Journal of Affective Disorders 190 (2016) 156161

Contents lists available at ScienceDirect

Journal of Affective Disorders

journal homepage: www.elsevier.com/locate/jad

Research report

Personality and social support as predictors of rst and recurrent

episodes of depression
Annemieke Noteboom a,n, Aartjan T.F. Beekman a,b, Nicole Vogelzangs b,
Brenda W.J.H. Penninx a,b
a
b

GGZinGeest, Amsterdam, The Netherlands

Department of Psychiatry, EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, The Netherlands

art ic l e i nf o

a b s t r a c t

Article history:
Received 7 May 2015
Received in revised form
5 September 2015
Accepted 10 September 2015
Available online 23 October 2015

Background: Depression is a prevalent psychiatric disorder with high personal and public health consequences, partly due to a high risk of recurrence. This longitudinal study examines personality traits,
structural and subjective social support dimensions as predictors of rst and recurrent episodes of depression in initially non-depressed subjects.
Methods: Data were obtained from the Netherlands Study of Depression and Anxiety (NESDA). 1085
respondents without a current depression or anxiety diagnosis were included. 437 respondents had a
prior history of depression, 648 did not. Personality dimensions were measured with the NEO-FFI, network
size, partner-status, negative and positive emotional support were measured with the Close Person
Questionnaire. Logistic regression analyses (unadjusted and adjusted for clinical variables and sociodemographic variables) examined whether these psychosocial variables predict a new episode of depression at two year follow up and whether this differed among persons with or without a history of
depression.
Results: In the unadjusted analyses high extraversion (OR:.93, 95% CI (.91.96), P o.001), agreeableness
(OR:.94, 95% CI (.90.97), Po .001), conscientiousness (OR:.93, 95% CI (.90.96), P o.001) and a larger
network size (OR:.76, 95% CI (.64.90), P .001) signicantly reduced the risk of a new episode of depression. Only neuroticism predicted a new episode of depression in both the unadjusted (OR:1.13, 95% CI
(1.101.15), P o.001) and adjusted analyses (OR:1.06, 95% CI (1.031.10), Po .001). None of the predictors
predicted rst or recurrent episodes of depression differently.
Limitations: we used a relatively short follow up period and broad personality dimensions.
Conclusions: Neuroticism seems to predict both rst and recurrent episodes of depression and may be
suitable for screening for preventive interventions.
& 2015 Elsevier B.V. All rights reserved.

Keywords:
Depression
Personality
Social support

1. Introduction
Depression is one of the most prevalent psychiatric disorders.
In a World Mental Health survey initiative in 18 countries Bromet
et al. (2011) found that the average lifetime prevalence of depression ranged from 11.1% to 14.6% and the twelve month prevalence ranged from 5.5% to 5.9%. Depression is highly recurrent:
at least 50% of those who recover from a rst episode of depression have one or more episodes in their lifetime and approximately 80% of those with a history of two episodes have another
recurrence (American Psychiatric Association, 2000). Depression
has major personal and public health consequences (Judd et al.,
n
Correspondence to: GGZinGeest, A.J. Ernststraat 1187, 1081 HL Amsterdam, The
Netherlands.
E-mail address: a.noteboom@ggzingeest.nl (A. Noteboom).

http://dx.doi.org/10.1016/j.jad.2015.09.020
0165-0327/& 2015 Elsevier B.V. All rights reserved.

2000; Greden, 2001). It is therefore important to identify the

factors that predict onset and recurrence so at-risk individuals can
be identied at an early stage and prevention steps can be taken.
Much is yet to be learned about factors that inuence the
course of depression, but there is growing evidence that different
risk factors are associated with onset and recurrence (Lewinsohn
et al., 1999). Onset of depression seems to be associated with demographic variables like gender (with a twofold increased risk for
women), marital status, age (Weissman et al., 1996; Wittchen
et al., 2000; Andrade et al., 2003; Van de Velde et al., 2010), the
presence of other psychiatric disorders (especially anxiety disorders), family history of depression and stressful life events
(Birmaher et al., 2004). Subclinical residual symptoms and the
number of previous episodes are the most important predictors of
recurrence (Burcusa and Iacono, 2007; Hardeveld et al., 2010).
Personality and social support dimensions have also been
identied as relevant risk factors for depression. The Five Factor

