Identifying Risk for Attrition during Treatment for Depression

Article (PDF Available)inPsychotherapy and Psychosomatics 78(6):372-9 · October 2009with33 Reads
DOI: 10.1159/000235977 · Source: PubMed
Abstract
Understanding patients' ambivalence about treatment persistence may be useful in tailoring retention interventions for individual patients with major depressive disorder. Participants (n = 265) with major depressive disorder were enrolled into an 8-week trial with a selective serotonin reuptake inhibitor. At baseline and week 2, the participants were asked about their intent to return for the next visit, complete the study and continue in the study should they experience side effects or no improvement. Dropouts were defined as participants who discontinued attending clinic visits before completing the trial. Participants who at baseline reported an uncertain/negative intent to continue if they experienced side effects or no improvement dropped out at a significantly higher rate by weeks 6 and 8. Uncertain/negative intent at week 2 predicted attrition at all following visits. Dropouts without side effects were more likely to have reported an uncertain/negative intent to attend at both baseline and week 2, while dropouts who experienced side effects were more likely to have reported an uncertain/negative intent to attend only at baseline. Positive intent to continue was associated with greater symptom improvement in both dropouts and completers despite the possibility of lack of efficacy. Participants' pretreatment concerns about continuing antidepressant treatment in the presence of side effects signals challenges to the completion of a full 8-week acute phase treatment, even if the participant does not develop side effects. Individualized review of concerns and tailoring appropriate interventions may be necessary to reduce attrition.
Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Regular Article
Psychother Psychosom 2009;78:372–379
DOI: 10.1159/000235977
Identifying Risk for Attrition during
Treatment for Depression
Diane Warden
a
Madhukar H. Trivedi
a
Stephen R. Wisniewski
c
Ira M. Lesser
d
Jeff Mitchell
e
G.K. Balasubramani
c
Maurizio Fava
f
Kathy Shores-Wilson
a
Diane Stegman
a
A. John Rush
a, b, g
Departments of
a
Psychiatry and
b
Clinical Sciences, University of Texas Southwestern Medical Center at Dallas,
Dallas, Tex. ,
c
Epidemiology Data Center, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pa. ,
d
Department of Psychiatry and Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center,
Torrance, Calif. ,
e
Laureate Psychiatric Clinic and Hospital, Tulsa, Okla. , and
f
Clinical Psychopharmacology Unit,
Massachusetts General Hospital, Boston, Mass. , USA;
g
Duke-National University of Singapore, Singapore
have reported an uncertain/negative intent to attend only at
baseline. Positive intent to continue was associated with
greater symptom improvement in both dropouts and com-
pleters despite the possibility of lack of efficacy. Conclu-
sions: Participants’ pretreatment concerns about continu-
ing antidepressant treatment in the presence of side effects
signals challenges to the completion of a full 8-week acute
phase treatment, even if the participant does not develop
side effects. Individualized review of concerns and tailoring
appropriate interventions may be necessary to reduce attri-
tion. Copyright © 2009 S. Karger AG, Basel
Introduction
Attrition, or leaving treatment prematurely, is a sub-
stantial problem in the treatment of major depressive dis-
order (MDD) [1] . The attrition rates in the first 12 weeks
o
f treatment can be as high as 65% in naturalistic settings
[24] and 36% in clinical trials [5] , and as many as 15% of
t
he patients never begin a prescribed antidepressant [6] .
T
hese high attrition rates are disturbing given that pa-
tients who leave treatment are less likely to reach remis-
sion [7–9] , have poorer functioning [10]
and are more
Key Words
Attrition Adherence Depression Antidepressant
Attitudes
Abstract
Background: Understanding patients’ ambivalence about
treatment persistence may be useful in tailoring retention
interventions for individual patients with major depressive
disorder. Methods: Participants (n = 265) with major depres-
sive disorder were enrolled into an 8-week trial with a selec-
tive serotonin reuptake inhibitor. At baseline and week 2, the
participants were asked about their intent to return for the
next visit, complete the study and continue in the study
should they experience side effects or no improvement.
