Full text of DOI:10.1093/geronb/56.4.S237

Journal of Gerontology: SOCIAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2
001, Vol. 56B, No. 4, S237–S248
Psychological Adjustment to Sudden and Anticipated
Spousal Loss Among Older Widowed Persons
1
1
2
1
3
Deborah Carr, James S. House, Camille Wortman, Randolph Nesse, and Ronald C. Kessler
1Institute for Social Research, University of Michigan, Ann Arbor.
2Department of Psychology, State University of New York, Stony Brook.
3Department of Health Care Policy, Harvard Medical School, Cambridge, Massachusetts.
Objectives. This study examined if older adults’ psychological adjustment to widowhood varies based on whether
the death was sudden or anticipated and if these effects are mediated by death context characteristics (e.g., predeath care-
giving, nursing home use, spouse’s age at death, and couple’s communication about the death).
Methods. The effects of forewarning on multiple indicators of mental health and grief were examined in a sample of
10 widowed persons who participated in the Changing Lives of Older Couples (CLOC) study. The CLOC is a proba-
2
bility sample of 1,532 married individuals aged 65 and older for whom baseline information was collected in 1987–88,
with widowed persons reinterviewed 6, 18, and 48 months after spousal loss.
Results. Forewarning did not affect depression, anger, shock, or overall grief 6 or 18 months after the loss. Pro-
longed forewarning was associated with elevated anxiety both 6 and 18 months after the death. Sudden spousal death el-
evated survivors’ intrusive thoughts at the 6-month follow-up only. Sudden death was associated with slightly higher
levels of yearning among women but significantly lower yearning among men both 6 and 18 months after the loss.
Discussion. The findings call into question the widespread belief that grief is more severe if death is sudden and sug-
gest a more complex relationship between bereavement and circumstances of spousal death.
IDOWHOOD is among the most stressful of life
events (Holmes & Rahe, 1967). The psychological
consequences of widowhood vary widely, however, based
on characteristics of the survivor (Matthews, 1991; Stroebe
first suggested in Lindemann’s (1944) classic article, “Symp-
tomatology and Management of Acute Grief.” Spouses of
the men serving in World War II experienced grief-like
symptoms (or anticipatory grief) before their spouses actu-
ally died, but they managed to emotionally disengage in an-
ticipation of their spouses’ deaths, and thus did not appear
to be highly grief-stricken upon the actual death. In contrast,
the relatives of young adults killed suddenly in a nightclub
fire suffered severe grief symptoms (Lindemann, 1944).
Building upon this work, many grief scholars have con-
cluded that individuals who anticipate their spouse’s death
will use the forewarning period to prepare psychologically
and practically for the transition to widowhood. Although
spouses may exhibit grief-like symptoms during the pre-
death period, their postloss adjustment is believed to be bet-
ter than those who experienced a sudden loss (Gerber, 1974,
p. 27; Rando, 1986; Vachon et al., 1982).
Empirical studies examining the effect of death forewarn-
ing on widowed persons’ psychological adjustment are in-
conclusive, however. Many studies have suggested that sud-
den spousal death is associated with poorer psychological
adjustment among widowed persons (Ball, 1977; Carey,
1979–80; Farberow, Gallagher-Thompson, Gilewski, &
Thompson, 1992; Glick, Weiss, & Parkes, 1974; Hill,
Thompson, & Gallagher, 1988; Jacobs, Kasl, & Ostfeld,
1986; Lundin, 1984; O’Bryant, 1990–91; Smith, 1978; Va-
chon et al., 1982; Wells & Kendig, 1997; Willis, Thomas,
Garry, & Goodwin, 1987; Zisook, Schuchter, & Lyons, 1987).
A smaller group of studies has found the reverse—that fore-
warning is linked to poorer adjustment among widowed
W
&
Stroebe, 1983; Umberson, Wortman, & Kessler, 1992),
the deceased (Parkes, 1985; Reed, 1998), and the marital re-
lationship (Carr et al., 2000). The context of the loss—includ-
ing whether the death was sudden or the endpoint of a long
and lingering illness—is also an important influence on pat-
terns of grief (Ball, 1977; Carey, 1979–80; Carnelley, Wort-
man, & Kessler, 1999; Lundin, 1984; O’Bryant, 1990–91;
Rando, 1986; Smith, 1978; Vachon et al., 1982). Numerous
studies have explored whether sudden or anticipated deaths
are more distressing for bereaved spouses, but this research
is inconclusive and does not specifically address the unique
circumstances of elderly bereaved persons. Understanding
the linkage between death forewarning and older widowed
persons’ well-being is critically important today because
chronic illnesses, or ongoing conditions for which there is no
cure, account for the majority of older adults’ deaths (McLe-
roy & Crump, 1994; Olshansky & Ault, 1986). Furthermore,
advances in medical technology that delay late-life mortality
mean that the interval between diagnosis and death is length-
ening. It is thus critically important that scholars and practi-
tioners understand how this “living-dying interval” affects
survivors’ adjustment (Pattison, 1977, 1978).
Theoretical Issues
The proposition that sudden deaths are more difficult for
family members to cope with than anticipated deaths was
S237

S238
CARR ET AL.
persons (Fengler & Goodrich, 1979; Gerber, Rusalem, Han-
non, Battin, & Arkin, 1975; Sanders, 1982–83; Schwab,
Chalmers, Conroy, Farris, & Markush, 1975). A third group
of studies has found no relationship between death fore-
warning and survivors’ psychological adjustment (Bowling &
Cartwright, 1982; Clayton, Halikas, Maurice, & Robins,
owhood as they observe their peers experience the loss of
spouse (Fooken, 1985). Consequently, sudden death may have
a more deleterious effect on the well-being of men than
women, because older women are more likely to anticipate
the death of their spouse (Hill et al., 1988).
1
973; Hill et al., 1988; Maddison & Walker, 1967; Mc-
Death Context and Forewarning
Gloshen & O’Bryant, 1988; Roach & Kitson, 1989; Sand-
ers, 1982–83).
We believe that there are four reasons for these inconclu-
sive findings. First, few studies have explicitly acknowl-
edged that the age of the deceased may condition the effects
of forewarning on widowed persons’ well-being. Second,
the larger context of the death has rarely been considered
A second reason for the inconclusiveness of past studies
is that empirical analyses have rarely controlled important
contextual factors related to the death, which may counter-
act or suppress the effects of death timing. For instance, al-
though prolonged forewarning periods may allow spouses
to prepare psychologically and practically for the loss (Ka-
lish, 1981; Rando, 1986), the warning period may also be ac-
companied by difficult caregiving duties (Anashensel, Pear-
lin, Mullan, Zarit, & Whitlach, 1995; George & Gwyther,
1986; Norris & Murrell, 1987; Wells & Kendig, 1997), emo-
tional isolation from family members and friends (Kramer,
1996–97), and the neglect of their own health (Fengler &
Goodrich, 1979; Rosenblatt, 1983; Sanders, 1982–83; Sie-
gel & Weinstein, 1983; Sweeting & Gilhooly, 1990). Stress
related to the “extended death watch” period (Gerber et al.,
1975) may be particularly difficult for older adults because
of their age and accompanying health problems (Averill
& Wisocki, 1981; Fengler & Goodrich, 1979; George &
Gwyther, 1986).
(
Sweeting & Gilhooly, 1990); stressors associated with
death forewarning, such as the strain of caregiving, may can-
cel out the gains presumably provided by the period of pre-
death anticipation (Rando, 1986; Siegel & Weinstein, 1983).
Third, a diverse set of outcome measures and inconsistent
definitions of sudden versus anticipated death have been
used, thus limiting the comparability of findings across stud-
ies. Finally, most past research has relied on cross-sectional
rather than longitudinal data; thus, baseline demographic and
socioeconomic characteristics that may affect both death
timing and psychological adjustment have been neglected.
Death Forewarning Among Older Adults
Contextual factors associated with death forewarning also
might be associated with better spousal adjustment (O’Bry-
ant, 1990–91). Caregiving may increase closeness with an ill
relative (Hinrichsen, Hernandez, & Pollack, 1992; Vachon et
al., 1982; Wilson, 1990) and may provide a lasting sense of
purpose to the caregiver (O’Bryant, Straw, & Meddaugh,
1990; Wright, 1991). Moreover, older individuals suffering
prolonged illnesses are more likely to reside in nursing homes
(Foner, 1994), and thus their spouses may make a smoother
transition to widowhood. The institutionalization process has
been characterized as “quasi-widowhood,” because spouses
have considerably reduced daily contact and communication,
may emotionally separate from one another, and are spared
the daily stresses of direct caregiving (DeSpelder & Strick-
land, 1992; Ross, Rosenthal, & Dawson, 1993). Thus, our
analyses included three important indicators of death context
that may mediate (or counteract and hence suppress) the ef-
fect of death timing on widowed persons’ adjustment: age of
spouse at death, whether the deceased was living in a nursing
home prior to death, and whether the survivor provided care
prior to their spouse’s death.
