Healing grief – an online self-help intervention programme for bereaved Chinese with prolonged grief: study protocol for a randomised controlled trial

ABSTRACT

Background: In China, mental health services do not currently meet the needs of bereaved people with symptoms of prolonged grief disorder (PGD). Internet-based grief interventions may help fill this gap, but such programmes have not yet been developed or evaluated in China. The proposed study aims to investigate the effectiveness, acceptability, and feasibility of an online self-help intervention programme named Healing Grief for bereaved Chinese with prolonged grief, and to explore the psychological mechanisms of potential improvements.

Methods: We designed a two-arm randomised controlled trial. At least 128 participants will be randomly assigned to either an Internet-based intervention group or a waitlist-control group. The Internet-based intervention will be developed based on the dual process model, integrating techniques of psychoeducation, behavioural activation, cognitive reappraisal, and meaning reconstruction, and will be delivered via expressive writing. The intervention comprises six modules, with two sessions in each module, and requires participants to complete two sessions per week and complete the intervention in 6 weeks. The primary outcomes include effectiveness, acceptability, and feasibility. The effectiveness will be assessed by measures of prolonged grief, posttraumatic stress, anxiety, and depressive symptoms. Acceptability and feasibility will be evaluated using survey and interview on user experience characteristics. Secondary outcomes include moderators and mediators, such as dual process coping, grief rumination, mindfulness, and continuing bond, to explore the psychological mechanisms of potential improvement. Assessments will take place at pre-intervention, post-intervention, and 3-month follow-up.

Conclusion: The proposed study will determine the effectiveness, acceptability, and feasibility of the newly developed online self-help intervention for bereaved Chinese with prolonged grief and clarify how the intervention helps with symptom improvements. Such an intervention may play an important role in easing the imbalance between the delivery and receipt of bereavement psychological services in China.

HIGHLIGHTS

• In China, mental health services are not widely available for bereaved people.

• The proposed study will be the first one to develop and evaluate an Internet-based self-help grief intervention for bereaved Chinese with prolonged grief.

• The proposed study will determine whether and how the intervention helps to improve the mental health of bereaved Chinese with prolonged grief.

Antecedentes: En China, los servicios de salud mental no satisfacen actualmente las demandas de las personas en duelo con síntomas de trastorno de duelo prolongado (PGD). Las intervenciones de duelo basadas en Internet pueden ayudar a llenar este vacío, pero dichos programas aún no se han desarrollado ni evaluado en China. El estudio propuesto tiene como objetivo investigar la efectividad, aceptabilidad y viabilidad de la intervención de duelo basada en Internet para chinos en luto con duelo prolongado, y explorar los mecanismos psicológicos de las posibles mejoras.

Métodos: Ensayo controlado aleatorizado de dos brazos. Al menos 128 participantes serán asignados aleatoriamente a un grupo de intervención basado en Internet o a un grupo de control en lista de espera. La intervención basada en Internet se desarrollará basándose en el modelo de proceso dual, integrando técnicas de psicoeducación, activación conductual, reevaluación cognitiva y reconstrucción de significado, y se realizará mediante escritura expresiva. La intervención consta de seis módulos, con dos sesiones en cada módulo, y requiere que los participantes completen dos sesiones por semana, en 6 semanas. El resultado primario incluye la efectividad, aceptabilidad y viabilidad. La efectividad del duelo prolongado, el estrés postraumático, la ansiedad y los síntomas depresivos. La aceptabilidad y viabilidad se evaluarán mediante encuestas y entrevistas sobre las características de la experiencia del usuario. Los resultados secundarios incluyen moderadores y mediadores, como el proceso dual de afrontamiento, la reflexión sobre el duelo, la atención plena y el vínculo continuo para explorar los mecanismos psicológicos de potenciales mejoras. Las evaluaciones se llevarán a cabo antes de la intervención, después de la intervención y a los 3 meses del seguimiento

Conclusión: El estudio propuesto determinará la efectividad, aceptabilidad y viabilidad de la intervención de autoayuda en línea para chinos en luto con duelo prolongado y aclarar cómo la intervención ayuda con la mejoría de los síntomas. Una intervención de este tipo puede desempeñar un papel importante para aliviar el desequilibrio entre la prestación y la recepción de servicios psicológicos para el duelo en China.

引言：在中国，心理健康服务目前还不能满足延长哀伤丧亲者的需求。网络化哀伤干预可能有助于填补这一需求，但此类项目尚未在中国进行开发或评估。本研究旨在探讨在线自助干预程序“哀伤疗愈”对中国延长哀伤丧亲者的有效性、可接受性和可行性，并探讨其潜在的改善机制。

方法：本研究采用双臂随机对照试验设计。至少 128 名参与者将被随机分配到网络化干预组或等待组。网络化干预将以双过程模型为基础，整合心理教育、行为激活、认知重评和意义重建等技术，并通过表达性书写进行递送。干预包括六个模块，每个模块包含两次任务，要求参与者每周完成两次任务，并在 6 周内完成干预。主要结果包含有效性、可接受性和可行性。有效性将由延长哀伤症状、创伤后应激症状、焦虑症状和抑郁症状相关量表进行评估。可接受性和可行性测量则包括用户体验问卷调查和访谈。次要结果为评估双过程应对、哀伤反刍、正念和持续性联结等调节因素和中介因素，探讨潜在改善的机制。评估将在干预前、干预后和 3 个月后的随访中进行。

结论：研究将确定在线自助干预对中国延长哀伤丧亲者的有效性、可接受性和可行性，并阐明干预是否以及如何帮助症状改善。这项干预可能对缓解中国丧亲心理服务的供给和需求失衡起到重要作用。

KEYWORDS:

• Bereavement
• prolonged grief
• Internet-based intervention
• randomised controlled trial
• study protocol
• Chinese

PALABRAS CLAVE:

• Duelo
• duelo prolongado
• Intervención basada en Internet
• ensayo controlado aleatorizado
• Protocolo de estudio
• Chino

关键词:

