See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281588585

Journal of Anxiety Disorders

Article  in  Journal of Anxiety Disorders · September 2015

DOI: 10.1016/j.janxdis.2015.07.004

CITATIONS READS

30 8,529

6 authors, including:

Margaret Mclafferty Cherie Armour

Ulster University Queen's University Belfast
25 PUBLICATIONS   273 CITATIONS    206 PUBLICATIONS   5,268 CITATIONS   

SEE PROFILE SEE PROFILE

Siobhan O ' Neill Sam Murphy

Ulster University Ulster University
151 PUBLICATIONS   5,354 CITATIONS    38 PUBLICATIONS   1,531 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Northern Ireland Veterans' Health and Wellbeing Study View project

Psychopathy in Chinese: validation of several psychopathic trait measurements in Chinese View project

All content following this page was uploaded by Margaret Mclafferty on 10 January 2018.

The user has requested enhancement of the downloaded file.

 Journal of Anxiety Disorders 35 (2015) 42–48

Contents lists available at ScienceDirect

Journal of Anxiety Disorders

Childhood adversity profiles and adult psychopathology in a

representative Northern Ireland study
Margaret McLafferty a,∗ , Cherie Armour b , Aine McKenna a , Siobhan O’Neill a ,
Sam Murphy a , Brendan Bunting a
a
School of Psychology, Ulster University, Magee Campus, Northland Road, Derry, Northern Ireland, United Kingdom
b
School of Psychology, Ulster University, Coleraine Campus, Cromore Road, Coleraine, Northern Ireland, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: Childhood adversities are key aetiological factors in the onset and persistence of psychopathology. The
Received 24 February 2015 aims of this study were to identify childhood adversity profiles, and investigate the relationship between
Received in revised form 10 June 2015 the adversity classes and psychopathology in Northern Ireland. The study utilized data from the Northern
Accepted 24 July 2015
Ireland Study of Health and Stress, an epidemiological survey (N = 1986), which used the CIDI to examine
Available online 12 August 2015
mental health disorders and associated risk factors. Latent Class Analysis revealed 3 distinct typologies;
a low risk class (n = 1709; 86%), a poly-adversity class (n = 122; 6.1%), and an economic adversity class
Keywords:
(n = 155; 7.8%). Logistic Regression models revealed that individuals in the economic adversity class had
Childhood adversity
Psychopathology
a heightened risk of anxiety and substance disorders, with individuals in the poly-adversity class more
Epidemiology likely to have a range of mental health problems and suicidality. The findings indicate the importance of
Northern Ireland considering the impact of co-occurring childhood adversities when planning treatment, prevention, and
intervention programmes.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction attachment figure impacts greatly on psychopathology (Carlson &

Epidemiological studies have found that childhood adversities
(CAs) can have a detrimental impact on psychological wellbeing.
Unfortunately, CAs are highly prevalent with more than two-thirds
of children experiencing at least one trauma before the age of
16 (Copeland, Keeler, Angold, & Costello, 2007). Using data from
the WHO World Mental Health (WMH) Survey Initiative, Kessler
et al. (2010) reported that CAs accounted for 29.8% of psychiatric
morbidities worldwide and that CAs were inter-correlated, had
a cumulative effect, and predicted further adversities; mirroring
findings from previous studies (Copeland et al., 2007; Dong et al.,
2004).
Adversities associated with maladaptive family functioning
(MFF) have, in particular, been strongly linked to the onset (Green
et al., 2010) and persistence of mental health problems (McLaughlin
et al., 2010), especially anxiety disorders. McLaughlin et al. (2010)
proposed that the detrimental impact may be due to the frequency
or duration of these types of adversities which are likely to be
enduring. Other studies have also found that trauma caused by an
∗ Corresponding author. Tel.: +00353 86 1551293.
E-mail address: mclafferty-m1@email.ulster.ac.uk (M. McLafferty).
Dalenberg, 2000). Fryers and Brugha (2013) proposed that parent-
ing style and quality, neglect, parental divorce, parental separation,
and exposure to inter-parental violence during childhood, all play
a role in adult psychiatric disorders. A recent systematic review
of adverse childhood experiences suggested that child abuse and
family violence have the greatest impact on future mental health
(DeVenter, Demyttenaere, & Bruffaerts, 2013).
Weich, Patterson, Shaw, and Stewart-Brown (2009) reported
that abusive and dysfunctional family relationships predicted a
range of disorders, including anxiety, PTSD and depression. CAs
have also been linked to substance use and externalizing disorders
(Slopen et al., 2010). Additionally, a strong association has been
found between CAs and suicidal behaviour (Bruffaerts et al., 2010).
While much of the epidemiological research into the associa-
tion between CAs and psychopathology is American based, WMH
surveys conducted in other countries have reported similar find-
ings. The Health 2000 project in Finland found that those who had
experienced CAs were almost twice as likely to have mental health
problems (Pirkola et al., 2005). The Mexican National Comorbidity
Survey found that family dysfunction and abuse were associ-
ated with psychopathology across the lifespan (Benjet, Borges, &
Medina-Mora, 2010). The Nigerian Survey of Mental Health and
Well-being (NSMHW) reported that adversities involving parental

