Introduction

Evidence suggests that most people will find ways to adapt to bereavement and that few will need specialist grief therapies [1–3]. Concurrently, however, bereaved people have reported difficulties accessing support, be that informally from family and friends, or formally from psychologists and grief counsellors [2,4–7]. Access difficulties were likely compounded by the COVID-19 pandemic due to travel restrictions and limited face-to-face interactions with family, friends, and formal health services. This led to calls from grief researchers to prepare for a ‘silent epidemic of grief’[8,9]. Internationally several studies have found that pandemic-related challenges have impacted on the support needs of the bereaved and the support sources available [10–13]. The support needs of bereaved people in Australia during the pandemic have not been examined.

Experiences in Australia during 2020 and 2021 differed markedly from the global situation, with considerably fewer infections and deaths from COVID-19. At the same time, however, Australia experienced some of the longest and most restrictive public health measures in Western countries, introduced to reduce the spread of COVID-19 and protect health workers and the community [14]. At various times, gatherings with family and friends who did not live in the same household were heavily restricted or prohibited, and many formal supports closed or shifted online, including mental health services. International borders closed in March 2020 for almost two years, citizens had to seek permission to leave or re-enter the country and state borders closed throughout 2020–2021 inhibiting travel within the country [15]. Demand for mental health services grew. In January and February 2022, psychologists reported a marked increase in wait times, worked an average of 17 additional hours unpaid a week to accommodate higher demand, and one-third closed their books to new clients [16]. Together, these factors suggest that many bereaved Australians might have been left with unmet support needs.

In the context of providing bereavement support, it is increasingly recognised that a public health approach is needed. This approach recognises that people may benefit from different types and sources of support depending on their risk of developing complex grief reactions. It involves tailoring support approaches to meet bereaved people’s needs at universal, selective, and indicated levels [17–19]. At a universal level, there is the provision of basic information about grief and compassion to all bereaved people, but with most support coming from existing social networks. Social support is modifiable and can helpfully influence the grieving process, making it an effective way to reduce the risk of deleterious grief outcomes [20,21]. When the recipient perceives it as helpful, social support is one of the strongest determinants of positive psychosocial outcomes after bereavement [22–24]. At a selective level, additional community support is suggested to reduce the likelihood of further complications. This could involve peer support from grief support groups and organisations or support from a general practitioner. At an indicated level, which refers to people showing signs of complicated reaction or high levels of mental health distress, it is recommended that supports involve targeted therapies delivered by relevant experts [17,25,26]. For example, prolonged grief disorder, a condition which affects about 10% of bereaved individuals and is associated with significant negative health and mental health impacts, responds selectively to psychotherapies designed specifically to address it [27–29]. Prior to the pandemic, one-third of bereaved Australians reported not receiving the support they would have liked [4]. Additionally, support when provided is not always perceived by the recipient as helpful or timely [2].

The National COVID-19 Bereavement Project was established to quantify the support needs and mental health effects of Australians bereaved from any cause during the COVID-19 global pandemic. Over 2000 Australians bereaved from any cause during the first two years of the pandemic completed the survey. Maccallum et al [30] reporting on the mental health outcomes, found that 53% of sample experienced elevated symptoms of grief and/or depression and anxiety. Almost 20% of the sample reported concurrent high levels of all three. Importantly, the experience of social isolation and loneliness was an independent correlate of high symptoms, relative to low symptoms.

In this article, we present primary findings from the National COVID-19 Bereavement Project related to bereavement support use. Our aims were to 1) identify the bereavement supports used by participants, 2) examine the perceived helpfulness of the supports used, 3) establish the prevalence of unmet needs, 4) determine the sociodemographic correlates and mental health measures associated with having unmet support needs, and 5) identify the nature of the unmet support needs. This information will enable identification of vulnerable groups and development of strategies to address unmet needs of bereaved Australians now, and in future pandemics.

Methods

Results

Characteristics of the sample

A total of 1,878 bereaved people (94.9% women; mean age 55.1 years) completed the survey (see Table 1 and S2 File). Most resided in major cities (68.4%), were university educated (87.7%), employed (59.8%), and partnered (59.7%). One-quarter reported living alone (25.9%). Most participants reported they had some unmet support needs (62.0%). People with unmet needs were significantly younger and indicated higher levels of distress on the mental health measures. They also reported different kinship relationships of the deceased and younger age of the decedent. Cause of death did not differ between groups, nor did relationship status or living alone. People with unmet support needs reported higher levels of anxiety, depression, and prolonged grief symptoms (see Table 1). For space, the experience of COVID-19 restrictions for the total sample and differences between groups are reported in S3 File.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participant characteristics, death characteristics, and mental health measures by support needs.

