The Utility of the Pandemic Grief Scale in Identifying Functional Impairment from COVID-19 Bereavement

Abstract

Background:

Meeting the needs of people bereaved by COVID-19 poses a substantial challenge to palliative care. The Pandemic Grief Scale (PGS) is a 5-item mental health screener to identify probable cases of dysfunctional grief during the pandemic.

Objective:

The PGS has strong psychometric and diagnostic features. The objective was to examine the incremental validity of the PGS in identifying mourners at risk of harmful outcomes.

Design:

A cross-sectional survey design involving sociodemographic questions and self-report measures of pandemic grief, generalized anxiety, depression, post-traumatic stress, separation distress, functional impairment, meaning-making difficulties, and substance use coping.

Setting/Subjects:

A sample of people bereaved through COVID-19 (N = 1065) in the United States.

Results:

Fully 56.6% of participants scored above the cut score of ≥7 on the PGS for clinically dysfunctional pandemic grief and 69.7% coped with their loss using drugs or alcohol for at least several days in past two weeks. PGS scores were not associated with time since loss. Hierarchical multiple regression models demonstrated that the PGS uniquely explained variance in functional impairment, meaning-making difficulties, and substance use coping, over relevant background factors, bereavement-related psychopathology, and separation distress. In the final model, the standardized regression coefficients for the PGS were 2–15 times larger than for the other competing measures in explaining each of the three outcomes.

Conclusions:

The findings underscore the clinical utility of this short and easy-to-use measure in identifying risk of deleterious outcomes across a range of functional and behavioral domains.

Introduction

Over 2.4 million people around the world have died from COVID-19.¹ Every death during this pandemic affects approximately nine people.² The unique circumstances associated with the COVID-19 pandemic, such as physical isolation from terminally ill family members and sources of support, restricted mourning and funeral practices, and economic hardships, underscore predictions that a greater number of people bereaved by this disease will experience preoccupying and protracted grief after the loss.^3–6 Recent research has shown that acute grief during the pandemic is more severe compared with acute grief from deaths before the pandemic.⁷ Furthermore, grief after a COVID-19 death is indeed more severe than grief after death from other natural causes.⁸ In a study of 307 people bereaved from COVID-19, almost two-thirds (63.2%) report clinically significant functional impairment.⁹ Clearly, the unique needs of mourners during the pandemic pose a substantial challenge to palliative care.^6,9

The Pandemic Grief Scale (PGS) is a short screener designed to identify probable cases of dysfunctional grief due to a COVID-19 loss.¹⁰ It was developed in a large sample (N = 831) of adults in the United States who had a close person die from COVID-19 and was shown to measure a unified COVID-19 grief construct, and was internally consistent (α = 0.86) and factorially invariant across gender, age, and race. The PGS also has solid diagnostic properties (e.g., sensitivity = 87%, specificity = 71%, and area under the curve = 0.87)¹⁰ that are comparable or superior to other psychiatric screeners.^11–13

The purpose of this study is to examine the incremental validity of the PGS by following established approaches.^14,15 To demonstrate incremental validity, the PGS should uniquely associate with multiple independent indicators of functioning (social/work impairments, difficulty making meaning of the loss, and substance use coping) above and beyond relevant participant characteristics (sociodemographics, loss-related factors, and COVID-19 diagnosis), bereavement-related psychopathology (i.e., generalized anxiety, depression, and post-traumatic stress), and separation distress symptoms (i.e., yearning, sorrow, preoccupation with deceased, and preoccupation with the death). Separation distress represents a core characteristic of disturbed grief.¹⁶

Materials and Methods

Participants and procedure

Ethics approval was granted by the Christopher Newport University Institutional Review Board. Online survey data from 1065 U.S. adults were collected in November 2020, yielding more than sufficient statistical power. Participants were recruited through Amazon MTurk in exchange for payment (USD $0.50) and were eligible if they experienced the death of a close person from COVID-19. MTurk provides data equivalent in quality to other methods of data collection, especially when data are screened appropriately.¹⁷ Only participants who furnished complete information were included in the final sample and all provided informed consent.

