Grief and delivering a statement in court: a longitudinal mixed-method study among homicidally bereaved people

ABSTRACT

Background: Participating in a criminal trial may increase the likelihood of developing psychopathology. In 2021, people bereaved by a plane disaster (flight MH17) had the opportunity to deliver a victim personal statement (VPS) in Dutch court.

Objective: This longitudinal mixed-method study examined different aspects of 84 bereaved people’s experiences with VPS delivery.

Method: Motivations to deliver, or not deliver, an oral VPS were examined qualitatively using thematic content analysis. Whether background and loss-related variables were related to the decision to deliver a VPS was examined using binary logistic regression analyses. Between-group (delivered VPS vs. did not) and within-group (pre- vs. post-VPS) comparisons were made regarding prolonged grief disorder (PGD), posttraumatic stress disorder (PTSD), and depression levels using t-tests and paired t-tests.

Results: Bereaved people were most frequently motivated to deliver an oral VPS to describe the impact of the incident, while those who did not deliver an oral VPS commonly wanted to protect themselves from the perceived emotional burden. None of the correlates – i.e. biological sex, age, level of education, number of losses, and (closest) relationship to the deceased – were related to the decision to deliver a VPS. Lastly, significantly higher PGD, PTSD, and depression levels were reported by people who delivered a VPS than those who did not, before and after the court hearing. No significant within-group differences were found over time.

Conclusions: Professionals may provide emotional support to bereaved people who want to deliver a VPS and manage their expectations if they want to deliver a VPS for the purpose of symptom reduction. Future research may benefit from examining other ways in which VPS delivery might have beneficial or detrimental effects for specific individuals. Overall, implementing VPS delivery in court on the basis of emotional restoration remains empirically unsupported, if defined as a reduction in psychopathological levels.

HIGHLIGHTS

• We are the first to examine if statement delivery changes grief-related distress.

• Statement delivery did not significantly change grief-related distress.

• Defining emotional restoration as a decrease in psychopathology remains unsupported.

Antecedentes: El participar en un juicio penal puede aumentar la probabilidad de desarrollo de psicopatología. En el 2021, las personas en duelo por un accidente aéreo (vuelo MH17) tuvieron la oportunidad de presentar una declaración personal de víctima (VPS por sus siglas en inglés) ante un tribunal holandés.

Objetivo: Este estudio longitudinal de métodos mixtos examinó los diferentes aspectos de la experiencia de 84 personas en duelo con la entrega de VPS.

Método: Las motivaciones para realizar o no una VPS oral se examinaron cualitativamente usando análisis de contenido temático. Se examinó mediante análisis de regresión logística binaria si las variables de antecedentes y relacionadas con las pérdidas estaban relacionadas con la decisión de realizar una VPS. Se realizaron comparaciones entre grupos (los que entregaron VPS vs los que no) e intragrupo (antes vs después de VPS) con respecto al trastorno por duelo prolongado (PGD por sus siglas en inglés), trastorno de estrés postraumático (TEPT) y niveles de depresión utilizando pruebas de t y pruebas de t pareadas.

Resultados: Las personas en duelo se sintieron con mayor frecuencia motivadas a realizar una VPS oral para describir el impacto del incidente, mientras que aquellos que no realizaron una VPS oral comúnmente querían protegerse de la carga emocional percibida. Ninguno de los correlatos por ej., sexo biológico, edad, nivel educacional, números de pérdidas y relación (más cercana) con el fallecido estuvieron relacionadas con la decisión de entregar una VPS. Por último, las personas que realizaron una VPS reportaron mayores niveles de PGD, TEPT y depresión que aquellas que no lo hicieron, antes y después de la audiencia judicial. No se encontraron diferencias significativas dentro del grupo a lo largo del tiempo.

Conclusiones: Los profesionales pueden brindar apoyo emocional a las personas en duelo que quieran realizar una VPS y manejar sus expectativas si desean realizar una VPS con el fin de reducir sus síntomas. Investigaciones futuras podrían beneficiarse al examinar otras formas en las que la entrega de VPS pudiera tener efectos beneficiosos o perjudiciales para individuos específicos. En general, la implementación de VPS en los tribunales sobre la base de la restauración emocional sigue sin tener respaldo empírico, si se define como una reducción en los niveles psicopatológicos.

KEYWORDS:

• Prolonged grief
• posttraumatic stress
• depression
• victim impact statement
• bereavement
• trauma
• homicide
• motivation

PALABRAS CLAVE:

• Duelo prolongado
• estrés traumático
• depresión
• declaración de impacto en la víctima
• duelo
• trauma
• homicidio
• motivación

1. Introduction

Losing a loved one as a result of a violent act, e.g. homicide (i.e. a person killed by another person; Smit et al., 2012), is considered a risk factor for developing bereavement-related psychopathology (Djelantik, Smid, et al., 2020; Kristensen et al., 2012). After non-violent bereavement, a relatively small number of people develop psychological complaints, such as prolonged grief disorder (PGD) (10%; Lundorff et al., 2017), posttraumatic stress disorder (PTSD), or depression (12–16% and 22%, respectively; O’Connor, 2010; Onrust & Cuijpers, 2006). After violent bereavement, however, almost half of all bereaved people develop PGD (49%; Djelantik, Smid, et al., 2020), 19–71% PTSD (with current PTSD rates ranging from 5 to 39%), and 8–54% report current depression (Milman et al., 2018; van Denderen et al., 2015). Certain characteristics of violent loss are positively related to the likelihood of developing psychopathology after loss, e.g. experiencing multiple losses or the loss of a close relation (i.e. child, partner/spouse, parent or sibling) compared to the loss of more distant relations, such as other family members or friends (Heeke et al., 2019; Kokou-Kpolou et al., 2020).

