphas, and Bivariate Correlations of Variables of Interest.

Variables 1 2 3 4 5 6 7 8 9 10 11

1. Self spirituality –

2. Nature spirituality .43* –

3. Human spirituality .30* .44*** –

4. MADE .03 .02 .06 –

5. Continuing bonds .13* .05 .17 .00 –

6. Personal growth .15 .17 .25* .20* .34* –

7. Sense of peace .06 .02 .00 −.06 .09 −.08 –

8. Emptiness & meaninglessness .13* .04 .01 −.10 −.13* −.29* .46* –

9. Valuing life .18* .22* .29*** .14* .38* .66* −.05 −.18 –

10. Complicated grief −.02 .02 .07 .22* .24* .37* −.42* .73* .26* –

11. Psychological distress −.07 −.01 −.04 .18 −.02 .08 −.11* .39* −.07 .47* –

Alpha .89 .96 .95 .96 .71 .82 .78 .73 .60 .90 .92

Mean 3.84 3.60 2.59 2.20 3.27 3.12 3.22 3.58 3.66 .67 1.57

Standard deviation 1.06 1.26 1.18 .87 .57 .67 .97 .78 .62 .53 .48

Range 1–5 1–5 1–5 1–6 1–5 1–5 1–5 1–5 1–5 0–4 1–4

Note. MADE = Measure of Atheist Discrimination Experiences.

*
p < .05, p < .01, p < .001, N = 355.

Predictors of Complicated Grief

A multiple regression model was used with three subscales of spirituality (i.e., self spirituality, nature spirituality, and human spirituality), anti-atheist discrimination, and the five subscales of the GRMI (i.e., continuing bonds, personal growth, sense of peace, emptiness and meaninglessness, and valuing life) as predictors of complicated grief. Results of the multiple regression analysis suggested that the model statistically significantly predicted complicated grief, F(9,345) = 58.25, p < .001, adj R² = .59. Contrary to expectations of Hypothesis A, none of the spirituality subscales were related to complicated grief. In line with expectations of Hypothesis B, anti-atheist discrimination did emerge as a significant predictor, with higher levels of discrimination predicting higher levels of complicated grief. (p < .001). Four of the five subscales of the GMRI were significant predictors of complicated grief. As expected with Hypothesis C, personal growth and (p = .046) and emptiness and meaningless (p < .001) were related to higher levels of complicated grief, with the latter emerging as a particularly strong predictor, and a sense of peace was related to lower levels of complicated grief. Unexpectedly, the valuing life subscale was not a significant predictor. Although no a priori hypothesis was made for the continuing bonds subscale, higher scores also emerged as a predictor of higher levels of complicated grief (p = .001). See Table 3 for regression coefficients and standard errors.

Predictors of Psychological Distress

A second multiple regression model was used with three subscales of spirituality (i.e., self spirituality, nature spirituality, and human spirituality), anti-atheist discrimination, and the five subscales of the GRMI (i.e., continuing bonds, personal growth, sense of peace, emptiness and meaninglessness, and valuing life) as predictors of psychological distress. Results of the multiple regression analysis suggested that the model statistically significantly predicted psychological distress, F(9,345) = 9.92, p < .001, adj R² = .19. Contrary to expectations of Hypothesis A, none of the spirituality subscales were related to psychological distress. Consistent with Hypothesis B, anti-atheist discrimination emerged as a significant predictor, with higher levels of discrimination predicting higher levels of psychological distress. (p < .001). Two of the five subscales of the GMRI were significant predictors of psychological distress. As expected with Hypothesis C, emptiness and meaningless (p < .001) was related to higher levels of psychological distress, again as a particularly strong predictor. The valuing life subscale also emerged as a significant predictor in this model, with higher levels related to less psychological distress (p = .003). Contrary to expectations, personal growth and sense of purpose were not related to psychological distress. Although no a priori hypothesis was made for the continuing bonds subscale, it was not a significant predictor. See Table 4 for regression coefficients and standard errors.

Table 3.
Regression Model Predicting Outcomes of Complicated Grief.