Measures

Bereavement-related psychopathology

Clinical symptoms of depression and generalized anxiety were measured using the Patient Health Questionnaire-4.¹⁸ Post-traumatic stress symptoms were measured using the National Stressful Events Survey Post-Traumatic Stress Disorder (PTSD) Short Scale.¹⁹ Participants indicated how frequently they experienced symptoms of depression with two items (e.g., “feeling down, depressed, or hopeless”; α = 0.74), generalized anxiety with two items (e.g., “feeling nervous, anxious, or on edge”; α = 0.72), and post-traumatic stress with nine items (e.g., “feeling jumpy or easily startled when you hear an unexpected noise?”; α = 0.92).

Separation distress

Separation distress symptoms of grief were measured with the first four items (e.g., “I felt a constant longing or yearning for the deceased”; α = 0.81) of the Persistent Complex Bereavement Inventory, which has established strong reliability as well as factorial and construct validity in previous research.^20,21

COVID-19 grief

COVID-19 grief was measured with the 5-item PGS (PGS; e.g., “I wished to die in order to be with the deceased”; α = 0.89), an efficient screening tool that identifies dysfunctional grief due to a COVID-19 loss with strong reliability as well as factorial and construct validity.¹⁰

Indicators of functioning

Functional impairment due to the loss was measured using the Work and Social Adjustment Scale.²² Participants were asked to rate how much impairment they experienced because of their COVID-19 loss with five items (e.g., “Because of this loss, my ability to work is impaired”; α = 0.94). Difficulty making meaning of the loss was measured using the six items (e.g., “This loss is incomprehensible to me”; α = 0.89) of the Integration of Stressful Life Experiences Scale-Short Form.²³ Substance use coping was measured by a single-item scale, “I used alcohol or other drugs to help me get through this loss.”

A 4-point time-anchored score (0 = not at all to 3 = nearly every day) that spanned the previous two weeks was applied to all of the measures except for the Work and Social Adjustment Scale,²² which used a 9-point severity scale (0 = not at all to 8 = very severely) and the Integration of Life Experiences Scale-Short Form,²³ which used a 5-point scale of agreement. Item scores within a measure were summed to form composite scores, with higher scores indicating higher levels of each construct.

Sociodemographic information

Participants were asked to report their age, gender, race, COVID-19 diagnosis, relationship with a significant person in their life who died from COVID-19, how long ago this person died, and whether the participant received professional help for this loss.

Statistical analyses

Descriptive statistics were used to characterize the sample. Independent t-tests and analyses of variance (ANOVA) were planned to determine differences in COVID-19 grief according to sociodemographic characteristics. However, due to violations of the assumption of homogeneity of variance in the relevant variables, Welch's t-tests were used in place of the planned t-tests. Pearson correlations were used to explore associations between measures. Three separate hierarchical multiple regression models were tested to detect the proportion of unique variance associated with COVID-19 grief for each model (i.e., functional impairment due to the loss, difficulty making meaning of the loss, and use of alcohol or drugs to cope with the loss). An initial screening of the data revealed no problems associated with singularity, multicollinearity, dependence of errors, violation of normality, linearity, or homoscedasticity of residuals. Based on the preliminary analyses, the following background and deceased-related variables were dummy coded and used as control variables in the first block: race (1 = White; 0 = non-White), gender (1 = men; 0 = women and other); diagnosis (1 = COVID-19 diagnosis; 0 = no COVID-19), relationship (1 = immediate family and romantic relationship; 0 = other), and help (1 = received professional help for loss; 0 = no help). The second block added the differential diagnostic variables of generalized anxiety, depression, and post-traumatic stress to the model. The third block added separation distress to the model. The last block added COVID-19 grief to the model.