• 丧亲
• 延长哀伤
• 网络化干预
• 随机对照试验
• 研究方案
• 中国人

1. Introduction

The death of a loved one represents an important turning point for bereaved people. According to the latest population data released by the National Bureau of Statistics, it is estimated that 11.1 million people died in China in 2023 (National Bureau of Statistics, 2024). On average, at least nine relatives are affected by the death of one person (Verdery et al., 2020). Thus, the number of bereaved people reached nearly 100 million in China in 2023. Bereavement carries an increased risk of mortality, impaired physical functioning, and deteriorating mental health (Stroebe et al., 2007). Although most people can recover from bereavement within six months or twelve months, some suffer from pervasive and persistent grief and impaired daily functioning for an unusually long time, which has been defined as prolonged grief disorder (PGD) in the International Classification of Diseases (11th edition) (ICD-11) (World Health Organization, 2018) and the text revision of the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-5-TR) (American Psychiatric Association, 2022). Meta-analysis revealed that 9.8% of bereaved people would develop prolonged grief disorder (Lundorff et al., 2017). In China, the prevalence of prolonged grief is 1.8–13.9% among bereaved adults (He et al., 2014; Li & Prigerson, 2016), and up to 35.5% in parents bereaved by losing their only child (Zhou et al., 2020) and to 37.8% among people bereaved due to COVID-19 (Tang & Xiang, 2021). Posttraumatic stress disorder and/or symptoms could be observed under certain circumstances, such as people who experienced unexpected death, with a prevalence of 5.2% (Atwoli et al., 2017), and whose loved one died from COVID-19, with a prevalence of 22% (Tang et al., 2021). Depression and anxiety could also occur after bereavement (Komischke-Konnerup et al., 2021; Tang et al., 2021). Despite the discouraging fact that bereavement could bring mental health concerns, seeking help from grief therapies and professional support seems to be an effective way of coping as it buffers the impact of grief on quality of life (Salisbury et al., 2022).

In China, most mental health services are provided by psychiatrists, psychotherapists, psychological counsellors, and registered clinical psychologists. There were 64,000 psychiatrists up to 2021 (National Health Commission of the People’s Republic of China, 2022a), 4819 psychotherapists, and 40,920 psychological counsellors in medical institutes up to 2020 (National Health Commission of the People’s Republic of China, 2022b), and nearly 4000 registered clinical psychologists up to 2020 (Chinese Psychological Society, 2021). With a population of over 1.40967 billion (National Bureau of Statistics, 2024), there are only about 8 mental health professionals per 100,000 population in China. For reference, about half of the countries have 13 or more mental health workers per 100,000 population (World Health Organization, 2022). Mental health professionals are dramatically insufficient in China (Lu et al., 2021), and very few of them have received systematic training in grief therapy and grief counselling (Xu et al., 2020), resulting in a lack of competency in providing grief intervention.

In addition to the dramatically insufficient nationwide mental health services and grief counsellors, bereaved people in China face similar obstacles during help-seeking as people with mental health concerns worldwide do. Even when mental health professionals are available, albeit few, bereaved Chinese people are prevented from accessing face-to-face mental health services by subjective perceptions and objective restrictions, including low mental health literacy, stigma, and a lack of individual financial resources (Wang et al., 2012). Moreover, many bereaved people believe that ‘time will heal grief’ (Tang et al., 2020) and thus are less likely to seek help from professionals than from families and friends (Tang et al., 2022). For the small group of bereaved people who sought professional help, they perceived professional support as less useful than that provided by family, friends, and people with similar bereavement experiences (Tang et al., 2022), perhaps because most professionals were incompetent at providing grief intervention.

Given the fact that existing mental health services cannot adequately help bereaved people in China, Internet-based interventions could potentially be a promising alternative option, as such interventions have the advantages of autonomy, flexibility, low cost of time and money, privacy protection, and anonymousness that help to reduce stigma and promote openness and frankness (Andrews & Williams, 2014). Therefore, bereaved Chinese people who are not willing to and/or not able to seek face-to-face professional help may benefit from Internet-based interventions. With the Internet penetration rate reaching 75.6% and 99.8% of the 1,067 million Internet users using mobile phones to access the Internet by the end of 2022 (China Internet Network Information Center, 2023), working and living online is becoming a lifestyle of Chinese people, which also opens up the possibility of accepting Internet-based interventions. As Internet-based interventions have been used in the treatment of depression (Ren et al., 2016; Ying et al., 2023), anxiety (Liu et al., 2020), and posttraumatic stress disorder (Wang et al., 2013; Zhao et al., 2023) in China, developing an Internet-based intervention for prolonged grief disorder is worth trying.

Internet-based grief intervention is still at a relatively early stage worldwide. Early interventions were mainly based on cognitive behavioural therapy (CBT) (Kersting et al., 2011; Kersting et al., 2013; Lange et al., 2000; Litz et al., 2014; van der Houwen et al., 2010; Wagner et al., 2006), including self-confrontation/exposure, cognitive restructuring/cognitive reappraisal, and social sharing as elements. Recently, some interventions were innovatively designed to combine CBT with grief theory such as the dual process model of coping with bereavement (DPM) (Brodbeck et al., 2019) and the Worden’s task model of grieving (Dominguez-Rodriguez et al., 2023). Internet-based grief interventions have been also expanded to art therapy based on solution-focused brief therapy and the meaning-making approach to bereavement (Park & Cha, 2023) and yoga intervention combining physical exercise and mindfulness cultivation (Huberty et al., 2020; Sullivan et al., 2022).

For intervention forms, both therapist-assisted and self-help intervention processes have been developed in which guidance is sent or shown to bereaved adults, who are asked to complete corresponding tasks. Therapist-assisted interventions involve therapists giving bereaved adults feedback, and/or research assistants reminding them to finish the tasks. Such interventions generally involve 10 sessions, conducted over approximately 5–8 weeks, 1–3 times per week, and 20–45 min each session. However, some interventions last longer, such as the Healthy Experience After Loss (HEAL) programme, which lasted for 12–18 weeks and involved 18 sessions (Litz et al., 2014). Self-help interventions do not include assistance from the therapists, but sometimes research assistants remind participants who have not finished their tasks to proceed with intervention (Reitsma et al., 2023; van der Houwen et al., 2010).

