European Psychiatry - Volume 41 - April 2017

580 / 916

S576 25th European Congress of Psychiatry / European Psychiatry 41S (2017) S521–S582
hospitalizations with psychosis increased from 5.23% to 14.28%
(P trend < 0.001). Utilization of atrial-cardioversion was lower in
patients with psychosis (0.76%v vs. 5.79%, P < 0.001). In-hospital
mortality was higher in patients with Psychosis (aOR 1.206; 95%CI
1.003–1.449; P < 0.001) and discharge to specialty care was sig-
nificantly higher (aOR 4.173; 95%CI 3.934–4.427; P < 0.001). The
median length of hospitalization (3.13 vs. 2.14 days; P < 0.001) and
median cost of hospitalization (16.457 vs. 13.172; P < 0.001) was
also higher in hospitalizations with psychosis.
Conclusions Our study displayed an increasing proportion of
patients with Psychosis admitted due to AF with higher mortal-
ity and extremely higher morbidity post-AF, and significantly less
utilization of atrial-cardioversion. There is a need to explore rea-
sons behind this disparity to improve post-AF outcomes in this
vulnerable population.
Disclosure of interest The authors have not supplied their decla-
ration of competing interest.
EV0527
Comparison between patients who
did not show up for their first visit
and the ones who did
B. Mata Saenz ∗, L. Nuevo Fernández , L.M. Asensio Aguerri ,
L. Beato Fernández , T. Rodríguez Cano
Hospital General Ciudad Real, Psychiatry, Ciudad Real, Spain
∗ Corresponding author.
Introduction Referrals to psychiatry from primary care has
increased in recent years. This can be the result of the global eco-
nomic situation and represents a problem for specialized care,
because patients can’t usually be correctly attended to. On the other
hand, patients who don’t come to visits make up other important
issues that we must analyze.
Objectives To analyze the differences between patients who did
not come for their first visit and those who did in order to try to
describe variables that could be affecting them.
Methods This is an epidemiological, analytic, prospective study of
patients referred to our department. The following variables were
collected: (1) referral protocol, (2) reason, (3) demographic data,
(4) attendance to appointment, (5) diagnosis impression and (6)
destination of referral. The SPSS 19.0 was used to analyze the data.
Results We studied a total of 1.048 patients for 15months, of
which 20.6% did not come to their first visit. A statistically sig-
nificant relationship between attendance and gender, year of the
appointment, adequate demand or not, previous follow-up and
diagnosis was found (Chi2). However, if a logistic regression was
carried out, only the adequacy of the demand was included in the
model.
Conclusions Coordination with general practitioners is essen-
tial to improve referrals and, most importantly, the attention
to patients. If we can agree on the referral criteria, a better-
personalized assistance can be offered to patients who have more
difficulties in coming (because of characteristics of illness, place of
residence, and other variables).
Disclosure of interest The authors have not supplied their decla-
ration of competing interest.
EV0528
Analysis of geographical distribution
of referrals to psychiatry from
primary care
B. Mata Saenz ∗, V. Mu˜noz Martinez , L. Nuevo Fernández ,
E. Lopez Lavela , L.M. Asensio Aguerri
Hospital General Ciudad Real, Psychiatry, Ciudad Real, Spain
∗ Corresponding author.
Introduction The distribution of the demand fromprimary care in
the mental health units could be a way of facilitating the coordina-
tion and improving the attention to patients. For this reason, in our
unit we have made a repartition of the areas among the different
psychiatrists.
Objectives To analyze if there was a correlation between the geo-
graphical origin of the patients or their primary care areas and the
referrals, and between them and their attendance.
Methods This is an epidemiological, analytic, prospective study
of patients referred to our department. The following variables
were collected: (1) referral protocol, (2) reason, (3) demographic
data (origin, gender, age), (4) Primary Care area, (5) attendance
to appointment, (6) diagnosis impression and (7) destination of
referral. The SPSS 19.0 was used to analyze the data.
Results A total of 1048 patients were sampled. A statistically
significant relationship hasn’t been found between place of resi-
dence, primary care area or areas of distribution in the Unit and
attendance (Chi 2). If we analyze the population of each distribu-
tion, we can describe similar percentages depending on the size of
these.
Conclusions Although a different distribution and a relationship
is thought between some areas and the attendance or the number
of referrals, we didn’t find out them in our sample.
Disclosure of interest The authors have not supplied their decla-
ration of competing interest.
EV0529
Identification of major depressive
disorder among the long-term
unemployed
K. Nurmela ∗, A. Mattila , V. Heikkinen , J. Uitti , A. Ylinen ,
P. Virtanen
1 University of Tampere, School of Social Sciences and Humanities,
University of Tampere, Finland
2 Tampere University Hospital, Department of Adult Psychiatry,
Tampere, Finland
3 Tampere University Hospital, Department of Neurosciences and
Rehabilitation, Tampere, Finland
4 University of Tampere, School of Medicine, University of Tampere,
Finland
5 University of Helsinki, Department of Neurological Sciences,
Helsinki, Finland
∗ Corresponding author.
Introduction Depression is a common disorder among the unem-
ployed, but research on identification of their depression in health
care (HC) is scarce.
Objectives The present study aimed to find out if the duration of
unemployment correlates to the risk for unidentifiedmajor depres-
sive disorder (MDD) in HC.
Methods Sample of the study consisted of long-termunemployed
who were in screening project diagnosed as having MDD (n = 243).
The diagnosis was found in the records of HC in 101 (42%) and not
found in 142 (58%) individuals. Binary logistic regression models
were used to explore the effect of the duration of unemployment
to the identification of MDD in HC.
Results The odds ratio (OR) for non-identified MDD in HC was
1.060 (95%CI 1.011–1.111, P = 0.016) per unemployment year and
when unemployment had continued, for example, five years the
OR for unidentified MDD was 1.336. The association remained sig-
nificant throughout adjustments for the set of background factors
(gender, age, occupational status, marital status, homelessness,
self-reported criminal records, suicide attempts, number of HC-
visits).
Conclusions This study among depressed long-termunemployed
indicates that the longer the unemployment period has lasted, the
greater the risk for non-identification of MDD is. HC services should