http://dx.doi.org/10.1016/j.janxdis.2015.07.004
0887-6185/© 2015 Elsevier Ltd. All rights reserved.
 M. McLafferty et al. / Journal of Anxiety Disorders 35 (2015) 42–48
maladjustment, impacted significantly on mental health in adult-
hood (Oladeji, Makanjuola, & Gureje, 2010).
Some surveys however have suggested that there may be cross-
national variations. The South Africa Stress and Health Study
reported that Africans were more likely to have higher prevalence
rates of CAs than Whites, with CAs impacting particularly on anx-
iety disorders (Slopen et al., 2010). While American based studies
have suggested that CAs predict psychopathology in general (Green
et al., 2010; McLaughlin et al., 2011), a Japanese epidemiological
study reported that certain disorders were associated with spe-
cific adversities (Fujiwara & Kawalkami, 2011). Mood disorder, for
example, was significantly related to parental mental health and
physical abuse; however, no specific childhood adversity was sig-
nificantly associated with the onset of anxiety disorders. While
Oladeji et al. (2010), in the NSMHW, found no significant relation-
ship with any specific adversity, they concluded that those who
experienced three or more MFF adversities were 12 times more
likely to develop anxiety disorders.
Traditionally, research assessing the link between CA and
psychopathology has focused on one particular adversity. Contem-
porary knowledge now indicates that CAs often co-occur (Armour,
Elklit, & Christoffersen, 2014; Dunn et al., 2011; Rosenman &
Rodgers, 2004). Accounting for co-occurrence rather than focus-
ing on the impact of single adversities is of paramount importance,
given it is probable that different combinations of adversities may
account for varying mental health outcomes. In addition, separat-
ing out the impact of a single adversity is problematic. Indeed,
Shevlin and Elklit (2008) suggested that research should focus on
individuals with similar adversity profiles.
Recent research has employed Latent Class Analysis (LCA)
to identify the co-occurrence of CA’s; also referred to as poly-
victimisation. Studies examining high risk individuals, such as
those in the juvenile justice system (Ford, Grasso, Hawke, &
Chapman, 2013) or those involved in mental health and social ser-
vices (Hazen, Connelly, Roesch, Hough, & Landsverk, 2009) reported
that individuals with childhood maltreatment profiles have more
psychological problems, with poly-victimised subgroups being par-
ticularly vulnerable (Ford et al., 2013). A number of studies have
utilized the latent profiles to examine associations with mental
health outcomes. Indeed, Dunn et al. (2011) found that different
adversity profiles were associated with a range of specific mental
health problems.
The aims of the current study were (1) to examine the nature
and frequency of CAs reported in NI, (2) to assess poly-victimization
across 12 independent childhood adversities and (3) to assess the
associations between childhood adversity classes and subsequent
mental health and suicidal outcomes in the NI population. Based
on previous literature, it was predicted that those categorized into
groups endorsing multiple childhood adversities would have an
increased likelihood of reporting psychiatric morbidity and suicidal
ideation and behaviour.
2. Method
2.1. Sample
As part of the WHO World Mental Health (WMH) Survey
Initiative (Kessler & Üstün, 2008), face-to-face household inter-
views were conducted by certified lay interviewers in NI between
February 2004 and August 2008. This study utilizes data obtained
from this epidemiological survey, known as the NI Study of Health
and Stress (NISHS). Full details on sampling methodology used in
the NISHS can be found in Bunting, Ferry, Murphy, O’Neill, and
Bolton (2013). The response rate for the NISHS was 68.4%. A total of
4340 participants (1899 males, 2441 females) completed part 1 of
43
the survey. Part 2 was then completed by a sub-sample of 1986 of
the original participants (950 males, 1036 females). All participants
were 18 years of age or older and were English speakers, residing
in NI. Non-English speakers, people residing in institutions, those
living in shared accommodation, including military barracks and
prisons, and those with learning disabilities were excluded. The
investigation was carried out in accordance with the latest version
of the Declaration of Helsinki. Ethical approval was obtained from
the University of Ulster Research Ethics Committee and informed
consent was obtained from all participants after the study had been
fully explained.
2.2. Diagnostic assessment
The WHO World Mental Health (WMH) Survey Initiative is a
collaborative, epidemiological, systematic study which assesses
the incidence, prevalence, and correlates of mental health prob-
lems using the WMH Composite International Diagnostic Interview
(CIDI) version 3.0 (Kessler & Üstün, 2004). This standardized
instrument retrospectively assesses mental health disorders in
accordance with ICD-10 and DSM-IV criteria and definitions, with
diagnostic hierarchy rules (Kessler et al., 2010). The current study
examines any mood disorders (major depressive, dysthymic, bipo-
lar I and II and sub-threshold bipolar disorders), any anxiety
disorders (GAD, PTSD, panic, agoraphobia without panic disorder,
social phobia, specific phobia, and separation anxiety disorders),
any substance disorders (drug and alcohol abuse, drug and alcohol
dependence with abuse) and any suicide ideation and behaviour
(gestures, plans, and attempts). In the NISHS all participants com-
pleted part 1 of the fully structured interview, which includes
a screening section, demographic variables and core diagnostic
assessment. Part 2, which is analysed in this study, contains a range
of diagnostic sections, along with risk factors such as childhood
adversities, consequences, and treatment, was completed by all
participants who responded positively to core mental health disor-
der screening questions, plus 50% of those who were sub-threshold
cases and 25% of other participants who did not meet either crite-
rion; this allowed for the calculation of sampling weights.
2.3. Childhood adversities assessment
CA questions are included in the childhood and PTSD sections of
the survey. This study utilized 12 CAs identified in previous WMH
surveys (cf. Kessler et al., 2010). The retrospectively reported CAs,
experienced before the age of 18 are presented in Table 1.
2.4. Demographic variables
Demographic information, based on Part 2 of the survey is pre-
sented in Table 3.
2.5. Data analysis
2.5.1. Data preparation
All WMH surveys use strict guidelines for cleaning and coding
data and preliminary files were reviewed to ensure data quality.
Imputations were performed on missing values for demographic
variables using hot-deck methods (Kessler & Üstün, 2008). Regres-
sion methods were used if the rates of missing data were high and
cases were deleted if the interview was not completed. There was
no missing data on the 12 CAs or on the covariates used in this
analysis. Statisticians working on the NISHS computed case-specific
weights (cf. Bunting et al., 2013). Part 2 weights, stratification units
and cluster units were utilized in the current study. Analyses were
implemented using SPSS version 21 and Mplus version 7.
44 M. McLafferty et al. / Journal of Anxiety Disorders 35 (2015) 42–48