Results

Approximately three quarters of the participants (M_age = 39.44) were White (76.4%), slightly more than half were male (56.3%), and over one-third had themselves been diagnosed with COVID-19 (34.9%). Most were within three months of bereavement (76.8%), and nearly 4 in 10 sought professional help for their loss (38.3%; Table 1). Overall, the sample was generally composed of individuals who were distressed and having difficulties adjusting to their loss (Table 2). Specifically, the total score means for anxiety, depression, COVID-19 grief, and functional impairment were just below clinically significant levels of distress.^10,18,21 Fully 56.6% of the sample scored above the cut score of ≥7 on the PGS for clinically dysfunctional pandemic grief established by previous research¹⁰ and 69.7% of the sample coped with their loss using drugs or alcohol for at least several days in the past two weeks.

Table 1.

Sample Characteristics (N = 1065)

Characteristics	n (%)
Gender
Male	600 (56.3%)
Female	463 (43.5%)
Other	2 (0.2%)
Race
White	814 (76.4%)
Black	118 (11.1%)
Asian	74 (6.9%)
Hispanic	49 (4.6)
Other	10 (0.9%)
COVID-19 diagnosis
No	693 (65.1%)
Yes	372 (34.9%)
Relationship with the deceased
Extended family	333 (31.3%)
Close friend	217 (20.4%)
Immediate family	209 (19.6%)
Acquaintance	187 (17.6%)
Other	10 (0.9%)
Time since loss in months
<1	99 (9.3%)
1	219 (20.6%)
2	289 (27.1%)
3	211 (19.8%)
4	122 (11.5%)
5	81 (7.6%)
6 or more months	44 (4.1%)
Professional help for loss
No	657 (61.7%)
Yes	408 (38.3%)
	Mean (SD)
Age	39.44 (11.75)

Table 2.

Zero-Order Correlations and Descriptive Statistics for Modeled Variables (N = 1065)

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1. Age	—
2. Race	0.21^***	—
3. Gender	−0.05	−0.08^*	—
4. Diagnosis	−0.05	−0.02	0.07^*	—
5. Time	0.03	−0.01	0.04	−0.06	—
6. Relationship	0.07^*	0.05	0.02	0.35^***	−0.03	—
7. Help	0.01	0.05	0.00	0.36^***	0.04	0.25^***	—
8. Depression	−0.03	0.05	0.03	0.32^***	0.02	0.27^***	0.29^***	—
9. Generalized anxiety	−0.02	0.08^*	−0.03	0.30^***	−0.01	0.24^***	0.27^***	0.70^***	—
10. Post-traumatic stress	0.00	0.06^*	0.05	0.39^***	0.03	0.36^***	0.38^***	0.77^***	0.74^***	—
11. Separation distress	0.06	0.04	0.02	0.29^***	−0.01	0.34^***	0.31^***	0.54^***	0.52^***	0.70^***	—
12. COVID-19 grief	0.01	0.05	0.10^**	0.38^***	0.04	0.39^***	0.41^***	0.61^***	0.56^***	0.79^***	0.76^***	—
13. Functional impairment	0.04	0.06^*	0.09^**	0.36^***	0.04	0.37^***	0.40^***	0.64^***	0.61^***	0.80^***	0.75^***	0.85^***	—
14. Meaning-making difficulties	0.03	0.09^**	0.05	0.35^***	0.00	0.32^***	0.38^***	0.58^***	0.58^***	0.72^***	0.70^***	0.78^***	0.81^***	—
15. Substance use coping	0.03	0.04	0.11^***	0.38^***	0.01	0.36^***	0.35^***	0.52^***	0.48^***	0.66^***	0.62^***	0.72^***	0.69^***	0.63^***	—
M	39.44	0.76	0.56	0.35	3.43	0.30	0.38	2.95	2.93	12.74	6.14	6.45	19.64	19.36	1.29
SD	11.75	0.42	0.50	0.48	1.55	0.46	0.49	1.73	1.64	6.75	2.82	4.23	11.21	5.84	1.05

Race (1 = White; 0 = non-White); gender (1 = men; 0 = women and other); diagnosis (1 = COVID-19 diagnosis; 0 = no COVID-19); time = time since loss in months; relationship (1 = immediate family and romantic relationship; 0 = other); help (1 = received professional help for loss; 0 = no help).

p < 0.05. ^**p < 0.01. ^***p < 0.001.

