Effect of caregiving at end of life on grief,quality of life and general health: A prospective,longitudinal,comparative study

Abstract

Background:

Determining the effect of caregiving and bereavement remains a challenge. To date, no study has employed a comparison group to investigate caregivers’ grief, quality of life and general health in relation to non-caregivers.

Aim:

We aimed to determine how caregivers’ grief, quality of life and general health changed following death compared to non-caregivers and whether pre-death grief predicted these outcomes.

Design:

A prospective, longitudinal study of family caregivers and a comparison group matched for age, gender and postcode was conducted. All participants completed questionnaires at four points – once pre-death and three times post-death (3–4 months, 6–7 months and 9–10 months).

Setting/participants:

Participants (N = 70) were family caregivers of persons receiving palliative care, mostly for cancer, recruited from three palliative care providers in Western Australia and matched comparisons recruited from advertisements.

Results:

There were significant differences between the caregivers’ and comparisons’ grief, general health and quality of life at pre-death, 3–4 months and 6–7 months post-death, but not at 9–10 months post-death. The rate of progression in these constructs following death was independent from the intensity of pre-death grief. However, caregiver prolonged grief score significantly predicted prolonged grief score at 6–7 and 9–10 months post-death.

Conclusion:

It took 9–10 months for the caregivers’ grief, general health and quality of life to correspond to the comparison group. These findings present an opportunity for palliative care research and practice to consider how best to support the majority of caregivers without grief complications so that their pre- and post-death support needs are realised.

Keywords

Bereavement grief family caregivers palliative care quality of life

What is already known about the topic?

Informal caregivers are central to the provision of end-of-life care.

The experience of caregiving is well-documented, but little is known about the effect of caregiving on bereavement outcomes.

Studies investigating the effects of caregiving and bereavement are limited by a lack of an adequate comparison group.

What this paper adds?

Caregiving significantly affected caregivers’ grief, quality of life, and general health and it took nearly a year for caregivers to adapt to the impact of caregiving and bereavement.

While caring, the caregivers’ quality of life and general health was impaired and their grief was comparable to that at 3–4 months post-death.

Implications for practice, theory or policy

Research and practice must consider the support needs of the majority of caregivers with normal grief.

Comprehensive, systematic and repeated assessment of grief and the provision of tailored support, both before and after death, are needed.

Introduction

Informal caregivers are integral to end-of-life care.¹ Family caregivers report physical and emotional exhaustion,^1,2 limited participation in social and recreational pursuits,³ difficulties preparing for the death of the person for whom they care,⁴ and positive aspects of caregiving.⁵ A small proportion of caregivers, about 6%–8%, may experience persistent and pervasive grief distress called Complicated Grief or Prolonged Grief Disorder (PGD).^6–8 PGD is associated with increased risk of suicide,⁹ serious illness,^10,11 sleeping problems,¹² social and occupational impairment,¹³ and lower quality of life (QoL).¹⁰ Although these outcomes of caregiving are well-documented, determining the effects of caregiving and bereavement remain a challenge.^14,15

There is an increasing emphasis on identifying pre-death risk factors of post-death outcomes in caregivers. Studies indicate that high pre-loss grief, determined using a range of measures, is associated with post-loss grief,¹⁶ depressive symptoms,¹⁷ subjective stress¹⁸ and PGD.^19–22 One study reported that grief and depressive symptoms were higher during caregiving than 6 months after the loss and showed that severe pre-loss grief and depressive symptoms predicted post-loss complicated grief and depressive symptoms.⁸ A systematic review of studies in end-of-life caregiving showed that higher pre-loss grief was associated with higher grief post-death.²³

Demographic factors are less strongly associated with PGD than is pre-loss grief.^8,19,20 Characteristics of patient illness, care at end-of-life and circumstances of death do not appear to be associated with PGD.²⁴ Pre-loss factors associated with post-loss PGD include poor preparation for death,^22,25,26 a short caregiving duration,²⁷ fewer years of education,^8,20,22,28 depressive symptoms,^16,28 being a partner/spouse,^8,20 a history of mental health issues²⁹ and limited social support.^22,27,30 Some studies indicate that caregiver age^20,28 and gender^31,32 are associated with PGD, but other studies report no such association.⁸