Table 1 It is important to also consider theoretical perspectives when

Prevalence of childhood adversities in Northern Ireland.
determining the final number of latent classes. Additionally, if
Type of childhood adversity n % classes are too small they may not be interpreted reliably. There-
Parental death 190 10.1 fore, the researcher should consider the interpretation of the classes
Parental divorce 105 5.2 in terms of practical value and conceptual criteria (Wang & Wang,
Other parental loss 58 2.6 2012). Thus model selection was based on a balance of fit indices,
Parental mental illness 159 6.1 parsimony, and substantive meaning.
Parental substance disorder 61 2.2
Parental criminal behaviour 48 1.9
Family violence 124 5.4 2.5.4. Logistic regression
Physical abuse 84 3.7
Logistic regression models were estimated in Mplus 7. Men-
Sexual abuse 61 2.3
Neglect 48 1.9
tal health disorders and suicidality were utilized as the dependent
Physical illness 74 2.8 variables (DVs) and a series of demographic, conflict traumas, and
Economic adversity 190 8.6 childhood adversity profiles were utilized as independent variables
(IVs). The IV’s examined in this study are detailed in Table 4.
Total number of childhood adversities

Experienced any childhood adversity 723 32.0 3. Results

One/anya 456
Two/any 154
Three/any 53
Four/any 35
Five or more/anyb 25
Note: n = raw un-weighted values, % part 2 weights used. Total sample N = 1986.
a
Prevalence estimates in the last five rows represent the proportion of all respon-
dents with any childhood adversity who have exactly one, two, three, four, five or
more.
b participants reporting adversities during childhood. Co-morbidity
Maximum number of childhood adversities experienced =9.
2.5.2. Examining type and prevalence of childhood adversities
The type and frequency of 12 dichotomously assessed CAs expe-
rienced in Northern Ireland were examined. Prevalence estimates
were calculated, using weights, to represent the proportion of all
respondents who experienced any childhood adversity.
2.5.3. Latent Class Analysis (LCA)
LCA categorises underlying homogenous, mutually exclu-
sive classes within a heterogeneous population (Hagenaars &
McCucheon, 2002). It is a person centred approach that identifies
unobserved subpopulations by clustering cases into latent classes
based on how they respond to observed categorical variables (Wang
& Wang, 2012). LCA is an exploratory process in which a series of
models ranging in number of classes are estimated. Models of 2-
5 latent classes, using the 12 CA’s as indicators were estimated,
implementing weight, stratification, and cluster variables, using the
robust maximum likelihood (MLR) estimator. Models were evalu-
ated using a series of model fit indices; the AIC (Akaike Information
Criterion), BIC (Bayesian Information Criterion), and the SSABIC
(sample size adjusted BIC). Lanza, Collins, Lemmon, and Schafer
(2007) recommend the use of the AIC or BIC. However, Nylund,
Bellmore, Nishina, and Graham (2007) propose that the adjusted
BIC should be the fit indice of choice. Other tests include the LRT
(Lo-Mendel-Rubin likelihood ratio test). Lower AIC, BIC and SSABIC
scores are indicative of the optimal model. If the LRT value is sig-
nificant this indicates preference for the model with one less class
(Wang & Wang, 2012). Entropy values which range from 0 to 1
are also examined, with higher values indicating more accurate
classification of latent class membership.
62.5
22.8
7.5 3.1. Examining type and prevalence of childhood adversities
3.8
3.4 Table 1 presents the prevalence of childhood adversities (CAs)
reported in NI, with rates ranging from 1.9% for neglect and parental
criminal behaviour, to 10.1% for parental death. The total number
of CAs experienced in NI are also presented in Table 1, with 32% of
of adversities was reported; however, of those who reported expe-
riencing an adversity, 62.5% suffered only one adversity.
3.2. Latent Class Analysis (LCA)
LCA fit indices are presented in Table 2. Following a thorough
examination of the fit indices and the substantive meaning of the
latent classes, the three class solution was deemed optimal. This
was based on lower AIC, BIC and SSABIC values for the three- com-
pared to the one- or two- class models. Improvements between
the three- and the four-class models on these fit indices were min-
imal, suggesting that the addition of a fourth class added little to
the overall fit of the model. Moreover, the values of the BIC rose
between the three and four- class models. The entropy value for
the three- class model was .936, indicating clear classification.
Fig. 1 shows the latent class profile plot for the three-class
model. Class 3, the largest class, comprised 86% of the sample and
was characterized by low probabilities of experiencing childhood
adversities; this was considered the normative or baseline group
and labelled the low risk class. Class 1 was characterized by a low
risk of experiencing most adversities with the exception of parental
death or divorce but had a very high probability of endorsing child-
hood economic adversity. This class was labelled the economic
adversity class, accounting for 7.8% of the sample. Class 2 accounted
for 6.1% of the sample. This group was characterized by endorsing
the highest probability of experiencing a wide range of adversities
in childhood, particularly those concerning parental maltreatment
and maladjustment and was labelled the poly-adversity class.
Table 3 shows the descriptive statistics for socio-demographics
and the prevalence of mental health, suicidal ideation and