Knowledge about the impending death also may affect
the couple’s interactions during the ill spouse’s last days.
Couples who anticipate a death may use the forewarning pe-
riod to make practical plans for the survivor’s economic and
social adjustment, thus enabling a smoother transition to
widowhood (Rando, 1986). Moreover, couples who use the
time between diagnosis and death to resolve emotional “un-
finished business” (Blauner, 1966) may find their relation-
ships strengthened in the final days. Bereaved persons who
were by their ailing spouse’s side at the moment they died
have been found to cope better with the loss than those who
did not reach such closure (Bennett & Vidal-Hall, 2000;
Fiewiger & Smilowitz, 1984–85; Hinds, 1985). We know of
Surprisingly few studies have examined the effects of
sudden versus anticipated death on older widowed persons’
well-being. Rather, most research has focused on premature
or “off-time” widowhood among young and midlife adults
(
1
Ball, 1977; Glick et al., 1974; Lindemann, 1944; Lundin,
984; Parkes & Weiss, 1983; Sanders, 1982–83; Vachon
et al., 1976). Other studies have focused solely on sudden
deaths occurring under particularly horrific circumstances
such as murders (Rynearson, 1984) or suicides (Calhoun &
Allen, 1991; Dunn & Morrish-Vidners, 1987; Reed, 1998;
van der Wal, 1989–90). Because deaths of young persons
and violent deaths are considered the most difficult to cope
with (Archer, 1999; Parkes & Brown, 1972; Reed, 1998;
Roach & Kitson, 1989; Smith, 1978; Vachon et al., 1976), it
is not clear whether the deleterious effects of such losses are
due to suddenness, life stage, or a combination thereof (i.e.,
sudden and premature death).
For older adults (i.e., aged 65 and older), even “sudden”
deaths may be anticipated and viewed as timely (Neugarten &
Hagestad, 1976). Given that predictable life transitions are
presumed to be less stressful than unexpected ones (George,
1
1
993; Pearlin, 1982; Pearlin & Lieberman, 1979; Thoits,
983), a sudden spousal death may be no more difficult for
the elderly survivor than death following a long period of
illness (DeSpelder & Strickland, 1992; Hyman, 1983).
Moreover, sudden spousal deaths are more likely to be an-
ticipated by older women than older men. Because of men’s
mortality disadvantage, women aged 65 and older outnum-
ber their male peers by roughly 1.5 to 1 (U.S. Bureau of the
Census, 1996). Nearly half of American women aged 65
and older are widowed, compared with 14% of men. Noting
the gender gap in mortality, Neugarten and Hagestad (1976)
observed that women may go through a rehearsal for wid-

DEATH FOREWARNING AND GRIEF IN WIDOWED PERSONS
S239
no studies that have empirically assessed whether these be-
haviors mediate the relationship between death forewarning
and psychological adjustment. Thus, our analyses included
indicators of whether the couple talked about how the survi-
vor would cope with being alone and whether the survivor
was with his or her spouse at the moment he or she died.
text of spousal death, as well as adjustment to the loss, hence
creating a spurious relationship between them. Because we
had baseline data prior to actual widowhood, we could con-
trol for important baseline (or prewidowhood) characteris-
tics. First, we controlled baseline mental health, to help dis-
tinguish the spouse’s affective state prior to the death and
change in affective state that occurred following the death
(Jacobs, 1993; Zisook & Schuchter, 1991). Second, we con-
trolled baseline demographic characteristics (age and sex)
and socioeconomic status (education, income, and home
ownership), because these characteristics have been linked
to both death context (e.g., younger persons and women are
more likely to have their spouses die suddenly) and later
psychological adjustment. Finally, respondent’s physical
health at baseline was controlled, because it may influence
both how the spouse manages stressors related to the death
timing and psychological adjustment following the death
(George & Gwyther, 1986).
In sum, our research had three objectives: first, to identify
the relationship between death forewarning and widowed
persons’ psychological adjustment 6 and 18 months after the
loss, adjusting for psychological adjustment and other charac-
teristics prior to widowhood; second, to assess whether and
how the relationship between death forewarning and wid-
owed persons’ well-being differs for men and women; and
third, to evaluate the extent to which the effect of death fore-
warning is mediated or suppressed by death context charac-
teristics. Analyses were based on data from the Changing
Lives of Older Couples (CLOC) survey, a prospective study
of a sample of married individuals aged 65 and older.
Measuring Death Forewarning and Adaptation to Loss
The discrepant results in past studies may reflect differ-
ences in the conceptualization and measurement of both in-
dependent and dependent variables (Sweeting & Gilhooly,
1
990). The operationalization of sudden versus anticipated
death has varied across studies. Sudden deaths have in-
cluded those where the deceased was dead on arrival at the
hospital (Carey, 1979–80), as well as those where the survi-
vor had less than 2 hr (Lundin, 1984), 1 week (Sanders,
1
982–83) or 2 weeks (Bowling & Cartwright, 1982) of
warning. As noted earlier, still other studies have consid-
ered sudden deaths to be synonymous with violent deaths
such as suicides, murders, or accidents (Calhoun & Allen,
1
991; Dunn & Morrish-Vidners, 1987; Rynearson, 1984;
van der Wal, 1989–90). Definitions of prolonged forewarn-
ing also vary, from 1 month (Bornstein, Clayton, Halikas,
Maurice, & Robins, 1973; Clayton et al., 1973) or 2 months
(
Vachon et al., 1982; Gerber et al., 1975) up to 6 months or
longer (Clayton et al., 1973).
The dependent variables used across studies also are di-
verse, ranging from self-report of medical symptoms (Ger-
ber et al., 1975) to depressive symptomatology (Clayton et al.,
1
973) and social isolation (Parkes & Weiss, 1983). Several
studies have focused on specific symptoms, such as anger,
guilt, and rumination (Glick et al., 1974; Sanders, 1982–83),
yet most have focused on global indicators of mental health
such as depression or broad grief scales (Carey, 1979–80;
Clayton, Desmarais, & Winokur, 1968; Fulton & Fulton,
METHODS
Sample
The CLOC study is a prospective study of a two-stage
area probability sample of 1,532 married individuals from
the Detroit Standardized Metropolitan Statistical Area. To be
eligible for the study, respondents had to be English-speak-
ing members of a married couple where the husband was
aged 65 or older. All sample members were noninstitution-
alized and were capable of participating in a 2-hr-long inter-
view. The original researchers oversampled women in an
effort to maximize the number of respondents who would
experience bereavement during the study period. Baseline
face-to-face interviews were conducted from June 1987
through April 1988. The response rate for the baseline inter-
view was 68%, which is consistent with the response rate
from other Detroit area studies in that period.
The CLOC researchers monitored spousal loss using
monthly death record tapes provided by the State of Michi-
gan and by reading the daily obituaries in Detroit-area
newspapers. Researchers used the National Death Index and
direct ascertainment of death certificates to confirm deaths
and obtain causes of death. Of the 335 respondents known
to have lost a spouse during the study, 316 were contacted
for possible interview (19 persons, or 6%, had died during
the interim). Of the 316 contacted, 263 persons (83%) par-
ticipated in at least one of the three follow-up interviews,
which were conducted 6 months (Wave 1), 18 months
(Wave 2), and 48 months (Wave 3) after the spouse’s death.
1
971). Consequently, findings purported to characterize the
general link between forewarning and adjustment instead
may be specific to particular outcomes.
We believe that psychological adjustment to loss encom-
passes a complex set of emotional, cognitive, and behav-
ioral reactions. Thus, we explored the effect of death fore-
warning and sudden death on five loss-related (i.e., shock,
anger, yearning, intrusive thoughts, and overall grief) and two
general (i.e., depression and anxiety) dimensions of psycho-
logical adjustment both 6 and 18 months after the death.
Moreover, rather than operationalizing sudden death and
prolonged forewarning a priori, we empirically evaluated
diverse measures of each construct to ascertain the specific
nature of the relationship between death forewarning and
adjustment of older adults. Drawing on past research and
theory, we evaluated three possible relationships: sudden
death effects only, linear effects of warning time, and curvi-
linear effects of warning time. Thus, our analyses evaluated
whether a particular parameterization of forewarning better
predicts widowed persons’ psychological adjustment.
Other Influences on Psychological Adjustment and
Spousal Loss
Finally, our research addressed the possibility that pre-
loss characteristics might affect both the suddenness and con-

S240
CARR ET AL.
Wave 1 analyses were based on the 250 widowed persons
35 men and 215 women) interviewed at the 6-month follow
up, or 79% of the 316 living respondents who lost a spouse.
