Research Trends

Suicide on Instagram –
${protocol}://econtent.hogrefe.com/doi/pdf/10.1027/0227-5910/a000529 - Tuesday, June 26, 2018 3:21:55 AM - Universitätsbibliothek LMU München IP Address:129.187.254.47

Content Analysis of a German

­Suicide-Related Hashtag
Florian Arendt

Department of Communication Science and Media Research, University of Munich (LMU), Germany

Abstract. Background: Suicide is the second leading cause of death among 15–29-year-olds globally. Unfortunately, the suicide-related con-
tent on Instagram, a popular social media platform for youth, has not received the scholarly attention it deserves. Method: The present study
provides a content analysis of posts tagged as #selbstmord, a German suicide-related hashtag. These posts were created between July 5 and
July 11, 2017. Results: Approximately half of all posts included words or visuals related to suicide. Cutting was by far the most prominent meth-
od. Although sadness was the dominant emotion, self-hate and loneliness also appeared regularly. Importantly, inconsistency – a gap between
one’s inner mental state (e.g., sadness) and one’s overtly expressed behavior (e.g., smiling) – was also a recurring theme. Conversely, help-seek-
ing, death wishes, and professional awareness–intervention material were very rare. An explorative analysis revealed that some videos relied
on very fast cutting techniques. We provide tentative evidence that users may be exposed to purposefully inserted suicide-related subliminal
messages (i.e., exposure to content without the user’s conscious awareness). Limitations: We only investigated the content of posts on one
German hashtag, and the sample size was rather small. Conclusion: Suicide prevention organizations may consider posting more awareness–
intervention materials. Future research should investigate suicide-related subliminal messages in social media video posts. Although tentative,
this finding should raise a warning flag for suicide prevention scholars.

Keywords: social media, youth, Instagram, cutting, subliminal messages

Suicide is the second leading cause of death among
15–29-year-olds globally (World Health Organization
[WHO], 2017a). There is a web of multiple risk factors in-
cluding emotional states (e.g., depressed mood), personal-
ity traits (e.g., low self-esteem), deficits in problem-solving
abilities, and stressors such as specific life events (e.g., a
breakdown of a close relationship; Orbach & Iohan-Barak,
2009). Among the myriad factors, the media are consid-
ered a key factor (Mann et al., 2005; Niederkrotenthaler &
Stack, 2017; WHO, 2017b).
Research has found that the media can have detrimental
and beneficial effects. On the one hand, there is evidence
that news reporting on suicide and depictions of suicide in
entertainment programming can increase suicidal behav-
ior – a phenomenon known as the Werther effect (Phillips,
1974; Stack, 2005). On the other hand, research has also
found that the media can decrease suicidal behavior by re-
sponsible reporting on suicide, for example, by publishing
stories on individuals who have overcome a suicidal crisis
– a phenomenon known as the Papageno effect (Nieder­
krotenthaler et al., 2010). Therefore, media effects on su-
icidal behavior depend on the way suicide is depicted (see
Niederkrotenthaler & Stack, 2017).
Much of our knowledge about media effects on suicid-
al behavior stems from studies conducted in the context
of traditional media environments, largely dominated by
printed newspapers and television. Importantly, the media
landscape has changed dramatically during recent years,
with social media becoming a key player. Instagram – a
prime example of a popular social media platform among
youth (see next section) – has not received the scholarly at-
tention it deserves. This social media platform distributes
a plethora of suicide-related content that is publicly avail-
able every day. The present study aims to improve our un-
derstanding by providing a content analysis of a German
suicide-related hashtag on Instagram.
Social Media and Suicidal Behavior
Although the Internet has increasingly gained the atten-
tion of suicide researchers (Biddle, Donovan, Hawton,
Kapur, & Gunnell, 2008), our knowledge of the role social
media play in terms of imitational and preventive effects
is limited. Nevertheless, research has shown that social
media are linked to both benefits and risks in regard to the
health of adolescents (Reid-Chassiakos, Radesky, Chris-
takis, Moreno, & Cross, 2016). Whereas the benefits may
include a reduction in social isolation, encouragement
toward recovery, and a decrease in self-harm urges, risks