A. Noteboom et al. / Journal of Affective Disorders 190 (2016) 156161

Model is among the most popular models of personality (McCrae

and Costa, 1990). This model includes ve important domains of
personality: neuroticism, extraversion, agreeableness, conscientiousness and openness to experience and can be seen as the
basic organizing framework for both normal and abnormal personality (Markon et al., 2005). Most research on personality and
depression has focused on the relationship between depression
and neuroticism or extraversion. Many studies have documented
the relationship between neuroticism and onset of depression
(Kendler et al., 2006; Fanous et al., 2007, Bagdy et al., 2008;
Grifth et al., 2010). Neuroticism also seems to be related to recurrence of depression (Burcusa and Lacono, 2007; Hill et al.,
2011), but evidence is still inconclusive (Hardeveld et al., 2010).
Prospective studies have found no evidence that extraversion can
predict future depressive episodes (Kendler et al., 2006). The association of the other personality traits with onset or recurrence of
depression has largely been unexamined.
Social support can be dened as a person's perceived belief that
help or empathy can be obtained when needed and the satisfaction with this available support (Sarason et al., 1987). The subjective dimension of social support-perceived support-has been
more consistently linked to depression than structural aspects of
support like network size (Finch et al., 1999; Haber et al., 2007). A
low level of perceived social support seems clearly linked with
depression in different patient groups (Stice et al., 2004; Travis
et al., 2004). However, the exact nature of this relationship is quite
unclear. There is evidence that greater social support protects
against the onset of depression (Kendler et al., 2005); although
other studies have failed to nd this effect (Burton et al., 2004;
Wade and Kendler, 2000). Several studies reported no relationship
between social support and recurrence of depression (Hardeveld
et al., 2010). A less investigated aspect of social support, the negative aspects of interactions, seems to be a risk factor for onset of
depression (Finch et al., 1999; Lincoln, 2008). To our knowledge
there are no studies examining the relationship between negative
aspects of interactions and the prediction of recurrence.
Personality plays an important role in the ability to develop and
maintain interpersonal relationships and in both the appraisal and
effectiveness of supportive interactions that take place in the
context of these relationships (Kendler et al., 2003; Lincoln, 2008).
Therefore this study aims to assess the predictive value of the
personality dimensions of the ve factor model and different social support dimensions for depression, both uniquely and in
concert. We aim to examine whether personality and social support are differently associated with the prediction of onset or recurrence of depression. Based on the literature, we hypothesize
that high neuroticism, low perceived support and a high level of
negative aspects of support will predict rst and potentially also
recurrent episodes of depression. Since there is some evidence
that social support may erode as a consequence of depression
(Coyne, 1976) and given the high comorbidity between depression
and anxiety we will statistically control the analyses for depression
severity at baseline and prior anxiety disorders.

2. Methods
2.1. Sample
Data for the present study were obtained from the Netherlands
Study of Depression and Anxiety (NESDA), an ongoing naturalistic
longitudinal cohort study examining the long-term course of depressive and anxiety disorders. NESDA has been designed to be
representative of those with depressive and anxiety disorders in
different health care settings and stages of the developmental
history. A total of 2981 respondents were recruited from different

157

health care settings (community, primary care and specialised

mental health care), including healthy controls and those with a
history or current diagnosis of a depressive and/or anxiety disorder. All respondents underwent a four-hour baseline assessment
that included an assessment of psychopathology, demographic
and personal characteristics, psychosocial functioning and biomarkers. Respondents were 18 through 65 years of age. The study
protocol was approved centrally by the Ethics Review Board of the
VU University Medical Centre Amsterdam and subsequently by the
local review boards of each participating centre. After full verbal
and written information about the study was provided, written
informed consent was obtained from all respondents. A detailed
description of the NESDA study has been published elsewhere
(Penninx et al., 2008). The present study drew on data from the
baseline and two-year follow-up assessments. We only included
respondents without a current diagnosis (dened as having no
depressive or anxiety disorder diagnosis in the prior six months)
at baseline (N 1244). Subsequently, we excluded respondents
with missing information on social network (N 11), personality
(N 1) or on one of the social support dimensions (N 57). Finally,
we excluded respondents without a two-year follow up measurement (N 90), leaving a sample of 1085 respondents. 648 respondents had no prior history of depression. 437 respondents did
have a history of depression. Excluded persons (n 159) did not
differ on sex, age or education compared to included persons.

3. Measurements
3.1. Depressive disorder diagnosis
Diagnoses of depressive disorders (dysthymia and major depressive disorder) were assessed at baseline and two-year follow
up and were dened according to the DSM-IV criteria. Depressive
disorder diagnoses were established by specially trained clinical
research staff with the Composite International Diagnostic Interview (CIDI, WHO version 2.1. Dutch version, ter Smitten et al.,
1997).The CIDI is a structured diagnostic interview with a high
inter-rater reliability (Wittchen et al., 1991), high testretest reliability (Wacker et al., 2006) and high validity for depressive and
anxiety disorders (Wittchen, 1994). The primary outcome variable
was the occurrence of an episode of depression between baseline
and the two-year follow-up. Based on the baseline CIDI interview
persons were categorized as those with or without a prior history
of depression.
3.2. Personality dimensions
We used the Dutch 60-item self-report NEO ve-factor inventory (NEO-FFI) to measure the ve personality dimensions,
neuroticism, extraversion, openness to experience, agreeableness
and conscientiousness, of the Five Factor Model (Hoekstra et al.,
1996). The NEO-FFI is a short version of the Revised NEO Personality Inventory (NEOPI-R, Costa and McCrae, 1995). The reliability,
internal structure and construct validity of the NEO-FFI are satisfactory (Hoekstra et al., 1996). The Cronbach's alpha for the
different subscales was good (N:.95; E:.87; O:.78; A:.87; C:.83). The
dimension sum score of each subscale was used for the statistical
analyses.
3.3. Social support
We used the Close Person Questionnaire (CPQ, Stansfeld and
Marmot, 1992) as a measure for two types of perceived support:
conding/emotional support and negative aspects of support. The
CPQ has been positively validated using the Self Evaluation and