Dropouts were defined as participants who discontinued at-
tending clinic visits before completing the trial. Results: Par-
ticipants who at baseline reported an uncertain/negative
intent to continue if they experienced side effects or no im-
provement dropped out at a significantly higher rate by
weeks 6 and 8. Uncertain/negative intent at week 2 predict-
ed attrition at all following visits. Dropouts without side ef-
fects were more likely to have reported an uncertain/nega-
tive intent to attend at both baseline and week 2, while
dropouts who experienced side effects were more likely to
Received: November 2, 2008
Accepted after revision: January 27, 2009
Published online: September 8, 2009
Diane Warden, PhD, MBA
Department of Psychiatry, University of Texas Southwestern Medical Center at Dallas
5323 Harry Hines Blvd., Dallas, TX 75390-9119 (USA)
Tel. +1 214 648 4614, Fax +1 214 648 0168
E-Mail Diane.Warden@UTSouthwestern.edu
© 2009 S. Karger AG, Basel
Accessible online at:
www.karger.com/pps
Identifying Risk for Attrition
Psychother Psychosom 2009;78:372–379
373
likely to relapse [11] . Clearly, it would be beneficial to de-
velop interventions to reduce attrition. The recent focus
on personalized medicine [12] highlights the importance
o
f asking not what is best for the hypothetical typical pa-
tient, but rather what is best for each individual [13] so
t
hat interventions may be tailored to meet the needs of
each patient. Unfortunately, there is currently a dearth of
strategies for identifying specific patients at risk for attri-
tion early in antidepressant treatment based on poten-
tially modifiable factors.
Specific beliefs and attitudes about illness and medica-
tions – such as concerns about overuse, harmfulness or
side effects have been associated with a lack of adher-
ence to dosing recommendations or persistence in treat-
ment [14 –17] . Attitudes about treatment are specific to
i
ndividuals and may also be modifiable during treat-
ment, making attitudes a potentially useful data source
for the identification of patients at risk for attrition.
Explicitly asking patients about their intent to contin-
ue in treatment may be the most direct method for ob-
taining data on patient attitudes [18, 19] . If intent to at-
t
end future visits is related to attrition, this factor could
be used as a covariate in statistical analyses to reduce out-
come and treatment group bias [18] . More importantly,
pa
tient-specific intent to continue coupled with the iden-
tification of possible reasons for future discontinuation
could be used to tailor evidence-based interventions to
retain individual patients.
Side effects and lack of efficacy are common and are
primary reasons for dropping out of antidepressant
treatment trials that last 6 weeks or longer (dropout
rates of 3.719.8 and 2.1–9.9%, respectively [5, 20] ). In a
na
turalistic setting, 55% of the patients on serotonin
reuptake inhibitors reported at least 1 bothersome side
effect within about 3 months of starting treatment and
these patients were about 3 times more likely to drop out
[6] .
This report evaluates participants’ intent to attend fu-
ture treatment visits overall and in the context of side ef-
fects or not getting better. It uses data from the Suicide
Assessment Methodology Study of the Depression Trials
Network of the National Institute of Mental Health
to ad-
dress the following questions: can attrition be predicted
using participants’ self-reported intent to (a) attend the
next treatment visit, (b) remain in the study until comple-
tion, or (c) remain in the study if experiencing side effects
or lack of efficacy?
We anticipated that participants who report a negative
or uncertain intent to continue in the study in response
to these questions are more likely to drop out. We also
explore how the actual experiences of side effects or lack
of efficacy during the study are related to participants’
earlier reported intent to remain in the study.
M e t h o d s
Study Description
The primary objective of the Suicide Assessment Methodolo-
gy Study was to develop easy-to-use clinician- and patient-friend-
ly measures of suicidality and associated symptoms. The Institu-
tional Review Boards at the National Coordinating Center (the
University of Texas Southwestern Medical Center), the Data Co-
ordinating Center and each of 15 regional centers approved and
oversaw the study protocol. The participants provided written in-
formed consent prior to study enrollment.
In all, 265 adult outpatients with nonpsychotic MDD, 18–75
years of age, were enrolled at 6 primary and 9 psychiatric care sites
across the USA and were self- or professionally referred. The re-
sults are therefore more generalizable to real-world clinical set-
tings than those from typical efficacy studies. Nonpsychotic
MDD was diagnosed clinically and confirmed with the Psychiat-
ric Diagnostic Screening Questionnaire [21, 22] and an MDD
c
hecklist from the Diagnostic and Statistical Manual of Mental
Disorders – fourth edition – text revision [23] . The participants
w
ere treated with a selective serotonin reuptake inhibitor (SSRI)
antidepressant and monitored for 8 weeks. The SSRIs used were
at the discretion of the treating physician and could include cital-
opram, escitalopram, fluoxetine, paroxetine, paroxetine-CR or
sertraline. The participants received medications and doses that
are routinely received in clinical practice for a duration of time
that reflects consensually recommended preferred practices.
