Abstract
Background:
Grief is a common reaction to the feeling of loss and it is considered a physiological and instinctive response. The ‘normal’ grief evolves into an ‘integrated’ phase within 1 year from death, and it is a non-pathological condition, that do not require specific therapeutic interventions. When this ‘integrated phase’ does not occur, the subject could reach pathological manifestations related to the grief. The Persistent Complex Bereavement Disorder (PCBD) is a new DSM5 clinical category characterized by symptoms related to the detachment and to the post-traumatic distress and it differs from normal and uncomplicated grief, for the disability caused by these reactions and their persistence and pervasiveness.
Aim:
The purpose of this work is the analysis of the pathways that led to this new definition, through a review of the main studies published in the last 20 years, with the aim to clarify the clinical utility of this new diagnostic category.
Method:
Relevant publications done in the last 20 years were identified via electronic searches of Pubmed, Embase, and Elsevier databases using the terms ‘complicated grief’ AND ‘persistent’, according to PRISMA guideline and PICO study design.
Results:
PCBD results a new important clinical category showing specific symptoms, diagnostic criteria, and treatment. It presents many differences with other pathologies, that goes into differential diagnosis with PCBD, and it and can be treated with targeted therapeutic approaches. Diagnostic criteria for PCBD could allow an early diagnosis and a correct treatment avoiding underdiagnosis and misdiagnosis.
Conclusion:
Further researches could focus on the evaluation of more neurobiological aspects, new psychometric tools, for assessing susceptibility to this pathology, and on the cultural aspects that may influence mourning reactions, in an ethno-psychiatric perspective.
Keywords
Introduction
The loss of significant ones is a natural occurrence of the life cycle. Everyone will experience this sooner or later, and yet it is one of the most difficult events that human beings experience.
Linked to the experience of loss, the terms grief, mourning, and bereavement have different shades of meaning, since they refer to different aspects of the phenomenon of loss, thus they are not interchangeable (Buckley et al., 2019; Carmassi et al., 2016; Geng et al., 2019; Levine, 2017; Shear, 2012; Tullis, 2017).
‘Grief’ refers to the emotional suffering concerning the loss, ‘mourning’ is the process by which people adapt to a loss and it is the external expression of grief, including rituals and being influenced by spiritual, religious, and cultural beliefs and practices; ‘bereavement’ is the period after having suffered the loss of a loved one, it is the time after a loss during which grief is experienced and mourning occurs.
However, in current literature all these terms indicate the phenomenon of loss, as we will do in the present work; in DSM-5 and in ICD-10, for example, ‘bereavement’ and ‘grief’ are used for clinical conditions that overlap in terms of symptoms, prevalence, and health correlates (Persistent and Complex Bereavement Disorder and Prolonged Grief Disorder) (DSM-5; ICD-11).
Grief is a common reaction to the feeling of loss; many people experience this kind of event during the course of their lives. It is a complex phenomenon and it transversally affects many cultures; while the experience of grief is unique the whole grief process is personally, relationally, and culturally defined.
In the greatest number people adequately address the loss, but a significant percentage of subjects develops a syndrome characterized by a prolongation of the psychological suffering related to the loss. For this reason, the Persistent Complex Bereavement Disorder (PCBD) appears as a separate pathological entity in the third section of the DSM 5 (2013), in order to study and analyze this new topic. A slightly different conceptualization, named Prolonged Grief Disorder (PGD), has been introduced in the 11th edition of the International Statistical Classification of Diseases and Related Health Problem.
PGD and PCBD strongly overlap in terms of symptoms, prevalence, and health correlates; pathological grief, treated in this article, so refers to PCBD and PGD in an overlapping manner. Loss is often a dramatic event that alters the psychic balance and the well-being of the subject, especially in the acute phase. Grief, as painful as it is, is considered a physiological and instinctive response (Carmassi et al., 2014). It is difficult to judge as pathological the reactions of a subject in the early stages of mourning, mainly considering that the experience of detachment is different from one to another, and it can also change in relation to the cultural group of belonging.
It could be useful to analyze the differences between ‘normal’ and ‘pathological’ grief in order to understand where the research that led to the definition of Persistent Complex Bereavement Disorder has started.