© 2018 Hogrefe Publishing Crisis 2018

https://doi.org/10.1027/0227-5910/a000529
 2 F. Arendt, Suicide on Instagram
may include a disturbing effect on (especially young) users
and a contagious triggering effect of self-harming behav-
iors (Baker & Lewis, 2013; Brown et al., 2018).
${protocol}://econtent.hogrefe.com/doi/pdf/10.1027/0227-5910/a000529 - Tuesday, June 26, 2018 3:21:55 AM - Universitätsbibliothek LMU München IP Address:129.187.254.47
Instagram is a prime example of a modern social me-
dia platform that is especially popular among youth (Pew
Research Center, 2015). Instagram is a web-based photo-
and video-sharing application that allows users to watch
and share pictures and short videos. Pictures or videos on
a specific topic can be identified by a so-called hashtag – a
word or phrase preceded by the symbol #. A hashtag allows
users to easily access posts with a specific theme or con-
tent and it enables users to access and feel connected to a
theme-related community (Moreno, Ton, Selkie, & Evans,
2016). Importantly for the present study, there are a num-
ber of suicide-related hashtags in different languages (e.g.,
#suicide, #suicidio), including the popular German hash-
tag that is the central focus of the present study (#selbst-
mord; English translation of the German word selbstmord
= suicide).
Unfortunately, the suicide-related content on Insta-
gram has not received the scholarly attention it deserves.
Scholars have only recently started investigating self-harm
on Instagram. Importantly, previous studies focused on
nonsuicidal self-injury (NSSI) – the deliberate destruc-
tion of body tissue in the absence of a conscious suicide
intention (Nock & Favazza, 2009). For example, Brown
and colleagues (2018) investigated posts on #cutting (and
related hashtags) and found that pictures of NSSI are fre-
quently posted. This finding is supported by a recent study
(Miguel et al., 2017). Notably, recovery-oriented resourc-
es are virtually absent, as Miguel and colleagues (2017)
have shown. Although Instagram has introduced its own
content advisory warning message – a message that pops
up explaining that you will possibly view emotionally dis-
turbing graphic content and the message also provides
helpful resources; individuals can nevertheless proceed
viewing the content – a previous study has shown that only
one third of relevant suicide-related hashtags trigger this
message (Moreno et al., 2016).
Key for the present study is that there is hardly any
knowledge related to self-injury with a conscious suicide
intention. Thus, the present study provides a content anal-
ysis of postings to #selbstmord. Due to this lack of previ-
ous studies, the present work purposefully employed a
descriptive and explorative approach. We derived a set of
variables from the suicide prevention literature (see Sh-
neidman, 1987; Sonneck, Kapusta, Tomandl, & Voracek,
2012) and coded publicly accessible posts on #selbstmord:
We placed special emphasis on the suicide method, bear-
ing possible method-specific imitation effects in mind (Et-
zersdorfer & Sonneck, 1998; Schmidke & Häfner, 1988).
We also coded for the presence of emotions (e.g., sadness,
self-hate), awareness–intervention materials including
Crisis 2018
recovery-oriented resources, overt expressions of death
wishes, and help-seeking.
The use of social media platforms such as Instagram has
increased in recent years (Pew Research Center, 2015).
Thus, it is important that scholars, mental health profes-
sionals, social media platform operators, parents, and
teachers are aware of problematic aspects of young peo-
ple’s social media use (Miguel et  al., 2017). Suicide pre-
vention research has a special responsibility in terms of
monitoring for potential problems.
Method
We conducted a content analysis of posts tagged with the
hashtag #selbstmord. We downloaded the 250 most re-
cent posts on July 11, 2017, and coded for the presence
or absence of specific content elements. These posts were
posted on Instagram and tagged with the hashtag #selbst-
mord between July 5 and July 11 (i.e., within 1 week). All
variables were coded dichotomously (0 = absent, 1 = pres-
ent) by one coder. For the reliability analysis, we random-
ly selected 10% of the material. The material was coded
by the same coder a second time. The analysis generally
indicated reliable measurements. Krippendorff ’s α values
were generally acceptable (all values, α > .74). There were
two exceptions: The measurement of the variables wom-
an (α = .58) and man (α = .35) were not reliable. Although
Krippendorff ’s α is very sensitive to inconsistent coding in
small samples where one code dominates (i.e., 0 or “ab-
sent” in the present study), these low α values indicate
that the sex-related variables could not be measured in a
reliable way. This was unexpected because sex is typically
one of the most reliable variables to code. However, in the
present sample, the images and videos often represented
young individuals. Sex-related cues were ambiguous and
thus hard to evaluate. Furthermore, as we will outline,
many videos used very fast cutting techniques. Because of
these techniques, sex-related cues were difficult to code as
well. Despite this low level of reliability, we decided to re-
port on our planned analyses regarding sex. However, we
emphasize that the sex-related results should be interpret-
ed with caution. We can only provide tentative evidence on
this aspect.
Variables
Reference to Suicide
This variable was coded as present when the post included
an explicit reference to suicidal thoughts, plans, or behav-
iors.
© 2018 Hogrefe Publishing
 F. Arendt, Suicide on Instagram 3