158

A. Noteboom et al. / Journal of Affective Disorders 190 (2016) 156161

Social Support Interview (Stansfeld and Marmot, 1992; Stansfeld

et al., 1998). In the present study we asked respondents how much
support they received from their partner (if present) and up to two
friends (over 18 years old, no housemates) with whom they have
important contact on a regular basis. Answers were rated on a
5-point scale ranging from never to very often. The conding/
emotional support subscale includes items about positive interaction and the possibility of conding in, or sharing problems
with, a partner or up to two condents. The negative aspects
subscale includes items about the stress or worry resulting from
the interaction with a partner or up to two condents, and dissatisfaction with the support available. From our total study population 793 respondents completed the questions about partner
support. 867 respondents had a rst condent and 682 had a
second condent. If a respondent didnot have a rst condent
questions about the second condent were skipped. Like Kendler
et al. (2005) we used the mean responses for the categories that
were available when respondents lacked a partner or a condent.
Cronbach's alpha for the subscales were: conding/emotional
support (.76) and negative aspects of support (.80).
3.4. Network size and partner status
Network size can be described as the number of people with
whom a person has regular contact and who provide support.
Network size was assessed by asking respondents with how many
relatives, friends or others over the age of 18 years they had regular and important contact. Answers were rated on a six point
scale ranging from 0 or 1 contact to more than 20 contacts. All
respondents were asked whether or not they had a husband or
wife or somebody they considered a partner.
3.5. Covariates
Sociodemographic variables included age, gender and years of
education. In addition, basic clinical characteristics were taken into
account including having a lifetime history of an anxiety disorder
(generalized anxiety disorder, social phobia, panic disorder with or
without agoraphobia and agoraphobia assessed with the CIDI interview) and depressive symptom severity. The severity of depressive symptoms was measured with the Inventory of Depressive Symptomatology (IDS), which is closely correlated with observer-rated scales and has satisfactory psychometric properties
(Rush et al., 1996). The scores ranged from 0 to 45.
3.6. Statistical analyses
First, we compared the group of respondents without a history
of depression with the group of respondents with a history of
depression in terms of socio-demographic and clinical characteristics, personality and social support variables at baseline. T-tests
for independent samples were used for continuous variables and
chi-square statistics for categorical variables. Next, univariable
logistic regression analyses (with the occurrence of a (new) episode of depression yes or no at follow up as the dependent
variable) were performed on the whole study sample to examine
the predictive value of each of our predictors. We repeated these
analyses for our main predictors (personality and social support)
while correcting them for demographic (age, gender, years of
education) and clinical (severity of depressive symptoms and
having a lifetime diagnosis of anxiety) characteristics. Next, a
multivariable model with all predictors included was tested.
Finally, to study whether our predictors were differently associated with rst or recurrent episodes of depression a yes/no prior
history*predictor interaction term was included in the logistic
regression analyses, separately for each predictor. In all analyses a

two-tailed Po .05 was considered statistically signicant The statistical software used was SPSS 18.0.

4. Results
4.1. General characteristics at baseline
Our total study sample (N 1085) contained 704 women
(64.9%), and the mean age was 42.4 years (SD 13.9). Most of our
respondents were married or had somebody they considered a
partner (N 804, 74.1%). 299 respondents reported a lifetime diagnosis of an anxiety disorder (27.6%) and 437 reported a lifetime
depression diagnosis (40.3%). Table 1 lists the baseline characteristics for the group of respondents with and without a prior history of depression at baseline. The group with a prior history
contained more women (P .001) and was signicantly older
(P .005). As expected, this group also reported more severe depressive symptoms (Po .001) and had more frequently a history of
anxiety disorders (P o.001). Persons with a depression history also
were more neurotic (P o.001), more introvert (P o.001), more
open to new experiences (P .002), less agreeable (P .015) and
less conscientious (P o.001). No difference was found between the
groups in the percentage of respondents with a partner (P .19).
Persons with a prior history of depression did have a smaller
network size (P o.001), reported less perceived emotional support
(P .043) and more negative aspects of support (P .001).
4.2. Predicting a (new) episode of depression at two year follow up
We used logistic regression analysis in the whole sample to
examine which predictors could signicantly predict a (new)
episode of depression at follow up (Table 2). In the univariable
uncontrolled analyses high neuroticism (OR: 1.13, 95% CI (1.10
1.15), P o.001) signicantly predicted a new episode of depression
Table 1
Baseline characteristics of the group with and the group without a lifetime depression diagnosis.

Demographics
Age (M,SD)
Gender female (N,%)
Years education (M,SD)
Clinical Characteristics
Severity of depressive
symptoms (M,SD)
Presence of lifetime anxiety diagnosis Yes (N,%)
Personality
Neuroticism (M,SD)
Extraversion (M,SD)
Openness (M,SD)
Agreeableness (M,SD)
Conscientiousness (M,SD)
Social support
Having a partner Yes (N,%)
Network size (N,%)
01
25
6 or more
Percieved emotional support (M,SD)
Percieved negative support (M,SD)

P-value

Group without
lifetime diagnosis
of depression
(N 648)

Group with lifetime diagnosis of

depression
(N 437)

41.4 (14.7)
396 (61.1)
12.9 (3.2)

43.8 (12.6)
308 (70.5)
12.6 (3.2)

8.5 (7.1)

14.3 (8.9)

o .001

96 (14.8)

203 (46.5)

o .001

21.1
26.8
27.3
32.6
31.6

o .001
o .001
.002
.015
o .001

15.4
29.9
26.2
33.3
33.2

(7.4)
(6.2)
(5.7)
(4.7)
(5.0)

487 (75.2)
9
198
441
15.6

(1.4)
(30.6)
(60.0)
(3.5)

7.9 (2.8)

P-value o.05 is considered statistically signicant.

