OMEGA, Vol. 68(3) 183-205, 2013-2014

A PARENT’S TEARS: PRIMARY RESULTS
FROM THE TRAUMATIC EXPERIENCES
AND RESILIENCY STUDY

JOANNE CACCIATORE
Arizona State University
JEFFREY R. LACASSE
Florida State University
CYNTHIA A. LIETZ
Arizona State University
JANE MCPHERSON
Florida State University

ABSTRACT

This study examined health/mental health status, family functioning, and
resiliency among a sample of bereaved parents (N = 503). Participants were
recruited from an online support community to complete an online survey
instrument (response rate = 51.75%). The questionnaire contained an array of
self-report instruments, such as the Hopkins Symptoms Checklist-25 (HSCL25), the Impact of Event Scale-Revised (IES-R), and the Family Assessment
Device (FAD), as well as open-ended questions. Many respondents scored
over the clinical cut-off for the HSCL-25 (51.3%, n = 258) and IES-R (42.3%,
n = 213). IES-R scores were negatively correlated with years-since loss (r =
–0.24, p < .05). In narrative responses, participants described a wide range of
deeply impactful mental and physical health problems. The results indicate
significant clinical distress in this sample of bereaved parents, with many
reporting enduring psychological, familial, and health consequences following the death of a child.

183
Ó 2013, Baywood Publishing Co., Inc.
doi: http://dx.doi.org/10.2190/OM.68.3.a
http://baywood.com

184 / CACCIATORE ET AL.

INTRODUCTION AND OVERVIEW
The death of a child is recognized across cultures as one of life’s greatest and most
interminable tragedies, and bereaved parents have been the subject of scientific
inquiry at an increasing rate. Using the PsychINFO database (Proquest version),
we searched the psychological literature for articles on bereaved parents (using
search terms “parents” AND “bereaved” or “bereavement” or “child death,” peerreviewed articles only, all fields) from 1960-present. From 1960-69, there were
only 97 articles published; from 1970-79, 350 articles; and from 1980-89, 846
articles. From 1990-99, the number of articles rose to 1250; and from 2000present, the number has skyrocketed, to 4,797 articles. Despite the burgeoning
interest in this subject, biopsychosocial adaptations remain dimly understood
within this population.
The psychological impact of bereavement can be substantial, especially in the
case of child death. Sanders (1979-1980) found that at 2.2 months post-loss,
grieving parents had higher grief reactions as compared to those who suffered
conjugal or parental loss, with despair a prominent feature (see also Leahy, 19921993). At long-term follow-up, mothers who lose a child to stillbirth, SIDS, or
neonatal death have higher rates of depression and anxiety compared to controls
(Boyle, Vance, Najman, & Thearle, 1996; Turton, Evans, & Hughes, 2009).
Rogers, Floyd, Seltzer, Greenberg, and Hong (2008) found an increase in longstanding depressive symptoms combined with a diminished sense of well-being
and purpose in bereaved parents, lasting several decades beyond the death of a
child. A Danish case-control study found an increased risk of suicide completion
in bereaved parents (Qin & Mortenson, 2003). This was strikingly high in the first
month after loss of a child; the odds ratio for completed suicide among men was
34.68 (95% CI = 19.31, 62.29) and for women was 76.05 (95% CI = 26.64,
217.08). While the increased risk of suicide fell over time, it remained elevated at
$5 years post-loss (e.g., for women, OR = 1.86, 95% CI = 1.25, 2.77). Another
observational study from Denmark (Li, Lauresen, Precht, Olsen, & Mortensen,
2005) found an increased risk of psychiatric hospitalization following the death of
a child (RR = 1.67, 95% CI = 1.53, 1.83).
Researchers have also frequently observed a negative health impact on parents
after child’s death (e.g., Murphy, Lohan, Braun, Johnson, Cain, & Beaton, 1999;
Rostilla, Saarela, & Kawachi, 2011), although there are also conflicting studies
that do not find this association (e.g., Birnbaum, Stewart, & Phillips, 1996; Li,
Johansen, & Olsen, 2002). Among the studies that do identify an association is a
large-scale epidemiological study reporting an increase in premature mortality
rates for bereaved mothers (HR = 1.43, 95% CI = 1.24, 1.64). The authors suggest
that “pathophysiological changes related to stress could increase susceptibility to
infectious diseases, affect the risk and prognosis of cancer, and lead to diseases of
the cardiovascular system . . . increasing smoking and alcohol intake, altering
dietary patterns, and reducing physical activity” (Li, Precht, Mortensen, & Olsen,

A PARENT’S TEARS / 185

2003, p. 366). Bereaved parents with physical health problems are more likely to
have co-occurring mental health difficulties as well; bereaved mothers with poor
selfreported health status were 4.6 times more likely to also report trauma
symptoms (Murphy et al., 1999).
While some studies refute the role of child death on divorce (Schwab, 1998),
other studies suggest that child death may, indeed, have a substantially negative
impact on family and marital functioning. Rogers et al. (2008) found that a group
of bereaved parents had higher rates of marital disruption (30.4%) than a comparison group (23.8%). One large study with a nationally represented sample of
3461 women found a significantly increased risk of marital dissolution in parents
experiencing stillbirth, infant, and child death that did not extend to infertility and
miscarriage (Shreffler, Hill, & Cacciatore, 2012). Beyond marital distress, family
functioning in general is adversely impacted by the death of a child (Lohan &
Murphy, 2002).
Researching parents who have experienced child death is sensitive, often logistically complex. Because they are a vulnerable population, care for clients must be
prioritized over data collection. In an increasingly electronic culture, those with
myriad conditions often seek social support in online support groups. The content
of such groups can be a rich source of data (e.g., Swartwood, Veach, Kuhne, Lee,
& Ji, 2011). Survey research that recruits participants from online support communities has many potential advantages; participants have already begun to
voluntarily share their experiences; they are personally invested in the topic, and
they are more likely to find the topic salient. A potential disadvantage of surveying
online support groups is that, in terms of demographics, participants tend to be
homogenous—largely white, relatively affluent, and educated (i.e., Eun-ok, Chee,
Lim, Guevara, Tsai, Clark, et al., 2007).
Although there are many available online support communities for bereavement, there are few surveys of participants from these communities. Feigelman,
Gorman, Beal, and Jordan (2008) found that 72.1% of participants in an online
support group for suicide survivors were currently depressed. Yet, there is a clear
gap in the literature in that there are no surveys of participants in Internet support
communities that focus specifically on parental bereavement. We thus conducted a
survey research project (TEARS: Traumatic Experiences and Resiliency Study)
examining this understudied population, surveying participants of support forums
of a non-profit organization which serves bereaved parents. Specifically, we
sought to describe this sample in terms of demographics, psychological distress,
family functioning, health status, and resiliency.
METHOD
We conducted a cross-sectional online survey of online support forum participants from June to October of 2010. This online forum launched in 1997 and now
offers 27 different forum boards moderated by nine volunteers trained by the

186 / CACCIATORE ET AL.