Internet-based grief interventions have been introduced to various populations, such as widow(er)s (Knowles et al., 2017), bereaved siblings (Wagner et al., 2022), people bereaved by suicide (Treml et al., 2021), parents who lost a child to cancer (Sveen et al., 2021), perinatal women (Huberty et al., 2020), and older adults (aged 55 and older) (Godzik et al., 2021). Meta-analyses of RCT demonstrated that Internet-based grief interventions are effective, with moderate to large effect sizes (Wagner et al., 2020; Zuelke et al., 2021). While studies included in the meta-analyses mostly adopted therapist-assisted interventions, Internet-based self-help grief interventions were also effective enough to improve mental health (Dominguez-Rodriguez et al., 2023; Huberty et al., 2020; Reitsma et al., 2023; Schrauwen, 2021; van der Houwen et al., 2010).

Although grief interventions were scarce either online or face-to-face in China, previous studies found that DPM was suitable for Chinese people. DPM was proposed by Stroebe and Schut (1999) and includes three components: loss-oriented coping (LO), restoration-oriented coping (RO), and oscillation. LO refers to strategies coping with stressors associated with the loss experience (e.g. yearning for the deceased), RO refers to strategies coping with stressors associated with the changes in life after bereavement (e.g. adjusting to new roles), and oscillation refers to the process that alternately uses the LO and RO coping to deal with both stressors. The theoretical framework of DPM has been demonstrated compatible with the Chinese cultural context and attitude toward things by analyzing the experiences of mothers bereaved due to the 2008 Wenchuan earthquake in China (Chen et al., 2019). During the grieving process, LO coping, respite, and RO coping interdependently oscillated in the daily lives of these bereaved mothers and worked together to facilitate their adaptation to the loss (Chen et al., 2019). This is in line with the Confucian philosophy Zhongyong (i.e. the Doctrine of the Mean; Zhu, 1983), representing the attitude and practice that a bereaved person should be neither too immersed in intense grief nor too avoidant of reminiscing the deceased in the context of bereavement.

When used in intervention protocols, DPM-based face-to-face group intervention was more effective than the control group that only focused on loss-oriented coping in improving complicated grief, depression, and social support among older adults who have been widowed for more than 2 years in Hong Kong (Chow et al., 2019). Developed under the DPM framework, a remote intervention combining group therapy and individual therapy, delivered through webinars, improved grief symptoms among people bereaved due to COVID-19 in Wuhan, China (Yu et al., 2022). Since DPM has been verified to be suitable for Chinese people from the perspectives of theory and practice, we chose DPM as our theoretical ground when designing the Internet-based grief intervention protocol.

To summarise, Internet-based grief interventions could potentially provide a significant alternative option to face-to-face grief interventions, representing a cost-effective and highly feasible approach. However, to the best of our knowledge, Internet-based grief intervention has not been developed for Chinese people yet. Thus, this study mainly aims to investigate the effectiveness, acceptability, and feasibility of an online self-help intervention programme, developed based on DPM, for bereaved Chinese with prolonged grief. The specific objectives are as follows.

Objective 1: To evaluate the effectiveness of the newly developed intervention by assessing the improvement in severity of prolonged grief, posttraumatic stress, anxiety, and depressive symptoms.

Objective 2: To evaluate the acceptability and feasibility of the newly developed intervention by exploring the user experience characteristics.

Additionally, few existing research on Internet-based grief interventions has clarified how the interventions exert positive effects on grief. Previous studies have verified that continuing bonds (Yu et al., 2016), dual process coping (Tang & Chow, 2017), grief rumination, and trait mindfulness (Tang et al., 2019) were related to the mental health of Chinese bereaved people, suggesting that these variables may have effects on the process of bereavement adjustment. Thus, we use these variables to investigate the psychological mechanisms of how Internet-based intervention works for bereaved Chinese with prolonged grief. The specific objective is listed below.

Objective 3: To explore the psychological mechanisms of how the intervention exerts positive effects by assessing the moderating and mediating effects of continuing bond, dual process coping, grief rumination, and trait mindfulness.

In sum, the proposed study will develop an online self-help intervention programme, based on DPM, for bereaved Chinese people, providing precise empirical data and practical experience in Chinese cultural settings. The proposed study aims to evaluate the effectiveness, acceptability, and feasibility of the newly developed Internet-based intervention in bereaved Chinese people and to explore the psychological mechanisms underlying any observed intervention effects.

2. Methods

2.1. Trial design

The proposed study is a randomised controlled trial with two groups of equal sample size: the Internet-based intervention group (IG) and the waitlist-control group (WCG). Participants will be randomly assigned to either group using a list generated using www.random.org. The IG will receive a self-help online intervention developed in this study immediately after screening. The intervention will take place twice a week over 6 weeks in total. The WCG will wait for 6 weeks and then start the online intervention after the end of the IG treatment period. Clinical structural interviews and self-constructed questions of personal information will be administered to screen for eligibility (T-1), and self-constructed sociodemographic and loss-related questions and self-reported measures on psychopathological symptoms and psychological variables will be used at baseline (T0), post-intervention (T1.0 and T1.1 for IG and WCG respectively), and follow-up (T2.0 and T2.1 for IG and WCG respectively; 3 months after intervention completion). Additionally, there will be a user experience survey with self-constructed questions and a semi-structured interview on experience during the intervention post-intervention. The study will be conducted following the Consolidated Standards of Reporting Trials (CONSORT) statement for a randomised trial (Schulz et al., 2010) and the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines (Chan et al., 2013).

2.2. Participants

2.2.1. Recruitment

Participants will be recruited via both online advertisement and offline flyers. Online advertisements with a link to the study website will be disseminated on social media platforms (e.g. Weibo, Zhihu, Baidu, Douban, WeChat), self-help forums, specific websites for bereaved people, online communities, and blogs in China. Offline flyers containing the address of the study website will be sent to a number of stakeholders, including schools, universities, hospitals, psychological counselling centres, social work organisations, professional associations of psychology, social work, and medicine, and professionals through the researchers’ personal networks.

The study website will provide thorough information about the study and the intervention. Interested individuals will be able to apply by submitting a screening questionnaire that determines whether they are eligible to participate in the study. Contact information of the research team will be provided on the study website to enable prospective participants to ask questions.

2.2.2. Eligibility criteria

All participants who meet the following criteria will be included in the study: (1) aged 18 years or older, (2) experienced the loss of a close person more than 6 months ago, (3) have access to the Internet through mobile phones, (4) have the ability to read and write in Chinese, and (5) rated 4 or 5 on at least one core symptom and at least one associated symptom of prolonged grief in the past month measured by the International ICD-11 Prolonged Grief Disorder Scale (IPGDS) (Killikelly et al., 2020). Participants with any of the following criteria will be excluded from the study: (1) psychiatric symptoms in the past month, (2) substance abuse in the past month, (3) suicidality in the past week, or (4) current psychotherapy.

