Critical appraisal of research
https://doi.org/10.1177/0020764019852655
International Journal of Social Psychiatry 1 –14 © The Author(s) 2019 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0020764019852655 journals.sagepub.com/home/isp
E CAMDEN SCHIZOPH
Introduction
Suicide represents a serious public health burden, with over 800,000 people dying as a result of suicide worldwide each year (Milner et al., 2017; World Health Organization, 2015). Suicide is the second most frequent cause of death among young people, with about 20 to 30 times as many suicide attempts (Wasserman et al., 2015). Suicide attempts and severe suicidal ideation have potentially serious consequences, including substantial psychological effects and death. Suicidal behaviour also has a profound negative impact on relatives and other people in the individual’s life (Wyman et al., 2008). A number of different preventive efforts have been evaluated in research, and a compre hensive review was recently published by Zalsman et al. (2016) presenting the effectiveness of existing preventive approaches. Education interventions (e.g. gatekeeper training or schoolbased education interventions) are
widely used strategies designed to improve early identification of persons at high risk of suicide (Robinson, Cox, & Malone, 2013; Schilling et al., 2014; Zalsman et al., 2016). At present, little is known regarding the collective effectiveness of these interventions in managing suicidality. The objective of this systematic review and metaanalysis is to provide a rigorous evaluation of the existing body of evidence for the effectiveness of educational interventions in preventing suicidal behaviour.
The effects of educational interventions on suicide: A systematic review and meta-analysis
Isabella Pistone , Ulrika Beckman, Erik Eriksson, Helena Lagerlöf and Morten Sager
Abstract Background: Suicide is a major public health problem. Educational interventions for preventing suicidal behaviour are widely used, although little is known regarding the collective effectiveness of these interventions. Aim: We evaluated the existing evidence for the effectiveness of education interventions in the prevention of suicidal behaviour. Methods: In this systematic review and meta-analysis, we searched multiple databases using terms related to suicide prevention. The articles were reviewed by two independent reviewers, and the quality of evidence was rated according to Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria. Data from individual studies were combined in meta-analyses. Results: We identified 41 studies evaluating two different types of interventions: school-based education interventions and gatekeeper training. Education interventions showed significant gains in terms of knowledge and attitudes, though the effects seem to vary depending on subjects’ personal characteristics. School-based education interventions significantly reduced suicidal ideation and suicide attempts in youths. Gatekeeper training showed no significant effect on suicide attempts or gatekeeper skills, though the quality of evidence for the estimate, according to GRADE criteria, was rated as very low. Conclusion: The results of this study indicate that school-based education interventions are effective in preventing suicidal ideation and suicide attempts. In clinical practice, as well as in research, the development and implementation of educational interventions should focus on participants’ individual characteristics.
Keywords Suicide prevention, systematic review, meta-analysis, educational intervention, school-based intervention, gatekeeper training
Department of Philosophy, Linguistics and Theory of Science, University of Gothenburg, Gothenburg, Sweden
Corresponding author: Isabella Pistone, Department of Philosophy, Linguistics and Theory of Science, University of Gothenburg, Olof Wijksgatan 6, 412 60 Gothenburg, Sweden. Email: isabella.pistone@gu.se
852655 ISP0010.1177/0020764019852655International Journal of Social PsychiatryPistone et al. review-article2019
Review Article
2 International Journal of Social Psychiatry 00(0)
Methods
Search strategy and selection criteria
This is a systematic review and metaanalysis. We searched Scopus, PubMed, PsycINFO and ASSIA in two separate searches for all relevant English, Swedish, Danish and Norwegian language studies published from database inception to 7 May 2017. The initial search used the Medical Subject Headings (MESH) identifiers for ‘suicide’, including subheadings as well as ‘suicid*’ in abstracts/titles. ‘Suicide’ was then combined with terms such as ‘suicide/ prevention & control’, ‘crisis intervention/education’, ‘gatekeeping’, ‘health education’, ‘patient education as topic’ and ‘suicide intervention’. Criteria for study inclusion were (1) studies using a randomised controlled trial (RCT) or quasiexperimental design, (2) studies evaluating education interventions and (3) studies including suicidal behaviour (suicide, attempts and ideation) and intermediate outcomes (knowledge, attitudes and skills). Criteria for study exclusion were (1) interventions developed for and conducted in the specific context of prisons, in health care or the national defence forces (universal manualbased interventions – for example, question, persuade, refer [QPR] – conducted in those settings were included), and (2) studies evaluating promotion, treatment and maintenance interventions. I.P. performed the searches. I.P. and E.E. also performed a number of manual searches to ensure that all relevant literature was identified.
Data analysis
Two reviewers independently screened the titles and abstracts of all the articles found. Half of the abstracts were screened by I.P. and H.L., and the other half by I.P. and E.E. Potentially relevant articles were retrieved, and the full texts were assessed for inclusion in the systematic review. Any disagreements regarding the inclusion of studies were resolved by discussion, sometimes using M.S. as a third reviewer. The external and internal validity of each study was assessed independently by the reviewers I.P., H.L., E.E. and U.B., using a checklist developed by the Swedish Council on Health Technology Assessment and Assessment of Social Services (SBU). The checklist contains a total of 31 questions that refer to standard criteria (e.g. mode of randomisation, level of blinding and dropout rate for the assessment of RCT). Studies were rated as having high, moderate or low risk of bias. We excluded studies rated as having a high risk of bias. GRADE methodology was used to assess the strength of evidence for the combined results for each outcome (Balshema et al., 2011; Guyatt et al., 2008). GRADE enables an overall rating of confidence in estimates of effect for each of the evaluated outcomes in a systematic review. According to GRADE methodology, the quality of evidence was classified as high: high confidence in the estimate, which with high probability will not be
changed by new research; moderate: the estimate is likely to be true, but could be changed by new research; low: the estimate is probably true but may be changed by new research; or very low: little confidence in the estimate, high probability that the estimate will be changed by the next published study (GRADE Working Group, 2018). In this systematic review, the preliminary confidence was set as high for the included RCT and low for the quasiexperimental studies. Thereafter, the confidence was judged with respect to five domains: overall risk of bias, inconsistency between studies (i.e. heterogeneity in results), width of the confidence intervals for the estimate (imprecision), indirectness of study results and the risk of publication bias. The overall rating was applied to the body of evidence and not to individual studies (Balshema et al., 2011; Higgins & Green, 2011). I.P., U.B. and E.E. extracted data that enabled evaluation of the primary outcomes: suicide attempts and suicide ideation, as well as the intermediate outcomes of knowledge, attitudes and skills (i.e. gatekeeper referrals of suicidal persons). Because the heterogeneity of populations and methodology in studies evaluating knowledge and attitudes did not permit a formal metaanalysis, we conducted a narrative analysis of those outcomes. Meta analyses were conducted on the outcomes: skills, suicide attempts and suicide ideation. Variables for which data were extracted were number of participants in the intervention group and control group. Data on the proportion of participants engaging in subsequent episodes of gatekeeper referral of suicidal persons, suicidal ideation and attempted suicide were summarised for each study using the odds ratio (OR) and 95% confidence intervals (CIs). Studyspecific effect sizes were combined using random effects meta analysis. Analyses were performed in RevMan for Mac, version 5.3 (Higgins & Green, 2011). I2 statistics were used to assess variability within studies, by convention, with I2 ⩾ 75% representing substantial betweenstudy heterogeneity (Higgins & Thompson, 2002).