Table 2
Fit indices for Latent Class Models 1–5.

Model Log-likelihood AIC BIC SSBBIC Entropy LRT (p)

1 −4122.050 8268.099 8335.226 8297.101 – –

2 −3799.191 7648.382 7788.229 7708.803 0.860 639.242 (.0154)
3 −3744.034 7564.067 7776.635 7655.907 0.936 109.209 (.0741)
4 −3706.032 7514.065 7799.353 7637.323 0.930 75.240 (.3427)
5 – – – – – –

Note: AIC = Akaike information criterion, BIC = Bayesian information criterion, SSABIC = sample size adjusted BIC, LRT = Lo-Mendell-Rubin adjusted likelihood ratio test. The
log-likelihood for the 5 class model was not replicated. Optimal model is highlighted in bold.
 M. McLafferty et al. / Journal of Anxiety Disorders 35 (2015) 42–48 45

Fig. 1. Latent profile plot of childhood adversity indicators.

behaviour and conflict related trauma variables across the three
classes.
3.3. Logistic regression analyses
3.3.1. Trauma profiles
Table 4 shows that when compared with the low risk class,
individuals from typologies characterized by a greater degree of
adversity during childhood were more likely to have a range of
mental health problems. Odds ratios and 95% confidence intervals
for all the variables of interest are presented in Table 4. In con-
trast to the low risk class, the economic adversity class were more
likely to suffer from an anxiety disorder (OR = 1.974, p < .01) or a
substance disorder (OR = 1.768, p < .05). When compared with the
low risk class, the poly-adversity class which endorsed high levels
of a wide range of childhood adversities were more likely to exhibit
suicidal ideation or behaviour (OR = 4.650, p < .001). Members of
this class also had the highest risk of suffering from an anxiety dis-
order (OR = 5.250, p < .001), any mood disorder (OR = 3.216, p < .001)
or any substance disorder (OR = 4.759, p < .001). The experience of
conflict traumas related to the Troubles in Northern Ireland was sig-
nificantly associated with a range of mental health problems and
suicidality, particularly anxiety disorders (OR = 3.122, p < .001).
3.3.2. Demographic associations
Males were significantly less likely to develop anxiety or
mood disorders but more likely to develop substance disorders
(OR = 3.972, p < .001). Those who were previously married were
significantly more likely to develop a range of mental health prob-
lems in comparison to people who were married/co-habiting, while

Table 3
Descriptive statistics for socio-demographics and prevalence of mental health, suicide ideation and behaviour and conflict related trauma variables across classes 1, 2, and 3.

Total sample Class 1 Class 2 Class 3

N = 1986 Econ adv Poly-adv Low Risk
n = 166 (%) n = 146 (%) n = 1674 (%)

Gender
Female 1164 (52.2%) 93 (45.9%) 93 (50.3%) 978 (52.9%)
Male 822 (47.8%) 73 (54.1%) 53 (49.7%) 696 (47.1%)

Age
Mean (SD) 45.4(17.6) 45.4(17.8) 40.8(13.1) 45.8(17.9)

Marital status
Married/Cohabitating 1174 (59.8%) 97 (56.8%) 77 (62.1%) 1000 (59.9%)
Previously married 342 (13.2%) 31 (15.3%) 31 (15.2%) 280 (12.9%)
Never Married 470 (27.0%) 38 (27.9%) 38 (22.7%) 394 (27.2%)

Years in education 12.5 (2.1) 12.4 (2.2) 12.6 (1.9) 12.5 (2.1)
Mean (SD)

Any anxiety disorder

No 1370 (77.0%) 99 (65.7%) 42 (38.0%) 1229 (80.8%)
Yes 616 (23.0%) 67 (34.3%) 104 (62.0%) 445 (19.2%)

Any mood disorder

No 1440 (79.7%) 106 (73.6%) 62 (51.7%) 1272 (82.3%)
Yes 546 (20.3%) 60 (26.4%) 84 (48.3%) 402 (17.7%)

Any substance disorder

No 1691 (85.9%) 133 (78.5%) 92 (58.6%) 1466 (88.5%)
Yes 295 (14.1%) 33 (21.5%) 54 (41.4%) 208 (11.5%)

Any suicidal ideation/behaviour

No 1731(90.7%) 140 (88.6%) 94 (66.4%) 1497 (92.6%)
Yes 255 (9.3%) 26 (11.4%) 52 (33.6%) 177 (7.4%)

Conflict trauma
No 1169 (61.0%) 80 (44.5%) 45 (27.8%) 1044 (64.9%)
Yes 817 (39.0%) 86 (55.5%) 101 (72.2%) 630 (35.1%)

Note: Class 1 = Economic adversity; Class 2 = Poly adversity; Class 3 = Baseline; OR = odds ratio; CI = confidence interval; Any Anxiety = any DSM Anxiety disorder; Any
Mood = any DSM Mood disorder; Any Substance = any DSM Substance disorder; any suicidal ideation and behaviour (gestures, plans, and attempts).
46 M. McLafferty et al. / Journal of Anxiety Disorders 35 (2015) 42–48

Table 4
Logistic regression analyses of socio-demographic, conflict trauma and latent class associations with mental health outcomes.