Variable B
95% CI for B SE B* β R ² Adj. R ² F df

LL UL

Model .60 .59 58.25 9,345

Constant 2.51* 2.16 2.87 .18

Self spirituality .02 −.02 .05 .02 .03

Nature spirituality .00 −.03 .04 .02 .00

Human spirituality .00 −.03 .04 .02 .01

MADE .08* .04 .12 .02 .13*

Continuing bonds .12 .05 .19 .04 .13**

Personal growth .08* .00 .15 .04 .10*

Sense of peace −.08 −.12 −.03 .02 −.14

Emptiness & meaninglessness .42* −.48 −.37 .03 .62*

Valuing life .01 −.07 .09 .04 .01

Note. MADE = Measure of Atheist Discrimination Experiences.

*
p < .05, p < .01, p < .001, N = 355.

Table 4.
Regression Model Predicting Outcomes of Psychological Distress.

Variable B
95% CI for B SE B* β R ² Adj. R ² F df

LL UL

Model .20 .19 9.92 9,345

Constant 2.64* 2.19 3.08 .23

Self spirituality .00 −.05 .05 .03 .00

Nature spirituality .02 −.03 .06 .02 .04

Human spirituality −.01 −.05 .04 .02 −.02

MADE .08* .03 .14 .03 .15*

Continuing bonds −.02 −.11 .07 .05 −.03

Personal growth .05 −.05 .14 .05 .07

Sense of peace .05 −.01 .10 .03 .10

Emptiness & meaninglessness .27* −.34 −.20 .04 .44*

Valuing life −.15 −.25 −.05 .05 −.20**

Note. MADE = Measure of Atheist Discrimination Experiences.

*
p < .05, p < .01, p < .001, N = 355.

Discussion

There is very little research related to bereavement processes for atheist individuals (Brewster et al., 2014) and this study is one of the first to apply thanatology concepts to an exclusively atheist sample. As atheist individuals often report various forms of stigma and discrimination (Brewster et al., 2016; Frazer et al., 2020; Hammer et al., 2012), results of this study can be useful for both researchers and mental health professionals for gaining a better understanding of how anti-atheist discrimination is related to bereavement outcomes, and, despite some assumptions to the contrary, how atheist individuals engage in meaning reconstruction after loss. This study also adds to a small, but growing body of research focused on death, dying, and bereavement in atheists and other nonbelievers.

The first hypothesis predicted that aspects of spirituality that may be applicable to atheist individuals (i.e., self spirituality, nature spirituality, and human spirituality) would be related to lower levels of complicated grief and psychological distress during bereavement. This hypothesis was not supported for either dependent variable. While there are no known studies linking these variables with bereavement outcomes in atheist individuals, past research has found that end-of-life preferences for atheist people include aspects of meaning and connectedness that are typically considered to be under the umbrella of spirituality (Smith-Stoner, 2007). Furthermore, the Sources of Spirituality Scale was chosen for this study based on its ability to assess for secular elements of spirituality (Davis et al., 2015). Nonsignificant findings for these variables could be due to the general lack of importance of spirituality for this sample (e.g., 88% identified as not religious or spiritual). Even so, it should be noted that those without spiritual beliefs do not necessarily lack the ability to cope effectively during bereavement, as prior studies have found that atheist individuals reported less complicated grief and psychological distress compared to those endorsing a belief in god(s) (Sawyer & Brewster, 2019).

In terms of anti-atheist discrimination, the hypothesis that greater discrimination would be associated with higher levels of complicated grief and psychological distress was supported. Although there were no prior studies linking these specific variables, these results correspond to those experiencing forms of what has been termed disenfranchised grief in prior research (e.g., Cordaro, 2012; Fenge, 2014; Houck, 2007; Valentine et al., 2016; Wheat & Thacker, 2019). People experience disenfranchised grief when their loss cannot be publicly mourned or supported (Doka, 1999). Atheist individuals who experience the loss of a loved one may lack outlets for validating their beliefs related to loss, particularly in strongly religious families or communities. As atheist writers have noted, those without a belief in god or gods do not believe their loved ones are in a better place or that they will see them again (Christina, 2014), and implicit or explicit messages to the contrary are likely detrimental to atheist people during times of bereavement. This would be consistent with findings that suggest many atheist people find that their lack of belief brings shame to others, are asked to pass as religious, and experience social ostracism (Brewster et al., 2016).