Results
Our literature searches identified a total of 35,709 articles. Screening of titles and abstracts resulted in 393 papers that were retrieved in full text. These papers were assessed for eligibility, and 351 were excluded due to irrelevance or high risk of bias. A total of 41 studies were included in the systematic review (Figure 1).
The included studies evaluated two types of education interventions: gatekeeper training and schoolbased education interventions targeting youths. The interventions were conducted in a variety of arenas, such as university, minority societies, schools and workplaces. All included studies were assessed as having moderate risk of bias (Table 1). The two types of education interventions did not seem to differ much with respect to content. They both included teaching material about suicide risk factors and
Pistone et al. 3
how to identify a suicidal person, and empowering the subjects to intervene when confronted with a person who is exhibiting these symptoms. The main difference between the intervention types is the potential directness of the intervention: gatekeeper training is provided to subjects who are intended to be gatekeepers for a specific population, for example, teachers who act as gatekeepers for pupils. In contrast, the schoolbased interventions target youths who can be either a gatekeeper for a friend or themselves be at risk of suicide and potentially helped directly by the education programme. While the former is mainly an indirect intervention, the latter is both indirect and potentially direct.
The effect on knowledge was evaluated in 25 studies representing 12,914 subjects. The heterogeneity between studies did not permit a formal metaanalysis. All studies showed an increase in knowledge about suicide and suicide prevention in the intervention group compared to the control group. The effect sizes of this increase varied across studies. The followup time varied widely across studies, from 3 to 18 months. Four studies (Chagnon, Houle, Marcoux, & Renaud, 2007; Ghoncheh, Gould, Twisk, Kerkhof, & Koot, 2016; Jacobson, Osteen, Sharpe, & Pastoor, 2012; Wyman et al., 2008) showed a reduction in the gains with respect to the subjects’ level of know ledge at 3 to 6month followup compared with the level attained immediately after training. Nevertheless, the improvement in knowledge remained significantly greater than the level observed prior to the intervention. One study (Shaffer, Garland, Vieland, Underwood, & Busner, 1991)
showed a significant gain in knowledge after the intervention compared to the control group, but that this gain did not remain significant at 18month followup (Vieland, Whittle, Garland, Hicks, & Shaffer, 1991), while another study (Svensson & Hansson, 2014) showed significant effects at 24month followup. One study evaluating gatekeeper training showed that education improved knowledge of suicide among teachers but showed no significant positive effect on knowledge in service staff (Wyman et al., 2008). The quality of evidence for knowledge was considered low due to the risk of bias and inconsistency in results between studies (Table 2).
The effects on attitudes were evaluated in 18 studies, representing 9,645 participants. Attitudes were measured using different scales across studies. The scales used to measure attitudes vary in quality and the results across studies were very heterogeneous. Due to this heterogeneity, it was hard to reach a conclusion concerning the overall effect of educational interventions on attitudes. The quality of evidence for the outcome attitude was considered very low due to the risk of bias and inconsistency in results between studies (Table 2).
The effects of gatekeeper skills were evaluated in 14 studies, representing 7,729 subjects, whereas 6 studies provided data that permitted metaanalysis (Figure 2). Gatekeeper skills were measured by asking the subjects in the studies if they had identified and referred a suicidal person to appropriate resources since the intervention. Metaanalysis (representing six studies, 1,348 subjects) showed no increase in gatekeeper skills in the intervention group compared to the control group at 3 to 6month followup (OR = 0.97, 95% CI = [0.77, 1.22], p = .80). There was no indication of heterogeneity between studies (I2 = 0%). There was a lack of studies evaluating skills at a longer followup. The quality of evidence for the estimates was rated as very low due to the risk of bias (short followup time) and imprecision of data.
Fifteen studies evaluated the effects on suicidal ideation, representing 18,338 subjects. Suicidal ideation was measured in the majority of studies by asking subjects or the population which gatekeepers were intended to help: ‘During the past 3 months, did you ever seriously consider attempting suicide (yes or no)?’ Seven studies measured suicide ideation using a suicidal scale, though these were not included in the metaanalysis. Suicidal ideation was measured in different populations depending on the type of education intervention, that is, studies evaluating gatekeeper training measure a population other than the subjects receiving the education, while studies evaluating school based education interventions measure the subjects receiving the education. Because of this difference, two separate metaanalyses were conducted for the two types of interventions. While a metaanalysis (including two studies, total of 5,552 subjects) shows no significant decrease in suicidal ideation in the intervention group at 3month
Figure 1. Flowchart.
4 International Journal of Social Psychiatry 00(0) T
a b
le 1
. In
cl ud
ed s
tu di
es .
A ut
ho rs
( ye
ar ), c
o un
tr y
In te
rv en
ti o n
St ud
y de
si gn
Lo ng
es t
fo llo
w -u
p af
te r
ba se
lin e
N um
be r
o f su
bj ec
ts (
n)
N am
e o f in
te rv
en ti o n
Ex te
nt o
f ed
uc at
io n
St ud
y su
bj ec
ts
O ut
co m
e m
ea su
re Ef
fe ct
s ES
( C
o he
n’ s
d) , [
95 %
C I]
b et
w ee
n gr
o up
s at
f o llo
w -u
p O
R [
95 %
C I]
R is
k o f
bi as
A bb
ey , M
ad se
n, a
nd P
o lla
nd
(1 98
9) , U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
O ne
m o nt
h n
= 7
3
N o n
am e
Le ct
ur es
in c
o m
bi na
ti o n
w it h
st ud
y m
at er
ia l d
ur in
g 1
m o nt
h U
ni ve
rs it y
st ud
en ts
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES M