Predictor Any anxiety Any mood Any substance Any suicidal ideation/behaviour
OR OR OR OR
(95% CI) (95% CI) (95% CI) (95% CI)

Gender
Female 1.00 1.00 1.00 1.00
Male 0.389*** 0.502*** 3.972*** 0.721
(0.282–0.538) (0.382–0.661) (2.861–5.515) (0.492–1.056)

Age 0.989* 0.980*** 0.970*** 0.986*

(continuous) (0.979–0.999) (0.973–0.987) (0.958–0.983) (0.974–0.998)

Marital status
Married/cohabit 1.00 1.00 1.00 1.00
Previously married 1.895** 1.997*** 2.278** 1.979**
(1.288–2.788) (1.419–2.811) (1.354–3.833) (1.284–3.051)
Never married 1.185 0.880 1.655* 1.018
(0.876–1.601) (0.637–1.218) (1.061–2.582) (0.658–1.574)

Education 0.973 0.998 0.975 0.979

(continuous) (0.900–1.052) (0.930–1.072) (0.896–1.061) (0.899–1.067)

Conflict trauma 3.122*** 2.215*** 2.054*** 2.265***

(2.448–3.982) (1.626–3.017) (1.424–2.962) (1.731–2.966)

Class 1 Economic adversity 1.974** 1.486 1.768* 1.381

(1.275–3.057) (0.948–2.329) (1.050–2.977) (0.722–2.643)
Class 2 Poly-adversity 5.250*** 3.216*** 4.759*** 4.650***
(2.802–9.834) (2.121–4.875) (2.418–9.365) (2.659–8.131)
Class 3 Baseline 1.00 1.00 1.00 1.00

Note: Class 1 = Economic adversity; Class 2 = Poly-adversity; Class 3 = Baseline; OR = odds ratio; CI = confidence interval; Any Anxiety = Any DSM Anxiety disorder; Any
Mood = Any DSM Mood disorder; Any Substance = Any DSM Substance disorder; any suicidal ideation and behaviour (gestures, plans, and attempts).
*
Significant difference compared to base category <.05.
**
Significant difference compared to base category <.01.
***
Significant difference compared to base category <.001.