According to the DSM-5-TR, PGD is characterized by symptoms such as severe longing for, or preoccupation with, the deceased (American Psychiatric Association [APA], 2022). PGD can be diagnosed if grief reactions persist for at least a year post-loss and significantly hinder daily life (APA, 2022). PGD, PTSD, and depression often co-occur, yet latent trajectory (Djelantik et al., 2022; Lenferink et al., 2020), latent class (Djelantik et al., 2017; Djelantik, Robinaugh, et al., 2020; Lenferink et al., 2017), confirmatory factor (Boelen et al., 2010), network (Djelantik, Robinaugh, et al., 2020), and cross-lagged analytic research (Lenferink et al., 2019) suggests they are distinct. What differentiates the symptomatology of these disorders is that PGD stems from separation distress (Boelen et al., 2023; Lenferink et al., 2023), PTSD from trauma (Shear, 2015), and depression likely from low self-esteem (Ottenbreit & Dobson, 2004; Shear, 2015).

PGD, PTSD, and depression levels of bereaved people may be altered by delivering a victim impact statement (VIS) in court. Since the 1970s, many common law countries, such as the United States, Australia, England and Wales, allow victims and their relatives to deliver a VIS about the impact of the crime (Roberts & Manikis, 2013). By contrast, most civil law countries¹ do not allow for VIS delivery. A notable exception is the Netherlands. Dutch legislation allows for written VIS delivery in all criminal cases and oral VIS delivery if the defendant is being prosecuted for crime sanctioned by at least eight years of imprisonment (Kunst et al., 2022). In 2016, VIS delivery was expanded to allow for the discussion of other topics, e.g. the victim’s opinion of the sentence (Kunst et al., 2022). Hence, the present study makes a distinction between VIS and Victim Personal Statement (VPS). VIS refers to the original version of the VIS, while VPS refers to the expanded version of the VIS.

Proponents of VIS delivery in the Netherlands have argued that relaying one’s thoughts and feelings concerning the crime in court may contribute to a person’s emotional restoration (Pemberton & Reynaers, 2011). In the Netherlands, a main reason for introducing VIS delivery in the Dutch Code of Criminal Proceedings was the opportunity for emotional restoration (Lens et al., 2010). However, no definition of ‘emotional restoration’ has been provided. Opponents of VIS delivery have warned that VIS delivery could have detrimental health effects for people, for example, when other trial participants respond unsympathetically to their statements (Pemberton & Reynaers, 2011). This unsympathetic response may result in secondary victimization. There is no consensus regarding the definition of secondary victimization (Pemberton, 2022). However, secondary victimization is usually assumed to involve harm caused by criminal proceedings in the field of victimology (Wemmers, 2013). Harm is often interpreted as an increase in psychopathological complaints in this context (e.g. Connolly & Gordon, 2015; Herman, 2003; Orth, 2002; Stretesky et al., 2010; Wemmers, 2013).

Despite the assumed emotional restorative potential of VPS delivery, it is largely unknown why homicidally bereaved people choose to deliver a VPS. Also, it is unclear what factors increase the likelihood of VPS delivery, and whether VPS delivery alters psychopathology levels. To the best of our knowledge, only one study addressed these issues (Lens et al., 2010). Lens et al.'s (2010) mixed-method study provided insights in motivations to deliver, or not deliver, an oral VIS reported by 119 Dutch crime victims exposed to various offence types. The four most frequently stated motivations to deliver an oral VIS were: (1) communicating the consequences (regarding various aspects of life), (2) influencing the punishment, (3) receiving recognition, and (4) (partially) processing the crime (i.e. stimulating closure). The four most frequently stated motivations to not deliver an oral VIS were: (1) self-protection, (2) fearing confrontation with the defendant, (3) low severity of the crime, and (4) perceiving no added value. Lens et al. also found that women more often than men have the intention to deliver a VIS. Lastly, they found that VIS delivery did not significantly change psychological complaints, contrary to the beliefs of proponents and opponents of VIS delivery. Specifically, the group that delivered a VIS reported significantly higher PTSD levels before and after the court hearing than the group that did not. For both groups a similar, significant decrease in PTSD levels was observed. Nonetheless, it is impossible to generalize these findings to all homicidally bereaved people, because Lens et al.’s study included only five homicidally bereaved people.

Arguably, bereaved crime victims may have other reasons to deliver, or not deliver, an oral VPS than non-bereaved crime victims. For example, homicidally bereaved people reported to perceive oral VIS delivery as a way to honour their deceased loved one(s) (Lens et al., 2010). Also, in a recent report the motivations of 10 bereaved crime victims to deliver an oral VPS were qualitatively examined. The following bereavement-specific motivations were mentioned: wanting to talk about the deceased, ensuring the victim is perceived as a person, addressing the victim’s innocence, and perceiving oral VPS delivery as the last act for the deceased (Kragting et al., 2022). Moreover, due to the expansion of the VIS, people might express more varied motivations to deliver, or not deliver, an oral VPS than an oral VIS. Furthermore, the one study (Lens et al., 2010) examining motivations of bereaved and non-bereaved crime victims to deliver, or not deliver, an oral VIS, as well as related topics, included only five bereaved crime victims. Thus, there is a need for more research examining motivations to deliver, or not deliver, an oral VPS, as well as related topics, in bereaved samples.

