key: cord-0074049-mjprqeu0
authors: Eisma, Maarten C.; Janshen, Antje; Lenferink, Lonneke I. M.
title: Content overlap analyses of ICD-11 and DSM-5 prolonged grief disorder and prior criteria-sets
date: 2022-01-26
journal: nan
DOI: 10.1080/20008198.2021.2011691
sha: 7f08fa14bdeda01a00da918d91845ef0a604ef9a
doc_id: 74049
cord_uid: mjprqeu0

BACKGROUND: The International Classification of Diseases eleventh edition (ICD-11) has recently included prolonged grief disorder (PGD), a diagnosis characterized by severe, persistent, and disabling grief. The text revision of the Diagnostic and Statistical Manual of Mental Disorders 5 (DSM-5-TR) is scheduled to include a similar but distinct diagnosis, also termed PGD. Concerns have been raised that these new diagnoses are qualitatively different from both prior proposed diagnoses for pathological grief and each other, which may affect the generalizability of findings obtained with different criteria sets. OBJECTIVE: We conducted a content overlap analysis of PGD(ICD-11), PGD(DSM-5-TR), and previous proposals for pathological grief diagnoses (i.e. PGD 2009; complicated grief (CG), PGD ICD-11 beta draft, persistent complex bereavement disorder (PCBD) per DSM-5). METHODS: Using the Jaccard’s Index, we established the degree of content overlap between core and accessory symptoms of PGD(ICD-11), PGD(DSM-5-TR), and prior proposals for pathological grief diagnoses. RESULTS: Main findings are that PGD(ICD-11) and PGD(DSM-5-TR) showed moderate content overlap with each other and with most prior proposed diagnoses for pathological grief. PGD(ICD-11) and PGD(DSM-5-TR) showed the strongest content overlap with their direct predecessors, PGD(ICD-11 beta draft) and PCBD, respectively. CONCLUSIONS: Limited content overlap between PGD(ICD-11) and PGD(DSM-5-TR) and preceding criteria sets may threaten generalizability of past research on phenomenological characteristics of pathological grief to current criteria sets. Similarly, findings obtained with instruments to assess PGD(ICD-11) may not generalize to PGD(DSM-5-TR) and vice versa. Researchers should aim to determine under which circumstances criteria sets for PGD yield similar or distinct characteristics. Convergence of criteria sets for PGD remains an important goal for the future.