‘Normal’ grief is a process that develops through different phases. It has been studied in psychology and there are several theories derived from different authors. ‘Normal’ grief could present severe psychological reactions (anguish, anger, shock, etc.), but it evolves into an ‘integrated’ phase within 1 year from death, configuring itself as a non-pathological condition that do not require specific therapeutic interventions. As observed by Lombardo et al. (2014), the integrated grief is a condition characterized by a quiet memory of the deceased and when this ‘integrated phase’ does not occur, the subject could reach grief-related clinical conditions.
In the last 20 years the scientific debate has led to a gradual definition of diagnostic criteria for various pathological patterns, named from time to time as ‘pathological grief’, ‘complicated grief’, and ‘traumatic grief’, up to the Persistent Complex Bereavement Disorder present in the DSM-5 as a new clinical category. This pathology differs from normal and uncomplicated grief, for the distress and disability caused by these reactions, and their persistence and pervasiveness.
The purpose of this work is the analysis of the pathways that led to this new definition, through a review of the main studies published in the last 20 years, with the aim to clarify the clinical utility of this new diagnostic category.
Materials and methods
Protocol and registration
This review has been registered at the PROSPERO database for systematic review with number 138976 on 03/07/2019. This review has been conducted according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guideline (Hutton et al., 2015; Liu et al., 2019; Tian et al., 2015) and PICO study design (Da Costa Santos et al., 2007).
Eligibility criteria
Inclusion and exclusion criteria are listed below:
Inclusion:
Studies about Grief and Complicated Grief
Diagnostic methods and therapies for CG
English language and full text
Exclusion:
Case report, case series, commentaries, PhD thesis
Older than 20 years study
Not accessible data or title or abstract
Preclinical studies and animal models
Information sources
The results of the review in analysis were obtained through research in the following scientific databases: Pubmed, Embase, Elsevier. The databases have automated the work and provided useful information for screening through integrated search filters.
Search
The keywords used in the scientific databases (see section ‘Information sources’) are the following: ‘complicated grief’ AND ‘persistent’. The keywords were chosen by the authors in order to provide the highest number of results possible and limit the risk of bias. Furthermore the main question of the study, according to the PICO (Population/Intervention/Comparison/Outcome) study design (Da Costa Santos et al., 2007) is:
Is there any the difference between normal and pathological grief on grief patients?
Study selection
A manual screening and study selection was performed to obtain 22 results with enough information as reported in PRISMA (Hutton et al., 2015; Liu et al., 2019; Tian et al., 2015) Flow Chart (Figure 1).

PRISMA flow diagram.
Data collection process
Two authors (CC, CM) independently assessed the results obtained and clarified any disputes in the presence of a third expert reviewer (MRAM). The results were subsequently re-evaluated by the auditors and the salient results were shown.
Data items
Authors (Year) – Authors and Year of publication
Sample – Sample size and type
Data Items – outcome of the study (showed in summary at Table 1)
Summary of evaluated measures in results.
Results – Main results
Statistic – Significant or not data (Table 2)
Synthesis of significant or not data.
Bias Risk – Bias Risk evaluation (Table 3)
Risk of bias table.
Note. U = unclear; L = low; M = moderate; h = High.
Summary measures
Results
Study selection
Authors initially obtained 49 results, on 10 years 45, 35 on humans, full text 31. Following a reading of the titles and abstracts, the authors selected, based on the information contained, a number of 21 articles (Figure 1).
Study characteristics
Kokou-Kpolou et al. (2018) in a cross sectional study on 98 bereaved mothers (mean age = 33.9 years) showed a significant interaction between depressive symptoms increasing or life cognitions in mothers and death of the child after or before the birth. Kliem et al. (2018), using an Inventory of Comlicated Grief-Revised (ICG-R) on 1,445 bereaved individuals showed a latent structure of PGD.
