Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=ijmh20

Journal of Mental Health

ISSN: 0963-8237 (Print) 1360-0567 (Online) Journal homepage: http://www.tandfonline.com/loi/ijmh20

Training and post-disaster interventions forthe psychological impacts on disaster-exposedemployees: a systematic review

Samantha K. Brooks, Rebecca Dunn, Richard Amlôt, Neil Greenberg & G.James Rubin

To cite this article: Samantha K. Brooks, Rebecca Dunn, Richard Amlôt, Neil Greenberg& G. James Rubin (2018): Training and post-disaster interventions for the psychologicalimpacts on disaster-exposed employees: a systematic review, Journal of Mental Health, DOI:10.1080/09638237.2018.1437610

To link to this article: https://doi.org/10.1080/09638237.2018.1437610

Published online: 15 Feb 2018.

Submit your article to this journal

View related articles

View Crossmark data

http://tandfonline.com/ijmhISSN: 0963-8237 (print), 1360-0567 (electronic)

J Ment Health, Early Online: 1–25� 2018 Informa UK Limited, trading as Taylor & Francis Group. DOI: 10.1080/09638237.2018.1437610

REVIEW ARTICLE

Training and post-disaster interventions for the psychological impactson disaster-exposed employees: a systematic review

Samantha K. Brooks1 , Rebecca Dunn1, Richard Amlot2 , Neil Greenberg1* , and G. James Rubin1*

1Department of Psychological Medicine, King’s College London, London, UK, and 2Emergency Response Department Science and Technology,

Health Protection Directorate, Public Health England, Salisbury, UK

Abstract

Background: When organisations are exposed to traumatic situations, such as disasters, oftenstaff are not prepared for the potential psychological impact which can negatively affect theirwellbeing.Aims: To conduct a systematic review of the literature on psychological interventions aimed atimproving staff wellbeing during or after disasters.Method: Four electronic literature databases were searched. Reference lists of relevant articleswere hand-searched.Results: Fifteen articles were included. Five studies suggested that pre-disaster skills trainingand disaster education can improve employee confidence. Ten studies on post-disasterinterventions revealed mixed findings on the effectiveness of psychological debriefing andlimited evidence for cognitive behavioural therapy, psychoeducation and meditation.Conclusions: Pre-disaster training and education can improve employees’ confidence in theirability to cope with disasters. The routine use of post-disaster psychological debriefings is notsupported; further research is needed to determine if debriefing interventions could be usefulin some circumstances. Further research is needed to provide more evidence on the potentialpositive effects of cognitive behavioural therapy, psychoeducation and meditation. Moreexperimental studies on psychological disaster interventions are needed.

Keywords

Disasters, interventions, wellbeing, work

History

Received 3 August 2017Revised 15 November 2017Accepted 18 January 2018Published online 13 February 2018

Introduction

Disasters and traumatic events are occurring more often

across the world (Guha-Sapir et al., 2013), with earthquakes,

hurricanes, floods, explosions and terrorist attacks affecting

increasingly large numbers of people. Many people who are

exposed to trauma are exposed as part of an occupational

group: for example a commercial organisation targeted by

terrorists or healthcare workers who assist with the emergency

response.

There has been much research on the prevalence of

psychological disorders – such as post-traumatic stress

disorder (PTSD), depression, anxiety, and alcohol misuse –

in employees following disaster exposure. For example a

review by Strohmeier & Scholte (2015) revealed high levels

of trauma-related psychological problems in humanitarian

workers, and research has demonstrated a risk of mental

health problems in frequently trauma-exposed employees,

such as fire-fighters (Chang et al., 2005) and police (West

et al., 2008). A number of recent reviews exploring organ-

isational responses to disasters (Brooks et al., 2015, 2016,

2017) have identified factors affecting the development of

psychological disorders in occupational groups, suggesting

that the extent of traumatic exposure, lack of social support,

lack of preparedness and greater impact of the disaster on

one’s personal and professional life may all be associated with

poor psychological outcomes. These factors appear to be

important in determining wellbeing regardless of occupation –

the same factors were identified in a study of rescue workers,

who are involved in disaster recovery as part of their role

(Brooks et al., 2015), and in other occupational groups who

would not expect to be exposed to trauma (Brooks et al.,

2017).

While the prevalence of psychological problems post-

disaster and factors affecting psychological wellbeing after a

traumatic incident have been explored, there is less evidence

regarding how to support trauma-exposed staff, and what kind

of interventions might be successful in reducing the distress

experienced during and after disasters.

Groups who are routinely exposed to traumatic situations

as part of their everyday work – such as rescue workers, the

military and emergency services personnel – are likely to have

some employee support systems in place and to be prepared to

support their staff or colleagues after being exposed to

trauma. For example a recent article (Dunn et al., 2015)

suggested that preparatory mental health briefings at the start

*These authors contributed equally to this work.

Correspondence: Samantha K. Brooks Department of PsychologicalMedicine, King’s College London, London, UK. E-mail:samantha.k.brooks@kcl.ac.uk

of their role, monitoring of employees’ wellbeing and

informal support, such as trauma risk management (TRiM;

Greenberg et al., 2008) or psychological first aid (PFA;

Pekevski, 2012) could enable staff to better support their

colleagues. However, as Dunn et al. noted, there has been

little evidence on how to best support trauma-exposed

employees, and it is unclear how many organisations apply

evidence to their practice. Often some form of psychological

debriefing, or ‘‘Critical Incident Stress Debriefing’’

(Mitchell, 1983), occurs. This differs from the structured

debriefing often offered in workplaces which typically take

the form of ‘‘after action reviews’’ and review performance

rather than wellbeing, in that psychological debriefing is

designed specifically to prevent the development of post-

traumatic stress and similar symptoms, is conducted by

mental health professionals, and offers a structured approach

for employees to discuss their emotions (Mitchell, 1983).

There has been mixed evidence on the effectiveness of this

kind of intervention, with research suggesting that debriefing

is unhelpful and can even have negative consequences

(Wessely et al., 2000). This has led the National Institute

for Health and Care Excellence (NICE) to suggest that

individual debriefing sessions focusing on the trauma experi-

ence should not be part of routine practice (NICE, 2005).

The aim of this article was to review the evidence on

effective interventions to support employees during or after

traumatic incident such as a disaster. We explored: (1) Can

pre-disaster training equip employees with better coping skills

if a disaster were to occur? (2) What kinds of post-disaster

interventions have been evaluated? (3) To what extent do pre-

disaster and post-disaster interventions protect the wellbeing

of employees and promote resilience?

Method

Search strategy

This study forms part of a wider review on the psychological

impact of disasters on employees (Brooks et al., 2015, 2016,

2017): our previous publications emerging from this review

focus on the factors affecting the psychological wellbeing of

disaster-exposed employees, and the data on interventions has

until now not been reviewed. One large, broad search was

used to locate articles for each strand of the study. This

combined three smaller searches. Search 1 included psycho-

logical terms, such as ‘‘wellbeing’’ and ‘‘PTSD’’. Search 2

included terms relating to extreme events, and was informed

by the emergency events database (Centre for Research on the

Epidemiology of Disasters, 2009). Search 3 included terms

relating to occupational groups, such as ‘‘employees’’ and

‘‘organisation’’. The full search strategy can be seen in

Appendix I.

One author (SKB) ran the searches on MEDLINE�,

Embase, PsycINFO� and Web of Science. Resulting citations

were downloaded to EndNote� software version X7

(EndNote, Philadelphia, PA), where duplicate items were

immediately removed. SKB then evaluated titles for relevance

and removed any clearly not relevant to the review. Next, two

authors (SKB, RD) used the inclusion criteria to screen

abstracts of the remaining citations and then finally to screen

full texts. Reference lists of all remaining articles were then

hand-searched for any additional, relevant studies: if our

included articles contained any citations which appeared to be

relevant and had not been found through our search, we

looked up these papers and assessed their relevance by

abstract and then by full-text.

Inclusion criteria

Articles were eligible for inclusion if they:

� Were written in English.

� Were published in peer-reviewed journals.

� Included employed participants (any occupational group;

any group of people that work together within a

hierarchical system to achieve some sense of group aim).

� Involved a ‘‘disaster’’ or potential disaster – we took an

inclusive approach to the definition of ‘‘disaster’’ and

largely relied on authors’ own assessment of whether to

categorise the incident in this way or not.

� Involved any form of psychological intervention designed

to help employees in any occupational group cope either

during or after a disaster.

� Was published post-1984 (30 years before the start of our

study in 2014 – reducing the risk of including research

conducted before post-traumatic stress disorder was

introduced as a diagnostic category in the DSM-III-R

(American Psychiatric Association, 1980)).

Data extraction and synthesis

We designed spreadsheets with the following headings: first

author; publication year; country of study; disaster; primary

aim; design; occupation of participants; participant demo-

graphics; details of intervention; outcomes measured; instru-

ments used; key results; conclusions; and limitations. These

data were extracted by two authors (SKB, RD) from all

studies which remained after the full-text screening.

During the data extraction, it became clear that due to the

varying types of interventions, instruments used and out-

comes measured, meta-analysis would not be possible.

Instead, we extracted all relevant outcomes, grouped articles

into pre-disaster and post-disaster interventions, and grouped

articles with similar outcomes together where possible.

Quality appraisal

This study used a quality appraisal tool developed for a

previous review (Brooks et al., 2015) which had been

informed by existing quality appraisal tools (Drummond &

Jefferson, 1996; EPHPP, 2009; National Institute for Health,

2014). This tool assessed quality in three areas – study design,

data collection and methodology and analysis and interpret-

ation of results. The appraisal tool can be seen in Appendix II.

Results

The initial search yielded 18 005 studies. One hundred and

seventy of these were relevant to the overall set of reviews on

the psychological impact of disasters on occupational groups,

and 15 were relevant for inclusion in this review. Details of

screening can be seen in Figure 1.

Seven studies were longitudinal, seven were cross-

sectional and one was a randomised controlled trial.

2 S. K. Brooks et al. J Ment Health, Early Online: 1–25

The disasters explored in the literature included earthquakes

(n¼ 3), the 11 September terrorist attacks in New York

(n¼ 3), potential pandemics (n¼ 2), hurricanes (n¼ 2) and

terrorism in general, a simulated submarine disaster, a rail

crash, a bank robbery and a bombing (each n¼ 1). A

summary of the characteristics of articles included in this

review is presented in Table 1, and more detailed description

of each study’s intervention is presented in Table 2.

Overall quality was mixed (see Figure 2). Studies typically

scored highly for design. Scores for data collection and

methodology were more mixed, with several articles not using

standardised measures and having participation rates of less

than 50%. Analysis and interpretation of results tended to be

the weakest areas; many did not use statistical tests or

consider confounding variables.

One-third of the relevant articles involved employees who

had not yet experienced a disaster and evaluated pre-disaster

training interventions designed to improve confidence and

ability to cope should an incident occur, while the remaining

two-third looked at interventions to improve wellbeing after a

disaster had occurred.