2.2.3. Sample size

Two previous meta-analyses reported that Internet-based interventions had at least a moderate effect size for grief symptoms (Wagner et al., 2020; Zuelke et al., 2021). Thus, assuming a between-subject effect size of d = 0.50, power of 0.80, and an alpha level of 0.05 (two-sided), the target sample size should be at least 128 participants (64 for each group). In addition, given that the high average dropout rate (about 27%) reported in the study of Wagner et al. (2020), a minimum sample size of 162 participants (81 per group) will be needed.

2.3. Procedure

The procedure from screening to follow-up assessments is shown in Figure 1.

Figure 1. Study flow from screening to follow-up. Monitoring refers to noticing whether participants have questions and assessing if participants need urgent help. Note: IG represents Internet-based intervention group and WCG represents waitlist-control group.

T-1, screening: After participants apply for the study via the screening questionnaire, research assistants will browse their responses and contact them. All research assistants are trained by the first author, a registered clinical psychologist of the Chinese Psychological Society, and thus certified to conduct clinical structural interviews. For participants who meet the inclusion criteria, research assistants will conduct a screening interview by telephone to estimate whether they are eligible for the study. For participants who are not eligible for the study, research assistants will make contact by email to explain the reasons for ineligibility and provide relevant psychological and grief resources. In particular, participants who indicate suicidal ideation in the screening questionnaire are not eligible for the study. They will be contacted by telephone to ensure their safety and provide immediate help if necessary.

Informed consent: After screening, research assistants will introduce the procedures of the study and obtain informed consent from eligible participants. An email with a user manual for the intervention and a consent form will be sent to eligible participants. Participants will be asked to read carefully and send back a written or electronically signed consent form within a day. They will be informed that there are two groups in the study, one group will receive the intervention once completing the baseline questionnaire, and the other group will wait for six weeks and then receive intervention. If they have any questions when reading the manual and consent form, they can ask questions to research assistants at any time. At last, after research assistants receive and check the signed consent form, they will send back a copy to participants.

T0, baseline (week 0): After informed consent, participants will receive a link to the baseline questionnaire. After completing the baseline questionnaire, participants will be randomised into one of the two groups (IG and WCG) as described below. All participants will be sent an email to inform which group they will be in and what they need to do next.

Treatment (week 1-week 6): IG participants will receive the self-help online intervention, which will take place twice a week over 6 weeks in total. Details of the intervention are described below. WCG participants are generally not allowed to receive any intervention during the period. However, if they have urgent needs for psychological intervention, information and services of other interventions will be provided.

T1, post-intervention (week 7/week 13): After completion of intervention, IG participants will receive a link to the post-intervention questionnaire and then take part in a semi-structured telephone interview (T1.0, week 7). WCG participants will receive a link to the second baseline questionnaire in the seventh week since the baseline assessment, then start the above intervention. The same post-intervention questionnaire used with IG participants will be administered to WCG participants after the intervention (T1.1, week 13).

T2, follow-up (week 19/week 25): Three months after completion of the intervention, a link to the follow-up questionnaire will be sent to IG (T2.0, week 19) and WCG (T2.1, week 25) participants.

2.4. Randomisation

After the baseline assessment, participants will be randomised into one of the two groups: IG or WCG. A permutated block randomisation with a block size of four and equal probabilities to be sampled into either group will be carried out. A research assistant who is not involved in participant recruitment, assessments, or intervention will execute the randomisation using a list generated by www.random.org so that the randomisation and allocation of participants will not be influenced.

2.5. Blinding

Because of the difference in the beginning time of the interventions of the two groups (week 0 versus week 7), it will not be possible to blind the participants and research assistants who are in contact with the participants during the study. However, because all assessments after randomisation will be carried out anonymously and automated via online questionnaire, this may minimise biases in the post-intervention and follow-up assessments. Moreover, the data analysis will be blinded. The research assistants who are in contact with the participants will not perform the data analysis. Thus, an independent research assistant will conceal participants’ identifications (IDs) and their group assignment in the data set until all results are finalised.

2.6. Intervention

Healing Grief is a 12-session online self-help intervention. This intervention is designed based on the DPM proposed by Dutch psychologists Stroebe and Schut (1999). The interventions based on DPM have been validated to be suitable for Chinese people (Chen et al., 2019; Chow et al., 2019; Yu et al., 2022). The DCM emphasises the oscillation between loss-oriented coping such as yearning for the deceased and restoration-oriented coping such as attending to life change. Thus, during the intervention, loss-oriented and restoration-oriented coping strategies will be alternated. Specifically, the intervention will include psychoeducation, behavioural activation, expressive writing, meaning reconstruction, and cognitive reappraisal as strategies structured into six modules: (1) understanding grief, (2) saying goodbye, (3) living in the moment, (4) continuing the bonds, (5) facing it together, and (6) searching for meaning. Each module will consist of two sessions. See Table 1 for an overview of the content of the intervention.

Table 1. Overview of Healing Grief.

Module	Session	Loss-oriented (LO)	Restoration-oriented (RO)	Approach
Understanding grief	1	L1: Monitoring grief		Psychoeducation Behavioral activation
	2	½ L2: Labelling emotions	½ R1: Self-care and mindfulness	Psychoeducation Behavioural activation
Saying goodbye	3	L3: Remembering the deceased		Expressive writing
	4	L4: Building a memory home		Expressive writing
Living in the moment	5		R2: Daily memorial rituals	Meaning reconstruction Behavioural activation
	6		R3: Dealing with survivor’s guilt	Psychoeducation Expressive writing
Continuing the bonds	7	L5: Writing to the deceased		Expressive writing Cognitive reappraisal
	8	L6: Building the continuing bond		Psychoeducation Expressive writing Cognitive reappraisal Meaning reconstruction
Facing it together	9		R4: Social support	Psychoeducation Behavioural activation
	10		R5: Assertiveness training	Psychoeducation Behavioural activation
Searching for meaning	11	½ L7: Writing to an imaginary friend	½ R6: Meaning reconstruction	Psychoeducation Expressive writing Cognitive reappraisal Meaning reconstruction
	12		R7: Benefit finding	Expressive writing Cognitive reappraisal Meaning reconstruction

Note: Each session begins with asking about the practice of the content of the last session and monitoring grief. L represents the loss-oriented component. R represents the restoration-oriented component.