nonprofit agency. As of 2010, 5,955 individuals had registered for these forums at
some point.
The online survey contained a combination of questions created to collect information on the respondent as well as several validated measures commonly used in
research on mental health, family functioning, and resiliency. Given the sensitivity
of the topic, we pilot-tested the survey by administering to a small group of individuals who had experienced bereavement, some of whom were academic
researchers, and incorporated the feedback into the final version of the survey. The
survey was then placed on the World Wide Web using Qualtrics survey software.
The survey utilized skip-logic, whereby participants were asked some questions
only if they met criteria set by a specific question. Thus, respondents were exposed
to a variable number of questions; depending on their answers, participants
answered from 167-209 individual items.
The forum membership database listed 1120 participants active from 20092010, with 972 valid e-mail addresses. We first sent an e-mail solicitation containing a link to the online survey, which resulted in 313 responses. We followed
with a series of further efforts, spaced several weeks apart; a reminder e-mail, a
video message from the agency, and the offer of a $20 gift card incentive for
participating. These follow-up efforts generated an additional 190 responses. We
received a total of 503 responses, resulting in a final response rate of 51.75%. All
but nine respondents (99.83%) were mothers or fathers who had suffered the death
of a child; seven respondents were grandparents, one was an aunt, and one
indicated “other.” Because of their close relationship with the child who died as
caregivers, we chose to include these individuals in our analyses and refer to the
overall sample as “parents.”
Measures
Demographics and Circumstances of Loss

In order to describe the characteristics of the sample, respondents were asked a
series of questions regarding demographics and the circumstances of their loss.
These included gender, age, race/ethnicity, income, educational level, and religion. Respondents were also asked if the child’s death was unexpected, whether
they were present when their child died, if they witnessed the death, if they saw the
child after death, and if they held/touched the child after he/she died. Several
temporal questions were asked which allowed the calculation of time-since loss.
Mental Health Functioning

As a gauge of psychiatric history, and to contextualize the other mental health
measures (see below), we asked “In the months before the loss, were you diagnosed with any mental health condition?” and “Were you taking psychiatric

A PARENT’S TEARS / 187

medications at the time of your loss?” Participants then took several validated
mental health measures.
Hopkins Symptoms Checklist (HSCL-25): The HSCL-25 (Derogatis, Lipman,
Rickels, Uhlenluth, & Covi, 1974) is a 25-item self-report instrument that contains
both anxiety and depression subscales. Respondents were asked to identify the
degree to which they had experienced each symptom on a 4-point scale ranging
from “not at all” (1) to “extremely” (4). In order to classify clinical cases, we used
a cut-off of > 1.75 (average score). This has performed well in identifying depression; research with women found a sensitivity of 0.81 and a specificity of 0.70, and
for panic and generalized anxiety disorder, a sensitivity of 0.67 and specificity of
0.73 (Sandanger, Moum, Ingebrigtsen, Dalgard, Sorenson, & Bruusgaard, 1998).
The Impact of Event Scale-Revised (IES-R): The IES-R is a self-report measure
which asks respondents to rate 22 difficulties (for example “I felt irritable and
angry”) to gauge how distressing each particular area has been over the past 7
days. Answers are recorded on a 5-point scale ranging from “not at all” (0) to
“extremely” (4), and the IES-R is scored by averaging item responses. There is no
agreed upon clinical cut-off for the IES-R (Weiss, 2004), although various cut
scores have been proposed and tested, and the IES-R is sometimes used to identify
clinical cases (e.g., Samuelson, Lundberg, & Fridlund, 2007). We chose to use an
IES-R total score cut-off of 33 (average score of 1.5). Using the 1.5 cut-off,
Creamer, Bell, and Faila (2003) found that the IES-R had “. . . a sensitivity of 0.91,
a specificity of 0.82, positive predictive power of 0.9, and negative predictive
power of 0.84” (p. 1494).
Drug and alcohol use: In order to assess for a post-loss increase in drug/alcohol
intake, we asked respondents, “Did you increase your drug or alcohol consumption after the death?” Those who answered “yes” to this question were administered the alcohol and drug use section of the self-report Addiction Severity Index
(ASI; see Rosen, Henson, Finney & Moos, 2000).
Physical Health Status

Participants were asked, “Has your physical health significantly changed since
your loss?” and could answer in response that their health: (a) has improved since
the loss, (b) has remained the same since the loss, or (c) has declined since the loss.
Participants were also asked an open-ended follow-up question, “Please explain
how your health has changed since the loss.”
Family Functioning

The General Functioning subscale of the Family Assessment Device (FAD),
was administered. This 12-item self-report instrument measures family functioning on a 4-point Likert scale according to the seven dimensions of the McMaster

188 / CACCIATORE ET AL.

Model of Family Functioning (Epstein, Bishop, & Levin, 1978). The FAD has
established psychometric properties (Miller, Ryan, Keitner, Bishop, & Epstein,
2000; Tutty, 1995), and discriminates between psychiatric and non-clinical families (Miller et al., 2000). In addition, one open-ended item asked, “How has your
loss affected your relationships with your spouse or partner and children?” to
allow respondents to provide more specific descriptions regarding their family
functioning since loss.
Resilience and Coping

The Posttraumatic Growth Inventory (PTGI) was administered to understand
how respondents may have reconstructed their lives in positive ways as a result of
the loss (Tedeschi & Calhoun, 1996). The PTGI is a 21-item instrument that measures respondents’ perceptions regarding their ability to positively reconstruct their
lives following a traumatic event. The PTGI uses a 6-point Likert scale, and the
instrument’s internal consistency was reported as .90 in one study (Tedeschi &
Calhoun, 1996). The scale measures perceived changes on the following components: new possibilities, relating to others, personal strength, spiritual change,
and appreciation of life. A limitation of the PTGI data in the present study relates
to the personal strength subscale; one item was not transferred to the online survey
correctly, leaving it out of the administration. The total score was calculated
without the self-reliance item. In addition to the PTGI, participants were asked
about their choices to engage in helping behaviors. Specifically, the survey asked
respondents about whether they attended support groups, counseling, and/or
engaged in volunteerism after the death of their child.
Statistical Analysis
The data were analyzed using PASW version 20.0 (SPSS Inc., Chicago, IL).
There was a low proportion of missing data; ~5% for HSCL-25 and IES-R, ~10%
for PTGI and FAD. A few respondents filled out the majority of the standardized
measures but did not answer every item (e.g., on the IES-R, 16 respondents missed
one item). Such missing values were replaced using mean imputation to generate
scale/subscale totals. In the overwhelming majority of cases, only one item was
replaced. Other than this use of within-scale mean imputation, we did not replace
other missing data, and we present available-case analyses (Pigott, 2001). The
years-since-loss variable was highly positively skewed and was normalized
through logarithmic transformation.
Our primary goal was to describe this unique sample in some detail. To accomplish this, we use descriptive statistics (primarily measures of central tendency),
proportions, and bivariate correlation, as well as narrative summaries of openended questions. Due to the nature of our data (a non-random survey of a
population, albeit one with a robust response rate), our results should be interpreted descriptively rather than inferentially (Berk & Freedman, 2009). We report

