International Journal of

Environmental Research
and Public Health

Review
Adjustment Disorder and Suicidal Behaviours
Presenting in the General Medical Setting:
A Systematic Review
Joanne Fegan and Anne M. Doherty *
Department of Psychiatry, University Hospital Galway, H91 YR71 Galway, Ireland
* Correspondence: [email protected]




Received: 28 June 2019; Accepted: 14 August 2019; Published: 18 August 2019


Abstract: Background: Adjustment disorder (AD) is a condition commonly encountered by clinicians

in emergency departments and liaison psychiatry settings and has been frequently reported among
patients presenting with suicidal behaviours. A number of previous studies have noted the strong
association between suicidal ideation and behaviours, and AD. In this paper, we aimed to explore this
relationship, by establishing the incidence of AD in patients who present with self-harm and suicidal
ideation, and the rates of self-harm among patients with a diagnosis of AD. Methods: We conducted
a review of the literature of well-established databases using specific key words then synthesised
the results into a descriptive narrative as well as representing it in table form. Results: Sample sizes
and study methods varied significantly across the review. A majority of studies were retrospective
chart-based reviews, and only three used structured diagnostic instruments. A high prevalence of
AD (ranging from 9.8 to 100%) was found, with self-poisoning representing the most common form
of suicide attempt in the majority of studies. Interpersonal difficulties were the main precipitant in
studies which examined this. Conclusions: This study suggests there is a strong association between
AD and suicidal behaviours. Given the paucity of research in the area, there is a need to build the
evidence base for effective treatment strategies.

Keywords: adjustment disorder; depressive episode; self-injurious behaviour; liaison psychiatry;

diagnosis; suicide/attempted

1. Introduction
Adjustment disorder (AD) is a condition which is characterised by the development of symptoms,
usually of depression or anxiety, in response to a stressful event [1]. This condition is frequently
diagnosed in patients attending Emergency Departments (EDs) and liaison psychiatry settings,
and in particular, has been commonly reported among patients presenting with suicidal behaviours,
including self-harm.
Suicidal ideation and behaviours may be a feature of a number of psychiatric disorders and are
an important symptom, indeed diagnostic criterion, in depression [2]. AD is defined by the World
Health Organisation in the International Classification of Diseases, 10th Edition (ICD-10) as a state
of “subjective distress and emotional disturbance, usually interfering with social functioning and
performance, and arising in the period of adaptation to a significant life change or to the consequences
of a stressful life event” [2]. A diagnosis of AD requires the identification of a precipitating stressor,
and symptoms must resolve within six months of the termination of the stressor. This diagnosis occurs
where the symptoms are not more appropriately attributed to another mental disorder. The ICD-10
diagnostic criteria do not specify the symptoms of AD beyond “those found in any of the affective
disorders”. However, some indications of typical symptoms are suggested by the subcategories of

Int. J. Environ. Res. Public Health 2019, 16, 2967; doi:10.3390/ijerph16162967 www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2019, 16, 2967 2 of 15

AD in ICD-10, which include “brief depressive reaction”, “prolonged depressive reaction”, “mixed
anxiety and depressive reaction”, indicating the common presentations of the condition [2]. Similarly,
the American Psychiatric Association’s classifcation system, the Diagnostic and Statistical Manual
of Mental Disorders, 5th Edition (DSM-5) categorises AD as presenting: “with depressed mood”,
“with anxiety” and “with mixed anxiety and depressed mood” [3], and has symptomatic overlap with
depression and anxiety.
AD has been described as a controversial disorder [4], especially with respect to its role in the
classifications systems. The key characteristics of AD have remained stable since it was first described
in the diagnostic classification systems, and include symptoms common to both depressive and anxiety
disorders. Although, unlike depressive episode, there is no prescribed list of clinical symptoms
required for the diagnosis of AD in ICD-10 and DSM-5, and there may be significant clinical overlap
between the two conditions in terms of symptomatology [5,6]. Some symptoms, including arguably
the most severe and life-threatening symptom, suicidal ideation, may be as common in AD as it is in
depressive episode [7].
The proposed ICD-11 will re-categorise AD under conditions specifically caused by stress, along
with other conditions such as post-traumatic stress disorder. It will define the diagnosis of AD in a
more positive manner, in describing two core symptoms: (a) Preoccupation with the stressor, and (b)
failure to adapt. Preoccupation with a stressor includes recurring distressing thoughts or ruminations
on the theme of the stressful situation, while failure to adapt is a more general difficulty, which brings
as a consequence of the preoccupation disturbance including those of sleep and concentration resulting
in an impairment of function across key domains, such as social or occupational functioning [8].
AD has not been included in the major epidemiological studies of mental disorders, and as a
consequence, the incidence and prevalence rates in the general population are unknown. The two
clinical areas which have come closest to providing epidemiological data on this condition are general
practice and liaison psychiatry. Huyse et al., in a European study of fifty-six consultation-liaison
psychiatry services in eleven countries found that adjustment disorder accounted for a significant
proportion of psychiatry morbidity in acute medical hospitals [9]. While self-harm, at 17%, was the most
common reason for referral, adjustment disorder and post-traumatic stress disorder was diagnosed in
12.4% of those patients referred. Unfortunately, this paper did not examine the relationship between
self-harm and diagnosis [9].
Our hypothesis is that AD is a common disorder in patients who present to emergency
departments of hospitals with suicidal ideation and behaviours, i.e., present for assessment by
liaison psychiatry services.
In this study we aimed to examine the association between suicidal ideation and behaviours in
AD in an acute medical hospital setting. The objective of this study was to establish the incidence of
AD in patients who present with self-harm and suicidal ideation, and the rates of suicidal ideation and
behaviours among patients with a diagnosis of AD.

2. Materials and Methods

A comprehensive search strategy was developed and used to search electronic databases (PubMed,
CINAHL, Medline, Psychinfo and the Cochrane Library) for published studies on suicidal behaviours
in adjustment disorders using the search terms, “adjustment disorder”, “suicide”, “adjustment
disorder AND suicide”, “adjustment disorder AND overdose”, “adjustment disorder AND self-harm”,
“adjustment disorder AND suicidal ideation and behaviours” and “adjustment disorder AND general
hospital psychiatry”. The search was confined to material published within the last thirty-five years.
A further filter requiring the published articles to be peer reviewed was also applied. Studies written
in a language other than English studies were excluded, in addition to letters, editorials, commentaries
and textbooks.
Int. J. Environ. Res. Public Health 2019, 16, 2967 3 of 15

We included studies which met the following criteria:

(a) Included patients diagnosed with AD, either clinical diagnosis or using structured
diagnostic instruments.
(b) Conducted in medical settings where the specialty of liaison psychiatry is to be found i.e.,
emergency departments/rooms, general medical wards, critical care units etc.
(c) Described patients presenting with self-harm (regardless of suicidal intent) or suicidal ideation.
(d) Studies that included at least one clinical characteristic in addition to diagnosis (self-harm
methods, previous attempts, etc.
(e) Observational studies with or without comparison groups.

The exclusion criteria were review papers, letters, editorials, commentaries, abstracts for which
there were no data available. We excluded all studies from the non-liaison psychiatry population, i.e.,
those who recruited from anywhere other than a general medical setting, where there was specialist
psychiatry input.
The study selection process was conducted in the first instance by one reviewer (JF) and
independently validated by a second reviewer (AMD). A meta-analysis of data was planned, but could
not be performed due to inherent heterogeneity in the studies. This heterogeneity may explain why no
previous meta-analysis of this kind was identified in the Cochrane Library.
For rating the methodological quality of the included studies, this study used the Quality
Assessment Tool for Observational, Cohort, and Cross-Sectional Studies of the National Institutes of
Health (NIH) [10], as modified by Troya [11]. For each study, the quality was assessed independently
by the reviewers separately, to give a rating of high, moderate or poor to each study.
Results of the review were synthesised into a descriptive narrative under specific headings
highlighting the prevalence of AD, the demographic profile and the suicide methods used, and were
also summarised in a descriptive table (Table 1).
Int. J. Environ. Res. Public Health 2019, 16, 2967 4 of 15

Table 1. Characteristics of included studies (n = 20).

Previous or
No. of Study
Study Type Setting Age Diagnosis Self-Harm Method Subsequent Death Influencing Factors/Precipitants
Participants Length
Attempt
Females more likely to have dx AD
AD 30.1%;
Retrospective Previous (p < 0.01), stressful life events (p <
AbuMadidi ED, Saudi personality disorder 78.7% poisoning; 26%
study (chart 398 13–74 years 6 years attempt in Not stated 0.001). Males more likely to have
et al. (2001) Arabia 32%, depression cutting
review) 21.5% substance misuse (p < 0.001),
8.6%
psychosis (p < 0.01)
Women more likely to poison, men
Liaison more likely to cut or violent act.
Retrospective AD 35.9%; major
Brakoulias et psychiatry 79.2% poisoning; 12.7% 12% prior Separated and divorced women
study (chart 1295 18–88 years depression %; 5 years Not stated
al. (2006) service, cutting; 4.7% violent self-harm 18–24 high risk. Violent group, AD
review) schizophrenia
Australia less common than depression or
schizophrenia.
AD 48% most
common diagnosis,
Briskman et Prospective 92.5% poisoning; 7.5% AD 46.7% aged 16–64, 57.8% in
1149 ED Israel 18–95 years personality 16%; 8 years 30% Not stated
al. (2017) cohort study hanging; cutting >65 years
depression; 15%
depression
Liaison
Mean age in AD
psychiatry AD 49.7%; Younger age, single marital status
Casey et al., Prospective with suicidal
348 services in 3 depressive episode 6 months Not stated Not stated None and greater severity of depressive
(2015) cohort study behaviour 36.5
Dublin 51.3% symptoms.
years
Hospitals
AD 84.3%; major Female—80.64%, childhood
Farzeneh et Cross-sectional depression 18%; adversity—48%,
248 ED, Iran 12–18 years Not stated Self-poisoning Not stated Not stated
al. (2010) cohort study personality disorder family psychiatric history—33%,
10% substances—11%
9.24% of the
Retrospective AD 33.7%; Self-poisoning— most
Galgali et al. Mean age 25 sample had a Not Substance abuse, epilepsy,
study (chart 119 ED, India depression 21%; One year common being
(1998) years previous stated/unknown co-morbid psychiatric illnesses
review) schizophrenia 4.3% pesticides
attempt
Self-poisoning by
Retrospective 15–55 and above. AD 13.5%; mood
Ghimire et various compounds, Gender, substance abuse,
study (chart 200 ED, Nepal 77% below the disorder 11%; 3 months Not reported Not reported
al. (2012) pesticides being the interpersonal conflict
review) age of 34 years substance abuse 7%
most common
Family conflict—30% suicidal
Retrospective
Grundikoff 265 ED, New ideation, 41% self-harm.
study (chart 0–17 years AD 417.7% 1 year Not reported 57 (22.4%) Not reported
et al. (2015) 93 self-harm York Peer conflict—30% suicidal
review)
ideation, 41% self-harm.
Liaison
1795 Mean age
psychiatry
Huyse et al. Cross-sectional self-harm presenting with Self-harm 17%; Self-harm 56% female, 24%
services in 1 year Not reported Not reported Not reported
(2001) cohort study Total in self-harm 38 AD 12.4% transferred to psychiatric ward
11 European
study 10560 years
countries
Int. J. Environ. Res. Public Health 2019, 16, 2967 5 of 15

Table 1. Cont.

Previous or
No. of Study
Study Type Setting Age Diagnosis Self-Harm Method Subsequent Death Influencing Factors/Precipitants
Participants Length
Attempt
Stressors included end of
Retrospective Medical relationship (18%), debt (18%) and
Lin et al. AD 41.1%;
study (chart 73 admissions, 16–83 years 10 years Charcoal burning Not reported Not reported illness (18%). Male patients had
(2012) depression 49.3%
review) Taiwan higher rates of AD, comorbid with
alcohol abuse.
Mean 45.8 years n = 17 87 (50%) total
AD n = 17
(SD20) (9.8%)—3(7.6%) 0 rodenticide
Retrospective Medical (9.8%)—2(3.2%) AD significantly associated with
Lin et al. rodenticide Self-poisoning by either rodenticide group; 87 (58%)
study (chart 174 admissions, rodenticide group; 12 years presentation with paraquat
(2018) group; 41.2 rodenticide or paraquat group; 45(30%) paraquat group.
review) Taiwan 15(14.1%) paraquat poisoning (high lethality group)
years (SD 14.9) paraquat No detail by
group
paraquat group group diagnosis
Death by suicide
1 participant Female—62.5%
Medical Acute stress was an
Lingeswaren Prospective had a previous Stressors included parenting issues
40 admissions, 10–30 years reaction/Adjustment 6 months Self-poisoning exclusion
et al. (2016) cohort study suicide 47.5%, interpersonal difficulties
India disorder in 100% criterion of this
attempt 30%, academic 7%
study
Tertiary
Retrospective
Magat et al. centre in AD 29%; depressive Gender (female) 86%, age 13–16
study (chart 65 5–18 years 2 years Self-poisoning 26% None
(2008) Honolulu, illness (45%) 68%
review)
Hawaii
Absence or
presence of
92.9% overdose; 1.4%
AD 35.78%; previous
Retrospective ED, rural each for drowning,
McCauley et depressive disorder suicidal Gender Female: Male 2:1, alcohol
study (chart 70 hospital, 10 to >60 years 1 year hanging, inhaling None
al. (2001) (28.6%); behaviour is implicated in 47% of cases
review) Ireland exhaust fumes,
schizophrenia 7.1% documented
laceration of wrists
in 47.7% of
charts.
Prospective Toxicological 20% had a
Mitrev AD 100%, no Interpersonal conflict—70%,
descriptive 140 unit, 15 to >60 years 2 years Self-poisoning prior suicide None
(1996) additional diagnosis occupational/ economic—25%
study Germany attempt
AD less educated, lower social
AD group poisoning n
status, unmarried. Majority
= 60 (70%: males 19,
unfavourable childhood events.
Prospective 22%, female 41, 48%);
Polyakova, AD 55.5%; Alcohol 3 times more likely to be
observational 155 ED, Moscow 18–65 years 9 months hanging 17 (20%: males Not reported None
1998 depression 44.5% involved in AD than depression,
study 12, 14%, female 5, 6%);
more impulsive. AD regretted
other 9 (10%: males 4,
(92%, compared with only 12% in
5%, female 5, 5%)
the depression group)
Int. J. Environ. Res. Public Health 2019, 16, 2967 6 of 15

Table 1. Cont.

Previous or
No. of Study
Study Type Setting Age Diagnosis Self-Harm Method Subsequent Death Influencing Factors/Precipitants
Participants Length
Attempt
32% had
AD 77% of the more
previous
serious suicide None–the study
suicide Gender—86% of participants were
attempts and 50% of was conducted
Suss et al. Cross sectional ED, New attempt, with female. Ethnicity 82% of
92 12–18 years the less serious 2 years Self-poisoning exclusively on
(2004) cohort study York 6% having participants were African
suicide attempts. non-fatal suicide
two or more American
Other diagnoses are attempts
previous
not listed
attempts
129 (83.8%)
prior
self-harm, 25
(16.4) >x2
AD—49 (31.8%)
Figures given
125 clinically; 12 (7.8)
Poisoning 104 (83.2%); for whole
Taggart et al. Prospective self-harm, of SCID 54.5% female, 45.5% male. 67.5%
ED, Belfast 13–77 years 1 year cutting 10 (8%); other 11 sample, Not reported
(2006) cohort study 167 patients Depression 30 previous psychiatric treatment.
(8.8%) figures for
in study (19.5%) clinically; 56
those
(36.4%) SCID
presenting
with self-harm
not described
separately
AD (53.5%), Major
ED and Family conflict 24.5%; conflict with
Retrospective depression (24.3%), 90% poisoning; 6%
Wai et al. medical friends 23.6%; school problems
study (chart 214 13–21 years Schizophrenia 4 years mixed; 4% violent incl Not reported Not reported
(1999) admissions, 11%; military service in 10% of
review) (1.9%) Substance defenestration
Singapore males.
misuse (0.5%)
Initial
Prospective
interviews 242 or 79% of
descriptive No age range AD 42.1%; major 27% of cases
over 2-year the study group Marital conflict (30%), family
Zhargami et study, Burns unit, given. Average depression 11%; had a previous
318 period, Self-immolation died as a result problems (12%), “love affair” (10%),
al. (2002) included Iran age of 27 years anxiety d/o 4.7%; suicide
follow up of conflict with spouse’s family (5%)
psychological stated schizophrenia 4.1% attempt
interviews 8 self-immolation
autopsy
years later
Int. J. Environ. Res. Public Health 2019, 16, 2967 7 of 15

Int. J. Environ. Res. Public Health 2019, 16, x 4 of 15

Figure 1. Study flow diagram.

Figure 1. Study flow diagram.
3.1. Description of Studies
3. Results
The 20 included studies were all from a general hospital setting, including EDs (n = 11; 55%),
The initialwards
medical search(n =yielded
3, 15%) or3395 articles.toxicology/burns
specialised Of these, 348 units
articles
(n = were identified
2; 10%). as duplicates
The remainder (n = 4; and
discarded. A further 3028 were excluded during the title screen as they were not related
20%) included patients across these settings. Over half of the studies were from countries where to AD or to
suicidalEnglish
behaviours. The
is not the firstremaining
language (n32 articles
= 11; were
55%) with screened
40% by abstract,
(n = 8) from at which
English-speaking stage and
countries 10 of these
1 study from
were excluded, a number
leaving of English-speaking
22 articles andscreening
for the full text non-English speaking
process. countries.
A further 10Half (n = 10;
articles were50%)excluded,
were retrospective chart reviews, and the remainder either cross-sectional (n = 3; 15%) or prospective
with 12 studies remaining from the search. A further 8 studies were found by hand-searching the
cohort studies (n = 7; 35%). The studies are further described in Table 1.
references of the included studies. A total of 20 full text articles were included in the final review
(Figure 1).

3.1. Description of Studies

The 20 included studies were all from a general hospital setting, including EDs (n = 11; 55%),
medical wards (n = 3, 15%) or specialised toxicology/burns units (n = 2; 10%). The remainder
(n = 4; 20%) included patients across these settings. Over half of the studies were from countries where
English is not the first language (n = 11; 55%) with 40% (n = 8) from English-speaking countries and
1 study from a number of English-speaking and non-English speaking countries. Half (n = 10; 50%)
were retrospective chart reviews, and the remainder either cross-sectional (n = 3; 15%) or prospective
cohort studies (n = 7; 35%). The studies are further described in Table 1.

3.2. Methodological Quality of Studies

The 19 included studies were assessed using the NIH Quality Assessment Tool for Observational,
Cohort, and Cross-Sectional Studies [10]. Figure 2a provides an overview of the quality of the included
studies, and Figure 2b highlights the areas of risk across the included studies as a whole, grouping as
Int. J. Environ. Res. Public Health 2019, 16, x 8 of 15

3.2. Methodological Quality of Studies

The 19Res.included
Int. J. Environ. Public Healthstudies were
assessed using the NIH Quality Assessment Tool8 offor
2019, 16, 2967 15
Observational, Cohort, and Cross-Sectional Studies [10]. Figure 2a provides an overview of the
quality of the included studies, and Figure 2b highlights the areas of risk across the included studies
high, low or unclear risk. Identified high risk areas ≥80% included measure of and adjustment for key
as a whole, grouping as high, low or unclear risk. Identified high risk areas ≥80% included measure
confounding variables and blinding of assessors. Low risk areas ≥60% included clear elucidation of
of and adjustment for key confounding variables and blinding of assessors. Low risk areas ≥60%
the study question, clear specification of the population studied, clearly prespecified inclusion criteria,
included clear elucidation of the study question, clear specification of the population studied, clearly
sample size description and description of time frame. Areas of unclear risk ≥60% included loss to
prespecified inclusion criteria, sample size description and description of time frame. Areas of
follow up.
unclear risk ≥60% included loss to follow up.

Study question

Loss to follow

Confounding
Participation

Sample size

Time frame
Population

Inclusion

outcomes
Repeated
assessors

exposure

Defined
Blinded
criteria
Study

rate

up
Abumadani et al. (2001) √ √ √ √ √ √ x x √ - x

Brackoulias et al. (2006) √ √ √ √ √ √ x √ √ - x

Briskman et al. (2017) √ √ √ √ √ √ x √ √ x √

Casey et al. (2015) √ x x √ √ √ √ √ √ x √

Farzeneh et al. (2010) √ √ √ √ √ √ x x √ - x

Galgali et al. (1998) √ √ √ √ √ √ x x √ x x

Ghimire et al. (2012) √ √ x √ √ √ x x ? - x

Grundikoff et al. (2015) √ √ √ √ √ √ x ? √ x x

Huyse et al. (2001) √ √ x √ √ √ x x ? - x

Lin et al. (2012) √ √ √ √ √ √ x ? √ - x

Lin et al. (2018) √ √ √ √ √ √ x ? ? - √

Lingeswaren et al. (2016) √ √ √ √ √ √ x - ? - x

Magat et al. (2008) √ √ x √ √ √ x √ √ - x

McCauley et al. (2001) √ √ √ √ √ √ x ? √ - x

Mitrev (1996) √ √ x x √ √ x ? √ - x

Polyakova et al. (1998) √ √ x √ √ √ x - ? - x

Suss et al. (2004) √ √ √ √ √ √ x x √ - x

Taggart et al. (2006) √ x √ √ √ √ x x √ - -

Wai et al. (1999) √ ? √ √ √ √ x x √ - ?

Zhargami et al. (2002) √ √ √ √ √ x x - √ ? x

√: Reported appropriately in study

x: Not reported: element not mentioned
?: Lack of clarity in paper
-: Not applicable to study design

(a)

Figure 2. Cont.
Int. J. Environ. Res. Public Health 2019, 16, 2967 9 of 15
Int. J. Environ. Res. Public Health 2019, 16, x 9 of 15

Confounding
Loss to follow up
Defined outcomes
Repeated exposure
Blinded assessors
Time frame
Sample size
Inclusion criteria
Participation rate
Population
Study Question

0 5 10 15 20 25

Low risk Unclear risk High risk

(b)

Figure 2. (a) Quality assessment within studies, (b) Quality assessment across studies.
Figure 2. (a) Quality assessment within studies, (b) Quality assessment across studies.
3.3. Socio-Demographic Factors
3.3. Socio-Demographic Factors
3.3.1. Age
3.3.1. Age
Four of the twenty (20%) studies included, focused exclusively on paediatric populations [12–14],
Four (40%)
and eight of thehadtwenty (20%)
a mixed studies included,
demographic which focused
includedexclusively
those under onthe
paediatric populations
age of eighteen [12–
as well as
14], andOfeight
adults. these(40%) had a mixed
Lingeswaran et al.demographic which included
described a population with agethose under
range theto
of ten age of eighteen
thirty as
years [15];
well as adults. Of these Lingeswaran et al. described a population with age range
Mitrev reported an age range of fifteen to over sixty years [16], and Ghimire et al. described a population of ten to thirty years
[15];
of twoMitrev
hundred reported an age range
presentations to an ED of fifteen
in Nepal to that
overranged
sixty years
in age[16],
from and Ghimire
fifteen et al. described
to fifty-five years (77% a
population
of whom were of two
aged hundred
under 34 presentations to an ED inetNepal
years) [17]; McCauley that ranged
al. described in age from
a population agedfifteen to fifty-
between 10
five years (77% of whom were aged under 34 years) [17]; McCauley et al. described
and 60 years in an ED in a hospital in rural Ireland, close to half of the sample were aged less than a population aged
between
thirty 10Galgali
[18]; and 60 years in an ED in
et al. reported a hospital
a mean age of in25rural
(SDIreland,
= 8.1) inclose
theirtosample,
half of the sample
without were aged
describing a
less than
range [19];thirty [18]; et
Zarghami Galgali et al. reported
al. reported an average a mean
age of age of 25(SD
27 years (SD==13.5),
8.1) again
in their sample,
without without
specifying
describing
the age range a range [19]; Zarghami
[20]; Abumadani et al.
et al. reported
report an average
an patients age of
in Saudi 27 years
Arabia aged (SD = 13.5),
13–74 [21]again
and Waiwithout
et al.
specifying
focused on the age range
a young [20]; Abumadani
adult/adolescent populationet al. all
report
agedan patients
under in Saudi Arabia aged 13–74 [21]
21 [22].
and Wai
Fouretstudies
al. focused
(22.2%) on afocused
young exclusively
adult/adolescent on anpopulation all agedCasey
adult population, underet21al. [22].
reported a mean
age of 36.5 years (SD = 10.1) in those presenting with AD and suicidal behavioursreported
Four studies (22.2%) focused exclusively on an adult population, Casey et al. in three aDublin
mean
age of 36.5
hospitals years
[23], (SD = 10.1)
Polyakova in those
recruited onlypresenting
patients aged with AD18–65
over and suicidal behaviours
years to their ED-based in three
studyDublin
of AD
hospitals [23], Polyakova recruited only patients aged over 18–65 years to their
and self-harm [7]. Brakoulias, in a larger study of self-harm presentations to a liaison psychiatry service ED-based study of
AD and self-harm [7]. Brakoulias, in a larger study of self-harm presentations
in Australia included adults only aged 18–88 years [24]. Briskman examined patients presenting with to a liaison psychiatry
service inaged
self-harm Australia
over 18, included
comparing adults
thoseonlyagedagedover 18–88
and belowyears65[24].
years,Briskman examined
the only included patients
study that
presenting with self-harm aged over 18, comparing those aged over and
specifically examined older patients [25]. The remaining four studies did not specify the age range below 65 years, the only
included
of study that specifically examined older patients [25]. The remaining four studies did not
participants.
specify the age range of participants.
3.3.2. Gender
3.3.2. Gender
Eighteen of the twenty studies (90%) showed a higher proportion of females than males in suicidal
Eighteenwith
populations, of the
fivetwenty
studies studies (90%) showed
having women a higher
representing moreproportion
than 80% of ofthefemales
sample: than
86% males
femalein
suicidal
in Magatpopulations,
et al.’s study, with83% infive studies et
Zhargami having women
al., 81.5% representing
in Suss et al., 80.4% more than 80%
in Farzeneh et of
al. the
andsample:
80% in
86% female in
Abumadani Magat
et al. et al.’s study,
[12–14,20,21]. The83%two in Zhargami
studies with aetmajority
al., 81.5% ofin Sussparticipants
male et al., 80.4%were in Farzeneh
both from et
al. and 80% in Abumadani et al. [12–14,20,21]. The two studies with a majority of male participants
were both from Taiwan: with 60% male participants in Lin’s 2012 study of charcoal poisoning, and
Int. J. Environ. Res. Public Health 2019, 16, 2967 10 of 15

Taiwan: with 60% male participants in Lin’s 2012 study of charcoal poisoning, and 75% male in Lin’s
2018 study of rodenticide and paraquat poisoning (78% in the paraquat subgroup were male) [26,27].

3.4. Frequency of Adjustment Disorder (AD) Diagnosis

The majority of the studies used a clinical diagnosis rather than a diagnosis based on a
semi-structured interview: Only three studies (15.7%) used a semi-structured interview. Zhargami
et al. used the Structured Clinical Interview DSM version 1 (SCID-I) a diagnostic semi-structured
interview based on DSM-III [20]. Taggart et al. also used SCID, and noted a significant difference
in the rates of diagnosis depending on whether clinical diagnosis or SCID diagnosis was used [28].
This study reported rates of AD of 32% when using clinical diagnosis, and 7.8% when using SCID.
Casey et al. similarly used both the Schedule for Clinical Assessment in Neuropsychiatry (SCAN)
a diagnostic semi-structured interview based on ICD-10, and clinical diagnosis [23]. Casey et al.
ultimately reported on the clinical diagnosis rather than the SCAN diagnosis, noting the inherent
weaknesses in the semi-structured schedules in the diagnosis of AD [23].
Three studies specifically selected patients with AD, the remainder examined a more general
cohort of patients presenting with suicidal ideation and behaviours. Mitrev’s study selected only
patients with AD attending a toxicology unit for emergency treatment of self-poisoning, and examined
the characteristics of these patients in terms of ongoing suicidal risk—they found a significantly higher
risk in those with chronic AD and pervious suicidal behaviours [16]. Polyakova and Casey both
selected patients with AD and compared them to patients with major depression [7,23]. Polyakova’s
study of 155 participants recruited from a Moscow ED had 55.5% AD and 44.5% depression [7]. Casey’s
study recruited 348 patients from three Dublin hospitals: 49.7% of whom had a clinical diagnosis of
AD, and 50.3% depression [23].
The remaining studies reported the percentage of patients with a diagnosis of AD where the
researchers were not specifically recruiting this diagnosis. A number of the studies reviewed found
AD to be a common diagnosis among those presenting for emergency assessment following self-harm.
The lowest proportion of AD found was reported by Lin et al. (9.8%) in patients presenting with
self-poisoning with either rodenticide or paraquat in Taiwan [27]. Ghimire et al. reported that 13.5%
of the patients presenting to a Nepalese ED with self-poisoning had a clinical diagnosis of AD [17].
Magat reported AD in 29% of patients presenting for treatment following self-poisoning in Hawaii [13].
Abumadani et al. found AD was the clinical diagnosis in 30.1% of their study population [21]. Taggart
at al. examined patients who presented to emergency departments in Belfast following self-harm and
found AD (31.8%) was 1.5 times as common as depression (19.5%) [28]. Galgali’s ED based study,
where ingestion of pesticides was the most common form of self-injury, found 33.7% of the 119 cases of
attempted suicides, referred for psychiatric assessment over a 12 month period, received a diagnosis of
AD—the most common diagnosis in this study [19]. AD was the most common diagnosis in McCauley
et al.’s study of self-harm in Ireland at 35.8% [18].
An Australian study of emergency referrals found that 35.9% of referrals to a new psychiatric
Emergency Care Centre in Sydney had a diagnosis of AD; furthermore, AD was the most common
diagnosis in those presenting with suicidal behaviours [24]. A 10-year retrospective study of attempted
suicide by charcoal burning in Taiwan, where this is a common method of suicide, found that 41% of
people presenting with attempted suicide by this method met the diagnostic criteria for a diagnosis of
AD [26]. Zhargami et al., in a study based in a burns unit in Iran found that 42.1% of patients referred
with self-immolation had a clinical diagnosis of AD [20].
Wai et al. reported a diagnosis of AD in 53.5% of patients presenting with self-injury to an ED
in Singapore [22]. Suss et al. reported AD in 77% of serious or high-risk suicide attempts, and 50%
of the lower-risk attempts [14]. Farzaneh et al. found 80% of a population of students presenting
with self-poisoning to a specialist poison centre in Tehran over a year, had a diagnosis of AD [12].
Lingeswaren et al. reported that 100% of the people seen with self-poisoning had a diagnosis of AD. It
Int. J. Environ. Res. Public Health 2019, 16, 2967 11 of 15

is not clear from the study whether a diagnosis of AD was a selection criterion or whether all patients
who presented during the timeframe of the studies happened to have a diagnosis of AD [15].

3.5. Suicide Attempt Method and Mortality Rates

The majority (eighteen) of the twenty studies in this review examine cohorts who have presented
with suicidal behaviours. Casey et al., and Grundikoff et al. both report on individuals presenting with
suicidal behaviours as well as suicidal ideation [23,29]. Lingeswaran et al. had death as an exclusion
criterion. This study reported that 60% of the sample acted impulsively i.e., within thirty minutes of the
suicidal thought and that interestingly, 97% had no previous attempt or family history of suicide [15].
Only two studies reported on patients who died by suicide following their presentation with
self-injury. As a result, this study cannot comment on mortality rates in this population. Zarghami et
al. reported that 79% of the cohort had died as a result of self-immolation [20], and Lin et al. reported
in their study of poisoning by rodenticide and paraquat that 50% of the cohort (all in the paraquat
group) died [27].
Overall, the most common form of suicide attempt was self-poisoning, the sole means of attempt
in almost half of the included studies (n = 8; 40%), and the most common method in a further seven
studies: 70%, 78.7%, 79.8%, 83.2%, 90%, 92.5% and 92.9%, respectively [7,18,21,22,24,28,29]. Suss
et al. examined adolescents attending a New York ED for treatment of non-fatal overdoses. They
found that the majority (77%) of the more serious suicide attempts received a diagnosis of AD from a
consultant psychiatrist [14]. Grundikoff et al. did not provide any detail on the suicidal behaviours in
the paediatric population studied [29].

3.6. Precipitants
Casey et al. compared suicidality in two groups, one with a diagnosis of AD and the other with
a depressive episode. They found that those with AD experienced more life events, higher rates of
personality disorders and higher rates of suicidal behaviours at a younger age and a lower depressive
symptom threshold, than those with a depressive episode. The possible role of personality disorder in
this finding was insignificant on multivariable analysis [23]. Farzeneh et al. found that almost a third
reported romantic disappointment as the main reason for attempting suicide whilst more than half
claimed family conflict [12]. A 1998 study of 308 people presenting with self-poisoning to a hospital in
Bangalore, found that more than a quarter cited problems within their primary support group (26%)
as the main stressor, whilst 58% had no identifiable trigger for their suicide attempt [19]. The most
common precipitant of the suicidal act in Mitrev et al.’s study, was problems in the primary support
group (in most cases, family) which was reported in 98 (70%) of the 140 cases [16]. Wai et al. found that
24.5% of patients who attended an ED in Singapore after a suicide attempt cited conflict with family as
their suicidal trigger whilst a further 23.6% alluded to conflict with friends [22]. Magat et al. found
that 22% of those attending an ED in Hawaii had had an argument with a family member whilst 11%
had experienced conflict with a significant other prior to a suicide attempt [13]. Ghimire et al. make a
distinction between interpersonal conflict and conflict within a marriage, and found that 72% of the
cohort (n = 200) presenting for medical treatment for deliberate self-harm, identified interpersonal
conflict as the trigger for suicidal behaviour, whilst 14.5% cited marital conflict. A further 3.5% claimed
romantic disappointment [17]. Grundikoff et al. reported family conflict in 41% and peer conflict in
20.4% of the patients presenting with suicidal behaviours in the paediatric population studied [29].

4. Discussion
AD is a common condition among patients presenting for treatment following suicidal behaviours,
across the studies where it is recorded as a diagnosis. AD is diagnosed with high frequency in suicidal
populations across multiple studies in differing nationalities and ethnic groups (Table 1). AD was the
exclusive diagnosis in two of the studies: Lingeswaran et al., retrospectively examined case notes of
adolescents presenting to an emergency department in India for treatment post self-poisoning [15],
Int. J. Environ. Res. Public Health 2019, 16, 2967 12 of 15

and Mitrev examined 140 patients attending a toxicology centre in Germany after self-poisoning in a
prospective study [16]. A majority of the included studies were retrospective reviews of case-notes.
This methodology brings with it some biases (selection bias, information bias).
The rates of diagnosis of AD among individuals presenting with self-harm are not dissimilar to
the rates reported in psychological autopsy studies, although are lower on average. Portzky et al.,
in a psychological autopsy study in Belgium, found AD to be the second most common diagnosis
in this group, accounting for 21.1% [30]. Likewise, Martunnen found that 21% of adolescent deaths
by suicide were related to a likely diagnosis of AD [31]. Lin’s national database study of Taiwanese
people admitted to medical and psychiatric hospitals with self-harm (n = 57,874) reported that AD is
associated with a significantly increased risk of repeated suicidal behaviours (OR 1.8) but a significantly
reduced risk of death by suicide (OR 0.12) [32].
The studies included in this review focused on a variety of age groups. From children and
adolescents only, adults only to the whole range of ages presenting with suicidal behaviours, with a
number of studies not identifying the age range included. The data included here suggests that AD is
an important diagnosis in young people and one that is associated with severe symptoms. Most of the
studies reported a majority of females presenting with suicidal behaviours.
In most studies included, the most common form of suicidal behaviour reported was self-poisoning,
with three-quarters (n = 15; 75%) of studies reporting that >70% of participants used this means.
Triggers or precipitants were varied, but interpersonal difficulties in various forms including family
and romantic were commonly reported as the precipitating stressors. Similarly, in a study of adolescent
inpatients in psychiatric hospitals, Chiou et al., found 25% of those who had attempted suicide cited
conflict with a parent as the main precipitant to suicidal behaviour, whilst 10% reported interpersonal
difficulty either within a romantic relationship or with a friend as the main stressor [33]. A systematic
review described psychological pain as a key factor identified in the suicide notes of people who died
by suicide [34]. The findings of this study, and in psychological autopsy studies suggest that there
may be overlap between this psychological pain and the diagnosis of AD, which is characterised by
significant distress regarding one or more stressors (causing psychological pain).
Perhaps the most striking finding of this paper is the small number of studies (of the great many
which have examined suicidal behaviours) which have included AD as a diagnosis. This may be
related to the inherent difficulties in diagnosing AD when relying on structured interviews, many of
which only include AD in an appendix, only to be used if the threshold for another disorder cannot
be met. This approach, ignoring context, has been criticised by many researchers in the area of
stress-related disorders [5,35]. The majority of the studies included have used clinical diagnosis, and as
a result have utilised the clinician’s clinical judgment about the role of context and stressors in the
patients’ presentations. This might be perceived as a weakness of the included studies, but given the
controversy around AD and its diagnosis using structured tools leading to researchers describing
clinical diagnosis for all its faults as the “gold standard”, it can be argued that this is instead a strength
of these studies [36]. In just two of the included studies, both from Ireland by Taggart et al. and Casey
et al., clinical diagnosis and semi-structured clinical interview schedules were used. In both cases the
semi-structured clinical interview schedules diagnosed depressive episode, where the clinical diagnosis
was AD [23,28]. A possible solution to this difficulty has been presented by the new classification
system of ICD-11, which gives a clearer framework to allow a diagnosis of adjustment to be made using
positive symptoms and accounting for clinical context [8]. This will allow future diagnostic schedules
to include AD in a more consistent and reproducible manner, and will strengthen the research in
this area.

4.1. Strengths and Limitations

This is the first systematic review of the association between the diagnosis of AD, and suicidal
ideation and behaviour, and encompasses all the literature published in the area as identified by the
literature search.
Int. J. Environ. Res. Public Health 2019, 16, 2967 13 of 15

The conclusions of this study are limited by the paucity of research in the area. We identified
14 studies, most of which were observational studies of small numbers of patients—the largest being
348 patients.
Another limitation to this study is absence of data in most of the studies of the degree of suicidal
ideation or intent underpinning the suicidal presentations. This is also related to the methodology of
retrospective review, used in the majority of the studies.

4.2. Further Research

This study identifies the need for further research into both AD as a diagnosis and into the
association of this diagnosis with suicidal behaviours. This systematic review suggests that there
is a strong association between suicidal ideations and behaviours and AD, especially in the general
hospital setting.

5. Conclusions
This study confirms the association of AD with suicidal ideation and behaviours in multiple
countries and once more highlights the increased risk in young adults, particularly females. Given the
high representation of self-poisoning as a method of suicide attempt, future public health campaigns
may need to consider stricter controls on over the counter medications and education of populations
regarding safer practices around storage of potentially toxic compounds like pesticides. AD represents
an important disorder to target in suicide prevention initiatives.

Author Contributions: Conceptualization: J.F., A.M.D.; methodology J.F., A.M.D.; software, J.F., A.M.D.;
validation, J.F., A.M.D.; formal analysis, J.F., A.M.D; investigation, J.F., A.M.D.; resources, J.F., A.M.D.; data
curation, J.F.; writing—original draft preparation, J.F.; writing—review & editing, J.F., A.M.D.; visualization, J.F.,
A.M.D.; supervision, A.M.D.; project administration, J.F., A.M.D.
Funding: This research received no external funding.
Acknowledgments: This study was unfunded. We would like to acknowledge the support of the librarians at
University Hospital Galway, especially Denise Duffy.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Silverstone, P.H. Prevalence of psychiatric disorders in medical inpatients. J. Nerv. Ment. Dis. 1996, 184,
43–51. [CrossRef] [PubMed]
2. World Health Organisation. International Classification of Diseases, 10th ed.; World Health Organisation:
Geneva, Switzerland, 1992.
3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 5th ed.;
American Psychiatric Association: Washington, DC, USA, 2013.
4. Zimmerman, M.; Martinez, J.H.; Dalrymple, K.; Martinez, J.H.; Chelminski, I.; Young, D. Is the distinction
between adjustment disorder with depressed mood and adjustment disorder with mixed anxious and
depressed mood valid? Ann. Clin. Psychiatry 2013, 25, 257–265. [PubMed]
5. Baumeister, H.; Maercker, A.; Casey, P. Adjustment disorder with depressed mood: A critique of its DSM-IV
and ICD-10 conceptualisations and recommendations for the future. Psychopathology 2009, 42, 139–147.
[CrossRef] [PubMed]
6. Casey, P. Diagnosing adjustment disorder with depressive features. Expert Rev. Neurother. 2008, 8, 1203–1208.
[CrossRef] [PubMed]
7. Polyakova, I.; Knobler, H.Y.; Ambrumova, A.; Lerner, V. Characteristics of suicidal attempts in major
depression versus adjustment reactions. J. Affect. Disord. 1998, 47, 159–167. [CrossRef]
8. Maercker, A.; Brewin, C.R.; Bryant, R.A.; Cloitre, M.; Van Ommeren, M.; Jones, L.M.; Humayan, A.; Kagee, A.;
Llosa, A.E.; Rousseau, C.; et al. Diagnosis and classification of disorders specifically associated with stress:
Proposals for ICD-11. World Psychiatry 2013, 12, 198–206. [CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2019, 16, 2967 14 of 15

9. Huyse, F.J.; Herzog, T.; Lobo, A.; Malt, U.F.; Opmeer, B.C.; Stein, B.; Jonge, P.; van Dijck, R.; Creed, F.;
Crespo, M.D.; et al. Consultation-Liaison psychiatric service delivery: Results from a European study. Gen.
Hosp. Psychiatry 2001, 23, 124–132. [CrossRef]
10. National Institutes of Health (NIH). Quality Assessment Tool for Observational Cohort and Cross-Sectional
Studies. NIH, 2018. Available online: https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-
tools (accessed on 12 August 2019).
11. Troya, M.I.; Babatunde, O.; Polidano, K.; Bartlam, B. Self-harm in older adults: Systematic review. Br. J.
Psychiatry 2019, 214, 186–200. [CrossRef]
12. Farzaneh, E.; Mehrpour, O.; Alfred, S.; Moghaddam, H.H.; Behnoush, B.; Seghatoleslam, T. Self-poisoning
suicide attempts among students in Tehran, Iran. Psychiatr. Danub. 2010, 22, 34–38.
13. Magat, R.C.; Guerrero, A.P.S. Suicidal behaviour trends in a pediatric population in Hawai’i. Hawaii Med. J.
2008, 673, 69–73.
14. Suss, A.; Homel, P.; Wilson, T.E.; Shah, B. Risk factors for nonfatal suicide behaviors among inner-city
adolescents. Pediatr. Emerg. Care 2004, 20, 426–429. [CrossRef] [PubMed]
15. Lingeswaran, A. Profile of Young Suicide Attempt Survivors in a Tertiary Care Hospital in Puducherry.
Indian J. Psychol. Med. 2016, 38, 533–539. [CrossRef]
16. Mitrev, I. A study of deliberate self-poisoning in patients with adjustment disorders. Folia Med. 1996, 38,
11–16.
17. Ghimire, S.; Devkota, S.; Budhathoki, R.; Sapkota, N.; Thakur, A. Psychiatric Comorbidities in Patients with
Deliberate Self-Harm in a Tertiary Care Center. J. Nepal. Med. Assoc. 2014, 52, 697–701. [CrossRef]
18. McCauley, M.; Russell, V.; Bedford, D.; Khan, A.; Kelly, R. Assessment following deliberate self-harm: Who
are we seeing and are we following the guidelines? Ir. J. Psychol. Med. 2001, 18, 116–119. [CrossRef]
[PubMed]
19. Galgali, R.B.; Rao, S.; Ashok, M.V.; Appaya, P.; Srinivasan, K. Psychiatric diagnosis of self poisoning cases:
A general hospital study. Indian J. Psychiatry 1998, 40, 254–259. [PubMed]
20. Zarghami, M.; Khalilian, A. Deliberate self-burning in Mazandaran, Iran. Burn. J. Int. Soc. Burn Inj. 2002, 28,
115–119. [CrossRef]
21. AbuMadini, M.S.; Rahim, S.I.A. Deliberate self harm in a Saudi university hospital: A case series over 6 years.
Arab J. Psychiatry 2001, 12, 22–35.
22. Wai, B.H.; Hong, C.; Heok, K.E. Suicidal behavior among young people in Singapore. Gen. Hosp. Psychiatry
1999, 21, 128–133.
23. Casey, P.; Jabbar, F.; O’Leary, E.; Doherty, A.M. Suicidal behaviours in adjustment disorder and depressive
episode. J. Affect. Disord. 2015, 174, 441–446. [CrossRef]
24. Brakoulias, V.; Ryan, C.; Byth, K. Patients with deliberate self-harm seen by a consultation-liaison service.
Australas Psychiatry 2006, 14, 192–197. [CrossRef]
25. Briskman, I.; Shelef, A.; Berger, U.; Baruch, Y.; Bar, G.; Asherov, J.; Lvovski, A.; Apter, A.; Barak, Y. Deliberate
Self-Harm in Older Adults: A General Hospital Emergency Department Survey. Isr. Med. Assoc. J. 2017, 19,
160–163.
26. Lin, C.; Yen, T.H.; Juang, Y.-Y.; Leong, W.C.; Hung, H.-M.; Ku, C.-H.; Li, J.-L.; Lee, S.-H. Comorbid psychiatric
diagnoses in suicide attempt by charcoal burning: A 10-year study in a general hospital in Taiwan. Gen.
Hosp. Psychiatry 2012, 34, 552–556. [CrossRef]
27. Lin, C.; Yen, T.H.; Juang, Y.Y.; Lee, C.P.; Lee, S.H. Distinct psychopathology of patients who attempted suicide
with rodenticide in Taiwan: A comparative study with patients of suicide with paraquat. Psychol. Res. Behav.
Manag. 2018, 29, 323–328. [CrossRef]
28. Taggart, C.; O’Grady, J.; Stevenson, M.; Hand, E.; McClelland, R.; Kelly, C. Accuracy of diagnosis at routine
psychiatric assessment in patients presenting to an accident and emergency department. Gen. Hosp.
Psychiatry 2006, 28, 330–335. [CrossRef]
29. Grudnikoff, E.; Soto, E.C.; Frederickson, A.; Birnbaum, M.L.; Saito, E.; Dicker, R.; Kane, J.M.; Correll, C.U.
Suicidality and hospitalization as cause and outcome of pediatric psychiatric emergency room visits. Eur.
Child. Adolesc. Psychiatry 2015, 24, 797–814. [CrossRef]
30. Portzky, G.; Audenaert, K.; van Heeringen, K. Suicide among adolescents: A psychological autopsy study of
psychiatric, psychosocial and personality-related risk factors. Soc. Psychiatry Psychiatr. Epidemiol. Int. J. Res.
Soc. Genet. Epidemiol. Ment. Health Serv. 2005, 40, 922–930. [CrossRef]
Int. J. Environ. Res. Public Health 2019, 16, 2967 15 of 15

31. Marttunen, M.J.; Aro, H.M.; Henriksson, M.M.; Lönnqvist, J.K. Adolescent suicides with adjustment disorders
or no psychiatric diagnosis. Eur. Child. Adolesc. Psychiatry 1994, 3, 101–110. [CrossRef]
32. Lin, Y.W.; Huang, H.C.; Lin, M.F.; Shyu, M.L.; Tsai, P.L.; Chang, H.J. Influential Factors for and Outcomes of
Hospitalized Patients with Suicide-Related Behaviors: A National Record Study in Taiwan from 1997–2010.
PLoS ONE 2016, 11, e0149559. [CrossRef]
33. Chiou, P.-N.; Chen, Y.-S.; Lee, Y.-C. Characteristics of adolescent suicide attempters admitted to an acute
psychiatric ward in Taiwan. J. Chin. Med Assoc. JCMA 2006, 69, 428–435. [CrossRef]
34. Pompili, M.; Belvederi Murri, M.; Patti, S.; Innamorati, M.; Lester, D.; Girardi, P. The communication of
suicidal intentions: A meta-analysis. Psychol. Med. 2016, 46, 2239–2253. [CrossRef]
35. Strain, J.J.; Diefenbacher, A. The adjustment disorders: The conundrums of the diagnoses. Compr. Psychiatry
2008, 49, 121–130. [CrossRef]
36. Doherty, A.M.; Kelly, B.D.; Jabbar, F.; Casey, P. Distinguishing between adjustment disorder and depressive
episode in clinical practice: The role of personality disorder. J. Affect. Disord. 2014, 168, 78–85. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).