Pre-disaster

Summary of studies

Five studies involved pre-disaster interventions aimed at

improving preparedness and ability to cope with disasters. Of

these, two involved Canadian hospital workers preparing for a

potential influenza epidemic (Aiello et al., 2011; Maunder

et al., 2010), one involved Norwegian Navy sailors participat-

ing in a simulated submarine disaster (Eid et al., 2004), one

involved occupational health workers preparing for terrorist

attacks in the USA (Gershon et al., 2004) and one involved

Records iden�fied through database searching

n = 17,999

Addi�onal records iden�fied through hand searching

n = 6

Total records found n = 18,005

Duplicates removed n = 6183

Titles screened n = 11,822

Abstracts screened n = 3655

Studies relevant for this review n = 15

Full-text ar�cles assessed for eligibility

n = 571

Studies relevant for broader set of reviews

n = 170

Excluded a�er �tle screening n = 8167

Excluded a�er abstract screening n = 3084

Excluded a�er full-text screening

n = 401

Relevant for other reviews but not interven�on-

related n = 155

Figure 1. Screening process.

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 3

Tab

le1

.S

um

mar

yo

fp

aper

s.

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

Aie

llo

etal

.(2

01

1)

Can

ada

Po

ten

tial

infl

u-

enza

pan

dem

icP

rean

dp

ost

-tra

inin

gsu

rvey

s1

25

0H

osp

ital

staf

fT

rain

ing

sess

ion

on

wh

atto

exp

ect

and

how

toco

pe.

Co

nfi

den

cein

abil

ity

tod

eal

wit

hp

an-

dem

icp

re-t

rain

-in

gan

db

elie

fth

ato

ne

wil

lb

eab

leto

cop

ed

uri

ng

ap

and

emic

po

st-

trai

nin

g;

asse

ssed

via

stu

dy-s

pec

ific

surv

ey.

On

ly3

4.9

%fe

ltco

nfi

den

tin

bei

ng

able

tod

eal

wit

hth

ep

and

emic

pre

-tr

ain

ing

;7

6.1

%fe

ltb

ette

rab

leto

cop

eas

are

sult

of

the

sess

ion

.8

7.5

%fo

un

dth

ese

ssio

n‘‘

use

ful’

’;9

1.3

%fo

un

dit

‘‘h

elp

ful

and

info

rmat

ive’

’an

d6

9.7

%fe

ltth

eyw

ere

pre

par

edfo

ra

pan

dem

icfo

l-lo

win

gth

ese

ssio

nB

osc

arin

oet

al.

(20

05

)U

SA

11

Sep

tem

ber

20

01

terr

ori

stat

tack

s.

Lo

ng

itu

din

al–

bas

elin

ed

ata

coll

ecte

d1

3–

15

mo

nth

sp

ost

-dis

aste

ran

dfo

llow

-up

dat

aco

llec

ted

24

mo

nth

sp

ost

-dis

aste

r.

16

81

Par

tici

pan

tsd

ivid

edin

tog

rou

ps

–n

ow

ork

site

sess

ion

atte

nd

ed(n¼

15

01

);o

ne

sess

ion

atte

nd

ed(n¼

82

);2

–3

sess

ion

sat

ten

ded

(n¼

68

)an

d4þ

sess

ion

sat

ten

ded

(n¼

30

).

Var

iou

sB

rief

emp

loyer

-pro

-v

ided

wo

rksi

tecr

isis

inte

rven

tio

ns.

Alc

oh

ol

abu

seas

sess

edb

yst

ud

y-s

pec

ific

qu

esti

on

san

dG

AG

Ecr

iter

ia;

sub

clin

ical

/par

tial

PT

SD

asse

ssed

by

mo

dif

ied

ver

sio

no

fP

TS

Dse

ctio

no

fN

IMH

-DIS

and

WH

Od

iag

-n

ost

icin

terv

iew

;d

epre

ssio

nas

sess

edb

yS

CID

’sd

epre

s-si

on

scal

e;p

sy-

chia

tric

syn

-d

rom

es(a

nx

iety

,so

mat

isat

ion

and

glo

bal

sever

ity

)as

sess

edb

yB

rief

Sy

mp

tom

Inven

tory

-18

.T

reat

men

tg

rou

ps

wer

eas

ked

tow

hat

exte

nt

ses-

sio

ns

hel

ped

them

dea

lw

ith

po

st-

dis

aste

rem

o-

tio

nal

pro

ble

ms

21

.2%

rep

ort

edth

ein

terv

enti

on

hel

ped

alo

t;2

2.8

%sa

idit

hel

ped

som

e;3

3.7

%sa

idit

hel

ped

ali

ttle

and

22

.4%

fou

nd

itn

ot

atal

lh

elp

ful.

Bet

wee

n1

and

3se

ssio

ns

app

eare

dto

be

effe

ctiv

ein

pro

tect

ing

wo

rk-

ers

fro

mb

inge

dri

nk

ing

(OR¼

0.2

6,

p5

0.0

5)

and

alco

ho

ld

epen

d-

ence

(OR¼

0.0

9,

p5

0.0

5).

2–

3se

ssio

ns

wer

ep

rote

ctiv

efo

rP

TS

D(O

R¼

0.3

6,

p5

0.0

5),

dep

res-

sio

n(O

R¼

0.2

3,

p5

0.0

5),

som

a-ti

sati

on

(OR¼

0.3

6,

p5

0.0

5),

anx

iety

(co

nti

nu

ed)

4 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

usi

ng

ast

ud

y-

spec

ific

scal

e.(O

R¼

0.1

7,

p5

0.0

1)

and

glo

bal

sever

ity

(OR¼

0.3

0,

p5

0.0

5)

Bo

scar

ino

etal

.(2

00

6)

US

A1

1S

epte

mb

er2

00

1te

rro

rist

atta

cks

Lo

ng

itu

din

al–

bas

elin

ed

ata

coll

ecte

d1

3–

15

mo

nth

sp

ost

-dis

aste

ran

dfo

llow

-up

dat

aco

llec

ted

24

mo

nth

sp

ost

-dis

aste

r.

11

21

Var

iou

sB

rief

emp

loyer

-pro

-v

ided

wo

rksi

tecr

isis

inte

rven

tio

ns.

As

above

Un

adju

sted

ou

t-co

mes

wer

egen

-er

ally

bet

ter

for

inte

rven

tio

ns

case

sfo

ral

lo

ut-

com

esw

ith

exce

pti

on

of

PT

SD

.A

fter

1:5

mat

chin

g,

wo

rk-

ers

wit

hem

plo

yer

inte

rven

tio

ns

had

less

sever

eP

TS

Dsy

mp

tom

s(�

0.8

)b

ut

no

dif

fere

nce

was

fou

nd

inP

TS

Dp

reval

ence

inp

ast

yea

r.P

rop

ensi

tysc

ore

resu

lts

bas

edo

n1

:5m

atch

ing

-in

terv

enti

on

wo

rker

sh

adb

ette

ro

utc

om

esat

foll

ow

-up

for

dep

ress

ion

(�2

.7%

,p5

0.0

5)

and

alco

ho

ld

epen

den

ce(�

4.8

%,

p5

0.0

1)

inp

ast

yea

ran

dre

du

ced

PT

SD

(�0

.8,

p5

0.0

5),

dep

ress

ion

and

anx

iety

sym

p-

tom

s(�

1.7

,p5

0.0

5)

inp

ast

mo

nth

Dif

ede

etal

.(2

00

7)

US

A1

1S

epte

mb

er2

00

1te

rro

rist

atta

cks

Ran

do

mis

edco

ntr

oll

edtr

ial

21

(7in

CB

Tin

terv

en-

tio

ng

rou

p,

14

intr

eatm

ent

asu

sual

(TA

U)

gro

up

).

Dis

aste

rw

ork

ers

Inte

rven

tio

ng

rou

p:

12

wee

kly

sess

ion

so

fco

g-

nit

ive

beh

avio

ura

lth

erap

y.T

AU

gro

up

:M

enta

l

PT

SD

,d

epre

ssio

n,

alco

ho

lp

rob

lem

san

dvar

iou

scl

in-

ical

dis

ord

ers;

mea

s-u

res

incl

ud

ed

An

AN

OV

Aex

am-

inin

gp

re/p

ost

chan

ges

inC

AP

Ssc

ore

sfo

un

da

sig

nif

ican

tm

ain

effe

cto

fti

me

(F

(co

nti

nu

ed)

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 5

Tab

le1

.C

on

tin

ued

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

hea

lth

trea

tmen

tco

mm

on

lyav

ail-

able

afte

ra

trau

-m

atic

even

t;p

arti

cip

ants

wer

eg

iven

resu

lts

of

thei

rP

TS

Dev

alu

atio

nan

dre

ferr

edb

ack

too

ccu

pat

ion

alh

ealt

hfo

ras

sist

-an

cew

ith

ob

tain

-in

gtr

eatm

ent

thro

ug

hco

mm

u-

nit

yp

rov

ider

s(t

ho

ug

hn

on

eto

ok

this

up

).

clin

icia

n-a

dm

inis

-te

red

PT

SD

Sca

le,

stru

ctu

red

clin

ical

inte

rvie

wfo

rD

SM

-IV

Ax

isI

dis

ord

ers,

life

stre

sso

rch

eck

list

-re

vis

ed,

PT

SD

chec

kli

st,

bec

kd

epre

ssio

nin

ven

-to

ry,

SC

L-9

0-R

,M

ich

igan

alco

ho

lsc

reen

ing

test

,so

cial

adju

stm

ent

scal

ean

dli

feev

ents

scal

e.

(1,

19

)¼

9.6

90

,p5

0.0

1);

asi

g-

nif

ican

tm

ain

effe

cto

fG

rou

p(F

(1,

19

)¼

6.3

17

,p5

0.0

5);

and

asi

gn

ific

ant

tim

eX

Gro

up

inte

ract

ion

(F(1

,1

9)¼

6.5

47

,p5

0.0

5).

Sig

nif

ican

td

ecre

ase

inC

AP

Ssc

ore

sfo

rC

BT

gro

up

com

par

edto

TA

Ug

rou

p(F¼

4.1

,p5

0.0

1).

No

ne

intr

eatm

ent

con

di-

tio

nin

crea

sed

inC

AP

Ssc

ore

bu

t2

1%

inT

AU

gro

up

did

incr

ease

.A

nA

NO

VA

exam

in-

ing

pre

/po

stch

anges

inB

DI

sco

res

fou

nd

atr

end

for

ati

me

XG

rou

pin

tera

ctio

n(F

(1,

16

)¼

3.4

57

,p5

0.0

81

)w

ith

no

sig

nif

ican

tm

ain

effe

cts.

An

AN

OV

Aex

amin

-in

gp

re/p

ost

chan

ges

inS

CL

-9

0G

SI

sco

res

fou

nd

atr

end

for

ati

me

XG

rou

pin

tera

ctio

n,

F(1

,1

8)¼

3.0

74

,p5

0.0

97

,an

da

sig

nif

ican

tm

ain

effe

cto

fG

rou

p,

F(1

,1

8)¼

7.0

62

,p5

0.0

16

,b

ut

no

(co

nti

nu

ed)

6 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

sig

nif

ican

tm

ain

effe

cto

fti

me.