Wave 2 analyses were based on the 184 widowed persons
trembling; (c) feeling suddenly scared for no reason; (d)
(
feeling fearful; (e) heart pounding or racing; (f) feeling
tense and keyed up; (g) spells of terror and panic; (h) feeling
so restless you could not sit still; (i) feeling that something
bad is going to happen to you; and (j) thoughts and images
of a frightening nature.
(
159 women and 25 men) who were also interviewed at the
8-month follow-up, representing 79% of the 246 persons
1
who were interviewed at Wave 1 and still alive at Wave 2.
We weighted the data to adjust for unequal probabilities of
selection and differential response rate at baseline. The final
weighted analytic sample for Waves 1 and 2 comprised 210
widowed persons (59 men and 151 women) and 155 wid-
owed persons (110 women and 45 men), respectively.
Psychological reactions specific to the loss, as well as an
overarching grief scale, were also considered. The four
components of grief we considered were shock, anger, yearn-
ing, and intrusive thoughts. Shock (ꢁ ꢂ .77) was evaluated
with three questions: In the last month, (a) have you felt as
though you were in a state of shock; (b) have you felt as
though you couldn’t believe what was happening; and (c)
have you felt emotionally numb? Anger (ꢁ ꢂ .68) was as-
sessed with three questions: In the past month, (a) have you
felt resentful or bitter about your spouse’s death; (b) have
you felt that the death of your spouse was unfair; and (c)
have you felt anger toward God? Yearning (ꢁ ꢂ .75) was
assessed with four questions: In the last month, (a) have you
found yourself longing to have your spouse with you; (b)
have you had painful waves of missing your spouse; (c)
have you experienced feelings of intense pain or grief over
the loss of your spouse; and (d) have you experienced feel-
ings of grief, loneliness, or missing your spouse? We evalu-
ated intrusive thoughts (ꢁ ꢂ .66) with three questions: In
the past month, (a) have you had difficulty falling asleep be-
cause thoughts about your spouse kept coming into your
mind; (b) have you tried to block out memories or thoughts
of your spouse; and (c) have you been unable to get
thoughts about your spouse out of your mind?
The issue of selective attrition deserves mention. If persons
who failed to participate in the followup interviews were sig-
nificantly different from those who did participate (in terms
of baseline characteristics) then caution should be taken in
generalizing findings to the larger population of elderly wid-
owed persons. Thus, we estimated logistic regression models
to predict the correlates of widowed persons’ nonparticipa-
tion in the Wave 1 and Wave 2 interviews. For models pre-
dicting Wave 1 nonparticipation, baseline demographic and
socioeconomic characteristics, physical and mental health,
and spouse’s physical health were evaluated as predictors.
Age and baseline anxiety increased risk of nonparticipation,
and home ownership significantly decreased one’s risk of
nonparticipation. For models predicting Wave 2 nonpartici-
pation (given that one participated in the Wave 1 interview),
sudden death; months of forewarning; baseline demographic,
physical, and mental health; and Wave 1 levels of grief were
evaluated as predictors. Not one variable was a significant
predictor (p ꢀ .05) of attrition at Wave 2. Caution should be
taken in generalizing findings to the population at large, be-
cause older, more anxious, and the residentially mobile per-
sons may be underrepresented in the analytic sample.
Grief (ꢁ ꢂ .85) was the average of the four subscale
scores. Response categories for all grief scale items were
“no, never”; “yes, but rarely”; “yes, sometimes”; and “yes,
often.” Items were drawn from widely used grief scales in-
cluding the Bereavement Index (Jacobs, Kasl, & Ostfeld,
1986), Present Feelings About Loss (Singh & Raphael,
Measures
1
981), and Texas Revised Inventory of Grief (Zisook,
Dependent variables.—Two general (i.e., depression
and anxiety) and five loss-related (i.e., shock, anger, yearn-
ing, intrusive thoughts, and overall grief) dimensions of
psychological adjustment at the 6-month and 18-month fol-
low-ups are considered. Depression (ꢁ ꢂ .83) was assessed
with a subset of nine negative items from the 20-item Cen-
ter for Epidemiologic Studies Depression (CES-D) scale
Devaul, & Click, 1982). Each of the dependent variables
was standardized for ease of interpretation and comparison
across indicators and thus had a mean of 0 and a standard
deviation of 1.
Independent variables.—The central independent vari-
able in the analysis was warning time prior to death, evalu-
ated retrospectively at the Wave 1 interview with the ques-
tion, “How long before your spouse’s death did you realize
that s/he was going to die?” Respondents could report the
duration in hours, days, weeks, months, years or “no warn-
ing/minutes.” The number of months of warning time was
used as the independent variable. Months were top-coded at
24, because more than 90% of widowed persons said that
they had fewer than 2 years of warning. Sudden death was a
dummy variable indicating those who had “no warning/
minutes” prior to their spouse’s death (33% of sample). We
did not separately consider the effects of violent sudden
deaths (e.g., suicide or murder), because fewer than 2% died
in this manner. Prolonged forewarning was a dummy vari-
able indicating those who had more than 6 months of warn-
ing prior to their spouse’s death (28% of sample). The refer-
ence group is persons with less than 6 months of warning.
(
Radloff, 1977). In the CES-D, respondents are asked to in-
dicate how often they experienced each symptom in the
week prior to interview. Response categories are “hardly
ever,” “some of the time,” or “most of the time.” The nine
symptoms are (a) I felt depressed; (b) I felt that everything I
did was an effort; (c) My sleep was restless; (d) I felt lonely;
(
e) People were unfriendly; (f) I did not feel like eating; My
appetite was poor; (g) I felt sad; (h) I felt that people dis-
liked me; and (i) I could not “get going.”
Anxiety (ꢁ ꢂ .86) was assessed with 10 items from the
Symptom Checklist 90 Revised (Derogatis & Cleary, 1977).
Respondents are asked to indicate how often they have ex-
perienced each of 10 symptoms in the week prior to inter-
view. Response categories are “not at all,” “a little bit,”
“
moderately,” “quite a bit,” and “extremely.” Symptoms in-
clude being bothered by (a) nervousness or shakiness; (b)

DEATH FOREWARNING AND GRIEF IN WIDOWED PERSONS
S241
Death context.—Because both the psychological conse-
quences of loss and warning time may be associated with
other conditions of the death, we controlled three additional
characteristics of the spouse’s death: (a) spouse’s age at
death (in years), and dichotomous variables indicating (b)
whether the respondent was providing care to his or her
spouse in the 6 months prior to the death, and (c) whether
the spouse was residing in a nursing home prior to death.
Communication about death was evaluated at Wave 1
with the question, “Did you and [your spouse] talk about
how you would deal with being on your own once she or he
was gone?” Closure was evaluated with the question, “Were
you there with your [husband/wife] at the moment when
Table 1. Means and Standard Deviations for Widowed Persons, by
Sex, Changing Lives of Older Couples Study, 1987–93
Women
Men
(n ꢂ 151)
(n ꢂ 59)
Variable
M or %
Dependent Variables
0.055
SD
M or %
SD
Anxiety, 6-month follow up
1.009 0.031
1.03
(
standardized)
Yearning, 6-month follow up
(standardized)
ꢃ.066
1.01
0.171
0.955
1.01
Intrusive thoughts, 6-month follow up
standardized)
Anxiety, 18-month follow up
standardized)
(
0.034
ꢃ.001
ꢃ.066
0.999 ꢃ.087
0.872 ꢃ.303
[
she/he] died?” Positive responses were coded as 1.
(
0.496
0.839
0.992
Yearning, 18-month follow up
(standardized)
Intrusive thoughts, 18-month follow up
Confounding factors.—Depression and anxiety at base-
1.06
0.149
line were measured exactly as described earlier (Derogatis &
Cleary, 1977; Radloff, 1977). Respondent’s physical health
at baseline was assessed with the question “How would you
rate your health at the present time? Would you say it is ex-
cellent, very good, good, fair, or poor?” Responses of “fair”
and “poor” were coded 1, and all others were coded as 0.
(
standardized)
0.005
0.981 ꢃ.033
Independent Variables
Warning Time Indicators
No warning time prior to death
6 or more months warning time
Warning time, months
0.373
0.287
5.09
0.485 0.319
0.454 0.288
0.47
0.457
9.3
8.00
5.95
Death Context
Demographic variables.—Control variables included
age, sex (1 ꢂ female), home ownership at baseline (1 ꢂ owns
home), total household income at baseline (natural log of in-
come), and education (a continuous measure ranging from 3
to 17 or more years of completed schooling). We measured
the total household income variable by having respondents
indicate which of 10 income categories most accurately char-
acterized their economic status. We derived a continuous
measure of income by taking the midpoint of each of the 10
income categories, with Pareto estimation of the mean for the
top income category. The natural log of income was used be-
cause the respondents’ income distribution was skewed, with
most respondents in the lower income categories. A final de-
mographic characteristic—race—was not considered in this
analysis. Preliminary analyses revealed that race (where 1 ꢂ
Black) was not a significant predictor of the outcomes vari-
ables, nor did the effects of forewarning differ by race.