Suicide Method Help-Seeking

We coded whether the post depicted a specific suicide We coded this variable when statements such as “Help
method: cutting, jumping, firearm, hanging, drowning, me,” “I need help,” or “Who can help me?” were present.
${protocol}://econtent.hogrefe.com/doi/pdf/10.1027/0227-5910/a000529 - Tuesday, June 26, 2018 3:21:55 AM - Universitätsbibliothek LMU München IP Address:129.187.254.47

poisoning, and railway. Each method was represented by

its own variable. Professional Awareness Material
We coded whether the post included professional aware-
Demographics ness material (e.g., a reference to a counseling service).
We coded whether the post visually (e.g., pictures, visuals)
or verbally (e.g., words, phrases) represented a female or
a male. We did not code for sex when the post represent- Data Analysis
ed drawings or animated characters. Furthermore, the age
of the individual was coded. We coded whether the post We calculated descriptive statistics (percentage values)
included young individuals (i.e., youth: children, adoles- and 95% confidence intervals based on bootstrapping
cents, young adults), older adults (including the elderly), techniques. The percentage values are significantly differ-
or held no age-related information. ent when the confidence intervals do not overlap. The con-
fidence interval lower and upper limits are reported after
Emotions the percentage value.
The following emotional expressions were coded when
they were visually or verbally present: sadness (e.g., tears,
unambiguously sad faces), loneliness (e.g., the presence of
the word lonely), and self-hate (e.g., verbal statements such Results
as “I hate myself ”).
There was an explicit reference to suicide in 45.6% (95%
Inconsistency CI = 39.6–51.6) of all posts. At least one specific method
We coded whether the post explicitly addressed a conflict was depicted in 34.4% (95% CI = 28.8–40.8) of all posts.
experienced when internal cognitive and emotional con- As can be seen in Figure 1, cutting was by far the most fre-
tent was inconsistent with overtly expressed behavior. In- quently depicted method, followed by jumping, firearm,
consistency was coded as present, for example, when the poisoning, railway, drowning, and hanging. Although both
post presented a verbal statement such as “I’m fine” along- women and men (or boys and girls) regularly appeared vis-
side a (visually or verbally represented) depressed mood. ually in posts, women (and girls) appeared in significantly
more posts (37.2%, 95% CI = 31.2–43.2), than men (or
Death Wish boys) did (16.8%, 95% CI = 12.4–21.6). Posts with pic-
E ON INSTAGRAM
Furthermore, we coded whether explicit death wishes tures of individuals
19 typically included youths (37.6%,
(e.g., “I want to die”) had been verbally communicated. 95% CI = 31.6–43.2). Conversely, older adults appeared
in only 0.4% (95% CI = 0.0–1.2) of all posts. In the major-
ity of the posts, however, age could not be coded because
P o s t s In c lu d in g M e t h o d [ % ]