COVID-19 grief was higher among men, t(962.46) = 3.08, p < 0.001, and those who were diagnosed with COVID-19, t(905.64) = 14.38, p < 0.001, as well as those who sought professional help with their loss, t(1002.71) = 15.36, p < 0.001, and lost either an immediate family member or romantic partner, F(4, 1050) = 54.83, p < 0.001. Notably, time since loss and age were not associated with COVID-19 grief. Only one race difference was found such that Whites reported slightly higher COVID-19 grief than non-Black minorities, F(2, 1062) = 4.04, p = 0.02. All the psychological measures and substance use coping were positively intercorrelated (Table 2).

Table 3 displays the results of the final hierarchical multiple regression models, one each for functional impairment, meaning-making difficulties, and substance use coping. COVID-19 grief accounted for 4–5% of unique variance in the models (see change in adjusted R²). The standardized regression coefficients for COVID-19 grief are 2–15 times larger in magnitude than the competing variables we tested. These coefficients show that, after controlling for all other variables, a 1 standard deviation increase in COVID-19 grief will result in a 0.45 standard deviation increase in functional impairment, a 0.43 standard deviation increase in meaning-making difficulties, and a 0.41 standard deviation increase in substance use coping.

Table 3.

Final Model Summary Explaining Functional Impairment, Meaning-Making Difficulties, and Substance Use Coping (N = 1065)

Variables	Functional impairment	Meaning-making difficulties	Substance use coping
Race	0.01	0.04^*	0.00
Gender	0.03^*	0.00	0.05^*
Diagnosis	−0.01	0.03	0.08^**
Relationship	0.02	−0.02	0.05^*
Help	0.05^*	0.06^**	0.03
Depression	0.04	0.04	0.02
Generalized anxiety	0.06^*	0.11^*	0.01
Post-traumatic stress	0.22^***	0.09^*	0.16^**
Separation distress	0.18^***	0.21^***	0.12^***
COVID-19 grief	0.45^***	0.43^***	0.41^***
Adjusted R²	0.78	0.66	0.55
Change in adjusted R²	0.05	0.05	0.04
Significant F change	p < 0.001	p < 0.001	p < 0.001

Values reflect standardized regression coefficients. Changes in R² and F values reflect the addition of pandemic grief in the final blocks of the models. Race (1 = White; 0 = non-White); gender (1 = men; 0 = women and other); diagnosis (1 = COVID-19 diagnosis; 0 = no COVID-19); relationship (1 = immediate family and romantic relationship; 0 = other); help (1 = received professional help for loss; 0 = no help).

p < 0.05. ^**p < 0.01. ^***p < 0.001.

Discussion

The most notable finding in this study is that the PGS uniquely explained work and social difficulties attributed to a COVID-19 loss above and beyond relevant background data (e.g., COVID-19 diagnosis, relationship with the deceased, help seeking), bereavement-related psychopathology (post-traumatic stress, depression, and anxiety), and general grief reactions (i.e., separation distress). This finding is significant because it demonstrates that the PGS meets basic¹⁵ and grief-specific¹⁴ criteria for incremental validity and instrument validation.²⁴ The finding that the PGS further explained substance use coping in a similar magnitude as functional impairment is equally noteworthy. Given that the pandemic has led to a higher than usual number of overdoses,²⁵ this study's results underscore the need to screen widely to identify those who suffer from problematic levels of grief who appear susceptible to substance-related misuse, addiction, and death. The fact that the PGS uniquely explained a substantially larger magnitude of each of the three major outcomes than the other competing measures in the final regression models underscores its practical relevance in identifying risk of deleterious outcomes across a range of functional and behavioral domains.