The final hypothesis related to meaning reconstruction and bereavement outcomes was partially supported. As expected, personal growth was positively associated with complicated grief, a sense of peace was negatively associated with complicated grief, valuing life was negatively related to psychological distress, and emptiness and meaninglessness was a strong predictor of higher levels of both complicated grief and psychological distress. Continuing bonds emerged as a significant predictor of complicated grief, though a priori* hypotheses were not made for this variable due to conflicting findings in past research (e.g., Boelen et al., 2006; Field et al., 1999). Findings related to the personal growth subscale are consistent with prior studies that suggest growth is at times positively related to grief and distress (Bellet et al., 2018). As a particularly strong predictor of complicated grief and psychological distress, the role of emptiness and meaninglessness appears to be an important variable that warrants additional exploration, especially in light of the nonsignificant findings of the spirituality subscales. Researchers and mental health professionals who focus solely on “spirituality” are likely missing the mark for many individuals for whom this term does not apply, including 18% of the U.S. population who identify as neither religious nor spiritual (Lipka & Gecewicz, 2017). This misapplication may be due to terminology (e.g., not identifying as spiritual) or conceptual factors (e.g., issues with conflating spirituality with religiosity). Furthermore, as the operational definition of spirituality varies widely, more secular terms such as meaning may be more relevant for atheist and other nonbelievers.

Limitations and Directions for Future Research

Conclusions from this study must be interpreted with consideration of several limitations. The first is that this was an exploratory study with a cross-sectional design. This does not allow for any causal interpretations of relationships between predictor and outcome variables. This is particularly relevant for the outcome variable of psychological distress. Since questions from this scale are not specific to specific changes due to bereavement, it is unclear to what extent the death of a close friend or family member influenced general distress in this sample without pre- and post-measures of this variable. Another limitation is the use of an exclusively online survey for data collection, limiting responses from those without internet access. A third limitation is that participants were recruited from atheist group listservs and social media pages with large memberships that may differ from a random sample of atheist individuals. In other words, individuals who would be more likely to come in contact with the survey may have a stronger identification to an atheist identity that someone who is not a follower of these listservs or social media pages. Another limitation is that the sample was predominately White. While this is typical of research focused on atheist individuals, researchers, advocates, and authors of atheist literature have increasingly called for an inclusion of the experiences of atheists of color (Abbott et al., 2020; Hutchinson, 2020). Future studies related to atheist bereavement should seek to highlight these experiences. Finally, this sample consisted only of participants from the U.S. in order to reduce confounding variables that might exist for discrimination based on the experiences of atheist people in other countries. This limits the generalizability of results to locations outside of the U.S.

Despite these limitations, this study is one of the first to examine variables typically included in thanatology research with an exclusively atheist sample. A primary goal of this study is to provide a foundation of data that can be used for a more conceptually and statistically refined examination of atheist reactions during bereavement. One way that future research can continue to examine this is by utilizing longitudinal study designs to examine changes in complicated grief and psychological distress over time to provide evidence for more causal connections between variables of interest. Longitudinal designs would also help to assess the degree to which psychological distress is a direct result of the loss, which cannot be assessed in this study. Additionally, while the concept of disenfranchised grief provides a solid conceptual rationale for the association between anti-atheist discrimination poorer bereavement outcomes, it is unclear what other factors are related to this. Future research could examine other potential mediating and moderating variables to obtain a better understanding of how and for whom bereavement processes are more complicated. Since bereavement is an experience that is unique to each individual, qualitative analyses could be utilized to assess themes in atheist individuals who have recently experienced the death of a loved one. Finally, since the term “spirituality” is nebulous and about 18% of those in the U.S. consider themselves neither religious nor spiritual (Lipka & Gecewicz, 2017; Oman, 2014), other measures examining specific existential or relational concepts using terminology that does not conflate meaning in life with religion and spirituality could be more appropriate.