A PARENT’S TEARS / 189

95% confidence intervals for all major quantitative findings using the definition of
confidence interval forwarded by Rothman (2003). To calculate confidence
intervals for proportions, we used the Wilson (1927) method with continuity correction (Newcombe, 1998). When reporting percentages for categorical variables,
we use the denominator of N = 503, our total sample; some variables do not sum to
100% due to missing data.
Procedure
The TEARS questionnaire was of substantial length, and space constraints
do not allow the presentation of all data in one manuscript; in order to report
the results in sufficient detail, other manuscripts will be generated. Here, we
concentrate on describing the characteristics of our sample, primarily in terms of
mental/ physical health and family functioning. The survey also contained
questions about the economic costs of bereavement (which all participants were
administered), as well as psychiatric medications (only administered to a
subsample). These data analyses, which are substantial enough to preclude including them here, are reported in separate manuscripts (i.e., Lacasse & Cacciatore, in
press; Fox, Cacciatore, & Lacasse, in press). Finally, this study was approved by
the Institutional Review Board of our academic institution, by the ethics
committee of the participating nonprofit organization, and all respondents gave
informed consent.

RESULTS
Description of the Sample
The sample was overwhelmingly female (95.0%, n = 478) and white (85.7%,
n = 431). Most responses came from the United States (n = 390, 77.53%) or from
United Kingdom, Australia, or Canada (n = 42, 8.34%). Almost all respondents
with missing data on this question connected through a U.S.-based server. Respondents resided mostly in suburban areas (n = 238, 47.3%), followed by rural
(n = 122, 24.3%) and then urban (n = 95, 18.9%) residency.
Educationally, 48.5% (n = 262) of the sample had completed college degrees at
the bachelors or graduate level, while 32.2% (n = 162) had attended some college or
technical school, and fewer than 3% had not completed high school (see Table 1).
Almost 70% (n = 350) identified as Christian, while 16.9% (n = 85) selected “no
religious affiliation.” In terms of categorical annual household income, 13.1%
(n = 66) earned < $28,000 annually, with 15.7% (n = 79) earning between $28,001$50,000, and 17.5% (n = 88) earning between $50,001-$75,000. The remaining
respondents (46.3%, n = 217) all earned > $75,000 a year (see Table 1).

478

350
12

Religious Affiliation
Christian
Jewish

431
5
18
3
1
9
1

n

Sex
Female

Age at loss
Age at survey
Years-since-loss

Race
Caucasian/White
African American
Latino
Native American
Asian
Mixed Race
Other

Variable

69.6
2.4

95.0

85.7
1.00
3.6
0.6
2.2
1.8
0.20

%

[65.3, 73.5]
[1.4, 4.1]

[92.7, 96.7]

[82.3, 88.6]
[0.4, 2.40]
[2.2, 5.7]
[0.16, 1.9]
[1.6, 4.0]
[0.9, 3.5]
[0.01, 1.3]

95% CI

17.0
18.07
00.43

Min

Table 1. Characteristics of Sample

67.00
70.22
41.48

Max

33.41 (8.50)
37.74 (8.80)
4.31 (4.31)

M (SD)

[32.64, 45.17]
[36.95, 38.53]
[3.94, 4.72]

95% CI

190 / CACCIATORE ET AL.

102
142
162
41
12

66
79
88
52
84
81

1
55
85

20.3
28.2
32.2
8.2
2.4

13.1
15.7
17.5
10.3
16.7
16.1

0.2
10.9
16.9

[16.0,
[24.4,
[28.2,
[6.0,
[1.3,

[10.4,
[12.7,
[14.3,
[7.9,
[13.6,
[13.1,

24.1]
32.4]
36.5]
11.0]
4.3]

16.5]
19.3]
21.2]
13.4]
20.3]
19.7]

[0.01, 1.3]
[8.5, 14.0]
[13.8, 20.5]

Note: CI = confidence interval; Min = Minimum value; and Max = Maximum value.

Level of Education
Graduate Degree
Bachelor Degree
Some College
H.S. Diploma
< H.S. Diploma

Current Household Income
Under $28,000
$28,001-$50,000
$50,001-$75,000
$75,001-$90,000
$90,001-$125,000
> $125,000

Muslim
Other
No affiliation

A PARENT’S TEARS
/ 191

192 / CACCIATORE ET AL.

Experiences with Loss
Many respondents (43.1%, n = 217) reported witnessing the death of their child,
while 16.3% (n = 82) reported that they were present at the time of death but did
not witness it. Most (88.3%, n = 444) reported that they saw their child after death,
and 84.7% (n = 426) reported that they had held or touched their child after death.
The death was unexpected for most respondents (77.7%, n = 391), while 11.3%
(n = 57) reported that the death was expected, and 7.8% (n = 39) reported that the
death was both expected and unexpected. The death of a baby to stillbirth was
commonly reported in this sample (n = 215, 42.7%). At the time of the survey,
respondents had a mean age of 37.74 years (SD = 8.8), with a mean time-since loss
of 4.31 years (SD = 4.31).
Mental Health Functioning
Few respondents (15.1%, n = 76) reported they were diagnosed with mental
health conditions in the months preceding the loss. The most common diagnoses
were depressive (11.7%, n = 59) and anxiety (7.8%, n = 39) disorders. Three
(0.6%) respondents reported attention-deficit hyperactivity disorder, and six
(1.2%) reported a diagnosis of bipolar disorder. Sixty-three participants (12.5%)
were taking psychiatric medications at the time of death.
The HSCL-25 was completed by 478 participants, resulting in a Cronbach’s
alpha of 0.96. The mean average score on the overall instrument was 1.96 (SD =
0.71); for the depression subscale, M = 2.10, SD = 0.77; and for the anxiety
subscale, M = 1.74, SD = 0.73. More than one-half (51.3%, n = 258) scored > 1.75
on the overall HSCL-25, while 57.9% (n = 291) scored > 1.75 on the depression
subscale, and 36.6% (n = 184) scored > 1.75 on the anxiety subscale. Among participants £ 12 months post-loss who had not been diagnosed with a mental disorder
at the time of loss (n = 61), 69.2% (n = 45) were clinical cases according to the
overall scale, 73.8% (n = 48) by the depression subscale, and 52.3% (n = 34) on the
anxiety subscale (see Table 2).
Four-hundred-eighty-three respondents completed the IES-R and the
instrument exhibited a Cronbach’s alpha of 0.95. The mean average score was
1.46 (SD = .90) for the overall instrument; for the subscales measuring intrusion,
avoidance, and hyperarousal, subscale means were, respectively, 1.92 (SD = 1.03),
1.07 (SD = .86), and 1.37 (SD = 1.19). Less than half of respondents (42.3%, n =
213) scored above the clinical cut-off of 1.5. Respondents with years-since-loss of
£ 12 months who were not diagnosed with mental disorder at the time of loss had a
63.1% (n = 41) rate of clinical caseness per IES-R of 36.5% (n = 125).
In order to examine the potential impact of a previous mental health diagnosis
preceding the loss on our results, we performed an influence analysis (Rothman, &
Greenland, 1998) by removing respondents who reported they were previously