The literature on the effects of caregiving and bereavement is hampered by a reliance on cross-sectional and retrospective research designs. Few studies have longitudinal designs with pre- and post-death data collection,²³ and few assessed pre-post grief over four or more time points; one study looked at dementia caregivers,³³ but not palliative care specifically. In addition, no study had a comparison group (matched to family caregivers) to determine the full effects of caregiving and bereavement, although one study did compare the caregivers’ QoL scores with age-matched normative data and showed that caregivers’ QoL was lower while caring and at 2 months post-death than the general population.³⁴ A comparison group is important because caregivers may be highly distressed upon bereavement due to the tasks of caring. Thus, even if a former caregiver’s distress decreases post-death, it might not be indicative of adaptive functioning.

The aim of the study was to determine the effects of caregiving and bereavement on family caregivers of patients receiving palliative care. Prospective, longitudinal studies are required to investigate caregivers’ needs along the caregiving and bereavement trajectory and to also inform the development and implementation of appropriate and targeted interventions.^8,35 In addition, comparing caregivers to a non-caregiving group (matched for age, gender and postcode) allows the caregivers’ general health and QoL to be compared to a normative group.¹⁴ We posed two research questions:

Does family caregivers’ pre-death grief predict post-death adjustment?

How do the caregivers’ grief, QoL and general health change over time compared to non-caregivers?

Method

The study was approved by the ethics committees of South Metropolitan Health Service (Ref. number: 12/284), Hollywood Private Hospital (Ref. number: HPH378) and Curtin University (Ref. number: HR131/2012).

Setting

The study was undertaken from 2013 to 2016 in three specialist palliative care services in Western Australia—a private hospital providing in-patient palliative care in a metropolitan area, a public hospital providing consultative palliative care in a metropolitan area and a community-based palliative care service in a public hospital in a regional area.

Design

The Caregiving and Bereavement study was a prospective, longitudinal study of family caregivers’ bereavement.³⁶ Family caregivers completed questionnaires at four data collection points: once before the death (Time 1) and three times post-death—3–4 months (Time 2), 6–7 months (Time 3) and 9–10 months (Time 4) post-death. At the first data collection point, caregivers were invited to participate in a face-to-face, semi-structured interview about preparedness for bereavement.⁴ A comparison group completed related questions at the same time points.

Participants

Caregivers were recruited from the three specialist palliative care services, in collaboration with healthcare staff. To be included, caregivers had to be aged 18 years and over, be an informal caregiver of a patient receiving palliative care and the patients’ health had to be described by the treating healthcare team as ‘stable’.³⁷ Caregivers were ineligible if they exhibited cognitive decline, as determined by the healthcare team and/or did not speak or understand English.

Comparisons were recruited via advertisements on a university radio station, advertisements in print media and flyers placed on community noticeboards. They were ineligible to join the study if currently, or in the previous 2 years, they had (1) cared for anyone with a life-limiting or terminal illness or (2) they had been bereaved.

Measures

Grief

The Hogan Grief Reaction Checklist (HGRC) is a 61-item measure of the multidimensional nature of grief and comprises six factors: despair, panic behaviour, personal growth, blame and anger, detachment, and disorganisation.³⁸ The measure does not require participants to evaluate symptoms specifically in relation to a deceased person and is therefore appropriate as a pre- and post-death measure of grief.

Prolonged grief

The Prolonged Grief-12 (PG-12) caregiver version is a 12-item measure of caregivers’ prolonged grief arising from the impending death of a loved one³⁹ (Time 1 only). The Prolonged Grief-13⁴⁰ is a 13-item measure of symptoms of PGD in bereaved people at least 6 months after death (Times 3 and 4 only). Scores of 36 or more or the presence of all five diagnostic criteria indicate the presence of PGD (personal communication with the developer).

QoL

The QOL Index is a two-item instrument measuring QoL.⁴¹

General health and well-being

The Short Form Health Survey (SF-12v2) is the standard 12-item measure of general health and well-being, with eight subscales: physical functioning, role limitations due to physical problems, role limitations due to emotional problems, bodily pain, general health perceptions, vitality and social function.⁴²

Procedure

Recruitment to the study occurred over a 20-month period, which was slower than anticipated. Data collection at Times 2–4 could only occur after the patients’ deaths and was completed 17 months after recruitment ceased. Date of patient death was provided by the relevant palliative care provider and the questionnaires for Times 2–4 posted from the university to the bereaved caregivers. Each questionnaire pack included an invitation letter, a detailed information document, consent form, questionnaire, list of support services and a reply-paid envelope. Participants had the options of completing it on their own, with the researcher over the telephone or in person at a mutually convenient venue. Pilot testing showed that the questionnaire took 20–30 min to complete.