(7.5)
(6.4)
(5.7)
(4.8)
(5.1)

317 (72.5)
19
172
246
15.2

(4.3)
(39.4)
(56.3)
(3.7)

8.5 (3.1)

.005
.001
.07

.19
o .001

.043
.001

A. Noteboom et al. / Journal of Affective Disorders 190 (2016) 156161

159

Table 2
Logistic regression analyses predicting a new episode of depression at 2 year follow up.
N 1085

Demographics
Gender
Age
Education
Clinical Characteristics
Symptom severity
Lifetime anxiety diagnosis
Personality
Neuroticism
Extraversion
Openness
Agreeableness
Conscientiousness
Social Support
Having a partner
Network size
Percieved emotional support
Negative aspects of support

Univariable (a)
Odds ratio (95% CI), P-value

Univariable (b)
Odds ratio (95% CI), P-value

Multivariable (c)
Odds ratio (95% CI), P-value

1.21 (.831.75), .33

.99 (.981.00), .07
.95 (.901.00), .05
1.10 (1.081.12), o .001
2.67 (1.873.82), o .001
1.13
.93
1.03
.94
.93

(1.101.15), o .001
(.91.96), o .001
(1.001.06), .09
(.90.97), o.001
(.90.96), o .001

.75
.76
.96
1.03

(.521.10), .14
(.64.90), .001
(.921.00), .06
(.971.09), .35

1.06
.98
1.02
.98
.99
.90
.90
.99
.97

(1.031.10), o .001
(.951.01), .20
(.991.06), .17
(.941.02), .36
(.951.02), .45

1.07 (1.031.11), .001

1.00 (.971.04), .84
1.03(.991.07), .13
.99 (.951.03), .60
1.00 (.961.04), .88

(.591.37), .62
(.761.07), .24
(.951.04), .79
(.911.03), .31

.92
1.02
1.01
.95

(.771.10), .34
(.641.63), .92
(.961.07), .61
(.881.02), .16

P-value o .05 is considered statistically signicant.

(a) Uncorrected.
(b) Corrected for age, gender,years of education, severity of depressive symptoms, presence of lifetime anxiety diagnosis (Y/N).
(c) Full model including all predictors.

at follow up. A high level of extraversion (OR: .93, 95% CI (.91.96),

P o.001), agreeableness (OR: .94, 95% CI (.90.97), Po .001), conscientiousness (OR: .93, 95% CI (.90.96), P o.001) and a larger
network size (OR: .76, 95% CI (.64.90), P .001) signicantly reduced the risk of a (new) episode of depression at follow up. In the
adjusted univariable analyses, adjusted for age, gender, years of
education and the clinical characteristics, only neuroticism predicted a (new) episode of depression (OR: 1.06, 95% CI (1.031.10),
P o.001). Except for neuroticism, the clinical characteristics
(symptom severity and having a lifetime anxiety disorder) in the
adjusted univariable analyses outweighed the predictive value of
the signicant predictors in the unadjusted analyses. The same
result was found in the multivariable analysis including all predictors in the model. Again, only high neuroticism (OR: 1.07, 95% CI
(1.031.11), P .001) predicted the development of a new episode
of depression.
4.3. Predicting rst episode depression or recurrence
Finally, we examined whether the predictive value of our predictors differed for rst or recurrent episodes of depression by
adding a yes/no prior history*predictor interaction term in the
logistic regression analyses separately for each predictor. In the
univariable adjusted analyses in the whole sample, which were
adjusted for the demographic and clinical characteristics, only
conscientiousness predicted a new episode of depression signicant (P .02). We repeated the logistic regression analysis for
conscientiousness in the group without a history of depression.
We found no signicant effect (OR: 0.95, 95% CI (0.891.01), Pvalue .09). We did the same for the group with a history of depression (with recurrence yes or no as a dependent variable)
and again found no signicant effect (OR: 1.01, 95% CI (0.971.06),
P-value .62). Based on these results we can conclude that none of
the personality nor the social support dimensions predict rst or
recurrent episodes of depression differently.

5. Discussion
In this study we examined the relationship between the personality dimensions of the Five Factor Model, several social