Accordingly, the first aim of the present study was to examine motivations to deliver, or not deliver, an oral VPS expressed by people bereaved by a large plane disaster. The second aim entailed exploring whether certain background and loss-related characteristics – i.e. biological sex, age, level of education, singular loss vs. multiple losses, and (closest) relationship to the deceased – were associated with the decision to deliver a VPS. The third aim concerned examining whether PGD, PTSD and depression levels (hereafter: psychopathology levels) differed between bereaved people who delivered a VPS and those who did not, as well as whether a significant change took place in psychopathology levels when comparing levels reported before and after the court hearing in both groups (i.e. delivered vs. did not) separately.

2. Material and methods

2.1. Procedure

On the 17th of July 2014, 298 civilians lost their lives, when passenger flight MH17 was hit by a missile above Ukraine (Dutch Safety Board, 2015). In 2022, three of the four suspects were sentenced to life in prison for intentionally shooting down flight MH17, and thereby killing the 283 passengers and 15 crew members (Ministry of Justice and Security, n.d.). After seven years, in 2021, people bereaved by the disaster could deliver a VPS concerning the crime.

The current longitudinal mixed-method study is part of an ongoing longitudinal survey study examining the psychological impact of losing a loved one due to the MH17 disaster (Buiter et al., 2022; Lenferink et al., 2017; Lenferink et al., 2019; Lenferink et al., 2020). The Ethical Committee of Psychology of the University of Groningen (EC-BSS) approved the study (ID: PSY-1920-S-0171). All participants gave written consent.

Participants were recruited at the start of the longitudinal study and once again before the start of the criminal trial (e.g. by Victim Support the Netherlands and the MH17 Disaster Foundation; for more details, see Buiter et al., 2022; Lenferink et al., 2017). For the present study, Dutch and non-Dutch participants of the ongoing longitudinal survey study were invited to participate if they had given permission to be contacted for follow-up measurement. The invitations were sent via e-mail or post. The online survey was available in Dutch and English.

Data were collected before and after the participants had the opportunity to deliver a VPS in court, hereafter referred to as pre-VPS and post-VPS respectively. VPSs were delivered in court from 6 September until 8 November 2021. At pre-VPS, data collection took place between 24 February and 25 May 2021. At post-VPS, data were collected between 22 November 2021 and 3 January 2022.

2.2. Participants

People could participate if they were an adult who lost a partner/spouse, relative, and/or friend due to the MH17 disaster, and had sufficient knowledge of Dutch or English. In total, 129 people participated at pre-VPS and 103 people at post-VPS. At post-VPS, the participants were asked: ‘Did the public prosecution service give you the opportunity to choose whether you want to deliver a VPS?’. Fifteen of the 103 people at post-VPS stated they had not been given the opportunity to deliver a VPS (could be due to e.g. a maximum of three people being allowed to speak per victim), and two participants dropped out before answering this question. Two of the remaining 86 post-VPS participants had not participated at pre-VPS. The final sample consisted of 84 participants who had the opportunity to deliver a VPS and completed both measurement occasions. Therefore, 84 participants were included in the analyses. Of the 84 participants, 24 participants delivered an oral VPS, 12 participants delivered a written VPS, and 48 participants delivered neither an oral nor a written VPS.

2.3. Measures

2.3.1. Motivations

At post-VPS, the participants were asked: ‘Did you yourself deliver an oral VPS or did you draw up a written VPS?’. The participants chose one of the following answer options: (1) yes, I drew up a written VPS, (2) yes, I delivered an oral VPS, (3) no, a loved one drew up a written VPS, (4) no, a loved one delivered an oral VPS, and (5) no, neither I nor my loved ones delivered a VPS, in written or in oral form. The answer options were dichotomized, with 1 and 2 equalling yes ( = 1), and 3 through 5 equalling no ( = 0). Subsequently, participants were asked to state their motivation to deliver, or not deliver, an oral VPS (‘I did/did not deliver an oral VPS, because … ’). The participants were instructed to state one motivation for their decision. Answers ranged from two words to multiple sentences.

2.3.2. Background and loss-related characteristics

The following background characteristics were measured: biological sex (1 = male, 2 = female), age (in years), and level of education (1 = primary school, 2 = secondary school, 3 = secondary vocational education, 4 = university (of applied sciences)). The loss-related characteristics that were measured were: the number of losses due to the plane disaster and the relationship to the deceased (1 = child, 2 = partner/spouse, 3 = parent, 4 = sibling, 5 = other).

2.3.3. Traumatic Grief Inventory Self-Report Plus (TGI-SR+)

The TGI-SR+ was used to assess DSM-5-TR PGD severity (Lenferink, Eisma, et al., 2022). The TGI-SR+ includes 22 items representing symptoms of four different diagnostic criteria sets for disturbed grief. For each diagnostic criteria set a different combination of items should be analysed. Eleven items were used to assess 10 DSM-5-TR PGD symptoms (APA, 2022). A PGD total score was calculated by adding the scores on nine items (items: 1, 3, 6, 9, 10, 11, 18, 19, 21) and the highest score indicated on items 2 and 8 (Lenferink, Eisma, et al., 2022). An example item is: ‘In the past month, I had trouble accepting the loss’. The participants were asked to score each item on a 5-point Likert scale ranging from 1 = never to 5 = always. The PGD total score ranges from 10 to 50. A score ≥ 33 is indicative of probable PGD (Lenferink, Eisma, et al., 2022). The instructions of the TGI-SR+ were modified to refer to loss(es) corresponding to the plane disaster. For participants who lost multiple loved ones, the following instruction was given:

Please fill in the questionnaire keeping in mind the loss that is most on your mind and/or is experienced as most stressful. In case multiple losses are equally often on your mind and/or are equally stressful, please select one person while completing the questionnaire.