those who never married were more likely to have substance dis-
orders (OR = 1.655, p < .05). Age was significantly associated with
having a range of mental health problems, with older respon-
dents displaying a decreased likelihood of having anxiety disorders
(OR = 0.989, p < .010), mood disorders (OR = 0.980, p < .001), sub-
stance disorders (OR = 0.970, p < .0001) and suicidal ideation and
behaviour (OR = 0.986, p < .01).
4. Discussion
The current study examined the prevalence of childhood adver-
sities (CAs) in NI and used a person centred approach to uncover
patterns of adversity during childhood. In addition to a low
risk class, a class which had a high probability of experiencing
economic adversity and a poly-adversity class were revealed. Sub-
sequently, associations between these adversity typologies and
socio-demographic factors and DSM mental health and suicidal
outcomes were assessed. In accordance with previous reports
(Hazen et al., 2009; Ford et al., 2013) individuals with childhood
adversity profiles were found to have elevated levels of psy-
chopathology.
Previous studies have reported high rates of CAs globally, how-
ever, this study found that the prevalence of CAs reported in NI were
generally lower than rates found in other countries (Kessler et al.,
2010), with the exception of economic adversity and parental men-
tal illness. While NI is classed as a high income country it has many
areas of deprivation which may partially account for these results.
In comparison with other WMH surveys (Kessler et al., 2010), peo-
ple in NI were less likely to report CAs associated with parental
maladjustment and maltreatment. For example, reported levels of
neglect, at 1.9%, were much lower than the rates found in other
countries. It should be noted nonetheless that nearly a third of par-
ticipants experienced adversities during childhood which is a great
cause for concern.
Other studies have reported that CAs do not happen in isola-
tion, with co-occurrence common (Armour et al., 2014). Using LCA
to examine the co-occurrence of CAs and identify patterns of adver-
sity, three childhood adversity typologies were revealed in the NI
population. Class 3, the largest group representing 86% of the pop-
ulation, had a low probability of experiencing CAs. Class 1, also had
a low probability of experiencing many adversities in childhood,
however, economic adversity was very high. This group also had
elevated levels of parental loss through death or divorce suggesting
that these are linked to economic adversity. Class 2, the poly-
adversity class, had a high probability of endorsing a wide range
of childhood adversities, particularly those involving parental mal-
treatment and maladjustment.
The findings are consistent with previous studies which found
that many CAs co-occur especially those termed maladaptive
family functioning (Green et al., 2010; McLaughlin et al., 2010).
However, the identification of a class endorsing high rates of eco-
nomic adversity indicates the magnitude of childhood economic
adversity in NI, which may be related to the respondents growing
up during the Troubles (Gallagher, Hamber, & Joy, 2012). According
to recent reports persistent child poverty remains high in compari-
son with other areas in the UK (Kelly et al., 2012). This could suggest
that the Troubles and the deprivation that went with it are still
impacting indirectly on children in NI today.
With the low risk group acting as the reference group, LR mod-
els indicated that individuals who experienced adversities during
childhood were more likely to have a range of mental health prob-
lems. A significant association was found between the class which
endorsed high levels of economic adversity and anxiety and sub-
stance disorders. This is in accordance with previous findings which
suggest that poverty increases sensitivity to stress, leaving a person
more susceptible to the development of mental health disorders
(McLaughlin et al., 2011). However, members of the poly-adversity
class displayed the greatest risk of suffering from an anxiety dis-
order and also had an elevated risk of a developing a DSM mood
 M. McLafferty et al. / Journal of Anxiety Disorders 35 (2015) 42–48
disorder, substance disorder or suicidal ideation and behaviour.
This is consistent with previous studies that reported that child
abuse and family violence had the greatest impact on mental health
(DeVenter et al., 2013).
Cross national variations have been found in other WMH sur-
veys examining the impact of CAs on psychological wellbeing. Some
studies reported that no specific adversity in childhood predicted
anxiety disorders but that people who experienced a number of
CAs were at a heightened risk of developing the disorder (Oladeji
et al., 2010). This study corroborates these findings, with those
who experienced multiple adversities being five times more likely
to have an anxiety disorder. In accordance with other studies
which reported significant associations between CAs and suicidal
behaviour (Bruffaerts et al., 2010), this study also found that those
who endured a range of adversities in childhood, the poly-adversity
class, were over four and a half times more likely to display suicidal
ideation and behaviour. In addition, members of this group were
also more than four and a half times more likely to have a DSM
substance disorder, and more than three times more likely to have
a DSM mood disorder, corroborating the findings of other WMH
surveys (Fujiwara & Kawalkami, 2011; Slopen et al., 2010).
Additionally, significant associations were found between the
demographic variables and psychopathology. In accordance with
previous studies, females were more likely to develop mood and
anxiety disorders and males were more likely to have substance dis-
orders (Eaton et al., 2012). Younger respondents were much more
likely to have mental health problems and suicidal ideation and
behaviour. In addition, people who experienced conflict related
traumas were significantly more likely to have negative mental
health outcomes as per previous findings (Priebe et al., 2010).
4.1. Limitations
One of the main limitations of the study was that the sample may
not be fully representative of the population as a number of people
were omitted, including those with learning disabilities, immi-
grants, homeless people, and those in institutions. As these groups
often have higher levels of mental health problems, their omis-
sion may impact on the findings. However this omission is common
practice in WMH surveys and this study followed the standardized