No

sig

nif

ican

td

iffe

r-en

ces

inal

coh

ol

sco

res.

Eid

etal

.(2

00

4)

No

rway

Sim

ula

ted

sub

mar

ine

dis

aste

r

Ex

per

imen

tal

stu

dy;

surv

eys

con

du

cted

pre

-tra

inin

gan

dat

the

end

of

the

6-d

trai

nin

g.

18

Sai

lors

Dis

aste

rsi

mu

lati

on

exer

cise

Per

son

alit

yh

ard

i-n

ess

asse

ssed

by

sho

rtfo

rmo

fth

ed

isp

osi

tio

nal

resi

lien

cysc

ale;

cop

ing

style

asse

ssed

by

Co

pin

gS

tyle

Qu

esti

on

nai

re;

PT

SD

asse

ssed

by

Imp

act

Of

Even

tS

cale

and

PT

SS

-1

0;

qu

alit

yo

fli

feas

sess

edb

yG

HQ

-12

.

On

asc

ale

of

1–

5,

med

ian

sco

reo

f5

for

wh

eth

ersi

mi-

lar

exer

cise

ssh

ou

ldb

eh

eld

on

are

gu

lar

bas

is;

med

ian

of

4/5

for

the

exer

cise

rein

-fo

rcin

gth

eir

con

-fi

den

cein

thei

rjo

bs.

Mea

sure

so

nIE

S,

PT

SS

-10

or

GH

Q-1

2n

ever

exce

ed’’

low

sym

pto

m,

sam

eas

usu

al’’

du

rin

gth

eex

erci

se.

QO

Lm

easu

red

by

GH

Qsi

gn

ifi-

can

tly

incr

ease

dover

tim

e[F

(6,

14

)¼

4.0

,p5

0.0

1].

Str

ess

sym

pto

ms

mea

-su

red

by

IES

sig

-n

ific

antl

yd

ecre

ased

over

tim

e[F

(6,

14

)¼

2.7

6,

p5

0.0

5].

Per

son

alit

yh

ard

i-n

ess

was

neg

a-ti

vel

yas

soci

ated

wit

hIE

Sto

tal

(r¼�

0.5

2,

p5

0.0

5),

IES

intr

usi

on

(r¼�

0.4

7,

p5

0.0

5)

and

IES

avo

idan

ce(r¼�

0.4

7,

p5

0.0

5)

wh

ile

avo

idan

ceco

pin

g

(co

nti

nu

ed)

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 7

Tab

le1

.C

on

tin

ued

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

was

po

siti

vel

yas

soci

ated

wit

hG

HQ

-12

ind

icat

-in

gre

du

ced

QO

L(r¼

0.5

3,

p5

0.0

5).

Ger

sho

net

al.

(20

04

)U

SA

Ter

rori

smin

gen

eral

Cro

ss-s

ecti

on

al8

4O

ccu

pat

ion

alh

ealt

hw

ork

ers

Ter

rori

smed

uca

-ti

on

alw

ork

pla

cep

rog

ram

me

Co

nfi

den

cein

clin

-ic

alsk

ills

;p

erce

pti

on

of

wo

rksi

tere

adi-

nes

s;m

oti

vat

ion

totr

eat

terr

ori

smv

icti

ms;

mo

tiv-

atio

nto

rece

ive

smal

lpox

vac

cin

e–

all

mea

sure

dby

stu

dy-s

pec

ific

qu

esti

on

nai

re.

Incr

ease

dco

nfi

-d

ence

inp

rote

ct-

ing

self

-wh

ilst

cari

ng

for

vic

tim

so

fb

iote

rro

rism

(59

.7%

)an

dch

emic

alte

rro

r-is

mex

po

sure

(59

.7%

);re

cog

-n

isin

gsy

mp

tom

so

fte

rro

rism

exp

osu

re(8

1.7

%)

and

trea

tin

gv

ic-

tim

so

fch

emic

alex

po

sure

(69

.5%

)K

enar

dy

etal

.(1

99

6)

Au

stra

lia

Ear

thqu

ake

Lo

ng

itu

din

al(�

27

,5

0,

86

and

11

4w

eek

sp

ost

-dis

aste

r)

19

5(6

2d

ebri

efed

and

13

3n

ot

deb

rief

ed)

Em

ergen

cyse

rvic

esan

dd

isas

ter

wo

rker

sG

rou

pst

ress

deb

rief

ing

How

stre

ssfu

lth

eyfo

un

dth

ed

isas

ter,

asse

ssed

by

stu

dy-s

pec

ific

scal

e;p

sych

o-

log

ical

mo

rbid

ity

asse

ssed

by

Imp

act

of

Even

tssc

ale

and

GH

Q-

12

.

No

corr

elat

ion

bet

wee

nn

um

ber

of

deb

rief

ing

ses-

sio

ns

and

per

-ce

ived

hel

pfu

l-n

ess.

No

sig

nif

ican

tfi

nd

-in

gs

inre

lati

on

tod

ebri

efin

gh

elp

-fu

lnes

sfo

rth

eIE

So

rG

HQ

-12

.D

ebri

efin

gw

asre

po

rted

tob

e‘‘

ver

y’’

or

‘‘ex

trem

ely

’’h

elp

ful

by

34

%o

fp

arti

cip

ants

,‘‘

som

ewh

at’’

hel

pfu

lby

46

%an

d‘‘

no

tat

all’

’h

elp

ful

by

20

%.

Th

ose

wh

ow

ere

deb

rief

edre

po

rted

sig

nif

i-ca

ntl

yh

igh

erover

all

GH

Q-1

2sc

ore

s.O

ver

all,

(co

nti

nu

ed)

8 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

sub

ject

sin

the

deb

rief

edg

rou

psh

ow

edle

ssim

pro

vem

ent

insy

mp

tom

sth

anth

ose

wh

ow

ere

no

td

ebri

efed

and

this

was

tru

efo

rb

oth

the

IES

(Mea

nst

and

ar-

dis

edch

ang

ein

dex

:�

0.3

1–

0.5

6v

)an

dth

eG

HQ

-12

(Mea

nst

and

ard

ised

chan

ge

ind

ex:

0.1

0–

0.2

8).

Lev

elo

fh

elp

erst

ress

was

un

rela

ted

tod

ebri

efin

gei

ther

over

all

or

inre

la-

tio

nto

chan

ge

over

tim

e.T

her

ew

ere

sig

nif

ican

tm

ain

effe

cts

for

hel

per

stre

sso

nb

oth

the

IES

(F(1

,1

90

)¼

6.7

8,

p5

0.0

1)

and

the

GH

Q-1

2(F

(1,

19

0)¼

8.6

9,

p5

0.0

1)

con

-fi

rmin

gth

efa

ctth

atth

ose

rep

ort

-in

gg

reat

erh

elp

erst

ress

had

hig

her

level

so

fm

ea-

sure

dp

sych

op

ath

olo

gy

Mau

nd

eret

al.

(20

10

)C

anad

aP

ote

nti

alin

flu

-en

zap

and

emic

A‘‘

do

se-f

ind

ing

’’st

ud

yco

mp

ared

pre

-po

stch

anges

inth

ree

dif

-fe

ren

td

ura

tio

ns

of

trai

nin

g.

15

8H

osp

ital

wo

rker

sC

om

pu

ter-

assi

sted

‘‘P

and

emic

Str

ess

Vac

cin

e’’

trai

nin

gco

urs

e.P

arti

cip

ants

ran

-d

om

lyas

sig

ned

tosh

ort

cou

rse

(7se

ssio

ns,

Co

nfi

den

cein

trai

n-

ing

and

sup

po

rt;

per

ceiv

edef

fi-

cacy

toad

apt

top

and

emic

con

di-

tio

ns;

inte

rper

-so

nal

pro

ble

ms;

cop

ing

via

Sig

nif

ican

tim

pro

ve-

men

tsfr

om

the

star

tto

the

end

of

the

cou

rse

inp

and

emic

self

-ef

fica

cy,

con

fi-

den

cein

trai

nin

gan

dsu

pp

ort

and

(co

nti

nu

ed)

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 9

Tab

le1

.C

on

tin

ued

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

med

ian

11

1m

in);

med

ium

cou

rse

(12

sess

ion

s,1

58

min

);o

rlo

ng

cou

rse

(17

ses-

sio

ns,

22

3m

in).

pro

ble

m-s

olv

ing

,se

ekin

gsu

pp

ort

or

esca

pe-

avo

id-

ance

.N

ine-

item

SA

RS

scal

ead

apte

dfo

rin

flu

-en

za;

Of

24

-ite

mP

and

emic

Sel

f-E

ffic

acy

Sca

led

eriv

edfo

rth

isst

ud

y;

IIP

-32

;w

ays

of

cop

ing

inven

tory

.

inte

rper

son

alp

rob

lem

s.D

imen

sio

ns

of

inte

rper

son

alp

rob

lem

sw

hic

him

pro

ved

incl

ud

edp

rob

-le

ms

wit

hb

ein

gso

cial

lyin

hib

ited

,n

on

-ass

erti

ve,

over

lyac

com

mo

-d

atin

g,

self

-sa

crif

icin

gan

din

tru

sive/

nee

dy.

Co

pin

gw

ith

stre

ssu

sin

gp

rob

-le

m-s

olv

ing

,se

ekin

gsu

pp

ort

fro

mo

ther

so

res

cap

e-av

oid

ance

did

no

tch

ange

over

the

cou

rse

over

all:

how

ever

,fo

rth

ose

no

tre

po

rtin

gp

rob

-le

m-s

olv

ing

and

seek

ing

sup

po

rt,

ther

ew

ere

pre

-p

ost

imp

rove-

men

ts.

Am

on

gth

ose

un

der

-uti

-li

sin

gp

rob

lem

-so

lvin

gat

T1

ther

ew

asa

sig

-n

ific

ant

pre

-po

stin

crea

se(9

5%

CI

of

dif

fere

nce

0.2

–0

.4,

p5

0.0

01

).A

mo

ng

tho

seu

nd

er-u

tili

sin

gse

ekin

gsu

pp

ort

,th

ere

was

agai

nan

incr

ease

(95

%C

Io

fd

iffe

ren

ce0

.1–

0.3

,p¼

0.0

03

).A

mo

ng

tho

sew

ho

rep

ort

edan

yu

se

(co

nti

nu

ed)

10 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

of

esca

pe-

avo

id-

ance

atT

1,

ther

ew

asa

dec

reas

ein

this

typ

eo

fco

pin

g(9

5%

CI

of

dif

fere

nce

�0

.1–

0.0

,p¼

0.0

1).

Eac

hco

urs

ele

ng

thre

sult

edin

imp

roved

self

-ef

fica

cyan

dco

n-

fid

ence

intr

ain

-in

gan

dsu

pp

ort

.T

he

med

ium

-le

ng

thco

urs

eap

pea

red

tob

eo

pti

mal

.M

ille

r-B

urk

eet

al.