A PARENT’S TEARS / 193

diagnosed with a mental disorder (n = 76) and then re-performed these data
analyses. There was no clinically significant change in any of the results; they
changed only slightly (e.g., IES-R > 1.5 changed from 42.3% to 40.3%).
Roughly one-quarter of respondents (25.6%, n = 129) reported an increase in
drug and alcohol use after the death. The mean number of days in which these
participants drank alcohol in the past 30 days was 9.84 (SD = 9.04); for drinking
alcohol to intoxication, 4.25 days (SD = 4.43); and for the number of days in which
alcohol was self-identified as a problem 2.94 days (SD = 3.94). Only eight
participants (1.6%) were “considerably” or “extremely” bothered by their
alcohol problems in the last 30 days, and, similarly, ten respondents reported that
treatment for alcohol problems was “considerably” or “extremely” important to
them. Eighteen (3.6%) bereaved parents reported that they had used cannabis in
the last 30 days; other drug use, such as amphetamines, was rare, with < 5
respondents reporting such use. Only three people (0.6%) reported that they were
“considerably” or “extremely” troubled/bothered by a drug problem.
One-third of participants (33.8%, n = 170) reported a decline in their health
since the loss, while 46.9% (n = 236) reported no change and 13.01% (n = 66)
reported improvement. Those reporting a post-loss degradation in health status
named a range of problems. Weight gain was common (10.5%, n = 53), and, like
many of the reported health conditions, was often presented in the context of
intermingled physical health symptoms and mental distress. Many reported a relationship between demoralization and weight gain and that in the wake of the loss, it
was difficult to continue healthy lifestyle habits. For instance, “I’ve gained lots of
weight and basically don’t care too much about my health,” while another wrote,
“I have stopped caring for my body in any significant way. I do not exercise or
even leave the house very often, whereas I used to be very active. . . . I have
gained a great deal of weight. . . .” Another reported physical health symptom
was daytime fatigue (6.6%, n = 33). Typical responses were “I’ve become
lethargic and weak . . . I’m tired a lot” and “I am more much more tired and feel
worse (physically) on a daily basis.” It was common for participants to describe
varying clusters of health problems that had significant impact on their
quality-of-life. For instance, “Developed high blood pressure, lost then gained
weight, developed psoriasis, heart palpitations, [and] headaches” and “Have
developed stress-induced migraines, a heart condition, and panic attacks.” Some
participants reported their immune system seemed to worsen post-loss, with
frequent illness a problem.
Family Functioning
Respondents on the General Functioning subscale of the FAD reported an
overall mean score of 1.82 (SD = 0.58; N = 437), suggesting a reasonable
level of family functioning for this sample. However, 31.6% (n = 138) scored
above the 2.0 cutting score suggesting those respondents perceived their families’

213

138

Clinical Cases IES-R
> 1.5 total avg. score

FAD
> 2.0
31.6

42.3

[27.3, 36.2]

[37.0, 45.8]

[46.9, 55.7]
[53.4, 62.2]
[32.4, 41.0]

7
5
3
2
3

0

22
25
15
10
15

3.75

4.0
3.9
4.0
4.0

4.00
4.00
4.00

Max

22.06 (6.58)
14.22 (5.07)
10.10 (3.36)
5.74 (2.74)
10.0 (3.16)

1.82 (0.58)

1.46 (0.90)
1.07 (0.85)
1.92 (1.03)
1.37 (1.19)

1.96 (0.71)
2.10 (0.77)
1.74 (0.73)

M (SD)

1.54]
1.15]
2.00]
1.48]

[21.46, 22.65]
[13.74, 14.68]
[9.71, 10.32]
[ 5.5, 5.98]
[10.21, 10.79]

[1.77, 1.87]

[1.38,
[1.00,
[1.82,
[1.27,

[1.90, 2.02]
[2.03, 2.17]
[1.68, 1.81]

95% CI

Note: All percentages are calculated using the denominator of N = 503, the total sample, except for the FAD, which uses N = 437, the number of
responders who lived in a family situation.

PTGI
Relating to Others
New Possibilities
Personal Strength
Spiritual Change
Appreciation of Life

258
291
184

Clinical Cases HSCL-25
> 1.75 Overall
> 1.75 Depression
> 1.75 Anxiety
51.3
57.9
36.6

Min

0
0
0
0

95% CI

Average Score-IES-R
Overall
Avoidance
Intrusion
Hyperarousal

%
1.00
1.00
1.00

n

Average Score-HSCL
Overall
Depression
Anxiety

Instrument

Table 2. Mental Health and Family Functioning Measures

194 / CACCIATORE ET AL.

A PARENT’S TEARS / 195

functioning to fall within the clinical range (Table 2). The responses to the
open-ended items (n = 420) regarding post-loss family relationships were consistent with the FAD results. The responses suggest that while many families were
functioning well, others have experienced negative effects to their relationships as
a result of the loss. Many respondents (45.5%, n = 191) reported the death of child
caused a negative effect on their relationships with their spouse or partner. One
participant stated, “My loss was heartbreaking for all of us, and we all dealt with it
on our own and not together. I feel it has caused distance between [me and] my
spouse.” Similarly, another stated, “I feel that losing our child has cost us
everything. Not only the incredible boy that we lost suddenly without warning, but
our roles in our marriage, in our family. I feel we are strangers now, we don’t know
how to be close.” Some responses suggested that couples had difficulty in
communication and intimacy after the loss.
Others spoke about the challenges the death brought about in their relationships
with their children. For example, one mother stated, “I tried to become closer to
them [the children], but it is difficult. I find I smother them instead. I have found
that we aren’t truly together. That each person in the world is truly alone and we
just happen to share space with other individuals.” Another parent shared, “I have
been somewhat leery of getting closer to my boys for the thought they, I may lose
them as well. I have since gotten over this feeling and now find I am extremely
overprotective of them for fear that something may happen to them.” The
comments related to increased challenges in parent-child relationships described
an escalation in overprotectiveness and the fear of losing another child as hindering closeness with their current children.
Despite these negative appraisals, many respondents (54.5%, n = 229) reported
that their relationships improved post-loss. For example, one respondent stated,
“My relationship with my husband has only gotten stronger. We communicate
more openly and have come to realize how important we are to one another.”
Similarly, another wife explained, “It has brought my husband and I closer,
because I feel that he’s the only other person in the whole wide world EVER who
will know EXACTLY how I feel.” A father agreed stating, “My son who died was
the closest companion that I have ever had, next to my wife. I believe that the death
of my son brought us [husband and wife] closer together.” Thus, while the loss
caused reduced communication and intimacy for some, for these couples, the
death of their child seemed to have the opposite effect. Several respondents also
noted improvements in parent-child relationships, noting an increased
appreciation for their children. One wrote, “My relationship with my children is
good. I feel closer to them. I feel a new sense of connection to my living children
since her death.” Similarly, another explained, “I value them [the children] more. I
realized how much of a miracle my first son is after losing my second. I cherish the
time spent with my family and try to make the most of it.” Overall, findings
suggest most of the families in this study perceived positive family functioning
despite the death of a child. Yet, a substantial subsample did identify negative

