Aarhus Prolonged Grief Disorder Scale (A-PGDS)

Abstract

The Aarhus Prolonged Grief Disorder Scale (A-PGDs; O’Connor et al., 2023) is a robust 20-item self-report instrument designed for the precise operationalization of Prolonged Grief Disorder (PGD) criteria as defined by both the ICD-11 and the DSM-5-TR. The scale development involved synthesizing diagnostic descriptions from both manuals, reviewing established PGD measures (Ito et al., 2012; Prigerson et al., 2009), and refining item wording based on feedback from focus groups of bereaved adults.

Validation data, collected from a sample of bereaved adults in Denmark, confirmed the psychometric soundness of the A-PGDs. Exploratory structural equation modeling revealed a two-factor structure for the 20 items. The scale demonstrated strong internal consistency and reliable test-retest stability, establishing it as a valid tool for assessing PGD symptomatology in research and clinical contexts.

Keywords

Antagonistic Feelings, Diagnostic and Statistical Manual, DSM-5-TR, Emotional Distress, ICD-11, International Classification of Diseases, Prolonged Grief, Separation Distress, Self-Report Instrument, Psychometrics, Bereavement, Grief Assessment.

Authors

O’Connor, Maja; Vang, Maria Louison; Shevlin, Mark; Elklit, Ask; Komischke-Konnerup, Katrine B.; Lundorff, Marie; Bryant, Richard.

Purpose

The A-PGDs was developed with the specific purpose of creating a unified and psychometrically sound scale capable of operationalizing the diagnostic criteria for Prolonged Grief Disorder (PGD) across the two primary global diagnostic systems: the World Health Organization’s ICD-11 and the American Psychiatric Association’s DSM-5-TR.

This dual focus ensures that the scale can be utilized effectively in multinational research efforts and clinical settings that adhere to different diagnostic standards, thereby promoting consistency in the measurement and identification of pathological grief reactions following the loss of a loved one.

Construct

The A-PGDs measures the psychological construct of Prolonged Grief Disorder (PGD). PGD is defined by a persistent and pervasive pattern of separation distress and preoccupation with the deceased, accompanied by specific cognitive, emotional, and behavioral symptoms that endure far beyond typical bereavement periods and cause significant impairment in daily functioning.

The 20 items are designed to capture both the core symptoms of PGD (such as yearning and preoccupation) and the associated symptoms (such as difficulty accepting the death, emotional numbness, and avoidance). The underlying structure, supported by factor analysis, validates its utility in distinguishing this specific form of persistent, distressing grief from other related conditions like major depression.

Validity

The validation study provided strong evidence for the structural and predictive validity of the A-PGDs.

Construct Validity: Results from the Exploratory Structural Equation Modeling (ESEM) confirmed the structural validity of the A-PGDs, supporting its function as a 20-item measure that loads onto two primary factors. This factor structure aligns with the theoretical distinctions between core and associated symptoms of PGD.

Content/Concurrent Validity: Analysis demonstrated that baseline PGD symptoms successfully predicted subsequent core symptoms. Importantly, among the identified factors, the associated symptoms cluster was found to be the exclusive predictor of functional impairment, emphasizing the clinical relevance of these secondary symptoms in assessing overall disability. Conversely, measures of depression were found to predict only associated symptoms, suggesting that the A-PGDs successfully measures the unique construct of prolonged grief rather than general depressive distress.

Reliability

Reliability testing confirmed that the A-PGDs provides consistent and stable measurements over time.

Internal Consistency: The scale demonstrated high internal consistency across both diagnostic frameworks. The Cronbach’s alpha coefficient for the ICD-11 PGD items was 0.88, indicating strong coherence. Similarly, the Cronbach’s alpha for the DSM-5-TR PGD items was 0.90, also reflecting robust internal consistency.

Test-Retest Reliability: Test-retest reliability was assessed in a subsample and found to be generally strong (p < .001), with large correlations (r ≥ 0.59) observed between scores across the initial test and retest phases. The stability was particularly pronounced for items belonging to the core-symptom cluster of both ICD-11 PGD (excluding self-blame) and DSM-5 PGD (r ≥ 0.72), demonstrating that the central features of prolonged grief are highly stable over time, even more so than the associated symptom clusters (r ≥ 0.59).

Factor Analysis

The factor structure of the A-PGDs was investigated using sophisticated statistical modeling techniques to ensure optimal fit to the empirical data.

Confirmatory Factor Analysis/Exploratory Structural Equation Modeling: For the PGD ICD-11 criteria, Models 2 through 4 exhibited acceptable error metrics (SRMR and RMSEA values below 0.08). While Model 4 achieved acceptable fit according to the Comparative Fit Index (CFI) and Tucker-Lewis Index (TLI), the Bayesian Information Criterion (BIC) suggested that the added complexity of Model 4 was not sufficiently justified by the gained precision (Δ14.2).

Regarding the PGD DSM-5-TR criteria, the analysis indicated a progressive improvement in model fit as the number of latent dimensions increased. Models 2 through 4 maintained acceptable error levels (SRMR and RMSEA < 0.08). Models 3 and 4 achieved acceptable fit based on CFI and TLI. The BIC suggested that the increased precision of Model 4 nearly compensated for its complexity (ΔBIC 10.9). Furthermore, a significant improvement in the four-factor model was identified by a ΔRMSEA of 0.019, supported by all other observed fit statistics.