Spouse’s age at death
Spouse lived in nursing home prior
to death
77.1
6.42 72.56
0.123 0.138
0.502 0.436
7.3**
0.347**
0.5
0.015
0.502
Provided care to spouse in 6
months prior to death
Communication About Death
R and spouse discussed how R will
cope with death
R was with spouse when he or she
died
Demographic Characteristics
Age, baseline
0.196
0.397
0.398 0.116
0.491 0.536
0.323
0.503*
69.43
11.33
0.898
20,480 16,308 22,511
1.29
6.99 73.46
2.69 11.11
0.304 0.966
5.92***
3.44
0.182
16,745
0.522
Years of education
Own home, baseline
Income, baseline
Natural log of income
Months between baseline and
Wave 1 interviews
0.523 1.37
36.55
18.15
35.83
19.21
Baseline Well-Being
Finally, we controlled for the duration (in months) between
the baseline and Wave 1 interviews. Although all Wave 1 in-
terviews were conducted 6 months following spousal death,
the duration between the baseline and Wave 1 interviews
ranged from 9 to 76 months because of variation in the timing
of spouse’s death. Thus, baseline assessments were more tem-
porally distant for those who lost their spouses at later dates.
Depression, baseline (standardized) 0.128
1.05
1.09
0.459 0.409
ꢃ.089
ꢃ.205
0.736
0.715*
0.497
Anxiety, baseline (standardized)
Fair or poor health, baseline
0.152
0.299
Notes: t tests were used to assess significant gender differences between
means. Ns are weighted Ns.
*p ꢀ .05; **p ꢀ .01; ***p ꢀ .001.
RESULTS
Men and women did differ in terms of the context sur-
rounding their spouse’s death. Men were significantly more
likely than women to report that their spouse was residing in
a nursing home prior to death (14% vs 2%) and that they
were with their spouse at the exact moment they died (54%
vs 40%). One fifth of women and 12% of men reported that
they had discussed with their spouse how they would cope
with being on their own. Similar proportions of men and
women (44% and 50%, respectively) reported providing
care to their spouse in the months prior to death. Few demo-
graphic characteristics differed by gender, except that men
in the CLOC were significantly older at baseline than the
Sample Characteristics
Descriptive statistics and t tests comparing means for
men and women are presented in Table 1. Men and women
did not differ from one another in terms of anxiety, yearn-
ing, or intrusive thoughts 6 or 18 months following their
loss. Death forewarning did not differ significantly by gen-
der; roughly one third of widowed persons experienced no
forewarning of their spouse’s death, and another one third
reported more than 6 months of warning. The average warn-
ing time was 5–6 months.

S242
CARR ET AL.
women and were also slightly more likely to own their own
homes. At baseline, women had significantly higher levels
of anxiety than men.
and a dichotomous indicator of sudden death; (c) curvilinear
effects of warning time, measured with a continuous indica-
tor of months of warning time, months of warning time
squared, and a dichotomous indicator of sudden death; and
(d) effects of sudden death and very prolonged warning
time, measured with dichotomous indicators representing
sudden death and warning time of more than 6 months;
warning time of less than 6 months was the reference cate-
gory. (In preliminary analyses, more fine-grained cutpoints
were evaluated. However, the model including just two di-
chotomous indicators for no warning versus more than six
months warning best fit the data—as evidenced by adjusted
Influence of Death Forewarning on Mental Health
Our first two objectives of the multivariate analysis were
a) to specify the relationship between death forewarning
(
and widowed persons’ psychological adjustment 6 and 18
months after their loss and (b) to assess whether these pat-
terns differ by gender. To address the first objective, we re-
gressed each of four different measures of forewarning on
two general (i.e., depression and anxiety) and five loss-
related (yearning, shock, anger, intrusive thoughts, and a
composite grief measure) indicators of psychological ad-
justment at both 6 and 18 months after the death. The four
models evaluated were (a) sudden death effects only, mea-
sured with a dichotomous indicator of sudden death; (b) lin-
ear effects of warning time, measured with a continuous in-
dicator of months of warning time (from 1 to 24 months)
2
r values.) To achieve the second objective, all models were
estimated including interaction terms of sex by death fore-
warning. All models controlled demographic characteristics
and respondent’s baseline physical and mental health.
(Complete models are not shown but are available from the
first author.) Models with significant effects of warning
time are presented in Tables 2–4.
Table 2. Ordinary Least Squares Regression Predicting Effects of Forewarning, Death Context, and Communication About Death on
Intrusive Thoughts, Changing Lives of Older Couples Study, 1987–93
Intrusive Thoughts, 6 Months
Intrusive Thoughts, 18 Months
Variable
Model 1
Model 2
Model 3
Model 1
Model 2
Model 3
Warning Time Indicators
No warning time prior to death
.316*
.145)
.390*
(.160)
.377*
(.160)
.285
(.161)
.242
(.179)
.186
(.179)
(
Demographic Characteristics
Sex (1 ꢂ female)
0.057
ꢃ.103
(.201)
ꢃ.005
(.015)
ꢃ.006
(.026)
0.139
(.256)
ꢃ.009
(.148)
ꢃ.163
(.200)
ꢃ.009
(.015)
ꢃ.005
(.026)
0.158
(.253)
ꢃ.041
(.146)
ꢃ.349
(.183)
ꢃ.003
(.012)
ꢃ.013
(.031)
ꢃ.327
(.304)
ꢃ.505**
(.166)
ꢃ.452
(.232)
0.005
(.017)
0.003
(.032)
ꢃ.284
(.306)
ꢃ.471**
(.165)
ꢃ.464*
(.231)
0.002
(.017)
0.002
(.031)
ꢃ.255
(.302)
ꢃ.501**
(.164)
(.162)
Age
0.002
(.011)
Years of education
Own home, baseline
Income (natural log), baseline
ꢃ.016
(.025)
0.034
(.251)
ꢃ.043
(.147)
Baseline Well-Being
Depression, baseline
0.037
0.064
(.085)
.155*
(.080)
0.145
(.161)
0.075
(.084)
.164*
(.079)
0.123
(.159)
0.076
(.095)
0.085
(.097)
ꢃ.295
(.180)
0.132
(.097)
0.092
(.096)
ꢃ.223
(.185)
0.127
(.096)
0.117
(.095)
ꢃ.233
(.183)
(
.083)
.162*
.081)
Anxiety, baseline
(
Fair or poor health, baseline
0.109
(.157)
Death Context
Spouse’s age at death
0.015
0.013
(.015)
ꢃ.320
(.355)
0.251
(.159)
ꢃ.006
(.018)
ꢃ.870*
(.359)
0.047
ꢃ.006
(.018)
ꢃ.834*
(.358)
0.091
(
.015)
ꢃ.437
.356)
0.228
Spouse lived in nursing home prior to death
(
Provided care to spouse in 6 months prior to death
(
.158)
(.176)
(.175)
Communication About Death
R and spouse discussed how R will cope with death
0.08
ꢃ.419*
(.207)
ꢃ.148
(.159)
0.137
1.49
(
.185)
ꢃ.389**
.143)
R was with spouse when he or she died
(
Adjusted R2
Constant
0.02
ꢃ.199
0.031
ꢃ1.09
(1.02)
210
0.057
ꢃ.501
(1.03)
210
0.095
1.36
(1.03)
155
0.117
ꢃ1.09
(1.02)
155
(.919)
(1.16)
155
N
210
Note: Number of months between baseline and Wave 1 interview controlled in all models.
p ꢀ .05; **p ꢀ .01; ***p ꢀ .001.