Clinical research coordinators (CRCs) at each clinical site sup-
ported participants and clinicians.
To provide appropriately vigorous yet tolerable dosing, clini-
cal management was informed by critical decision point dosing
tables and measurement-based care
[24–26] . Measurement-based
c
are included measurement at each clinic visit of (1) depressive
symptom severity with the Quick Inventory of Depressive Symp-
tomatology – Clinician-rated (QIDS-C
16
) [27–29] and (2) side ef-
fects and medication tolerability using the Systematic Assessment
for Treatment-Emergent Events – Systematic Inquiry [30] , a 55-
i
tem self-report that rates the most commonly reported side ef-
fects expected with the study medications, the 3-item Frequency,
Intensity and Burden of Side Effects Rating (FIBSER) [31] , a self-
re
port measure which provides global ratings of frequency, inten-
sity and overall burden due to side effects attributable to the an-
tidepressant treatment, and a self-rated medication treatment ad-
herence questionnaire to assess compliance with the prescribed
antidepressant. Measurement-based care was used successfully in
clinical practice settings in the Sequenced Treatment Alternatives
to Relieve Depression trial
[32, 33] .
Protocol visits were to occur at weeks 0, 2, 4, 6 and 8. In addi-
tion, the QIDS-C
16
and FIBSER were collected by phone at weeks
1, 3, 5 and 7. The participants were also contacted by telephone on
Mondays, Wednesdays and Fridays during the first 2 weeks fol-
lowing medication initiation and following a dose increase (week
4 or later) to evaluate the presence of suicidal ideation and emer-
gence of associated symptoms.
Warden et al.
Psychother Psychosom 2009;78:372–379
374
S t u d y P o p u l a t i o n
A total of 265 eligible outpatients provided written informed
consent and were enrolled and treated from July 2007 through
February 2008. Eligible participants had a score 6 14 on the base-
line 17-item Hamilton Rating Scale for Depression [34, 35] . Par-
t
icipants with general medical conditions (GMCs) were eligible as
long as their GMCs did not contraindicate the use of an SSRI. Pa-
tients were ineligible if they had bipolar disorder; schizophrenia;
schizoaffective disorder; MDD with psychotic features (lifetime);
a current primary diagnosis of anorexia nervosa, bulimia nervo-
sa, or obsessive-compulsive disorder; current substance abuse or
dependence; required inpatient treatment at the time of study en-
try; or had a well-documented history of nonresponse (in the cur-
rent major depressive episode) to two adequately delivered SSRI
treatments. Patients were also ineligible if they were breast-feed-
ing, pregnant, or intending to become pregnant, had taken an
antipsychotic medication within 4 months of study entry, or had
taken antidepressants in the 2 weeks prior to screening (4 weeks
for fluoxetine and 6 weeks for MAOIs). Suicidality was acceptable
as long as inpatient treatment was not indicated at the baseline
visit.
A s s e s s m e n t s
At the screening/baseline visit, CRCs collected clinical and
sociodemographic information and completed the 17-item Ham-
ilton Rating Scale for Depression. At baseline, the participants
also completed a 125-item forced-choice (symptom present or ab-
sent) self-report DSM-IV axis I screening questionnaire, the Psy-
chiatric Diagnostic Screening Questionnaire and the Self-Ad-
ministered Comorbidity Questionnaire
[36] , a 40-item self-report
t
hat assesses the presence of a range of common medical condi-
tions. During all clinic visits, including baseline, the CRCs also
collected the QIDS-C
16
, FIBSER and Systematic Assessment for
Treatment-Emergent Events – Systematic Inquiry.
At baseline and at the week 2, 4 and 6 clinic visits, the partici-
pants responded to the following questions in a self-report for-
mat: (1) How likely is it that you will come back for the next treat-
ment visit? (2) How likely is it that you will complete all 8 weeks
of the study? (3) If your depression is not getting better, how like-
ly are you to stay in the study? (4) If you are experiencing bother-
some side effects, how likely are you to stay in the study?
Each question was rated on a Likert-type scale with 5 possible
responses ranging from very unlikely to very likely. Questions 1
and 2 were derived from the approach suggested by Leon et al. [18]
and Demirtas and Schafer [19] .