Beyond the identification of these common clinical outcomes, PGS scores were also associated with a struggle for meaning in the context of loss, an important outcome identified in research on bereavement.²⁶ Both contemporaneous and longitudinal research have documented difficulties in making sense of a loss as a major risk factor for intense and prolonged courses of bereavement.²⁷ The link between disrupted meaning making and pandemic grief evidenced in this study reinforces concerns that their conjunction could forecast the long-term debilitation associated with prolonged grief disorder, a distinctive form of pathological grief recognized by both the World Health Organization²⁸ and the American Psychiatric Association.²⁹ This prospect clearly deserves research in future longitudinal studies.

Palliative care is often challenged by questions concerning when and what bereavement support to provide, to whom, and on what basis.^30,31 A recent editorial called for improvements in supporting the bereaved in greatest need, including improved screening efforts.³² The length and complexity of existing grief measures mean that they are unlikely to be feasible for use in palliative settings,^33,34 particularly during a pandemic. The results of this study show that the PGS has unique value in mental health screening to identify several maladaptive outcomes of COVID-19 bereavement, namely functional impairment, meaning-making difficulties, and substance use coping. Responding to the mental health effects of the pandemic adds a substantial challenge to already overburdened services.³⁵ Thus, the PGS is likely to help palliative services make ethical and cost-effective decisions about resource allocation, especially where resources are already stretched.

In addition, the surge of postpandemic grief worldwide will have the greatest impact in communities under-resourced to start with,³⁶ leading to well-organized advocacy for a White House Office of Bereavement Care in the United States to address the grief of these and all communities (https://live-evermore.org). However, this call for universal bereavement support and selective availability of specialized grief therapy when indicated imply a global need for greater professional training in grief assessment and intervention, as advocated and provided by leading organizations in end-of-life and bereavement care worldwide (e.g., Hospice Foundation of America, National Hospice and Palliative Care Organization, Australian Centre for Grief and Bereavement, Portland Institute for Loss and Transition, National Bereavement Partnership and Cruse Bereavement Care in the United Kingdom, Fundacion Hospital in Spain, Quatro Estações in Brazil, and others). Likewise, there are efforts underway to curate a comprehensive index of Internet resources of all kinds for the bereaved themselves (Wortman, C. B., personal communication, March 15, 2021). The emerging application of implementation science to bereavement care³⁷ will be especially fruitful in improving the lives of bereaved people beyond the pandemic.

Notably, neither the PGS nor other screeners used in this study to assess depressive, anxious, or post-traumatic symptomatology correlated even modestly with time across the six or more months of bereavement assessed in this research. This is striking, as a comprehensive meta-analysis of bereavement outcomes has documented the normative trend for symptomatology to decline as a function of time since loss.³⁸ The fact that no such trend was observed for >1000 adults bereaved by COVID-19 in this study is, therefore, all the more alarming. This is especially the case in light of the finding that over half of the present sample scored in the clinically dysfunctional range of the PGS, raising the specter of prolonged and preoccupying grief for the majority of those losing loved ones to COVID-19 unless adequately treated. The related finding that the two highest correlations of the PGS were with post-traumatic stress on the one hand and separation distress on the other is compatible with a conceptualization of COVID-19 grief as a traumatic form of grief.⁵

Conclusions

The results of this study support the utility of the PGS as a short, easy-to-use, and clinically relevant screener that could enable palliative settings to identify mourners at risk of functional impairment, meaning-making difficulties, and substance use coping. However, the results of this study are qualified by two limitations. First, we relied on a convenience sample and online self-report measures. Although this is common in bereavement research, replication of this study using random sampling and in-depth clinical interviews of the participants should yield more representative and detailed results. Given that MTurk samples tend to report significantly lower mean scores on measures of post-traumatic stress, depression, and generalized anxiety symptoms than undergraduate, community, and treatment-seeking samples,³⁹ our estimates might be conservative. Second, because we used a cross-sectional design in our study, we could not determine the causal relationships between the variables. Therefore, a replication and extension of this study using a longitudinal design would be desirable to emphasize COVID-19 grief as a unique syndrome specific to the current pandemic, and likely to be relevant to similar pandemics yet to come.

Footnotes

Authors' Contributions

S.A.L and R.A.N. conceived the study. S.A.L. analyzed the data. S.A.L. and L.J.B. drafted the initial article. All authors contributed to data interpretation, writing, and critically reviewing the article, and approved the final article. All agree to be accountable for all aspects of the study.

Funding Information

No funding was received for this article.

Author Disclosure Statement

No competing financial interests exist.

References

World Health Organization: WHO Coronavirus Disease (COVID-19) Dashboard 2021. https://covid19.who.int (Last accessed February 17, 2021).

Verdery

, Smith-Greenaway

, Margolis

, Daw

: Tracking the reach of COVID-19 kin loss with a bereavement multiplier applied to the United States. PNAS, 2020; 117:17695–17701.

Neimeyer

, Lee

: Circumstances of the death and associated risk factors for severity and impairment of

COVID-19 grief

. Death Stud (In Press).

Silver

: Surviving the trauma of COVID-19. Science, 2020; 369:11.

Stroebe

, Schut

: Bereavement in times of COVID-19: A review and theoretical framework. OMEGA J Death Dying, 2020; 82:500–522.

Wallace

, Wladkowski

, Gibson

, White

: Grief during the COVID-19 pandemic: Considerations for palliative care providers. J Pain Symptom Manage, 2020; 60:e70–e76.

Eisma

, Tamminga

: Grief before and during the COVID-19 pandemic: Multiple group comparisons. J Pain Symptom Manage, 2020; 60:e1–e4.

Eisma

, Smid

, Boelen

: Acute grief after deaths due to COVID-19, natural causes and unnatural causes: An empirical comparison. J Affect Dis, 2021; 278:54–56.

Breen

, Lee

, Neimeyer

: Psychological risk factors of functional impairment following

COVID-19 deaths

. J Pain Symptom Manage 2021. [Epub ahead of print]; DOI: 10.1016/j.jpainsymman.2021.01.006.

10.

Lee

, Neimeyer

: Pandemic grief scale: A screening tool for dysfunctional grief due to a

COVID-19 loss

. Death Stud 2020. [Epub ahead of print]; DOI: 10.1080/07481187.2020.1853885.

11.

Djelantik

AAAMJ

, Smid

, Kleber

, Boelen

: Early indicators of problematic grief trajectories following bereavement. Eur J Psychotraumatol, 2017; 8:1–6.

12.

Guldin

, O'Connor

, Sokolowski

, et al.: Identifying bereaved subjects at risk of complicated grief: Predictive value of questionnaire items in a cohort study. BMC Palliat Care, 2011; 10:1–7.

13.

Weinstein

, Berwick

, Goldman

, et al.: A comparison of three psychiatric screening tests using receiver operating characteristics (ROC) analysis. Med Care, 1989; 27:593–607.

14.

Bonanno

: Is complicated grief a valid construct?. Clin Psychol, 2006; 13:129–134.

15.

Hunsley

, Meyer

: The incremental validity of psychological testing and assessment: Conceptual, methodological, and statistical issues. Psychol Assess, 2003; 15:446–455.

16.

Lenferink

LIM

, Boelen

, Smid

, Paap

MCS

: The importance of harmonising diagnostic criteria sets for pathological grief. Br J Psychiatry 2019. [Epub ahead of print]; DOI: 10.1192/bjp.2019.240.

17.

Chmielewski

, Kucker

: An MTurk crisis? Shifts in data quality and the impact on study results. Soc Psychol Pers Sci, 2020; 11:464–473.

18.

Kroenke

, Spitzer

, Williams

, Löwe

: An ultra-brief screening scale for anxiety and depression: The PHQ–4. Psychosomatics, 2009; 50:613–621.

19.