Eisma (2018), evaluated 379 adult patients from general population by a vignette-based experiment, aimed at evauating stigmatizing reactions (i.e. negative attributions, negative emotional reactions, and larger preferred social distance) in response to people diagnosed with a PGD diagnosis, especially when developed in response to the loss of a friend, instead of a partner. Bartl (2018), in a RCT on 51 patients, evaluated Integrative Cognitive-Behavioral Therapy (CBT) of PGD. CBT is significantly useful in PGD patients with a Cohen’s d = 0.60, the effect of therapy remained stable for 1.5 years of follow up. Mauro et al. (2017) diagnosed 70% of Complicated Grief patients using PCBD criteria; 59.6% using PGD criteria and 99.6% using CG criteria. This study evaluated rate of diagnosis on two groups of patients. Kotoucova (2017) evaluated the effect of a United States member’s death on family member. This study has been conducted on 232 military family members, comparing these results with a non-military-related sample. Despite demographic characteristics similarity was present between groups. Lee et al. (2016) tried to recognize neuroticism as a risk for CG. Authors obtained data from 99 students finding that social isolation and spiritual struggles could have contributed to the grief for a percentage of students. LeBlanc et al. (2016) recruited 49 patients (two groups with CG or healthy bereaved) and conducted an investigation between CG and heart rate, respiratory sinus arrhythmia and skin conductance levels. According to authors, despite the modest sample size, were rare to highlight differences between patients. According to Cozza et al. (2016) DSM-5 helped to identify 53% of PGD, and Complicated Grief Criteria identified 90% of clinical cases. All criteria accurately excluded non clinical grief cases. In this study, Cozza et al. (2016) recruited 1,732 patients as US family member. Arizmendi et al. (2016) through a Functional Magnetic Resonance Imaging methods evaluated cortical activity on three groups of patients. Activity of the Rostral anterior cingulate cortex was higher on CG sample. Bui et al. (2015) standardized the Structured Clinical Interview for CG (SCI-CG); the 31-item instrument revealed a satisfactory internal consistency (alpha = .78). According to Rosner et al. (2014), RCT, CBT could improve PG symptoms, this therapy produced a 89% improving on sample patients with a follow up of 1.5 years and a 8 to 14 dropouts. Zetumer et al. (2015) enrolled 345 patients evaluated by the Healing Emotion After Loss (HEAL). Parents who experienced the loss of a child demonstrated higher levels of CG (p = .025) when compared with caregivers (p = .007) and non-parents with CG. Parents who had lost younger children were exposed to a with of dead (p = 0.041). Hall et al. (2014) with their exploratory and descriptive study, evaluated global and specific cognitive functioning in a sample of 335 patients compared with a control group of 250 subjects by the Montreal Cognitive Assessment (MoCA). CG partecipants had lower MoCA values thus. Robinaugh et al. (2014), conducted a review on CG, evaluating physiological reactivity to reminder of loss. Authors examined results of individual studies whose results showed that anxiety sensitivity (AS) could contribute to CG development. The review on differential diagnosis and management of CG by Simon (2013), supported the efficacy of targeted psychotherapy for CG healing. Furthermore, preliminary study had suggested that antidepressant medications may be helpful. According to a study by Melhem et al. (2013), on parentally bereaved children aged between 8 and 17 years old, Complicated Grief-Revised for Children had an higher sensitivity (0.942) and specificity (0.965) in differentiating PG reaction from healthy cases at 9 months follow-up. DSM-5 criteria, according to authors, differentiated only 20% to 41.7% of cases with PGR at different time points (9, 21, and 33 months). O’Connor et al. (2013) presented two studies about catecholamines in plasma pre- and post-psychotherapeutic treatment, and another one about immunologic and neuroimaging biomarkers O’Connor et al. (2013). They showed that CG patients had high catecholamines levels and how the dorsal anterior cingulate cortex, insula, and posterior cingulate cortex were frequently activated in CG patients. Meert et al. (2011) evaluated 138 parents and 106 children with CG symptoms using the ICG, attachment, caregiving, demographic, and social support. According to Meert et al. (2011) ICG scores were 33.4 ± 13.6 at 6 month and 28 at 18 month, representing an improvement in ICG (p < .001). McCarthy et al. (2010) evaluated 58 patients who had lost child for cancer with self-report questionnaires for the assessment of CG (ICG-R) and depression (Beck Depression Inventory-Second Edition (BDI-II)). Twenty-one percent of parents reported depressive symptoms.
Synthesis of results
Risk of bias
Risk of bias analysis has been conducted according to Whiting (2018), Savovic (2018), and Higgins (2011) and results have been classified in the Table 3.