Remission was defined as a score of ^ 5 on the last available
QIDS-C
16
. The presence of side effects was determined by a FIB-
SER score of 6 3, reflecting a moderate burden of side effects.
Definition of Attrition
Dropouts were defined as participants who discontinued at-
tending clinic visits at any point after the baseline visit. Those
who left treatment for medical reasons, such as development of a
medical condition that contraindicated the study medication,
were not considered dropouts.
Analytic Methods
Descriptive statistics, means and standard deviations for con-
tinuous variables, and percentages for discrete variables, were
used to characterize the sample.
2
tests and Fisher’s Exact Test
Table 1. Baseline characteristics of the sample
Race
White 170 (64.2)
Black 64 (24.1)
Other 31 (11.7)
Gender
Male 77 (29.1)
Female 188 (70.9)
Employment
Unemployed 95 (36.0)
Employed 155 (58.7)
Retired 14 (5.3)
Marital status
Never married 85 (32.2)
Married 99 (37.5)
Divorced/separated 70 (26.5)
Widowed 10 (3.8)
Insurance
Private 115 (44.4)
Public 50 (19.3)
None 94 (36.3)
Hispanic
No 234 (88.3)
Yes 31 (11.7)
Age at onset (≥18 years)
No 96 (36.6)
Yes 166 (63.4)
Family history of alcohol/drug
No 144 (54.5)
Yes 120 (45.5)
Family history of suicide
No 256 (97.0)
Yes 8 (3.0)
Number of depressive episodes (≥2)
No 79 (33.8)
Yes 155 (66.2)
Chronic depression (≥2 years)
No 174 (71.0)
Yes 71 (29.0)
Anxious features
No 90 (34.0)
Yes 175 (66.0)
Age, years
265 [41.2813.5]
Education, years
264 [13.582.8]
HRSD
17
(without SI item)
264 [21.584.5]
QIDS-C
16
(without SI item)
265 [14.883.2]
GMC severity
264 [3.283.7]
Figures are numbers of cases with percentages in parentheses
and means 8 SD in square brackets. HRSD
17
= 17-item Hamilton
Rating Scale for Depression; QIDS-C
16
= Quick Inventory of De-
pressive Symptomatology (clinician-rated); GMC = general med-
ical condition.
Identifying Risk for Attrition
Psychother Psychosom 2009;78:372–379
375
were applied to test the equality of the probability of attrition
among those who provided positive and uncertain/negative re-
sponses to the assessment of the intent to attend future visits.
2
tests were also used to evaluate the association of attrition, the
presence of side effects and intent to continue with side effects, as
well as attrition, symptom improvement and intent to continue if
not improving. All tests were conducted with a 2-sided alternative
hypothesis.
R e s u l t s
The characteristics of the sample are reported in ta-
ble 1 . About a fifth of the sample dropped out (56/264);
4.6% (n = 12/264) by week 2, 9.9% (n = 26/264) by week
4, 15.5% (n = 41/264) by week 6 and 21.2% (n = 56/264)
by week 8. Remission was significantly less likely for
dropouts (23%) than for completers (47%) (p = 0.0013).
Given the frequency distribution of responses to the
questions about future intent ( table 2 ), the positive re-
sponses ‘very likely’ and ‘likely’ to stay were collapsed to
the category ‘positive’ and the uncertain/negative re-
sponses ‘not sure,’ ‘unlikely’ and ‘very unlikely’ to stay
were collapsed to the category ‘uncertain/negative’ for
analysis.
Uncertain/negative responses to the questions about
coming back for the next treatment visit or completing
the study predicted attrition at several time points, but
very few participants (24%) responded in this manner
( table 3 ). However, a substantial number of participants
were willing to signal negative/uncertain intent at base-
line (2336%) and at week 2 (15–25%) in response to the
questions about side effects or lack of efficacy. An uncer-
tain/negative response at baseline to either of these ques-
tions was associated with attrition at weeks 6 and 8, while
an uncertain/negative response to either of these ques-
tions at week 2 predicted attrition at weeks 4, 6 and 8.
There was no association between baseline severity
and attrition (p = 0.9235). The odds ratio for the QIDS-
C
16
with a 5-unit increase was 1.02, and using the good-
ness of fit test, the model fits well.