(19

99

)U

SA

Ban

kro

bb

erie

sC

ross

-sec

tio

nal

14

1B

ank

emp

loyee

sC

riti

cal

inci

den

tst

ress

deb

rief

ing

.W

ork

pro

du

ctiv

ity,

stre

ss,

physi

cal

hea

lth

,w

ork

rela

-ti

on

ship

s,p

er-

son

alre

lati

on

-sh

ips,

des

ire

tokee

pw

ork

ing

,d

ays

off

wo

rk;

psy

cho

log

ical

hea

lth

sym

pto

ms

–su

rvey

des

ign

edfo

rth

est

ud

y.

Of

47

%at

ten

ded

afo

rmal

lyp

re-

sen

ted

CIS

Dse

s-si

on

;al

mo

st8

of

10

bel

ieved

itto

be

hel

pfu

l(t

hes

era

tin

gs

wer

en

ot

asso

ciat

edw

ith

asp

ects

of

the

rob

ber

yco

nte

xt)

.7

7%

also

rate

din

terv

enti

on

sw

ith

co-w

ork

ers

assu

cces

sfu

l;6

4%

inte

rven

tio

ns

wit

hsu

per

vis

ors

;5

4%

wit

hfa

mil

y/

frie

nd

san

d4

8%

wit

hover

all

wo

rken

vir

on

men

t.O

f2

0%

con

sid

ered

over

all

wo

rken

vir

on

men

tto

hav

em

ade

thei

rre

cover

yw

ors

e.O

f1

1%

rep

ort

edth

atqu

esti

on

sfr

om

cust

om

ers,

po

lice

and

med

ia

(co

nti

nu

ed)

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 11

Tab

le1

.C

on

tin

ued

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

wer

eo

bst

acle

sto

reco

ver

yN

ort

het

al.

(20

02

)U

SA

Ok

lah

om

aC

ity

bo

mb

ing

Cro

ss-s

ecti

on

al1

81

Fir

e-fi

gh

ters

wh

ose

rved

asre

scu

ean

dre

cov-

ery

wo

rker

s.

Wo

rkp

lace

deb

rief

ing

sP

sych

iatr

icd

iag

-n

ose

s;d

iag

no

stic

inte

rvie

wsc

hed

ule

.

Of

92

%h

adp

arti

ci-

pat

edin

wo

rk-

pla

ced

efu

sin

gs/

deb

rief

ing

s.T

wo

-th

ird

was

sati

sfie

dw

ith

this

and

on

e-th

ird

dis

sati

sfie

d;

how

ever

,8

9%

said

they

wo

uld

reco

mm

end

the

inte

rven

tio

n.

PT

SD

was

no

tre

late

dto

sati

s-fa

ctio

no

rre

com

-m

end

atio

no

fth

ed

ebri

efin

gs.

How

ever

,p

arti

ci-

pan

tsw

ith

oth

er(n

on

-PT

SD

)d

is-

ord

ers

wer

ele

ssli

kel

yto

rep

ort

sati

sfac

tio

n(5

0v

s.7

1%

,x

2¼

6.9

3,

p5

0.0

1)

or

rec-

om

men

dth

em(8

2vs.

92

%,

x2¼

4.2

4,

p5

0.0

5)

Rei

det

al.

(20

05

)U

SA

Hu

rric

anes

Cro

ss-s

ecti

on

al5

3F

lori

da

Dep

artm

ent

of

Hea

lth

emp

loyee

sT

rain

ing

on

kn

ow

-le

dge,

skil

lsan

dp

rep

ared

nes

s.

Imp

act

of

trai

nin

go

nk

now

led

ge,

skil

lsan

daw

aren

ess;

per

cep

tio

ns

of

pre

par

edn

ess

and

kn

ow

led

ge/

skil

lap

pli

cati

on

rela

-ti

ve

totr

ain

ing

;in

vo

lvem

ent

inth

e2

00

4h

urr

i-ca

nes

.O

pen

-en

ded

qu

esti

on

so

nre

com

men

da-

tio

ns,

over

all

imp

act

of

trai

n-

ing

,su

gges

tio

ns

for

imp

rov

ing

Lar

ge

maj

ori

tyre

po

rted

the

trai

nin

gin

crea

sed

thei

rk

now

led

ge

of

the

core

pro

-g

ram

con

cep

ts.

Mo

stin

dic

ated

the

trai

nin

gen

han

ced

thei

rab

ilit

yo

rp

re-

par

edth

emto

imp

lem

ent

the

13

spec

ific

skil

lsad

dre

ssed

inth

etr

ain

ing

.O

f9

0%

ind

icat

edth

etr

ain

ing

was

(co

nti

nu

ed)

12 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

del

iver

yan

dad

d-

itio

nal

com

men

ts.

All

asse

ssed

via

stu

dy-s

pec

ific

surv

ey.

use

ful

for

thei

rjo

bs

and

98

%fe

ltb

ette

rp

rep

ared

tore

spo

nd

toa

maj

or

trau

ma

asa

resu

lto

fth

etr

ain

ing

.O

f8

9%

had

inco

rpo

rate

dth

etr

ain

ing

and

its

rela

ted

con

-ce

pts

/sk

ills

inth

eir

per

son

alli

fe;

83

%in

thei

rp

rofe

ssio

nal

life

Sey

leet

al.

(20

13

)In

do

nes

iaE

arth

qu

ake

4-y

ear

lon

git

ud

inal

stu

dy

43

Ele

men

tary

sch

oo

lte

ach

ers

Psy

cho

edu

cati

on

Po

st-t

rau

mat

icd

is-

tres

ssy

mp

tom

s;d

epre

ssio

n;

per

-ce

pti

on

so

fte

ach-

ing

effi

cacy

;p

erce

pti

on

so

fcl

assr

oo

mb

ehav

-io

ur.

PC

L;

CE

S-

D;

Tea

cher

Eff

icac

yS

cale

;m

od

ifie

dver

sio

no

fch

ild

beh

avio

ur

chec

kli

st.

Co

mp

ared

top

aren

tsfr

om

the

com

mu

-n

ity

:th

eyd

idn

ot

dif

fer

inte

rms

of

deg

ree

of

eart

h-

quak

eex

po

sure

or

dep

ress

ion

,b

ut

teac

her

sre

po

rted

low

erP

TS

Dsc

ore

s.P

re-i

nte

rven

tio

n,

CE

S-D

sco

res

wer

ea

sig

nif

ican

tp

red

icto

ro

fre

po

rted

neg

ativ

est

ud

ent

beh

avio

ur

(bet

a¼

0.3

54

,p¼

0.0

37

).P

CL

was

no

ta

sig

nif

i-ca

nt

pre

dic

tor

of

teac

her

s’p

erce

p-

tio

ns

of

neg

ativ

ecl

assr

oo

mb

ehav

-io

ur.

Pre

-in

terv

enti

on

,P

CL

sco

res

pre

-d

icte

din

crea

sed

gen

eral

teac

hin

gef

fica

cy(b

eta

¼0

.34

3,

p¼

0.0

25

).P

ost

-in

terv

enti

on

,C

ES

-Dan

dP

CL

sco

res

wer

en

ot

(co

nti

nu

ed)

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 13

Tab

le1

.C

on

tin

ued

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

pre

dic

tive

of

per

-so

nal

teac

hin

gef

fica

cy.

Pai

red

-sam

ple

tte

sts

show

eda

sig

nif

i-ca

nt

dro

pin

PC

Lsc

ore

s(t

(42

)¼

1.9

5,

p¼

0.0

29

)an

dC

ES

-Dsc

ore

s(t

(42

)¼

2.3

07

,p¼

0.0

13

)fr

om

bef

ore

toaf

ter

the

inte

rven

tio

nT

ehra

ni

etal

.(2

00

1)

UK

Pad

din

gto

ntr

ain

cras

hO

ne

wee

kaf

ter

cras

h;

foll

ow

edu

p4

mo

nth

saf

ter

cras

h.

12

Sai

nsb

ury

’sem

plo

yee

s(t

he

sup

erm

arket

was

nea

rto

the

cras

hsi

tean

dth

eyw

ere

som

eo

fth

efi

rst

on

the

scen

e,p

rov

idin

gla

d-

der

sto

hel

pp

asse

n-

ger

scl

imb

up

fro

mth

etr

ack

and

giv

ing

imm

edia

tefi

rst

aid

).

Deb

rief

ing

PT

SD

dep

ress

ion

.E

xte

nd

edIm

pac

to

fE

ven

tsS

cale

;G

old

ber

gA

nx

iety

and

Dep

ress

ion

Sca

le.

Rec

ord

so

fsi

ckn

ess

abse

nce

.

On

ew

eek

afte

rth

eev

ent,

ther

ew

asa

hig

hle

vel

of

po

st-

trau

ma.

Fo

ur

mo

nth

sfo

llow

ing

the

deb

rief

ing

,th

ere

was

asi

g-

nif

ican

tre

du

ctio

nin

the

sym

pto

ms

mea

sure

dby

the

IES

-Ean

dG

old

ber

gin

stru

-m

ents

.B

efo

resc

ore

s,af

ter

sco

res,

mea

nd

if-

fere

nce

bet

wee

nsc

ore

s,T

-sco

re,

sig

nif

ican

ce,

resp

ecti

vel

y:

An

xie

ty:

5.9

,3

.1,

2.8

,3

.75

,0

.00

3D

epre

ssio

n:

4.8

,2

.2,

2.6

,4

.08

,0

.00

1A

vo

idan

ce:

18

.3,

9.7

,8

.6,

7.7

,0

.00

1A

rou

sal:

15

.8,

6.3

,9

.5,

4.3

,0

.00

1R

e-ex

per

ien

ce:

24

.2,

9.3

,1

5.1

,7

.5,

0.0

01

Th

ep

erfo

rman

cein

dic

ato

rssh

ow

edth

atal

lth

e

(co

nti

nu

ed)

14 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

emp

loyee

sd

emo

nst

rate

dan

imp

rovem

ent

inp

erfo

rman

ceb

etw

een

Novem

ber

19

99

and

Ap

ril

20

00

wit

hth

eg

reat

est

imp

rovem

ent

bei

ng

inth

eir

chee

rfu

lnes

s,ab

ilit

yto

con

cen

-tr

ate

and

per

-fo

rman

ceo

nth

ejo

b.

Th

em

an-

ager

sas

sess

edth

ep

erfo

rman

ceo

fm

ost

of

the

emp

loyee

sas

exce

llen

to

rver

yg

oo

din

Ap

ril

20

00

.T

her

ew

asa

red

uc-

tio

nin

the

level

of

sick

nes

sab

sen

ceb

etw

een

the

firs

tq

uar

ter

of

19

99

and

20

00

.In

the

wh

ole

gro

up

,1

4fe

wer

day

sab

sen

cew

asre

c-o

rded

in2

00

0th

anh

adb

een

reco

rded

in1

99

9.