format of the WMH-CIDI. Another limitation of the study could be
the use of retrospective studies which cannot determine causality
and can be problematic due to recall bias. The rates of some child-
hood adversities were lower than would be expected, particularly
those related to abuse, which could suggest under reporting in the
Northern Ireland population. This may be due to recall bias or a
reluctance to share personal information. Additionally, in line with
previous WMH papers, harsh physical punishment was excluded
from the current study (Kessler et al., 2010). Further research exam-
ining the impact of this adversity would be beneficial. In addition,
the findings in the current study may not be generalizable to other
samples.
4.2. Implications and future research
While the study has some limitations, it nevertheless adds to
previous literature on CAs. The study provides valuable epidemi-
ological information on a range of adversities and the detrimental
impact they have on psychopathology in NI. It informs policy mak-
ers and practice, highlighting the need to target children at risk of
multiple adversities. The study emphasizes the importance of sta-
ble intra-family relationships and provides support for initiatives to
improve parenting skills. In addition it shows that when planning
interventions it is important to consider co-occurrence of adversi-
ties. While McLaughlin et al. (2011) found that economic adversity
in childhood predicted the onset of a range of mental health
47
problems it did not significantly predict the persistence or sever-
ity of disorders. Given the elevated rates reported in NI, further
research examining the long term impact of economic adversity
on psychopathology may be particularly beneficial. Additionally,
as it has been reported that the Troubles have had a very detrimen-
tal impact on mental health it would be useful to explore patterns
of adversities in childhood in conjunction with experiences of the
Troubles in the NI population.
4.3. Conclusions
The prevalence and impact of economic adversity on mental
health was found to be particularly high in NI. Previous studies
based in NI also reported that increased rates of psychopathology
were associated with social deprivation (McConnell, Bebbington,
McClelland, Gillespie, & Houghton, 2002). Associations between
mental disorders and economic adversity need to be considered in
terms of social and economic policies. Changes in the welfare bene-
fits for people with mental disorders will have a significant impact
on this population in NI and may impact further upon their men-
tal health. Overall the study found that individuals from typologies
characterized by a greater degree of adversity during childhood
were more likely to report negative mental health and suicidal
outcomes, compared to typologies reporting fewer adversities. In
particular individuals who experienced a range of CAs were more
than 5 times more likely to have an anxiety disorder. The findings
highlight the importance of considering the impact of co-occurring
CAs when planning treatment, prevention, and intervention pro-
grammes.
Role of funding source
The survey was carried out in conjunction with the World
Health Organization WMH Survey Initiative, which is supported by
the National Institute of Mental Health (NIMH; R01 MH070884),
the John D. and Catherine T. MacArthur Foundation, the Pfizer
Foundation, the US Public Health Service (R13-MH066849, R01-
MH069864, and R01 DA016558), the Fogarty International Center
(FIRCA R03-TW006481), the Pan American Health Organization, Eli
Lilly and Company, Ortho-McNeil Pharmaceutical, GlaxoSmithK-
line, and Bristol-Myers Squibb. The NISHS was supported by a grant
from the Research and Development (R&D) Division in NI. The
authors have support from the Commission for Victims and Sur-
vivors in Northern Ireland (CVSNI) for the submitted work; they
have no relationships with the R&D Division NI or CVSNI that might
have an interest in the submitted work. None of the funders had any
role in the design, analysis, interpretation of results, or preparation
of this paper.
Conflict of interest
None.
Contributors
Margaret McLafferty contributed to the development of the
theoretical rationale of the study, conducted literature searches,
carried out analyses, interpreted the results and wrote the first
draft of this manuscript. Dr Cherie Armour supervised the study,
contributed to the development of the theoretical rationale, con-
ducted analyses and assisted in the editing of the paper. Dr Aine
McKenna contributed to the development of the theoretical ratio-
nale of the study, conducted analyses and assisted in the editing of
the final paper. Professor Siobhan O’Neil, Dr Sam Murphy and Pro-
fessor Brendan Bunting coordinated the NISHS and were involved in
48 M. McLafferty et al. / Journal of Anxiety Disorders 35 (2015) 42–48
the design, supervision and editing of the final paper. All the afore-
mentioned authors contributed to the paper and approved the final
manuscript.
Acknowledgements
We would like to thank the staff of the WMH Data Collection
and Data Analysis Coordination Centres for assistance with instru-
mentation, fieldwork, and consultation on data analysis.
References
Armour, C., Elklit, A., & Christoffersen, M. N. (2014). A latent class analysis of
childhood maltreatment: identifying abuse typologies. Journal of Loss and
Trauma: International Perspectives on Stress and Coping, 19(10), 23–29. http://
dx.doi.org/10.1080/15325024.2012.734205
Benjet, C., Borges, G., & Medina-Mora, M. E. (2010). Chronic childhood adversity
and onset of psychopathology during three life stages: Childhood, adolescence
and adulthood. Journal of Psychiatric Research, 44(11), 732–740.
Bunting, B. P., Ferry, F. R., Murphy, S. D., O’Neill, S. M., & Bolton, D. (2013). Trauma
associated with civil conflict and posttraumatic stress disorder: Evidence from
the Northern Ireland Study of Health and Stress. Journal of Traumatic Stress, 26,
134–141.
Bruffaerts, R., Demyttenaere, K., Borges, G., Haro, J. M., Chiu, W. T., Hwang, I., et al.
(2010). Childhood adversities as risk factors for onset and persistence of
suicidal behaviour. The British Journal of Psychiatry, 197, 20–27. http://dx.doi.
org/10.1192/bjp.bp.109.074716
Carlson, E. B., & Dalenberg, C. J. (2000). A conceptual framework for the impact of
traumatic experiences. Trauma, Violence & Abuse, 1, 4–28.
Copeland, W. E., Keeler, G., Angold, A., & Costello, E. J. (2007). Traumatic events and