o de
ra te
A se
lt in
e, J am
es , S
ch ill
in g,
a nd
G
la no
vs ky
( 20
07 ), U
SA Sc
ho o l- ba
se d
ed uc
at io
n R
C T
T hr
ee m
o nt
hs n
= 4
,1 33
Si gn
s o f su
ic id
e T
w o d
ay s
ed uc
at io
n Pu
pi ls
in s
ch o o l
D oe
s no
t pr
ov id
e ef
fe ct
s ad
ju st
ed f or
th
e cl
us te
r ra
nd om
is at
io n
K no
w le
dg e
ES =
0 .4
5 [0
.4 1,
0 .4
9] At
tit ud
e ES
= 0
.2 5
[0 .2
3, 0
.2 7]
G at
ek ee
pe r
sk ill
s O
R =
1 .0
3 [0
.8 6,
1 .2
4] Su
ic id
al id
ea tio
n O
R =
0 .8
4 [0
.6 9,
1 .0
3] Su
ic id
e at
te m
pt O
R =
0 .6
6 [0
.4 7,
0 .9
1]
M o de
ra te
B ur
ns e
t al
. ( 20
17 ), A
us tr
al ia
G at
ek ee
pe r
tr ai
ni ng
R C
T T
w o m
o nt
hs n
= 1
81
M H
FA T
w o d
ay s
ed uc
at io
n U
ni ve
rs it y
st ud
en ts
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES
M o de
ra te
C ha
gn o n,
H o ul
e, M
ar co
ux ,
an d
R en
au d
(2 00
7) , U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
Si x
m o nt
hs n
= 7
1
N o n
am e,
d es
ig ne
d by
M o nt
re al
S ui
ci de
C en
tr e
O ne
t ra
in in
g da
y/ w
ee k
du ri
ng 3
w ee
ks Pe
o pl
e w
o rk
in g
w it h
yo ut
hs
K no
w le
dg e
ES =
2 .0
2 [1
.7 1,
2 .3
3] At
tit ud
e ES
= 0
.7 4
[0 .4
7, 1
.0 1]
M o de
ra te
Eg ge
rt , T
ho m
ps o n,
H er
ti ng
, an
d N
ic ho
la s
(1 99
5) , U
SA Sc
ho o l- ba
se d
ed uc
at io
n Q
ua si
-e xp
er im
en ta
l T
en m
o nt
hs n
= 1
05
N o n
am e
55 m
in ut
es /d
ay in
6 m
o nt
hs Pu
pi ls
in s
ch o o l
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
Eg ge
rt , T
ho m
ps o n,
R el
l, an
d Pi
ke (
20 02
), U
SA Sc
ho o l- ba
se d
ed uc
at io
n R
C T
T en
w ee
ks n
= 2
38
C -C
A R
E T
w o -h
o ur
in di
vi du
al c
o un
se lli
ng , 2
-h o ur
c ri
si s
in te
rv en
ti o n
an d
12 h
o ur
s ed
uc at
io n
Pu pi
ls in
s ch
o o l
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
Fi tz
pa tr
ic k,
W it te
, a nd
Sc
hm id
t (2
00 5)
, U SA
Sc ho
o l- ba
se d
ed uc
at io
n R
C T
T w
o m
o nt
hs n
= 1
10
Pr o bl
em -s
o lv
in g
in te
rv en
ti o n,
n o n
am e
35 m
in ut
es v
id eo
e du
ca ti o n
Pu pi
ls in
s ch
o o l
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
Pistone et al. 5
(C on
tin ue
d)
A ut
ho rs
( ye
ar ), c
o un
tr y
In te
rv en
ti o n
St ud
y de
si gn
Lo ng
es t
fo llo
w -u
p af
te r
ba se
lin e
N um
be r
o f su
bj ec
ts (
n)
N am
e o f in
te rv
en ti o n
Ex te
nt o
f ed
uc at
io n
St ud
y su
bj ec
ts
O ut
co m
e m
ea su
re Ef
fe ct
s ES
( C
o he
n’ s
d) , [
95 %
C I]
b et
w ee
n gr
o up
s at
f o llo
w -u
p O
R [
95 %
C I]
R is
k o f
bi as
G ho
nc he
h, G
o ul
d, T
w is
k,
K er
kh o f,
an d
K o o t
(2 01
6) ,
N et
he rl
an ds
G at
ek ee
pe r
tr ai
ni ng
R C
T T
hr ee
m o nt
hs n
= 1
90
M en
ta l h
ea lt h
o nl
in e
Ei gh
t In
te rn
et m
o du
le s
Pe o pl
e w
o rk
in g
w it h
yo ut
hs
K no
w le
dg e
ES =
0 .5
6 [0
.1 2,
1 .0
1] M
o de
ra te
H en
ni g,
C ra
bt re
e, a
nd B
au m
(1
99 8)
, U SA
Sc ho
o l- ba
se d
ed uc
at io
n Q
ua si
-e xp
er im
en ta
l Se
ve n
w ee
ks n
= 3
96
W he
n fr
ie nd
s ar
e hu
rt in
g 90
m in
ut es
le ct
ur e
Pu pi
ls in
s ch
o o l
G at
ek ee
pe r
sk ill
s O
R =
0 .9
0 [0
.3 7,
2 .1
9] M
o de
ra te
H o o ve
n, W
al sh
, P ik
e, a
nd
H er
ti ng
( 20
12 ), U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
15 m
o nt
hs n
= 4
72
C -C
A R
E pl
us P
-C A
R E
T w
o d
ay s
ed uc
at io
n an
d a
fo llo
w -u
p te
le ph
o ne
c al
l Pa
re nt
s o f hi
gh -r
is k
yo ut
hs
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s Su
ic id
e at
te m
pt D
o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s
M o de
ra te
Ja co
bs o n,
O st
ee n,
S ha
rp e,
an
d Pa
st o o r
(2 01
2) , U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
Fo ur
m o nt
hs n
= 7
3
Q PR
90 m
in ut
es e
du ca
ti o n
du ri
ng o
ne t
ra in
in g
o cc
as io
n U
ni ve
rs it y
st ud
en ts
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES G
at ek
ee pe
r sk
ill s
O R
= 1
.0 5
[0 .3
9, 2
.8 3]
M o de
ra te
Je ns
en , M
o rt
ho rs
t,
V en
ds bo
rg , H
jo rt
ho j,
an d
N o rd
en to
ft (
20 16
), D
en m
ar k
G at
ek ee
pe r
tr ai
ni ng
R C
T Si
x m
o nt
hs n
= 5
66
M H
FA T
w o d
ay s
ed uc
at io
n Pe
o pl
e w
o rk
in g
w it h
so ci
al s
er vi
ce s
K no
w le
dg e
ES =
0 .4
4 [0
.1 5,
0 .7
4] At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES G
at ek
ee pe
r sk
ill s
ES =
0 .1
2 (d
o n
o t
pr o vi
de d
at a
o n
C I)
M o de
ra te
Jo rm
, K it ch
en er
, F is
ch er
, a nd
C
ve tk
o vs
ki (
20 10
), A
us tr
al ia
G at
ek ee
pe r
tr ai
ni ng
R C
T Si
x m
o nt
hs n
= 2
62
M H
FA T
w o d
ay s
ed uc
at io
n G
en er
al p
ub lic
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES G
at ek
ee pe
r sk
ill s
ES =
0 .2
2 (d
o n
o t
pr o vi
de d
at a
o n
C I)
M o de
ra te
Jo rm
, K it ch
en er
, S aw
ye r,
Sc
al es
, a nd
C ve
tk o vs
ki
(2 01
0) , A
us tr
al ia
G at
ek ee
pe r
tr ai
ni ng
R C
T Si
x m
o nt
hs n
= 4
23
Y -M
H FA
T w
o d
ay s
ed uc
at io
n T
ea ch
er s
in s
ch o o l
K no
w le
dg e
ES =
0 .5
4 [0
.1 48
, 0 .9
24 ]
At tit
ud e
D o n
o t
pr o vi
de d
at a
o n
ES
M o de
ra te
T a b
le 1
. (C
o nt
in ue
d)
6 International Journal of Social Psychiatry 00(0)
A ut
ho rs
( ye
ar ), c
o un
tr y
In te
rv en
ti o n
St ud
y de
si gn
Lo ng
es t
fo llo
w -u
p af
te r
ba se
lin e
N um
be r
o f su
bj ec
ts (
n)
N am
e o f in
te rv
en ti o n
Ex te
nt o
f ed
uc at
io n
St ud
y su
bj ec
ts
O ut
co m
e m
ea su
re Ef
fe ct
s ES
( C
o he
n’ s
d) , [
95 %
C I]
b et
w ee
n gr
o up
s at
f o llo
w -u
p O
R [
95 %
C I]
R is
k o f
bi as
K al
af at
a nd
E lia
s (1
99 4)
, U SA
Sc ho
o l- ba
se d
ed uc
at io
n Q
ua si
-e xp
er im
en ta
l T
hr ee
w ee
ks n
= 2
53
N o n
am e
45 m
in ut
es e
du ca
ti o n
o n
th re
e o cc
as io
ns Pu
pi ls
in s
ch o o l
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES
M o de
ra te
K it ch
en er
a nd
J o rm
( 20
04 ),
A us
tr al
ia G
at ek
ee pe
r tr
ai ni
ng R
C T
Si x
m o nt
hs n
= 3
01
M H
FA T
w o d
ay s
ed uc
at io
n Pe
o pl
e w
o rk
in g
w it hi
n so
ci al
s er
vi ce
s
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES G
at ek
ee pe
r sk
ill s
O R
= 1
.1 3
[0 .7
1, 1
.8 1]
M o de
ra te
K lin
gm an
a nd
H o ch
do rf
(1
99 3)
, I sr
ae l
Sc ho
o l- ba
se d
ed uc
at io
n R
C T
Si x
m o nt
hs n
= 2
37
C o pi
ng s
tr at
eg ie
s, n
o n
am e
O ne
h o ur
f o r
12 w
ee ks
Pu pi
ls in
s ch
o o l
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
La nc
as te
r et
a l.