Dropouts who experienced side effects were 3 times
more likely to have reported uncertain/negative rather
than positive intent about staying in the study with both-
ersome side effects at baseline (7 vs. 2%). Dropouts who
did not report side effects were still twice as likely to have
reported uncertain/negative rather than positive intent at
baseline about staying in the study with side effects (20
vs. 10%) and 3 times as likely to have reported uncertain/
negative rather than positive intent at week 2 (20 vs. 7%).
Completers, whether or not they actually experienced
side effects during treatment, showed little difference in
uncertain/negative versus positive intent to continue at
either baseline or week 2.
Among both dropouts and completers, participants
who reported an uncertain/negative intent to continue
with lack of improvement had significantly less improve-
ment in symptom severity by the QIDS-C
16
than those
who reported a positive intent to continue (23 vs. 38%
change in dropouts and 43 vs. 56% change in completers).
There was about a 2-point difference in improvement on
the QIDS-C
16
in both of these comparisons, a clinically
meaningful difference.
Discussion
Our results indicate that self-reported intent to con-
tinue if experiencing side effects and intent to continue
even with a lack of initial efficacy are very good markers
for attrition. A substantial percentage of participants
were willing to express concern or ambivalence at base-
line about continuing to attend with side effects (36%) or
Table 2. Responses to intent-to-attend questions at baseline (n = 264) and at week 2 (n = 213)
Response Likely to attend next
visit, %
Likely to complete
8 weeks, %
Stay if not improving,
%
Stay if having side
effects, %
baseline week 2 baseline week 2 baseline week 2 baseline week 2
Very likely 84.5 86.4 81.1 85.0 53.0 60.1 34.9 50.7
Likely 13.3 10.8 14.8 12.6 24.2 24.4 28.8 23.9
Not sure 1.5 1.4 3.4 1.9 21.2 12.7 28.4 22.1
Unlikely 0.0 0.5 0.0 0.0 1.2 2.3 6.4 1.9
Very unlikely 0.7 0.9 0.7 0.5 0.4 0.5 1.5 1.4
Warden et al.
Psychother Psychosom 2009;78:372–379
376
lack of efficacy (23%). An uncertain/negative response to
either of these questions at baseline and at week 2 pre-
dicted attrition, with improved prediction at week 2 for
both questions.
Since 58% of the patients experience at least moderate
side effects as early as 1 week after SSRI initiation [37] ,
th
is improved prediction at week 2 with either the side
effect or lack of improvement question may be due to par-
ticipants basing their retention predictions on actual ex-
perience with side effects with the current treatment and/
or a beginning perception of the drugs efficacy.
Since only 2–4% of the participants expressed a nega-
tive or uncertain intent to continue with treatment
when
asked the more general questions (i.e. likely to attend next
visit, likely to complete 8 weeks), questions addressing
specific concerns as opposed to general ambivalence are
more likely to be useful in clinical practice. Participants
may have good intentions at treatment initiation to com-
plete treatment or may be concerned about disappointing
their clinicians by expressing a general lack of commit-
ment. Patients who actually experienced side effects prior
to dropping out were more likely, at baseline, to have re-
ported an uncertain/negative response about continuing
with side effects. Even more strikingly, however, drop-
outs who did not report side effects were still twice as
likely at baseline and 3 times as likely at week 2 to have
reported they were uncertain or negative about continu-
ing treatment. In addition to predicting attrition, uncer-
tain or negative responses to the question about continu-
ing to attend if not improving were also associated with
a meaningfully lower rate of symptom improvement in
both dropouts and completers. This result may be similar
to studies that found positive expectations of treatment
effectiveness to be associated with improved treatment
outcomes in depression [3840] as well as a possible
m
echanism in the placebo effect [41] .
