Content uploaded by Sarah Welchans Craun
Author content
All content in this area was uploaded by Sarah Welchans Craun on Jun 01, 2022
Content may be subject to copyright.
https://doi.org/10.1177/10887679211013071
Homicide Studies
1 –16
© 2021 SAGE Publications
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/10887679211013071
journals.sagepub.com/home/hsx
Article
Homicide or Suicide: How
Nudity Factors into This
Determination
Sarah W. Craun1, Leah Tanner1, Victoria Clausen1,
Melissa A. Merola2, Leonard Opanashuk1,
and Timothy G. Keel1
Abstract
Anecdotal reports of deceased celebrities being found nude abound, yet research
is lacking regarding the frequency of nudity at death. Moreover, it is unknown if
nudity at the time of death is a useful investigative clue or a distracting non-factor
in equivocal death cases. This study used data from 119,145 homicides and suicides
reported to the Centers for Disease Control to explore victim nudity, prior life
stressors, and demographics on the likelihood of a death being a homicide or a
suicide. Logistic regression results indicate that a female victim being found nude is
a strong indicator of homicide.
Keywords
equivocal deaths, investigation, policing, crime scene, policing, nudity, suicides
Determining whether a violent death is the result of homicide or suicide can be
challenging for medico-legal death investigators.1 The motivation to end a life, may
be clear, but whether the death was self-inflicted or perpetrated by another may not
be. Moreover, the existence of multiple motivations on the part of both the victim
and possible offenders may obfuscate the true nature of the manner of death. This
can be further confounded by characteristics of the death scene that defy expecta-
tions-such as when the victim is found nude and nudity does not seem congruent
1Federal Bureau of Investigation, Quantico, VA, USA
2Bureau of Alcohol, Tobacco, Firearms and Explosives, Huntsville, AL, USA
Corresponding Author:
Leah Tanner, CIRG/IOSS/NCAVC, Federal Bureau of Investigation, Quantico1 Range Road, VA 22135, USA.
Email: ltanner@fbi.gov
1013071HSXXXX10.1177/10887679211013071Homicide StudiesCraun et al.
research-article2021
2 Homicide Studies 00(0)
with the rest of the scene. A review of popular press found several instances of
equivocal death cases involving a nude victim. For example, in one such case a
woman was found nude, hands bound, and hanging from a balcony. The death was
determined to be suicide, yet the family went on to win a wrongful death suit against
the brother of the victim’s boyfriend (Pelisek, 2019). In an article titled “Is It
Common for Women to Commit Suicide in the Nude?” the woman’s family argues
she would have never committed suicide, let alone commit suicide naked (Hawkins,
2019). In another case, a woman was found deceased, nude, and with a clothesline
around her neck. It was initially classified as a suicide, changed to undetermined,
and then finally classified as a homicide years later (Saavedra, 2020). As seen by
the aforementioned case anecdotes, the presence of nudity is often not congruent
with other crime scene characteristics and is therefore worthy of empirical study.
The current study aims to explore the possible relationship between decedent nudity
and homicides to determine whether nudity can serve as a distinguishing factor for
medico-legal death investigators in differentiating between the two manners of
death, which may be helpful in equivocal death investigations.
Literature Review
Lacks et al. (2008) define an undetermined death, also known as an equivocal death,
as “any death where there is uncertainty as to the circumstances of the death, creating
the relatively equal probability of two or more possible manners of death” (p. 150). In
an equivocal death, the cause of death may be known, but the manner (natural death,
accident, suicide, homicide, or undetermined) is not (Davis & Ogloff, 2014). For
example, a medical examiner may determine that the cause of death is from a drug
overdose, but label the death as undetermined, as the manner of death could not be
concluded to be an accident or suicide. This occurs more frequently than one may
expect. The Centers for Disease Control and Prevention (CDC) reported that 9.5%
were listed as undetermined, meaning that the evidence indicating that one manner of
death was not more compelling than another (Jack et al., 2018).
Victim and Crime Scene Characteristics in Equivocal Death Analyses
In reviewing suicide and undetermined deaths in Sweden, Lindqvist and Gustafsson
(2002) found that alcohol-related deaths were more commonly classified as unde-
termined deaths. Researchers, using samples from across the globe, have also
found that those who completed suicides were more likely to have substance abuse
issues than people in the general population (Cavanagh et al., 2003; Cheng et al.,
2000).
Determining that a suicide occurred can be difficult because “it is rare for there to
be clear objective evidence that the deceased took their own life” (Canter et al., 2004,
p. 234). Cases are clearer when there is suicidal communication, either verbally or
through a note, as researchers found that previous suicidal communication was highly
predictive of a death being classified as a suicide (Lindqvist & Gustafsson, 2002).
Craun et al. 3
In fact, previous research finds that only 34% of decedents who committed suicide left
a note (Safe States Alliance, 2017). In the absence of such objective evidence investi-
gators may turn to examination of the life events preceding the victim’s death.
However, some stressful life situations that are risk factors for suicide are also risk
factors for victimization and homicide. For example, intimate partner problems lead-
ing to separating are risk factors for homicide, yet at the same time intimate partner
problems were also in the lives of 43% of people who committed suicide (Brown &
Seals, 2019; Johnson & Hotton, 2003).
Conducting an Equivocal Death Analysis
To start, medico-legal death investigators review the injury patterns observed on the
deceased to help determine the manner of death. For instance, if the deceased was
wounded with a knife, the directionality and axes of the wounds have been found to be
a discerning factor in determining homicides versus suicides, and superficial stab
wounds and clothing lacerations are also considered in suicide and homicide determi-
nations (De-Giorgio et al., 2015; Terranova et al., 2020). Similarly, in the case of
homicide by gunshot, the type of gun used, the location and number of the injuries to
the body, the direction of the gunshot, and if it was a contact shot, are all utilized in the
determination of suicide or homicide (Cave et al., 2014; Molina & DiMaio, 2008;
Molina et al., 2007, 2013a, 2013b). Observable characteristics of both injury to the
body and conditions present in the context in which the victim was found are inher-
ently vital to determining cause of death.
The assessment of an attending doctor or coroner will have a role in the law enforce-
ment investigation, “a vision of the event can be strongly influenced by the initial
orientation of the medical examiner” with ambiguous death scenes (Visentin et al.,
2019, p. 148). For example, in equivocal death scenarios, Crepeau-Hobson (2010)
found that 31.5% of medical examiners stated that a lack of suicidal history was a
specific piece of information that would direct them toward an accident determination
as compared to a suicide determination for a deceased child.
In equivocal death analyzes, assessors consider the likelihood of various sce-
narios in determining the manner of death (Davis & Ogloff, 2014). To do this,
medico-legal death investigators review the body discovery scene, the autopsy, and
tests for substance use while also analyzing victimology—the behavioral and psy-
chological history of the deceased (Lacks et al., 2008). This can be done through a
psychological autopsy—a thorough review where death investigators and mental
health professionals determine the psychological state of the deceased (Dattilio &
Sadoff, 2011). Individuals surrounding the deceased are asked questions about the
deceased’s stress levels, coping abilities, personal problems, or crises, and behav-
ioral and emotional changes (Gavin & Rogers, 2006; Lacks et al., 2008). In com-
pleting an equivocal death analysis Davis and Ogloff (2014) urge investigators, in
the Checklist of Equivocal Death Information, to consider overarching categories
such as personal experiences, health history, family, relationships, and the death
scene. Within each of those categories the investigators then dive deeper to
4 Homicide Studies 00(0)
examine topics such as financial issues, mental health diagnoses, previous writings,
history of violence victimization or perpetration, among other topics that could
influence suicidal actions. The death investigation is made even more challenging
by the nature of suicide where “suicidal individuals may perform non-suicidal acts
and non-suicidal individuals may end by performing suicidal acts” (Silverman &
De Leo, 2016, p. 84). The information collected can be used to assist in the classi-
fication of the deceased’s manner of death (Davis & Ogloff, 2014).
Psychological autopsies have received a fair amount of criticism including the lack
of standardization in how and when they should be conducted (Canter, 2000; Dattilio
& Sadoff, 2011; Davis & Ogloff, 2014; Hjelmeland et al., 2012; Pouliot & De Leo,
2006). The challenge with psychological autopsies, however, is those most likely to be
interviewed for a deceased’s psychological autopsy may be unable or unwilling to
provide an accurate picture of the victim (Dattilio & Sadoff, 2011; Lacks et al., 2008).
Interviewees may provide inaccurate information for organic reasons, such as genuine
difficulty in the recollection of events due to the trauma of the death or the age of the
person being interviewed (Pouliot & De Leo, 2006). However, from a malicious stand-
point, those consulted may have reasons to conceal, distort, or suppress evidence about
the manner of death (Canter, 2000). These reasons could include their own biases
about suicide, financial/insurance incentives for a death not to be labeled a suicide, or
covering their own involvement in the death (Canter, 2000; Hjelmeland et al., 2012;
Pouliot & De Leo, 2006). Ferguson and Petherick (2016) found that in a sample of
sixteen homicides staged as suicide cases, 50% of the offenders were friends or non-
domestic family members, while the other 50% were cohabitating spouses or partners,
in other words, those who would most likely be consulted for a psychological autopsy.
Deception by family and friends of the deceased happens frequently with offenders in
equivocal death cases that are actually homicides (Ferguson & Sutherland, 2018;
Pettler, 2016). In a review of misclassified, unsolved homicide cases, the fact the man-
ner of death was equivocal allowed some offenders to avoid detection (Ferguson &
McKinley, 2019).
Moving beyond psychological autopsies, some argue equivocal death analyses
in general lack empirical research to support their use (Canter, 2000). While there
is limited empirical research, Lacks et al. (2008) notes that anecdotes are subse-
quently necessary to support the use of psychological autopsies. More recently,
Visentin et al. (2019) developed a five item scale for use in equivocal death analy-
ses: (a) statistical frequency of manner used to cause the death, (b) victim’s history
of mental illness, (c) circumstantial data, (d) if multiple modes of causing death
were used (i.e., cutting wrists and consuming pills), and (e) compatibility of the
injury pattern with suicide. Testing of the newly developed scale revealed that a
score of four or lower out of 10 successfully categorized 87% of the suicides as
such, while excluding all homicides and 97% of accidental deaths (Visentin et al.,
2019). Since psychological autopsies are somewhat criticized, and structured
equivocal death analyses is in its infancy, systematic study of additional distin-
guishing factors of homicides and suicides, such as nudity at death, becomes even
more important to consider.
Craun et al. 5
Nudity at Time of Death
One topic that has appeared within several equivocal death cases but has not been
empirically examined in the research literature is nudity at the time of death. As
Hawkins (2019) clearly notes, the academic literature provides no insight into the
frequency with which suicide is completed in the nude. Using three case examples,
Simon (2008) theorized why a person may commit suicide nude, but the article pro-
vided little evidence for use in equivocal death investigations. The theoretical basis for
suicidality does not provide much assistance either. Joiner’s (2005) theory of suicidal-
ity contained three main factors which paved the way for a successful suicide attempt.
He posited that perceived burdensomeness, a lack of belonging, and acquired capabil-
ity allowed a person to follow through on their suicidal thoughts. Feeling as though
one is a burden and lacks connections to others, as well as the capability to commit
suicide, does not have a clear-cut connection with the level of undress at death, how-
ever. Additionally, while Chan and Heide (2009), as well as Smith et al. (2011), men-
tion nudity within the context of sexual homicides, we were unable to find any research
literature that explores how frequent nudity is in homicides and suicides, which could
indicate if nudity might be a discerning factor in undetermined deaths.
Equivocal death analysis is made more challenging because the research has not
empirically examined victim and situational factors that those close to the victim see
as important in understanding whether a death is a homicide or a suicide, such as when
the family of one victim stated she would never commit suicide naked (Hawkins,
2019). The current research examines victim nudity, victim life stressors, and victim
demographics among suicide and homicide cases. These analyses fill the omission in
the literature by incorporating nudity at time of death, relationship and family issues,
personal crises, mental health issues, and a variety of other variables related to the
deceased. This allows investigators to integrate empirical findings into their decision-
making processes in equivocal death cases. Therefore, the present study will answer
the following research questions:
Research question #1: Is there a relationship between the victim nudity at the time
of death and the likelihood of the death being a homicide or a suicide?
Research question #2: Is there an interaction effect between victim gender and
nudity on the likelihood of the death being a homicide or a suicide?
Research question #3: If there is a relationship between nudity and the likelihood of
a death being a homicide as compared to a suicide, how does the strength of the
relationship compare to other variables such as personal problems or crises experi-
enced by the victim?
Method
Sample
The National Violent Death Registry System (NVDRS) is a state-based reporting sys-
tem maintained by the CDC that houses over 600 variables pertaining to homicides,
6 Homicide Studies 00(0)
suicides, and undetermined deaths from around the country.2 For this analysis, the
authors obtained the NVDRS’ data from 2014 through 2017. While the NVDRS rou-
tinely captured stressors within the deceased’s past, in 2014, the NVDRS began cap-
turing crisis variables, which are those stressors that happened to the deceased within
2 weeks of their death. This is an important benefit as the longitudinal status of stress-
ors and temporal changes of those stressors were a valued contribution of the work of
Davis’ Checklist of Equivocal Death Information (Davis & Ogloff, 2014). The total
sample for the present study was 119,145 violent deaths from 37 states, with 74.7% of
deaths classified as suicides and the remaining 25.3% classified as homicides.
Measures
The dependent binary variable was whether the violent death was classified by the
local jurisdiction as a suicide (0) or a homicide (1). To assist in equivocal death
investigations, the independent variables were situational variables that could be
measured both with the suicides and homicides. The independent variables were all
binary and coded as yes (1) or no/unknown/not marked (0). The NVDRS dataset did
not contain a variable to measure if the victim was found nude, so we searched the
included law enforcement and medical examiner narratives for the following words
(and their variations): nude, naked, unclothed, unclad, and undress. If a case was
flagged, the third author reviewed the narrative to ensure that the victim was found
either completely nude or partially nude (with genitals or breasts revealed). In con-
junction with nudity, we used the following binary variables contained within the
NVDRS to measure victim stressors: (a) if a mental health diagnosis was noted, (b)
if the deceased person experienced a mental health crisis, (c) if there was a family
relationship problem noted (with someone other than the intimate partner), (d) if
there was a family crisis noted (with someone other than an intimate partner), (e) if
the deceased had been the victim of interpersonal violence within the last month, (f)
if the deceased had an alcohol or drug problem, (g) if the deceased had a drug/alco-
hol crisis, and (h) if an argument led to the deceased’s death. Crises were classified
as such if they were a severe problem in the 2 weeks prior to death, while problems
extending beyond two-weeks prior to death were considered more long term. Also,
an argument that led to death could have been coded yes for multiple reasons. For
example, the deceased person having a fight with his father and then committing
suicide or the deceased having a fight with his father and then his father killing him
could both be situations in which an argument led to death. The control variables of
the victim’s gender, age, and race were also included in the study.
Data Analyses
Initially, the authors ran univariate and bivariate frequencies. Following this initial
descriptive analysis, the authors conducted a multivariate logistic regression. Due to
the large sample size, all the independent variables, control variables, and the inter-
action term (female victim × nudity) were included in the model without concern of
Craun et al. 7
overfitting the model. The logistic regression model was tested for influential obser-
vations, specification errors, goodness of fit, and multicollinearity. Approximately
180 observations were noted to be influential with a Pregibon’s dbeta of greater than
two, which could indicate undue influence on the model (UCLA Institute for Digital
Research and Education, 2020). We ran the model without the 180 observations and
the significance and strength of association of the independent variables remained
the same. Therefore, for parsimony, all the observations are included in the pre-
sented logistic regression.
Results
For the bivariate analyses, all the independent variables and control variables were
significant with the dependent variable at p < .001, which is not unexpected due to the
large sample size. Table 1 illustrates that across approximately 30,000 homicides, in
0.8% of the cases, the victim was found nude. This is twice as frequent as nudity in
suicides, as only 0.4% of suicide victims were found nude at the time of death (χ2[df = 1,
N = 119,145] = 72.3, p < .001). Focusing on the victim stressor variables, an argument
prior to the death (χ2[df = 1, N = 119,145] = 2600, p < .001) and the deceased being a
victim of interpersonal violence in the past month (χ2[df = 1, N = 119,145] = 417.3,
p < .001) were both associated with a higher likelihood of homicide.
The bivariate association between family relationship problems and manner of
death (χ2[df = 1, N = 119,145] = 774.7, p < .001), along with family relationship crises
and manner of death (χ2[df = 1, N = 119,145] = 139.0, p < .001), demonstrated that both
were more common in suicides. The same held true for mental health problems
(χ2[df = 1, N = 119,145] = 17,000, p < .001) and mental health crises (χ2[df = 1,
N = 119,145] = 310.9, p < .001). Finally, at the bivariate level, substance abuse prob-
lems (χ2[df = 1, N = 119,145] = 2,600, p < .001) and substance abuse crises (χ2[df = 1,
N = 119,145] = 479.3, p < .001) were more common in suicides.
Combining the independent variables into one multivariate logistic regression
model (see Table 2), it becomes clear that victim race strongly influences the like-
lihood of a death being a homicide or a suicide. The odds of homicide victimiza-
tion were 18 times higher among black individuals compared to white deceased
persons. Deceased individuals with Hispanic, American Indian, and Asian back-
grounds all had higher odds ratios indicating homicide as compared to white
deceased individuals.
With regards to stressors experienced by the victim, being the victim of interper-
sonal violence increased the odds 5.5 times that the death was a homicide. Closely
following was the interaction between victim gender and being found nude. Female
victims who were found nude were 5.4 higher odds of the death being a homicide. The
presence of mental health problems and crises, substance abuse problems and crises,
and as well as family relationship problems were all related to an increased likelihood
of death being a suicide. The only stressor variable unrelated to homicide and suicide
determination was experiencing a family crisis within the 2 week period preceding
death if the deceased had experienced a family crisis in the past 2 weeks.
8 Homicide Studies 00(0)
Using this data and based on the presented logistic regression model, the authors
explored the predicted probabilities of a death being a homicide or a suicide. Only
considering the gender of the victim, the likelihood that a female victim was a homi-
cide victim was 27.0%. However, if it was a female victim who was found nude this
probability more than doubled, jumping to 57.2%, with all other factors in the model
held constant (See Figure 1 for the interaction effect between victim gender and
nudity). In comparison, with male victims found clothed the probability of homicide
was 25% and jumped to 32% with male, nude victims. If a nude female had an argu-
ment prior to death, the probability of homicide increased to 66.8%, and if she was
also the victim of interpersonal violence in the past 30 days, the likelihood of homicide
increased to 85.6%.
Discussion
In almost 10% of violent death, the manner of death was labeled as undetermined,
which indicates providing empirical support to law enforcement in their equivocal
death investigations is crucial (Jack et al., 2018). Therefore, researchers examined
those cases where manner of death was determined and utilized factors recognized
Table 1. Victim and Situational Characteristics in Homicides and Suicides.
Homicides
(25.3%; n = 30,163)
Suicides
(74.7%; n = 88,982) p
Victim demographics
Average victim age 33.7 years 46.5 years <.001
Female victim 16.5% 23.0% <.001
Victim race <.001
American Indian 1.7% 1.3%
Asian/Pacific Islander 1.2% 2.3%
Black 55.7% 5.9%
Hispanic 14.1% 6.1%
Other/unknown 0.6% 0.4%
Two or more races 1.0% 0.9%
White 25.8% 83.0%
Situational characteristics
Argument led to death 27.2% 14.3% <.001
Family relationship problem 3.8% 8.6% <.001
Family relationship crisis 1.1% 2.2% <.001
IPV victim within past month 1.6% 0.4% <.001
Mental health problem 3.9% 45.5% <.001
Mental health crisis 0.1% 1.1% <.001
Substance abuse problem 12.0% 26.6% <.001
Substance abuse crisis 0.2% 1.9% <.001
Victim found nude 0.8% 0.4% <.001
Craun et al. 9
within to apply the extrapolated analysis to better understand or evaluate more dif-
ficult death scenes. To our knowledge, this is the first empirical study to examine
the relationship between nudity at the time of death and manner of death. We fur-
ther expanded this work by integrating other stressors in the lives of the deceased
with the goal of informing law enforcement’s equivocal death investigations.
In answering our first research question, we found that there was an increased
chance that a death was a homicide if the victim was found nude. As questioned in the
title of the Hawkins (2019) article “Is it common for women to commit suicide in the
nude?” and our second research question exploring interaction effects between gen-
der, nudity, and the likelihood of homicide, this study demonstrated it is relatively
rare for women to commit suicide while nude, and female nudity at the time of death
is strongly related to the death being a homicide. In relation to our third research
question about the strength of association of nudity in relationship to homicides, life
stressors and victim demographics are also significantly associated with homicide-
related deaths. However, in this sample, only two were stronger predictors of homi-
cide than a female victim found nude: being an African-American victim and having
Table 2. Victim and Situational Variables Likelihood of Death being a Homicide.
(n = 119,104).
b
95% CI
Odds ratio p LL UL
Victim age −0.02 −0.02 −0.02 0.98 <.001
Female victim 0.20 0.15 0.25 1.22 <.001
Victim racea
American Indian 0.93 0.82 1.05 2.55 <.001
Asian/Pacific Islander 0.29 0.16 0.41 1.33 <.001
Black 2.93 2.89 2.98 18.74 <.001
Hispanic 1.64 1.59 1.69 5.16 <.001
Other/unknown 0.99 0.79 1.20 2.72 <.001
Two or more races 0.94 0.79 1.09 2.56 <.001
Argument led to death 0.87 0.83 0.92 2.39 <.001
Family relationship problem −0.77 −0.86 −0.67 0.46 <.001
Family relationship crisis −0.13 −0.30 0.04 0.88 .14
IPV victim within month 1.71 1.52 1.89 5.51 <.001
Mental health problem −2.66 −2.73 −2.60 0.07 <.001
Mental health crisis −0.71 −1.26 −0.17 0.49 .01
Substance abuse problem −0.36 −0.41 −0.31 0.70 <.001
Substance abuse crisis −1.74 −2.05 −1.44 0.17 <.001
Victim found nude 0.70 0.38 1.01 2.01 <.001
Female victim & found nude 1.69 1.21 2.18 5.44 <.001
Constant −0.81 −0.87 −0.76 0.44 <.001
aWhite—reference category.
10 Homicide Studies 00(0)
a previous experience of interpersonal violence (within 30 days of death). Previous
mental health problems and crises, substance abuse problems and crises, and family
relationship problems were all associated with higher risk of suicide. Family relation-
ship crises were not related to either homicide or suicide. Crises were classified as a
severe problem in the 2 weeks prior to death, while a corresponding problem was
considered more long-term.
Operational Considerations
Our findings suggest that law enforcement should consider adopting an equivocal
investigation perspective when the deceased is found nude and suicide is alleged by
those surviving the decedent. There are several structured guides for psychological
autopsies and equivocal death investigations, such as the previously mentioned
Checklist of Equivocal Death Information (Davis & Ogloff, 2014). The validity of
the checklists is outside the scope of this paper. We, therefore, do not recommend
one specific tool over another, nor do we argue that structured tools will provide a
definitive answer on the manner of death. Rather, we encourage the use of structured
tools to ensure that all investigative avenues and perspectives are explored and to aid
investigators in achieving a systematic and semi-standardized approach in such
cases. This functions optimally when homicide investigators are called in from the
beginning with a possible nude suicide. Doing so allows for the scene of the death to
25%
32%
27%
57%
0%
25%
50%
75%
100%
Males Found
Clothed
Males Found Nude Females Found
Clothed
Females Found
Nude
Figure 1. Gender and nudity interaction predicted probabilities of homicide (n = 119,104)*.
*Produced using the margins command; controlling for other independent variables in the logistic
regression.
Craun et al. 11
be preserved for evidentiary purposes. For example, with nude equivocal deaths, law
enforcement should pay attention to items that could be of evidentiary value, such as
books, journals, phones, and laptops. Moreover, as with a psychological autopsy,
law enforcement should interview witnesses and key relationships from various con-
nections in the deceased’s life while remaining mindful of who may have a stake in
covering up relationship strains or misdeeds, and whether any acquaintances may
hold biases for or against labeling a death a suicide (Canter, 2000; Hjelmeland et al.,
2012; Pouliot & De Leo, 2006). This also involves understanding the deceased’s
psychological state and social relationships through investigative avenues such as a
scrubbing (or exploitation) of all available social media applications. The deceased’s
social circle can possibly provide insight into medical or psychological care obtained
by the decedent, which allows the investigator to understand the possibility of prior
suicide attempts or ideations. Finally, as nudity is rare in violent deaths, it necessi-
tates the examination of possible staging. Ferguson (2015) found 5% of staged
homicides involved offenders setting up the crime scene to appear as if the motiva-
tion for the death was sexual when it was not.
While the current study demonstrated that nudity is strongly associated with
violent death via homicide, it is not a perfect predictor, as there were over 300
deaths in the dataset that were nude suicides. For possible suicides that involve
nudity, we suggest law enforcement incorporate Joiner’s (2005) concept of suicid-
ality factors—perceived burdensomeness, lack of belonging, and acquired capabil-
ity—into their interviewing. Law enforcement should investigate if the deceased
suffered from the perception that they were a burden to others and that they would
always be a burden (Joiner, 2005). In addition, Joiner stressed a key suicidality fac-
tor is lack of connection to others, or the perception that an individual did not
belong in some way. Finally, law enforcement should look for evidence in the
deceased’s life that increased the capability for suicide. In other words, law enforce-
ment should examine if the deceased was actively working against the natural ten-
dency toward self-preservation by mental practice or previously aborted suicide
attempts (Joiner, 2005).
Limitations and Future Directions
While the NVDRS dataset is quite extensive, it did not cover the entire nation with the
2017 data, and we were limited to case descriptions within law enforcement and medi-
cal examiner narratives. Therefore, we relied on the assumption that the deceased
being found nude was unique enough that it would have been noted in the narratives.
However, we cannot discount that there are cases where the deceased was found nude,
but it was not noted in the narratives. Additionally, we did not search the narratives to
determine if there were indications of staging that the medico-legal death investigators
noted, which is something particularly important to consider for potentially staged
homicides (Ferguson & Petherick, 2016; Miller et al., 2020). Also, due to the nature of
the data, NVDRS collapses no/unknown/not available into a single category which
limits our ability to distinguish between “no” responses and true missing data.
12 Homicide Studies 00(0)
Moreover, there was no way for us to determine if the deaths that were labeled as a
homicide or a suicide were at one time considered an equivocal death case, so it is
possible that they may be different in unknown ways from equivocal death cases that
were eventually labeled as homicides or suicides. Finally, due to the NVDRS’ coding
rules autoerotic fatalities are excluded, so in-depth examination into this specific type
of nude death was not possible.
The present study is the first to demonstrate the existence of the relationship
between nudity and the likelihood of homicide. Future work should aim to see if these
findings can be replicated within additional datasets and populations. Additionally,
while this was the first study to examine the interaction between gender and nudity in
violent death, future research should explore potential interaction effects between vic-
tim nudity and other relevant variables such as age, sexual orientation, drug/alcohol
use, victim’s body image and religious practices. Finally, case characteristics, such as
the location of the body and expectations of nudity within that location (such as in the
shower vs. in the backyard), should also be explored in order to fully examine the
diagnostic value of the nudity variable in the context of violent deaths and law enforce-
ment investigations.
Equivocal death investigations are difficult from the outset as there may be multiple
ways investigators could interpret the scene upon arrival. Depending upon the situa-
tional factors, context and evidence at the scene, the death may appear to be a suicide
when in fact what is observed is has been staged by the offender. One variable of
importance which had not been previously examined was that of nudity. With some
acknowledged limitations, the results of this study suggest that there is a higher prob-
ability a homicidal act has occurred when a female victim is discovered nude.
Assessment of the level of dress or undress of a possible suicide victim should serve
as one aspect of a holistic approach that includes not just a scene assessment, forensics
and pathological exam, but also a psychological assessment of the victim in determin-
ing manner of death.
Acknowledgments
We want to acknowledge Sarah A. Shaffer with her assistance reviewing and editing this
manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship,
and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of
this article.
ORCID iD
Sarah W. Craun https://orcid.org/0000-0002-3736-636X
Craun et al. 13
Notes
1. Although death investigations may be conducted by both law enforcement and medical
examiners or coroners, for the purpose of this literature review, we do not differentiate
these professions. Rather, we use the term medico-legal death investigators to include any
profession which may be involved in such investigations.
2. We chose not to include accidental deaths in our analysis as the dataset only includes a
restricted range of accidental violent deaths and is therefore not a representative sample of
all accidental deaths that occur.
References
Brown, S., & Seals, J. (2019). Intimate partner problems and suicide: Are we missing the vio-
lence? Journal of Injury and Violence Research, 11(1), 53–64. https://doi.org/10.5249/jivr.
v11i1.997
Canter, D. V. (2000). Psychological autopsies. In J. Siegel & P. Saukko (Eds.), Encyclopedia
of forensic sciences. Elsevier. http://eprints.hud.ac.uk/id/eprint/8669/1/canterpsycho-
logical.pdf
Canter, D. V., Giles, S., & Nicol, C. (2004). Suicide without explicit precursors: A state of
secret despair? Journal of Investigative Psychology and Offender Profiling, 1(3), 227–248.
https://doi.org/10.1002/jip.4
Cavanagh, J. T. O., Carson, A. J., Sharpe, M., & Lawrie, S. M. (2003). Psychological autopsy
studies of suicide: A systematic review. Psychological Medicine, 33(3), 395–405. https://
doi.org/10.1017/s0033291702006943
Cave, R, DiMaio, V.J. & Molina, D.K. (2014). Homicide or suicide? Gunshot wound interpreta-
tion: A Bayesian approach. American Journal of Forensic Medicine and Pathology, 35(2),
118–123.
