Risk Factors for Complicated Grief in Older Adults

Abstract

Background:

Older adults grieving the death of a spouse have been found to have a higher risk of complicated grief compared with younger adults.

Objective:

The study objective was to find out whether personal characteristics of the patient and the bereaved partner, or characteristics of the patient's illness, end-of-life care, and the nature of death are risk factors for complicated grief in older adults.

Methods:

We performed a nested case-control study within the Rotterdam Study. We selected 100 couples of which one person had deceased and the other person experienced “complicated grief,” and 100 control couples of which one person had deceased and the other person experienced “normal grief.” Complicated grief was assessed with a 17-item Inventory of Complicated Grief (ICG). Determinants were assessed using several sources of information that were available for all participants of the Rotterdam Study. Additionally, medical files of the deceased were manually screened. Logistic regression analysis was performed.

Results:

Only depression at baseline was significantly associated with complicated grief. Bereaved partners with depression at baseline had a higher risk of complicated grief compared to bereaved partners without depression (OR=3.48; 95% CI=1.40–8.68).

Conclusions:

Our results suggest that complicated grief in older adults is not clearly related to the circumstances of dying of the deceased partner. Preexisting conditions such as depression seem to be more important in explaining the occurrence of complicated grief.

Introduction

Bereavement is a common experience in older adults. Although the majority of adults recover after the loss of a loved one, a portion continues to grieve for an extended period of time and develops symptoms of a state known as “complicated grief”¹ or Prolonged Grief Disorder.² This type of grief is distinct from “normal grief,” as the person cannot accept the death and instead experiences disbelief and preoccupations with the deceased person.^1,3 Complicated grief can be associated with a number of negative health outcomes, such as hypertension, sleep impairment, and suicidality.^4–6 Among older adults, complicated grief is underdiagnosed, minimized as a factor affecting mental health and function, and undertreated.^3,7,8 Although the bereavement experience of older adults has been associated with less mourning,⁹ secondary consequences such as social isolation may lead to grief of longer duration and poorer health and mental health outcomes than observed in younger persons. Older adults grieving the death of a spouse have been found to have a higher risk of complicated grief compared with younger adults.^7,8 A Dutch study showed that the prevalence of complicated grief in older adults in the general population is 4.8%.¹

Knowledge about vulnerability to complicated grief can provide opportunities to target care and resources appropriately.¹⁰ The literature abounds with factors that are considered useful in identifying who is at risk of complicated grief.^10–14 A great deal of attention is paid to the characteristics of the bereaved, such as age, sex, and physical and mental illness.¹⁰ Another area of interest is the deceased's illness, the characteristics of end-of-life care, and the nature of the patient's death.¹⁰ It has been shown that there is an increased risk of complicated grief if the duration of the patient's terminal illness was either very short or very long;¹² if the patient had suffered from a cognitive impairment;¹² if the patient died in a hospital setting;¹³ if the patient died after euthanasia;¹⁴ and if death occurred suddenly or unexpectedly.¹⁰ Multiple care transitions and hospitalizations in the last phase of life could also influence grief experiences.¹⁵ However, little is known about these characteristics as potential risk factors for complicated grief in older adults. Since the bereavement experience of older adults is different from younger adults, this potentially also holds true for factors that contribute to this bereavement experience. To examine whether complicated grief in older adults can be explained by predeath information, this study includes both personal and situational factors in a comprehensive analysis. The aim of the present study is to find out whether personal characteristics of the patient and the bereaved partner, or characteristics of the patient's illness, end-of-life care, and the nature of death are risk factors for complicated grief in older adults.

Methods

This study was based on the Rotterdam Study, an ongoing prospective cohort of older adults to examine the occurrence and risk factors of chronic diseases.¹⁶

The Rotterdam Study¹

The Rotterdam Study comprises two cohorts. The first stems from the original study which commenced in 1990–1993. At this time all inhabitants aged over 55 years living in the Ommoord district of Rotterdam were invited to participate; 7983 persons (78%) participated. In 2000, people who had become 55 years of age, or who were 55 years or over and had moved into the study district after the start of the study, were added as a second cohort; 3011 (67%) participated. A detailed description of the design of the Rotterdam study has been published elsewhere.¹⁷

Nested case control

Loss of a spouse is very common and accounts for a large proportion of losses among older adults.¹⁸ Therefore, we selected 200 (married/partnership) from the Rotterdam cohorts, in a nested case-control design:¹⁹ 100 couples of which one person had deceased and of which the other experienced complicated grief, and 100 control couples of which one person had deceased and the other person experienced normal grief. Group matching was used to increase statistical power.