Research demonstrated that the TGI-SR+ has good psychometric properties (Kokou-Kpolou et al., 2022; Lenferink, Eisma, et al., 2022). Cronbach’s alphas of the DSM-5-TR PGD items in this study were .93 at pre-VPS and .92 at post-VPS.

2.3.4. The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5)

The 20-item PCL-5 was used to assess DSM-5 PTSD severity (Blevins et al., 2015; Boeschoten et al., 2014; Weathers et al., 2013). An example item is: ‘In the past month, how much were you bothered by repeated, disturbing dreams of the stressful experience?’. Every item was scored on a scale from 0 = not at all to 4 = extremely. A PTSD total score was calculated by summing the scores on all items (range: 0–80). A score > 32 is indicative of probable PTSD (Krüger-Gottschalk et al., 2017). The instructions of the PCL-5 were modified to refer to the loss(es) corresponding to the plane disaster. The PCL-5 has excellent psychometric properties (Blevins et al., 2015). Cronbach’s alphas at pre- and post-VPS were both .95.

2.3.5. Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR₁₆)

The 16-item QIDS-SR₁₆ was used to assess DSM-IV depression severity (Rush et al., 2003). An example item is: ‘sleeping too much’, which the participant had to rate from zero (e.g. I sleep no longer than 7–8 hr/night, without napping during the day) to three (e.g. I sleep longer than 12 hr in a 24-hour period including naps), based on experiences of the past week (Rush et al., 2003). The scores on nine items – representing symptom criterion domains – were summed to create a total depression score (range: 0–27) (Rush et al., 2003). A score ≥ 16 is indicative of probable depression (Rush et al., 2003). The QIDS-SR₁₆ has good psychometric properties (Rush et al., 2003). Cronbach’s alphas in this study were .83 at pre-VPS and .82 at post-VPS.

2.4. Analyses

Following prior research (Lenferink et al., 2018), thematic content analysis (Corbin & Strauss, 2008) was employed to analyse the open-ended question concerning the participant’s motivation to deliver, or not deliver, an oral VPS. In doing so, irrelevant information was first deleted from the response (i.e. unrelated to what is being investigated). Subsequently, the remaining sentence(s) were split up in smaller, relevant and logical components (i.e. units). Next, each unit was analysed separately and labelled with a specific sub-theme. In this context, the label indicates the unit’s contents (open coding; Corbin & Strauss, 2008) and was created based on the words of the participant(s) (i.e. inductively). Then, labels were compared based on their resemblances and differences (re-analyzation; Corbin & Strauss, 2008). To increase coherency, re-labelling was allowed. Re-labelling was when new labels were created for the sub-themes in the words of the researcher (where needed), replacing the old ones. The last step (axial coding; Corbin & Strauss, 2008) involved the creation of overarching themes. Regarding missing data, one participant did not state his/her motivation and two participants misinterpreted the question.

To determine whether the participant’s decision to deliver a VPS is associated with background and loss-related characteristics, binary logistic regression analyses were performed using IBM SPSS statistics (version 28.0.1.0; IBM Corp, 2021). Biological sex was coded as 0 = male and 1 = female. The remaining background and loss-related categorical variables that have more than two categories were dichotomized to avoid computational difficulties, due to the small sample size. Level of education of the participant was recoded as 0 = other than university and 1 = university (of applied sciences), and the number of losses due to the plane disaster as 0 = single loss and 1 = multiple losses. Following prior research (Lenferink et al., 2020), (closest) relationship to the deceased was recoded as 0 = other than partner/spouse or child and 1 = partner/spouse or child. Univariate binary logistic regression analyses were performed to investigate possible associations between someone’s decision to deliver a VPS (0 = no, 1 = yes) and biological sex, age, level of education, number of losses, as well as the (closest) relationship to the deceased. Furthermore, if more than one variable was found to be significantly associated with someone’s decision to deliver a VPS in the univariate analyses (p < .05), then these variables would be included simultaneously in a multivariate logistic regression model.

It was initially planned to perform ANOVAs and ANCOVAs to examine differences between groups regarding psychopathology levels, as well as differences over time within groups using IBM SPSS statistics (version 28.0.1.0; IBM Corp, 2021). However, as several assumptions of ANCOVA were violated (e.g. homogeneity of regression slopes), t-tests were performed to examine differences between groups (delivered vs. did not) regarding pre- and post-VPS psychopathology levels. After this, paired t-tests were performed to investigate whether a significant change took place in psychopathology levels when comparing pre- and post-VPS levels within each group (delivered vs. did not). There were no missing data regarding PGD, PTSD, and depression levels at either measurement occasion. The p-value indicating significance (p < .05) was adjusted for multiple testing (p < .05/12 = .004). Cohen’s d effect sizes were calculated, with a distinction being made between small (0.2 ≤ d < 0.5), medium (0.5 ≤ d < 0.8), and large (d ≥ 0.8) effect sizes (Cohen, 1988).

3. Results

3.1. Participants

Sample characteristics are shown in Table 1. The majority of participants was female and obtained a university degree. Most participants experienced multiple losses, of which the (closest) relationship to the deceased was most commonly sibling or child.

Table 1. Sample characteristics (N = 84).