https://doi.org/10.1371/journal.pone.0304025.t001

Use of bereavement supports

On average, participants used a mean of 2.3 (SD = 1.8) supports (see Table 2). Overall, family and friends (85.1%) and self-help resources (27.2%) (e.g., books, websites) were used most. One-fifth visited a general practitioner (21.5%) and 19.6% visited a psychologist. Approximately 17.6% accessed online support groups (e.g., via Facebook). Those with unmet needs sought support from more sources (see Table 2; mean = 2.5, SD = 1.9) than those whose support needs were met (mean = 2.1, SD = 1.8). Those with unmet support needs were less likely to report using family and friends for support (p<0.001) but more likely to report using self-help resources (p<0.001), internet/online community groups (p<0.001), informal bereavement support groups (p = 0.018), formal grief support groups (p = 0.034), general practitioners (p = 0.043), psychologists (p = 0.006), and psychiatrists (p = 0.008).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Bereavement supports used, categorised by unmet support needs.

https://doi.org/10.1371/journal.pone.0304025.t002

Perceived helpfulness of bereavement supports

Most participants indicated that family and friends were the most helpful support (66.7%), followed by psychologists (8.9%), self-help resources (4.9%), and grief counselling (4.9%) (see S4 File). In the open-ended question, the largest proportion of participants (27.9%) reported that nothing was unhelpful for them. The next largest proportion of participants indicated family and friends (23.9%) were unhelpful, followed by psychologists (4.7%). Open-ended responses presented in Table 3 provided some insight into why family and friends, community, and formal supports were perceived either as helpful or unhelpful.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Reasons why supports were perceived as helpful or unhelpful as reported by participants in open-ended responses.

https://doi.org/10.1371/journal.pone.0304025.t003

Reasons for family and friends being the most helpful included the shared experience of the death; the pre-existing nature of the relationship which facilitated feelings of support in their grief; being the most readily available support due to proximity or availability during lockdowns; and their ability to offer emotional and practical support in different ways. In contrast, family and friends were perceived as unhelpful due to lack of face-to-face availability, escalating family tensions compounded by COVID-19, platitudes, or other unhelpful comments; lack of sustained offers of support; the perception that emotional capacity of family and friends was already stretched due to COVID-19; and perceived pressure from friends and family to move on with life.

Community support groups and online supports, such as Facebook and Reddit were perceived as helpful as they allowed participants to interact with others who had similar experiences where participants felt they could express their emotions without judgement. In contrast, participants described community supports as unhelpful due to lack of specificity and increased distress from being exposed to others’ grief. Seeing/reading/hearing about others’ grief through community supports was perceived to worsen mental health. For similar reasons, social media was noted as unhelpful.

Formal supports were perceived to be helpful due to ability to provide specialist care, including targeted strategies for managing emotions; objectivity and ability to offer non-biased advice. Pre-existing relationships were once again noted, where trust had already been established. Participants felt that they could express emotions freely, where they did not feel comfortable doing so with family and friends. On the contrary, formal supports were perceived as unhelpful due to not being specialised in trauma/grief counselling, or the perception that the practitioner was trying to “fix” their grief, while the participants viewed grief as being a natural process. Other reasons included the out-of-pocket cost of formal supports, delay in appointments with some waiting up to 15 months, and others found telehealth unsuitable.

Correlates of unmet support needs

Results of the multiple logistic regression analysis are summarised (see Table 4). Older age was associated with having needs met (OR = 0.96; 95%CI 0.95 to 0.96). Four COVID-19-related challenges were positively associated with unmet support needs; COVID-19 impacted their ability to care for the deceased (OR = 1.84; 95%CI 1.47 to 2.32); to know what was happening to them (OR = 1.58; 95%CI 1.16 to 2.16); to say goodbye (OR = 1.48; 95%CI 1.16 to 1.88); and, limited close contact with family and friends after the death (OR = 1.81; 95%CI 1.45 to 2.24).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Associations of unmet needs with independent variables: Multiple logistic regression.