30
of an absence of age-related information (62.0%, 95% CI
= 56.4–68.0).
Concerning emotional states, we found that sadness
20
was depicted in 26.8% (95% CI = 21.6–32.4) of all posts
and thus represented the most frequently depicted emo-
tion. Self-hate (95% CI = 9.2%, 5.6–12.8), and loneliness
10
(95% CI = 5.2%, 2.8–8.4), were represented in signifi-
cantly fewer posts, but also appeared regularly. Inconsist-
ency was apparent in 12.4% (95% CI = 8.4–16.8) of posts
0 – an unexpectedly high number of appearances.
Only a few posts included death wishes (2.8%, 95% CI
g

y
g

g
g

rm

a
in

in

in

in
in

= 0.8–5.2) or help-seeking (2.4%, 95% CI = 0.8–4.4). No-

w
a
p
tt

g
il
e

o
m

n
w
u

a
ir

is

a
C

ro
u

tably, there was only one post that included professional

F

H
J

D
P

M e th o d awareness material (0.4%, 95% CI = 0.0–1.2). Interest-

ingly, this content was not posted by a suicide prevention
Figure 1. Percentages of posts showing a specific suicide method. Er-
ror bars indicate bootstrapped 95% confidence intervals. Reading ex- organization, but included a screenshot made by a user
ample: 22% of all posts (N = 250) depicted cutting. showing Instagram’s advisory content warning message.

igure 1. Percentages of posts showing a specific suicide method. Error bars indicate
© 2018 Hogrefe Publishing Crisis 2018