Background and loss-related characteristics	Total sample	Delivered VPS (written or oral)	Did not deliver VPS
Biological sex, N (%)
Male	35 (42)	18 (50)	17 (35)
Female	49 (58)	18 (50)	31 (65)
Age, M (SD)	59.5 (13.5)	58.5 (13.9)	60.3 (13.3)
Level of education, N (%)
University (of applied sciences)	59 (70)	26 (72)	33 (69)
Secondary vocational education	9 (11)	3 (8)	6 (13)
Secondary school	16 (19)	7 (19)	9 (19)
Primary school	0 (0)	0 (0)	0 (0)
Number of losses due to MH17 plane disaster, N (%)
One	33 (39)	18 (50)	15 (31)
Two	32 (38)	11 (31)	21 (44)
Three	7 (8)	2 (6)	5 (10)
Four	9 (11)	3 (8)	6 (13)
Five	3 (4)	2 (6)	1 (2)
Closest deceased relative is my … ^a, N (%)
Child	31 (37)	15 (42)	16 (33)
Partner/spouse	6 (7)	4 (11)	2 (4)
Parent	11 (13)	6 (17)	5 (10)
Sibling	32 (38)	10 (28)	22 (46)
Other	4 (5)	1 (3)	3 (6)
Mode of delivery, N (%)
Oral VPS	24 (29)
Written VPS	12 (14)
Someone else, oral VPS	14 (17)
Someone else, written VPS	5 (6)
None	29 (35)

Note: M = Mean; SD = Standard Deviation; VPS = Victim Personal Statement.

^a Only the participant’s closest relationship to the deceased was included in the analysis in case of multiple losses, ordered from child, partner/spouse, parent, sibling to other.

At pre-VPS, participants scored on average 24.6 (SD = 8.8) on PGD, 17.1 (SD = 15.9) on PTSD, and 6.9 (SD = 5.2) on depression. In total, 18 participants (21%) scored above the cut-off score indicating probable PGD, 12 (14%) probable PTSD, and eight (10%) probable depression. Twenty participants (24%) scored at least above one cut-off score. Of these 20 participants, the scores of 12 participants (14%) indicated probable comorbidity. One participant (1%) scored above the cut-off scores indicating probable PGD and depression, five (6%) scored above the cut-off scores indicating probable PGD and PTSD, and six (7%) scored above all cut-off scores indicating probable PGD, PTSD, and depression.

At post-VPS, participants scored on average 24.0 (SD = 8.5) on PGD, 16.2 (SD = 15.2) on PTSD, and 6.9 (SD = 5.0) on depression. In total, 13 participants (16%) scored above the cut-off score indicating probable PGD, nine (11%) probable PTSD, and six (7%) probable depression. Fifteen participants (18%) scored at least above one cut-off score. Of these 15 participants, the scores of eight participants (10%) indicated probable comorbidity. Three participants (4%) scored above the cut-off scores indicating probable PGD and PTSD, and five (6%) scored above all cut-off scores indicating probable PGD, PTSD, and depression.

3.2. Motivations to deliver an oral VPS

Twenty-four participants stated their motivation to deliver an oral VPS. Table 2 displays the number of participants who stated a certain motivation to deliver an oral VPS and the number of units that received a specific label. An example of each sub-theme is provided in Appendix A. The main themes were ‘expression-related’ (i.e. topics related to the consequences of the incident) and ‘deceased-related’ (i.e. wanting to deliver an oral VPS for their loved one(s)). The most frequently mentioned sub-themes were ‘describing the impact of the incident’ (e.g. male, 54 years old, who lost at least their parent; ‘I wanted to explain what impact bringing down the plane has’), ‘honouring the deceased’ (e.g. a female, 69 years old, who lost their child; ‘[It] was the last thing we could do for our daughter’), and ‘spokesperson of the deceased’ (e.g. female, 43 years old, who lost their spouse; ‘I wanted my voice to be heard for my husband’).

Table 2. Motivations to deliver an oral victim personal statement (N = 24).

Main theme	Sub-theme	Participants		Units
		n	%	n	%
Expression-related	Describing the impact of the incident	14	58	19	35
	Addressing the behaviour of defendants/third parties	2	8	2	4
	Directly addressing the court	1	4	1	2
	Expressing gratitude towards the court	1	4	1	2
Deceased-related	Honouring the deceased	6	25	6	11
	Spokesperson of the deceased	5	21	5	9
	Obligation to the deceased	3	13	3	6
	Remembering the deceased	2	8	3	6
Trial-related	Influencing the punishment	4	17	4	7
	Seeking justice	1	4	1	2
	Contributing to the criminal trial	1	4	1	2
Receiving recognition	Recognition of the participant’s story	4	17	5	9
	Recognition of the harm	1	4	1	2
Adjustment to the loss	Facilitating the mental processing of the incident	1	4	1	2
	Acceptance	1	4	1	2

3.3. Motivations to not deliver an oral VPS

Sixty participants did not deliver an oral VPS. Table 3 displays the number of participants who stated a certain motivation as a response to why they decided not to deliver an oral VPS, as well as the number of units that received a specific label. An example of each sub-theme is provided in Appendix B. The main themes were ‘no added value’ (i.e. stating how useless oral VPS delivery is in different ways) and ‘protection’ (i.e. perceiving oral VPS delivery as an emotional burden for themselves or others). The most commonly mentioned sub-themes were ‘self-protection’ (e.g. a female, 35 years old, who lost her parent; ‘Because I was afraid of the emotional consequences’), ‘no added value to the participant’² (e.g. a male, 41 years old, who lost (at least) his parent; ‘It would not be of benefit to me’), and ‘loss/crime remains unchanged’ (e.g. a male, 65 years old, who lost (at least) his child; ‘We do not get anyone back by doing it’).

Table 3. Motivations to not deliver an oral victim personal statement (N = 60).