Variable
Age	M = 43.77 SD = 13.45
Race
White	90%
Multiracial	4%
Latina/o or Hispanic	2%
Asian American	1%
Black/African American	1%
Native/Indigenous American	1%
Another race not listed	1%
Gender
Woman	68%
Man	30%
Another gender not listed (e.g., “genderqueer”)	1%
Highest level of education
Four-year college degree	30%
Some college	26%
Master’s degree	18%
Two-year college degree	14%
Doctorate	7%
High school degree	6%
Less than high school degree	1%
Additional self-identified labels
Humanist	57%
Secular	45%
Freethinker	35%
Agnostic	23%
Another identity not listed (e.g., bright, anti- theist, Satanist, skeptic)	12%
Spiritual identification
Neither religious nor spiritual	88%
Spiritual but not religious	12%
Time self-identifying as atheist in years	M = 22.27 SD = 14.90

Variables	1	2	3	4	5	6	7	8	9	10	11
1. Self spirituality	–
2. Nature spirituality	.43***	–
3. Human spirituality	.30***	.44***	–
4. MADE	.03	.02	.06	–
5. Continuing bonds	.13*	.05	.17**	.00	–
6. Personal growth	.15**	.17**	.25***	.20***	.34***	–
7. Sense of peace	.06	.02	.00	−.06	.09	−.08	–
8. Emptiness & meaninglessness	.13*	.04	.01	−.10	−.13*	−.29***	.46***	–
9. Valuing life	.18***	.22***	.29***	.14*	.38***	.66***	−.05	−.18**	–
10. Complicated grief	−.02	.02	.07	.22***	.24***	.37***	−.42***	.73***	.26***	–
11. Psychological distress	−.07	−.01	−.04	.18**	−.02	.08	−.11*	.39***	−.07	.47***	–
Alpha	.89	.96	.95	.96	.71	.82	.78	.73	.60	.90	.92
Mean	3.84	3.60	2.59	2.20	3.27	3.12	3.22	3.58	3.66	.67	1.57
Standard deviation	1.06	1.26	1.18	.87	.57	.67	.97	.78	.62	.53	.48
Range	1–5	1–5	1–5	1–6	1–5	1–5	1–5	1–5	1–5	0–4	1–4

Variable	B	95% CI for B	SE B	β	R ²	Adj. R ²	F	df
Model						.60	.59	58.25	9,345
Constant	2.51***	2.16	2.87	.18
Self spirituality	.02	−.02	.05	.02	.03
Nature spirituality	.00	−.03	.04	.02	.00
Human spirituality	.00	−.03	.04	.02	.01
MADE	.08***	.04	.12	.02	.13***
Continuing bonds	.12**	.05	.19	.04	.13**
Personal growth	.08*	.00	.15	.04	.10*
Sense of peace	−.08**	−.12	−.03	.02	−.14**
Emptiness & meaninglessness	.42***	−.48	−.37	.03	.62***
Valuing life	.01	−.07	.09	.04	.01

Variable	B	95% CI for B	SE B	β	R ²	Adj. R ²	F	df
Model						.20	.19	9.92	9,345
Constant	2.64***	2.19	3.08	.23
Self spirituality	.00	−.05	.05	.03	.00
Nature spirituality	.02	−.03	.06	.02	.04
Human spirituality	−.01	−.05	.04	.02	−.02
MADE	.08***	.03	.14	.03	.15***
Continuing bonds	−.02	−.11	.07	.05	−.03
Personal growth	.05	−.05	.14	.05	.07
Sense of peace	.05	−.01	.10	.03	.10
Emptiness & meaninglessness	.27***	−.34	−.20	.04	.44***
Valuing life	−.15**	−.25	−.05	.05	−.20**

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jacob S. Sawyer

Author Biography

Jacob S. Sawyer, PhD is an assistant professor of psychology at the Pennsylvania State University, Mont Alto campus. His research interests include experiences of nonreligious/atheist individuals; psychological factors related to death, dying and bereavement; and college student mental health.

References

Abbott

D. M.

Mollen

Mercier

Anaya

E. J.

Rukus

V. A.

(2020). “Isn't atheism a white thing?”: Centering the voices of atheists of color. Journal of Counseling Psychology, 67(3), 275–287. https://doi.org/10.1037/cou0000399

Bellet

B. W.

Jones

P. J.

Neimeyer

R. A.

McNally

R. J.

(2018). Bereavement outcomes as causal systems: A network analysis of the co-occurrence of complicated grief and posttraumatic growth. Clinical Psychological Science, 6(6), 797–809. https://doi.org/10.1177/2167702618777454

Boelen

P. A.

Stroebe

M. S.

Schut

H. A. W.

Zijerveld

A. M.

(2006). Continuing bonds and grief: A prospective analysis. Death Studies, 30(8), 767–776. https://doi.org/10.1080/07481180600852936