Case definition

In case a participant has lost a spouse, grief was assessed in the original cohort in the fourth follow-up examination (2002–2004) and in the added cohort in the second follow-up examination (2004–2005).¹ Complicated grief was assessed with a 17-item Dutch version of the Inventory of Complicated Grief (ICG), constructed by Prigerson et al. (1995).²⁰ The ICG is the most widely used instrument to measure complicated grief, and items represent the array of symptoms attributed to complicated grief. The measure has high internal consistency and convergent and criterion validity and it is considered the gold standard for measurement of complicated grief in older adults.¹ A summary score for the ICG was calculated by adding up each individual item score (responses from 0=never to 4=always) across the 17-items, providing a potential total score range of 0 to 68. Participants with a total score greater than 21 and with symptoms reported to have been present for at least six months were considered to have complicated grief. In total, 1089 (19%) participants reported that they were experiencing grief at the time of the assessment and of these, 277 participants were assessed as having complicated grief. A detailed description of the measurement of complicated grief has been published elsewhere.¹

Assessment of determinants

Several sources of information were available for all participants of the Rotterdam Study: home interviews, examinations at the research center, Nationwide Medical Registry, and general practitioners' records. A detailed description of the methods for data collection of the Rotterdam study has been published elsewhere.²¹ These sources were used for the assessment of personal characteristics as well as characteristics of patient's illness, end-of-life care, and the nature of death as potential risk factors for complicated grief.

Personal characteristics at baseline

Information was gathered on sex, age, and ethnicity (Caucasian or non-Caucasian). Working status was recorded and recoded into retired from full-time work/unemployed or working. Education was grouped according to the Dutch Standard Classification of Education (Dutch Central Bureau of Statistics, 1989).²² The ratings for low (1) to high education (6) were recoded into low, intermediate, and high education. Cognitive capacity was assessed with the Mini Mental State Examination (MMSE), which assesses six broad areas of daily cognitive functions.²³ Activities of daily living performance was assessed with the Stanford Health Assessment Questionnaire²⁴ and the Instrumental Activities of Daily Living scale.²⁵ Depression was evaluated with the use of the Centre for Epidemiological Studies Depression (CESD) scale (score ≥16).²⁶

Patient's illness, end-of-life care, and the nature of death

Health events were coded according to the 10th edition of the International Classification of Disease (ICD).²⁷ The underlying causes of death were recorded and classified as neoplasms, diseases of the circulatory system, and other causes of death. Information was gathered on the date and place of death. Place of death was recorded as at home; in hospital; community living (i.e., home for the elderly/nursing home); or other places. Further, information was obtained on hospital admissions in the last year of life (number and median number of days).

Medical files of the deceased were manually screened for additional information on the patient's illness, characteristics of end-of-life care, and the nature of the patient's death. Based on the literature, a checklist was developed by the research team. The checklist was piloted before use: 10 files were independently assessed by Sophie Bruinsma (SB) and Judith Rietjens (JR) and discussed. This led to some small changes in the checklist. A second pilot was performed to test these changes; no disagreement was found. As a next step, all medical files were checked by SB with the use of this checklist. Several topics were covered. The duration of the illness that was the underlying cause of death was registered as a continuous variable. The nature of death was classified as ‘completely unexpected,’ ‘patient was ill, but death occurred unexpectedly,’ and ‘expected.’ ‘The number of transitions between care settings in the last three months of life’ was registered. End-of-life decision making was classified as ‘withholding or withdrawal of potential life-prolonging treatments,’ ‘euthanasia,’ and ‘other end-of-life decisions.’

Data analysis

To increase the comparability of analyses and reduce bias, missing values (if >5%) were imputed with randomized single imputation.²⁸ This was the case for ‘the duration of the illness’ (30%), ‘the expected nature of death’ (9%), and ‘the number of transitions between care settings in the last three months of life’ (14%). To determine the association between potential risk factors—personal characteristics, characteristics of the deceased person's illness, end-of-life care, nature of death—and the grief response, logistic regression was performed. Those variables that showed a p-value ≤0.20 in the univariate regression analysis were included in the multivariate regression model.

Results

The characteristics of the deceased and the bereaved partner are presented in Appendix 1. The majority of the deceased persons were male (66%). They were on average 74 years old at the time of death. Almost all were Caucasian (99%). At baseline, most deceased persons had been retired or unemployed (85%) and the highest education attained was predominantly low or intermediate (91%). Bereaved partners were predominantly female (66%). They were on average 73 years old at the time of their spouse's death. All partners were Caucasian (100%). At baseline, 89% of the partners had been unemployed or retired, and the highest education attained was predominantly low or intermediate (93%). Of all partners, 19% suffered from depression at baseline. On the activities of daily living scale, partners scored on average 24 (‘little difficulties’) and on the cognitive status scale on average 28 (‘normal cognition’). The median of the time passed between a person's death and the interview with the bereaved partner was 34 months (IQR 14–68).

Table 1 describes the characteristics of the deceased persons and the bereaved partners in relation to the occurrence of normal and complicated grief. Very few differences were found between the types of grief with regards to personal characteristics. Of the bereaved partners with normal grief, 9% suffered from depression at baseline, compared to 25% of the bereaved partners with complicated grief. The median time between a person's death and the interview with the bereaved partner was 39 months (IQR 16–67) for normal grief, compared to 33 (IQR 13–68) for complicated grief. Univariate regression analysis showed that bereaved partners with depression at baseline had a higher risk of complicated grief (OR=3.86; 95% CI=1.60–9.33).

Table 1.

Differences Between Older Adults Experiencing Normal Grief or Complicated Grief: Characteristics of Deceased and Bereaved Partners (N=200)

			Difference between groups, unadjusted ^a		Difference between groups, adjusted for sex and age ^b
Deceased	Normal grief (n=100)	Complicated grief (n=100)	Overall p-value	OR (95% CI)	Overall p-value	OR (95% CI)
Sex
Male	65 (65%)	67 (67%)	0.765	1.00 (ref)	0.148	1.69 (0.83–3.44)
Female	35 (35%)	33 (33%)		0.92 (0.51–1.64)
Age at death (mean±SD)	74.88±7.11	74.08±7.21	0.428	0.98 (0.95–1.02)	0.064	0.93 (0.86–1.00)
Working status (baseline)
Working	14 (14%)	16 (16%)	0.626	1.00 (ref)	0.702	1.00
Unemployed or retired	86 (86%)	81 (81%)		0.82 (0.38–1.80)		0.86 (0.39–1.90)
Missing	-	3
Education (baseline)
Low	49 (49%)	51 (51%)	0.864	1.00 (ref)	0.314	1.00
Intermediate	40 (40%)	38 (38%)		0.91 (0.51–1.65)		0.82 (0.44–1.51)
High	11 (11%)	8 (8%)		0.70 (0.25–1.88)		0.62 (0.22–1.70)
Missing	-	3
Bereaved partners
Sex
Male	36 (36%)	33 (33%)	0.655	1.00 (ref)	0.580	1.00
Female	64 (64%)	67 (67%)		1.14 (0.64–2.05)		1.18 (0.65–2.15)
Age at death of partner (mean±SD; n=198)	72.98±6.76	72.94±6.48	0.966	0.99 (0.96–1.04)	0.949	1.00 (0.96–1.05)
Working status (baseline)
Working	11 (11%)	10 (10%)	0.818	1.00 (ref)	0.750	1.00
Unemployed or retired	89 (89%)	90 (90%)		0.90 (0.36–2.22)		1.16 (0.46–2.94)
Education (baseline)
Low	63 (63%)	62 (62%)	0.645	1.00 (ref)	0.787	1.00
Intermediate	28 (28%)	34 (34%)		1.23 (0.67–2.27)		1.31 (0.68–2.51)
High	9 (9%)	4 (4%)		0.45 (0.13–1.54)		0.48 (0.13–1.74)
Depression (CESD at baseline)
No depression	81 (81%)	66 (66%)	0.004	1.00 (ref)	0.003	1.00
Depression	9 (9%)	25 (25%)		3.49 (1.49–7.81)		3.86 (1.60–9.33)
Missing	10	9
ADL^c (at baseline; n=190)	24.44±3.03	24.08±3.41	0.444	0.97 (0.88–1.06)	0.441	0.97 (0.88–1.06)
Cognitive status^d (MMSE at baseline; mean±SD; n=194)	28.24±1.40	27.93±1.46	0.130	0.85 (0.70–1.05)	0.132	0.85 (0.70–1.05)
Time since death of partner, months,^e median (IQR); n=196	39 (16–67)	33 (13–68)	0.812	0.99 (0.99–1.01)	0.678	1.00 (0.99–1.01)

Difference test is based on univariate logistic regression. Reference group is normal grief.