*

DEATH FOREWARNING AND GRIEF IN WIDOWED PERSONS
S243
Table 3. Ordinary Least Squares Regression Predicting Effects of Forewarning, Death Context, and Communication About Death on
Anxiety, Changing Lives of Older Couples Study, 1987–93
Anxiety, 6 Months
Model 2
Anxiety, 18 Months
Model 2
Variable
Model 1
.003
Model 3
Model 1
Model 3
Warning Time Indicators
No warning time prior to death
ꢃ.002
(.163)
.473**
(.171)
ꢃ.010
(.163)
.522**
(.177)
ꢃ.107
(.139)
.241
ꢃ.116
(.146)
.318*
(.159)
ꢃ.110
(.147)
.336*
(.162)
(.155)
6
or more months warning time
.372*
(.165)
(.152)
Demographic Characteristics
Sex (1 ꢂ female)
ꢃ.040
ꢃ.243
(.189)
ꢃ.018
(.014)
.048*
(.024)
0.417
(.241)
ꢃ.086
(.138)
ꢃ.230
(.190)
ꢃ.019
(.014)
.048*
(.024)
0.443
(.242)
ꢃ.092
(.138)
0.151
(.141)
ꢃ.029***
(.009)
0.004
(.024)
ꢃ.005
(.217)
0.037
ꢃ.058
(.176)
ꢃ.036**
(.012)
0.014
(.024)
0.052
(.223)
0.063
ꢃ.032
(.177)
ꢃ.036**
(.012)
0.015
(.024)
0.034
(.223)
0.066
(
.152)
ꢃ.012
.009)
0.038
.024)
0.384
.237)
ꢃ.127
.139)
Age
(
Years of education
Own home, baseline
Income (natural log), baseline
(
(
(
(.124)
(.124)
(.124)
Baseline Well-Being
Depression, baseline
0.092
.079)
.144ꢄ
0.138
(.081)
.146*
(.075)
.445**
(.150)
0.133
(.081)
.152*
(.076)
.448**
(.151)
ꢃ.083
(.071)
.192**
(.071)
.351**
(.133)
ꢃ.05
ꢃ.046
(.072)
.193**
(.070)
.351*
(.137)
(
(.072)
.200**
(.069)
.350*
(.137)
Anxiety, baseline
(.076)
Fair or poor health, baseline
.435**
(.148)
Death Context
Spouse’s age at death
0.013
0.013
(.014)
ꢃ.941**
(.350)
0.004
0.014
(.013)
ꢃ.680*
(.290)
ꢃ.006
(.131)
0.015
(.013)
ꢃ.733*
(.293)
ꢃ.024
(.133)
(
.014)
ꢃ.912**
.346)
ꢃ.013
.151)
Spouse lived in nursing home prior to death
(
Provided care to spouse in 6 months prior to death
(
(.152)
Communication About Death
R and spouse discussed how R will cope with death
ꢃ.219
ꢃ.046
(.159)
0.166
(.121)
0.156
0.897
(.859)
155
(
.181)
ꢃ.021
.136)
R was with spouse when he or she died
(
Adjusted R2
Constant
0.129
ꢃ.243
0.150
ꢃ.695
(.954)
210
0.148
ꢃ.599
(.980)
210
0.137
1.61
(.749)
155
0.156
1.09
(.845)
155
(.868)
N
210
Note: Number of months between baseline and Wave 1 interview controlled in all models.
*p ꢀ .05; **p ꢀ .01; ***p ꢀ .001.
Three important findings emerged from the regression re-
sults. First, the effect of death forewarning on survivors’
well-being was far weaker than past research has suggested.
Considering only those models that parameterized fore-
warning with two dummy variables, we tested 14 models
18-month follow-up. Even when potential counterbalancing
or suppressor variables (i.e., death context and communica-
tion) were added to the regression equation, warning time
was not a significant predictor of any of the four outcomes.
Sudden death significantly increased intrusive thoughts 6
months after the loss, although the effect was no longer sig-
nificant at the 18-month follow-up (Table 2). Prolonged
forewarning (i.e., more than 6 months of forewarning) was
associated with elevated anxiety both 6 and 18 months after
the loss (Table 3).
Third, gender differences in the effect of forewarning
were significant for just one of the seven outcomes: yearn-
ing. At both the 6- and 18-month follow-ups, sudden death
(compared with anticipated death) was associated with
slightly higher yearning scores among women and consider-
ably lower yearning scores among men at both interviews
(Table 4).
(
i.e., seven dependent variables each at two follow-ups)
with two forewarning indicators in each model. Of these,
only two models produced significant effects—each includ-
ing one of the forewarning dummy variables. These effects
are probably greater than expected by chance but indicate
very limited effects of forewarning on reactions to widow-
hood among older adults.
Second, when death forewarning did significantly affect
psychological adjustment, its effect differed across out-
comes. Death forewarning—regardless of how operational-
ized—was unrelated to four (i.e., depression, grief, shock,
and anger) of the seven possible outcomes at both the 6- and

S244
CARR ET AL.
Table 4. Ordinary Least Squares Regression Predicting Effects of Forewarning, Death Context, and Communication About Death on
Yearning, Changing Lives of Older Couples Study, 1987–93
Yearning, 6 Months
Model 2
Yearning, 18 Months
Model 2
Variable
Model 1
Model 3
Model 1
Model 3
Warning Time Indicators
No warning time prior to death
ꢃ.668*
ꢃ.559*
(.293)
.758*
ꢃ.538
(.297)
.733*
(.327)
ꢃ.524
(.322)
.695
ꢃ.653
(.353)
.884*
(.387)
ꢃ.645
(.359)
.898*
(.393)
(
.274)
.770*
.318)
No Warning Time ꢅ Sex
(
(.323)
(.377)
Demographic Characteristics
Sex (1 ꢂ female)
ꢃ.437*
.190)
0.012
ꢃ.463*
(.227)
0.014
(.014)
0.001
(.025)
.527*
(.249)
0.268
(.144)
ꢃ.468*
(.229)
0.014
(.014)
0.001
(.025)
.527*
(.251)
0.263
(.145)
ꢃ.396
(.237)
0.006
(.012)
ꢃ.046
(.033)
0.294
(.317)
ꢃ.214
(.174)
ꢃ.396
(.282)
0.029
(.018)
ꢃ.031
(.033)
0.295
(.319)
ꢃ.185
(.173)
ꢃ.391
(.283)
0.031
(.018)
ꢃ.031
(.033)
0.284
(.321)
ꢃ.174
(.174)
(
Age
(.010)
Years of education
Own home, baseline
Income (natural log), baseline
ꢃ.009
(.025)
0.447
(.243)
0.243
(.143)
Baseline Well-Being
Depression, baseline
0.039
0.046
(.083)
0.099
(.078)
0.135
(.156)
0.048
(.083)
0.101
(.079)
0.131
(.156)
0.062
(.099)
0.008
(.101)
ꢃ.201
(.188)
0.107
(.101)
0.003
(.010)
ꢃ.087
(.193)
0.109
(.102)
ꢃ.007
(.101)
ꢃ.084
(.194)
(.080)
Anxiety, baseline
0.107
(.079)
Fair or poor health, baseline
0.079
(.152)
Death Context
Spouse’s age at death
0.001
0.001
(.014)
ꢃ.180
(.356)
0.276
(.157)
ꢃ.026
(.019)
ꢃ.727
(.385)
0.14
ꢃ.026
(.019)
ꢃ.752*
(.391)
0.121
(
.014)
ꢃ.209
.351)
0.273
Spouse lived in nursing home prior to death
(
Provided care to spouse in 6 months prior to death
(
.155)
(.185)
(.188)
Communication About Death
R and spouse discussed how R will cope with death
0.048
0.133
(.220)
0.087
(.169)
0.032
ꢃ.008
(1.25)
155
(
.182)
ꢃ.076
.143)
R was with spouse when he or she died
(
Adjusted R2
Constant
0.082
ꢃ1.49
0.086
ꢃ2.13
(1.00)
210
0.078
ꢃ2.04
(1.03)
210
0.016
0.099
(1.09)
155
0.041
0.175
(1.22)
155
(.898)
N
210
Note: Number of months between baseline and Wave 1 interview controlled in all models.
*p ꢀ .05; **p ꢀ .01; ***p ꢀ .001.
Death Context and Communication as Mediators of
Death Forewarning Effects
1), sudden death was associated with a .316 standard devia-
tion increase in intrusive thoughts (p ꢀ .05). When death
context variables were adjusted, the effect of sudden death
increased to .39 standard deviations. Few of the death con-
text variables had direct effects on intrusive thoughts, al-
though having been with one’s spouse at the moment of
death protected against high levels of intrusive thoughts (b ꢂ
ꢃ.389, p ꢀ .001). Of the baseline variables, only anxiety
was positively and significantly related to intrusive thoughts;
those with higher levels of anxiety prior to their spouse’s
death had elevated levels of intrusive thoughts 6 months af-
ter the death.
To better understand the effect of death forewarning on
anxiety, yearning, and intrusive thoughts, we expanded base-
line models (Model 1) to include two sets of possible media-
tor (or suppressor) variables: death context (Model 2) and
communication (Model 3) variables. Tables 2–4 show re-
gression results for intrusive thoughts (Table 2), anxiety
(
Table 3), and yearning (Table 4). Separate models were es-
timated for the 6- and 18-month outcomes.