IG participants will receive access to the online intervention after giving informed consent. They will first create an account using their IDs, then complete two 20–30-minute sessions per week with a recommended time interval of two or three days between each session. The sessions will be presented in order, meaning that participants will need to complete one session before moving on to the next. The whole process is intended to last for 6 weeks and is entirely self-help without the guidance of therapists. However, trained research assistants will be assigned to track participants’ progress weekly. A research assistant will send a reminder email if a participant does not complete the two sessions within a week and will provide encouragement when a participant achieves the weekly goal set by the intervention. In addition, if participants complete more than two sessions within a week or complete two sessions within a shorter time interval than the recommended one, they will be reminded to complete the remaining sessions as recommended. The reminders and encouragement scripts will be standardised. Participants will also be able to initiate contact with the research assistant at any time as needed. WCG participants will receive the same intervention 6 weeks after baseline assessment.

The intervention will be text-based, mainly in the form of questions and answers. Participants will primarily answer the questions in dialogue boxes or by following written instructions to complete tasks. In addition, brief examples, audio recordings, reading materials, and links to keep diaries will be included in the sessions to enrich the content and enhance understanding (see Figure 2). At the beginning of each session, participants will be asked to indicate whether and how they have practiced the content of the last session, to monitor their grief by rating their highest and lowest levels of grief during the day on a rating scale (score 0–10), and to input where they are, what they are doing, and who they are with at the time when they experience the highest and lowest levels of grief. All of these are developed via a mini programme nested in WeChat, one of the most commonly used messaging and calling applications in China, connecting a billion people (https://www.wechat.com/en/). The mini programme has almost the same functions as applications, and because it is nested in WeChat, it will be easily accessible without installing anything if there is a WeChat on smartphones and tablets.

Figure 2. Sample screenshots of Healing Grief include (a) the home page, (b) the interface for monitoring grief, (c) the interface for a session, and (d) the interface for keeping diaries.

2.7. Measures

Screening variables, sociodemographic and loss-related variables, primary outcomes (effectiveness measures and user experience characteristics), secondary outcomes (moderators and mediators) are summarised in Table 2. All the questionaries and interview will be provided in Chinese.

Table 2. Measures.

Construct	Instrument (abbreviation)	Item No.	Rating Likert scale wording (scores)	Time frame	Reliability	T-1	T0	T1	T2
Screening variables
Psychiatric symptoms	DIPS	3	‘YES’ or ‘NO’	Last month	/	X
Substance abuse	CAGE Questionnaire	4	‘YES’ or ‘NO’	Last month	/	X
Suicidal ideation	BSI-CV	19	Anchor differs (0–2)	Last week & Lifetime	/	X
Sociodemographic and loss-related variables					/
Sociodemographic variables	/	6	/	Present			X
Loss-related variables	/	9	/	Present			X
Primary outcome
Effectiveness measure
PGD	IPGDS	14	Almost never–always (1–5)	Last month	α = 0.97		X	X	X
PTSD	PCL-5	20	Not at all–extremely (0–4)	Last month	α = 0.94		X	X	X
Anxiety	HADS-anxiety subscale	7	Anchor differs (0–4)	Last month	α = 0.77		X	X	X
Depression	HADS-depression subscale	7	Anchor differs (0–4)	Last month	α = 0.76		X	X	X
User experience characteristics
Usage of the intervention	/	3	/	Last 6 weeks	/			X
Perceptions of the intervention form	PCTQ-P (adapted version)	16	Strongly disagree-strongly agree (1–7)	Last 6 weeks	α = 0.92			X
Acceptability of the intervention content and intervention settings	/	10	Anchor differs (1–5)	Last 6 weeks	/			X
Future use intention	/	3	Not at all-extremely (1–5)
User experience interview	/	11	/	Last 6 weeks	/			X
Secondary outcomes
Posttraumatic growth	PTGI	21	Did not experience this change–experienced this change to a very great degree (0–5)	Since loss	α = 0.93		X	X	X
Dual coping process	DCI; Time spent; Oscillation	10	DCI: Not at all–very much (1–5) Time spent: no time–all of the time (1–100) Oscillation: extremely difficult to do so–extremely easy to do so (1–5)	Last week	α = 0.81 (LO) α = 0.75 (RO)		X	X	X
Grief rumination	UGRS	15	Never–very often (1–5)	Last month	α = 0.90		X	X	X
Mindfulness	MAAS	15	Almost always–almost never (1–6)	Present	α = 0.92		X	X	X
Continuing bond	CBS	19	Almost never–always (1–4)	Last month	α = 0.92		X	X	X

Note: PGD = Prolonged Grief Disorder; PTSD = Posttraumatic Stress Disorder; DIPS = German Diagnostic Interview for Psychiatric Symptoms (Margraf, Cwik, Pflug, et al., 2017; Margraf, Cwik, Suppiger, et al., 2017); CAGE Questionnaire = Cutting down, Annoyance by criticism, Guilty feelings, Eye-openers questionnaire (Ewing, 1984); BSI-CV = Beck Scale for Suicide Ideation-Chinese Version (Li et al., 2010); IPGDS = International ICD-11 Prolonged Grief Disorder Scale (Killikelly et al., 2020); PCL-5 = Posttraumatic Stress Disorder Checklist for DSM-5 (Cheng et al., 2020); HADS = Hospital Anxiety and Depression Scale (Ye & Xu, 1993); PTGI = Posttraumatic Growth Inventory Scale (Wang et al., 2011); DCI = Dual Coping Inventory (Tang & Chow, 2017); UGRS = Utrecht Grief Rumination Scale (Tang et al., 2019); MAAS = Mindful Attention Awareness Scale (Deng et al., 2012); CBS = Continuing Bond Scale (Yu et al., 2016); PCTQ-P = Perceptions of Computerized Therapy Questionnaire-Patient (Carper et al., 2014); LO = loss-oriented coping; RO = restoration-oriented coping; Time spent = Time spent in LO and RO; Oscillation = the oscillation between LO and RO; α = Cronbach's alpha.