(2 01
4) ,
A us
tr al
ia G
at ek
ee pe
r ed
uc at
io n
R C
T N
o f o llo
w -u
p n
= 1
05
Q PR
90 m
in ut
es e
du ca
ti o n
V o lu
nt ee
rs
K no
w le
dg e
ES =
1 .1
3 [0
.8 0,
1 .4
6] M
o de
ra te
La Fr
o m
bo is
e an
d H
o w
ar d-
Pi tn
ey (
19 95
), U
SA Sc
ho o l- ba
se d
ed uc
at io
n Q
ua si
-e xp
er im
en ta
l Ei
gh t
m o nt
hs n
= 1
28
T he
Z un
i L ife
S ki
lls D
ev el
o pm
en t
C ur
ri cu
lu m
90 m
in ut
es /w
ee k
du ri
ng 3
0 w
ee ks
Pu pi
ls in
s ch
o o l
K no
w le
dg e
ES =
0 [
–0 .2
5, 0
.2 5]
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s
M o de
ra te
Li ps
o n,
S pe
er , B
ru nw
as se
r,
H ah
n, a
nd E
is en
be rg
( 20
14 ),
U SA
G at
ek ee
pe r
tr ai
ni ng
R C
T T
hr ee
m o nt
hs n
= 5
53
M H
FA T
w o d
ay s
ed uc
at io
n C
o lle
ge s
tu de
nt s
At tit
ud e
ES =
0 .0
2 [–
0. 64
, 0 .0
28 ]
G at
ek ee
pe r
sk ill
s O
R =
0 .8
7 [0
.6 1,
1 .2
4]
M o de
ra te
M cL
ea n
an d
Sw an
br o w
B
ec ke
r (2
01 7)
, U SA
G at
ek ee
pe r
tr ai
ni ng
R C
T Fo
ur m
o nt
hs n
= 1
62
N o n
am e
90 m
in ut
es e
du ca
ti o n
U ni
ve rs
it y
st ud
en ts
G at
ek ee
pe r
sk ill
s D
o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
M o ha
tt , B
o ec
km an
n, W
in ke
l, M
o ha
tt , a
nd S
ho re
( 20
17 ),
U SA
G at
ek ee
pe r
tr ai
ni ng
R C
T Ei
gh t
m o nt
hs n
= 1
76
M H
FA Ei
gh t
ho ur
s ed
uc at
io n
M ili
ta ry
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES G
at ek
ee pe
r sk
ill s
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s
M o de
ra te
T a b
le 1
. (C
o nt
in ue
d)
Pistone et al. 7
(C on
tin ue
d)
A ut
ho rs
( ye
ar ), c
o un
tr y
In te
rv en
ti o n
St ud
y de
si gn
Lo ng
es t
fo llo
w -u
p af
te r
ba se
lin e
N um
be r
o f su
bj ec
ts (
n)
N am
e o f in
te rv
en ti o n
Ex te
nt o
f ed
uc at
io n
St ud
y su
bj ec
ts
O ut
co m
e m
ea su
re Ef
fe ct
s ES
( C
o he
n’ s
d) , [
95 %
C I]
b et
w ee
n gr
o up
s at
f o llo
w -u
p O
R [
95 %
C I]
R is
k o f
bi as
O rb
ac h
an d
B ar
-J o se
ph
(1 99
3) , I
sr ae
l Sc
ho o l- ba
se d
ed uc
at io
n R
C T
T hr
ee w
ee ks
n =
2 15
N o n
am e
T w
o h
o ur
s/ w
ee k
du ri
ng s
ev en
w ee
ks Pu
pi ls
in s
ch o o l
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
O ’R
ei lly
, B el
l, K
el ly
, a nd
C
he n
(2 01
1) , A
us tr
al ia
G at
ek ee
pe r
tr ai
ni ng
R C
T Si
x w
ee ks
n =
2 72
M H
FA 12
h o ur
s ed
uc at
io n
U ni
ve rs
it y
st ud
en ts
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
es D
o n
o t
pr o vi
de d
at a
o n
ES
M o de
ra te
Sa nd
le r,
T ei
n, W
o lc
hi k,
a nd
A
ye rs
( 20
16 )
G at
ek ee
pe r
tr ai
ni ng
R C
T 15
m o nt
hs n
= 2
44
Fa m
ily b
er ea
ve m
en t
pr o gr
am m
e U
nc le
ar Fa
m ily
Su ic
id al
id ea
tio n
O R
= 0
.4 2
[0 .1
6, 1
.0 8]
M o de
ra te
Sa re
en e
t al
. ( 20
13 ), C
an ad
a G
at ek
ee pe
r tr
ai ni
ng R
C T
Si x
m o nt
hs n
= 5
5
A SI
ST T
w o d
ay s
ed uc
at io
n du
ri ng
a w
ee ke
nd Pu
pi ls
in s
ch o o l
G at
ek ee
pe r
sk ill
s O
R =
0 .3
2 [0
.0 7,
1 .4
5] M
o de
ra te
Sh an
no nh
o us
e, L
in , S
ha w
, an
d Po
rt er
( 20
17 ), U
SA G
at ek
ee pe
r tr
ai ni
ng Q
ua si
-e xp
er im
en ta
l 18
m o nt
hs n
= 1
49
A SI
ST T
w o d
ay s
ed uc
at io
n Sc
ho o l p
er so
nn el
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES
M o de
ra te
Sc hm
id t, C
ap ro
n, R
ai ne
s, a
nd
A lla
n (2
01 4)
, U SA
Sc ho
o l- ba
se d
ed uc
at io
n R
C T
O ne
m o nt
h n
= 1
08
C A
ST 45
m in
ut es
e du
ca ti o n
Pu pi
ls in
s ch
o o l
Su ic
id al
id ea
tio n
D o n
o t
pr o vi
de d
at a
o n
O R
o r
ca se
s M
o de
ra te
Sh af
fe r,
G ar
la nd
, V ie
la nd
, U
nd er
w o o d,
a nd
B us
ne r
(1 99
1) , U
SA
Sc ho
o l- ba
se d
ed uc
at io
n Q
ua si
-e xp
er im
en ta
l 18
m o nt
hs n
= 1
,4 38
N o n
am e
90 m
in ut
es e
du ca
ti o n
du ri
ng o
ne t
ra in
in g
o cc
as io
n Pu
pi ls
in s
ch o o l
K no
w le
dg e
D o n
o t
pr o vi
de d
at a
o n
ES M
o de
ra te
Sc hi
lli ng
e t
al . (
20 14
), U
SA Sc
ho o l- ba
se d
ed uc
at io
n R
C T
T hr
ee m
o nt
hs n
= 3
86
Si gn
s o f su
ic id
e T
w o d
ay s
ed uc
at io
n Pu
pi ls
in s
ch o o l
K no
w le
dg e
ES =
0 .4
0 (d
o n
o t
pr o vi
de d
at a
o n
C I)
At tit
ud e
D o n
o t
pr o vi
de d
at a
o n
ES Su
ic id
e at
te m
pt O
R =
0 .7
4 [0
.4 7,
1 .1
7]
M o de
ra te
T a b
le 1
. (C
o nt
in ue
d)
8 International Journal of Social Psychiatry 00(0)
A ut
ho rs
( ye
ar ), c
o un
tr y
In te
rv en
ti o n
St ud
y de
si gn
Lo ng
es t
fo llo
w -u
p af
te r
ba se
lin e
N um
be r
o f su
bj ec
ts (
n)
N am
e o f in
te rv
en ti o n
Ex te
nt o
f ed
uc at
io n
St ud
y su
bj ec
ts
O ut