Intrusive Thoughts
Sudden death was associated with elevated levels of in-
trusive thoughts 6 months after the loss. The effect of death
suddenness increased when death context and communica-
tion indicators were controlled. When only demographic
and baseline health characteristics were controlled (Model
In general, death forewarning, context, and demographic
factors were relatively weak predictors of intrusive thoughts;
the final model explained only 6% of the variance in Wave
1 intrusive thoughts. By Wave 2, intrusive thoughts were no
longer significantly linked to sudden death. Although the

DEATH FOREWARNING AND GRIEF IN WIDOWED PERSONS
S245
effect of sudden death faded over time, other death con-
text variables were significantly associated with intrusive
thoughts 18 months after the loss. Persons who discussed
the death with their spouse and those whose spouses resided
in nursing homes had significantly lower levels of intrusive
thoughts.
roughly .2 standard deviations higher than widows who had
been forewarned. Gender differences in the effect of fore-
warning remained significant at the p ꢀ .05 level.
DISCUSSION
Researchers have conducted many studies to determine
whether sudden or anticipated deaths are more distressing to
the bereaved. Although findings have been inconsistent
across studies, the majority of such work has concluded that
sudden deaths are more difficult for the survivor than antici-
pated deaths (Ball, 1977; Carey, 1979–80; Glick, Weiss &
Parkes, 1974; Hill et al., 1988; Jacobs et al., 1986; Lundin,
1984; O’Bryant, 1990–91; Smith, 1978; Vachon et al.,
1982; Wells & Kendig, 1997; Willis et al., 1987; Zisook et al.,
1987). Our study of grief among elderly widowed persons
has shown that the protective effects of death forewarning
on survivors’ mental health—found in earlier research—
may be overstated. Our findings also suggest that research-
ers should no longer ask the question, “Does death fore-
warning affect psychological adjustment among widowed
persons?” Rather, the more appropriate questions are “Which
dimensions of psychological adjustment are affected by
death forewarning?” and “For whom does death forewarn-
ing affect psychological well-being?”
Anxiety
As shown in Table 3, having more than 6 months of fore-
warning was a positive and significant predictor of anxiety
levels both 6 and 18 months after the death. The effect of
prolonged forewarning was not mediated by death context
or communication variables; rather, the effect was sup-
pressed by these factors. The effects of death forewarning,
death context, and demographic factors on anxiety were
similar at both the 6- and 18-month follow-ups, and the
amount of variance explained was the same at both time
points.
The baseline model (Model 1) revealed that prolonged
forewarning was associated with a .37 standard deviation
increase in anxiety levels at Wave 1, and this effect in-
creased when death context variables were considered. The
baseline model predicting Wave 2 anxiety showed a posi-
tive (though not statistically significant) effect of prolonged
death on anxiety. However, when death context factors
were considered, this effect increased from .24 to .32 and
was significant at the p ꢀ .05 level. Individuals whose
spouses resided in a nursing home prior to death had signif-
icantly lower anxiety levels at both waves. Communication
about the death and being with one’s spouse at death were
unrelated to anxiety levels both 6 and 18 months after the
death.
First, our analyses reveal that death forewarning is not a
significant predictor of broad mental health outcomes such
as depression and overall grief either 6 or 18 months follow-
ing the loss. These findings are consistent with past studies
showing that broad measures of negative affect such as de-
pression are unrelated to death forewarning (Ball, 1977;
Bornstein et al., 1973; Bowling & Cartwright, 1982; O’Bry-
ant, 1990–91). As noted earlier, depression and grief com-
prise distinctive emotional, cognitive, physiological, and
behavioral symptoms, and these distinct symptoms may
respond in very different ways to death forewarning. Our
subscale analyses reveal that sudden death is associated
with elevated intrusive thoughts yet reduced anxiety. Con-
sequently, these competing effects may cancel out one an-
other when an aggregate scale such as grief is considered as
a dependent variable.
Our analyses also show that death forewarning is not a
significant predictor of postloss shock and anger. Because
shock and anger are believed to be immediate reactions to
loss (Glick et al., 1974; Parkes, 1970; Zisook et al., 1987),
measurements obtained 6 months after the death may be too
late to evidence effects. Death forewarning may also be un-
related to shock and anger in an older population, given that
widowhood is a normative and anticipated transition among
older adults (Neugarten & Hagestad, 1976).
Second, sudden deaths are associated with elevated levels
of intrusive thoughts. Survivors whose spouses died sud-
denly have elevated levels of intrusive thoughts 6 months
after the loss, although the effect fades by the 18-month fol-
low-up. These findings are consistent with research examin-
ing symptoms of posttraumatic stress disorder (PTSD) among
the bereaved. (Intrusive thoughts are believed to be similar
conceptually to PTSD, where unprovoked painful thoughts
about the deceased plague the survivor of a sudden or
shocking loss; Archer, 1999.) Recent studies have revealed
that individuals who lose family members to violent deaths
Yearning
As shown in Table 4, sudden deaths were associated with
reduced yearning among men and somewhat elevated
yearning among women at both the 6- and 18-month inter-
views. The first row of coefficients represents the effect of
sudden death among men, and the second row represents the
interaction term or the difference between women and men
in the effect of sudden death. When only demographic and
health variables were controlled (Model 1), sudden death
was associated with a .67 standard deviation reduction in
men’s yearning levels and a .10 standard deviation increase
(
6
i.e., from .67 to .77) in women’s yearning levels at the
-month follow-up (compared with those whose spouses
died after a warning period). Interestingly, the effect of sud-
den death on men’s yearning was partially mediated by
death context and communication characteristics, yet the ef-
fect of sudden death on women’s yearning was suppressed
by these factors. Of the three death context variables, only
one was (marginally) significant: Having provided care for
one’s spouse prior to death was associated with an increase
(
b ꢂ .27) in yearning 6 months after the loss.
At the 18-month follow-up, the general patterns docu-
mented at Wave 1 persisted although effects were weaker.
At Wave 2, men whose wives died suddenly continued to
have yearning scores roughly .5 standard deviations lower
than widowed men who had some forewarning. Women
whose spouses died suddenly had yearning levels that were

S246
CARR ET AL.
(
accident, suicide, or homicide) are much more likely to
(Neugarten & Hagestad, 1976). The harmful effects of death
forewarning on men’s yearning may instead reflect gender
differences in how individuals respond to their partner’s end
of life experiences. For men, the forewarning period may be
associated with increased closeness to the spouse and isola-
tion from others. During this time, men may become even
more emotionally bonded to their spouse, at the expense of
relationships with others. Given gender differences in mor-
tality, men may have few same-sex peers who are also
awaiting an ill wife’s death. In contrast, women may rely on
their female friends’ direct experience with spousal illness
to help them through the difficult dying process and thereaf-
ter (Fooken, 1985).
meet the criteria for PTSD than those whose relatives die of
other conditions (Zisook, Chentsova-Dutton, & Schuchter,
1
998; Kaltman & Bonanno, 1999). Likewise, family mem-
bers of murder victims report higher levels of intrusive
thoughts than relatives of persons who die naturally (Ry-
nearson & McCreery, 1993). Our analysis shows further
that the effect of sudden death on intrusive thoughts is no
longer significant by the 18-month follow-up, consistent
with the observation that intrusive thoughts will fade over
time, as the widowed person becomes enmeshed in other ac-
tivities and relationships (Parkes, 1985; Parkes & Brown,
1
972).
Third, prolonged forewarning increases survivors’ anxi-
Why do the effects of forewarning differ by sex for yearn-
ing only? This pattern may reflect the fact that yearning is
the only subdimension of grief that is explicitly relational
and thus may reflect gender differences in how spousal rela-
tionships unfold during the predeath period. Intrusive thoughts,
in contrast, reflect the cognitive process of stimulus-inde-
pendent thoughts (Archer, 1999, p. 65), and anxiety reflects
worries about coping with daily experiences.
In sum, our findings suggest that among older adults,
sudden spousal death does not have far-reaching deleterious
effects on mental health, except for increasing intrusive
thoughts during the first 6 months after widowhood. Pro-
longed periods of anticipating a spouse’s death are also not
generally deleterious, but they do seem to increase anxiety
for both men and women and yearning among men only.
These gender-specific effects may be due to the different
experiences of men and women in caring for and relating to
a dying spouse. Future research is needed to more ade-
quately understand the limited but real effects found here
and to explore the degree to which other characteristics of
the deceased, the surviving spouse, the marital relationship,
and the social context condition the ways in which older
adults respond to widowhood and its forewarning.
ety levels both 6 and 18 months following the death. The
harmful effects of advanced forewarning (i.e., more than 6
months) are suppressed by death context and communica-
tion characteristics; thus past studies that omitted controls
for death context characteristics may have underestimated
the effect of advanced forewarning on elderly survivor’s
anxiety. Our analyses also show that the harmful effects of
advanced forewarning cannot be explained away by care-
giving, couple communication, spouse’s age at death, or
whether one’s spouse resided in a nursing home prior to
death. Grief scholars thus face the task of identifying why
and how prolonged forewarning periods lead to elevated
anxiety among older adults.