Chan, H.-C., & Heide, K. M. (2009). Sexual homicide: A synthesis of the literature. Trauma
Violence & Abuse, 10(1), 31–54. https://doi.org/10.1177/1524838008326478
Cheng, A. T. A., Chen, T. H. H., Chen, C.-C., & Jenkins, R. (2000). Psychosocial and psychiat-
ric risk factors for suicide: Case-control psychological autopsy study. The British Journal
of Psychiatry, 177(4), 360–365. https://doi.org/10.1192/bjp.177.4.360
Crepeau-Hobson, F. (2010). The psychological autopsy and determination of child suicides:
A survey of medical examiners. Archives of Suicide Research, 14(1), 24–34. https://doi.
org/10.1080/13811110903479011
Dattilio, F. M., & Sadoff, R. L. (2011). Psychological autopsy. In E. Y. Drogin, F. M. Dattilio,
R. L. Sadoff, & T. G. Gutheil (Eds.), Handbook of forensic assessment: Psychological and
psychiatric perspectives (pp. 593–611). John Wiley & Sons.
Davis, M. R., & Ogloff, J. R. P. (2014). Psychological autopsy of equivocal deaths. In G. J.
N. Bruinsma & D. L. Weisburd (Eds.), Encyclopedia of criminology and criminal justice
(pp. 4098–4107). Springer.
De-Giorgio, F., Lodise, M., Quaranta, G., Spagnolo, A. G., D’Aloja, E., Pascali, V. L., &
Grassi, V. M. (2015). Suicidal or homicidal sharp force injuries? A review and critical
analysis of the heterogeneity in the forensic literature. Journal of Forensic Sciences, 60(s1),
S97–S107. https://doi.org/10.1111/1556-4029.12673
Ferguson, C. (2015). Staged homicides: An examination of common features of faked burglar-
ies, suicides, accidents, and car accidents. Journal of Police and Criminal Psychology,
30(3), 139–157. https://doi.org/10.1007/s11896-014-9154-1
14 Homicide Studies 00(0)
Ferguson, C., & McKinley, A. (2019). Detection avoidance and mis/unclassified, unsolved
homicides in Australia. Journal of Criminal Psychology, 10(2), 113–122. https://doi.org
/10.1108/jcp-09-2019-0030
Ferguson, C., & Petherick, W. (2016). Getting away with murder: An examination of detected
homicides staged as suicides. Homicide Studies, 20(1), 3–24. https://doi.org/10.1177
/1088767914553099
Ferguson, C., & Sutherland, T. (2018). Murder by pushing: An exploratory analysis of hom-
icidal falls from a height. American Journal of Forensic Medicine & Pathology, 39(3),
192–200. https://doi.org/10.1097/paf.0000000000000386
Gavin, M., & Rogers, A. (2006). Narratives of suicide in psychological autopsy: Bringing lay
knowledge back in. Journal of Mental Health, 15(2), 135–144. https://doi.org/10.1080
/09638230600608735
Hawkins, E. (2019, June 14). Is it common for women to commit suicide in the nude? Oxygen.
https://www.oxygen.com/death-at-the-mansion-rebecca-zahau/crime-time/nude-female-
suicide-cleopatra-marilyn-monroe-rebecca-zahau
Hjelmeland, H., Dieserud, G., Dyregrov, K., Knizek, B. L., & Leenaars, A. A. (2012).
Psychological autopsy studies as diagnostic tools: Are they methodologically flawed?
Death Studies, 36(7), 605–626. https://doi.org/10.1080/07481187.2011.584015
Jack, S. P. D., Petrosky, E., Lyons, B. H., Blair, J. M., Ertl, A. M., Sheats, K. J., & Betz,
C. J. (2018). Surveillance for violent deaths—National Violent Death Reporting System,
27 states, 2015. MMWR Surveillance Summaries, 67(11), 1–32. https://doi.org/10.15585/
mmwr.ss6711a1
Johnson, H., & Hotton, T. (2003). Losing control: Homicide risk in estranged and intact intimate
relationships. Homicide Studies, 7(1), 58–84. https://doi.org/10.1177/1088767902239243
Joiner, T. (2005). Why people die by suicide. Harvard University Press.
Lacks, R. D., Westveer, A. E., Dibble, A., & Clemente, J. (2008). Equivocal death investiga-
tion: Case study analyses. Victims & Offenders, 3(2-3), 150–164. https://doi.org/10.1080
/15564880801938292
Lindqvist, P., & Gustafsson, L. (2002). Suicide classification—clues and their use: A study of
122 cases of suicide and undetermined manner of death. Forensic Science International,
128(3), 136–140. https://doi.org/10.1016/s0379-0738(02)00188-3
Miller, M. L., Merola, M. A., Opanashuk, L., Robins, C. J., Chancellor, A. S., & Craun, S. W.