2.7.1. Screening variables

The research assistants will screen participants by asking a few questions about personal information (e.g. age and time since loss in months) to exclude individuals with ineligible sociodemographic and loss-related characteristics and via a telephone interview to exclude the presence of current psychiatric symptoms, substance abuse, acute suicidality, and current psychotherapy.

German Diagnostic Interview for Psychiatric Symptoms (DIPS) (Margraf, Cwik, Pflug, et al., 2017; Margraf, Cwik, Suppiger, et al., 2017). Psychiatric symptoms will be assessed by three items taken from the German Diagnostic Interview for Psychiatric Symptoms (DIPS). In the current study, participants will be asked if they have had strange or unusual experiences in the past month, including (i) Hearing or seeing things that other people didn’t notice. (ii) Hearing voices or conversations when no one was around. (iii) Feeling that something strange was going on around them, that people were doing things to test, antagonise, or hurt them so that they felt like they had to be on guard all the time. The three items can be answered with ‘YES’ or ‘NO’ responses. Only participants who answer ‘NO’ for all the three items will be included. The Chinese-translated versions of these three items have been used for screening in the study of Internet-based intervention (Wang et al., 2013).

CAGE Questionnaire (Ewing, 1984) (C-Cutting down, A-Annoyance by criticism, G-Guilty feelings, E-Eye-openers) (L’akoa et al., 2013) The CAGE questionnaire is designed to be a screening instrument that can be answered with ‘YES’ or ‘NO’ responses instead of a diagnostic instrument. The CAGE questionnaire has been translated into Chinese (Zhang, 2013). It includes four questions: (i) Have you ever felt the need to cut down your drinking? (ii) Have you ever felt annoyed by criticism of your drinking? (iii) Have you ever had guilty feelings about drinking? (iv) Have you ever taken a morning eye opener (O’Brien, 2008)? In the present study, we will only exclude participants with substance abuse issues in the past month. Thus, the time frame of assessment will be revised to the past month. Participants who answer ‘NO’ to at least three questions will be included.

Beck Scale for Suicide Ideation-Chinese Version (BSI-CV) (Li et al., 2010) The BSI-CV was developed based on the Beck Scale for Suicide Ideation (BSI) (Beck, 1991), the Scale for Suicide Ideation-Current (SSI-C) (Beck et al., 1979), and the Scale for Suicide Ideation-Worst (SSI-W) (Beck et al., 1999). The Scale is made up of 19 items and can be completed in 5–10 min. Clinicians can examine suicidal ideation in patients on a 3-point rating scale ranging from 0 to 2. Higher scores indicate higher levels of suicidal ideation and a higher risk of suicide. Items 1–5 of the BSI-CV will be used to examine the presence of the risk of suicide. Participants will only be included if they score 0 (i.e. ‘I do not want to die’, ‘I want to live rather than die’, and ‘I have no intention to attempt suicide’) on items 2–4 based on the feelings within a week. For example, participants will be excluded if they answer that the extent to which they want to die (item 2) is moderate or more within a week and may be included if they answer that they have not thought about dying within a week. If participants score 1 or 2 in at least one of the first five items, research assistants will ask them the following 14 questions to understand their current situation and will provide them with crisis intervention hotlines if necessary.

Participants who are receiving current psychotherapy will be excluded based on their answers for one question (i.e. Are you receiving stable counseling or systematic psychotherapy?). Only participants who are not receiving stable counselling or systematic psychotherapy will be included.

2.7.2. Sociodemographic and loss-related variables

Sociodemographic and loss-related information will be assessed with self-developed items mainly in the baseline questionnaire. Sociodemographic information includes sex, age, education level, occupation, religious belief, and marital status. Loss-related information includes relationship to the deceased, sex of the deceased, age of the deceased, time since the loss in months, and cause of death. In addition, participants rate the unexpectedness of death, traumatic level of the loss, closeness with the deceased, and conflict with the deceased on a 5-point Likert scale ranging from 1 (not at all) to 5 (very much). All of these data will be collected to describe the characteristics of the sample.

2.7.3. Primary outcome

2.7.3.1. Effectiveness measure

International ICD-11 Prolonged Grief Disorder Scale (IPGDS) The severity of prolonged grief symptoms is measured using the International ICD-11 Prolonged Grief Disorder Scale (IPGDS) (Killikelly et al., 2020). The IPGDS was developed for assessing PGD symptoms. It contains 13 items about yearning, preoccupation, emotional distress, and functioning impairment of a bereaved individual after the death of a close person (the standard scale), and one cultural screening item. The cultural screening item allows participants to rate the degree to which their grief would be considered worse (e.g. more intense, severe and/or of longer duration) compared with that of others in their community or culture. Participants indicate how often they experienced these symptoms in the past month on a 5-point rating scale ranging from 1 (almost never) to 5 (always). The total score of all items excluding the cultural screening item represents the levels of PGD symptoms; higher scores indicate higher levels of symptoms. The IPGDS has been validated in Chinese (α = 0.97) samples, yielding a reliability value of 0.93 on the standard scale (13 items) (Killikelly et al., 2020).

Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5) The PCL-5 is a 20-item self-report measure assessing the presence and severity of PTSD symptoms in the DSM-5 (Blevins et al., 2015). In this measure, participants rate how much they had been bothered by each item in the past month on a 5-point Likert scale ranging from 0 (not at all) to 4 (extremely). The total score of all items represents the severity of PTSD symptoms, with higher scores indicating more severe PTSD symptoms. The internal consistency of the PCL-5 scores was reported to be excellent among Chinese healthcare workers during the outbreak of coronavirus disease 2019, α = 0.91 (Cheng et al., 2020).

Hospital Anxiety and Depression Scale (HADS) The HADS is a questionnaire used to assess anxiety and depression (Zigmond & Snaith, 1983). It contains an anxiety subscale and a depression subscale, each consisting of seven items. In this questionnaire, participants indicate how often or to what extent they experienced various symptoms in the past month on a 4-point Likert scale ranging from 0 to 4 (anchor differs). The sum score on all 7 items per scale suggest more severe symptoms. The Chinese version of the HADS was reported to be reliable and valid (Ye & Xu, 1993).

2.7.3.2. User experience characteristics

Usage of the intervention will be measured by the login frequency, the number of minutes per session, and the number of completed sessions on the basis of objective user data from the backend of the mini-programme and self-reported data from the post-intervention questionnaire. Among them, the login frequency and the number of completed sessions can also reflect treatment adherence.