Sociological research on chronic stressors provides a
starting point for this inquiry. Chronic (or long-term) stres-
sors, such as caregiving or watching a spouse suffer from a
debilitating illness, are believed to be more difficult for psy-
chological adjustment than stressors of shorter duration
(
Avison & Turner, 1988; see also Pearlin & Skaff, 1995, for
a review). The psychological effects of a chronic stressor
may be compounded when experienced in conjunction with
concurrent or successive stressors (Holmes & Rahe, 1967;
Johnson & Catalano, 1983). Spouses who spend prolonged
periods anticipating their spouse’s death are presumably at a
greater risk of experiencing concurrent stressors than those
whose anticipation periods are confined to a shorter time
frame. By incorporating indicators of other prewidowhood
stressors into future analyses, the pathways linking pro-
longed forewarning and survivors’ elevated anxiety levels
may become more apparent.
Acknowledgment
This research was supported by Grants R01-AG15948 and P01-
AG05561 from the National Institute of Aging and from the Nancy Pritzker
Research Network.
Address correspondence to Deborah Carr, Department of Sociology,
University of Michigan, 500 S. State Street, Ann Arbor, MI 48109. E-mail:
carrds@umich.edu
Fourth, the relationship between sudden death and yearn-
ing differs significantly for men and women. Sudden death
predicts a slight increase in women’s yearning, a finding
that is generally consistent with past clinical research re-
vealing that sudden death is linked to poor adjustment to
loss (Lindemann, 1944; Rando, 1986). It is not surprising
that this finding converges with past research on the effects
of forewarning, given that the overwhelming majority of
studies examining the psychological consequences of late-
life widowhood focus on samples of women only (e.g. Lo-
pata, 1973; O’Bryant, 1990–91).
However, it is surprising that sudden death reduces
yearning for men, yet has only a small harmful effect for
women. We had expected the reverse: Because women are
more likely than men to be widowed, they may be better
prepared for the transition—even if it occurs suddenly
References
Anashensel, C. S., Pearlin, L. I., Mullan, J. T., Zarit, S., & Whitlach, C.
(
1995). Profiles in care giving: The unexpected career. New York: Ac-
ademic Press.
Archer, J. (1999). The nature of grief: The evolution and psychology of re-
actions to loss. London: Routledge Press.
Averill, J. R., & Wisocki, P. A. (1981). Some observations on behavioral
approaches to the treatment of grief among the elderly. In H. J. Sobel
(
Ed.), Behavioral therapy in terminal care (pp. 125–150). Cambridge,
England: Ballinger.
Avison, W. R., & Turner, R. J. (1988). Stressful life events and depressive
symptoms: Disaggregating the effects of acute stressors and chronic
strains. Journal of Health and Social Behavior, 29, 253–264.
Ball, J. F. (1977). Widows’ grief: The impact of age and mode of death.
Omega, 7, 307–333.
Bennett, K. M., & Vidal-Hall, S. (2000). Narratives of death: A qualitative
study of widowhood in later life. Ageing and Society, 20, 413–428.
Blauner, R. (1966). Death and social structure. Psychiatry, 25, 378–399.

DEATH FOREWARNING AND GRIEF IN WIDOWED PERSONS
S247
Bornstein, P. E., Clayton, P. J., Halikas, J. A., Maurice, W. L., & Robins,
E. (1973). The depression of widowhood after thirteen months. British
Journal of Psychiatry, 122, 561–566.
Bowling, A., & Cartwright, A. (1982). Life after death: A study of the el-
derly widowed. London: Tavistock.
Calhoun, L. G., & Allen, B. G. (1991). Social reactions to the survivor of a
suicide in the family: A review of the literature. Omega, 20, 265–272.
Carey, R. G. (1979–80). Weathering widowhood: Problems and adjust-
ment of the widowed during the first year. Omega, 10, 163–174.
Carnelley, K. B., Wortman, C. B., & Kessler, R. C. (1999). The impact of
widowhood on depression: Findings from a prospective survey. Psy-
chological Medicine, 29, 1111–1123.
Johnson, C. L., & Catalano, D. J. (1983). A longitudinal study of family
supports to impaired elderly. The Gerontologist, 23, 612–618.
Kalish, R. A. (1981). Death, grief, and caring relationships. Monterey,
CA: Brooks/Cole.
Kaltman, S., & Bonanno, G. (1999). Trauma and bereavement: Examining
the role of sudden and violent deaths. Unpublished manuscript.
Kramer, D. (1996–97). How women relate to terminally ill husbands and
their subsequent adjustment to bereavement. Omega, 34, 93–106.
Lindemann, E. (1944). Symptomatology and management of acute grief.
American Journal of Psychiatry, 151, 155–160.
Lopata, H. Z. (1973). Widowhood in an American city. Cambridge, MA:
Schenkman.
Carr, D., House, J. S., Kessler, R. C., Nesse, R. M., Sonnega, J., & Wort-
man, C. (2000). Marital quality and psychological adjustment to
widowhood among older adults: A longitudinal analysis. Journal of
Gerontology: Social Sciences, 55B, S197–S207.
Clayton, P., Desmarais, L., & Winokur, G. (1968). A study of normal be-
reavement. American Journal of Psychiatry, 125, 168.
Lundin, T. (1984). Morbidity following sudden and expected bereavement.
British Journal of Psychiatry, 144, 84–88.
Maddison, D., & Walker, W. (1967). Factors affecting the outcome of con-
jugal bereavement. British Journal of Psychiatry, 113, 1057–1067.
Matthews, A. M. (1991). Widowhood in late life. Toronto, Canada:
Buttersworth.
Clayton, P. L., Halikas, J. A., Maurice, W. L., & Robins, E. (1973). Antici-
patory grief and widowhood. British Journal of Psychiatry, 122, 47–51.
Derogatis, L. R., & Cleary, P. A. (1977). Confirmation of the dimensional
structure of the SCL-90: A study in construct validation. Journal of
Clinical Psychology, 33, 981–989.
DeSpelder, L. A., & Strickland, A. L. (1992). The last dance: Encounter-
ing death and dying (3rd Ed.). Mountain View, CA: Mayfield.
Dunn, R. G., & Morrish-Vidners, D. (1987). The psychological and social
experience of suicide survivors. Omega, 18, 175–215.
McGloshen, T. H., & O’Bryant, S. L. (1988). The psychological well-
being of older, recent widows. Psychology of Women, 12, 99–116.
McLeroy, K., & Crump, C. (1994, Spring). Health promotion and disease
prevention: A historical perspective. Generations, 18, 9–17.
Neugarten, B., & Hagestad, G. O. (1976). Age and the life course. In G.
Binstock, & E. Shanas (Eds.), Handbook of aging and the social sci-
ences (pp. 35–55). New York: Van Nostrand Reinhold.
Norris, F. H., & Murrell, S. A. (1987). Older adult family stress and adapta-
tion before and after bereavement. Journal of Gerontology, 42, 606–612.
O’Bryant, S. L. (1990–91). Forewarning of husband’s death: Does it make
a difference? Omega, 22, 227–239.
Farberow, N. L., Gallagher-Thompson, D., Gilewski, M., & Thompson, L.
(
1992). Changes in grief and mental health of bereaved spouses of
older suicides. Journal of Gerontology: Psychological Sciences, 47,
P357–P366.
Fengler, A. P., & Goodrich, R. (1979). Wives of elderly, disabled men: The
hidden patients. The Gerontologist, 19, 175–183.
Fiewiger, M., & Smilowitz, M. (1984–85). Relational conclusion through
interaction with the dying. Omega, 15, 161–172.
Foner, N. (1994). The caregiving dilemma: Work in an American nursing
home. Berkeley: University of California Press.
O’Bryant, S. L., Straw, L. B., & Meddaugh, D. I. (1990). Contributions of
the caregiving role to women’s development. Sex Roles, 23, 645–658.
Olshansky, S. J., & Ault, A. B. (1986). The fourth stage of the epidemio-
logic transition: The age of delayed degenerative diseases. Millbank
Memorial Fund Quarterly, 64, 355–391.
Parkes, C. M. (1970). The first year of bereavement: A longitudinal study
of the reaction of London widows to the death of their husbands. Psy-
chiatry, 33, 444–467.
Fooken, I. (1985). Old and female: Psychosocial concomitants of the aging
process in a group of older women. In J. Munniches, P. Mussen, E.
Olbrich, & P. G. Coleman (Eds.), Life span and change in a geronto-
logical perspective (pp. 7–101). Orlando, FL: Academic Press.
Fulton, R., & Fulton, J. A. (1971). A psychosocial aspect of terminal care:
Anticipatory grief. Omega, 2, 91–100.