196 / CACCIATORE ET AL.

effects to partnered and parent-child relationships which they attribute to the death
of their child.
Resilience and Coping
The mean scores on the PTGI subscales are as follows: relating to others (M =
22.06, SD = 6.6); new possibilities (M = 14.22, SD = 5.07); personal strength
(M = 10.02, SD = 3.36); spiritual change (M = 5.74, SD = 2.64); and appreciation
for life (M = 10.0, SD = 3.16). These scores suggest that, overall, respondents
experienced positive changes after loss to a moderate degree. The highest mean
score was on the appreciation of life subscale, suggesting that bereavement can
bring about a positive shift in perspective for some.
Despite the challenges these parents faced, evidence of resilience and efforts to
engage in positive coping activities were also commonly discussed. Help-seeking
behaviors in this sample were notable. Sixty-three percent of the sample indicated
that they were aware of grief support groups offered in their communities, while
34.4% were unaware of them. Many participants (42.3%) indicated that they
attended a grief support group. Reported attendance at support groups ranged from
1 to 200 sessions, with a sample mean of 9.4 sessions (SD = 22.15).
Most bereaved parents in this study (53.5%; n = 269) reported attending at least
one counseling session with a trained mental health professional. Reported counseling attendance ranged from 1 to 250 sessions with a sample mean of 20.8 sessions (SD = 40.18). Participants were asked to rate the “helpfulness” of counseling
on a 6-point scale where higher scores indicated more helpfulness. The mean score
of this item was 4.49 (SD = 1.42), with 56.7% of respondents indicating that counseling was “helpful” or “very helpful,” and only 17.2% indicating that it was “very
unhelpful” or “unhelpful.”
In addition to support groups and counseling, volunteerism and other prosocial
behaviors are activities that can help others while also helping oneself. In this
study, 182 respondents (36.2%) indicated that they had volunteered in their communities before their loss, while 225 (44.7%) reported volunteering after their
loss. Specific volunteer activities included leading a support group, providing peer
counseling or support, public speaking, community event organizing, fundraising,
and/or financial contributions to help others.
Bivariate Correlations
Table 3 lists bivariate correlations between primary study variables. There was a
statistically significant correlation between years-since-loss and both IES-R and
HSCL-25 scores. These correlations were all negative; for instance, –0.24 for the
total IES-R (p < .05). This equates to a r2 value of 0.04, or 4% of variance in
HSCL-25 scores explained by years since-loss. FAD scores had almost no correlation with years-since loss, (r = – 0.01, p > .05) while the PTGI was positively
correlated. The FAD was positively correlated with the IES-R and HSCL-25.

—
–.24*
.31*
–.19*
–.20*
–.20*
–.21*
–.14*
–.01
.19*
—
.92*
.81*
.93*
.80*
.78*
.72*
.23*
–.16

IES-R

—
.57*
.84*
.74*
.73*
.64*
.16*
–.11*

IES-R-I

—
.64*
.56*
.55*
.51*
.22*
–.15*

IES-R-A

—
.84*
.80*
.77*
.25*
–.16*

IES-R-H

—
.97*
.91*
.29*
–.17*

HSCL

—
.78*
.32*
–.23*

HSCL-D

—
.20*
–.05

HSCL-A

—
.26*

FAD

—

PTGI

Note: N ranges from 459-483. YSL-Log = Logarithmic transformation of years-since-loss; IES-R-I = intrusion subscale; IES-R-A = avoidance
subscale; IES-R-H = hyperarousal subscale; HSCL-D = depression subscale; HSCL-A = anxiety subscale. For the General Functioning Subscale
of the FAD, we used the final score, where lower numbers indicate better family functioning; for all other measures, we used scale and subscale
totals.
*p < .05 (2-tailed).

YSL-Log
IES-R
IES-R-I
IES-R-A
IES-R-H
HSCL
HSCL-D
HSCL-A
FAD
PTGI

YSL-Log

Table 3. Pearson Correlations Between Primary Study Variables

A PARENT’S TEARS
/ 197

198 / CACCIATORE ET AL.

DISCUSSION
We have described in detail a sample of bereaved parents (N = 503) who participate in an online support community. The sample is overwhelmingly white and
female, not uncommon in bereavement research (e.g., Cacciatore, 2007) especially when the sample is largely composed of women who suffered a baby’s death
to stillbirth as in our data (n = 215, 42.7%). These online bereaved parents, an
understudied population, have been described across a broad array of biopsychosocial variables, including health, mental health, and family functioning.
One strength of our study is the use of validated mental measures. The HSCL-25
and IES-R results indicate that although few were diagnosed with mental disorder
prior to their loss, many in this sample have enduring and significant emotional
and mental health difficulties. The mean HSCL-25 overall score was 1.96 (SD =
0.71), with higher scores on the depression subscale (M = 2.1, SD = 0.77) than on
the anxiety subscale (M = 1.74, SD = 0.73). The majority of the sample (n = 291,
57.9%) scored above the clinical cut-off of > 1.75 on the depression subscale, and
assuming that the HSCL-25 performs adequately as a screening tool (Sandanger
et al., 1998) this indicates a high prevalence of depression in this sample using this
standardized measure. By comparison, the 12-month prevalence of depression is
~8.3% in the general population of the United States (Kessler, Bimbaum, Shahly,
Bromet, Hwang, McLaughlin, et al., 2010).
In previous studies (Zisook, Chentsova-Dutton, & Schuchter, 1998), depressive
symptoms co-occurred with anxiety and trauma. A substantial minority (42.3%,
n = 213) scored > 1.5 on the IES-R. Although the IES-R is not a measure of
DSM-IV defined PTSD, this cut-off score has been found to perform reasonably
well diagnostically, and IES-R scores are highly correlated with the PTSD
Checklist (0.84; see Creamer, et al., 2003). In a Japanese study, a slightly lower
cut-off score (1.36) was validated through clinical interviews (Asukai, Kato,
Kawamura, Kim, Yamamoto, Kishimoto, et al., 2002), and it is possible that using
a cut-off of 1.5 led to reduced sensitivity in our study, as 21 cases (4.1%) had
IES-R average scores between 1.36 and 1.49. Despite this arguably conservative
approach, a substantial proportion of respondents were clinical cases according to
their IES-R scores. This high prevalence exceeds that found in studies of burn
victims (Palmu, Suominen, Vuola, & Isometsa, 2010) and ICU survivors (Griffith,
Fortune, Barber, & Young, 2007, p. 1514). It is also higher than the 27.7% rate of
PTSD found in mothers who lost a child to violent death at 5-year follow-up
(Murphy, Johnson, Chung, & Beaton, 2003) and a study of low-income mothers
impacted by Hurricane Katrina which found that 32.7% had IES-R scores > 1.5 at
~4-year follow up.
We examined a subsample of respondents without a mental health diagnosis at
the time of loss at £ 1 year since the loss. On the HSCL depression subscale, 73.8%
were clinical cases, while on the IES-R, 63.1% exceeded the clinical-cut off. The
12-month incidence rates of PTSD and depression in the general population are