Difference test is based on multivariate logistic regression. Reference group is normal grief.

Total score with a minimum of 0 (much difficulties with ADL) and 27 (little difficulties with ADL).

Any score ≥27 points (out of 30) indicates a normal cognition. Below this, scores can indicate severe (≤9 points), moderate (10–18 points), or mild (19–24 points) cognitive impairment.

Time in months between death of the deceased and the interview where type of grief was assessed. ADL, activities of daily living; CESD, Centre for Epidemiological Studies Depression scale; CI, confidence interval; IQR, interquartile range; MMSE, Mini Mental State Examination; OR, odds ratio; ref, SD, standard deviation.

Table 2 describes the characteristics of deceased persons' illness, end-of-life care, and nature of death in relation to the occurrence of normal and complicated grief. For normal grievers, in 35% of the cases the partner had died from cancer, in 32% from diseases of the circulatory system, and in 32% from other causes. For complicated grievers, this was respectively 45%, 35%, and 21%. Out of nine partners of patients who had died from euthanasia, eight experienced complicated grief. Univariate analysis showed that partners of patients who had died from cancer had a higher risk of complicated grief than partners of patients who had died from diseases of the circulatory system (OR=0.76; 95% CI=0.39–1.49) or other diseases (OR=0.47; 95% CI=0.23–0.97). Further it showed that if the deceased died after the use of euthanasia, complicated grief was more likely (OR=8.38; 95% CI=1.02–68.49).

Table 2.

Differences Between Normal Grief and Complicated Grief: Characteristics of Illness and End-of-Life Care, and Nature of Death (N=200)

					Difference between groups, unadjusted ^a		Difference between groups, adjusted for sex and age ^b
			Normal grief (n=100)	Complicated grief (n=100)	Overall p-value	OR (95% CI)	Overall p-value	OR (95% CI)
Cause of death(ICD-10)		Neoplasms	35 (35%)	45 (45%)	0.066	1.00 (ref)	0.042	1.00
		Diseases of the circulatory system	32 (32%)	35 (35%)		0.80 (0.42–1.55)		0.76 (0.39–1.49)
		Other ^c	32 (32%)	21 (21%)		0.51 (0.25–1.03)		0.47 (0.23–0.97)
		Missing	1	1
Number of comorbidities		0	7 (7%)	5 (5%)	0.407	1.00 (ref)	0.438	1.00
		1	27 (27%)	28 (28%)		1.45 (0.41–5.14)		1.51 (0.42–5.39)
		2	18 (18%)	25 (25%)		1.94 (0.53–7.12)		2.02 (0.54–7.59)
		3	14 (14%)	18 (18%)		1.80 (0.47–6.90)		1.87 (0.47–7.47)
		4	13 (13%)	10 (10%)		1.07 (0.26–4.43)		1.22 (0.29–5.16)
		≥5	20 (20%)	14 (14%)		0.98 (0.26–3.73)		1.02 (0.26–4.00)
Place of death		Home	36 (36%)	37 (37%)	0.918	1.00 (ref)	0.918	1.00
		Hospital	42 (42%)	40 (40%)		0.93 (0.49–1.74)		0.91 (0.48–1.75)
		Community living	16 (16%)	16 (16%)		0.97 (0.42–2.23)		1.00 (0.42–2.38)
		Other ^d	3 (3%)	3 (3%)		0.97 (0.18–5.14)		0.91 (0.17–4.86)
		Missing	3	4
Number of hospitalizations last year of life		0	20 (20%)	24 (24%)	0.515	1.00 (ref)	0.975	1.00
		1	26 (26%)	21 (21%)		0.64 (0.28–1.47)		0.64 (0.28–1.48)
		2	16 (16%)	14 (14%)		0.64 (0.27–1.77)		0.69 (0.27–1.77)
		3	10 (10%)	11 (11%)		0.87 (0.31–2.48)		0.88 (0.31–2.50)
		4	5 (5%)	11 (11%)		1.74 (0.52–5.88)		1.75 (0.52–5.96)
		≥5	22 (22%)	19 (19%)		0.68 (0.29–1.62)		0.70 (0.29–1.65)
Duration of hospitalizations last year of life, median (IQR); n=156			25 (9–49)	26 (14–50)	0.827	1.01 (0.99–1.01)	0.950	1.00 (0.99–1.01)
Duration of illness ^e		0–6 months	40 (40%)	42 (42%)	0.264	1.00 (ref)	0.247	1.00
		>6 months–1 year	23 (23%)	27 (27%)		1.12 (0.55–2.26)		1.14 (0.56–2.31)
		>1–4 years	17 (17%)	22 (22%)		1.23 (0.57–2.65)		1.21 (0.56–2.61)
		>4 years	17 (17%)	22 (22%)		0.45 (0.18–1.11)		0.44 (0.18–1.10)
		Missing	2	0
Sudden death ^e		Completely unexpected	20 (20%)	14 (14%)	0.164	1.00 (ref)	0.155	1.00
		Deceased ill, but death unexpected	29 (29%)	26 (26%)		1.28 (0.54–3.04)		1.34 (0.56–3.19)
		Expected	51 (51%)	60 (60%)		1.68 (0.77–3.66)		1.73 (0.79–3.80)
Number of transitions last 3 months ^e,f		0	26 (26%)	25 (25%)	0.243	1.00 (ref)	0.222	1.00
		1	35 (35%)	30 (30%)		0.93 (0.45–1.92)		0.91 (0.43–1.91)
		2	21 (21%)	19 (19%)		0.98 (0.43–2.23)		0.96 (0.42–2.21)
		≥3	17 (17%)	26 (26%)		1.65 (0.73–3.74)		1.71 (0.74–3.93)
End-of-life decision making	Withholding or withdrawal of potential life-prolonging treatments	No/not in file	76 (76%)	81 (81%)	0.390	1.00	0.367	1.00
		Yes	24 (24%)	19 (19%)		0.74 (0.38–1.46)		0.73 (0.37–1.45)
	Euthanasia	No/not in file	99 (99%)	92 (92%)	0.044^*	1.00	0.047^*	1.00
		Yes	1 (1%)	8 (8%)		8.61 (1.06–70.17)		8.38 (1.02–68.49)
	Other ^g	No/not in file	84 (84%)	90 (90%)	0.211	1.00	0.179	1.00
		Yes	16 (16%)	10 (10%)		0.58 (0.25–1.36)		0.56 (0.24–1.31)