Kilpatrick

, Resnick

, Friedman

: Severity of posttraumatic stress symptoms—Adult. 2013. https://www.psychiatry.org/psychiatrists/practice/dsm/dsm-5/online-assessment-measures (Last accessed October 15, 2020).

20.

Lee

: The persistent complex bereavement inventory: A measure based on the DSM-5. Death Stud, 2015; 39:399–410.

21.

Lee

: Factorial structure of the Persistent Complex Bereavement Inventory: Testing a hierarchical factor model. Death Stud, 2018; 42:356–361.

22.

Mundt

, Marks

, Shear

, Griest

: The work and social adjustment scale: A simple accurate measure of impairment in functioning. Br J Psychiatry, 2002; 180:461–464.

23.

Holland

, Currier

, Neimeyer

: Validation of the integration of stressful life experiences scale–Short form in a bereaved sample. Death Stud, 2014; 38:234–238.

24.

Boelen

, Djelantik

MJAAA

, de Keijser

, et al.: Further validation of the Traumatic Grief Inventory-Self Report (TGI-SR): A measure of persistent complex bereavement disorder and prolonged grief disorder. Death Stud, 2019; 43:351–364.

25.

American Medical Association: Issue Brief: Reports of Increases in Opioid-Related Overdose and Other Concerns During COVID Pandemic. 2020, pp. 1–15.

26.

Neimeyer

: Meaning reconstruction in bereavement: Development of a research program. Death Stud, 2019; 43:79–91.

27.

Milman

, Lee

, Neimeyer

: Social isolation and the mitigation of coronavirus anxiety: The mediating role of meaning. Death Stud 2020. [Epub ahead of print]; DOI: 10.1080/07481187.2020.1775362.

28.

World Health Organization International Classification of

Disease

, 11th ed. 2019. https://icd.who.int/browse11/l-m/en (Last accessed December 28, 2020).

29.

Moran

: Board approves new prolonged grief disorder for

DSM

. Psychiatric News 2020. [Epub ahead of print]; DOI: 10.1176/appi.pn.2020.11a12.

30.

Breen

, Aoun

, O'Connor

, Rumbold

: Bridging the gap in palliative care bereavement support: An international perspective. Death Stud, 2014; 38:54–61.

31.

Aoun

, Rumbold

, Howting

, et al.: Bereavement support for family caregivers: The gap between guidelines and practice in palliative care. PLoS One, 2017; 12:e0184750.

32.

Lichtenthal

: Supporting the bereaved in greatest need: We can do better. Palliat Support Care, 2018; 16:371–374.

33.

Sealey

, Breen

, O'Connor

, Aoun

: A scoping review of bereavement risk assessment measures: Implications for palliative care. Palliat Med, 2015; 29:577–589.

34.

Roberts

, Jankauskaite

, Slivjak

, et al.: Bereavement risk screening: A pathway to psychosocial oncology care. Psychooncology 2020. [Epub ahead of print]; DOI: 10.1002/pon.5526.

35.

Rauch

SAM

, Simon

, Rothbaum

: Rising tide: Responding to the mental health impact of the COVID-19 pandemic. Depress Anxiety, 2020; 37:505–509.

36.

Moore

, Jones-Eversley

, Toliver

, et al.: Six feet apart or six feet under: The impact of COVID-19 on the Black community. Death Stud 2020. [Epub ahead of print]; DOI: 10.1080/07481187.2020.1785053.

37.

Breen

, Moullin

: The value of implementation science in bridging the evidence gap in bereavement care. Death Stud 2020. [Epub ahead of print]; DOI: 10.1080/07481187.2020.1747572.

38.

Currier

, Neimeyer

, Berman

: The effectiveness of psychotherapeutic interventions for the bereaved: A comprehensive quantitative review. Psychol Bull, 2008; 134:648–661.

39.

Engle

, Talbot

, Samuelson

: Is Amazon's Mechanical Turk (MTurk) a comparable recruitment source for trauma studies?. Psychol Trauma, 2020; 12:381–388.