A PARENT’S TEARS / 199

estimated to be 3.5% and 1.75-2.5%, respectively (Breslau, 2009; Bromberger,
Kravtiz, Matthews, Yuok, Brown, & Feng, 2009). Our estimate is not a 12-month
incidence proportion, as it is based on a cross-sectional assessment. However, it
does suggest that, in our sample, the incidence proportion is no lower than what we
have reported. While the sensitivity and specificity of the instrumentation should
be considered, these data confirm that child death often leads to significant distress
in both the short and long-term.
The relationship between traumatic bereavement and negative mental health
consequences is a long-standing finding in the bereavement literature. Leahy
(1992-1993) found that bereaved mothers scored significantly higher on depression measures than did bereaved adult daughters or widows, reporting symptoms
such as sadness, sense of failure, guilt, suicidal ideation, self-image changes, work
difficulty, and negative self-appraisal. Sanders’ (1979-1980) found that bereaved
parents experience intense and enduring somatic, psychological, and social consequences. The death of a child is a disaster, and human reactions to disaster are very
complex, with emotional, cognitive, and biological effects. Shalev (2000) suggests that disaster, coupled with separation from the object of love and affection,
can incite features of depressed mood, intrusive thoughts, pining, numbing, and
severe separation distress. These features, as exemplified in the data reported here,
are congruent with previous studies, suggesting that for many bereaved parents,
emotional, somatic, and behavioral symptoms may endure long-term (Arnold,
Gemma, & Cushman, 2005; Gillis, Moore, & Martinson, 1997; McCarthy, Clarke,
Ting, Conroy, Anderson, & Heath, 2010). At the same time, we did find a correlation between years-since-loss and mental health status, although the clinical significance could be argued, since the coefficient of non-determination was ³ 0.95.
A clinically significant impact of time upon anxiety and depressive symptoms in
parental bereavement has been noted; in a population-based follow-up study,
Kreicbergs, Valdimarsdottir, Onelov, Henter, and Steineck (2004) found the risk
of anxiety and depression was equivalent to controls at 7-9 years post-loss.
The high prevalence of depression and anxiety in this sample based on these
measures should be contextualized. Surveying members of an online support
group entails an unavoidable selection bias (Rothman & Greenland, 1998). Individuals who are struggling with mental health issues following bereavement are
potentially more likely to participate in an online support community.
Interestingly, the prevalence of depression in our sample (73.8%) is similar to that
self-reported by participants in an online support group for suicide survivors
(72.1%, see Feigelman et al., 2008). Our mental health findings document the
difficult struggles of an important and understudied subpopulation, but should not
be generalized.
A minority of participants self-reported that their health had declined since the
loss of their child (33.8%, n = 170), while most reported either level or improved
health (60.0%, n = 302). These results should be interpreted in light of the
fact that the mean time-since-loss was 4.31 years, with a mean respondent age of

200 / CACCIATORE ET AL.

36.69 years. Health status often declines with age and the mean age of participants
indicates many respondents moving toward middle age. Thus, it is not possible
from these data to isolate a casual impact of child loss upon health status.
However, many of the respondents reported clinically important health problems
subsequent to their loss, often co-occurring with anxiety, trauma, and depression.
Researchers continue to document an association between psychological trauma
and poor physical health (e.g., Boscarino, 2008), and it is likely that this association exists in our sample as well.
One-quarter of participants reported increasing their alcohol and/or drug use
after the loss; however, with a mean of 4.31 years-since-loss, it appears many such
increases were self-limiting in that few current substance abuse problems were
reported. However, the subsample that reported increasing drug/alcohol use drank
to intoxication 4.25 days a month; surveys of women in the general population find
that only 1.73% of white women drink to intoxication > 3 times a month (Caetano,
Baruah, Ramisetty-Mikler, & Ebama, 2010). This suggests that alcohol consumption among bereaved parents is perhaps a matter of clinical concern. Indeed,
self-medication with alcohol among traumatized individuals is a well-recognized
phenomenon (McFarlane, 1998), with potential ramifications for both physical
and mental health.
Findings related to family functioning are consistent with the family resilience
literature as they suggest some families are able to cope effectively and even
grow stronger with this devastating loss while others continue to struggle (Lietz,
2012; Walsh, 2002). Although the death of a child represents one of the most
painful experiences a family can have, this study demonstrates variability in functioning. While some families are clearly struggling, as indicated in the open-ended
responses, many families are able to mitigate the risks. Family resilience literature
suggests this variability can be explained by the presence of protective factors that
allow family systems to cope effectively. Previous research has identified social
support, spirituality, appraisal, flexibility, humor, and communication as protective factors that can explain how some families survive trauma and loss better than
others (DeFrain & Asay, 2007; Lietz, 2007, 2012; Walsh, 1996, 2002). Further
research should test models of family resilience to identify family strengths
specifically indicated for bereaved families.
When taken together, our data demonstrate the significant and frequently enduring negative impact of losing a child. What is striking is that this has occurred to
this sample in the context of having access to resources that are ideally of help.
Respondents were well-educated, white, middle-class women with good incomes
and sufficient health insurance, who seek and receive help through support groups
and therapists. More than one-half of respondents reported benefits of counseling.
Our results may indicate that our current system of care for and understanding of
bereaved parents is inadequate, and that providers could be doing more to help
grieving parents cope. There are no easy remedies, but problem-solving efforts
should include addressing macro-level problems in mental health (Gomory,