pped 95% confidence intervals. Reading example: 22% of all posts (N = 250) depicted
 4 F. Arendt, Suicide on Instagram
Explorative Analysis
Videos were included in 26.4% (95% CI = 21.2–32.0) of
${protocol}://econtent.hogrefe.com/doi/pdf/10.1027/0227-5910/a000529 - Tuesday, June 26, 2018 3:21:55 AM - Universitätsbibliothek LMU München IP Address:129.187.254.47
all posts. During the coding process, an in-depth analysis
of the video content revealed an unexpected finding: Many
videos relied on very fast cutting techniques (i.e., a film
editing technique that relies on the use of several consec-
utive shots of very brief duration). In these videos, it was
sometimes almost impossible to get the depicted content
by viewing the videos at a regular speed. When analyzed
in a frame-by-frame fashion, some videos presented sui-
cide-related content (i.e., an explicit depiction of suicid-
al behavior) in these extremely fast segments. The sui-
cide-related content was represented visually (e.g., a short
clip of a girl slicing her throat) and by words. Regarding
the latter, the words (e.g., dead, suicidal, depression) were
presented as short flashes. To gain a general impression of
some of these possible suicide-related subliminal messag-
es, sample videos can be obtained from the author upon
request. We return to this important explorative finding in
the Discussion section.
Discussion es in advertising are forbidden in many countries owing
Social media platforms such as Instagram have only re-
cently begun to gain scholarly attention in the suicide pre-
vention domain. As already noted, previous research on
suicide-related content on Instagram has focused on NSSI.
We extended this body of research by investigating posts
with a German suicide hashtag. We found that cutting was
the most prominent method. Importantly, many posts pre-
sented an explicit graphic depiction of the method. This is
a significant finding when bearing possible method-spe-
cific imitation effects in mind (Etzersdorfer & Sonneck,
1998; Schmidke & Häfner, 1988). Sadness was the dom-
inant emotion. Explicit death wishes and help-seeking
were rare. Interestingly, a gap between the Instagrammer’s
inner mental state (e.g., sadness, self-hate) and his or her
overtly expressed behavior (e.g., smiling) – a phenomenon
termed inconsistency in the present study – was a recurring
theme.
Instagram is an important social media channel, espe-
cially for youth. Thus, there is great potential for suicide
prevention to reach young, vulnerable individuals via this
platform. It is important to note that Instagram already
uses warning messages when individuals want to expose
themselves to specific hashtags. Unfortunately, a recent
study found that only one third of problematic suicide-re-
lated hashtags generated Instagram’s own content adviso-
ry message (Moreno et al., 2016). Warning messages are
important, but not enough. Awareness–intervention mate-
Crisis 2018
rials including recovery-oriented resources should be post-
ed by professional suicide prevention organizations more
frequently. As noted by Brown and colleagues (2018),
preventive work might benefit from advancements in ma-
chine learning that can help to identify vulnerable users
and posts. When users are identified, access to counseling
services should be provided.
Subliminal Messages
Explorative analyses revealed that many videos relied on
very fast cutting techniques and presented suicide-relat-
ed content within these fast-paced scenes. Most notably,
some of this content is presented at subthreshold dura-
tions. Thus, it is very hard, sometimes impossible, to be
consciously aware of this suicide-related content when
these videos are viewed at a regular speed.
The advertising literature has coined the term sublim-
inal messages for similar advertising content. Subliminal
messages describe “a technique of exposing consumers to
product pictures, brand names, or other marketing stim-
uli without the consumers having conscious awareness”
(Trappey, 1996, p. 517). Importantly, subliminal messag-
to their deceptive and manipulative character. Although
we cannot categorically prove whether the producers of
these suicide-related videos had the persuasive intention
of detrimentally changing other users’ behavior, it is nev-
ertheless highly questionable from an ethical point of view.
Although the effects of subliminally presented stimuli are
complex, we know from previous psychological research
that subliminally presented stimuli can elicit substantial
effects on human information processing (Olsen & Fazio,
2002). A short exposure to these stimuli can even lead
to automatic behavior changes (Bargh, Chen, & Burrows,
1996). The general idea is that subliminally presented
stimuli can prime (i.e., activate) corresponding concepts
in memory (even outside of conscious awareness), which
in turn activate mentally associated behavioral schemata.
Subliminal messages may thus increase the accessibility of
suicide-related behavioral schemata (e.g., related to a spe-
cific method). Although the mere automatic (re-)activation
of behavioral schemata in memory does not fully deter-
