f, E
b
Department of Psychiatry, Hospital Nostra Senyora de Meritxell, Andorra la Vella, Andorra
c
Department of Psychiatry, Clinical Institute o
y, Instit
Prospective study
d toxicity were significantly associated with psychiatric disorder risk.
Cancer is a life-threatening disease, and its psychological im-
regarding risk factors for psychiatric disorders include retro-
spective or cross-sectional designs, sampling bias, only focus-
ing on a limited number of risk factors, lack of assessment by
multivariate statistical methods, or small sample size. More-
over, most of the published studies have used patient-rated
views and/or standardized diagnostic criteria.
1–5
Clinician
standard for detecting psychiatric disorders.
To the best of our knowledge only one study
6
has investi-
gated multivariate risk factors for psychiatric disorders dur-
ing hospitalization for stem cell transplantation (SCT).
Sasaki and colleagues
6
diagnosed a mental disorder according
oses, Spain. Tel.: +34 972 153 102; fax: +34 972 510 411.
EUROPEANJOURNALOFCANCER42 (2006) 514–520
available at www.sciencedirect.com
: w* Corresponding author: Present address: Espronceda 43 B, 17480 Rpact on patients has been an important aspect of clinical
oncology. In most cancer patients with a positive psychiatric
condition depression and/or anxiety are the central symp-
toms.
1–7
Methodological shortcomings in the cancer literature
interviews and standardized diagnostic criteria such as the
‘Diagnostic and Statistical Manual for Mental Disorders, 4th
ed.’ (DSM-IV
8
) or The World Health Organization International
Classification of Disorders have long been held to be the gold
1–5Psychiatric disorder
Risk factors
Our study findings may help to improve identification of the patients most at risk for psy-
chiatric disturbances during hospitalization for stem cell transplantation.

2005 Elsevier Ltd. All rights reserved.
1. Introduction
depression or anxiety scale scores at a level suggestive of a
clinical diagnosis, without using structured clinical inter-Risk factors
women, a past psych
higher regimen-related
SCT Unit, Department of Haematolog
Barcelona, Spain
ARTICLE INFO
Article history:
Received 15 June 2005
Accepted 6 July 2005
Available online 19 January 2006
Keywords:
Cancer
Haematopoietic stem cell
transplantation0959-8049/$ - see front matter
2005 Elsevi
doi:10.1016/j.ejca.2005.07.037
E-mail address: jmprieto@comg.es (J.M. Pf Psychiatry and Psychology, Hospital Clı´nic, University of Barcelona, Barcelona, Spain
ute of Haematology and Oncology, IDIBAPS, Hospital Clı´nic, University of Barcelona,
ABSTRACT
The aim of this study was to determine the risk factors for psychiatric disorder in haema-
tological cancer patients during hospitalization for stem cell transplantation. In this 3-year
prospective study, 220 patients received stem cell transplantation at a single institution.
Structured psychiatric interviews applying standardized diagnostic criteria were performed
at hospital admission and weekly during hospitalization until discharge or death, yielding a
total of 1062 interviews. Psychiatric disorder (any depressive, anxiety, or adjustment disor-
der) prevalence at the time of hospital admission was 21% and psychiatric disorder inci-
dence during post-admission follow-up was 22%. After adjusting for multiple
confounders in multivariate logistic regression analyses, we found that younger age,
iatric history, lower functional status, pain, smoking cessation, andEsteve Cirera
c
, Cristobal Gasto´
c
a
Department of Psychiatry, Institut d’Assiste`ncia Sanitaria, Girona, SpainStem cell transplantation: Risk
psychiatric morbidity
Jesu´ s M. Prieto
a,
*, Jordi Blanch
c
, Jorge Atala
b
journal homepageer Ltd. All rights reserved
rieto).actors for
nric Carreras
d
, Montserrat Rovira
d
,
ww.ejconline.com.

ANto DSM-IV criteria in 16 (41%) of 39 allogeneic SCT patients.
Higher anxiety prior to isolation, unrelated donor, and female
sex predicted the occurrence of psychiatric disorders during
isolation. However, their findings were limited by the small
sample size and by the use of a very heterogeneous sample
of psychiatric disorders for risk factor analysis.
Depression and/or anxiety may have a deleterious effect in
many ways: it may impair quality of life;
9,10
increase symptom
burden
2
and pain intensity;
2,4,5,11
lower compliance with med-
ical treatment;
12
reduce overall survival time;
10
and increase
health care costs
13
and hospital stay.
7
The high prevalence of
depression or anxiety during hospitalization for SCT,
6,7
the
associated complications mentioned above, and the fact that
anxiety and depression tend to be under recognized in oncol-
ogy patients
14
highlight the critical importance of identifying
and treating these disorders in transplant patients.