co m
e m
ea su
re Ef
fe ct
s ES
( C
o he
n’ s
d) , [
95 %
C I]
b et
w ee
n gr
o up
s at
f o llo
w -u
p O
R [
95 %
C I]
R is
k o f
bi as
Sc hi
lli ng
, A se
lt in
e, a
nd J am
es
(2 01
6) , U
SA Sc
ho o l- ba
se d
ed uc
at io
n R
C T
T hr
ee m
o nt
hs n
= 1
,0 46
Si gn
s o f su
ic id
e T
w o d
ay s
ed uc
at io
n Pu
pi ls
in s
ch o o l
K no
w le
dg e
ES =
0 .4
2 [0
.3 4,
0 .5
0] At
tit ud
e ES
= 0
.2 0
[0 .1
6, 0
.2 4]
Su ic
id al
id ea
tio n
O R
= 0
.7 4
[0 .4
7, 1
.1 7]
Su ic
id e
at te
m pt
O nl
y pr
o vi
de s
ba se
lin e
m ea
su re
s –
in te
rv en
ti o n:
n =
5 , c
o nt
ro l:
n =
0
M o de
ra te
St ru
nk , S
o rt
er , K
in g,
a nd
V
id o ur
ek (
20 14
), U
SA Sc
ho o l- ba
se d
ed uc
at io
n R
C T
N o ne
n =
1 ,5
47
Su rv
iv in
g th
e te
en s
Fo ur
d ay
s ed
uc at
io n
Pu pi
ls in
s ch
o o l
K no
w le
dg e
ES =
0 .2
0 [–
0. 02
, 0 .4
3] M
o de
ra te
Sv en
ss o n
an d
H an
ss o n
(2 01
4) , S
w ed
en G
at ek
ee pe
r tr
ai ni
ng R
C T
T w
o y
ea rs
n =
4 06
M H
FA 12
h o ur
s ed
uc at
io n
Pe o pl
e w
o rk
in g
w it hi
n so
ci al
s er
vi ce
s
K no
w le
dg e
ES =
0 .6
1 [0
.3 5,
0 .8
7] At
tit ud
e D
o n
o t
pr o vi
de d
at a
o n
ES G
at ek
ee pe
r sk
ill s
ES =
0 .2
( do
n o t
pr o vi
de d
at a
o n
C I)
M o de
ra te
T o m
pk in
s an
d W
it t
(2 00
9) ,
U SA
G at
ek ee
pe r
tr ai
ni ng
Q ua
si -e
xp er
im en
ta l
Si x
m o nt
hs n
= 1
46
Q PR
60 m
in ut
es e
du ca
ti o n
du ri
ng o
ne t
ra in
in g
o cc
as io
n U
ni ve
rs it y
st ud
en ts
K no
w le
dg e
ES =
0 .3
6 [0
.2 9,
0 .4
4] G
at ek
ee pe
r sk
ill s
ES =
0 .0
2 (d
o n
o t
pr o vi
de d
at a
o n
C I)
M o de
ra te
T o m
pk in
s, W
it t, a
nd
A jb
ra ib
es h
(2 01
0) , U
SA G
at ek
ee pe
r tr
ai ni
ng Q
ua si
-e xp
er im
en ta
l T
hr ee
m o nt
hs n
= 1
44
Q PR
60 m
in ut
es e
du ca
ti o n
du ri
ng o
ne t
ra in
in g
o cc
as io
n U
ni ve
rs it y
st ud
en ts
K no
w le
dg e
ES =
0 .1
0 [–
0. 10
, 0 .3
1] At
tit ud
e ES
= 0
.5 7
[0 .3
9, 0
.7 5]
M o de
ra te
T o um
bo ur
o u
an d
G re
gg
(2 00
2) , U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
T hr
ee m
o nt
hs n
= 5
77
PA C
E Se
ve n
ed uc
at io
n o cc
as io
ns Pa
re nt
s an
d pe
o pl
e w
o rk
in g
w it h
yo ut
hs
Su ic
id al
id ea
tio n
O R
= 0
.7 2
[0 .2
9– 1.
77 ]
M o de
ra te
T sa
i, Li
n, C
ha ng
, Y u,
a nd
C
ho u
(2 01
1) , T
ai w
an G
at ek
ee pe
r tr
ai ni
ng R
C T
N o f o llo
w -u
p n
= 1
95
SA P
90 m
in ut
es e
du ca
ti o n
N ur
se s
w o rk
in g
in h
o sp
it al
K no
w le
dg e
ES =
0 .8
0 [0
.7 0,
0 .9
0] M
o de
ra te
T a b
le 1
. (C
o nt
in ue
d)
Pistone et al. 9
A ut
ho rs
( ye
ar ), c
o un
tr y
In te
rv en
ti o n
St ud
y de
si gn
Lo ng
es t
fo llo
w -u
p af
te r
ba se
lin e
N um
be r
o f su
bj ec
ts (
n)
N am
e o f in
te rv
en ti o n
Ex te
nt o
f ed
uc at
io n
St ud
y su
bj ec
ts
O ut
co m
e m
ea su
re Ef
fe ct
s ES
( C
o he
n’ s
d) , [
95 %
C I]
b et
w ee
n gr
o up
s at
f o llo
w -u
p O
R [
95 %
C I]
R is
k o f
bi as
W as
se rm
an e
t al
. ( 20
15 ), 1
0 Eu
ro pe
an c
o un
tr ie
s G
at ek
ee pe
r tr
ai ni
ng R
C T
12 m
o nt
hs n
= 4
,5 75
Q PR
90 m
in ut
es e
du ca
ti o n
du ri
ng o
ne t
ra in
in g
o cc
as io
n T
ea ch
er s
w o rk
in g
in s
ch o o ls
Su ic
id al
id ea
tio n
O R
= 0
.7 6
[0 .4
5, 1
.2 8]
Su ic
id e
at te
m pt
O R
= 0
.5 9
[0 .3
1, 1
.1 2]
M o de
ra te
Sc ho
o l- ba
se d
ed uc
at io
n R
C T
12 m
o nt
hs n
= 4
,5 75
Y A
M T
hr ee
-h o ur
r o le
-p la
y, 2
-h o ur
le ct
ur e
an d
se lf-
he lp
m
at er
ia l
Pu pi
ls in
s ch
o o l
Su ic
id al
id ea
tio n
O R
= 0
.5 5
[0 .2
9, 1
.0 1]
Su ic
id e
at te
m pt
O R
= 0
.4 6
[0 .2
5, 0
.8 7]
W ilc
o x
et a
l. (2
00 8)
, U SA
Sc ho
o l- ba
se d
ed uc
at io
n R
C T
15 y
ea rs
n =
5 00
G B
G T
w o y
ea rs
w it h
re gu
la r
pl ay
m ee
ti ng
s Pu
pi ls
in s
ch o o l
Su ic
id al
id ea
tio n
O R
= 0
.5 3
[0 .3
0, 0
.9 2]
Su ic
id e
at te
m pt
O R
= 0
.7 5
[0 .4
2, 1
.3 2]
M o de
ra te
W ym
an e
t al
. ( 20
08 ), U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
12 m
o nt
hs n
= 2
50
Q PR
90 m
in ut
es e
du ca
ti o n
du ri
ng o
ne t
ra in
in g
o cc
as io
n T
ea ch
er s
an d
se rv
ic e
st af
f w
o rk
in g
in s
ch o o ls
K no
w le
dg e
ES =
0 .3
7 [–
1. 12
, 1 .8
8] At
tit ud
e ES
= 0
.7 5
[0 .5
3, 0
.9 7]
G at
ek ee
pe r
sk ill
s O
R =
1 .2
2 [0
.6 8,
2 .1
8]
M o de
ra te
W ym
an e
t al
. ( 20
10 ), U
SA G
at ek
ee pe
r tr
ai ni
ng R
C T
Fo ur
m o nt
hs
So ur
ce o
f st
re ng
th Fo
ur t
o s
ix h
o ur
s ed
uc at
io n
du ri
ng o
ne t
ra in
in g
o cc
as io
n Pu
pi ls
( pe
er le
ad er
s) in
s ch
o o l
G at
ek ee
pe r
sk ill
s ES
= 0
.2 1
[– 0.