Wae

lde

etal

.(2

00

8)

US

AH

urr

ican

eK

atri

na

8-w

eek

pro

gra

mm

eb

egin

nin

g1

0w

eek

sp

ost

-dis

aste

r;m

id-

trea

tmen

tan

dp

ost

-tr

eatm

ent

qu

esti

on

-n

aire

sco

mp

lete

dat

3an

d8

wee

ks

po

st-

wo

rksh

op

.

Tw

enty

atb

asel

ine;

15

atfo

llow

-up

.M

enta

lh

ealt

hw

ork

ers

Fo

ur-

ho

ur

med

ita-

tio

nw

ork

sho

pfo

llow

edby

8-

wee

kh

om

est

ud

yp

rog

ram

me.

PT

SD

sym

pto

ms;

dep

ress

ion

;an

x-

iety

;fo

llow

-up

qu

esti

on

nai

reas

ked

them

tora

teim

pro

vem

ent

inw

ellb

ein

g,

stre

ssco

pin

g,

fru

stra

-ti

on

tole

ran

ce,

acti

vit

yle

vel

,d

epre

ssio

n,

anger

,p

hysi

cal

pai

n,

slee

pp

rob

lem

san

dfa

tig

ue

asa

Sco

res

for

tota

lP

TS

D,

re-e

xp

eri-

enci

ng

,h

yp

erar

-o

usa

lan

dst

ate

anx

iety

sig

nif

i-ca

ntl

yd

ecre

ased

over

tim

e.S

lop

esfo

rav

oid

ance

and

dep

ress

ion

wer

ein

the

exp

ecte

dd

irec

tio

ns

bu

tn

ot

stat

isti

call

yd

if-

fere

nt

fro

mze

ro.

All

trea

tmen

t

(co

nti

nu

ed)

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 15

Tab

le1

.C

on

tin

ued

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

resu

lto

fin

ter-

ven

tio

n.

PC

L-S

;C

ES

-D;

Sta

te-

Tra

itA

nx

iety

Inven

tory

;st

ud

y-

spec

ific

qu

esti

on

s.

effe

ctsi

zes

wer

ein

the

med

ium

ran

ge,

exce

pt

dep

ress

ion

,w

hic

hw

asin

the

smal

lra

nge.

Of

93

%re

po

rted

feel

ing

som

ewh

ato

rm

uch

bet

ter

asa

resu

lto

fth

ein

terv

enti

on

.M

ore

than

60

%re

po

rted

imp

rovem

ents

inst

ress

cop

ing

,fr

ust

rati

on

tole

r-an

ce,

acti

vit

yle

vel

and

dep

res-

sio

n.

Of

47

%re

po

rted

imp

roved

abil

ity

tod

eal

wit

han

ger

.T

he

slo

pes

of

tota

lP

TS

Dan

dan

x-

iety

wer

eco

rre-

late

dw

ith

the

tota

ln

um

ber

of

min

ute

so

fm

edi-

tati

on

pra

ctic

eac

ross

the

8w

eek

s(r¼�

0.4

0,

p5

0.0

5an

dr¼�

0.3

8,

p5

0.0

5,

resp

ect-

ivel

y)

ind

icat

ing

that

mo

rem

edit

a-ti

on

pra

ctic

ew

asas

soci

ated

wit

hg

reat

erim

pro

vem

ents

.W

uet

al.

(20

12

)C

hin

a2

00

8S

ich

uan

eart

hqu

ake

On

em

on

th;

foll

ow

-up

com

ple

ted

by

7m

on

ths

po

st-d

isas

ter.

12

67

Mil

itar

yre

scu

ers

51

2P

IM,

1-m

on

thp

ost

-dis

aste

rA

nx

iety

,d

epre

ssio

nan

dP

TS

D.

Ho

spit

alA

nx

iety

and

Dep

ress

ion

Sca

le(H

AD

S);

Str

uct

ure

d

Bas

elin

ean

aly

sis

sug

ges

ted

no

sig

-n

ific

ant

dif

fer-

ence

sb

etw

een

the

stu

dy

gro

up

s.S

ever

ity

of

PT

SD

,an

xie

ty

(co

nti

nu

ed)

16 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Stu

dy

Co

un

try

Dis

aste

rD

esig

nP

arti

cip

ants

(n)

Ro

leo

fp

arti

cip

ants

Inte

rven

tio

nO

utc

om

esas

sess

edR

esu

lts

Inte

rvie

wfo

rP

TS

D(S

I-P

TS

D).

and

dep

ress

ion

dec

reas

edover

tim

ein

all

thre

eg

rou

ps,

wit

hsi

g-

nif

ican

td

iffe

r-en

ces

bet

wee

nth

eg

rou

ps

insy

mp

-to

ms

of

PT

SD

(p\0

.01

).C

om

par

edw

ith

the

deb

rief

ing

and

con

tro

lg

rou

p,

sig

nif

ican

tly

low

ersc

ore

so

fP

TS

Dan

dp

osi

-ti

ve

effi

cacy

inim

pro

vin

gsy

mp

-to

ms

of

re-

exp

erie

nci

ng

,av

oid

ance

and

hyp

erar

ou

sal

wer

efo

un

din

the

‘‘5

12

PIM

’’g

rou

p.

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 17

Florida Department of Health employees preparing for

hurricanes (Reid et al., 2005).

Three studies involved interactive training workshops

providing education and allowing for discussion of key

issues (Aiello et al., 2011; Gershon et al., 2004; Reid et al.,

2005); one study involved interactive computer-based training

(Maunder et al., 2010) and one entailed a disaster simulation

exercise (Eid et al., 2004). Training courses lasted between

one hour (Aiello et al., 2011) and six days eighteen hours (Eid

et al., 2004). All five interventions were designed to educate

employees about what they may experience during a disaster

and how best to cope with this. The education that was

provided covered various aspects of disasters including

disaster planning (Eid et al., 2004; Gershon et al., 2004;

Maunder et al., 2010; Reid et al., 2005); common concerns of

employees during a disaster (Aiello et al., 2011); the potential

psychological impact of disasters (Aiello et al., 2011;

Gershon et al., 2004; Maunder et al., 2010); coping strategies

(Aiello et al., 2011; Maunder et al., 2010) and resources for

further support (Aiello et al., 2011; Maunder et al., 2010).

Further details of what each intervention entailed can be seen

in Table 2.

Outcomes

All five studies considered the effect of the intervention on

participants’ confidence. Aiello et al. (2011) found that pre-

intervention, only 34.9% of hospital workers felt confident in

being able to deal with a potential influenza pandemic.

Following the session, 69.7% agreed with the statement ‘‘The

session prepared me for a pandemic’’, while 76.1% felt more

confident in their ability to cope as a result of the session,

agreeing with the statement ‘‘Following today’s session, I

believe that I will be better able to cope in the event of a

pandemic’’. Reid et al. (2005) found that 98% of Department

of Health workers felt more confident and better prepared to

respond to a major trauma as a result of the skills training;

most of those who were involved in hurricane response in

2004, having received training in 2003, indicated that the

training had given them confidence and they had used the

acquired skills in their response efforts. Gershon et al. (2004)

found increased confidence in recognising symptoms of

terrorism exposure, treating victims of chemical exposure and

protecting oneself whilst caring for victims. Maunder et al.

(2010) found that participants showed significant improve-

ments from the start to the end of the training course in

‘‘pandemic self-efficacy’’ (i.e. confidence in one’s ability to

carry out their job effectively during a pandemic) and

significantly improved confidence in training and support

received. Eid et al. (2004) found that their submarine disaster

simulation exercise reinforced confidence in Navy sailors: on

a scale of 1–5 (with 5 being high) there was a median score of

4/5 when participants were asked to rate whether the exercise

had reinforced their confidence in the submarine service.

Maunder et al. (2010) found that participants showed

significant improvements from the start to the end of the

course in some interpersonal problems, including social

inhibition and non-assertive, overly accommodating, self-

sacrificing and intrusive/needy behaviours. The same

study found that the intervention led to better coping

strategies: those participants who did not report coping with

stress by using problem-solving or seeking support pre-course

tended to be more likely to use these positive coping

strategies post-course, and less likely to use negative coping

strategies, such as escape-avoidance.

Participants’ perceived usefulness of intervention

Aiello et al. (2011) reported that 87.5% of hospital workers

found the training session ‘‘useful’’ while 91.3% found it

‘‘helpful and informative’’. Reid et al. (2005) found that 90%

of Department of Health employees found the hurricane

training useful, and as a result 89% had incorporated the

training and its related concepts or skills into their personal

lives while 83% had incorporated these into their professional

lives. Eid et al. (2004) reported a median score of 5 (out of 5)

for whether similar exercises should be held on a regular

basis, showing that sailors did indeed feel the exercises were

useful and should be held regularly.

No studies reported any detrimental effects of pre-disaster

training on confidence and the one study (Maunder et al.,

2010) which considered the impact of the training on

interpersonal problems found only positive effects. No studies

measured whether there was any detrimental effect of pre-

disaster training on other mental health outcomes.

Post-disaster

Summary of studies

Ten studies explored the effects of interventions designed to

improve psychological wellbeing after disasters. Two studies

from the USA involved a worksite crisis intervention for

employees of various organisations following the 11

September terrorist attacks in New York (Boscarino et al.,

2005, 2006). Kenardy et al. (1996) examined the effects of

group stress debriefing on emergency services personnel and

disaster workers following an earthquake in Australia; North

et al. (2002) looked at the effects of debriefing on fire-fighters

after a bombing in the USA; Tehrani et al. (2001) examined the

effects of debriefing on supermarket workers in the UK after a

train crash; and Miller-Burke et al. (1999) examined the effects

of Critical Incident Stress Debriefing (CISD) for bank

employees following bank robberies in the USA. It is important

to note that though these articles all refer to ‘‘debriefing’’, the

procedures involved and focus of the intervention may have

differed between studies (see Table 2 for specific details).

Other interventions included cognitive behavioural therapy for

disaster workers who met full or subthreshold PTSD criteria

after the 11 September attacks in the USA (Difede et al., 2007);

psycho-education for elementary school teachers in Indonesia

who experienced an earthquake (Seyle et al., 2013); meditation

for mental health workers following a hurricane in the USA

(Waelde et al., 2008); and a psychological intervention called

the ‘‘512 PIM’’ (512 Psychological Intervention Model)

including discussions, training on cohesion and stress man-

agement, delivered to military rescuers following an earth-

quake in China (Wu et al., 2012).

Outcomes

Boscarino et al. (2005) reported that between 1 and 3 worksite

crisis sessions appeared to protect workers from binge

18 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Table 2. Overview of interventions.

Study Details of intervention

Qualityappraisalscore (%)

Aiello et al. (2011) One-hour training session on predicted stressors associated with influenza; commonconcerns; normal responses to extraordinary stress; organisational approaches tobuilding resilience and reducing stress; coping and resources for further support.Sessions focused on principles such as ‘‘hoping for the best but preparing for the worst’’and making implicit coping strategies explicit. Interaction, in terms of discussion,identification of problems and suggested solutions, was encouraged.

66.7

Boscarino et al. (2005) Brief employer-provided worksite crisis interventions. These could be any brief sessionsrelated to coping with the 9/11 disaster received shortly after the event, arranged by theemployer and led by mental health professionals or counsellors. These included criticalincident stress management, psychological debriefing or any other short-term inter-ventions designed to provide emergency mental health support following trauma.Interventions were different for different workplaces, but 60–70% reported that theintervention involved coping/relaxation techniques and strategies for positive thinking.

93.8

Boscarino et al. (2006) See above 93.8Difede et al. (2007) Twelve weekly sessions of cognitive behavioural therapy modified for disaster workers.

Including psychoeducation; treatment rationale and contracting; breathing exercises;cognitive reprocessing; relapse prevention and homework involving imaginal exposureusing audio-taped imaginal exposure from treatment sessions, graduated in vivoexposure and cognitive reprocessing.

75

Eid et al. (2004) Simulated submarine disaster exercise lasting 6 d 18 h. Participants were seated in a sectionof the submarine in seawater of 35 m depth with lighting, ventilation and heating shutoff. Activities and schedules were organised as closely as possible to SUBSUNKprocedures – an operation carried out to search for a submarine that is known to havesunk.

80

Gershon et al. (2004) Half-day terrorism educational workplace programme, including: role of emergencymanagement in a disaster; information on reporting guidelines, diagnosis, treatment,infection control and bioterrorism effects; methods to diagnose and treat chemical andradiological agents; an overview of disaster mental health and treatment options; keyelements in workplace disaster planning.

73.3

Kenardy et al. (1996) Group stress debriefing 87.5Maunder et al. (2010) ‘‘Pandemic Stress Vaccine’’, a computer-assisted course comprised of audio and video

mini-lectures, onscreen notes, printed fact sheets, quizzes and games and audio modulesfor relaxation skills. Topics addressed included what to expect during a pandemic; whatis resilience; normal stress responses; psychological first aid; working outside of one’scomfort zone; moral and ethical dilemmas, coping approaches; active listening;expressing oneself constructively; balancing work and family; talking to children aboutdisasters/emergencies; personal and home preparation; managing drugs and alcohol;danger signals; and resources for getting help. Participants were randomly assigned toeither the short course (7 sessions, median 111 min total); medium course (12 sessions,158 min); or long course (17 sessions, 223 min).

87.5

Miller-Burke et al. (1999) Critical Incident Stress Debriefing emphasising containment of feelings and expression offeelings in a safe and supportive environment, aimed at moving participants from‘‘shock’’ to ‘‘acceptance’’. This was delivered by an external employee-assistanceservice.

66.7

North et al. (2002) Workplace debriefings/defusings. Debriefings were carried out in groups of 15–20 workers,while ‘‘mental health defusing’’ was carried out individually, after each shift in whichparticipants worked at the bomb site.

80

Reid et al. (2005) Bioterrorism Trauma Intervention Specialist Training (BTIST): Providing skills inassessment, triage, trauma stabilisation, individual and group defusing, debriefing, stressmanagement, compassion fatigue, grief intervention, cultural competence, criticalincident stress management and team development. The 5-d training course includedinteractive teaching methods, such as case studies, role-playing and scenarios. Moduleswere also available on CDs and DVDs.

60

Seyle et al. (2013) Psychoeducational material about impact of disasters on thoughts, emotions and behavioursin adults; relaxation and coping exercises such as deep breathing; psychoeducationalinformation about impact of disasters on children and basic information aboutsupporting children’s emotional stability; education on child misbehaviour andsupporting children’s attention in class.

75

Tehrani et al. (2001) Group debriefing one-week post-disaster, beginning with a short presentation on post-trauma stress giving the opportunity to ask questions and learn about the commonreactions/symptoms experienced after a traumatic event. The debriefing then allowedemployees to share their experiences of what they went through, focusing on theirbehaviours and things that were seen, heard, smelled, touched and tasted. This allowedthem to examine what had happened during the day of the crash, to fill gaps in the story,check understanding and share knowledge.

50

Waelde et al. (2008) Four-hour meditation workshop, including instruction and guided practice in meditation;breathing; guided breathing-focused imagery; mantra repetition and letting go of

81.3

(continued )

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 19

drinking or alcohol dependence, while 2–3 sessions were

protective for PTSD, depression, somatisation, anxiety and

global severity. A follow-up study using propensity scores to

match intervention cases to non-intervention controls

(Boscarino et al., 2006) found that outcomes tended to be

better for intervention cases: they were less likely to show

symptoms of depression and alcohol dependence in the past

year and less likely to show symptoms of PTSD, depression

and anxiety over the past month. Similarly, Tehrani et al.

(2001) found a high level of trauma symptoms immediately

post-disaster, but four months after the debriefing intervention

participants reported reduced anxiety, depression, and PTSD

symptoms – however, there was no control group to determine

whether this recovery was due to the intervention. Tehrani

et al. (2001) also found that post-intervention, all employees

demonstrated an improvement in ‘‘cheerfulness’’ and ability

to concentrate. However, one study did not find any

improvement in psychological outcomes following debrief-

ing: in Kenardy et al.’s (1996) study of group stress

debriefing, the participants who were debriefed reported

significantly higher scores on the GHQ-12 than those who

were not debriefed, and showed less improvement in PTSD.

Even when level of exposure was taken into account, there

was no evidence of debriefing improving psychological

wellbeing.

North et al. (2002) measured PTSD symptoms but not as

an effect of the intervention; instead, they explored whether

PTSD symptoms were associated with satisfaction with, or

recommendation of, the intervention. They found no such

association, though participants with other, non-PTSD dis-

orders were less likely to report satisfaction or recommend the

intervention.

In Wu et al.’s (2012) study of the ‘‘512 Psychological

Intervention Model’’, anxiety, depression and PTSD symp-

toms decreased over time for all three groups (the ‘‘512 PIM’’

group, the debriefed group and the control group), with

significant differences between groups with regard to PTSD

symptoms. The ‘‘512 PIM’’ group had significantly lower

scores of PTSD and positive efficacy in improving symptoms

of re-experiencing, avoidance and hyperarousal.

The study which explored the effects of CBT (Difede et al.,

2007) found a significant decrease in PTSD scores for the

CBT group compared to the treatment-as-usual group, 21% of

which actually increased in PTSD scores. However, there

were no significant differences between groups in terms of

alcohol misuse.

One study looked at the effects of a psycho-educational

intervention (Seyle et al., 2013). This article reported a

significant decrease in depression and PTSD symptoms post-

intervention.

The study exploring meditation as an intervention (Waelde

et al., 2008) found that scores for total PTSD, re-experiencing,

hyperarousal and state anxiety significantly decreased over

time. Total PTSD and anxiety scores were correlated with the

total number of minutes of meditation practiced across the

8 weeks, suggesting more meditation was associated with

greater improvements. The majority of the participants (93%)

in this study reported feeling somewhat or much better as a

result of the intervention. More than 60% reported improve-

ments in stress coping, frustration tolerance, activity level and

depression, while 47% reported improved ability to deal with

anger.

Participants’ perceived usefulness of intervention

In Boscarino et al.’s (2005) study, 21.2% reported that the

worksite crisis intervention helped a lot; 22.8% said it helped

some; 33.7% said it helped a little and 22.4% found it not at

all helpful. North et al. (2002) found that two-third of

participants were ‘‘satisfied’’ with the workplace

Table 2. Continued

Study Details of intervention

Qualityappraisalscore (%)

thoughts, feelings and sensations as they arise. Followed by 8-week home studyprogramme including manual and 4 audio recordings of guided meditations.

Wu et al. (2012) One month post-disaster, one group of participants received the 512 PIM. This included:introduction (explaining goals of the session and developing a friendly and trustfulatmosphere); facts and thoughts (participants describe the facts of the trauma as they seethem, and then recall their first thoughts during the trauma); reaction and symptoms(participants reconstruct the trauma and accompanying emotions in detail, and thendescribe stress symptoms they experienced during and just after the event and currently);stress management (tips and advice about ways of coping with the trauma or the stresssymptoms); training of cohesion (e.g. participants were instructed to play games whichneed team co-operation). Comparison groups included a debriefing group and a no-treatment group.

100

0

2

3

4

Num

ber o

f ar�

cles

0.51

1.52

.53

.54

.55

41-50% 51-60% 61-70%Qualit

71-80% 81y %

-90% 91-100%

Figure 2. Quality appraisal scores.

20 S. K. Brooks et al. J Ment Health, Early Online: 1–25

defusings/debriefings they had taken part in; although one-

third were dissatisfied, 89% still said they would recommend

the intervention. Miller-Burke et al. (1999) found that almost

80% of participants who attended a formally presented CISD

session found it to be helpful. Interventions with co-workers

were also perceived as successful by 77%; interventions with

supervisors were found to be helpful by 64%; and 54% found

interventions with family or friends to be helpful. In Kenardy

et al.’s (1996) study – which showed no positive effect of

debriefing – there was no correlation between the number of

debriefing sessions and the perceived helpfulness of the

intervention; debriefing was reported to be ‘‘very’’ or

‘‘extremely’’ helpful by 34% of participants, ‘‘somewhat’’

helpful by 46% and ‘‘not at all’’ helpful by 20%.

Discussion

This review aimed to explore the literature on interventions

designed to improve the wellbeing of employees exposed to

disasters. We aimed to assess the success of such interven-

tions by looking at their effect on psychological constructs,

such as confidence and ability to cope, and on symptoms of

psychological disorders, such as PTSD and anxiety. We also

aimed to explore how helpful or useful the participants found

the interventions.

We found only a limited amount of relevant literature. Five

studies considered the impact of pre-disaster training courses,

focusing mainly on whether they improved confidence and

whether participants found them useful. Overall, we found

that pre-disaster training was found to be helpful in increasing

confidence in all five studies. Three of the five studies asked

participants to rate the usefulness of the intervention, with all

three reporting positive results. There was very little explor-

ation of the impact of training on psychological outcomes,

though one study reported that it led to improvements in

‘‘inter-personal’’ problems and more effective coping mech-

anisms (i.e. problem-solving or seeking support, rather than

avoidance).

It appears that evidence-based approaches, providing

education and focusing on mediators of distress, may be

relevant and useful in terms of improving employees’

confidence in their own abilities to act appropriately during

a disaster and their ability to cope. Realistic simulated crisis

training, such as the rehearsal training explored by Eid et al.

(2004) appears to also be valuable in improving confidence.

While these are positive findings, confidence on its own is not

an especially useful outcome for pre-disaster training as it is

likely to be a temporary outcome, especially if measured

directly after training. It would perhaps be more useful for

studies to focus on how participants rate the usefulness of the

training, and whether they plan to or succeed in implementing

their learning into their work – as in the Reid et al. (2005)

study which reported that over 80% of participants had

incorporated training concepts into their professional lives.

This is more likely to be a long-lasting outcome than

increased confidence.