The comparisons responded to advertisements and were asked to confirm that they were not currently, or in the previous 2 years, caring for anyone with a life-limiting or terminal illness or had been bereaved. They then were asked to provide their date of birth, gender, and postcode, and contact details, and were asked to join the study if their age, gender and postcode matched a caregiver (some were matched to two caregivers). The comparisons received their questionnaires in the mail and posted their completed questionnaires back to the university in reply-paid envelopes. All participants received an AUD$50 voucher after completing the final questionnaire.

Statistical analyses

The caregivers’ PG-12 scores were calculated at Time 1.⁴³ The PG-13 was used to determine PGD diagnostic criteria and the total symptom score at Times 3 and 4. To answer the research questions, data were analysed using linear mixed models (LMMs) as implemented in the ‘nlme’ package in R 3.5.1. To improve the likelihood of convergence, a separate LMM was estimated for each outcome (HGRC for grief, QoL for quality life and SF-12v2 for general health). Each model was specified with a fixed effect of ‘Time’ (assessment time point), ‘Group’ (Comparisons or Caregivers) and the interaction of the two. The main effect of ‘Group’ represents the difference in mean between comparisons and caregivers at Time 1. The effect of ‘Time’ represents the average amount of change between each time point in the comparison group. Finally, the ‘Time × Group’ interaction is then the difference in the average amount of change over time between comparisons and caregivers. A random intercept and a random effect of ‘Time’ were included for each participant. Given that the data represents repeated measurements of the same individuals, an autoregressive structure was applied to the residuals (AR1). This assumes that measurements that are closer together in time will be more strongly correlated than those further apart. Statistical power for this approach was estimated using the ‘powerlmm’ package⁴⁴ with an alpha level of 0.05. An LMM estimated with a caregiver group of n = 38 and comparison group of n = 32 measured at four time points had a power level of 0.82 to detect a medium effect (equivalent to Cohen’s d = 0.5).

For the QoL and general health models, the pre-death grief (HGRC) score and its interactions were included as fixed effects to assess how pre-death levels of grief affect the caregivers’ post-death adjustment. For the grief model, analogous interpretations can be made by inspecting the correlation between the random intercept and slope. This correlation also represents the relationship between the caregivers’ pre-death levels of grief (their intercept) and their post-death changes in grief (their slope). Additional models with just the caregiver data were computed to determine if their PG-12 score (Time 1) predicted changes in grief, quality life and general health at Times 2–4 or predicted prolonged grief score at Times 3 and 4.

Results

Characteristics of the sample

Participants (N = 70) were caregivers (n = 38) and matched comparisons (n = 32). The caregivers were aged 41–77 years (M = 59.25, SD = 10.79). The comparisons were aged 41–82 years (M = 60.53, SD = 9.41). See Table 1 for demographics by group. Fisher’s exact tests showed there were no significant differences between the two groups on any variable. For caregivers, the time between completing the Time 1 questionnaire and the date of death ranged from 1 to 265 days (Mdn = 34). None of the caregivers met criteria or clinical cut-off scores for caregiver prolonged grief or PGD. The mean values and SDs for each measure are shown in Table 2.

Table 1.

Demographics for caregivers (n = 38) and matched comparisons (n = 32).