01 , 0
.4 1]
M
o de
ra te
ES : e
ffe ct
s iz
e; C
I: co
nf id
en ce
in te
rv al
; O R
: o dd
s ra
ti o ; R
C T
: r an
do m
is ed
c o nt
ro lle
d tr
ia l;
M H
FA : m
en ta
l h ea
lt h
fir st
a id
; Q PR
: q ue
st io
n, p
er su
ad e,
r ef
er ; A
SI ST
: a pp
lie d
su ic
id e
in te
rv en
ti o n
sk ill
t ra
in in
g;
C A
ST : c
o gn
it iv
e an
xi et
y se
ns it iv
it y
tr ea
tm en
t; S
A P:
s ui
ci de
a w
ar en
es s
pr o gr
am m
e; G
B G
: g o o d
be ha
vi o ur
g am
e; C
-C A
R E:
c o un
se lo
rs c
ar e,
a ss
es s,
r es
po nd
, e m
po w
er ; P
-C A
R E:
p ar
en ts
c ar
e; Y
-M H
FA :
yo ut
h m
en ta
l h ea
lt h
fir st
a id
; P A
C E:
p ar
en ti ng
a do
le sc
en ts
, a c
re at
iv e
ex pe
ri en
ce ; Y
A M
: y o ut
h aw
ar e
o f m
en ta
l h ea
lt h.
T a b
le 1
. (C
o nt
in ue
d)
10 International Journal of Social Psychiatry 00(0)
followup (OR = 0.82, 95% CI = [0.56, 1.19], p = .29, I2 = 21%) (Figure 3), results from another metaanalysis (Figure 4) using data from 12month followup show a
significant decrease in suicidal ideation (OR = 0.54, 95% CI = [0.35, 0.81], p = .003, I2 = 0%). The quality of evidence for the estimates at 12month followup was rated as
Table 2. Summary of findings using Grading of Recommendations Assessment, Development and Evaluation (GRADE).
Outcome Number of studies; subjects Follow-up
Effect [95% CI] Confidence in estimates
Conclusion
Knowledge 25; 12,914 0–18 months
ES varying between 0.10 and 2.02 across studies
Low It is possible that education interventions can increase knowledge
Attitude 18; 9,645 0–24 months
ES varying between 0.15 and 0.79 across studies
Very low High probability that the estimate will be changed by the next published study
Gatekeeper skills (calculated on the estimate of the meta-analysis)
6; 1,348 0–12 months
OR = 0.97, 95% CI = [0.77, 1.22], p = .80
Very low High probability that the estimate will be changed by the next published study
Suicidal ideation (calculated on the estimate of the meta-analysis) School-based education interventions Gatekeeper training
2; 4,922 12 months 3; 5,360 3–12 months
OR = 0.54, 95% CI = [0.35, 0.81], p = .003 OR = 0.68, 95% CI = [0.45, 1.01], p = .06
Moderate Very low
School-based education probably do decrease suicidal ideation High probability that the estimate will be changed by the next published study
Suicide attempts (calculated on the estimate of the meta-analysis) School-based education interventions Gatekeeper training
2; 4,913 12 months 2; 5,047 12–15 months
OR = 0.60, 95% CI = [0.38, 0.95], p = .03 OR = 0.59, 95% CI = [0.31, 1.12], p = .229 RC = –1.033 (not significant)
Moderate Very low
School-based education interventions probably decrease suicide attempts in youths High probability that the estimate will be changed by the next published study
CI: confidence interval; ES: effect size; OR: odds ratio; RC: rate of change.
Figure 2. Meta-analysis of gatekeeper skills.
Figure 3. Meta-analysis of suicidal ideation, school-based education at 3-month follow-up.
Pistone et al. 11
moderate. Three studies (Sandler, Tein, Wolchik, & Ayers, 2016; Toumbourou & Gregg, 2002; Wasserman et al., 2015) evaluating gatekeeper training (5,360 subjects) generated data that allowed a metaanalysis (Figure 5). There was no significant decrease in suicidal ideation in the intervention group compared to the control group at 3month followup (OR = 0.68, 95% CI = [0.45, 1.01], p = .06). There was no indication of heterogeneity between studies (I2 = 0%), and the quality of evidence for the estimate of gatekeeper education was rated as very low due to the risk of bias (short followup time) and imprecision of data.
Six studies evaluated the effect on suicide attempts. Suicide attempts were measured by asking subjects or the population which the gatekeepers were supposed to help: ‘During the past X months, did you actually attempt suicide (yes or no)?’ Appropriate aggregate data could be extracted from three studies (Aseltine, James, Schilling, & Glanovsky, 2007; Schilling, Aseltine, & James, 2016; Wasserman et al., 2015) evaluating schoolbased interventions aimed at youths (total of 9,661 subjects), thus enabling metaanalysis.
Results showed a significant decrease in suicide attempts in the intervention group compared to the control group at 3month followup (OR = 0.56, 95% CI = [0.39, 0.80], p = .001) (Figure 6) as well as for 12month followup (OR = 0.60, 95% CI = [0.38, 0.95], p = .03) (Figure 7). There was no indication of high heterogeneity between studies in any of the metaanalyses.
Two studies not included in the metaanalysis evaluated gatekeeper training. One of those showed a significant positive effect on 3month followup and showed no decrease in suicide attempts at 3 or 12month followup (Hooven, Walsh, Pike, & Herting, 2012; Wasserman et al., 2015). The quality of evidence for the estimates of the schoolbased interventions was considered moderate, while the quality of evidence for the estimates of the gatekeeper intervention was rated very low due to the risk of bias and imprecision of data.
Discussion
The results of this systematic review and metaanalysis show that educational interventions were effective with respect to
Figure 4. Meta-analysis of suicidal ideation, school-based education at 12-month follow-up.
Figure 5. Meta-analysis of suicidal ideation, gatekeeper training.
Figure 6. Meta-analysis of suicidal attempt, school-based education at 3-month follow-up.