In this 3-year prospective study carried out during hospi-
talization for SCT, we evaluated psychiatric disorders (depres-
sive, anxiety, and adjustment disorders) based on structured
psychiatric interviews and standardized DSM-IV diagnostic
criteria. Weekly interviews were carried out from hospital
admission until discharge or death. In an earlier report from
our cohort,
7
we reported the general prevalence of DSM-IV
psychiatric disorders and its association with a longer hospi-
tal stay. The purpose of the current paper was to identify risk
factors associated with existing psychiatric disorders at the
time of hospital admission or with new psychiatric disorders
occurring during post-admission follow-up.
2. Patients and methods
2.1. Study population
Patients were consecutively recruited from the SCT Unit, Hos-
pital Clı´nic, Barcelona, between July 21, 1994, and August 8,
1997. Inclusion criteria were haematological malignancy, at
least 16 years of age, patient’s first SCT, and verbal informed
consent. Of 253 patients that received an SCT, 235 met the eli-
gibility criteria. Due to scheduling difficulties, 15 patients
could not be interviewed at the first assessment and were ex-
cluded from the study. All patients who were approached
agreed to be interviewed. Thus, the final study cohort in-
cluded 94% of the eligible population (220/235). There were
no differences in age, sex, haematological diagnosis, or dis-
ease risk status between the 220 patients who participated
in the study and the 15 who were excluded (P > 0.20).
2.2. Study procedures
Detailed information on transplant regimens, graft-versus-
host disease prophylaxis and patient care has been published
elsewhere.
7
Briefly, patients were assessed in a first struc-
tured interview within 48 h of hospital admission (day
9to
day
4, depending on the conditioning regimen), and subse-
quently on a weekly basis from day of transplant (day 0) until
discharge or death (day +7; day +14; day +21 and so on). The
first interview lasted 15–45 min and included sociodemo-
graphic data, assessment of past and current psychiatric
EUROPEANJOURNALOFCstatus with structured interview and DSM-IV criteria, and
the Nottingham Health Profile.
15
In the following weeklyassessments, we administered a brief psychiatric structured
interview with DSM-IV criteria lasting 5–15 min. At hospital
admission a Karnofsky score
16
was obtained from the haema-
tologist. After discharge, using a standardized form, J.M.P ab-
stracted pertinent clinical data required to rate the regimen
toxicity scale.
17
After discharge, using a standardized form,
J.M.P abstracted pertinent clinical data required to rate the
regimen toxicity scale.
17
For each particular patient, rating
of the post-admission risk factors (regimen toxicity, graft-ver-
sus-host-disease, and documented infection) was obtained
from the same in-hospital follow-up period used to rate the
post-admission psychiatric disorder cases. For a patient who
received the last psychiatric assessment at day +14 and was
discharged on day +20, the rating of the post-admission risk
factors was obtained from the period between the start of
the conditioning regimen and day +14.
2.3. Psychiatric assessment
Three interviewers participated in the study: two psychia-
trists (J.M.P. and J.B.) and a 4th year psychiatric resident
(J.A.). Psychiatric information from the patient interviews
was complemented with information from the family and
medical and nursing staff. Psychiatric diagnoses were as-
signed at a diagnosis meeting held every two months, at
which a consensus diagnosis was reached on each patient.
No interrater reliability assessment was carried out.
2.3.1. Current psychiatric status
A complete description of the psychiatric assessment has
been published elsewhere.
7
Briefly, the psychiatric interview
followed a structured format with psychiatric diagnoses being
defined according to DSM-IV criteria. Our aim was to conduct
a relatively short psychiatric interview focusing on depres-
sive, anxiety, and adjustment disorders known to be common
in cancer patients.
1–5
The alterations in some depressive
symptoms such as anorexia, and fatigue as a direct result of
the neoplastic process or cytotoxic treatment present a meth-
odological problem for the diagnosis of depression in cancer
patients.
2,3,9,18
In our study setting, in which intensive condi-
tioning treatment is used, most of the patients present fatigue
and anorexia. The DSM-IV requires a symptom to be counted
toward the diagnosis of depression only if it is thought not to
be due to cancer or its treatment, with a consequent risk for
under diagnosis in the SCT setting. As in our previous report,
7
we used the model of the Sloan-Kettering Cancer Institution
group to diagnose major depression. For research purposes,
this method is the best of the four possible diagnostic models
available, as it maximizes specificity.
18
It ensures the most
homogeneous depressed group possible, with the fewest con-
founding variables, thereby increasing the clinical and statis-
tical significance of the research data.
18
The Sloan-Kettering
method eliminates anorexia and fatigue from the list of nine
major depression criteria, and requires only four (instead of
five) of the remaining seven symptoms for diagnosis.
2.3.2. Psychiatric rates by time of diagnosis and overall
prevalence rates
CER42 (2006) 514–520 515Depending on the time of psychiatric diagnosis, we made a
distinction between admission prevalence and post-admis-