Brewster

M. E.

Hammer

Sawyer

J. S.

Eklund

Palamar

(2016). Perceived experiences of atheist discrimination: Instrument development and evaluation. Journal of Counseling Psychology, 63(5), 557–570. https://doi.org/10.1037/cou0000156

Brewster

M. E.

Robinson

M. A.

Sandil

Esposito

Geiger

(2014). Arrantly absent: Atheism in psychological science from 2001 to 2012. The Counseling Psychologist, 42(5), 628–663. https://doi.org/10.1177/0011000014528051

Brewster

M. E.

Velez

B. L.

Geiger

E. F.

Sawyer

J. S.

(2020). It’s like herding cats: Atheist minority stress, group involvement, and psychological outcomes. Journal of Counseling Psychology, 67(1), 1–13. https://doi.org/10.1037/cou0000392

Christina

(2014). Comforting thoughts about death that have nothing to do with god. Pitchstone Publishing.

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155–159. https://dx-doi-org.web.bisu.edu.cn/10.1037/0033-2909.112.1.155

Cordaro

(2012). Pet loss and disenfranchised grief: Implications for mental health counseling practice. Journal of Mental Health Counseling, 34(4), 283–294. https://doi.org/10.17744/mehc.34.4.41q0248450t98072

10.

Corr

C. A.

(1999). Enhancing the concept of disenfranchised grief. OMEGA—Journal of Death and Dying, 38(1), 1–20. https://doi.org/10.2190/LD26-42A6-1EAV-3MDN

11.

Davis

D. E.

Rice

Hook

J. N.

Van Tongeren

D. R.

DeBlaere

Choe

Worthington

E. L.

(2015). Development of the sources of spirituality scale. Journal of Counseling Psychology, 62(3), 503–513. https://doi.org/10.1037/cou0000082

12.

Doka

K. J.

(1999). Disenfranchised grief. Bereavement Care, 18(3), 37–39. https://doi.org/10.1080/02682629908657467

13.

Fenge

L.-A.

(2014). Developing understanding of same-sex partner bereavement for older Lesbian and gay people: Implications for social work practice. Journal of Gerontological Social Work, 57(2–4), 288–304. https://doi.org/10.1080/01634372.2013.825360

14.

Field

N. P.

(2006). Continuing bonds in adaptation to bereavement: Introduction. Death Studies, 30(8), 709–714. https://doi.org/10.1080/07481180600848090

15.

Field

N. P.

Gal-Oz

Bonanno

G. A.

(2003). Continuing bonds and adjustment at 5 years after the death of a spouse. Journal of Consulting and Clinical Psychology, 71(1), 110–117. https://doi.org/10.1037/0022-006X.71.1.110

16.

Field

N. P.

Nichols

Holen

Horowitz

M. J.

(1999). The relation of continuing attachment to adjustment in conjugal bereavement. Journal of Consulting and Clinical Psychology, 67(2), 212–218. https://doi.org/10.1037/0022-006x.67.2.212

17.

Frazer

El-Shafei

Gill

A. M.

(2020). Reality check: Being nonreligious in America. American Atheists.

18.

Gervais

W. M.

Najle

M. B.

(2018). How many atheists are there? Social Psychological and Personality Science, 9(1), 3–10. https://doi.org/10.1177/1948550617707015

19.

Gillies

J. M.

Neimeyer

R. A.

Milman

(2015). The grief and meaning reconstruction inventory (GMRI): Initial validation of a new measure. Death Studies, 39(2), 61–74. https://doi.org/10.1080/07481187.2014.907089

20.

Green

D. E.

Walkey

F. H.

McCormick

I. A.

Taylor

A. J. W.

(1988). Development and evaluation of a 21-item version of the Hopkins symptom checklist with New Zealand and United States respondents. Australian Journal of Psychology, 40(1), 61–70. https://doi.org/10.1080/00049538808259070

21.

Hammer

J. H.

Cragun

R. T.

Hwang

Smith

J. M.

(2012). Forms, frequency, and correlates of perceived anti-atheist discrimination. Secularism and Nonreligion, 1, 43. https://doi.org/10.5334/snr.ad

22.