12 International Journal of Social Psychiatry 00(0)
gains in knowledge about suicide and suicide prevention, though the quality of evidence for the outcome was rated as low, indicating that the conclusions are probably correct, but may be changed as a consequence of new research. The effects of educational interventions on attitudes varied widely across studies, and the quality of evidence for the outcome was rated as very low. The heterogeneity of effect sizes across studies as well as subgroup analysis conducted by included studies (Tompkins & Witt, 2009; Wyman et al., 2008) indicate that the effect was probably dependent on individual aspects of the study subjects. These findings correspond to a previously published review by Burnette, Ramchand and Ayer (2015) evaluating gatekeeper training. They concluded that individual aspects such as sex, age, job role and prior training were of relevance for the effects on knowledge and attitudes. Although education has been statistically shown to increase knowledge and attitudes, the practical relevance of this increase is largely unknown at present. A gain in knowledge is only relevant if it leads to an actual change in behaviour. There is also a lack of evidence supporting the suggestion that a gain in knowledge and attitude leads to an actual decrease in suicidal behaviour. Future studies should focus on investigating the role of individual aspects of subjects and how the intermediary outcomes correlate to primary outcomes.
Metaanalysis showed no significant effect on gatekeeper skills, though the sample size of the included stu dies might not be sufficient to detect small effect sizes, which is probably the case when evaluating gatekeeper skills. Also, the short followup time is probably not sufficient to allow appropriate evaluation of this outcome. The quality of evidence was therefore rated very low, indicating a need for larger studies with longer followup times to evaluate the effect of education studies on gatekeeper skills. The metaanalysis of suicidal ideation at 12 months indicates that schoolbased education was effective in decreasing suicidal ideation in youth. The quality of evidence was rated as moderate, indicating that the estimate is most likely correct and will probably not be changed as a consequence of future studies.
From this we can conclude that schoolbased education interventions probably do have a preventive effect on suicidal ideation. The metaanalysis of suicide attempts indicated that
schoolbased education interventions have a positive effect on the outcome, with the quality of evidence rated as moderate. In contrast, the estimate for suicide attempts in gatekeeper training showed no significant positive effect. These findings suggest that changes in suicidal behaviour are perhaps more likely if the targeted popu lation is personally involved in the intervention as both gatekeepers and target population, as in schoolbased interventions, than with interventions where the gatekeeper receiving training is merely supposed to work as a mediator for another population, for example, teachers as gatekeepers for pupils. The effectiveness of schoolbased interventions on suicidal behaviour has previously also demonstrated effects in other systematic reviews (KlimesDougan, Klingbeil, & Meller, 2013; Robinson et al., 2013; Schilling et al., 2014). School based education interventions therefore merit serious consideration by policy makers and practitioners when considering preventive strategies that target youths.
No study reporting on suicide was identified by the literature search, and the paucity of such research is a major limitation in the research literature. Future studies should attempt to evaluate this primary outcome, though this may be difficult due to the large study sample size required to detect the small number of suicide events. The main limitation of this systematic review is the absence of evaluation or comparison of specific educational programmes. While it would probably be of interest to practitioners and policymakers to compare the effects of the different schoolbased educational programmes, this was beyond the scope of this systematic review. Despite this, the systematic review and metaanalysis were conducted using a rigorous evaluation design. The use of GRADE is a strength in this systematic review, providing clear conclusions based on the body of evidence (Guyatt et al., 2008), in order to help guide future development of education interventions, research and inform policy decisions.
Authors’ note
The authors of this report had full access to all data in the study.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article:
Figure 7. Meta-analysis of suicidal attempt, school-based education at 12-month follow-up.
Pistone et al. 13
The study was initiated and funded by the Public Health Agency of Sweden.
ORCID iD
Isabella Pistone https://orcid.org/0000000152728958
References
Abbey, K. J., Madsen, C. H., & Polland, R. (1989). Shortterm suicide awareness curriculum. Suicide and Life-Threatening Behavior, 19, 216–227.
Aseltine, R. H., Jr., James, A., Schilling, E. A., & Glanovsky, J. (2007). Evaluating the SOS suicide prevention program: A replication and extension. BMC Public Health, 7, 161. doi:10.1186/147124587161
Balshema, H., Helfanda, M., Holger, J., Schunemann, H. J., Oxman, A. D., Kunz, R., ... Guyatt, G. H. (2011). GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology, 64, 401–406.
Burnette, C., Ramchand, R., & Ayer, L. (2015). Gatekeeper training for suicide prevention: A theoretical model and review of the empirical literature. Santa Monica, LA: RAND National Defense Research Institute.
Burns, S., Crawford, G., Hallett, J., Hunt, K., Chih, H. J., & Tilley, P. J. (2017). What’s wrong with John? A randomised controlled trial of Mental Health First Aid (MHFA) training with nursing students. BMC Psychiatry, 17, 111. DOI 10.1186/s1288801712782
Chagnon, F., Houle, J., Marcoux, I., & Renaud, J. (2007). Controlgroup study of an intervention training program for youth suicide prevention. Suicide and Life-Threatening Behavior, 37, 135–144.
Eggert, L. L., Thompson, E. A., Herting, J. R., & Nicholas, L. J. (1995). Reducing suicide potential among highrisk youth: Tests of a schoolbased prevention program. Suicide and Life-Threatening Behavior, 25, 276–296.
Eggert, L. L., Thompson, E. A., Rell, B. P., & Pike, K. C. (2002). Preliminary effects of brief schoolbased prevention approaches for reducing youth suiciderisk behaviors, depression, and drug involvement. Journal of Child and Adolescent Psychiatric Nursing, 15, 48–64.
Fitzpatrick, K. K., Witte, T. K., & Schmidt, N. B. (2005). Randomized controlled trial of a brief problemorientation intervention for suicidal ideation. Behavior Therapy, 36, 323–333.
Ghoncheh, R., Gould, M. S., Twisk, J. W., Kerkhof, A. J., & Koot, H. M. (2016). Efficacy of adolescent suicide prevention elearning modules for gatekeepers: A randomized controlled trial. JMIR Mental Health, 3, e8. doi:10.2196/ mental.4614
GRADE Working Group. (2018, August). Retrieved from www. gradeworkinggroup.org
Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck Ytter, Y., AlonsoCoello, P., & Schunemann, H. J. (2008). GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. British Medical Journal, 336, 924–926.
Hennig, C. W., Crabtree, C. R., & Baum, D. (1998). Mental health CPR: Peer contracting as a response to potential suicide in adolescents. Archives of Suicide Research, 4, 169–187.
Higgins, J. P. T. & Green, S. (Eds.). (2011). Cochrane handbook for systematic reviews of interventions. The Cochrane Collaboration. Available from www.handbook.cochrane. org
Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying heterogeneity in a metaanalysis. Statistics in Medicine, 15, 1539–1558.
Hooven, C., Walsh, E., Pike, K. C., & Herting, J. R. (2012). Promoting CARE: Including parents in youth suicide prevention. Family & Community Health, 35, 225–235.
Jacobson, J. M., Osteen, P. J., Sharpe, T. L., & Pastoor, J. B. (2012). Randomized trial of suicide gatekeeper training for social work students. Research on Social Work Practice, 22, 270–281.
Jensen, K. B., Morthorst, B. R., Vendsborg, P. B., Hjorthoj, C., & Nordentoft, M. (2016). Effectiveness of Mental Health First Aid training in Denmark: A randomized trial in waitlist design. Social Psychiatry and Psychiatric Epidemiology, 51, 597–606.
Jorm, A. F., Kitchener, B. A., Fischer, J. A., & Cvetkovski, S. (2010). Mental Health First Aid training by elearning: A randomized controlled trial. Australian and New Zealand Journal of Psychiatry, 44, 1072–1081.
Jorm, A. F., Kitchener, B. A., Sawyer, M. G., Scales, H., & Cvetkovski, S. (2010). Mental Health First Aid training for high school teachers: A cluster randomized trial. BMC Psychiatry, 10, 51. doi:10.1186/1471244X1051
Kalafat, J., & Elias, M. (1994). An evaluation of a school based suicide awareness intervention. Suicide and Life- Threatening Behavior, 24, 224–233.