support dimensions and the development of a new episode of

depression after two year follow up in subjects with or without a
history of depression. Our study has three major ndings. First,
personality and social support dimensions are not differently related to rst or recurrent episodes of depression. Second, only
neuroticism and none of the other dimensions of the ve factor
model predicted new episode of depression consistently after already considering clinical factors. We found some indication that
high extraversion, agreeableness and conscientiousness may reduce the likelihood of a new episode of depression, but these effects disappeared after adjustment for clinical characteristics
(symptom severity and having a lifetime diagnosis of an anxiety
disorder). Indicating that these clinical characteristics outweigh
the personality dimensions as predictors. Third, contrary to our
expectations, we found no evidence that low perceived support or
any of the other social support dimensions, neither the structural
(network size or having a partner) nor the perceived aspects
(perceived emotional support or negative aspects of support) are
associated with the development of a new episode of depression
after accounting for baseline clinical characteristics.
How can we understand these ndings? Our baseline analyses
are largely in line with earlier described ndings (Kotov et al.,
2010; Klein et al., 2011). Compared to a group without a lifetime
diagnosis of depression we found that the group with a lifetime
diagnosis reported a higher level of neuroticism and lower levels
of extraversion and conscientiousness. Contrary to their ndings in
our study we also found that the group with a lifetime history of
depression reported a lower level of agreeableness. Also our
baseline ndings concerning social support are largely in line with
earlier research (Finch et al., 1999; Lincoln, 2008; Oddone et al.,
2011). The group with a lifetime history of depression reported
less emotional support, more negative aspects of support and a
smaller social network.
Longitudinal data examining the predictive value of personality
and social support dimensions as risk factors for depression are
sparse. Neuroticism is most consistently described as a risk factor
for depression and some researchers even suggest that it might be
the core of internalizing psychopathology (Grifth et al., 2010),
although it's role in recurrence is still inconclusive (Hardeveld
et al., 2010). Results are inconsistent concerning social support
dimensions as risk factors for depression. Consistent with our

160

A. Noteboom et al. / Journal of Affective Disorders 190 (2016) 156161

ndings Monroe and Steiner (1986) describe how the magnitude

of the association between support and depressive symptoms diminishes and even disappears when controlling for the confounding inuence of the disorder itself. Clinical characteristics
seem to be stronger predictors. There are some recent longitudinal
studies that did nd evidence that lower levels of perceived support predict depression even after accounting for baseline depression (Nasser and Overholser, 2005; Leskela et al., 2006), but in
their review Lakey and Cronin (2008) conclude that there is a
relative lack of evidence that low perceived support prospectively
predicts depression onset. In our study neuroticism was the only
predictor that predicted rst and recurrent episodes of depression
even beyond the effect of clinical characteristics. In the unadjusted
analyses a larger social network and a higher level of agreeableness, conscientiousness and extraversion reduced the chance of a
new episode of depression. These protective effects disappeared in
the adjusted analyses. One could hypothesize that those with a
more extraverted, conscientious and agreeable personality style
are more likely to have a large network and resist impulses
to withdraw from relationships when feeling sad due to a highly
felt responsibility in relationships. This tendency towards prosocial behaviour may lead to more social support and behavioural
activation through social activities (Cukrowicz et al., 2008) and
reduce the chance of developing a new episode of depression.
Possibly the protective effects of these personality traits are
related to the severity of the depressed symptoms. When the
depressive symptoms are too severe they possibly outweigh one's
personality style. Further research is needed to examine this
dynamic interplay between severity of depression, personality and
social interactions.
There are several major models that aim to explain the association between personality and depression (Bagdy et al., 2008).
The vulnerability model postulates that personality scores can
predict who develops the condition among previously unaffected
individuals. In contrast, the scar model argues that psychopathology changes personality, whereas the complication model
posits that personality changes are due to the illness and temporary in nature. Our results are in line with the vulnerability
model: also in the group of respondents without a history of
psychopathology a high level of neuroticism predicted the development of a rst episode of depression at two year follow up. The
stress generation hypothesis (Hammen, 1991) offers another explanation for the association between personality and depression.
There is a substantial amount of support for this hypothesis, which
claims that depression-prone individuals are not simply passive
respondents to stressful events in their lives but active agents in
the creation of depressogenic life stressors especially in the interpersonal domain (Lui and Alloy, 2010). This can be seen as a
source of interpersonal stress that might increase the risk of recurrence of depression (Wingate and Joiner, 2004). Neuroticism
plays an important role in the ability to develop and maintain
interpersonal relationships and in both the appraisal and effectiveness of supportive interactions that take place in the context of
these relationships (Kendler et al., 2003; Lincoln, 2008). So one
could hypothesize that these patients are, due to their more
neurotic personality, more sensitive to, or more prone to elicit,
negative aspects in their relationships and less capable of eliciting
the level of emotional support they need. At baseline we did nd
that the group with a prior history of depression were more
neurotic and reported a smaller network size and less emotional
support, but these social support factors had no predictive value in
the longitudinal data. So other explanations have to be considered.
Perhaps depression has a great impact on the ability to invest in
social relationships and reduced social support should be viewed
as a consequence of depression instead of as a risk factor, consistent with the scar hypothesis (Bagdy et al., 2008). Otherwise, a