It is important to resolve adverse events that occur ear-
ly in treatment [37] and maximize the chances of rapid
e
fficacy with aggressive treatment [42] to help reduce at-
t
rition. It is even more proactive, however, to inoculate
patients who directly express concerns or ambivalence
against discontinuation. Uncertain/negative intent to
Table 3. Associations between intent to attend and attrition
Response n Attrition by treatment week
week 2 week 4 week 6 week 8
% p value % p value % p value % p value
Likely to attend next visit
Baseline No 6 0.0 0.5887 0.0 0.4128 16.7 0.9380 16.7 0.7829
Yes 258 4.7 10.1 15.5 21.3
Week 2 No 6 16.7 0.0622 16.7 0.3115 50.0 0.0067
Yes 207 2.9 6.3 12.1
Likely to complete 8 weeks
Baseline No 11 18.2 0.0266 27.3 0.0476 36.4 0.0513 36.4 0.2092
Yes 253 4.0 9.1 14.6 20.6
Week 2 No 5 20.0 0.0339 40.0 0.0023 60.0 0.0017
Yes 208 2.9 5.8 12.0
Likely to stay if not getting better
Baseline No 60 8.3 0.1091 15.0 0.1276 30.0 0.0004 36.7 0.0009
Yes 204 3.4 8.3 11.3 16.7
Week 2 No 33 12.1 0.0020 21.2 0.0002 30.3 0.0015
Yes 180 1.7 3.9 10.0
Likely to stay with bothersome side effects
Baseline No 96 7.3 0.1054 13.5 0.1279 24.0 0.0043 31.3 0.0026
Yes 168 3.0 7.7 10.7 15.5
Week 2 No 54 7.4 0.0493 14.8 0.0047 24.1 0.0059
Yes 159 1.9 3.8 9.4
No = Uncertain/unlikely to continue; Yes = likely to continue.
Identifying Risk for Attrition
Psychother Psychosom 2009;78:372–379
377
continue with side effects or lack of efficacy may reflect
concern about the treatment’s possible harmfulness, con-
sideration that depression is not serious enough to toler-
ate side effects, lack of awareness of the time needed to
reach remission, prior negative experiences with side ef-
fects or any of a host of other possibilities. However, a
noncommittal statement from a patient about continuing
in treatment generally, or in the context of side effects or
lack of treatment efficacy, is an excellent signal that indi-
cates a potentially less resilient patient. With these pa-
tients, it may be helpful for clinicians to conduct an indi-
vidualized review of patient-specific thoughts regarding
side effects and lack of efficacy as well as other concerns
about treatment or medication to personalize the focus of
educational efforts.
There is evidence that such inoculation may be help-
ful. Studies have found that patients who discussed ad-
verse events with physicians or pharmacists at the initia-
tion of therapy and those told to continue medication for
at least 6 months were less likely to discontinue antide-
pressant treatment [6] , and patients told by their clinician
t
hat they might not see a benefit for 2–4 weeks were more
likely to be taking their antidepressants 1 month into
treatment [4] . These interventions may have modified pa-
t
ient beliefs or their intent to continue, or both. Studies
of collaborative care or educational interventions indi-
cate that the provision of educational or supportive inter-
ventions that can address topics such as what to do about
side effects or the time expected to reach remission may
be useful in improving adherence or persistence in taking
medications as well as outcomes [4, 6, 4346] . Personal-
i
zed responses to each patient’s own expressed concerns
may be even more useful in retaining less committed, at-
risk patients in treatment for a sufficient time to maxi-
mize the chances of reaching and sustaining remission
with this often chronic or recurrent illness.
The current study has several limitations that may af-
fect its generalizability. The participants received moni-
tored dosing of medications while a CRC supported both
clinician and patient. Side effects may not have been iden-
tified if they occurred following a visit but prior to drop-
out. Symptom improvement for completers and dropouts
was based on the last available measurement occasion.
Data about prior treatment experience were not available.
We do not have retention data beyond 8 weeks. Some
comparisons had small numbers, making generalizing
from the findings difficult. The current study also did not
address factors related to clinicians (e.g. experience), clin-
ics (e.g. access and availability of appointments), medica-
tion (e.g. complexity of the regimen) or participants (e.g.
specific beliefs or attitudes about medication or treat-
ment).
In summary, questions regarding intent to attend fu-
ture antidepressant treatment visits in the context of side
effects or lack of efficacy may be useful clinical tools.
They may identify patients with concerns about these is-
sues, as well as patients who may be less resilient to the
challenges of completing a course of treatment or have
other specific concerns that can be addressed. An indi-
vidualized review of patient concerns and the individual-
ized tailoring of educational or other interventions may
reduce attrition and increase the chances for remission.
A c k n o w l e d g m e n t s
This project was funded by the National Institute of Mental
Health under Contract N01MH90003 to UT Southwestern Med-
ical Center at Dallas (P.I.: M.H. Trivedi). The content of this pub-
lication does not necessarily reflect the views or policies of the
Department of Health and Human Services, nor does mention of
trade names, commercial products or organizations imply en-
dorsement by the US Government. This analysis was also sup-
ported in part by a National Alliance for Research on Schizophre-
nia and Depression Young Investigator Award (D. Warden). We
would also like to acknowledge the editorial support of Jon Kilner,
MS, MA.