Hargrave

P. A.

Leathem

J. M.

Long

N. R.

(2012). Peritraumatic distress: Its relationship to posttraumatic stress and complicated grief symptoms in sudden death survivors. Journal of Traumatic Stress, 25(3), 344–347. https://doi.org/10.1002/jts.21703

23.

Houck

J. A.

(2007). A comparison of grief reactions in cancer, HIV/AIDS, and suicide bereavement. Journal of HIV/AIDS & Social Services, 6(3), 97–112. https://doi.org/10.1300/J187v06n03_07

24.

Hunsberger, B., & Altemeyer, B. (2006). Atheists: A groundbreaking study of America's nonbelievers. Prometheus Books.

25.

Hutchinson

(2020). Humanists in the hood: Unapologetically Black, feminist, and heretical. Pitchstone Publishing.

26.

(2013). Little’s test of missing completely at random. Stata Journal, 13(4), 795–809.

27.

Lipka

Gecewicz

(2017). More Americans now say they’re spiritual but not religious. Pew Research Center. https://www.pewresearch.org/fact-tank/2017/09/06/more-americans-now-say-theyre-spiritual-but-not-religious/

28.

Mansfield, E. R., & Helms, B. P. (1982). Detecting multicollinearity. The American Statistician, 36(3a), 158--160. https://doi.org/10.1080/00031305.1982.10482818

29.

Miller

(2015). The spiritual child: The new science on parenting for health and lifelong thriving. St. Martin’s Press.

30.

Oman

(2013). Defining religion and spirituality. In Paloutzian

R. F.

Park

C. L.

(Eds.), Handbook of the psychology of religion and spirituality (2nd ed., pp. 23–48). Guilford Press.

31.

Parent

M. C.

(2013). Handling item-level missing data: Simpler is just as good. The Counseling Psychologist, 41(4), 568–600. https://doi.org/10.1177/0011000012445176

32.

Pew Research Center. (2019, October 17). In U.S., decline of Christianity continues at rapid pace. https://www.pewforum.org/2019/10/17/in-u-s-decline-of-christianity-continues-at-rapid-pace/

33.

Prigerson

H. G.

Maciejewski

P. K.

Reynolds

C. F.

Bierhals

A. J.

Newsom

J. T.

Fasiczka

Frank

Doman

Miller

(1995). Inventory of complicated grief: A scale to measure maladaptive symptoms of loss. Psychiatry Research, 59(1–2), 65–79. https://doi.org/10.1016/0165-1781(95)02757-2

34.

Sawyer

J. S.

Brewster

M. E.

(2019). Assessing posttraumatic growth, complicated grief, and psychological distress in bereaved atheists and believers. Death Studies, 43(4), 224–234. https://doi.org/10.1080/07481187.2018.1446061

35.

Shear

Frank

Houck

P. R.

Reynolds

C. F.

(2005). Treatment of complicated grief: A randomized controlled trial. Journal of the American Medical Association, 293(21), 2601–2608. https://doi.org/10.1001/jama.293.21.2601

36.

Smith-Stoner

(2007). End-of-life preferences for atheists. Journal of Palliative Medicine, 10(4), 923–928. https://doi.org/10.1089/jpm.2006.0197

37.

Stroebe

Schut

(2005). To continue or relinquish bonds: A review of consequences for the bereaved. Death Studies, 29(6), 477–494. https://doi.org/10.1080/07481180590962659

38.

Tedeschi

R. G.

Calhoun

L. G.

(2004). Posttraumatic growth: Conceptual foundations and empirical evidence. Psychological Inquiry, 15(1), 1–18. https://doi.org/10.1207/s15327965pli1501_01

39.

Valentine

Bauld

Walter

(2016). Bereavement following substance misuse: A disenfranchised grief. OMEGA—Journal of Death and Dying, 72(4), 283–301. https://doi.org/10.1177/0030222815625174

40.

Wheat

L. S.

Thacker

N. E.

(2019). LGBTQ+ loss experiences and the use of meaning reconstruction with clients. Journal of LGBT Issues in Counseling, 13(3), 232–251. https://doi.org/10.1080/15538605.2019.1627973

41.

Wortmann

J. H.

Park

C. L.

(2008). Religion and spirituality in adjustment following bereavement: An integrative review. Death Studies, 32(8), 703–736. https://doi.org/10.1080/07481180802289507