George, L., & Gwyther, L. (1986). Caregiver well-being: A multidimen-
sional examination of family caregivers of demented adults. The
Gerontologist, 26, 253–259.
Parkes, C. M. (1985). Bereavement. British Journal of Psychiatry, 146,
11–17.
Parkes, C. M., & Brown, R. (1972). Health after bereavement: A controlled
study of young Boston widows and widowers. Psychosomatic Medi-
cine, 34, 449.
Parkes, C. M., & Weiss, R. S. (1983). Recovery from bereavement. New
York: Basic Books.
Pattison, E. M. (1977). The experience of dying. Englewood Cliffs, NJ:
Prentice-Hall.
George, L. K. (1993). Sociological perspectives on life transitions. Annual
Review of Sociology, 19, 353–373.
Gerber, I. (1974). Anticipatory bereavement. In B. Schoenberg, A. C. Carr,
A. H. Kutscher, D. Peretz, & I. K. Goldberg (Eds.), Anticipatory grief
Pattison, E. M. (1978). The living-dying process. In C. A. Garfield (Ed.),
Psychosocial care of the dying patient (pp. 133–168). New York:
McGraw-Hill.
Pearlin, L. I. (1982). The social contexts of stress. In L. Goldberger & S.
Breznitz (Eds.), Handbook of stress (pp. 367–379). New York: Free
Press.
(
pp. 26–31). New York: Columbia University Press.
Gerber, I., Rusalem, R., Hannon, N., Battin, D., & Arkin, A. (1975). Antic-
ipatory grief and aged widows and widowers. Journal of Gerontology,
Pearlin, L. I., & Lieberman, M. (1979). Social sources of emotional dis-
tress. In R. Simmons (Ed.), Research in community and mental health
(pp. 217–248). Greenwich, CT: JAI Press.
Pearlin, L. I., & Skaff, M. M. (1995). Perspectives on the family and stress
and late life. In J. Loranz (Ed.), Handbook of aging and mental health:
An integrative approach (pp. 323–340). New York: Plenum Press.
Radloff, L. (1977). The CES-D scale: A self-report depression scale for re-
search in the general population. Applied Psychological Measurement,
1, 381–401.
3
0, 225–229.
Glick, I. O., Weiss, R. S., & Parkes, C. M. (1974). The first year of be-
reavement. New York: Wiley.
Hill, C. D., Thompson, L. W., & Gallagher, D. (1988). The role of antici-
patory bereavement in older women’s adjustment to widowhood. The
Gerontologist, 28, 792–796.
Hinds, C. (1985). The needs of families who care for patients with cancer
at home: Are we meeting them? Journal of Advanced Nursing, 10,
5
75–581.
Rando, T. A. (1986). A comprehensive analysis of anticipatory grief: Per-
spectives, processes, promises and problems. In T. A. Rando (Ed.),
Loss and anticipatory grief (pp. 3–38). Lexington, MA: Lexington
Books.
Hinrichsen, G. A., Hernandez, N. A., & Pollack, S. (1992). Difficulties and
rewards in family care of the depressed older adult. The Gerontologist,
3
2, 486–492.
Holmes, J. H., & Rahe, R. H. (1967). The social readjustment scale. Jour-
nal of Psychosomatic Research, 11, 213–228.
Hyman, H. H. (1983). Of time and widowhood: Nationwide studies of en-
during effects. Durham, NC: Duke University Press.
Jacobs, S. (1993). Pathologic grief: Maladaptation to loss. Washington,
DC: American Psychiatric Association Press.
Reed, M. D. (1998). Predicting grief symptomatology among the suddenly
bereaved. Suicide & Life-Threatening Behavior, 28, 285–301.
Roach, M. J., & Kitson, C. (1989). Impact of forewarning on adjustment to
widowhood and divorce. In D. A. Lund (Ed.), Older bereaved spouses:
Research with practical applications (pp. 185–200). New York:
Hemisphere.
Jacobs, S., Kasl, S., & Ostfeld, A. (1986). The measurement of grief: Be-
reaved versus non-bereaved. The Hospice Journal, 2, 21–36.
Rosenblatt, P. (1983). Bitter, bitter tears: Nineteenth century diarists and
twentieth century grief theories. Minneapolis: University of Minnesota.

S248
CARR ET AL.
Ross, M., Rosenthal, C., & Dawson, P. (1993). Spousal caregiving follow-
ing institutionalization: The experience of elderly wives. Journal of
Advanced Nursing, 18, 1531–1539.
Rynearson, E. K. (1984). Bereavement after homicide: A descriptive study.
American Journal of Psychiatry, 141, 1452–1454.
1996 (116th Ed.). Washington, DC: U.S. Government Printing Office.
Vachon, M. L., Formo, S. A., Freedman, K., Lyall, A., Rogers, J., & Free-
man, S. J. (1976). Stress reactions to bereavement. Essence: Issues in
the Study of Aging, Dying and Death, 1, 23–33.
Vachon, M. L., Rogers, J., Lyall, W. A., Lances, W. J., Sheldon, A. R., &
Freeman, S. J. (1982). Predictors and correlates of adaptation to conju-
gal bereavement. American Journal of Psychiatry, 139, 998–1002.
van der Wal, J. (1989-90). The aftermath of suicide: A review of empirical
evidence. Omega, 20, 149–171.
Rynearson, E. K., & McCreery, J. M. (1993). Bereavement after homicide:
A synergism of trauma and loss. American Journal of Psychiatry, 150,
2
58–261.
Sanders, C. M. (1982–83). Effects of sudden vs. chronic illness on be-
reavement outcome. Omega, 13, 227–241.
Schwab, J. J., Chalmers, J. M., Conroy, S. J., Farris, P. N., & Markush, R. E.
Wells, Y. D., & Kendig, H. L. (1997). Health and well-being of spouse care-
givers and the widowed. The Gerontologist, 37, 666–674.
(
1975). Studies in grief: A preliminary report. In B. Schoenberg, A. C.
Willis, L., Thomas, P., Garry, P. J., & Goodwin, J. (1987). A prospective
study of response to stressful life events in initially healthy elders.
Journal of Gerontology, 42, 627–630.
Carr, A. H. Kutscher, D. Peretz, & I. K. Goldberg (Eds.), Anticipatory
grief (pp. 78–87). New York: Columbia University Press.
Siegel, K., & Weinstein, L. (1983). Anticipatory grief reconsidered. Jour-
nal of Psychosocial Oncology, 1, 61–73.
Singh, B., & Raphael, B. (1981). Postdisaster morbidity of the bereaved: A
possible role for preventive psychiatry? The Journal of Nervous and
Mental Disease, 169, 203–212.
Wilson, V. (1990). The consequences of elderly wives caring for disabled
husbands: Implications for practice. Social Work, 35, 417–421.
Wright, L. K. (1991). The impact of Alzheimer’s disease on the marital re-
lationship. The Gerontologist, 31, 224–237.
Zisook, S., Chentsova-Dutton, Y., & Schuchter, S. R. (1998). Post-trau-
matic stress disorder following bereavement. Annals of Clinical
Psychiatry, 10, 157–163.
Smith, W. J. (1978). The etiology of depression in a sample of elderly wid-
ows. Journal of Geriatric Psychiatry, 11, 81–83.
Stroebe, M. S., & Stroebe, W. (1983). Who suffers more? Sex differences
in health risks of the widowed. Psychological Bulletin, 93, 279–301.
Sweeting, H. N., & Gilhooly, M. L. M. (1990). Anticipatory grief: A re-
view. Social Science and Medicine, 30, 1073–1080.
Zisook, S., Devaul, R., & Click, N. (1982). Measuring symptoms of grief
and bereavement. American Journal of Psychiatry, 139, 1590–1593.
Zisook, S., & Schuchter S. R. (1991). Early psychological reaction to the
stress of widowhood. Psychiatry, 54, 320–332.
Thoits, P. A. (1983). Dimensions of life events that influence psychologi-
cal distress: An evaluation and synthesis of the literature. In H. B.
Kaplan (Ed.), Psychosocial stress: Trends in theory and research (pp.
Zisook, S., Schuchter, S. R., & Lyons L. E. (1987). Adjustment to widow-
hood. In S. Zisook (Ed.), Biopsychosocial aspects of bereavement (pp.
51–74). Washington, DC: American Psychiatric Association Press.
3
3–103). New York: Academic Press.
Umberson, D., Wortman, C., & Kessler, R. C. (1992). Widowhood and de-
pression: Explaining long-term gender differences in vulnerability.
Journal of Health and Social Behavior, 33, 10–24.
Received November 2, 2000
Accepted January 26, 2001
U.S. Bureau of the Census. (1996). Statistical abstract of the United States: