European Psychiatry - Volume 41 - April 2017

99 / 916

25th European Congress of Psychiatry / European Psychiatry 41S (2017) S69–S105 S95
depending on the age of the participants. Data was then analyzed
using multiple linear regressions.
Results The total sample comprised of 1027 participants; 675
persons aged 25–50 years and 352 persons aged 51–65 years.
The sample contained roughly equal number of men (52.8%) and
women (47.2%) The full model explained 59,79% variance and was
highly significant F (18,1008) = 85,76, P = 001. Some factors that par-
ticipants feel like could help them reduce the stress in workplace
and subsequently reduce the burnout are longer holidays, lowering
the administration burden, better work place conditions and lastly
increasing the authority a person has in a given work place.
Conclusion The study has shown an association between work-
stress and burnout and thus in order to prevent burnout with it
related job absence certain precaution steps should be made. The
reoccurring theme that would seem to improve the situation is
decreasing the administrative work that is unrelated to the profes-
sion as well as increasing the powers the employees have in their
position.
Disclosure of interest The authors have not supplied their decla-
ration of competing interest.
O073
Influence of gender in patients
attended in emergency rooms for
suicidal tendencies
B. Mata Saenz ∗, E. Segura Escobar , E. Lopez Lavela ,
L.-M. Asensio Aguerri , L. Nuevo Fernandez , T. Rodriguez Cano ,
L. Beato Fernández
Hospital General Ciudad Real, Psychiatry, Ciudad Real, Spain
∗ Corresponding author.
Introduction The first timewhenpeople attempt suicide first con-
tact is critical. Psychiatrists must decide to hospitalize them or
follow-up in mental health units and the bases of a doctor-patient
relationship are formed.
Objectives An analysis of referrals to psychiatry from the emer-
gency room (ER) was developed. Our objective was to discover if
there was a statistical correlation between gender and other vari-
ables, especially repeated visits and admissions.
Methods Our sample was composed of patients who visited the
ER for suicidal tendencies for 20 months. We carried out an obser-
vational retrospective study. The variables collected were: age,
gender, cause, repeated visit (visit to the ER in the following two
months), previous attempts, previous follow-up, method used, use
of toxic substances during the attempt, intentionality, referral from
the ER, later follow-up and diagnostic impression at the ER.
Results A total of 620 patients were sampled. The relationship
between gender and repeated visit, previous attempts, dysfunc-
tional personality traits, use of substances and later follow-up was
found (Chi2). Although the relationship between admissions and
gender were not statistically significant, influence by gender (over
all in males) can be observed in logistic regression models. As well
as, in patients who visited the ER several times, dysfunctional per-
sonality traits seem to be the most common but gender marks
significant differences between groups.
Conclusions The data obtained is consistent with those reported
in previous studies. To know who the riskier groups are can allow
professionals to plan protocols and unify admission criteria.
Disclosure of interest The authors have not supplied their decla-
ration of competing interest.
O074
Risk of mental disorders and
difficulties or conflict in relationships
in young adults
S. Gorbe˜na , P. Penas∗, E. Calvete , I. Crespo , I. Iraurgi
University of Deusto, Department of Personality–Assessment and
Psychological Treatment, Bilbao, Spain
∗ Corresponding author.
Introduction Higher risk of mental health problems has been
linked with problems in relationships, including the experience
of relational conflict with significant others and peers. Conversely,
positive relations with others have been established as a key factor
of psychological well being.
Objectives We hypothesized that psychological maladjustment
will be related to the number, nature and severity of relational
stressors. Furthermore, there would be a higher likelihood of risk of
mental disorders for those who experience more relational hard-
ships and of greater severity. Positive relations with others will
protect from risk of mental health problems.
Method A total of 4461 university students completed a health
and well-being survey, including the GHQ-12 (centesimal and 3-
point cut-off scores), Ryff psychological well-being scale and a scale
of 25 life stressors. Indexes of number and severity of difficulties
in relationships were calculated with 10 items including romantic
partners, friends, family, and classmates.
Results Correlations were significant. Logistic regression showed
a risk effect for all stressors with OR values above 1.32. Over-
all perceived severity had the highest value (OR = 2.38, 95%
CI = 2.16–2.61) and amongst the 10 stressors, gender related
abuse/violence was also the highest (OR = 1.90, 95% CI = 1.73–2.09).
Positive relations showed a protective effect (OR = 0.60, 95%
CI = 0.56–0.54).
Conclusions Findings can inform health promotion, prevention
and therapeutic interventions so as to improve the quality of
personal relationship and conflict management skills, and to
strengthen well-being associated with positive relations with oth-
ers. Academic institutions committed to student welfare and the
promotion of healthy environments should play a major role in
young adults’ mental health.
Disclosure of interest The authors have not supplied their decla-
ration of competing interest.
O075
Clinical prediction of suicide attempt
in schizophrenia using a machine
learning approach
V. De Luc ∗, A. Bani Fatemi , N. Hettige
CAMH, psychiatry, Toronto, Canada
∗ Corresponding author.
Objective Suicide is a major concern for those afflicted by
schizophrenia. Identifying patients at the highest risk for future
suicide attempts remains a complex problem for psychiatric inter-
vention. Machine learningmodels allow for the integration of many
risk factors in order to build an algorithm that predicts which
patients are likely to attempt suicide. Currently, it is unclear how
to integrate previously identified risk factors into a clinically rel-
evant predictive tool to estimate the probability of a patient with
schizophrenia for attempting suicide.
Methods We conducted a cross-sectional assessment on a sam-
ple of 345 participants diagnosed with schizophrenia spectrum
disorders. Suicide attempters and non-attempters were clearly
identifiedusing the Columbia Suicide Severity Rating Scale (C-SSRS)
and the Beck Suicide Ideation Scale (BSS). We developed two clas-
sification algorithms using a regularized regression and random