The prevalence of prolonged grief disorder in bereaved individuals following unnatural losses: Systematic review and meta regression analysis

In the paper

“Theoretically, unnatural losses may cause more mental health difficulties because of the disruption of positive and self-evident assumptions about the world (i.e., 'the world is a safe place') and due to more intrusive and negative memories.”

What the sources say

• Partly supported

  The Introduction (page unavailable) says unnatural or violent losses carry elevated risks of disturbed grief, PTSD, and depression, and discusses maladaptive cognitions and negative beliefs about the world and safety as possible contributors. It does not say that more intrusive and negative memories explain the greater difficulties; the paper’s reported findings concern negative cognitions, not memories. (LLM, conf=0.85)

  cites Boelen et al., 2016 sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites Boelen et al., 2006 source

In the paper

“Furthermore, in comparison with existing treatments for MDD and PTSD, PGD appears uniquely responsive to grief-specific treatments (Shear et al., 2005;Zisook and Shear, 2009).”

What the sources say

• Partly supported

  The Results report a higher response rate to complicated grief treatment than to interpersonal psychotherapy (51% versus 28%), which the paper identifies as an established treatment for depression. The paper did not compare the grief-specific treatment with an existing PTSD treatment, so it does not support that part of the comparison. (LLM, conf=0.86)

  cites Shear et al., 2005 sourceExperimental — accuracy not yet measured
• Partly supported

  The “complicated Grief” section reports that targeted complicated grief treatment produced better outcomes than standard psychotherapy. It does not compare grief-specific treatment with existing PTSD treatments, and its discussion of antidepressants does not establish the claimed comparison with MDD treatments. (LLM, conf=0.86)

  cites Zisook and Shear, 2009 sourceExperimental — accuracy not yet measured

In the paper

“Previous research has indicated that one out of ten naturally bereaved individuals develops prolonged grief disorder (PGD).”

• Also stated in: § 1. Introduction · plain

  “A recent meta-analysis by Lundorff et al. (2017) showed that one out of ten bereaved adults following a non-violent loss is at risk of developing PGD.”

What the sources say

• Supported

  The Discussion (p. 13) reports a pooled PGD prevalence of 9.8% and describes it as affecting one out of ten adults after a non-violent loss. The Conclusion (p. 20) likewise says one out of ten adults experiencing bereavement will exhibit clinically significant PGD symptoms. (LLM, conf=0.92)

  cites Lundorff et al. (2017) source

In the paper

“The core underlying characteristic of PGD is 'yearning for the deceased', whereas 'anxiousness' and 'depressed mood' are central to PTSD or MDD (Maercker and Znoj, 2010;Shear, 2015).”

What the sources say

• Partly supported

  Table 2 in “Communalities and differences of CGD and PTSD” identifies yearning as the core symptom group for complicated grief, supporting the PGD portion of the claim. It identifies intrusive symptoms—not anxiousness—as the core symptom group for PTSD, and the paper does not establish depressed mood as central to MDD. Page numbers are unavailable in our copy. (LLM, conf=0.85)

  cites Maercker and Znoj, 2010 sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites Shear, 2015 source

In the paper

“Following both the title and abstract screening, as well as the fulltext screening, the interrater agreement was evaluated using Cohen's Kappa statistic, κ (McHugh, 2012).”

What the sources say

• Supported

  The 'Measurement of interrater reliability' section identifies Cohen’s kappa as a measure of agreement between two raters. The cited paper supports the choice of statistic; the screening stages describe the citing authors’ own procedure. (LLM, conf=0.93)

  cites McHugh, 2012 source

In the paper

“Recently, PGD has been included in the International Classification of Diseases, 11th revision (ICD-11) (Killikelly and Maercker, 2017;Prigerson et al., 2009;World Health Organization, 2018) with a time criterium of 6 months following bereavement.”

What the sources say

• Supported

  Table 1 (Introduction, ‘PGD diagnosis and its precursors’) lists the ICD-11 PGD time criterion as persistence for more than six months after the loss, while also requiring that the duration exceed cultural and social norms. The ‘International applicability’ section explains that six months is an approximate time frame, not an inflexible cutoff. (LLM, conf=0.90)

  cites Killikelly and Maercker, 2017 source
• Partly supported

  Phase 5: proposing criteria for PGD specifies that diagnosis should not be made until at least six months after the death. The Conclusion describes these as proposed criteria for ICD-11 and calls for PGD’s consideration for inclusion; this 2009 paper does not establish that PGD was later included in ICD-11. (LLM, conf=0.88)

  cites Prigerson et al., 2009 sourceExperimental — accuracy not yet measured
• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites World Health Organization, 2018

In the paper

“There is also an ongoing discussion about which criteria best describe the phenomenon of pathological grief (Boelen et al., 2018;Cozza et al., 2019;Maciejewski et al., 2016;Mauro et al., 2017).”