https://doi.org/10.1371/journal.pone.0304025.t004

Categories of unmet needs

Participants responded to a multiple-choice question asking what supports they would have wanted but did not or could not access. In the order of increasing need, 11.6% indicated that information would have helped, 28.9% selected community-based supports, such as grief support groups, as well as community and peer support groups, and 29.3% selected professional supports. One-quarter were unsure what would have helped (25.6%) and 23.4% said nothing would have helped. A total of 1,233 participants (66%) provided an open-text response to a question inquiring about additional unmet needs (6% of these responses were excluded from analysis as they involved descriptions of grief reactions or the compounding difficulties they experienced during the pandemic, rather than specific unmet needs). Of the remaining responses, twelve categories of unmet needs were identified (see Table 5). The most common reported unmet need was for social support after the death. Participants described the need to give and receive in-person support from their family/friends living within and outside Australia; the need for family togetherness; physical comfort; and someone to ‘check in’ initially and periodically after the death. Linkage and access to professional mental health support during the pandemic was the second most reported unmet need. Participants also expressed a range of unmet needs in relation to the time of death. When the death occurred in an inpatient setting, many participants reported a need for compassion and flexibility in visiting rules before, during, and after death so they could be present at the time of death; say goodbye; and spend time with the body. The need for clear, timely and consistent communication from healthcare professionals and government was often noted, particularly during periods of public health restrictions where family/friends had limited or no access to inpatient settings. Participants reported the need for staff to inform family of prognosis and approaching death; prepare family for the dying process; answer questions; provide information about the cause of death; involvement in achieving last wishes; and ensure compassionate delivery of death notice to each family member. Clear provision of information and compassion from the government regarding public health restrictions was also reported, including the need for timely communication of changing restrictions; clear instructions on how to obtain travel exemptions; and clear communication regarding rules for restrictions, such as rules surrounding compassionate visits. Confusion over public health restrictions resulted in worry and concern.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Categories of unmet needs as reported by participants to open-ended items.

https://doi.org/10.1371/journal.pone.0304025.t005

Discussion

This study found that Australians bereaved during the COVID-19 pandemic at all levels of need (universal, selective, and indicated) used multiple avenues of support during this time, that no support appeared universally helpful, and that those with unmet support needs were more likely to be younger; have experienced the loss of a partner or child; and were impacted by COVID-19 in ways that affected their ability to care for the dying, understand what was happening to them, be with friends and family, and say goodbye.

The most common and helpful support was indicated to be family and friends, which is consistent with pre-pandemic literature [2] and aligns with public health recommendations for the provision of universal bereavement support [17–19]. Importantly however, for many, friends and family were also described as the least helpful source of support. Some reasons for this were analogous to pre-pandemic reasons, such as limited empathy, insensitivity, and poor advice. However, other reasons for unhelpfulness were specific to the circumstances of public health measures, including limited face-to-face contact, the perception that others could not tolerate their grief during a global pandemic, and disrupted relationships due to opposing views on the pandemic, vaccination, and compliance with COVID-19 public health measures. These reasons could compound feelings of social disconnection where the bereaved are unwilling or unable to express their feelings in a social context due to the perceived negative consequences of doing so [23]. We also found evidence of this with participants reporting friends and family as too emotional themselves to be able to provide support, and thus did not seek their support or perceived them as unhelpful (see unhelpful supports table–see Table 3). Combined, these factors may contribute to people masking their grief [8] contributing to increased loneliness and social isolation, which has been linked with increased psychological distress in this sample [30] and others [22,23]. The Irish Hospice Foundation and the European Grief Model have recently proposed an additional tier to the public health model, where the three above are underpinned by grief literacy for all [18,19]. Grief literacy serves as a way to improve social support by helping people, communities and society to identify grief more readily and have the knowledge and skills to seek out relevant information and supports to avoid negative outcomes associated with grief [37,38] In a grief literate society, people would openly speak about their losses and be comfortable hearing of others’ rather than showing discomfort and masking grief [37].

It is important to recognised that despite the literature highlighting the potential value of social support for grieving persons, a systematic review on social support following bereavement concluded that the body of research is highly fragmented, with methodological flaws and omissions (i.e., student samples as proxies for the general community, numerous biases, limited controls, poor quality) [39]. Thus, calls to transfer grief support to social networks and communities unprepared to provide it must be avoided due to its potential to cause further harm [37]. These challenges identified here with social support are important to consider in pandemic planning for bereavement support, as these complicating factors might hinder use and perceived helpfulness, which are typically considered sufficient buffers against the deleterious effects of bereavement [24].

Perhaps it is not surprising that self-help resources were the second most used source of support given the limited availability of face-to-face social supports. Participants described the helpfulness of self-help resources in terms of its availability (e.g., book, online resources) and the comfort they drew from being able to connect with and read about others’ experiences. This might have facilitated the sense of belonging to a group with shared experiences and concerns. Pre-pandemic, a sense of belonging and shared experiences was a top reason why people felt family and friends were helpful [2]; however, during the pandemic, it appears self-help resources such as books and information on grief-related websites may have been able to address some of these needs. Self-help resources such as these reflect bereavement support recommended at the universal and selected level.