Difference test is based on univariate logistic regression. Reference group is normal grief.

Difference test is based on multivariate logistic regression. Reference group is normal grief.

Other causes of death: diseases of the respiratory system; diseases of the nervous systems, mental and behavioral disorders; certain infectious and parasitic diseases; diseases of the digestive system; diseases of the genitourinary system; symptoms, signs, and abnormal clinical and laboratory findings not elsewhere classified; injury, poisoning, and certain other consequences of external causes; external causes of morbidity and mortality.

Other=e.g., outside on the street, family's home.

Imputed results.

First transition mainly from home to hospital; second transition mainly from hospital to home.

Palliative sedation, intensified alleviation of symptoms, do not resuscitate.

ICD-10, International Classification of Diseases and Related Health Problems; IQR, interquartile range.

Following the univariate analyses (see Tables 1 and 2), multivariate logistic regression was performed (for all variables with p<.02) (no table presented). Included in the model were the underlying cause of death; the nature of death; end-of-life decisions (including euthanasia); sex and the age at death of the deceased person; depression (CESD); and cognitive status (MMSE) at baseline. Only depression was significantly associated with type of grief. Bereaved partners with depression at baseline had a higher risk of complicated grief compared to bereaved partners without depression (OR=3.48; 95% CI=1.40–8.68).

Discussion

In the present study we found that complicated grief in older adults is significantly associated with depression, and not with characteristics of the patient's illness, end-of-life care, and the nature of death.

Multivariate analysis showed that only depression at baseline was significantly associated with an increased risk of complicated grief. Apparently, complicated grief in older adults cannot be explained by circumstances surrounding the patient's death (situational factors), but predominantly by factors related to the bereaved person itself (personal factors). Depression has been closely associated with grief in the literature. While studies have focused on depression as an outcome of grief,²⁹ or as a syndrome following a spousal death,³⁰ there are few studies that examined the etiologic relevance of depression for the onset of complicated grief.^31–33 Horowitz et al., who studied 70 individuals who had experienced the death of a spouse when they were between the ages of 21 and 55 years, found that those bereaved with a history of major depressive disorder were more vulnerable to complicated bereavement.³¹ This may be explained by a lack of ability to cope with loss.³⁴ Apparently, this especially holds true for older bereaved partners.