Characteristic	Caregivers		Comparisons
Characteristic	N	%	N	%
Gender
Male	10	26.3	10	31.3
Female	26	68.4	22	68.8
Missing	2	5.3	0	0
Marital status
Single/Never married	2	5.3	2	6.3
Widowed	3	7.9	4	12.5
Divorced/separated	4	10.5	4	12.5
Married/de facto	27	71.1	22	68.8
Missing	2	5.3	0	0
Cultural background
Non-Aboriginal Australian	20	52.6	14	43.8
Aboriginal and/or Torres Strait Islander	0	0	5	15.6
Other English speaking	14	36.8	8	25.0
Non-English speaking	2	5.3	5	15.6
Missing	2	5.3	0	0
Usual employment
Paid employment	15	39.5	14	43.8
Pensioner	14	36.8	10	31.3
Self-funded retiree	4	10.5	5	15.6
Student	1	2.6	0	0
Other (e.g. carer’s allowance)	2	5.3	3	9.4
Missing	2	5.3	0	0
Current employment
Paid employment	12	31.6	13	40.6
Pensioner	14	36.8	10	31.3
Self-funded retiree	4	10.5	4	12.5
Other (e.g. carer’s allowance)	5	13.2	5	15.6
Extended sick leave	1	2.6	0	0
Missing	2	5.3	0	0
Patient
Spouse/partner	18	47.4
Parent	14	36.8
Sibling	1	2.6
Child	1	2.6
Other (friend)	2	5.3
Missing	2	5.3
Length of palliative care
Less than 1 week	0	0
Between 1 week and 1 month	2	5.3
Between 1 month and 6 months	7	18.4
Between 6 months and 1 year	7	18.4
More than 1 year	18	47.4
Missing	4	10.5
Broad diagnostic category
Cancer	31	81.6
Other diagnoses (e.g. scleroderma and heart disease)	5	13.1
Missing	2	5.3

Table 2.

Descriptive statistics for each measure across time.

Measure	Time 1		Time 2		Time 3		Time 4
	Caregivers	Comparisons	Caregivers	Comparisons	Caregivers	Comparisons	Caregivers	Comparisons
	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
HGRC	130.71 (29.28)	113.88 (28.73)	130.53 (30.74)	114.29 (32.01)	124.45 (30.29)	117.66 (31.12)	120.26 (22.63)	118.41 (32.28)
PG-12	26.60 (8.33)
PG-13					24.04 (5.62)		21.23 (4.44)
QOL	9.66 (4.16)	15.77 (3.94)	12.59 (3.79)	15.66 (3.85)	12.72 (3.82)	15.96 (4.47)	14.04 (3.88)	15.56 (4.72)
SF-12v2	36.09 (4.09)	39.91 (3.11)	37.36 (3.32)	39.71 (3.42)	37.84 (3.15)	38.95 (3.12)	38.62 (3.09)	38.87 (3.72)

SD: standard deviation; HGRC: Hogan Grief Reaction Checklist, PG-12: Prolonged Grief-12 caregiver version; PG-13: Prolonged Grief Disorder Scale; QOL: Quality of Life Index; SF-12v2: The Short Form Health Survey.

Grief

The results of the LMM for grief indicated that, at Time 1, caregivers had significantly higher levels of grief than the comparisons (β = 17.56 (4.00, 31.12), p = 0.012). As would be expected, the level of grief in the comparison group did not significantly change over time (β = 1.59 (–0.90, 4.09), p = 0.209). In contrast, the level of grief in caregivers significantly declined over time (β = –4.51 (–8.14, –0.87), p = 0.015). Figure 1 shows the mean grief score for each group at each time point. To test if there was a difference between the two groups at the final time point, the ‘Time’ variable was rescaled in the LMM so that the intercept refers to the final time point. With this adjustment, the ‘Group’ coefficient then represents the difference between the caregiver and comparison groups at the final measurement, rather than the first. This re-scaled LMM indicates that the difference in grief between caregivers and comparisons at 9–10 months post-death was not significant (β = 4.03 (–10.87, 18.94), p = 0.591). The correlation between the intercept and slope parameters, however, was very weak: r = –0.059 (–0.53, 0.45), suggesting that the rate of change in grief in the post-death period was not related to pre-death grief. In addition, caregiver prolonged grief score (Time 1) did not predict changes in grief score (HGRC) over time. However, caregiver prolonged grief score (PG-12 at Time 1) significantly predicted PG-13 score at Time 3 (β = 0.56, p =< 0.001) and at Time 4 (β = 0.31, p = 0.014).

Figure 1.

Caregivers’ and comparisons’ grief over time.