smaller network size and less perceived support could be manifestations of an underlying vulnerability that also predisposes to
depression. More research is necessary to examine the possible
complex dynamic interplay between depression, social support
and neuroticism.
The results from this study have important clinical implications. Neuroticism can be regarded as a risk factor for both rst
and recurrent episodes of depression. In the diagnostic phase of
treatment of patients with a history of depression or even just
subclinical depressive symptoms the level of neuroticism should
be assessed. Patients with a high level of neuroticism can be regarded as a high-risk group for developing a new episode of depression and might prot from preventive interventions. Metaanalytic evidence indicates that preventive interventions can reduce the incidence of depressive disorders by 25% (Cuijpers et al.,
2008). Cuijpers et al. (2010) also advise to develop interventions
focusing on neuroticism itself instead of its specic negative
outcomes (like depression).
There are also several limitations in our study. We examined a
relatively short follow-up period of two years and possibly as a
result the absolute number of rst episodes and recurrences in the
follow-up period per group was relatively low. On the other hand,
Gopinath et al. (2007) had a shorter follow up period and found a
slightly higher recurrence rate in primary care (31%). Our study
group was selected from both primary care and community which
might explain the difference. A longer follow-up period could
probably provide more information. It is possible that more specic associations between personality traits and depression were
not evident because we focused on the broad personality dimensions of the Five Factor Model instead of on the facet-level traits
that comprise these broad dimensions. Another limitation is that
we did not include the age of onset of the depressive disorder in the
analysis. There is evidence that early onset depression is more frequently associated with personality disorders compared to late onset
depression (Fava et al., 1996). Our study also has some important
strengths. The longitudinal design allowed for an examination of
the predictive value of our main predictors. The study population
is large, including respondents who develop a rst episode of
depression and respondents with recurrent depression. We only
included respondents without a current diagnosis (dened as having
no depressive or anxiety disorder diagnosis in the prior six months)
and controlled for symptom severity at baseline to reduce the risk of
earlier described confounding of the personality measurements by the
presence of depressive symptoms (Fava et al., 1994,, 2002). Also, we
included all dimensions of the ve factor model.
In summary, our data indicate that neuroticism is possibly an
important predictor of both rst and recurrent episodes of depression. Neither structural or perceived social support dimensions nor the other personality dimensions of the ve factor model
predicted depression at follow up. Neuroticism can be assessed
relatively easily and efciently and thus is suitable for screening
for individuals who can benet from preventive interventions.

Contributors
Annemieke Noteboom formulated the hypothesis of this study,
analyzed and interpreted the data and wrote the report.
Brenda Penninx is a principal investigator of the NESDA study.
She supervised the rst author in the analysis and interpretation
of the data and in writing the report .
Nicole Vogelzangs supervised the rst author in SPSS usage and
participated in writing the report.
Aartjan Beekman is a principal investigator of the NESDA study.
He also supervised the rst author in the analysis and interpretation of the data and writing the report.

A. Noteboom et al. / Journal of Affective Disorders 190 (2016) 156161

Role of funding source

The infrastructure for the NESDA study (www.nesda.nl) is funded through the
Geestkracht program of the Netherlands Organisation for Health Research and Development (ZON-MW, Grant number 10-000-1002) and is supported by participating
universities and metal health organizations (VU University Medical Center, GGZ InGeest, Arkin, Leiden University Medical Center, GGZ Rivierduinen, University Medical
Center Groningen, Lentis, GGZ Friesland, GGZ Drenthe, Scientic Institute for Quality of
Healthcare (IQ Healthcare), Netherlands Institute for Health Services Research (NIVEL)
and Netherlands Institute of Mental Health and Addiction (Trimbos)).

Acknowledgements
We thank all mental health care organizations who supported the NESDA study
for their assistance in the data collection and all patients fort heir participation in
this study.