Antecedentes: La Decimoprimera Clasificación Internacional de Enfermedades (CIE-11) ha incluido recientemente el Trastorno Por Duelo Prolongado (PGD por sus siglas en ingles), un diagnóstico caracterizado por un duelo severo, persistente e incapacitante. La versión revisada del Manual Diagnóstico y Estadístico de los Trastornos Mentales (DSM-5-TR) tiene agendado incluir un diagnóstico similar pero diferente, también llamado PGD. Ha existido preocupación de que ambos diagnósticos sean cualititativamente diferentes de aquellos propuestos previamente para duelo patológico y también entre sí, lo que puede afectar la posibilidad de generalización de los hallazgos obtenidos con cada conjunto de criterios diagnósticos. Objetivo: Conducimos un análisis de solapamiento de contenido de los criterios diagnósticos del PGD de acuerdo a la CIE-11, del PGD de acuerdo al DSM-5-TR y de propuestas previas para diagnósticos de duelo patológico [como el PGD de Prigerson y colaboradores, publicado el 2009, Duelo complicado (CG por sus siglas en inglés) del borrador beta de la CIE 11, el Trastorno por Duelo Complejo Persistente (PCBD por sus siglas en inglés) del DSM-5]. Métodos: Usando el Índice de Jaccard, establecimos el grado de solapamiento del contenido entre los síntomas principales y accesorios de los criterios diagnósticos del PGD de acuerdo a la CIE-11, del PGD de acuerdo con el DSM-5-TR y de propuestas previas para diagnósticos de duelo patológico. Resultados: Los resultados principales son que los criterios diagnósticos del PGD de acuerdo a la CIE-11 y PGD de acuerdo al DSM-5-TR mostraron un solapamiento de contenido moderado entre ellos y también con la mayoría de los diagnósticos de duelo patológico previamente propuestos. Ambos diagnósticos mostraron el mayor solapamiento de contenidos con sus predecesores directos, el Duelo Complicado del borrador beta de la CIE-11 y el PCBD respectivamente. Conclusiones: el solapamiento limitado de contenidos entre los criterios diagnósticos del PGD de acuerdo a la CIE-11 y PGD de acuerdo al DSM-5-TR y los criterios precedentes pueden amenazar la generalización de investigación pasada de las características fenomenológicas del duelo patológico en los criterios diagnósticos actuales. En forma similar, los hallazgos obtenidos con instrumentos para evaluar el PGD de acuerdo a la CIE-11 pueden no ser generalizables al PGD de acuerdo al DSM-5-TR. Los investigadores debiesen determinar bajo qué circunstancias los criterios diagnósticos de PGD muestran características distintas o similares. La convergencia de los criterios diagnósticos de PGD sigue siendo una importante meta para el futuro. A concern regarding the development of new criteriasets for pathological grief is that they are, as a rule, qualitatively different from preceding criteria-sets (e.g. Boelen & Prigerson, 2012; Djelantik et al., 2021; Eisma & Lenferink, 2017; Stelzer, Zhou, Maercker, O'Connor, & Killikelly, 2020) . Criteria-sets differ in number of included symptoms, symptom content, and diagnostic algorithms (Eisma, Rosner, & Comtesse, 2020; Lenferink, Boelen, Smid, & Paap, 2021) . Consequently, the phenomenological characteristics of different pathological grief criteria-sets vary. For example, PGD ICD-11 has limited diagnostic agreement with prior proposed criteriasets, such as PCBD and PGD 2009 (e.g. Boelen, Lenferink, Nickerson, & Smid, 2018; Comtesse et al., 2020; Cozza et al., 2020) , although the extent of agreement partially depends on the chosen diagnostic algorithm . Therefore, previous findings on important clinical issues, ranging from dimensionality of diagnoses to treatment efficacy, may not generalize to newer criteria-sets. Additionally, since PGD ICD-11 and PGD DSM-5-TR also differ in symptom count, content, and diagnostic algorithms, findings obtained with one version of PGD may not generalize to the other.

Therefore, clarifying to what extent criteria-sets capture the same content and whether and when criteria-sets of pathological grief yield similar or different results appears important. The aim of the present contribution is to assess the comparability of different criteria-sets using a mathematical approach. Specifically, we will establish the extent to which the content of core and accessory symptoms in PGD ICD-11 , PGD DSM-5-TR , and preceding criteria-sets overlap. We derive these methods from Fried (2017) , who used a similar approach to illustrate the limited content overlap between items from seven frequently used self-report measures of depression.

We estimated content overlap between criteria-sets using the Jaccard Index, a similarity coefficient for binary data ranging from 0 (no overlap among criteria-sets) to 1 (complete overlap). It is calculated with the following formula: J =s/(u1+u2+s), where J is the Jaccard Index, s is the number of items that two criteria-sets share, and u1 and u2 are the number of symptoms unique to each criteria set. Since there is no established guideline on categorizing the strength of overlap using the Jaccard Index, we will apply the rule by Evans (1996) for the correlation coefficient as an indicator: very weak 0.00-0.19, weak 0.20-0.39, moderate 0.40-0.59, strong 0.60-0.79, and very strong 0.80-1.0 (Fried, 2017) .

Supplemental Table S1 shows the results. A first main finding is that there is moderate overlap between the most recent criteria-sets PGD ICD-11 and PGD DSM-5-TR (J =0.47). PGD shows the strongest overlap with the PGD ICD-11 beta draft (J =0.58), whereas PGD DSM-5-TR shows the strongest overlap with PCBD (J = 0.63), illustrating that they most closely resemble their direct predecessors. Both PGD ICD-11 and PGD DSM-5-TR show least overlap with CG (J =0.22 and 0.37, respectively). Overall, the mean overlap between PGD ICD-11 and PGD DSM-5-TR with all other criteria-sets is moderate (J =0.41 and 0.48, respectively). CG stands out as the diagnosis showing the least overlap with all other criteria-sets (J =0.35), whereas PCBD shows most content overlap with other criteriasets (J =0.49).