Discussion
Summary of evidence
The data collected according to the materials and methods section show different results, for which it is not possible to perform a univocal statistical analysis. In this section the results were discussed in such a way as to offer an overall picture. Kokou-Kpolou et al. (2018), showed how negative cognitions are associated with the symptoms of post-perinatal loss and change if death occurs before or after birth, the main limitation of this article is the small sample size. Kliem et al. (2018), conducted article has shown how a latent dimensional structure of PGD is evident. Eisma (2018), evidenced that stigma and its negative consequences seem to be a valid concern for the establishment of pathological mourning disorders in diagnostic manuals. Bartl (2018) showed as PG-CBT could improve posttraumatic growth in PGD patients, this could be an useful instrument for the clinician. Mauro et al. (2017), in their study concluded that PCBD actual criteria are not adequate, since they fail to capture this pathology; in the Authors’ opinion, this could be resolved by using Complicated Grief criteria. According to Lee et al. (2016), social isolation and spiritual struggles contribute negatively to the pain of emotionally sensitive students. The results also support the effectiveness of a biopsychosocial-spiritual approach to the study of adolescent pain. LeBlanc et al. (2016), observed evidence of blunted parasympathetic nervous system reactivity. Cozza et al. (2016) shows that DSM-5 persistent complex mourning disorder criteria accurately exclude non-clinical regulatory pain, but also exclude almost half of clinical cases, while complicated pain criteria exclude non-clinical cases identifying more than 90% of cases clinicians. The authors conclude that a significant modification is needed to improve the identification of cases by persistent diagnostic criteria of DSM-5 complex mourning disorder. Complicated mourning criteria are superior in accurately identifying clinically impaired pain. Arizmendi et al. (2016), Post hoc analysis evidenced activity in the dorsal ACC in the Complicated Grief. Bui et al. (2015), presented a SCI-CG for CG diagnosis, with significant results. This Interview could help clinicians during CG diagnosis. Rosner et al. (2014) showed how integrative cognitive behavioral therapy for prolonged grief could be useful in these patients with clinical relevance; however, in this study, no long term follow-up was performed. Zetumer et al. (2015) in their study showed that parents who had lost younger children were near to suicide or death thoughts. Comparison between parents who lost younger children with parents who lost older children was limited by a small sample size. Hall et al. (2014), investigated cognitive functioning in complicated grief. Authors demonstrated lower cognitive and attentional performances in patients with CG. Robinaugh et al. (2014), examined anxiety sensitivity (AS, i.e. the fear of anxiety-related sensations) in two studies of grieving adults with and without CG. In both studies, mourning adults with CG showed elevated AS values compared to those without CG. Simon (2013), demonstrated that individuals with CG have a greater risk of adverse health conditions, so it should be diagnosed for suicide or comorbid condition preventing. Melhem et al. (2013) proposed a short screening scale that, if validated, could help to identify PGR children and adolescents, being useful for the development of prevention and intervention strategies. O’Connor et al. (2013), with this study provided evidence of the hypothesis that catecholamine levels are affected by grief and, in turn, can affect the ability of those who have complicated grief to benefit from psychotherapy. Cervino (2019) supposed to find biomarkers useful for early diagnosis as in other medicine fields. According to O’Connor (2012), biomarkers could be helpful to predict CG and, possibly, also to orient treatments toward effective strategies. Meert et al. (2011) showed how complicated grief symptoms decrease among parents between 6 and 18 months after their child’s death. Complicated grief will allow parents at risk for persistent distress to receive professional support. McCarthy et al. (2010) showed how time since death is a significant predictor of distress and how they may be need palliative care to ensure best outcomes. According to the results of the individual studies discussed above, it is clear that the therapeutic techniques developed for the CG are different. However, the most adequate diagnostic technique is discussed. Many studies are of a comparative type, and tend to highlight which is the instrument with greater sensibility and sensitivity for CG diagnoses. In addition, instrumental diagnostic methods are still present, experimental, which show different serum levels in patients affected by CG. Other methods involve instrumental imaging investigations, these have shown abnormalities in different brain areas in patients with or susceptible to CG. Surely the future prospects of this pathology concern a rapid, safe, and fast diagnostic approach, and a therapeutic tool, with as many features; the articles in question repeatedly propose CBT, as a treatment for the patient with CG.
Limitations
The main limitation of this study is represented by the fact that many results cannot be compared with each other, and therefore a univocal statistical analysis cannot be performed.