Themes	Sub-themes	Participants		Units
		n	%	n	%
No added value	No added value to the participant	13	22	15	13
	Loss/crime remains unchanged	8	13	9	8
	No added value to the case	7	12	8	7
	No added value in general	4	7	5	4
	No added value to the punishment	2	3	2	2
Protection	Self-protection	23	38	27	24
	Protection of others	1	2	1	1
Other means of making a statement	Another representative spoke	6	10	6	5
	Personal view shared by someone else	3	5	3	3
Moving past the incident	Focus on present/future	4	7	4	4
	Downward comparison	2	3	2	2
	Not wanting revenge	2	3	2	2
	Distancing oneself from the criminal trial	2	3	2	2
Execution obstacles	Media attention/protecting one’s privacy	4	7	6	5
	Difficulty expressing oneself	3	5	4	4
	Public speaking	2	3	2	2
	Not knowing what to address	1	2	1	1
Boundaries to participation	Restrictions related to the country of residence	3	5	3	3
	Restrictions related to personal/professional circumstances	3	5	3	3
	Physical complaints	2	3	2	2
The defendants	Unwilling to share emotions/consequences with the defendants	1	2	3	3
	Absence of defendants	1	2	1	1
Other	Negative perspective on the victim personal statement	2	3	2	3

Note: One participant did not state his/her motivation and two participants misinterpreted the question.

3.4. Background and loss-related correlates of the decision to deliver a VPS

The results of the univariate binary logistic regression analyses can be found in Appendix C. None of the background or loss-related characteristics were significantly related to the likelihood of delivering a VPS. Consequently, no multivariate binary logistic regression analysis was performed.

3.5. Between-group differences in psychopathology levels for people who did versus did not deliver a VPS

The results of the t-tests can be found in Table 4. Significantly higher psychopathology levels were reported by the participants who delivered a VPS than participants who did not deliver a VPS at pre- and post-VPS. The effect sizes of all t-tests were medium.

Table 4. Between-group differences in psychopathology levels for people who did versus did not deliver a VPS (N = 84).

Levels of …	Did not deliver VPS (N = 48)		Delivered VPS (written or oral) (N = 36)		t	p	Cohen’s d
	M	SD	M	SD
Pre-VPS
PGD	22.02	7.24	28.03	9.68	−3.125	.003	.70
PTSD	12.13	12.17	23.75	17.87	−3.362	.001	.76
Depression	5.31	4.30	9.03	5.55	−3.334	.001	.75
Post-VPS
PGD	21.40	6.59	27.44	9.47	−3.283	.002	.74
PTSD	11.46	10.67	22.58	17.97	−3.303	.002	.75
Depression	5.48	4.08	8.86	5.42	−3.266	.002	.71

Note: VPS = Victim Personal Statement; M = Mean; SD = Standard Deviation; PGD = Prolonged Grief Disorder; PTSD = Posttraumatic Stress Disorder.

3.6. Within-group differences in pre- and post-VPS psychopathology levels for the group that did and the group that did not deliver a VPS

The results of the paired t-tests can be found in Table 5. None of the within-group comparisons were significant, meaning that there were no significant differences in psychopathology levels within either group (delivered vs. did not) over time (pre- vs. post-VPS).

Table 5. Within-group differences in pre- and post-VPS psychopathology levels for the group that did and the group that did not deliver a VPS (N = 84).

Levels pre- vs. post-VPS	M diff	SD	t	df	p	Cohen’s d
PGD
Did not deliver VPS	.63	4.38	.989	47	.328	.14
Delivered VPS (written or oral)	.58	3.92	.892	35	.378	.15
PTSD
Did not deliver VPS	.67	6.84	.675	47	.503	.10
Delivered VPS (written or oral)	1.17	7.53	.930	35	.359	.16
Depression
Did not deliver VPS	−.17	2.88	−.401	47	.691	−.06
Delivered VPS (Written or oral)	.17	2.47	.405	35	.688	.07

Note: VPS = Victim Personal Statement; M diff = Mean difference; SD = Standard Deviation; df = degrees of freedom; PGD = Prolonged Grief Disorder; PTSD = Posttraumatic Stress Disorder.

4. Discussion

Little research has examined why homicidally bereaved people deliver an oral VPS, what factors increase the likelihood of VPS delivery, and whether VPS delivery is beneficial in any way. The present study aimed to reduce this knowledge gap by examining (1) motivations to deliver, or not deliver, an oral VPS, (2) correlates of VPS delivery, and (3) changes in psychopathology levels. In doing so, the present study relied on a sample of 84 people who lost loved ones due to a plane disaster that took place on July 17th 2014.

Concerning the first aim, bereaved people often delivered an oral VPS to discuss the impact of the crime or speak for/about their loved one(s). Bereaved people who did not deliver an oral VPS did not perceive any added value (i.e. no personal benefits) or wanted to protect themselves (or others) from the emotional burden they associated with oral VPS delivery. In the present study, a larger variation in motivations was found when compared to prior research (Lens et al., 2010). This could be explained by the expansion of the VIS, as it allows victims and bereaved people to discuss whatever they want (Kunst et al., 2022). Previously they were limited to discuss the impact of the crime (Roberts & Manikis, 2013). The four most frequently mentioned motivations to deliver an oral VIS found in a previous study, i.e. communicating the consequences, influencing the punishment, receiving recognition, and (partially) processing the crime (Lens et al., 2010), were mentioned in the present study. Moreover, two of the four most frequently mentioned motivations to not deliver an oral VIS in prior research, i.e. self-protection and perceiving no added value (Lens et al., 2010), were also mentioned in the present study. However, fearing confrontation with the defendant and low severity of the crime were not mentioned as motivations to not deliver an oral VPS in the present study. It is possible that these were not applicable. To elaborate, the defendants were absent, and all participants were bereaved by the same, objectively severe, crime. Furthermore, we are the first to find that bereaved people have their own category of motivations for delivering an oral VPS. Notably, in a recent report motivations of 10 bereaved crime victims to deliver an oral VPS were qualitatively explored (Kragting et al., 2022). The deceased-related motivations observed in the present study were similar to those stated in the report. Future studies may benefit from in-depth interviews in which motivations to deliver, or not deliver, a VPS in general are examined. During in-depth interviews, bereaved people can elaborate on their perceptions of, and experiences with, oral and written VPS delivery. This may provide insight into what factors of VPS delivery might be related to psychopathology.