QoL

The LMM for QoL showed a similar pattern of results. At Time 1, caregivers demonstrated significantly lower levels of QoL than the matched comparisons (β = –4.85 (–6.61, –3.08), p < 0.001). The QoL in comparisons did not significantly change over time (β = –0.06 (–0.51, 0.40), p = 0.806), whereas it significantly improved over time in caregivers (β = 1.48 (0.83, 2.13), p < 0.001; Figure 2). As with the grief model, the mean level of pre-death grief was not significantly associated with the rate of change in post-death QoL in either the comparisons (β = –0.01 (–0.02, 0.01), p = 0.400) or the caregivers (β = –0.01 (–0.02, 0.02), p = 0.798). As with grief, this pattern of change would suggest that the differences in QoL between the two groups would no longer be significant at the final follow-up. The re-scaled LMM indicates that at 9–10 months post-death, there is no significant difference between caregivers and comparisons (β = 0.40 (–2.32, 1.51), p = 0.675). In addition, caregiver prolonged grief score (Time 1) did not predict changes in QoL over time.

Figure 2.

Caregivers’ and comparisons’ quality of life over time.

General health

The results for general health were consistent with those of grief and QoL. At Time 1, caregivers showed significantly poorer general health and well-being than the matched comparisons (β = –2.93 (–4.43, –1.42), p < 0.001). The comparisons again displayed no significant change over time (β = –0.34 (–0.76, 0.08), p = 0.109), while the health and well-being of caregivers significantly improved (β = 1.15 (0.59, 1.75), p < 0.001; Figure 3). There was also no relationship between the mean level of pre-death grief and the rate of change in post-death general health scores in either the comparisons (β = 0.01 (–0.01, 0.02), p = 0.721) or the caregivers (β = 0.01 (–0.01, 0.03), p = 0.568). Finally, the time-rescaled LMM showed no significant difference in general health and well-being between caregivers and matched comparisons at 9–10 months post-death (β = 0.53 (–1.03, 2.08), p = 0.501). In addition, caregiver prolonged grief score (Time 1) did not predict changes in general health over time.

Figure 3.

Caregivers’ and comparisons’ general health over time.

Discussion

Caregiving has a significant impact on caregivers. While caring, the caregivers’ QoL and general health was low in comparison to their post-death scores. Their grief while caring was comparable to that at 3–4 months post-death, and it took 9–10 months post-death for caregivers’ grief, QoL and general health to return to a normative standard. These novel findings present a challenge to research and practice that has moved towards focusing upon the small minority of bereaved caregivers with PGD and not on the more common experiences of grief.⁴⁵

The rate of change in grief, QoL and general health following death was unrelated to the intensity of pre-death grief, whether measured by the HGRC or the PG-12. Caregiver prolonged grief (PG-12) predicted post-death prolonged grief (PG-13) scores, and this finding aligns with previous studies showing the predictive utility of the PG-12.^8,19–21 However, several factors contribute to prolonged grief symptoms post-death,^{20,22,25–27} which is why comprehensive, systematic and repeated assessment of grief, both before and after death, is recommended in research,^20,23,46,47 policy⁴⁸ and practice guidelines.⁴⁹ Pre- and post-death support that is tailored to the needs of ‘normal’ grievers, who are the majority of bereaved caregivers, might best work in partnership with explicit and careful attention to promoting informal support,⁵⁰ and could focus on bereaved caregivers’ wishes for more information on preparing for bereavement, how to harness their social networks and developing strategies for dealing with grief.⁵¹

This study demonstrates the need for palliative care services to support family caregivers while caring and after the death, including those with normal grief. Palliative care services are moving towards harnessing a suite of strategies to meet the varied support needs of caregivers pre-death. These strategies have benefits pre- and post-death,⁵² but we should not overlook the importance of post-death support. Palliative care services already have contact with family caregivers, yet do not always systematically assess caregivers’ grief support needs prior to or after the patient’s death. One study showed that only 75% of caregivers were asked about their coping, 39.4% were asked about emotional or psychological problems and 62.9% were offered information about grief and bereavement services.⁵³ Following the death, only 50.9% received contact from staff about coping or questions they would like answered.⁵³ Services could therefore capitalise on the pre-death ‘window of opportunity’ to support caregivers⁵⁴ and to build relationships in order to offer appropriate post-death support.