Instrument

Test Type: Original

Format: Items are rated using a 5-point Likert-scale and supplementary yes/no response options.

Language Available: English and Danish.

Population Group: Human (Male and Female).

Age Group: Adulthood (18 years and older).

Population Details: The study participants were bereaved adults located in Denmark.

Test Methodology: The methodology included Test Validity, Concurrent Validity, Construct Validity, Content Validity, Test Reliability, Internal Consistency, Test-Retest Reliability, Factor Analysis, Confirmatory Factor Analysis, and Structural Equation Modeling.

Keywords

Prolonged Grief Disorder (PGD), Bereavement, Psychopathology, Structural Equation Modeling, Psychometrics, Self-Report, Denmark, Grief Assessment, Internal Consistency, Diagnostic Instrument.

Authors

Author ORCID Identifier: 0000-0001-7031-4070 (Vang, Maria Louison); 0000-0003-0162-5750 (Lundorff, Marie)

Affiliation Email addresses: [email protected]

Correspondence Address: Aarhus University, Department of Psychology, Unit for Bereavement Research, Aarhus, Denmark.

Permissions & Fee and Test Year

Permissions: May be used for Research/Teaching.

Fee: No fee.

Test Year: 2023.

Website: creativecommons.org/licenses/by/4.0/deed.en

Reference’s

• Ito, M., Nakajima, S., Fujisawa, D., Miyashita, Y., Konishi, S., Akechi, T., & Holland, J. C. (2012). The Inventory of Complicated Grief-Japanese Version: reliability and validity study. Psychiatry and Clinical Neurosciences, 66(2), 159-166.

• O’Connor, M., Vang, M. L., Shevlin, M., Elklit, A., Komischke-Konnerup, K. B., Lundorff, M., & Bryant, R. A. (2023). Development and validation of the Aarhus Prolonged Grief Disorder Scale (A-PGDs). Journal of Affective Disorders, 342, 199-205.

• Prigerson, H. G., Maciejewski, P. K., & Maercker, A. (2009). Prolonged grief disorder: Psychometric properties of an interview-based measure. Journal of Affective Disorders, 115(1-2), 183-188.

Items of the Aarhus Prolonged Grief Disorder Scale (A-PGDs)

The following items are part of the Aarhus Prolonged Grief Disorder Self-Report Scale, designed for use with ICD-11 and DSM-5 Tr PGD (The Aarhus PGD-scale).

Please consider the name of the person you lost and respond to the questions in relation to your loss.

Each item is scored on the following 5-point scale:

1. Not at all

2. A little

3. To some extent

4. Very much

5. Overwhelmingly

Item no.	Item formulation
1.	Have you longed for [deceased’s name] during the past month?
2.	Have you during the past month found yourself preoccupied with thoughts of [deceased’s name] even when you did not want to be thinking about them?
3.	Have you had feelings of sadness or sorrow during the past month?
4.	Have you felt guilty during the past month?
5.	Have you felt angry during the past month?
6.	During the past month, has it been hard for you to believe that [deceased’s name] is dead?
7.	Have you blamed yourself for your loss during the past month?
8.	During the past month, have you had trouble accepting that [deceased’s name] is dead?
9.	During the past month, have you felt that you have lost a part of yourself? (e.g., feeling as though a part of you has died)
10.	During the past month, have you been unable to experience positive emotions?
11.	During the past month, have you felt emotionally numb? (e.g., having difficulties with feeling emotions as you used to do, being emotionally stunned)
12.	Have you had difficulty engaging in social or other activities during the past month?
13.	Have you felt loneliness during the past month?
14.	During the past month, have you tried to avoid reminders that [deceased’s name] is dead? (e.g., avoiding certain thoughts, feelings, places, music, conversation topics, etc. or keeping yourself constantly going)
15.	During the past month, have you felt that life is meaningless since [deceased’s name] has died?
16.	Have you felt bitterness during the past month?
17.	Overall, have these difficulties led to a decline in your level of functioning? (i.e., your ability to function in everyday life)

If you answered “A little” (2) or higher to question 17, have you then experienced this every day or almost every day? Yes/No

Item no.	Item formulation	Response
17a.	Does this apply in relation to your work/study/daily tasks?	Yes/No
17b.	Does this apply to your social life?	Yes/No
17c.	Does this apply to your family life/domestic obligations?	Yes/No
17d.	Does this apply to other areas than those mentioned?	Yes/No
18)	Have any of your acquaintances expressed concern about your grief reaction? (e.g., that they feel that it exceeds what they consider normal in relation to your social, cultural or religious norms).	Yes/No
19)	Are you worried about your own grief reaction, including that it is more severe or intense than you expected? (e.g., compared to the people you surround yourself with or what you think is normal).	Yes/No
20)	Would you say that you felt this way during the last 6 months? (Please answer this question in relation to your total response to the questions above).	Yes/No

Note on Marking:

• Grey: only ICD-11.

• Bold: only DSM-5-TR.

• No marking: symptoms included in both ICD-11 and DSM-5-TR.