What the sources say

• Supported

  The Introduction (page not supplied) compares competing criteria sets for disturbed grief, reports differing findings about their performance, and says more work is needed to identify a psychometrically sound conceptualization. This supports the claim that which criteria best describe pathological grief remains under discussion. (LLM, conf=0.93)

  cites Boelen et al., 2018 source
• Supported

  The Introduction (page not supplied) describes four proposed criteria sets and says the study aims to inform the development of consensus criteria. The Discussion (page not supplied) compares how well those criteria identify cases, supporting the claim that the best criteria for pathological grief remain under discussion. (LLM, conf=0.93)

  cites Cozza et al., 2019 source
• ?Not checked — no full text of the sourcecites Maciejewski et al., 2016 source

In the paper

“According to Prigerson et al. (2009), the PGD diagnosis can be set after 6 months' post-bereavement.”

What the sources say

• Supported

  In “Phase 5: proposing criteria for PGD,” the paper specifies that a diagnosis should not be made until at least six months after the death. This supports the claim that PGD can be diagnosed after six months post-bereavement, provided the other diagnostic criteria are met. (LLM, conf=0.95)

  cites Prigerson et al. (2009) source

In the paper

“Twenty-five articles met eligibility criteria.”

• Also stated in: § 3.3. Meta-analysis · plain

  “In total, twenty-five articles with thirty prevalence estimates were included in the overall meta-analysis.”

In the paper

“The mean age of the included samples ranged from 20 (Morina et al., 2011) to 72 years (Kristensen et al., 2012a).”

In the paper

“These studies comprised a total of 4774 participants (range n = 10 to 803) of whom 2296 (48%) were identified as PGD cases.”

In the paper

“Seventeen studies were performed in OECD membership countries: six in North America (USA), four in Norway, one in the Netherlands, one in Belgium, one in Portugal, one in Korea, one in Japan, and two in Australia.Eight studies were performed in non-OECD membership countries: three in Kosovo, three in China, one in Cambodia, and one in Rwanda.”

In the paper

“With regard to the nature of traumatic events, we found that suicide, intentional killing, and accidents were not significant predictors for PGD.”

In the paper

“Of the study participants 68% was female.”

In the paper

“The mean time post-loss in the studies varied from less than 1 month (Mitchell et al., 2004) to 30 years (Stammel et al., 2013) (Table 2).”

In the paper

“Two studies (Neria et al., 2007;Shear et al., 2006) used selfconstructed questionnaires based on criteria of PGD.”

What the sources say

• ?Not checked — no full text of the sourcecites Neria et al., 2007 source
• ?Not checked — no full text of the sourcecites Shear et al., 2006 source

In the paper

“The studies assessed PGD using either the Prolonged Grief-13 (PG-13), the Inventory of Complicated Grief (ICG) or the Inventory of Complicated Grief-Revised (ICG-R) (Tables 1 and 2).”

In the paper

“Previous research has clearly shown that the experience of losing a child is one of the most painful and stressful events that an adult can experience (Keesee et al., 2008;Rubin and Malkinson, 2001;Wijngaards-de Meij et al., 2005).”

What the sources say

• ?Not checked — no full text of the sourcecites Keesee et al., 2008 source
• ?Not checked — no full text of the sourcecites Rubin and Malkinson, 2001 source
• ?Not checked — no full text of the sourcecites Wijngaards-de Meij et al., 2005 source

In the paper

“This finding accords with previous research which showed that grief symptoms normally decrease over time, and that most bereaved individuals eventually adjust to the loss and thereby return to adaptive functioning (Bonanno, 2004;Kristensen et al., 2012b).”

What the sources say

• ?Not checked — no full text of the sourcecites Bonanno, 2004 source
• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Kristensen et al., 2012b

In the paper

“During the full-text screening, the researchers disagreed on 18 (5%) papers yielding an interrater reliability of κ = 0.89 (p < .001),indicating almost perfect agreement.”