Limitations

Due to attrition rates, gatekeeping and the passive recruitment of participants, which most often relies on clinician invitation, participant numbers in palliative care research are relatively small.⁵⁵ Our sample was smaller than expected, despite doubling the study’s recruitment duration and attempts to recruit from additional services. We have no information about caregivers who were not invited, or declined, to participate. To avoid participant burden, we did not systematically gather information about the supports the caregivers received to assist them with care. Given that invitations to participate came from service staff, it is possible that staff were more inclined to invite caregivers who were coping better and had less pre-death grief than those who were not invited; this could explain why none of the caregivers met criteria for pre-death prolonged grief or for PGD at 6–7 or 9–10 months post-death. This contrasts with population-based studies of bereaved people,^6,7 a cohort study of bereaved caregivers in Denmark,⁸ as well as an Australian study of palliative care caregivers, which reported that 6.7% of the sample at 6 months post-death and 11.3% at 13 months post-death met diagnostic criteria for PGD.²⁰

Patients’ imminent death has been reported by caregivers as a factor for declining study participation,⁸ where in the aforementioned study, the median time from recruitment to patient death was 58 days compared to the much shorter period in our study (34 days), which could explain why recruitment was challenging. Our sample size and heterogeneity were hampered by a lack of systematic data collection relating to caregivers from their enrolment in palliative care to post-death follow-up, unlike some contexts where systematic data collection is the norm (e.g. Thomas et al.²⁰). A larger sample would also allow more complex statistical analyses related to identifying additional factors that could predict caregivers’ post-death outcomes.

Conclusion

The majority of family caregivers adapt well to bereavement. The number of prospective studies encompassing pre and post-death time points remains small, and this is the first to have a matched comparison group to determine the ‘true’ impact of caregiving and bereavement. Increasingly, attention is paid to predicting which caregivers may develop prolonged grief and less is known about the course of normative grief in this population. Our data show that caregiving significantly affects caregivers’ grief, QoL and general health, and it takes nearly a year for caregivers to accommodate and adapt to the impact of caregiving and bereavement. These findings may assist in developing the evidence base concerning bereavement care so that the pre- and post-death support needs of the majority of caregivers are addressed.

Footnotes

Acknowledgements

The authors acknowledge and express deepest gratitude to the participants of this study. The authors thank the palliative care services for their interest in this research and for assisting in recruiting participants.

Authorship

All authors made substantial contributions to the conception (L.J.B.) or design (L.J.B., S.M.A. and M.O.) of the study, to the acquisition (L.J.B. and D.H.), analysis (A.R.J.) or interpretation of data (L.J.B., S.M.A. and M.O.), or to the drafting and critical revision of the manuscript (L.J.B., S.M.A., M.O., A.R.J. and D.H.). All authors approved the final version of the manuscript and participated sufficiently in the work to take public responsibility for appropriate portions of the content.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was funded by an Australian Research Council Discovery Early Career Researcher Award (DE120101640).

ORCID iDs

Lauren J Breen

Samar M Aoun

Moira O’Connor

References

Aoun

Kristjanson

Currow

, et al. Caregiving for the terminally ill: at what cost. Palliat Med 2005; 19(7): 551–555.

Dumont

Mongeau

End-of-life care and the grieving process: family caregivers who have experienced the loss of a terminal-phase cancer patient. Qual Health Res 2008; 18(8): 1049–1061.

Hayman

Langa

Kabeto

, et al. Estimating the cost of informal caregiving for elderly patients with cancer. J Clin Oncol 2001; 19(13): 3219–3225.

Breen

Aoun

O’Connor

, et al. Family caregivers’ perspectives on preparing for bereavement: a qualitative analysis. J Pain Symptom Manage 2018; 55: 1473–1479.

Boerner

Schulz

Horowitz

Positive aspects of caregiving and adaption to bereavement. Psychol Aging 2004; 19: 668–675.

Aoun

Breen

Howting

, et al. Who needs bereavement support? A population based survey of bereavement risk and support need. PLoS ONE 2015; 10(3): e0121101.

Kersting

Brahler

Glaesmer

, et al. Prevalence of complicated grief in a representative population-based sample. J Affect Disord 2011; 131(1–3): 339–343.

Nielsen

Neergaard

Jensen

, et al. Predictors of complicated grief and depression in bereaved caregivers: a nationwide prospective cohort study. J Pain Symptom Manage 2017; 53(3): 540–550.

Latham

Prigerson

HG.

Suicidality and bereavement: complicated grief as psychiatric disorder presenting greatest risk for suicidality. Suicide Life Threat Behav 2004; 34(4): 350–362.