mine behavior, it increases the likelihood of its execution
(see Bargh, 2017).
The problem becomes even more apparent when con-
sidering the following fact: As revealed by Moreno and col-
leagues (2016), some suicide-related hashtag terms over-
lap with existing non-suicide terms such as #cat. Moreno
and colleagues (2016) argued that children or adolescents
who want to see pictures or videos of cute cats may be in-
advertently exposed to posts with suicide-related content
© 2018 Hogrefe Publishing
 F. Arendt, Suicide on Instagram
because of the overlap. In the context of subliminal mes-
sages, they may be exposed to suicide-related content
even without consciously being aware of it – an alarming
${protocol}://econtent.hogrefe.com/doi/pdf/10.1027/0227-5910/a000529 - Tuesday, June 26, 2018 3:21:55 AM - Universitätsbibliothek LMU München IP Address:129.187.254.47
phenomenon.
We want to emphasize that the empirical evidence and
our post hoc theorizing on this phenomenon are tenta-
tive. Despite its tentative nature, it should definitely raise
a warning flag for both suicide researchers and mental
health professionals. More research is undoubtedly need-
ed on this possible alarming subliminal phenomenon and
its effects.
Limitations the social media domain.
This study has several limitations. First, we only investi-
gated the content of posts on one German hashtag. This
decreases the generalizability of our findings. However,
many posts use English content. This is especially true
for the video posts. Second, the study used a descriptive
and explorative approach without a focus on a deductive
test of hypotheses. Nevertheless, this approach enabled us
to reveal an important phenomenon (subliminal suicide
messages). It is up to future research to provide a more sys-
tematic assessment. Third, the sample of posts was rather
small (compared with the number of posts that are availa-
ble on that hashtag). Furthermore, these posts were posted
on Instagram within a period of only 1 week. This decreas-
es the generalizability of the findings. Fourth, the coding
of sex-related content (i.e., the variables woman and man)
was not reliable (see explanation in Method section). Fu-
ture studies should provide replications of sex-related find-
ings. Fifth, we did not investigate variables related to the
source of the posts (i.e., who posted the content). Sixth, we
investigated only three emotions (i.e., sadness, loneliness,
and self-hate). Future research should definitely broaden
the perspective.
Conclusion Psychological Medicine, 48, 337–346. https://doi.org/10.1017/
In the context of traditional media such as newspapers
and television, media guidelines have been developed to
increase responsible reporting (Bohanna & Wang, 2012;
Pirkis, Blood, Beautrais, Burgess, & Skehan, 2006). Sim-
ilar guidelines should be developed for social media plat-
forms as well. If future studies can support our explorative
finding, these guidelines should include a passage on sub-
liminal messages. There is clearly a need to monitor sui-
cide-related content on social media platforms.
Suicide prevention scholars as well as mental health pro-
© 2018 Hogrefe Publishing
5
fessionals should be aware of the possible consequences
of social media platforms such as Instagram. Owing to its
high number of younger users, this social media platform
clearly deserves more attention. In a first step, research has
to provide more empirical evidence: What kind of (sublim-
inal) messages do users post? Who posts them? What is the
content of the posts? How often are they posted? In a sec-
ond step, possible policy interventions should be discussed
by suicide experts. The goal is to increase awareness
among health professionals, journalists, parents, teachers,
and social media companies. Subliminal messages in ad-
vertising are forbidden in many countries because of their
deceptive and manipulative character. Discussions are
needed on whether a ban is also an appropriate solution in
Furthermore, professional suicide prevention organiza-
tions may consider posting more awareness–intervention
materials including help sources on Instagram, a prime ex-
ample of modern social media platforms that is especial-
ly popular among youth. This allows professional suicide
prevention organizations to provide preventive materials
especially to this (young) segment of the population.
References
Baker, T., & Lewis, S. (2013). Responses to online photographs of
non-suicidal self-injury: A thematic analysis. Archives of Suicide
Research, 17, 223–235. https://doi.org/10.1080/13811118.201
3.805642
Bargh, J. (2017). Before you know it: The unconscious reasons we do
what we do. London, UK: William Heinemann.
Bargh, J., Chen, M., & Burrows, L. (1996). Automaticity of social be-
havior: Direct effects of trait construct and stereotype activa-
tion on action. Journal of Personality and Social Psychology, 71,
230–244. https://doi.org/10.1037/0022-3514.71.2.230
Biddle, L., Donovan, J., Hawton, K., Kapur, N., & Gunnell, D. (2008).
Suicide and the internet. BMJ, 336, 800–802. https://doi.org/­
10.1136/bmj.39525.442674.AD
Bohanna, I., & Wang, X. (2012). Media guidelines for the respon-