(2020). “911 What’s your emergency?”: Deception in 911 homicide and suicide staged
as homicide calls. Homicide Studies. Advance online publication. https://doi.org/10.1177
/1088767920948242
Molina, D. K., & DiMaio, V. J. M. (2008). Rifle wounds: A review of range and location as
pertaining to manner of death. American Journal of Forensic Medicine & Pathology, 29(3),
201–205. https://doi.org/10.1097/paf.0b013e31818345a5
Molina, D. K., DiMaio, V. J. M., & Cave, R. (2013a). Gunshot wounds: A review of firearm
type, range, and location as pertaining to manner of death. American Journal of Forensic
Medicine & Pathology, 34(4), 366–371. https://doi.org/10.1097/paf.0000000000000065
Molina, D. K., DiMaio, V. J. M., & Cave, R. (2013b). Handgun wounds: A review of range
and location as pertaining to manner of death. American Journal of Forensic Medicine &
Pathology, 34(4), 342–347. https://doi.org/10.1097/paf.0000000000000048
Molina, D. K., Wood, L. E., & DiMaio, V. J. M. (2007). Shotgun wounds: A review of range
and location as pertaining to manner of death. American Journal of Forensic Medicine &
Pathology, 28(2), 99–102. https://doi.org/10.1097/01.paf.0000257415.82728.d7
Craun et al. 15
Pelisek, C. (2019, May 20). Calif. woman’s mansion death was ruled a suicide 8 years ago– but
family believes it was murder. People. https://people.com/crime/rebecca-zahau-family-
suspects-murder-police-say-death-by-suicide/
Pettler, L. G. (2016). Crime scene staging dynamics in homicide cases. CRC Press.
Pouliot, L., & De Leo, D. (2006). Critical issues in psychological autopsy studies. Suicide
and Life-Threatening Behavior, 36(5), 491–510. https://doi.org/10.1521/suli.2006.36
.5.491
Saavedra, T. (2020). Justice delayed: O.C. woman’s 1974 death reclassified from suicide to
homicide. The Orange County Register. https://www.ocregister.com/justice-delayed-o-c-
womans-1974-death-reclassified-from-suicide-to-homicide
Safe States Alliance. (2017). Deaths from suicide: A look at 18 states. A special report with
data from the national violent death reporting system, 2013–2014. https://cdn.ymaws.
com/www.safestates.org/resource/resmgr/nvdrs/Suicide_Report_2017/NVDRS_Special_
Report_Final_2.pdf
Silverman, M. M., & De Leo, D. (2016). Why there is a need for an international nomenclature
and classification system for suicide. Crisis, 37(2), 83–87. https://doi.org/10.1027/0227-
5910/a000419
Simon, R. I. (2008). Naked suicide. Journal of the American Academy of Psychiatry and the
Law Online, 36(2), 240–245.
Smith, S. G., Basile, K. C., & Karch, D. (2011). Sexual homicide and sexual violence-associated
homicide: Findings from the National Violent Death Reporting System. Homicide Studies,
15(2), 132–153. https://doi.org/10.1177/1088767911406236
Terranova, C., Doro, L., Zancaner, S., Zampini, T., Mazzarolo, C., Bonvicini, B., Viero, A.,
& Montisci, M. (2020). Criminological and medico-legal aspects in homicidal and sui-
cidal sharp force fatalities. Journal of Forensic Sciences, 65(4), 1184–1190. https://doi.
org/10.1111/1556-4029.14285
UCLA Institute for Digital Research and Education. (2020). Logistic regression diagnostics.
https://stats.idre.ucla.edu/stata/webbooks/logistic/chapter3/lesson-3-logistic-regression-
diagnostics/
Visentin, S., Massaro, L., Viel, G., Cecchetto, G., & Montisci, M. (2019). Suicide identifica-
tion during on-site inspection. Proposal and application of an interpretative method for
death scene investigation. Forensic Science International, 297, 148–155. https://doi.org
/10.1016/j.forsciint.2019.01.029
Author Biographies
Sarah W. Craun, PhD is a senior research associate at the United States Sentencing
Commission. Prior she was a research coordinator for the Behavioral Analysis Unit at the
Federal Bureau of Investigation.
Leah Tanner is a supervisory special agent in the FBI’s Behavioral Analysis Unit focusing on
crimes against adults. Prior to joining the Behavioral Analysis Unit she was an agent with the
FBI’s Los Angeles field office.
Victoria Clausen is a crime analyst assigned to the Violent Criminal Apprehension Program
(ViCAP) at the FBI. Previously she worked as a crime analyst for the FBI’s Behavioral Analysis
Unit focusing on research and program management.
16 Homicide Studies 00(0)
Melissa A. Merola, MA, MSEd, is a supervisory special agent with ATF assigned to the FBI’s
Behavioral Analysis Unit focusing on crimes against adult victims. Her focus is on cases with a
nexus to fire fatalities.
Leonard Opanashuk, JD is a supervisory special agent in the FBI’s Behavioral Analysis Unit
focusing on research and program management. Prior he worked in the crimes against adults
Behavioral Analysis Unit.
Timothy G. Keel is a major case specialist with the FBI’s Behavioral Analysis Unit. Prior to
joining the FBI, he was a homicide detective with the Baltimore, MD police department.