Overall, our analysis demonstrated modest content overlap between prior proposed criteria-sets and both PGD ICD-11 and PGD DSM-5-TR . Moreover, the two newest criteria-sets showed limited content overlap with each other. These findings complement prior empirical research demonstrating differences between the characteristics of different criteria-sets for pathological grief (e.g. Cozza et al., 2020) . Using a single validated instrument, such as the recently developed Traumatic Grief Inventory-Self Report Plus, to assess symptoms of different criteria-sets, may be instrumental to further elucidate when criteria-sets behave similarly or differently (Lenferink, Eisma, Smid, de Keijser, & Boelen, Lenferink, et al., 2022) .

Together, results suggest that limited content overlap could partly explain differences in findings across different criteria-sets. Two courses of action may help reduce this problem of generalizability in the future. First, we should strive for greater convergence of future diagnostic criteria-sets with presently used criteria-sets (Lenferink et al., 2021) . Second, since PGD ICD-11 uses a descriptive diagnosis without a formal diagnostic algorithm, we could investigate which PGD ICD-11 algorithm yields to the greatest convergence with past criteria-sets, and, more importantly, with PGD DSM-5-TR criteria . Some limitations warrant mention. First, this work is a mathematical exercise that complements but does not substitute empirical studies of similarities and differences between pathological grief criteria-sets. Second, we only compared core and accessory symptoms of criteria-sets. For example, we did not take into account differences in time criteria or diagnostic algorithms between proposed diagnoses. A third limitation is that for CG we split up some compound symptoms (e.g. 'Frequent intense feeling of loneliness or like life is empty or meaningless without the person who died' was separated into 'loneliness' and 'feeling life is empty/meaningless') because these symptoms were also separated in other criteriasets (see Lenferink et al., 2021 for details) . This may have led us to overestimate content overlap between CG and other criteria-sets. Fourth, one grief researcher assessed overlap between criteria sets (cf. Lenferink et al., 2021) . Multiple assessors may have yielded more reliable and replicable classifications of symptoms.

Notwithstanding these limitations, our analyses have demonstrated suboptimal comparability in the content of past and current pathological grief criteriasets. We have highlighted how this may result in problems of generalizability of findings obtained with past and current criteria-sets. This work illustrates the need for further convergence of diagnoses and empirical investigations of the similarities and differences between pathological grief diagnoses and related phenomenological characteristics.

No potential conflict of interest was reported by the author(s). 

Diagnostic and statistical manual of mental disorders

Prolonged grief disorder in section II of DSM-5: A commentary

Comparison of six proposed diagnostic criteria-sets for disturbed grief

Evaluation of the factor structure, prevalence, and validity of disturbed grief in DSM-5 and ICD-11

Commentary on the inclusion of persistent complex bereavement-related disorder in DSM-5

When does grief become pathological? Evaluation of the ICD-11 diagnostic proposal for prolonged grief in a treatment-seeking sample

Optimizing the clinical utility of four proposed criteria for a persistent and impairing grief disorder by emphasizing core, rather than associated symptoms

Traumatic grief research and care in the aftermath of the COVID-19 pandemic

Response to: Prolonged grief disorder for ICD-11: The primacy of clinical utility and international applicability

ICD-11 prolonged grief disorder criteria: Turning challenges into opportunities with multiverse analyses

Straightforward statistics for the behavioral sciences

The 52 symptoms of major depression: Lack of content overlap among seven common depression scales

The importance of harmonising diagnostic criteria-sets for pathological grief

Valid measurement of DSM-5 persistent complex bereavement disorder and DSM-5-TR and ICD-11 prolonged grief disorder

Diagnosis and classification of disorders specifically associated with stress: proposals for ICD-11

Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11

History and status of prolonged grief disorder as a psychiatric diagnosis

Complicated grief and related bereavement issues for DSM-5

Prolonged grief disorder and the cultural crisis

International statistical classification of diseases and related health problems

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.