Regarding the second aim, biological sex, age, level of education, number of losses, and (closest) relationship to the deceased were not significantly related to the decision to deliver a VPS. The findings suggest that none of the examined background and loss-related characteristics increased or decreased the chance of bereaved people delivering a VPS. Contrastingly, past research found that female non-bereaved crime victims more often than male non-bereaved crime victims have the intention to deliver a VIS (Lens et al., 2010). Our findings need to be replicated with a larger, more diverse bereaved sample, to draw firm conclusions regarding how background and loss-related characteristics are related to VPS delivery.

With respect to the third aim, bereaved people who delivered a VPS reported higher psychopathology levels before and after the court hearing, than those who did not. The findings suggest that bereaved people who deliver a VPS experience more psychological complaints than those who do not. This mirrors findings from a prior study examining mostly non-bereaved crime victims. To clarify, crime victims who delivered a VIS reported higher PTSD levels before and after the court hearing than people who did not (Lens et al., 2010). The present study is the first to suggest that these prior findings, relating psychopathology to VIS delivery, may generalize to bereaved people. Furthermore, no significant within-group differences (delivered vs. did not) in psychopathology levels were found when comparing pre- and post-VPS levels. Thus, VPS delivery does not seem to result in a significant increase or decrease of psychological complaints. These findings are not completely in line with past research (Lens et al., 2010), which observed a significant decrease in PTSD levels in both groups (delivered vs. did not). This may be partly explained by differences in sample composition. To elaborate, the present study included only homicidally bereaved people, while Lens et al. (2010) included (mostly) non-bereaved crime victims of various offence types (e.g. stalking and sexual offences). An alternative explanation is time since victimization. In the present study, time since victimization was more than seven years. In the study by Lens et al., however, time since victimization ranged (on average) from half a year to one-and-a-half years. When comparing the findings, it seems worthwhile to consider that psychopathology levels may stabilize after an amount of time. To elaborate, prior latent trajectory research suggests that the majority of MH17 bereaved people experienced an initial decrease in PGD and PTSD levels within 31 months following bereavement, after which levels seem to stabilize (Lenferink et al., 2020). Depression levels, however, seem relatively stable post-loss (Lenferink et al., 2020). Nonetheless, limited research has examined these symptom trajectories in exclusively violently bereaved people (Kristensen et al., 2020; Lenferink et al., 2020), especially in the context of a criminal trial. Future research may benefit from examining whether participating in a criminal trial (e.g. delivering a VPS and requesting compensation) may alter trajectories. Overall, more research is needed to determine how VPS delivery is related to psychopathology levels in non-bereaved and bereaved crime victims.

A strength of the present study is that all participants were bereaved by the same crime. Thus, crime- and trial-related characteristics could not confound the results, such as offence type and time since victimization. Although the sample size was small, between-group (delivered vs. did not) comparisons could be made regarding psychopathology levels. To the best of our knowledge, this has never been done before in a sample of exclusively bereaved people. Moreover, the longitudinal design of our study allowed us to examine changes in psychopathology over time. Lastly, by using both qualitative and quantitative methods a more elaborate picture of bereaved people’s experiences with VPS delivery has been created.

Although the uniqueness of the sample is a strength, it limits the generalizability of the findings to all people bereaved by violent loss. For example, the MH17 disaster received a lot of media attention, and a monument was created to commemorate the victims. Therefore, findings likely do not generalize to people bereaved by other causes which received less public attention. Also, in our sample, probable PGD, PTSD, and depression rates varied between 7% and 21% for pre- and post-VPS. These percentages are notably lower than some of the percentages found in previous research in people confronted with unnatural or violent deaths of loved ones (Djelantik, Smid, et al., 2020; Milman et al., 2018; van Denderen et al., 2015). Differences in prevalence rates may be related to differences between our study and prior research in terms of time since loss, criteria used to define disturbed grief reactions, and measures used to assess these reactions, among others. Therefore, generalizability of our findings to other violently bereaved people may be limited. Generalizability of the findings to other MH17 bereaved people may also be limited, due to self-selected sampling. Furthermore, the present study compared means of psychopathology levels at pre- and post-VPS within each group (delivered vs. did not). However, changes could have occurred on an individual level. Zooming in on differences in response patterns might be relevant in this regard. For instance by examining if latent classes emerge on the basis of psychopathology levels, whether transitions take place in class membership from pre- to post-VPS, and what factors predict class transitions. One way to investigate this is by applying latent transition analysis (see e.g. Lenferink, Liddell, et al., 2022). Unfortunately, such an analysis could not be conducted in the present study due to the small sample size. Moreover, we cannot rule out that additional potentially traumatic events may have occurred in the period between the crime and the VPS, which may have exacerbated psychopathology levels. Additionally, self-report measures were used to measure psychopathology levels, which tend to result in overestimation of symptom levels (Kramer et al., 2023). Lastly, one researcher (LN) coded the motivations, while preferably two independent coders should analyse qualitative data to reduce the risk of bias.