Social support is important in the bereavement period, because it protects against physical and psychological illness and helps to maintain quality of life.³⁵ A study performed among older adults who experienced complicated grief showed that they often rely on available interpersonal support to help them manage their grief, but that such support is not always experienced as sufficient.³⁶ Professional support may therefore be especially important for older persons seeking bereavement support.^35,36 Health care professionals who care for terminal patients and their partners should pay particular attention to partners with pre-loss depression.¹¹ Potentially, physicians could be trained to perform evidence-based assessments for depression and to link those who could benefit to bereavement or mental health specialists.³⁷

In the present study, eight out of nine relatives of patients who had died from euthanasia experienced complicated grief. A previous Dutch study showed that the bereaved family of patients with cancer who died by euthanasia coped better with respect to grief symptoms and posttraumatic stress reactions than the bereaved of patients with cancer who died a natural death.^14,38 Our results should be interpreted with caution due to the small numbers, but the difference with the previous study may be explained by the fact that patients receiving euthanasia in our study may have had a relatively difficult and protracted dying process, whereas in the previous study, patients who received and did not receive euthanasia died from comparable disorders. Van den Boom,³⁹ who described the consequences of euthanasia on grief among the bereaved family and friends, previously found that a complicated euthanasia process was associated with complicated grief and added distress to the bereaved family and friends.

No associations were found between sociodemographic characteristics of the bereaved and the type of grief they experience. While demographic factors are consistently identified as relating to bereavement outcomes, it is likely that they are of little importance in determining an individual's specific risk of complicated bereavement outcomes.¹⁰ Demographic factors such as age, gender, and socioeconomic status may affect health independently of bereavement.^10,40 Age, for instance, may be more of an indicator of differences in grieving style than a specific indicator of risk.⁴¹ Also the (younger) age of the deceased is often cited as a risk factor for complicated bereavement in surviving relatives.¹⁰ However, this counts particularly in relation to the death of child.¹⁰ As the current study focuses on spousal loss, this could potentially explain the lack of a significant association.

Our study has several strengths. First, it was conducted within a population-based setting.¹ Second, a large sample was employed, which enhances the generalizability of the findings.¹ However, the study also has some limitations. First, grief was dichotomized. Dichotomization of a continuous outcome variable may lead to a loss of power. However, complicated grief is designated as a disorder with distinct characteristics and adverse outcomes. Unfortunately, no information was found in the medical files on relatives' involvement in the patient's care. Intensity of care provided has been found to be a risk for complicated grief.¹² Future studies should assess whether characteristics of the caregiving experience result in distinctive risk factors for developing complicated grief in older adults.¹² This also holds true for other bereaved related characteristics, such as a history of previous losses and high pre-death distress, or factors concerning interpersonal relationships, such as the availability of social support and the level of family functioning.¹⁰ Social support has been shown to be an important protective factor against the negative effects of complicated grief.^42–45 Finally, other mental health factors than depression were not taken into account but may have been of influence, such as anxiety and panic disorders.

In conclusion, our results suggest that complicated grief in older adults is not related to the circumstances of dying of the deceased spouse. Preexisting conditions such as depression seem to be more important in explaining the occurrence of complicated grief.

Footnotes

Acknowledgments

We would like to thank everyone involved in the data collection process of the Rotterdam Study and the inhabitants of Ommoord for their time and effort.

Author Disclosure Statement

No competing financial interests exist. This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Appendix 1.

Characteristics of Deceased and Bereaved Partners (n=200)

Deceased	N (%)
Sex
Male	132 (66%)
Female	68 (34%)
Age at death (mean±SD)	74.48±7.15
Ethnicity
Caucasian	192 (99%)
Non-Caucasian	1 (1%)
Missing	7
Working status (baseline)
Working	30 (15%)
Unemployed or retired	167 (85%)
Missing	3
Education^a (baseline)
Low	100 (50%)
Intermediate	78 (40%)
High	19 (10%)
Missing	3
Bereaved partners
Sex
Male	69 (34%)
Female	131 (66%)
Age at death deceased (mean±SD)	72.96±6.60
Ethnicity
Caucasian	194 (100%)
Non-Caucasian	0
Missing	6
Working status (baseline)
Working	21 (11%)
Unemployed or retired	179 (89%)
Education^a (baseline)
Low	125 (62%)
Intermediate	62 (31%)
High	13 (17%)
Depression (CESD) (baseline) (n=181)
No depression	147 (81%)
Depression	34 (19%)
Missing
ADL^b (baseline) (n=192)	24.26±3.22
Cognitive status^c (MMSE) At baseline (n=194) (mean±SD)	28.08±1.43
Time since bereavement^d (in months) (median (IQR)	34 (14–68)