References
American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental
Disorders Text Revision-Fourth. American Psychiatric Association, Washington DC.
Andrade, L., Caraveo-Anduaga, J.J., Berglund, P., Bijl, R.V., Dragomericka, E., Kohn, R.,
Keller, M.B., Kessler, R.C., Kawakami, N., Kilic, C.O.D., Ustun, T.B., Vicente, B.,
Wittchen, H., 2003. The epidemiology of major depressive episodes: results from the
International Consortium of Psychiatric Epidemiology (ICPE) Surveys. Int. J. Methods
Psychiatr. Res. 21 (1), 321.
Bagdy, R.M., Quilty, L.C., Ryder, A.C., 2008. Personality and depression. La. Rev. De. Psychiatr. 53 (1), 1425.
Birmaher, B., Williamson, D.A., Dahl, R.E., Axelson, D.A., Kaufman, J., Dorn, L.D., Ryan, N.
D., 2004. Clinical presentation and course of depression in youth: does onset in
childhood differ from onset in adolescence? J. Am. Acad. Child. Adolesc. Psychiatry
43 (1), 6370.
Bromet, E., Andrade, H.L., Hwang, I., Sampson, N.A., Alonso, J., de Girolamo, G., de Graaf,
R., Demyttenaere, K., Hu, C., Iwata, N., Karam, N., Kaur, J., Kostyuchenko, S., Lepine, L.
P., Levinson, D., Matschinger, H., Mora, M.E.M., Oakley Brown, M., Posada-Villa, J.,
Viana, M.C., Williams, D.R., Kessler, R.C., 2011. Cross-National epidemiology of DSMIV depressive episode. BioMedCentral Med. 9, 90.
Burcusa, S.L., Lacono, W.G., 2007. Risk for recurrence in depression. Clin. Psychol. Rev. 27
(8), 959985.
Burton, E., Stice, E., Seeley, J.R., 2004. A prospective test of the stress-buffering model of
depression in adolescent girls: no support once again. J. Consult. Clin. Psychol. 72,
689697.
Costa, P.T., McCrae, R.R., 1995. Domains and facets: hierarchical personality assessment
using the revised NEO personality inventory. J. Pers. Assessment 64, 2150.
Coyne, J.C., 1976. Depression and the response of others. J. Abnorm. Psychol. 85, 186193.
Cuijpers, P., van Straten, A., Smit, F., Mihalopoulos, C., Beekman, A., 2008. Preventing the
onset of depressive disorders: a meta-analytic review of psychological interventions.
Am. J. Psychiatry 165, 12721280.
Cuijpers, P., Smit, F., Penninx, B.W.J.H., de Graaf, R., ten Have, M., Beekman, A.T.F., 2010.
Economic Costs of neuroticism: a population-based study. Arch. Gen. Psychiatry 67,
10861093.
Cukrowicz, K.C., Franzese, A.T., Thorp, S.R., Cheavens, J.S., Lynch, T.R., 2008. Personality
traits and perceived support among depressed older adults. Aging Ment. Health 12
(5), 662669.
Fanous, A.H., Neale, M.C., Aggen, S.H., Kendler, K.S., 2007. A longitudinal study of personality and major depression in a population-based sample of male twins. Psychol.
Med. 37, 11631172.
Fava, M., Boufdes, E., Pava, J.A., McCarthy, M.K., Steingard, R.J., Rosenbaum, J.F., 1994.
Personality disorder comorbidity with major depression and response to uoxetine
treatment. Psychother. Psychosom. 62, 160167.
Fava, M., Alpert, J.E., Borus, J.S., Andrew, B.A., Nierenberg, A.A., Pava, J.A., Rosenbaum, J.F.,
1996. Patterns of personality disorder comorbidity in early-onset versus late-onset
major depression. Am. J. Psychiatry 153, 13081312.
Fava, M., Farabaugh, A.H., Sickinger, A.H., Wright, E., Alpert, S., Sonawalla, S., Nierenberg,
A.A., Worthington, J.J., 2002. Personality disorders and depression. Psychol. Med. 32
(6), 10491057.
Finch, J.F., Okun, M.A., Pool, G.J., Ruehlman, L.S., 1999. A comparison of the inuence of
conictual and supportive social interactions on psychological distress. J. Pers. 67 (4),
581621.
Gopinath, S., Katon, W.J., Russo, J.E., Ludman, E.J., 2007. Clinical factors associated with
relapse in primary care patients with chronic or recurrent depression. J. Affect.
Disord. 101, 5763.
Greden, F.J., 2001. The burden of recurrent depression: causes, consequences and future
prospects. J. Clin. Psychiatry 62 (22), 59.
Grifth, J.W., Zinbarg, R.E., Craske, M.G., Mineka, S., Rose, R.D., Waters, A.M., Sutton, J.M.,
2010. Neuroticism as a common dimension in the internalizing disorders. Psychol.
Med. 40 (7), 11251136.
Haber, M.G., Cohen, E.J.L., Lucas, E.T., Baltes, B.B., 2007. The relationship between selfreported received and perceived social support: a meta-analytic review. Am. J.
community Psychol. 39, 133144.
Hammen, C., 1991. Generation of stress in the course of unipolar depression. J. Abnorm.
Psychol. 100, 555561.
Hardeveld, F., Spijker, J., de Graaf, R., Nolen, W.A., Beekman, A.T.F., 2010. Prevalence and
predictors of recurrence of major depressive disorder in the adult population. Acta
Psychiatr. Scand. 122, 184191.
Hill, J., Holcombe, C., Clark, L., Boothby, M.R., Hincks, A., Fisher, J., Tufail, S., Salmon, P.,
2011. Predictors of onset of depression and anxiety in the year after diagnosis of

161

breast cancer. Psychol. Med. 41 (7), 14291436.