10.

Boelen

Prigerson

HG.

The influence of symptoms of prolonged grief disorder, depression, and anxiety of quality of life among bereaved adults. Eur Arch Psychiatry Clin Neurosci 2007; 257: 444–452.

11.

Prigerson

Bierhals

Kasl

, et al. Traumatic grief as a risk factor for mental and physical morbidity. Am J Psychiatry 1997; 154(5): 616–623.

12.

Boelen

Lancee

Sleep difficulties are correlated with emotional problems following loss and residual symptoms of effective prolonged grief disorder treatment. Depress Res Treat 2013; 2013: 739804.

13.

Simon

Shear

Thompson

, et al. The prevalence and correlates of psychiatric comorbidity in individuals with complicated grief. Compr Psychiatry 2007; 48(5): 395–399.

14.

Breen

LJ.

The effect of caring on post-bereavement outcome: research gaps and practice priorities. Prog Palliat Care 2012; 20: 27–30.

15.

Stroebe

Boerner

Caregiving and bereavement research: bridges over the gap. Palliat Med 2015; 29(7): 574–576.

16.

Romero

Ott

Kelber

ST.

Predictors of grief in bereaved family caregivers of persons with Alzheimer’s disease: a prospective study. Death Stud 2014; 38: 395–403.

17.

Levy

Martinowski

Derby

JF.

Differences in patterns of adaptation in conjugal bereavement: their sources and potential significance. Omega: J Death Dying 1994; 29: 71–87.

18.

Butler

Field

Busch

, et al. Anticipating loss and other temporal stressors predict traumatic stress symptoms among partners of metastatic/recurrent breast cancer patients. Psychooncology 2005; 14(6): 492–502.

19.

Nanni

Biancosino

Grassi

Pre-loss symptoms related to risk of complicated grief in caregivers of terminally ill cancer patients. J Affect Disord 2014; 160: 87–91.

20.

Thomas

Hudson

Trauer

, et al. Risk factors for developing prolonged grief during bereavement in family carers of cancer patients in palliative care: a longitudinal study. J Pain Symptom Manage 2014; 47(3): 531–541.

21.

Lichtenthal

Nilsson

Kissane

, et al. Underutilization of mental health services among bereaved caregiver with prolonged grief disorder. Psychiatric Serv 2011; 62: 1225–1229.

22.

Tsai

Prigerson

, et al. Longitudinal changes and predictors of prolonged grief for bereaved family caregivers over the first 2 years after the terminally ill cancer patient’s death. Palliat Med 2016; 30(5): 495–503.

23.

Nielsen

Neergaard

Jensen

, et al. Do we need to change our understanding of anticipatory grief in caregivers? A systematic review of caregiver studies during end-of-life caregiving and bereavement. Clin Psychol Rev 2016; 44: 75–93.

24.

Bruinsma

Tiemeier

Verkroost-Van Heemst

, et al. Risk factors of complicated grief in older adults. J Palliat Med 2015; 18: 438–446.

25.

Barry

Kasl

Prigerson

HG.

Psychiatric disorders among bereaved persons: the role of perceived circumstances of death and preparedness for death. Am J Geriatr Psychiatry 2002; 10(4): 447–457.

26.

Schulz

Boerner

Klinger

, et al. Preparedness for death and adjustment to bereavement among caregivers of recently placed nursing home residents. J Palliat Med 2015; 18(2): 127–133.

27.

Burton

Haley

Small

, et al. Predictions of well-being in bereaved former hospice caregivers: the role of caregiving stressors, appraisals, and social resources. Palliat Support Care 2008; 6: 149–158.

28.

Allen

Haley

Small

, et al. Bereavement among hospice caregivers of cancer patients one year following loss: predictors of grief, complicated grief, and symptoms of depression. J Palliat Med 2013; 16(7): 745–751.

29.

Smith

Kalus

Russell

, et al. Exploring current service provision for prolonged grief disorder and suggestions for future advances. Mortality 2009; 14: 53–69.

30.

Kelly

Edwards

Synott

, et al. Predictors of bereavement outcome for family carers of cancer patients. Psychooncology 1999; 8(3): 237–249.

31.

Ferrario

Cardillo

Vacario

, et al. Advanced cancer at home: caregiving and bereavement. Palliat Med 2004; 18: 129–136.