sible reporting of suicide. Crisis, 33, 190–198. https://doi.
org/10.1027/0227-5910/a000137
Brown, R., Fischer, T., Goldwich, A., Keller, F., Young, R., & Plener, P.
(2018). #cutting: Non-suicidal self-injury (NSSI) on Instagram.
S0033291717001751
Etzersdorfer, E., & Sonneck, G. (1998). Preventing suicide by in-
fluencing mass-media reporting. The Viennese experience
1980–1996. Archives of Suicide Research, 4, 67–74. https://doi.
org/10.1023/A:1009691903261
Mann, J., Apter, A., Bertolote, J., Beautrais, A., Currier, D., Haas, A.,
… Hendin, H. (2005). Suicide prevention strategies: A system-
atic review. JAMA, 294, 2064–2074. https://doi.org/10.1001/
jama.294.16.2064
Miguel, E., Chou, T., Golik, A., Cornacchio, D., Sanchez, A., DeSerisy,
M., & Comer, J. (2017). Examining the scope and patterns of de-
liberate self-injurious cutting content in popular social media.
Depression and Anxiety, 34, 786–793. https://doi.org/10.1002/
da.22668
Crisis 2018
 6 F. Arendt, Suicide on Instagram
Moreno, M., Ton, A., Selkie, A., & Evans, Y. (2016). Secret society 123:
Understanding the language of self-harm on Instagram. Jour-
nal of Adolescent Health, 58, 78–84. https://doi.org/10.1016/j.
jadohealth.2015.09.015
${protocol}://econtent.hogrefe.com/doi/pdf/10.1027/0227-5910/a000529 - Tuesday, June 26, 2018 3:21:55 AM - Universitätsbibliothek LMU München IP Address:129.187.254.47
Niederkrotenthaler, T., & Stack, S. (2017). Media and suicide. Inter-
national perspectives on research, theory, and policy. New York,
NY: Transaction Publishers.
Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss,
M., Etzersdorfer, E., … Sonneck, G. (2010). The role of media
reports in completed and prevented suicide: Werther versus
Papageno effects. British Journal of Psychiatry, 197, 234–243.
https://doi.org/10.1192/bjp.bp.109.074633
Nock, M., & Favazza, A. (2009). Understanding nonsuicidal self-in-
jury: Origins, assessment and treatment. In M. Nock (Ed.),
Nonsuicidal self-injury: Definition and classification (pp. 9–18).
Washington, DC: American Psychological Association.
Olsen, M. & Fazio, R. (2002). Implicit acquisition and manifesta-
tion of classically conditioned attitudes. Social Cognition, 20,
89–102. https://doi.org/10.1521/soco.20.2.89.20992
Orbach, I., & Iohan-Barak, M. (2009). Psychopathology and risk
factors for suicide in the young: theoretical and empirical. In D.
Wasserman & C. Wasserman (Eds.), Oxford textbook of suicidol-
ogy and suicide prevention: A global perspective (pp. 633–642).
Oxford, UK: Oxford University Press.
Pew Research Center. (2015). Teens, social media and technology
overview 2015. Retrieved from http://www.pewinternet.org/­
2015/04/09/teens-social-media-technology-2015/
Phillips, D. P. (1974). The influence of suggestion on suicide:
Substantive and theoretical implications of the Werther ef-
fect. American Sociological Review, 39, 340–354. https://doi.
org/10.2307/2094294
Pirkis, J. E., Blood, R. W., Beautrais, A., Burgess, P. M., & Skehan, J.
(2006). Media guidelines on the reporting of suicide. Crisis, 27,
82–87. https://doi.org/10.1027/0227-5910.27.2.82
Reid-Chassiakos, Y., Radesky, J., Christakis, D., Moreno, M., & Cross,
C. (2016). Children and adolescents and digital media. Pediat-
rics, 138, e30162593. https://doi.org/10.1542/peds.2016-2593
Schmidke, A., & Häfner, H. (1988). The Werther effect after tel-
evision films: New evidence for an old hypothesis. Psycho-
logical Medicine, 18, 665–676. https://doi.org/10.1017/S00
33291700008345
Crisis 2018
Shneidman, E. (1987). A psychological approach to suicide. In G.
Vandenbos & B. Bryant (Eds.), Cataclysms, crises, and catastro-
phes (pp. 151–183). Washington, DC: American Psychological
Association.
Sonneck, G., Kapusta, N., Tomandl, G., & Voracek, M. (2012). Kri­
senintervention und Suizidverhütung [Crisis intervention and
suicide prevention]. Vienna, Austria: Facultas.
Stack, S. (2005). Suicide in the media: A quantitative review of
studies based on nonfictional stories. Suicide & Life-Threat-
ening Behavior, 35, 121–133. https://doi.org/10.1521/suli.35.2.
121.62877
Trappey, C. (1996). A meta-analysis of consumer choice and sublim-
inal advertising. Psychology & Marketing, 13, 517–530. https://
doi.org/10.1002/(SICI)1520-6793(199608)13:5<517::AID-
MAR5>3.0.CO;2-C
World Health Organization. (2017a). Suicide. Fact sheet. Retrieved
from http://www.who.int/mediacentre/factsheets/fs398/en/
World Health Organization. (2017b). Preventing suicide: A resource
for media professionals. Update 2017. Retrieved from http://
www.who.int/mental_health/suicide-prevention/resource_
booklet_2017/en/
Received July 27, 2017
Revision received December 21, 2017
Accepted December 22, 2017
Published online June 21, 2018
Florian Arendt, PhD, is a postdoctoral researcher at the Department of
Communication Science and Media Research, LMU Munich, Germany.
His primary research interests lie in the field of health communication.
Florian Arendt
Department of Communication Science and Media Research
University of Munich (LMU)
Oettingenstraße 67
80538 Munich
Germany
florian.arendt@ifkw.lmu.de
© 2018 Hogrefe Publishing