Importantly, no negative effects associated with pre-

disaster skills training were reported in any of the five pre-

disaster training studies. No studies reported a detrimental

effect on confidence and the study which considered

interpersonal problems found only positive effects of training.

Participants typically rated training as useful and something

they would recommend. This is a positive finding especially

since it could be possible for any disaster training to increase

feelings of worry or anxiety in employees, as it requires them

to consider worst-case scenarios and perhaps causes them to

think about stressful situations they may not have considered

before. However, it is possible that the lack of findings

regarding detrimental effects is due to the nature of the post-

training questions asked; future studies should include a

measure of worry or anxiety to further explore whether

training could have a detrimental effect.

Ten studies explored post-disaster interventions, focusing

mainly on their impact on psychological symptoms. Three

studies suggested that workplace interventions such as

debriefings can have long-term positive effects, with ideally

more than one session being most helpful (Boscarino et al.,

2005). However, one study found that debriefing was not

helpful and could in fact lead to poorer outcomes. Other

interventions such as the ‘‘512 Psychological Intervention

Model’’ (Wu et al., 2012), CBT, psychoeducation and

meditation were all found to be useful in improving psycho-

logical symptoms. It is also important to note that the studies

exploring these interventions were among the highest-scoring

studies in the quality appraisal, in particular the 512

Psychological Intervention Model study which described the

appropriate statistical tests fully and considered confounding

variables and was the only study to receive 100% on our

quality assessment. However, these interventions were only

explored in one study each, therefore, we cannot be sure that

the results are generalisable to other employee populations.

Four studies explored how useful participants rated the

interventions, and the results were not especially positive

overall. The most positive result came from Miller-Burke

et al.’s (1999) study of bank employees who participated in

CISD – 80% reported finding it useful. Three studies on

workplace debriefings suggested that participants did not find

them particularly helpful. The other studies did not include

any measure of helpfulness of the intervention, and, therefore,

we do not know whether the 512 Psychological Intervention

Model, CBT, psychoeducation or meditation were rated as

helpful by the employees who participated in them.

Overall, we found little evidence of what kind of post-

disaster interventions are most useful. The fact that we found

such a small number of relevant studies, many of which

looked at different interventions, means we cannot identify

which intervention is most successful. It does appear that

debriefing led to an improvement in psychological symptoms

but, as one study found that debriefing did not improve

wellbeing at all and in fact led to poorer outcomes, it is likely

that the effectiveness of debriefing is dependent on factors,

such as how and when it is carried out, what it involves, and

how many sessions are provided. It is also important to note

that there was little consistency within the articles regarding

what constituted debriefing, and so it would be inappropriate

to generalise from this review. The lack of longitudinal

studies also means that there was no evidence on the potential

long-term effects of psychological debriefing.

Literature reviews focusing on the mental health impact of

disasters on communities in general have shown similar

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 21

findings. For example a review of the mental health response

to community disasters (North & Pfefferbaum, 2013)

identified PFA, debriefing and ‘‘crisis counselling’’ as the

three most commonly applied early psychosocial interven-

tions. Their discussion of PFA was fairly positive, suggesting

it should be embedded into emergency response systems,

although they point out that it has not been empirically tested.

They concluded that debriefing was ineffective for PTSD

prevention or treatment and that it could actually be harmful

in the long term, suggesting that instead, individuals at risk

for psychopathology should be identified and referred for

psychiatric services. Finally, they found that ‘‘crisis counsel-

ling’’ was poorly defined in the literature and that overall it

appeared to be insufficient in meeting the needs of individuals

who require formal psychiatric treatment. Another review on

disaster impact at the community level (Goldmann & Galea,

2014) supported the ideas that debriefing is not recommended

and that PFA is the preferred post-disaster intervention,

although empirical studies evaluating its effectiveness are still

needed. They also found empirical support for CBT, in

particular exposure therapy where the individual is asked to

recollect the traumatic experience, though they point out that

there are few randomised controlled trials for CBT in disaster

survivors and few studies looking at the long-term effects.

This review also proposes that pre-disaster, it is helpful to

develop and test response methods within communities that

are adaptable to different disaster situations, and to build on

knowledge gained from previous disasters. They suggest that

addressing modifiable stressors and helping victims return to

pre-disaster routines may be effective in reducing post-

disaster distress; they also cite an association between media

exposure of the disaster and PTSD, suggesting that commu-

nities should look into ways of communicating information in

such a way as to reduce fear and promote calm.

Limitations

Two-third of the included articles (n¼ 10) were from North

America. It may have been useful to consider widening the

search and translating foreign language articles to provide

further evidence about the effectiveness of interventions. The

decision to restrict the search to articles published in peer-

reviewed journals may have limited the results; future

research may consider searching grey literature for relevant

studies. The quality appraisal tool we used was the same one

we used to assess articles in our previous reviews (Brooks

et al., 2015, 2016, 2017); it was a deliberate choice to use the

same tool throughout our reviews to ensure consistency.

However, this particular tool does not specifically assess the

quality of intervention studies, and it may have been useful to

include questions specifically relating to the potential for bias

which are relevant only for intervention studies.

There were also a number of key limitations of the studies

included in this review. The pre-disaster intervention studies

tended to limit their analysis to exploring confidence and

perceived self-efficacy, and did not measure other psycho-

logical outcomes. Most of the pre-disaster studies were cross-

sectional rather than prospective or randomised controlled

trials. Often, confidence pre-training was measured retro-

spectively, so while participants may have felt their

confidence improved, it is difficult to ascertain whether

this is true and by how much it improved. The articles

reviewed here also failed to explore the impact this

reported increase in confidence may have on coping and

other psychological outcomes. Follow-up periods tended to

be short, with participants completing measures immedi-

ately following their training sessions, and so long-term

effects on confidence and competency are not known.

Furthermore, while participants clearly reported feeling

more confident after training, there is no evidence to

suggest whether this would translate to better coping during

an actual disaster. A useful way of assessing this might

involve having simulated crisis events (i.e. emergency

preparedness exercises or drills), where employees have to

act and respond exactly as they would in a disaster

situation. This would give a better idea of whether the

training has actually been successful in terms of improving

skills, though it would still not be clear – without

experiencing a real disaster – whether the training really

does improve employees’ ability to cope psychologically.

Though several of the post-disaster intervention studies

showed positive effects of interventions, such as workplace

debriefings, some studies (e.g. Boscarino et al., 2005, 2006)

looked at many different workplaces, all of whom had

received their own workplace interventions. These presum-

ably differed in terms of how they were offered, the timing of

the intervention, who led the sessions, the length of the

sessions and what exactly was involved. As there was no

attempt to differentiate the results of different workplaces, it

is difficult to ascertain which aspects of debriefing are

important in terms of improving outcomes. Many of the post-

disaster studies failed to ask participants to rate the usefulness

of, or their satisfaction with, the interventions. Many had no

control groups or ‘‘treatment-as-usual’’ groups to compare

results with, and many had no follow-up or pre-disaster

comparisons. Therefore, even when results were positive and

indicated that employees coped well after disasters, it is

impossible to know whether this is due to the interventions:

perhaps the employees involved were particularly resilient, or

perhaps symptoms would have improved over time even

without an intervention.

In addition, many studies had small sample sizes and high

dropout rates, making it difficult to generalise from the

results. Often no standardised measures were used, and the

questionnaires given to participants were brief. Several

studies presented few statistical details and did not measure

psychological outcomes. Very few studies considered other

factors which may affect the effectiveness of interventions,

such as prior psychiatric history or previous exposure to

trauma.

Implications

The paucity of good quality work in this area is striking. More

and better quality, studies are required before any firm

recommendations can be made on how best to prepare

employees for a possible disaster or to intervene following a

disaster. Future studies should include comparison groups.

Reliance on non-randomised study techniques, especially

ones without a suitable control group and accounting for

22 S. K. Brooks et al. J Ment Health, Early Online: 1–25

known confounders, can lead to misguided practice. A review

on psychological debriefing (Greenberg, 2001) highlights

this, with six of eight randomised trials of debriefing showing

it to be at best unhelpful and at worst harmful, while the

uncontrolled trials showed mixed results, sometimes finding

debriefing to be unhelpful but sometimes finding it to be

beneficial.

The small number of studies examining pre-disaster

training courses appeared to show that these courses were

successful in increasing confidence and feelings of self-

efficacy; however, it is unclear whether this would translate to

better coping if faced with a real disaster, or whether

increased confidence would be protective of mental health

post-incident. It is also unclear what impact training may have

on psychological wellbeing, as no studies measured this,

focusing instead on confidence and satisfaction with the

training course. Taking account of the limitations of the

included studies is particularly important given that whilst

some of the studies we found reported that psychological

debriefing was helpful, there is good evidence from

randomised controlled studies of debriefing techniques in

other populations that not only is it ineffective in improving

mental health status in the longer term but also has the

potential to cause harm (NICE, 2005; Rose et al., 2009).

However, some suggest that the evidence that debriefing is

unhelpful or even harmful is unreliable and that, if protocols

are followed correctly, debriefing can have positive effects on

wellbeing (Hawker & Hawker, 2015). This is an area which

merits further research within disaster-exposed organisations.

From the results of this review, it appears possible that

CBT, psychoeducation and meditation may be useful in

improving the psychological wellbeing of employees after a

disaster. However, these findings come from a very limited

number of studies, without randomised controlled trials, and,

therefore, any suggestion that these could be helpful is

extremely tentative. Future research is needed to support these

findings.

Based on the findings of this review as well as the

findings of reviews on interviews for disaster-exposed

communities, our suggestions for organisations would be as

follows:

� Pre-disaster training appears to be helpful for employees;

prepare them for both the practical issues involved and

the potential psychological distress they may experience.

Develop and test response methods adaptable to different

emergency situations; simulated crisis training may be

useful in helping them to understand how to respond

should a disaster occur. Use any previous experience of

emergency situations to inform these response methods

and crisis training. Educate employees on the symptoms

of distress they should look out for post-disaster and

where they should go to seek help should they need to.

Workshops focusing on appropriate coping skills may be

useful – for example encouraging problem-solving rather

than avoidance, and encouraging them to seek support

from others. PFA training for employees may help them

feel able to listen to and support their colleagues. These

steps are likely to enhance their confidence both in their

abilities to act appropriately during a disaster and their

ability to cope with it psychologically.

� Communicate calmly and effectively about the emer-

gency situation with employees in a way as to not

produce fear.

� Do not ‘‘debrief’’ following a disaster as this has the

potential to be harmful. Instead, encourage a supportive

atmosphere within the workplace where employees feel

able to discuss their feelings with their colleagues.

Managers who are educated in the possible risks of

disaster exposure and the symptoms to look out for may

be able to identify employees who are particularly

vulnerable and may need more support than the organ-

isation can provide and these employees should be

signposted on to the appropriate psychiatric services.

Declaration of interest

NG runs a psychological health consultancy which provides

disaster support training and support.