Perceptions of the intervention form will be assessed using a questionnaire with a self-translated version of 16 items adapted from the Perceptions of Computerized Therapy Questionnaire-Patient (PCTQ-P, Carper et al., 2014). The items will be retained and adapted according to the characteristics of the intervention programme, with seven items from the relative advantage (RA) dimension, five items from the complexity (CMPX) dimension, and four items from the compatibility (CMPT) dimension. RA assesses the advantages of the intervention via a WeChat mini-programme in comparison to other forms (e.g. Intervention via WeChat mini-programme makes it easier to get therapy); CMPX assesses the extent to which the intervention via the WeChat mini-programme is easy to use (e.g. Learning to use an intervention via the WeChat mini-programme would be easy for me); CMPT assesses the compatibility of the intervention via the WeChat mini-programme with individual beliefs and preferences (e.g. Intervention via WeChat mini-programme fits with my lifestyle). Responses will be rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The internal consistency reliability of the original scale was found to be excellent (α = 0.92) in a sample of undergraduates in America (Carper et al., 2014).

Acceptability of the intervention content and intervention settings will be evaluated with questions on the perceptions of intervention content and the achievement of treatment goals. Perceptions of the intervention content will be measured by questions on perceived quality, amount of information, degree of difficulty, overall helpfulness, and satisfaction. The questions can be answered on a 5-point scale ranging from 1 to 5, with higher scores indicating that the intervention content is considered to be high-quality, informative, easy to understand, helpful, and satisfying. The helpfulness of each session will be assessed on a 5-point scale with an additional option for participants that did not use the sessions (0 = I didn’t use the session). Moreover, participants will report which sessions they found especially impressive using multiple-choice questions so that the acceptability of each session can be further evaluated. Achievement of treatment goals will be measured using a self-developed questionnaire with 28 items, with seven subscales consisting of 3–5 items each, used to assess the goal achievement of the study procedure and each module. Participants will indicate the extent to which they agree with each statement on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Higher average scores are reflective of getting closer to the goals.

Future use intention includes three parts: intentions of recommending the intervention to others who have similar experiences, continuing to participate in the online self-help intervention via mini-programme, and receiving online self-help intervention via other platforms (e.g. website, application). The questions on future use intention are answered on a 5-point scale ranging from 1 to 5, with higher scores indicating a greater willingness to use in the future.

User experiences interview will be conducted via telephone to explore the experience of participants. The interview will be semi-structured, including 11 questions across the following aspects: (1) experience of forms (e.g. How do you feel about the experience of using the intervention provided by the mini-programme?), (2) experience of content (e.g. What benefits /discomfort [if any] did you experience from the intervention?), (3) reasons for persistence (e.g. Did you complete all of the modules of the intervention [why/why not]?), (4) effect on intentions (e.g. Would you recommend this mini-programme to someone who is in the same situation as you? [why/why not?]), (5) effect on cognition (e.g. How has your understanding of grief/psychological adjust spacing interventions changed [if at all]?) and (6) Improvement advice (e.g. What do you think that need to be improved of our programme?). In addition, adverse events will be recorded by an inventory for the assessment of the negative effects of psychotherapy (Ladwig et al., 2014) if participants indicate any difficulty or uncomfortable experiences.

2.7.4. Secondary outcomes

Posttraumatic Growth Inventory Scale (PTGI) The PTGI is a 21-item self-report measure designed for assessing positive outcomes reported by people who have experienced adverse life events including bereavement (Tedeschi & Calhoun, 1996). Participants report to what extent things have changed because of the loss on a 6-point Likert scale ranging from 0 (I did not experience this change as a result of my loss) to 5 (I experienced this change to a very great degree as a result of my loss). Total scores calculated by summing up all items in each subscale and higher scores indicating greater positive changes after the loss. more severe symptoms. The Chinese version of the PTGI demonstrated good psychometric properties (Wang et al., 2011), and the Cronbach's alpha was 0.93 in a bereaved Chinese sample (Chen & Tang, 2021).

Dual Coping Inventory (DCI) The DCI is a seven-item self-report inventory used to assess two different coping orientations in bereaved people (Meij et al., 2011). The two subscales, loss-oriented (LO) coping (e.g. I am occupied by the loss) and restoration-oriented (RO) coping (e.g. I direct my thoughts towards the future) contain three and four items, respectively (Tang & Chow, 2017). In this measure, participants report how applicable the descriptions of each item have been to them in the past week on a 5-point Likert scale ranging from 1 (not at all) to 5 (very much). The Chinese version of the DCI has been validated (Tang & Chow, 2017), and the reported Cronbach’s alpha values in a sample of bereaved Chinese people were 0.81 for the LO subscale and 0.75 for the RO subscale.

Time spent in LO and RO coping will also be assessed by two self-reported questions. One is ‘How much time did you spend in dealing with thoughts and emotions related to the loss over the past week’, and the other is ‘How much time did you spend in dealing with thoughts and emotions related to the present and future life over the past week’. Participants can rate on a 100-point scale ranging from 1 (no time) to 100 (all of the time). In addition, participants need to answer a question on the oscillation between LO and RO coping, namely ‘I was able to shift my attention between the loss-related thoughts and emotions and the restoration-related thoughts and emotions over the past week’, rating on a 5-point Likert scale that ranges from 1 (extremely difficult to do so) to 5 (extremely easy to do so).

Utrecht Grief Rumination Scale (UGRS) The UGRS is a 15-item questionnaire designed to assess grief rumination (Eisma et al., 2014). Participants report how often they have experienced certain types of ruminative thoughts in the past month on a five-point Likert scale ranging from 1 (never) to 5 (very often). The UGRS measures ruminative thoughts with three-item subscales across five topics: (a) reactions (e.g. How often in the past month did you try to analyze your feelings about this loss?), (b) injustice (e.g. How often in the past month did you wonder why this had to happen to you and not someone else?), (c) counterfactuals (e.g. How often in the past month did you analyze whether you could have prevented the death?), (d) meaning (e.g. How often in the past month did you analyze what the personal meaning of the loss is to you?), and (e) reactions of others (e.g. How often is the past month did you think about how you would like others to react to your loss?). Higher sum scores (overall scale and subscales) indicate higher levels of grief rumination and grief rumination subtypes. The Chinese version of the UGRS has been validated, and the Cronbach’s α was reported to be 0.90 for a sample of bereaved Chinese people (Tang et al., 2019).