Highest education attained (completed or not completed): Low=primary education; primary education, plus higher not completed education; lower vocational education; lower secondary education. Intermediate=intermediate vocational education; general secondary education. High=higher vocational education; university.

Total score with a minimum of 0 (much difficulties with ADL) and 27 (little difficulties with ADL).

Any score greater than or equal to 27 points (out of 30) indicates a normal cognition. Below this, scores can indicate severe (≤9 points), moderate (10–18 points), or mild (19–24 points) cognitive impairment.

Time between the death of deceased and the interview where type of grief was assessed. ADL, activities of daily living; CESD, Centre for Epidemiological Studies Depression scale; MMSE, Mini Mental State Examination.

References

Newson

, Boelen

, Hek

, et al.: The prevalence and characteristics of complicated grief in older adults. J Affect Dis, 2011; 132:231–238.

Workman

: Prolonged grief disorder: A problem for the past, the present, and the future. PLoS Med, 2009; 6:e100–e122.

Boelen

, van den Bout

: Complicated grief, depression, and anxiety as distinct postloss syndromes: A confirmatory factor analysis study. Am J Psychiatry, 2005; 162:2175–2177.

Prigerson

, Bierhals

, Kasl

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

Prigerson

, Frank

, Kasl

, et al.: Complicated grief and bereavement-related depression as distinct disorders: Preliminary empirical validation in elderly bereaved spouses. Am J Psychiatry, 1995; 152:22–30.

Szanto

, Prigerson

, Houck

, et al.: Suicidal ideation in elderly bereaved: The role of complicated grief. Suicide Life Threat Behav, 1997; 27:194–207.

Ott

, Lueger

, Kelber

, et al.: Spousal bereavement in older adults: Common, resilient, and chronic grief with defining characteristics. J Nerv Ment Dis, 2007; 195:332–341.

Supiano

, Luptak

: Complicated grief in older adults: A randomized controlled trial of complicated grief group therapy. Gerontologist, 2014; 54:840–856.

Parkes

: Hospice heritage: Introduction. Omega (Westport), 2007; 56:1–5.

10.

Aranda

, Milne

: Guidelines for the Assessment of Complicated Bereavement Risk in Family Members of People Receiving Palliative Care. Melbourne: Centre for Palliative Care,. 2000.

11.

Chiu

, Huang

, Yin

, et al.: Determinants of complicated grief in caregivers who cared for terminal cancer patients. Support Care Cancer, 2010; 18:1321–1327.

12.

Ghesquiere

, Haidar

, Shear

: Risks for complicated grief in family caregivers. J Soc Work End Life Palliat Care, 2011; 7:216–240.

13.

Bernard

, Guarnaccia

: Husband and adult-daughter caregivers' bereavement. J Death Dying, 2002; 45:153–166.

14.

Jacobs

, Mazure

, Prigerson

: Diagnostic criteria for traumatic grief. Death Stud, 2000; 24:185–199.

15.

Abarshi

, Echteld

, Van den Block

, et al.: Transitions between care settings at the end of life in the Netherlands: Results from a nationwide study. Palliat Med, 2010; 24:166–174.

16.

Hofman

, Breteler

, van Duijn

, et al.: The Rotterdam Study: 2010 objectives and design update. Eur J Epidemiol, 2009; 24:553–572.

17.

Hofman

, van Duijn

, Franco

, et al.: The Rotterdam Study: 2012 objectives and design update. Eur J Epidemiol, 2011; 26:657–686.

18.

Federal Interagency Forum on Aging-Related Statistics: Older Americans 2012: Key Indicators of Well-Being. Washington, DC: Federal Interagency Forum on Aging-Related Statistics, 2012.

19.

Ernster

: Nested case-control studies. Preventive Med, 1994; 23:587–590.

20.

Prigerson

, Horowitz

, Jacobs

, et al.: Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Med, 2009; 6:e100–e121.

21.