In the paper

“However, several other researchers argue that the 6month time frame is too short and may lead to over-inclusion of the recently bereaved (Rubin et al., 2008).”

What the sources say

• ?Not checked — no full text of the sourcecites Rubin et al., 2008 source

In the paper

“All disagreements were discussed until both researchers agreed about the in-or exclusion.”

In the paper

“Therefore, to examine the publication bias, Egger's t statistics was used (Egger et al., 1997).”

What the sources say

• ?Not checked — no full text of the sourcecites Egger et al., 1997 source

In the paper

“Finally, we tested the robustness of our findings with a permutation test in 1000 iterations.”

What the sources say

• ?Not checked — no full text of the sourcecites Higgins and Thompson, 2004 source

In the paper

“Similar concepts with slightly different criteria are called persistent complex bereavement disorder (PCBD), included in the Fifth Edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), (American Psychiatric Association, 2013) and complicated grief (Shear, 2015), and include time criteria of 12 and 6 months following bereavement, respectively.”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites American Psychiatric Association, 2013
• ?Not checked — no full text of the sourcecites Shear, 2015 source

In the paper

“During the title and abstract screening, the researchers disagreed on 326 (9%) papers yielding an interrater reliability of κ = 0.64 (p < .001),indicating substantial agreement.”

In the paper

“In one of the most recent reviews about correlates of PGD following a violent loss, a meta-analysis was conducted in which significant effect sizes were found with regard to the following correlates: comorbid psychopathology, suicidality, rumination, the relationship with the deceased, exposure to earlier traumatic events, age and prior history of counseling (Heeke et al., 2019).”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Heeke et al., 2019

In the paper

“Therefore, before we included these predictors in the meta-regression analysis, we assessed the multicollinearity with an inter correlation matrix (Harrer et al., 2019).”

What the sources say

• ?Not checked — no full text of the sourcecites Harrer et al., 2019 source

In the paper

“In the correlation matrix we did not find correlations of more than 0.8 and all predictors could be included in the meta-regression analysis (Supplementary Materials C).”

In the paper

“Then, the eligibility of both the abstracts and full-text articles were independently reviewed by MD and AM using this software.”

In the paper

“The protocol was registered in PROSPERO -an international prospective register for review protocols -in January 2018 (registration number CRD42018084631).”

In the paper

“The risk of bias was evaluated and double-checked independently by MD and AM.”

In the paper

“In a prior study, we have found that acceptance in the first year of the loss predicted a more favorable grief trajectory across the second year (Djelantik et al., 2017a).”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Djelantik et al., 2017a

In the paper

“The meta-analysis was performed and reported in accordance with the PRISMA guidelines for systematic reviews and meta-analyses (Moher et al., 2010).”

What the sources say

• ?Not checked — no full text of the sourcecites Moher et al., 2010 source

In the paper

“According to the quality assessment scores, four studies scored low (score = 9-10), twelve studies scored moderate (score = 7-8), and nine studies scored high (score = 6 or less), on Risk of Bias (See Supplementary Material B for the detailed scores for the RoB of each study).”

In the paper

“The data-extraction was performed and double-checked independently by MD and AM.”

In the paper

“In the present meta-analysis, only two studies included participants in the first year after loss (Mitchell et al., 2004;Tsutsui et al., 2014), therefore we were not able to compare PGD rates between 0-6, 6-12 and more than 12 months after the loss.”

In the paper

“Death of only child was a strong predictor for experiencing PGD in our model.However, this predictor was not considered robust in the permutation test, most likely because this predictor was only assessed in one included study (Xu et al., 2014).”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Xu et al., 2014

In the paper

“Our results imply that bereaved parents who lost their only child, and bereaved individuals following violent killings such as suicide, accidents, homicide, and war-related deaths are most vulnerable for developing PGD.”

In the paper

“Other core symptoms of PGD include 'avoidance of reminders of the death', 'difficulties accepting the loss', and 'significant impairment of daily functioning' (Boelen and Prigerson, 2007;Silverman et al., 2000).”

What the sources say

• ?Not checked — no full text of the sourcecites Boelen and Prigerson, 2007 source
• ?Not checked — no full text of the sourcecites Silverman et al., 2000 source

In the paper

“our search syntax was aimed to finding studies with a reported prevalence for PGD and not to finding all possible correlates for PGD.”

In the paper

“Because of the limitations on review level due to varying sample sizes, sets of criteria and sampling bias in the included studies, future studies on the prevalence of PGD are highly recommended.”