Mindful Attention Awareness Scale (MAAS) The MAAS is a 15-item questionnaire used to assess mindfulness (Brown & Ryan, 2003; Deng et al., 2012). Participants rated how frequently they currently have a mindful state of mind described in 15 statements on a six-point scale rated from 1 (almost always) to 6 (almost never). Higher average scores reflect higher levels of dispositional mindfulness. The internal consistency of the MAAS scores in a bereaved Chinese sample was 0.92 (Tang et al., 2019).

Continuing Bond Scale (CBS) The CBS is a 19-item self-report questionnaire designed to measure different ways of maintaining a connection with the deceased person (Field et al., 2003; Field & Filanosky, 2009). Participants indicate how frequently they felt the deceased was a part of their life in the past month on a 4-point Likert scale from 1 (almost never) to 4 (always). The Hong Kong version of the CBS was adapted by Ho et al. (2013) and showed psychometric validity in a mainland Chinese sample. The internal consistencies of its two subscales in the mainland Chinese sample were adequate (externalised: α = 0.78, internalised: α = 0.93) (Yu et al., 2016).

2.8. Statistical analysis

Sociodemographic data and main outcomes will be reported using descriptive statistics. Chi-square tests, t-tests, and analysis of variance (ANOVA) will be performed to examine whether the characteristics of participants of IG are different from that of participants from WCG at baseline.

To test the treatment effect (i.e. a significantly greater decrease in PGD and other mental health outcomes from baseline to post-intervention and follow-up in the IG compared with the WCG), a 3 × 2 repeated-measures ANOVA will be conducted with time as the within-subjects factor (i.e. baseline, 7-week, 19-week) and group as the between-subjects factor (IG vs. WCG), including an interaction effect between time and group. Cohen’s d will be calculated to reflect effect sizes. To determine the psychological mechanisms underlying improvements in outcomes, potential mediators and moderators will be included in indirect effect analyses and moderation effect analyses. To identify the predictors of dropout, logistic regression analyses will be conducted, with the characteristics of participants at baseline as independent variables.

All analyses will be based on the per-protocol (PP) principle. Intent-to-treat (IT) analysis will be also used to make sure the IG and WCG are statistically comparable after randomisation, regardless of who drops out during the intervention. Participants who drop out of the study after randomisation will be asked to state their reasons for withdrawing from the study and continue the assessments. Missing values will be handled using complete case analysis for the PP analysis and the last observation carried forward approach for the ITT analysis. Interim analyses will be calculated to present the first results before the study is finished. All analyses will be conducted using SPSS, with an alpha level of 0.05.

2.9. Analysis of user experience interview

After the transcribed verbatim of the interview recording, NVivo Version 12 Plus qualitative data analysis software will be used for facilitating the data coding. A thematic analysis approach will be used to flexibly identify, analyze, and interpret the data by organising the codes into the core themes and subthemes.

3. Discussion

Although a large number of bereaved Chinese people suffer from mental health problems, few of them have access to mental health services, particularly face-to-face grief therapy and grief counselling, greatly due to the dramatic lack of professionals who are competent in working with bereaved people. Recently, Internet-based grief interventions have been found to provide helpful alternative options to face-to-face grief interventions in developed Western countries, exerting a positive effect on reducing psychological distress (Wagner et al., 2020) in a way that is cost-effective and easily accessible (Andrews & Williams, 2014). Therefore, the proposed study aims to determine the effectiveness, acceptability, and feasibility of an Internet-based grief intervention developed for bereaved Chinese adults through a randomised controlled trial, which is designed to help adults who suffer from prolonged grief disorder in China.

To the best of our knowledge, the proposed study will be the first one to develop and evaluate an Internet-based self-help grief intervention for bereaved Chinese people. One advantage of the proposed study is that it will be suitable for Chinese bereaved people. On one hand, the intervention is developed based on DPM, a validated theory that is compatible with the Confucian philosophy Zhongyong embedded in Chinese people’s daily life and helps improve the mental health of Chinese bereaved people (Chen et al., 2019; Chow et al., 2019; Yu et al., 2022). On the other hand, the contents of the intervention are designed to be close to Chinese habits of expression and reading, and the illustrations of psychoeducation and use of coping strategies reflect the experiences of bereaved Chinese people. Another advantage is that, in addition to describing the effectiveness, acceptability, and feasibility of the intervention, the proposed study will explore the underlying psychological mechanisms involved in the improvement process. Moreover, although a systematic review reported that interventions that are based on the dual process model may be more effective than other traditional grief therapy (Fiore, 2021), only one online treatment programme developed based on DPM has been evaluated in a randomised controlled trial (Brodbeck et al., 2019). The proposed study will further verify the effect of the dual process model on the Internet-based grief intervention by examining whether the intervention facilitates oscillation between LO and RO coping and whether the changes in the oscillation could predict the improvements in the mental health of the bereaved Chinese people. In conclusion, the proposed study will provide new and in-depth practical experience to contribute to the development of Internet-based grief interventions.

A limitation of the proposed study is that it may be difficult for older bereaved people to accept or use the Internet-based grief intervention, potentially reducing the generalizability of the findings. In addition, because of the high average dropout rates in previous Internet-based grief interventions (approximately 27%, Wagner et al., 2020), dropout rates have been taken into account when recruiting participants, and thus a minimum of 17 additional participants for each group will be recruited.

Availability of data and material

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Health Science Center, Shenzhen University (reference number PN-2020-024) on 8 September 2020.

Written informed consent will be obtained from all participants prior to their participation in the study.

Disclosure statement

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

Data availability statement

As this manuscript describes a study protocol, provision of a dataset is not applicable.

Additional information

Funding

This work was supported by the National Nature Science Foundation of China under Grant [number 32100890]; Humanities and Social Science Youth Foundation, Ministry of Education of China under Grant [number 21YJC840022]; Shenzhen Fundamental Research General Programme under Grant [number JCYJ20230808105905010]; Natural Science Foundation of Guangdong Province under Grant [number 2022A1515011097]; and The Shenzhen Humanities & Social Sciences Key Research Bases of the Center for Mental Health, Shenzhen University.