Kitchener, B. A., & Jorm, A. F. (2004). Mental Health First Aid training in a workplace setting: A randomized controlled trial [ISRCTN13249129]. BMC Psychiatry, 4, 23. doi:10.1186/1471244X423
KlimesDougan, B., Klingbeil, D. A., & Meller, S. J. (2013). The impact of universal suicideprevention programs on the help seeking attitudes and behaviors of youths. Crisis, 34, 82–97.
Klingman, A., & Hochdorf, Z. (1993). Coping with distress and self harm: The impact of a primary prevention program among adolescents. Journal of Adolescence, 16, 121–140.
LaFromboise, T., & HowardPitney, B. (1995). The Zuni life skills development curriculum: Description and evaluation of a suicide prevention program. Journal of Counseling Psychology, 42, 479–486.
Lancaster, P. G., Moore, J. T., Putter, S. E., Chen, P. Y., Cigularov, K. P., Baker, A., & Quinnett, P. (2014). Feasibility of a web based gatekeeper training: Implications for suicide prevention. Suicide and Life-Threatening Behavior, 44, 510–523.
Lipson, S. K., Speer, N., Brunwasser, S., Hahn, E., & Eisenberg, D. (2014). Gatekeeper training and access to mental health care at universities and colleges. Journal of Adolescent Health, 55, 612–619.
McLean, K., & Swanbrow Becker, M. A. (2018). Bridging the gap: Connecting resident assistants and suicidal residents through gatekeeper training. Suicide and Life-Threatening Behavior, 48, 218–229.
Milner, A., Witt, K., Pirkis, J., Hetrick, S., Robinson, J., Currier, D., ... Carter, G. L. (2017). The effectiveness of suicide prevention delivered by GPs: A systematic review and metaanalysis. Journal of Affective Disorders, 210, 294–302.
14 International Journal of Social Psychiatry 00(0)
Mohatt, N. V., Boeckmann, R., Winkel, N., Mohatt, D. F., & Shore, J. (2017). Military Mental Health First Aid: Development and preliminary efficacy of a community training for improving knowledge, attitudes, and helping behaviors. Military Medicine, 182, e1576–e1583.
Orbach, I., & BarJoseph, H. (1993). The impact of a suicide prevention program for adolescents on suicidal tendencies, hopelessness, ego identity, and coping. Suicide and Life- Threatening Behavior, 23, 120–129.
O’Reilly, C. L., Bell, J. S., Kelly, P. J., & Chen, T. F. (2011). Impact of Mental Health First Aid training on pharmacy students’ knowledge, attitudes and selfreported behaviour: A controlled trial. Australian and New Zealand Journal of Psychiatry, 45, 549–557.
Robinson, J., Cox, G., & Malone, A. (2013). A systematic review of schoolbased interventions aimed at preventing, treating, and responding to suiciderelated behavior in young people. Crisis, 34, 164–182.
Sandler, I., Tein, J. Y., Wolchik, S., & Ayers, T. S. (2016). The effects of the family bereavement program to reduce suicide ideation and/or attempts of parentally bereaved children six and fifteen years later. Suicide and Life-Threatening Behavior, 46, 32–38.
Sareen, J., Isaak, C., Bolton, S. L., Enns, M. W., Elias, B., Deane, F., ... Katz, L. Y. (2013). Gatekeeper training for suicide prevention in First Nations community members: A randomized controlled trial. Depression & Anxiety, 30, 1021–1029.
Schilling, E. A., Aseltine, R. H., Jr., & James, A. (2016). The SOS Suicide Prevention Program: Further evidence of efficacy and effectiveness. Prevention Science, 17, 157–166.
Schilling, E. A., Lawless, M., FauBuchanan, L., Buchanan, L., Fau Aseltine, R. H., Jr., & Aseltine, R. H. (2014). ‘Signs of Suicide’ shows promise as a middle school suicide prevention program. Suicide and Life-Threatening Behavior, 44, 653–667.
Schmidt, N. B., Capron, D. W., Raines, A. M., & Allan, N. P. (2014). Randomized clinical trial evaluating the efficacy of a brief intervention targeting anxiety sensitivity cognitive concerns. Journal of Consulting and Clinical Psychology, 82, 1023–1033.
Shaffer, D., Garland, A., Vieland, V., Underwood, M., & Busner, C. (1991). The impact of curriculumbased suicide prevention programs for teenagers. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 588–596.
Shannonhouse, L., Lin, Y. W. D., Shaw, K., & Porter, M. (2017). Suicide intervention training for K12 schools: A quasi experimental study on ASIST. Journal of Counseling & Development, 95, 3–13.
Strunk, C. M., Sorter, M. T., King, K. A., & Vidourek, R. A. (2014). Effectiveness of the surviving the Teens(R) suicide prevention and depression awareness program: An impact evaluation utilizing a comparison group. Health Education & Behavior, 41, 605–613.
Svensson, B., & Hansson, L. (2014). Effectiveness of Mental Health First Aid training in Sweden. A randomized controlled trial with a sixmonth and twoyear followup. PLoS ONE, 9, e100911. doi:10.1371/journal.pone.0100911
Tompkins, T. L., & Witt, J. (2009). The shortterm effectiveness of a suicide prevention gatekeeper training program in a college setting with residence life advisers. Journal of Primary Prevention, 30, 131–149.
Tompkins, T. L., Witt, J., & Ajbraibesh, N. (2010). Does a gatekeeper suicide prevention program work in a school setting? Evaluating training outcome and moderators of effectiveness. Suicide and Life-Threatening Behavior, 39, 671–681.
Toumbourou, J. W., & Gregg, M. E. (2002). Impact of an empowermentbased parent education program on the reduction of youth suicide risk factors. Journal of Adolescent Health, 31, 277–285.
Tsai, W. P., Lin, L. Y., Chang, H. C., Yu, L. S., & Chou, M. C. (2011). The effects of the Gatekeeper SuicideAwareness Program for nursing personnel. Perspectives in Psychiatric Care, 47, 117–125.
Vieland, V., Whittle, B., Garland, A., Hicks, R., & Shaffer, D. (1991). The impact of curriculumbased suicide prevention programs for teenagers: An 18month followup. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 811–815.
Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, G., ... Carli, V. (2015). School based suicide prevention programmes: The SEYLE clusterrandomised, controlled trial. The Lancet, 385, 1536–1544.
Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., & Anthony, J. C. (2008). The impact of two universal randomized first and secondgrade classroom interventions on young adult suicide ideation and attempts. Drug and Alcohol Dependence, 95, 60–73.
World Health Organization. (2015). WHO suicide data. Retrieved from http://www.who.int/mediacentre/factsheets/fs398/en/
Wyman, P. A., Brown, C. H., Inman, J., Cross, W., Schmeelk Cone, K., Guo, J., & Pena, J. B. (2008). Randomized trial of a gatekeeper program for suicide prevention: 1year impact on secondary school staff. Journal of Consulting and Clinical Psychology, 76, 104–115.
Wyman, P. A., Brown, H., LoMurray, M., SchmeelkCone, K., Petrova, M., Yu, Q., ... Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. American Journal of Public Health, 100, 1653– 1661.
Zalsman, G., Hawton, K., Wasserman, D., van Heeringen, K., Arensman, E., Sarchiapone, M., ... Zohar, J. (2016). Suicide prevention strategies revisited: 10year systematic review. Lancet Psychiatry, 3, 646–659.