The research was funded by the National Institute for

Health Research Health Protection Research Unit (NIHR

HPRU) in Emergency Preparedness and Response at King’s

College London in partnership with Public Health England

(PHE). The views expressed are those of the author(s) and not

necessarily those of the NHS, the NIHR, the Department of

Health or Public Health England. The funding body did not

play a role in the design, collection, analysis or interpretation

of data; the writing of the manuscript; or the decision to

submit the manuscript for publication.

ORCID

Samantha K. Brooks http://orcid.org/0000-0003-3884-

3583

Richard Amlot https://orcid.org/0000-0003-3481-6588

Neil Greenberg http://orcid.org/0000-0003-4550-2971

G. James Rubin http://orcid.org/0000-0002-4440-0570

References

Aiello A, Young-Eun Khayeri M, Shreyshree R, et al. (2011). Resiliencetraining for hospital workers in anticipation of an influenza pandemic.J Contin Educ Health Prof, 31, 15–20.

American Psychiatric Association. (1980). Diagnostic and statisticalmanual of mental disorders. 3rd ed. Washington, DC: Author.

Boscarino JA, Adams RE, Figley CR. (2005). A prospective cohort studyof the effectiveness of employer-sponsored crisis interventions after amajor disaster. Int J Emerg Ment Health, 7, 9–22.

Boscarino JA, Adams RE, Foa EB, Landrigan PJ. (2006). A propensityscore analysis of brief worksite crisis interventions after the WorldTrade Center disaster: Implications for intervention and research. MedCare, 44, 454–62.

Brooks SK, Dunn R, Sage CAM, et al. (2015). Risk and resilience factorsaffecting the psychological wellbeing of individuals deployed inhumanitarian relief roles after a disaster. J Ment Health, 24, 385–413.

Brooks SK, Dunn R, Amlot R, et al. (2016). Social and occupationalfactors associated with psychological distress and disorder amongdisaster responders: A systematic review. BMC Psychol, 4, 18.

Brooks SK, Dunn R, Amlot R, et al. (2017). Social and occupationalfactors associated with psychological wellbeing among occupationalgroups affected by disaster: A systematic review. J Ment Health, 26,373–384.

Centre for Research on the Epidemiology of Disasters (2009). Disasterlist. Available from: http://www.emdat.be/disaster-list.

Chang CM, Connor KM, Lai TJ, et al. (2005). Predictors ofposttraumatic outcomes following the 1999 Taiwan earthquake. JNerv Ment Dis, 193, 40–6.

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 23

Difede J, Malta LS, Best S, et al. (2007). A randomized controlledclinical treatment trial for World Trade Center attack-related PTSD indisaster workers. J Nerv Ment Dis, 195, 861–5.

Drummond M, Jefferson T. (1996). Guidelines for authors and peerreviewers of economic submissions to the BMJ. TheBMJ Economic Evaluation Working Party. British Med J, 313,275–83.

Dunn R, Brooks SK, Rubin GJ, Greenberg N. (2015). Psychologicalimpact of traumatic events: Guidance for trauma-exposed organisa-tions. Occup Health Work, 12, 17–21.

Effective Public Health Practice Project (EPHPP) (2009). Qualityassessment tool for quantitative studies. Available from: http://www.ephpp.ca/tools.html.

Eid J, Johnsen BH, Saus E-R, Risberg J. (2004). Stress and coping in aweek-long disabled submarine exercise. Aviat Space Environ Med, 75,616–21.

Gershon RRM, Gemson DH, Qureshi K, McCollum MC. (2004).Terrorism preparedness training for occupational health professionals.J Occup Environ Med, 46, 1204–9.

Goldmann E, Galea S. (2014). Mental health consequences of disasters.Annu Rev Public Health, 35, 169–83.

Greenberg N. (2001). A critical review of psychological debriefingand a proposal for the future: The management of psycho-logical health after traumatic experiences. J R Nav Med Serv,87, 158–61.

Greenberg N, Langston V, Jones N. (2008). Trauma riskmanagement (TRiM) in the UK Armed Forces. J R Army MedCorps, 154, 124–7.

Guha-Sapir D, Hoyois P, Below R. (2013). Annual Disaster StatisticalReview 2013. Available from: http://cred.be/sites/default/files/ADSR_2013.pdf.

Hawker D, Hawker D. (2015). What can be learned from the debriefingcontroversy? In: Yule W, Hawker DD, Alexander D, Holowenko H,Tehrani N, Durkin J, Hacker Hughes J, eds. Early Interventions forTrauma. Proceedings from symposia; 2014 Nov 25 and 2015 Jan 8.Crisis, Disaster and Trauma Section; Leicester, UK: BritishPsychological Society, 6–17 p.

Kenardy JA, Webster RA, Lewin TJ, et al. (1996). Stress debriefing andpatterns of recovery following a natural disaster. J Trauma Stress, 9,37–49.

Maunder RG, Lancee WJ, Mae R, et al. (2010). Computer-assistedresilience training to prepare healthcare workers for pandemicinfluenza: A randomized trial of the optimal dose of training. BMCHealth Serv Res, 10, 72.

Miller-Burke J, Attridge M, Fass PM. (1999). Impact of traumatic eventsand organizational response – A study of bank robberies. J OccupEnviron Med, 41, 73–83.

Mitchell JT. (1983). When disaster strikes . . . the critical incidentdebriefing process. JEMS, 8, 36–9.

National Institute for Health and Care Excellence (NICE) (2005). Post-traumatic stress disorder (PTSD): The management of PTSD in adultsand children in primary and secondary care. Available from: http://www.nice.org.uk/guidance/cg26/chapter/guidance#the-treatment-of-ptsd.

National Institute for Health (2014). Quality assessment tool forobservational cohort and cross-sectional studies. Available from:http://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/cardiovas-cular-risk-reduction/tools/cohort.

North CS, Pfefferbaum B. (2013). Mental health response to communitydisasters: A systematic review. JAMA, 310, 507–18.

North CS, Tivis L, McMillen C, et al. (2002). Coping, functioning andadjustment of rescue workers after the Oklahoma City bombing.J Trauma Stress, 15, 171–5.

Pekevski J. (2012). First responders and psychological first aid. J EmergManag, 11, 39–48.

Reid WM, Ruzycki S, Haney ML, et al. (2005). Disaster mental healthtraining in Florida and the response to the 2004 hurricanes. J PublicHealth Manag Pract, 11, S57–62.

Rose S, Bisson J, Churchill R, Wessely S. (2009). Psychologicaldebriefing for preventing post traumatic stress disorder (PTSD). In:The Cochrane Library. Oxford, UK: Update Software.

Seyle DC, Widyatmoko CS, Cohen Silver R. (2013). Coping with naturaldisasters in Yogyakarta, Indonesia: A study of elementary schoolteachers. Sch Psychol Int, 34, 387–404.

Strohmeier H, Scholte WF. (2015). Trauma-related mental healthproblems among national humanitarian staff: A systematic review ofthe literature. Eur J Psychotraumatol, 6, 28541.

Tehrani N, Walpole O, Berriman J, Reilly J. (2001). A specialcourage: Dealing with the Paddington rail crash. Occup Med(Lond), 51, 93–9.

Waelde LC, Uddo M, Marquett R, et al. (2008). A pilot study ofmeditation for mental health workers following Hurricane Katrina. JTrauma Stress, 21, 497–500.

Wessely S, Bisson J, Rose S, et al. (2000). A systematic review of briefpsychological interventions (‘debriefing’) for the treatment of imme-diate trauma related symptoms and the prevention of post traumaticstress disorder. In: Oakley-Browne M, Churchill R, Gill D, eds.Depression, Anxiety and Neurosis Module of the Cochrane Databaseof Systematic Reviews. Oxford, UK: Update Software.

West C, Bernard B, Mueller C, et al. (2008). Mental health outcomes inpolice personnel after Hurricane Katrina. J Occup Environ Med, 50,689–95.

Wu S, Zhu X, Zhang Y, et al. (2012). A new psychological intervention:‘‘512 Psychological Intervention Model’’ used for military rescuers inWenchuan earthquake in China. Soc Psychiatry Psychiatr Epidemiol,47, 1111–9.

Appendix I – Search strategy

Search: EMBASE 1980–2015; EMBASE 1974–1979; EMBASE Classic1947–1973; Ovid Medline 1946–2015; PsycINFO 1806–2015; Web ofScience.

Search 1 (psychological wellbeing);Well-being; anxiety; panic; posttraumatic stress; PTSD; stress;

‘‘mental health’’; depress*; neurosis; adjustment disorder*; distress;

psychological; resilience; coping; ‘‘mental disorder*’’; ‘‘positive psych-

ology’’; ‘‘satisfactory life’’; mindfulness; flourish; pleasure; flow;

growth¼COMBINE WITH ORSearch 2 (disasters);Anthrax; avalanche; avian influenza; bioterrorism; bird flu; blizzard;

bomb*; chemical spill; Chernobyl; cyclone; drought; disaster*; earth-

quake; Ebola; emergenc*; explosion; fire; flood; Fukushima; H1N1;

H5N1; hurricane; industrial accident; landslide; massacre; mass killing;

MERs; Middle East respiratory syndrome; pandemic; nuclear radiation;

radiological; SARs; severe acute respiratory syndrome; 11 September;

shooting*; storm; swine flu; terroris*; Three Mile Island; tidal wave;

tornado; tsunami; typhoon; volcanic eruption; volcano; World Trade

Center.¼COMBINE WITH ORSearch 3 (occupational search terms);Organisation*; organisation*; occupation*; employee*; employer*;

workforce*; worker*; business; team; emergency response; healthcareprovider*; healthcare worker*; construction work*; fire-fighter*; fireofficer*; paramedic*; doctor*; nurse*; police; first aid responder*;personnel; hospital administrator; military.¼COMBINE WITH ORCombine Search 1 AND Search 2 AND Search 3

Appendix II – Quality appraisal form

All questions are answered with ‘‘yes’’ or ‘‘no’’.

Section 1: Study design(1) Was the research question/objective clearly stated?(2) Were all subjects selected or recruited from the same or similar

populations (including the same time period)?(3) Were the inclusion and exclusion criteria for being in the study pre-

specified and applied uniformly to all participants?(4) Was the study population and size clearly specified and defined?

24 S. K. Brooks et al. J Ment Health, Early Online: 1–25

Section 2: Data collection and methodology(1) Were standardised measures used, or where measures are designed

for the study, attempts to ensure reliability and validity were made?(2) Were the data collected in a way that addressed the research issue?(3) Was the participation rate stated and at least 50%?(4) Was the number of participants described at each stage of the

study?(5) If the study followed participants up, were reasons for loss to

follow-up explained?

Section 3: Analysis and interpretation of results(1) Were details of statistical tests sufficiently rigorous and described?(2) Were details of confidence intervals given?(3) Were potential confounding variables measured and adjusted

statistically for their impact on the relationship between exposure(s)and outcome(s)?

(4) Was the answer to the study question provided?(5) Are the findings related back to previous research?(6) Do conclusions follow from the data reported?(7) Are conclusions accompanied by the appropriate caveats?

DOI: 10.1080/09638237.2018.1437610 Psychological disaster interventions 25