A PARENT’S TEARS / 201

Wong, Cohen, & Lacasse, 2011), improving clinical care in the immediate wake of
child death (Cacciatore, 2010), and improved education and training for aspiring
clinicians (Kirchberg & Neimeyer, 1991; Wass, 2004).
We are concerned that these data could lead to systemic pathologizing of
bereaved parents. The common degree of distress may be viewed by some as
indicative of a mental disorder, or, alternatively, the normal and common expressions of a profoundly traumatic and life altering loss (Cacciatore & Thieleman,
2012). We favor the latter interpretation and believe critical thinking regarding
diagnosis is earnestly necessary—indeed, the frequency and intensity of reported
distress raises the question of what is “normal” for a bereaved parent. We prefer
these results not be interpreted as a call for more aggressive psychiatric diagnosis
or drug treatment, the utility of which is questionable (Bui, Nadal-Vicens, &
Simon, 2012).
As with any study, interpretation of these data should include a consideration of
our limitations. The measures utilized for this study are self-reports and subject to
the potential shortfalls of this type of data acquisition (Rust & Golombok, 1999).
We used self-report instrumentation, but in-person clinical assessment is preferable, if much more costly. Non-response error is a potential concern; if nonrespondents differ substantially on the variables of interest, our estimates are
biased. Our survey was also cross-sectional; recall may be compromised for some
respondents who experienced the death of a child many years ago. The sample is
demographically homogenous and thus future research should focus on ethnic,
religious, and socioeconomic diversity in the respondents. Despite the aforementioned methodological limitations, this study adds to what we know about
bereaved parents participating in an online support community, and perhaps offers
insight into the experiences of other bereaved parents who suffer enduring distress
after the death of their child.
ACKNOWLEDGMENTS
The authors acknowledge Andrea N. Cimino for assistance with data management. This study is dedicated to the memory of the many children who died young
and their parents who mourn in their absence.
REFERENCES
Arnold, J., Gemma, B. P., & Cushman, L. F. (2005). Exploring parental grief: Combining
quantitative and qualitative measures. Archives of Psychiatric Nursing, 19(6),
245-255.
Asukai, N., Kato, H., Kawamura, N., Kim, Y., Yamamoto, K., Kishimoto, J., et al. (2002).
Reliability and validity of the Japanese language version of the Impact of Event Scale–
Revised. The Journal of Nervous & Mental Disease, 190(3), 175-182.

202 / CACCIATORE ET AL.

Berk, R., & Freedman, D. A. (2009). Statistical assumptions as empirical commitments. In
D. Collier, J. S. Sekhon, & P. B. Stark (Eds.), Statistical models and causal inference:
A dialogue with the social sciences. New York, NY: Cambridge University.
Birnbaum, L. K., Stewart, B. J., & Phillips, D. S. (1996). Health status of bereaved parents.
Nursing Research, 45, 105-109.
Boscarino, J. A. (2008). A prospective study of PTSD and early-age heart disease mortality
among Vietnam veterans: Implications for surveillance and prevention. Psychosomatic
Medicine, 70(6), 668-676.
Boyle, F. M., Vance, J. C., Najman, J. M., & Thearle, M. J. (1996). The mental health
impact of stillbirth, neonatal death or SIDS: Prevalence and patterns of distress among
mothers. Social Science and Medicine, 43(8), 1273-1282.
Breslau, N. (2009). The epidemiology of trauma, PTSD, and other posttraumatic disorders.
Trauma, Violence and Abuse, 10(3), 198-210.
Bromberger, J. T., Kravtiz, H. M., Matthews, K., Youk, A., Brown, C., & Feng, W. (2009).
Predictors of first lifetime episodes of major depression in midlife women. Psychological Medicine, 39(1), 55-64.
Bui, B., Nadal-Vicens, B., & Simon, N. M. (2012). Pharmacological approaches to the
treatment of complicated grief. Dialogues in Clinical Neuroscience, 14, 149-157.
Cacciatore, J. (2007). Effects of support groups on post traumatic stress response in women
experiencing stillbirth. Omega: Journal of Death and Dying, 55(1), 71-90.
Cacciatore, J. (2010). Stillbirth: Patient-centered psychosocial care. Clinical Obstetrics
and Gynecology, 53(3), 691-699.
Cacciatore, J., & Thieleman, K. (2012). Pharmacological treatment following traumatic
grief. Journal of Loss and Trauma, 17(6), 557-579.
Caetano, R., Baruah, J., Ramisetty-Mikler, R., & Ebama, M. S. (2010). Sociodemographic
predictors of pattern and volume of alcohol consumption across Hispanics, Blacks, and
Whites: 10-year trend (1992-2002). Alcoholism: Clinical and Experimental Research,
34(10), 1728-1792.
Creamer, M., Bell, R., & Faila, S. (2003). Psychometric properties of the Impact of Event
Scale- Revised. Behaviour Research and Therapy, 41, 1489-1496.
DeFrain, J., & Asay, S. (2007). Strong families around the world: The family strengths
perspective. New York, NY: Haworth Press.
Derogatis, L. R., Lipman, R. S., Rickels, K., Uhlenluth, E. H., & Covi, L. (1974). The Hopkins Symptoms Checklist (HSCL): A self-report symptom inventory. Behavioral
Science, 19, 1-15.
Epstein, N. B., Bishop, D. S., & Levin, S. (1978). The McMaster Model of family
functioning. Journal of Marital and Family Therapy, 4(4), 19-31.
Eun-Ok, I., Chee, W., Lim, H. J., Guevara, E., Tsai, H., Clark, M., et al. (2007). Characteristics of cancer patients in internet cancer support groups. Computer Informed
Nursing, 25(6), 334-343.
Feigelman, W., Gorman, B. S., Beal, K. C., & Jordan, J. R. (2008). Internet support groups
for suicide survivors: A new mode for gaining bereavement assistance. Omega: Journal of Death and Dying, 57(3), 217-243.
Fox, M., Cacciatore, J., & Lacasse, J. (in press). Child death in the U.S.: Productivity and
the economic burden of parental grief. Death Studies.
Gillis, L., Moore, I. M., & Martinson, C. (1997). Measuring parental grief after childhood
cancer: Potential use of the SCL-90R. Death Studies, 21(3), 277-291.