5. Conclusions

On the basis that future studies will replicate the findings, the following implications are made. First, a reason for expanding the VIS was to satisfy the need of crime victims to express themselves more (Kragting et al., 2022). The expansion seems to (partially) fulfil this need for homicidally bereaved people, as evidenced by the larger variety of motivations to deliver a VPS observed in the present study than observed in previous research (Lens et al., 2010) conducted prior to the expansion. Second, the findings do not support the belief that VPS delivery results in either emotional restoration or secondary victimization, if both refer to changes in psychopathology levels. Future studies may benefit from examining other ways in which VPS delivery might be beneficial or detrimental to specific bereaved people. For example, in terms of experiencing agency and communion (Kragting et al., 2022), anger (Buiter et al., 2022; Winkel, 2020), recognition (Wemmers, 2008), or well-being. Third, professionals may provide emotional support to bereaved people who want to deliver a VPS and manage their expectations if they want to deliver a VPS for the purpose of symptom reduction. To conclude, bereaved people state a multitude of reasons to deliver, or not deliver, an oral VPS, and while those who delivered a written or an oral VPS report higher psychopathology levels than those who did not, using VPS delivery to decrease psychopathology levels seems ineffective.

Acknowledgements

Our sincerest gratitude to all people who took part in the study, as well as the MH17 Disaster Foundation, family liaison officers of the Dutch police, and Victim Support the Netherlands for the effort they put in supporting us with recruiting participants. Additionally, we would like to thank research assistant Marjel Buiter for her support with data collection.

Disclosure statement

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

Data availability statement

The dataset associated with this study can be found on the Open Science Framework repository via this link (https://osf.io/xej78/).

Additional information

Funding

This research was funded by Fund Victim Support (in Dutch: Fonds Slachtofferhulp), the Dutch Ministry of Justice and Safety [20002853] and Stichting Achmea Slachtoffer en Samenleving (SASS) [201230]. The funding agencies were not involved in the execution of the study.

Notes

1 The main difference between the two systems is that in common law countries, case law – in the form of published judicial opinions – is of primary importance, whereas in civil law systems, codified statutes predominate.

2 The participants who stated ‘no added value to the participant’ as a motivation to not deliver an oral VPS had not delivered a written VPS.

Appendix A

Table 6. Examples of motivations to deliver an oral victim personal statement.

Sub-theme	Example
Describing the impact of the incident	I wanted the world to know what the perpetrators did to us
Addressing the behaviour of defendants/third parties	[To] address that the attitude of the Russian Federation is disrespectful
Directly addressing the court	So that I could talk about certain matters with the court
Expressing gratitude towards the court	Look in the eyes of the judges and thank them myself, for putting in so much effort to seek justice
Honouring the deceased	The last thing we could do for our daughter
Spokesperson of the deceased	I wanted my voice to be heard for my husband
Obligation to the deceased	I feel that I owe it to my parents
Remembering the deceased	So they will get an image of who she was
Influencing the punishment	[To ensure] the offenders will receive their punishment
Seeking justice	So people know I didn’t stop fighting for [name] and justice
Contributing to the criminal trial	By doing this, I want to make a contribution to the process
Recognition of the participant’s story	[We] had to listen for years, now they listen to us
Recognition of the harm	Increase understanding for the consequences
Facilitating the mental processing of the incident	Forcing myself to recall painful memories, by [delivering the victim personal statement]
Acceptance	Letting go and saying goodbye

Appendix B

Table 7. Examples of motivations to not deliver an oral victim personal statement.

Sub-theme	Example
No added value to the participant	I did not get the impression it would make me feel any better
Loss/crime remains unchanged	It does not change a thing
No added value to the case	Mine would not have added anything
No added value in general	I didn’t see the point, or any added value
No added value to the punishment	[No added value] when it concerns increasing the chance of the defendants being convicted
Self-protection	I thought I would not be able to handle the emotional circumstances
Protection of others	[Because I was afraid of the emotional consequences] for people close to me
Another representative spoke	My father used his right to speak
Personal view shared by someone else	We created the statement together
Focus on the present/future	I put a focus on the future
Downward comparison	A lot of people have gone through many things at my age
Not wanting revenge	[In my family] less of a focus is put on feelings of revenge
Distancing oneself from the criminal trial	I wanted to distance myself in that area
Media attention/protecting one’s privacy	[I] did not want to be approached by the media
Difficulty expressing oneself	I have no way to describe the emptiness I feel when I think about my loss
Public speaking	Public speaking is not my thing
Not knowing what to address	[I did not have] an overview [to create a victim personal statement]
Restrictions related to the country of residence	I live in a foreign country
Restrictions related to personal/professional circumstances	It is not easy to take leave
Physical complaints	I have heart problems
Unwilling to share emotions/consequences with the defendants	The perpetrators do not have the right to look that deep into my life and my feelings
Absence of defendants	The suspects were not there to listen to it
Negative perspective on the victim personal statement	The victim personal statement was commonly used to express self-pity

Appendix C

Table 8. Background and loss-related correlates of the decision to deliver a VPS (N = 84).

Characteristics of the participant	Constant (SE)	B (SE)	95% CI for Odds Ratio
			Odds Ratio	Lower	Upper
Background
Biological sex (0 = male)	0.06 (0.34)	-0.60 (0.45)	0.55	0.23	1.32
Age (in years)	0.28 (1.00)	-0.01 (0.02)	0.99	0.96	1.02
Level of education (0 = other than university)	-0.41 (0.41)	0.17 (0.49)	1.18	0.46	3.06
Loss-related
Number of losses due to MH17 plane disaster (0 = single)	0.18 (0.35)	-0.79 (0.46)	0.46	0.19	1.11
(Closest) Relationship to the deceased (0 = other than partner/spouse or child)	-0.57 (0.30)	0.62 (0.45)	1.86	0.78	4.48

Notes: VPS = Victim Personal Statement; SE = Standard Error; CI = Confidence Interval.