Hoekstra, H.A., Ormel, J., de Fruyt, F., 1996. Handleiding NEO persoonlijkheidsvragenlijsten NEO-PI-R en NEO-FFI. Swets Test Services, Lisse.
Judd, L.L., Akiskal, H.S., Zeller, P.J., Paulus, M., Leon, A.C., Maser, J.D., Endicott, J., Coryell,
W., Kunovac, J.L., Mueller, T.I., Rice, J.P., Keller, M.B., 2000. Psychosocial disability
during the long-term course of unipolar major depressive disorder. Arch. Gen. Psychiatry 57 (4), 375380.
Kendler, K.S., Gardner, C.O., Prescott, C.A., 2003. Personality and the experience of environmental adversity. Psychol. Med. 7, 11931202.
Kendler, K.S., Myers, J., Prescott, C.A., 2005. Sex differences in the relationship between
social support and risk for major depression: a longitudinal study of opposite-sex
twin pairs. Am. J. Psychiatry 162, 250256.
Kendler, K.S., Gatz, M., Gardner, C.O., Pedersen, N.L., 2006. Personality and major depression: a Swedish longitudinal, population-based twin study. Arch. Gen. Psychiatry
63, 11131120.
Klein, D.N., Kotov, R., Bufferds, S.J., 2011. Personality and depression: explanatory models
and review of the evidence. Annu. Rev. Clin. Psychol. 7, 269295.
Kotov, R., Gamez, W., Schmidt, F., Watson, D., 2010. Linking big personality traits to
anxiety, depressive and substance abuse disorders: a meta analysis. Psychol. Bull.
136, 768821.
Lewinsohn, P.M., Allen, N.B., Seeley, J.R., Gotlib, I.H., 1999. First onset versus recurrence of
depression: differential processes of psychosocial risk. J. Abnorm. Psychol. 108 (3),
483489.
Lincoln, K.D., 2008. Personality, negative interactions and mental health. Soc. Serv. Rev.,
223252.
Lui, R.T., Alloy, L.B., 2010. Stress generation in depression: a systematic review of the
empirical literature and recommendations for future study. Clin. Psychol. Rev. 30 (5),
582593.
Markon, K.E., Krueger, R.F., Watson, D., 2005. Delineating the structure of normal and
abnormal personality: an integrative hierarchical approach. J. Pers. Soc. Psychol. 88
(1), 139157.
McCrae, R.R., Costa, P.T., 1990. Personality in Adulthood: A Big Five Theory Perspective.
Guilford, New York.
Monroe, S.M., Steiner, S.C., 1986. Social support and psychopathology: interrelations with
preexisting disorder, stress and personality. J. Abn. Psychol. 95, 2539.
Nasser, E., Overholser, J.C., 2005. Recovery from major depression: the role of support
from family, friends and spiritual beliefs. Act. Psych. Scan. 111, 25132.
Oddone, C.G., Hybels, C.F., McQuoid, D.R., Steffens, D.C., 2011. Social support modies the
relationship between personality and depressive symptoms in older adults. Am. J.
Geriatr. Psychiatry 19 (2), 123130.
Penninx, B.W., Beekman, A.T., Smit, J.H., Zitman, F.G., Nolen, W.A., Spinhoven, P., Cuipers,
P., de Jong, P.J., van Marwijk, H.W.J., Assendelft, W.J.J., van der Meer, K., Verhaak, P.,
Wensing, M., de Graaf, R., Hoogendijk, W.J., Ormel, J., van Dyck, R., 2008. The
Netherlands Study of Depression and Anxiety (NESDA): rationale, objectives and
methods. Int. J. Methods Psychiatr. Res. 17, 121140.
Rush, A.J., Gullion, C.M., Basco, M.R., Jarrett, R.B., Trivedi, M.H., 1996. The Inventory of
Depressive Symptomatology (IDS): psychometric properties. Psychol. Med. 26,
477486.
Sarason, I.G., Sarason, B.R., Shearin, E.N., Pierce, G.R., 1987. A brief measure of social
support: practical and theoretical implications. J. Soc. Pers. Relatsh. 4, 497510.
ter Smitten, M.H., Smeenk, R.M.W., van den Brink, W., 1997. Composite International
Diagnostic Interview (CIDI). Basis versie 2.1. Interviewer handleiding. World Health
Organization.
Stansfeld, S., Marmot, M., 1992. Deriving a survey measure of social support: the reliability and validity of the close persons questionnaire. Soc. Sci. Med. 35 (8),
10271035.
Stansfeld, S.A., Bosma, H., Hemingway, H., Marmot, M.G., 1998. Psychosocial and work
characteristics and social support as predictors of SF-36 health functioning: the
Whitehall II study. Psychosom. Med. 60 (3), 247255.
Stice, E., Ragan, J., Randall, P., 2004. Prospective relations between social support and
depression: differential direction of effects for parent and peer support? J. Abnorm.
Psychol. 113, 155159.
Travis, L.A., Lyness, J.M., Shields, C.G., King, D.A., Cox, C., 2004. Social support, depression,
and functional disability in older adult primary-care patients. Am. J. Geriatr. Psychiatry 12 (3), 265271.
Van de Velde, S., Bracke, P., Levecque, K., 2010. Gender differences in depression in 23
European countries. Cross-national variation in the gender gap in depression. Soc.
Sci. Med. 71 (2), 305313.
Wacker, H.R., Battegay, R., Mullejans, R., Schlosser, C., 2006. Using the CIDI-C in the
general population. In: Stefanis, C.N., Rabavilas, A.D., Soldatos, C.R. (Eds.), Psychiatry:
A World Perspective. Elsevier Science Publishers, Amsterdam, pp. 138143.
Wade, T.D., Kendler, K.S., 2000. Absence of interactions between social support and
stressful life events in the prediction of major depression and depressive symptomatology in women. Psychol. Med. 30, 965974.
Weissman, M.M., Bland, R.C., Canino, G.J., Faravelli, C., Greenwald, S., Hwu, H.G., Joyce, P.
R., Karam, E.G., Lee, C.K., Lellouch, J., Lepine, J.P., Newman, S.C., Rubio-Stipec, M.,
Wells, J.E., Wickramaratne, P.J., Wittchen, H., Yeh, E.K., 1996. Cross-National epidemiology of major depression and bipolar disorder. J. Am. Med. Assoc. 276 (4),
293299.
Wingate, L.R., Joiner, T.E., 2004. Depression-related stress generation: a longitudinal
study of black adolescents. Behav. Ther. 35, 247261.
Wittchen, H.U., Kessler, R.C., Pster, H., Lieb, M., 2000. Why do people with anxiety
disorders become depressed? A prospective-longitudinal community study. Acta
Psychiatr. Scand. 102, 1423.
Wittchen, H.U., 1994. Reliability and validity studies of the WHO-Composite International
Diagnostic Interview (CIDI): a critical review. J. Psychiatr. Res. 28, 5784.
Wittchen, H.U., Robins, L.N., Cottler, L.B., Sartorius, N., Burke, J.D., Regier, D., 1991. Crosscultural feasibility, reliability and sources of variance of the Composite International
Diagnostic Interview (CIDI). The multicentre WHO/ADANHA eld trials. Br. J. Psychiatry 159, 645653.