A PARENT’S TEARS / 203

Gomory, T., Wong, S. E., Cohen, D., & Lacasse, J. R. (2011). Clinical social work and the
biomedical industrial complex. Journal of Sociology and Social Welfare, 38(4),
135-165.
Griffith, J., Fortune, G., Barber, V., & Young, J. D. (2007). The prevalence of post
traumatic disorder in survivors of ICU treatment. Intensive Care Medicine, 33,
1506-1518.
Kessler, R. C., Bimbaum, H., Shahly, V., Bromet, E., Hwang, I., McLaughlin, K. A., et al.
(2010). Age differences in the prevalence and comorbidity of DSM-IV major depressive episodes. Depression and Anxiety, 27(4), 351-364.
Kreicbergs, U., Valdimarsdottir, U., Onelov, E., Henter, J. I., & Steineck, G. (2004).
Anxiety and depression in parents 4-9 years after the loss of a child owing to a malignancy: A population-based follow-up. Psychological Medicine, 34(8), 1431-1441.
Kirchberg, T. M., & Neimeyer, R. A. (1991). Reactions of beginning counselors to situations prolonging death and dying. Death Studies, 15(6), 603-610.
Lacasse, J., & Cacciatore, J. (in press). Prescribing of psychiatric medications to bereaved
parents following peri/neonatal death. Death Studies.
Leahy, J. (1992-1993). A comparison of depression in women bereaved of a child, spouse,
or parent. Omega: Journal of Death and Dying, 26(3), 207-217.
Li, J., Johansen, S. P., & Olsen, J. (2002). Stroke in parents who have lost a child: A
nationwide follow-up study in Denmark. Neuroepidemiology, 22, 211-216.
Li, J., Laursen, T. M., Precht, D. H., Olsen, J., & Mortensen, P. B. (2005). Hospitalization
for mental illness among parents after the death of a child. New England Journal of
Medicine, 352, 1190-1196.
Li, J., Precht, D. H., Mortensen, P. B., & Olsen, J. (2003). Mortality in parents after the
death of a child in Denmark: A nationwide followup study. Lancet, 361(9355), 363.
Lietz, C. A. (2007). Uncovering stories of family resilience: A mixed methods study of
resilient families, part 2. Families in Society, 88(1), 147-155.
Lietz, C. A. (2012). Family resilience in the context of high risk situations. In D. Becvar
(Ed.), Handbook of family resilience. New York, NY: Springer Publishing.
Lohan, J. A., & Murphy, S. A. (2002). Family functioning and family typology after an
adolescent or young adult’s sudden violent death. Journal of Family Nursing, 8(32),
32-49.
McCarthy, M. C., Clarke, N. E., Ting, C. L., Conroy, R., Anderson, V. A., & Heath, J. A.
(2010). Prevalence and predictors of parental grief and depression after the death of a
child from cancer. Journal of Palliative Medicine, 13(11), 1321-1326.
McFarlane, A. C. (1998). Epidemiological evidence about the relationship between PTSD
and alcohol use: The nature of the association. Addictive Behaviors, 23(6), 813-825.
Miller, I., Ryan, C., Keitner, G., Bishop, D., & Epstein, N. B. (2000). The McMaster
approach to families: Theory, assessment, and research. Journal of Family Therapy,
22(2), 168-189.
Murphy, S. A., Johnson, L. C., Chung, I., & Beaton, R. D. (2003). The prevalence of PTSD
following violent death of a child and predictors of change 5 years later. Journal of
Traumatic Stress, 16(1), 17-25.
Murphy, S. A., Lohan, J., Braun, T., Johnson, L. C., Cain, K. C., & Beaton, R. D. (1999).
Parents’ health, health care utilization, and health behaviors following the violent
deaths of their 12- to 28-year old children: A prospective longitudinal analysis. Death
Studies, 23, 589-616.

204 / CACCIATORE ET AL.

Newcombe, R. G. (1998). Two-sided confidence intervals for the single proportion:
Comparison of seven methods. Statistics in Medicine, 17, 857-872.
Palmu, R., Suominen, K., Vuola, J., & Isometsa, E. (2010). Mental disorders after burn
injury: A prospective study. Burns, 37, 601-609.
Pigott, T. D. (2001). A review of methods for missing data. Educational Research and
Evaluation, 7, 353-383.
Qin, P., & Mortenson, P. B. (2003). The impact of parental status on the risk of completed
suicide. Archives of General Psychiatry, 60(8), 797-802.
Rogers, C. H., Floyd, F. J., Seltzer, M. M., Greenberg, J., & Hong, J. (2008). Long-term
effects of the death of a child on parents’ adjustment in midlife. Journal of Family
Psychology, 22(2), 203-211.
Rosen, C. S., Henson, B. R., Finney, J. W., & Moos, R. H. (2000). Consistency of
self-administered and interview-based Addiction Severity Index composite scores.
Addiction, 95(3), 419-425.
Rostila, M., Saarela, J., & Kawachi, I. (2011). Mortality in parents following the death of a
child: A nationwide follow-up study from Sweden. Journal of Epidemiology and
Community Health. doi:10.1136/jech-2011-200339
Rothman, K. J. (2003). Epidemiology: An introduction. New York, NY: Oxford University.
Rothman, K. J., & Greenland, S. (1998). Modern epidemiology (2nd ed.).
Lippincott-Raven: Philadelphia, PA.
Rust, J., & Golombok, S. (1999). Modern psychometrics: The science of psychological
assessment, 2nd edition. London, UK: Routledge.
Samuelson, K. A. M., Lundberg, D., & Fridlund, B. (2007). Stressful memories and psychological distress in adult mechanically ventilated intensive care patients—A
2-month follow-up study. Acta Anaesthesiologica Scandinavica, 51, 671-678.
Sandanger, I., Moum, T., Ingebrigtsen, G., Dalgard, O. S., Sorensen, T., & Bruusgard, D.
(1998). Concordance between symptom screening and diagnostic procedure: Social
Psychiatry and Psychiatric Epidemiology, 33, 345-354.
Sanders, C. (1979-1980). A comparison of adult bereavement in the death of a spouse,
parent, and child. Omega: Journal of Death and Dying, 10(4), 303-321.
Schwab, R. (1998). A child’s death and divorce: Dispelling the myth. Death Studies, 22,
445-468.
Shalev, A. (2000). Biological responses to disaster. Psychiatric Quarterly, 7(3), 277-288.
Shreffler, K., Hill, T., & Cacciatore, J. (2012). The impact of infertility, miscarriage,
stillbirth and child death on marital dissolution. Journal of Divorce and Remarriage,
53(2), 91-107.
Swartwood, R. M., Veach, P. M., Kuhne, J., Lee, H. K., & Hi, K. (2011). Surviving grief:
An analysis of the exchange of hope in online grief communities. Omega: Journal of
Death and Dying, 63(2), 161-181.
Tedeschi, R., & Calhoun, L. (1996). The Posttraumatic Growth Inventory: Measuring the
positive legacy of trauma. Journal of Traumatic Stress, 9(3), 455-471.
Thielman, K., & Cacciatore, J. (2013). The DSM-5 and the bereavement exclusion: A call
for critical evaluation. Social Work, 58(3), 277-280.
Turton, P., Evans, C., & Hughes, P. (2009). Long-term psychosocial sequelae of stillbirth:
Phase II of a nested case-control study. Archives of Women’s Mental Health, 12(1),
35-41.

A PARENT’S TEARS / 205

Tutty, L. M. (1995). Theoretical and practical issues in selecting a measure of family
functioning. Research on Social Work Practice, 5(1), 80-106.
Walsh, F. (1996). The concept of family resilience. Family Process, 35(3), 261-281.
Walsh, F. (2002). A family resilience framework: Innovative practice applications. Family
Relations, 51, 130-137.
Wass, H. (2004). A perspective on the current state of death education. Death Studies, 28,
289-304.
Weiss, D. S. ( 2004). The impact of Event Scale-Revised. In J. P. Wilson & T. M. Keane
(Eds.), Assessing psychological trauma and PTSD (2nd ed., pp. 168-189). New York,
NY: Guilford.
Wilson, E. B. (1927). Probable inference, the law of succession, and statistical inference.
Journal of the American Statistical Association, 22, 209-212.
Zisook, S., Chentsova-Dutton, Y., Schuchter, S. R. (1998). PTSD following bereavement.
Annals of Clinical Psychiatry, 10(4), 157-163.

Direct reprint requests to:
Joanne Cacciatore
School of Social Work
College of Public Programs
Arizona State University
411 N. Central Avenue
Phoenix, AZ 85004-0689
e-mail: joanne.cacciatore@asu.edu

Copyright of Omega: Journal of Death & Dying is the property of Baywood Publishing
Company, Inc. and its content may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.