32.

Gilbar

Ben-Zur

Bereavement of spouse caregivers of cancer patients. Am J Orthopsychiatry 2002; 72(3): 422–432.

33.

Hebert

Dang

Schulz

Preparedness for the death of a loved one and mental health in bereaved caregivers of patients with dementia: findings from the REACH study. J Palliat Med 2006; 9(3): 683–693.

34.

Götze

Brähler

Gansera

, et al. Anxiety, depression and quality of life in family caregivers of palliative care patients during home care and after the patient’s death. Eur J Cancer Care 2018; 27: e12606.

35.

Grande

Stajduhar

Aoun

, et al. Supporting lay carers in end of life care: current gaps and future priorities. Palliat Med 2009; 23(4): 339–344.

36.

Breen

Aoun

O’Connor

The effect of caregiving on bereavement outcome: study protocol for a longitudinal, prospective study. BMC Palliat Care 2015; 14: 6.

37.

Eagar

Green

Gordon

An Australian casemix classification for palliative care: technical development and results. Palliat Med 2004; 18(3): 217–226.

38.

Hogan

Greenfield

Schmidt

LA.

Development and validation of the Hogan Grief Reaction Checklist. Death Stud 2001; 25(1): 1–32.

39.

Kiely

Prigerson

Mitchell

SL.

Health care proxy grief symptoms before the death of nursing home residents with advanced dementia. Am J Geriatr Psychiatry 2008; 16(8): 664–673.

40.

Prigerson

Horowitz

Jacobs

, et al. Prolonged grief disorder: psychometric validation of criteria proposed for DSM-IV and ICD-11. PLoS Med 2009; 6(8): 1–12.

41.

Graham

Longman

Quality of life in persons with melanoma: preliminary model testing. Cancer Nurs 1987; 10: 338–346.

42.

Sanderson

Andrews

The SF-12 in the Australian population: cross-validation of item selection. Aust N Z J Public Health 2002; 26(4): 343–345.

43.

Chiambretto

Moroni

Guarnerio

, et al. Prolonged grief and depression in caregivers of patients in vegetative state. Brain Inj 2010; 24(4): 581–588.

44.

Magnusson

. powerlmm: power analysis for longitudinal multilevel models (R package version 0.4.0) (Computer software), 2018, https://CRAN.R-project.org/package=powerlmm

45.

Rumbold

Aoun

SM.

Bereavement and palliative care: a public health perspective. Bereave Care 2014; 22(3): 131–135.

46.

Burns

Prigerson

Quinn

, et al. Moving on: factors associated with caregivers’ bereavement adjustment using a random population-based face-to-face survey. Palliat Med 2018; 32(1): 257–267.

47.

Prigerson

Maciejewski

PK.

Predicting prolonged grief disorder: caregiver prodrome turns bereaved survivor syndrome. Commentary on Thomas et al. J Pain Symptom Manage 2014; 47(3): 516–517.

48.

Palliative Care Australia. National palliative care standards, 5th ed. Canberra, ACT, Australia: Palliative Care Australia, 2018.

49.

Hudson

Hall

Boughey

, et al. Bereavement support standards and bereavement care pathway for quality palliative care provision. Palliat Support Care 2018; 16: 375–387.

50.

Rumbold

Aoun

SM.

An assets-based approach to bereavement care. Bereave Care 2015; 34(3): 99–102.

51.

Breen

Aoun

Rumbold

, et al. Building community capacity in bereavement support: lessons learnt from bereaved former caregivers. Am J Hospice Palliat Med 2017; 34: 275–281.

52.

Aoun

Ewing

Grande

, et al. The impact of supporting family caregivers before bereavement on outcomes after bereavement. J Pain Symptom Manage 2018; 55(2): 368–378.

53.

Aoun

Rumbold

Howting

, et al. Bereavement support for family caregivers: the gap between guidelines and practice in palliative care. PLoS ONE; 12(10): e0184750.

54.

Sealey

O’Connor

Aoun

, et al. Exploring barriers to assessment of bereavement risk in palliative care: perspectives of key stakeholders. BMC Palliat Care 2015; 14: 49.

55.

Aoun

Slatyer

Deas

, et al. Family caregiver participation in palliative care research: challenging the myth. J Pain Symptom Manage 2017; 53(5): 851–861.