Leening

, Kavousi

, Heeringa

, et al.: Methods of data collection and definitions of cardiac outcomes in the Rotterdam Study. Eur J Epidemiol, 2012; 27:173–185.

22.

Dutch Central Bureau of Statistics: Standaard onderwijs indeling 1978, codenlijst van opleidingen, alfabetisch gerangschrikt. Voorburg: Dutch Central Bureau of Statistics, 1989.

23.

Folstein

, Folstein

, McHugh

: “Mini-mental state:” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res, 1975; 12:189–198.

24.

Fries

, Spitz

, Young

: The dimensions of health outcomes: The health assessment questionnaire, disability and pain scales. J Rheumatol, 1982; 9:789–793.

25.

Lawton

, Brody

: Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist, 1969; 9:179–186.

26.

Radloff

: The CESD scale: A self-report depression scale for research in the general population. Applied Psychological Measurement., 1977; 1:385–401.

27.

World Health Organization:. International Statistical Classification of Diseases and Health Related Health Problems, 10th revision (ICD-10). Geneva: WHO, 1992.

28.

Donders

, van der Heijden

, Stijnen

, et al.: Review: A gentle introduction to imputation of missing values. J Clin Epidemiol, 2006; 59:1087–1091.

29.

Parkes

: Bereavement: Studies of Grief in Adult Life, 3rd ed. Madison, WI: International Universities Press, 1998.

30.

Zisook

, Shuchter

: Uncomplicated bereavement. J Clin Psychiatry, 1993; 54:365–372.

31.

Horowitz

, Siegel

, Holen

, et al.: Diagnostic criteria for complicated grief disorder. Am J Psychiatry, 1997; 154:904–910.

32.

Lazare

: Unresolved Grief: Outpatient Psychiatry: Diagnosis and Bereavement. Baltimore, MD: Wiliams and Wilkins, 1979, pp. 498–512.

33.

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; 16:745–751.

34.

Hirsch

, Allison

, Shaw

; Harvard Medical School: Coping with Grief and Loss: A Guide to Healing. A Harvard Medical School Special Health Report. Boston: Harvard Medical School, 2010.

35.

Benkel

, Wijk

, Molander

: Family and friends provide most social support for the bereaved. Palliat Med, 2009; 23:141–149.

36.

Ghesquiere

: “I was just trying to stick it out until I realized that I couldn't:” A phenomenological investigation of support seeking among older adults with complicated grief. Omega (Westport), 2013; 68:1–22.

37.

Ghesquiere

, Patel

, Kaplan

, et al.: Primary care providers' bereavement care practices: Recommendations for research directions. Int J Geriatr Psychiatry, 2014; 29:1221–1229.

38.

Swarte

, van der Lee

, van der Bom

, et al.: Effects of euthanasia on the bereaved family and friends: A cross sectional study. BMJ, 2003; 327:189.

39.

Van den Boom

: AIDS, euthanasia and grief. AIDS Care, 1995; 7:S175–S185.

40.

Sanders

: Risk factors in bereavement outcome. In: Stroebe

, Stroebe

, Hansson

(eds): Handbook of Bereavement: Theory, Research, and Intervention. Cambridge: Cambridge University Press, 1993.

41.

Sanders

: Comparison of younger and older spouses in bereavement outcome. J Death Dying, 1980–1981; 11:217–232.

42.

Burke

, Neimeyer

, McDevitt-Murphy

: African American homicide bereavement: Aspects of social support that predict complicated grief, PTSD, and depression. Omega (Westport), 2010; 61:1–24.

43.

Macias

, Jones

, Harvey

, et al.: Bereavement in the context of serious mental illness. Psychiatr Serv, 2004; 55:421–426.

44.

Ogrodniczuk

, Joyce

, Piper

: Changes in perceived social support after group therapy for complicated grief. J Nerv Ment Dis, 2003; 191:524–530.

45.

Nam

, Hyun

: Effects of satisfaction with social support on complicated grief in bereaved dementia caregivers. Korean J Health Promot, 2014; 14:43–49.

Risk Factors for Complicated Grief in Older Adults

Abstract

Abstract

Background:

Objective:

Methods:

Results:

Conclusions:

Introduction

Methods

The Rotterdam Study 1

Nested case control

Case definition

Assessment of determinants

Personal characteristics at baseline

Patient's illness, end-of-life care, and the nature of death

Data analysis

Results

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

The Rotterdam Study¹