Formal supports such as primary health practitioners and psychologists, recommended at the indicated level for people showing signs of prolonged grief disorder, were used by approximately one in five participants. The levels of psychologist use in this study are higher than reported pre-pandemic [1], suggesting a higher level of distress during the pandemic, or greater awareness and acceptance of seeking help. People perceived formal support as helpful due to the objectivity they provided; however, in some cases they were perceived as unhelpful due to limited specialist skills in grief counselling and interventions. Several studies have shown that such knowledge is not included or well-integrated in the preparation of health professionals [40,41] and that concerns about the training and skill of providers is a reason why people hold negative attitudes about grief counselling [42]. Strategies to increase workforce capacity to support the increasing number of people bereaved now and for future pandemics are urgently needed. There are however multiple challenges facing the field in providing care under normal circumstances and gaps in the literature regarding treatment for prolonged grief. Whilst we may be able to identify prolonged grief earlier than 12 months and identify candidates of preventative care [43], it is less clear what effective early care for prolonged grief looks like. Preliminary work investigating an internet-based therapist-assisted prevention intervention for prolonged grief disorder has shown promising results [44], however there remains a scarcity of research in this area. In addition to different needs at different levels, people will also have different needs and different times. This is an area to be explored in analyses of longitudinal data from this study.

Concerningly, during the COVID-19 pandemic, twice as many people had unmet bereavement support needs compared to pre-pandemic reports [1,4]. Needs ranged from practical to emotional support and emerged before the death, at the time of death, and after the death. This highlights the complexity of supporting the bereaved and the need to develop a comprehensive action plan in collaboration with a wide range of stakeholders. These include people involved in the care of the dying (where relevant), and people involved in care of the family after the death. In other words, work is needed to actualise a continuum of bereavement support that begins with end-of-life care [45]. Additionally, a key area of unmet need specific to the pandemic context was clarity of communication from government, clinicians, and public health announcements about restrictions. People reported confusion about what the restrictions were at a given time and sought clarification for compassionate visits and exemptions without success which added stress. The rationale for public health protections and clarity and simplicity of processes for seeking and being granted exemptions must be addressed in planning for future pandemics.

The strongest correlates of unmet support needs were experiencing the death of a partner or child. A close relationship to the deceased was also a significant correlate of poorer mental health outcomes in the sample [30]. Inability to spend time with the dying person, being unaware of what was happening to them, and being unable to say goodbye due to restrictions were also independent correlates of unmet needs in this study. Public health measures necessarily consider possible impacts of a pandemic at a population level, rather than the individual level, and set regulations, accordingly. In circumstances, such as the first 12 months of the global pandemic, when much was unknown about COVID-19 and its mutations (e.g., transmissibility, likelihood of severe disease/death across age groups, effectiveness of vaccination), the precautionary principle drove more restrictive settings. At a population level, this approach in Australia was successful by international comparisons [46]. For individuals, however, (in this case, the bereaved), there were significant negative consequences related to those settings. Those responding to future pandemics might specifically consider carers of people in the final stages of life, their families, and the recently bereaved for possible modifications to some of the restrictions. At a practical level, spaces could be redesigned to allow for safe access to enable togetherness. For example, the built environment of nursing homes (many communal facilities) as well as the care provision required (close contact) are among several factors that enhance risk of transmission in this setting. Evidence suggests that nursing homes that experienced larger COVID outbreaks were those accommodating a larger number of residents, with more shared rooms, and more likely to report non-compliance with regulations [47]. This supports the redesign of the built environment of care spaces as well as adapted operations (compliance with sanitary measures) in the short and long term to better prepare us for future outbreaks and pandemics and enable opportunities for togetherness that are safe [47].

Finally, the Productivity Commission’s Inquiry into Mental Health 2022 emphasised clinical services over the importance of family and social networks in providing support [48]. Given that social supports were relied on most, perceived as most helpful, and lack thereof independently accounted for variance in unmet support needs, a clear recommendation for a bereavement support action plan is to bolster social networks’ ability to provide support in times of loss; although must not be used as the only strategy. As suggested by the public health model of bereavement support, formal specialist grief therapy is required for some, but all levels of support are needed [1].

Conclusions

This study demonstrated the complexity of bereavement support needs during a pandemic. Specialised grief therapy needs to be more readily available to the minority of grievers who would benefit from it. In tandem, because two-thirds of participants reported unmet support needs, the urgency and necessity of bolstering bereavement support at every level—universal, selective, indicated—is highlighted. Support in the wake of bereavement is essential to help the increasing number of people bereaved now and for future pandemics.

Supporting information

Acknowledgments

The authors would like to thank the bereaved participants who completed our survey. We would also like to acknowledge the research assistants who were employed on this project (blinded for review).