posttraumatic stress in childhood. Archives of General Psychiatry, 64, 577–584.
http://dx.doi.org/10.1001/archpsyc.64.5.577
DeVenter, M., Demyttenaere, K., & Bruffaerts, R. (2013). The relationship between
adverse childhood experiences and mental health in adulthood. A systematic
literature review. Tijdschrift Voor Psychiatrie, 55(4), 259–268.
Dong, M., Anda, R. F., Felitti, V. J., Dube, S. R., Williamson, D. F., Thompson, T. J., Loo,
C. M., & Giles, W. H. (2004). The interrelatedness of multiple forms of childhood
abuse, neglect, and household dysfunction. Child Abuse & Neglect, 28, 771–784.
Dunn, V. J., Abbott, R. A., Croudace, T. J., Wilkinson, P., Jones, P. B., Herbert, J., et al.
(2011). Profiles of family-focused adverse experiences through childhood and
early adolescence: The ROOTS project a community investigation of adolescent
mental health. BMC Psychiatry, 11:109, 1–16. http://www.biomedcentral.com/
1471-244X/11/109
Eaton, N. R., Keyes, K. M., Krueger, R. F., Balsis, S., Skodol, A. E., Markon, K. E., et al.
(2012). An Invariant Dimensional Liability Model of gender differences in
mental disorder prevalence: Evidence from a national sample. Journal of
Abnormal Psychology, 121(1), 282–288.
Ford, J. D., Grasso, D. J., Hawke, J., & Chapman, J. F. (2013). Poly-victimization
among juvenile justice-involved youths. Child Abuse and Neglect, 37, 788–800.
Fryers, T., & Brugha, T. (2013). Childhood determinants of adult psychiatric
disorder. Clinical Practice & Epidemiology in Mental Health, 9, 1–50.
Fujiwara, T., & Kawalkami, N. (2011). Association of childhood adversities with the
first onset of mental disorders in Japan: Results from the World Mental Health
Japan, 2002–2004. Journal of Psychiatric Research, 45, 481–487. http://dx.doi.
org/10.1016/j.jpsychires.2010.08. 002
Gallagher, E., Hamber, B., & Joy, E. (2012). Perspectives and possibilities: Mental
health in post-agreement Northern Ireland. Shared Space: A Research Journal on
Peace, Conflict and Community Relations in Northern Ireland, 13, 63–78.
Green, J. G., McLaughlin, K. A., Berglund, P. A., Gruber, M. J., Sampson, N. A.,
Zaslavsky, A. M., et al. (2010). Childhood adversities and adult psychiatric
View publication stats
disorders in the national comorbidity survey replication 1: Associations with
first onset of DSM-IV disorders. Archives of General Psychiatry, 67(2), 113–123.
Hagenaars, J. A., & McCucheon, A. L. (Eds.). (2002). Applied latent class analysis.
Cambridge: Cambridge University Press.
Hazen, A. L., Connelly, C. D., Roesch, S. C., Hough, R. L., & Landsverk, J. A. (2009).
Child maltreatment profiles and adjustment problems in high-risk
adolescents. Journal of Interpersonal Violence, 24, 361–378.
Kelly, G., Tomlinson, M., Daly, M., Hillyard, P., Nandy, S., & Patsios, D. (2012). The
necessities of life in Northern Ireland. Poverty and Social exclusion in the UK,.
http://www.poverty.ac.uk/sites/default/files/attachments/
WP%20Analysis%20No.1%20-%20Kelly%20et%20al%20May2012 0.pdf
Kessler, R. C., McLaughlin, K. A., Green, J. G., Gruber, M. J., Sampson, N. A., Zaslavsky,
A. M., et al. (2010). Childhood adversities and adult psychopathology in the
WHO World Mental health Surveys. British Journal of Psychiatry, 197,
378–385.
Kessler, R. C., & Üstün, T. B. (2004). The World Mental Health (WMH) Survey
Initiative Version of the World Health Organisation (WHO) Composite
International Diagnostic Interview (CICI). International Journal of Methods in
Psychiatric Research, 13, 93–121.
Kessler, R. C., & Üstün, T. B. (2008). The WHO world mental health surveys: global
perspectives on the epidemiology of mental disorders. New York: Cambridge
University Press.
Lanza, S. T., Collins, L. M., Lemmon, D. R., & Schafer, J. L. (2007). PROC LCA: A SAS
procedure for latent class analysis. Structural Equation Modelling, 14(4),
671–694.
McConnell, P., Bebbington, P., McClelland, R., Gillespie, K., & Houghton, S. (2002).
Prevalence of psychiatric disorder and the need for psychiatric care in
Northern Ireland. British Journal of Psychiatry, 181, 214–219.
McLaughlin, K. A., Breslau, J., Green, J. G., Lakoma, M. D., Sampson, N. A., Zaslavsky,
A. M., et al. (2011). Childhood socio-economic status and the onset, persistence,
and severity of DSM-IV mental disorders in a US national sample. Social Science
Medicine, 73, 1088–1096. http://www.ncbi.nlm.nih.gov/pubmed/21820781
McLaughlin, K. A., Green, J. G., Gruber, M. J., Sampson, N. A., Zaslavsky, A. M., &
Kessler, R. C. (2010). Childhood adversities and adult psychopathology in the
National Comorbidity Survey Replication (NCS-R) II: Associations with
persistence of DSM-IV disorders. Archives of General Psychiatry, 67(2),
124–132.
Nylund, K., Bellmore, A., Nishina, A., & Graham, S. (2007). Subtypes, severity, and
structural stability of peer victimization: What does latent class analysis say?
Child Development, 78, 1706–1722.
Oladeji, B. D., Makanjuola, V. A., & Gureje, O. (2010). Family-related adverse
childhood experiences as risk factors for psychiatric disorders in Nigeria. British
Journal of Psychiatry, 186–191. http://dx.doi.org/10.1192/bjp.bp.109.063677
Pirkola, S., Isometsa, E., Aro, H., Jestuka, L., Hamalainen, J., Veijola, J., et al. (2005).
Childhood adversities as risk factors for adult mental disorders. Results from
the Health 2000 study. Social Psychiatry & Psychiatric Epidemiology, 40,
769–777. http://dx.doi.org/10.1007/s00127-005-0905-x
Priebe, S., Bogic, M., Ajdokovic, D., Farnciskovic, T., Galeazzzi, G. M., Kucukalic, A.,
et al. (2010). Mental disorders following war in the Balkans: A study in 5
countries. Archives of General Psychiatry, 67(5), 518–528.
Rosenman, S., & Rodgers, B. (2004). Childhood adversity in an Australian
population. Social Psychiatry and Psychiatric Epidemiology, 39(9), 695–702.
Shevlin, M., & Elklit, A. (2008). A latent class analysis of adolescent adverse life
events based on a Danish national youth probability sample. Nordic Journal of
Psychology, 62, 218–224.
Slopen, N., Williams, D. R., Seedat, S., Moomal, H., Herman, A., & Stein, D. J. (2010).
Adversities in childhood and adult psychopathology in the South Africa Stress
and Health Study: Associations with first-onset DSM-IV disorders. Social
Science & Medicine, 71(10), 1847–1854.
Wang, J., & Wang, X. (2012). Structural equation modelling: Applications using Mplus.
Oxford: Wiley Blackwell.
Weich, S., Patterson, J., Shaw, R., & Stewart-Brown, S. (2009). Family relationships
in childhood and common psychiatric disorders in later life: Systematic review
of prospective studies. The British Journal of Psychiatry, 194, 392–398.