Disclosures
D. Warden, PhD, MBA currently owns stock in Pfizer Inc. and
has owned stock in Bristol-Myers Squibb Company within the
last 5 years.
M.H. Trivedi., MD has been a consultant for Abbott Labora-
tories Inc., Akzo (Organon Pharmaceuticals Inc.), AstraZeneca,
Bayer, Bristol-Myers Squibb Company, Cephalon Inc., Cyberon-
ics Inc., Eli Lilly & Company, Fabre-Kramer Pharmaceuticals
Inc., Forest Pharmaceuticals, GlaxoSmithKline, Janssen Pharma-
ceutica Products, LP, Johnson & Johnson PRD, Meade Johnson,
Neuronetics, Parke-Davis Pharmaceuticals Inc., Pfizer Inc., Phar-
macia & Upjohn, Sepracor, Solvay Pharmaceuticals Inc., Vantage-
Point and Wyeth-Ayerst Laboratories. He has served on speaker
bureaus for Abdi Brahim, Akzo (Organon Pharmaceuticals Inc.),
Bristol-Myers Squibb Company, Cephalon Inc., Cyberonics Inc.,
Forest Pharmaceuticals, GlaxoSmithKline, Janssen Pharmaceu-
tica Products, LP, Eli Lilly & Company, Pharmacia & Upjohn,
Solvay Pharmaceuticals Inc., and Wyeth-Ayerst Laboratories. He
has also received grant support from Bristol-Myers Squibb Com-
pany, Cephalon Inc., Corcept Therapeutics Inc., Cyberonics Inc.,
Eli Lilly & Company, Forest Pharmaceuticals, GlaxoSmithKline,
Janssen Pharmaceutica, Merck, National Institute of Mental
Health, National Alliance for Research in Schizophrenia and De-
pression, Novartis, Pfizer Inc., Pharmacia & Upjohn, Predix
Pharmaceuticals, Solvay Pharmaceuticals Inc. and Wyeth-Ayerst
Laboratories.
Warden et al.
Psychother Psychosom 2009;78:372–379
378
S.R. Wisniewski., PhD has been a consultant for Cyberonics
Inc. (2005–2006), ImaRx Therapeutics Inc. (2006), Bristol-Myers
Squibb Company (2007), Organon (2007) and Case-Western Uni-
versity (2007).
I.M. Lesser, MD has received grant support from the National
Institute of Mental Health and Aspect Medical Systems, and has
served on the speaker bureau of the Medical Education Speakers
Network.
J. Mitchell, MD has received research support from Bristol-
Myers Squibb, Jazz Pharmaceuticals, Eli Lilly and Ortho McNeil,
has been on the advisory boards of and/or a consultant for Eli
Lilly, has been on the speaker bureaus of Eli Lilly and Forest Lab-
oratories, and has equity holdings (excluding mutual funds/
blinded trusts) in Eli Lilly, Forest, Pfizer and Sanofi-Aventis.
M. Fava, MD has received research support from Abbott Lab-
oratories, Alkermes, Aspect Medical Systems, Astra-Zeneca,
Bristol-Myers Squibb Company, Cephalon, Forest Pharmaceuti-
cals Inc., GlaxoSmithKline, J & J Pharmaceuticals, Lichtwer
Pharma GmbH, Eli Lilly & Company, Lorex Pharmaceuticals,
Novartis, Organon Inc., PamLab, LLC, Pfizer Inc., Pharmavite,
Roche, Sanofi/Synthelabo, Solvay Pharmaceuticals Inc. and
Wyeth-Ayerst Laboratories. He has provided advisory/consulting
services to Aspect Medical Systems, Astra-Zeneca, Bayer AG,
Biovail Pharmaceuticals Inc., BrainCells Inc., Bristol-Myers
Squibb Company, Cephalon, Compellis, Cypress Pharmaceuti-
cals, Dov Pharmaceuticals, EPIX Pharmaceuticals, Fabre-Kra-
mer Pharmaceuticals Inc., Forest Pharmaceuticals Inc., Glaxo-
SmithKline, Grünenthal GmbH, J & J Pharmaceuticals, Janssen
Pharmaceutica, Jazz Pharmaceuticals, Knoll Pharmaceutical
Company, Eli Lilly & Company, Lundbeck, MedAvante Inc., No-
vartis, Nutrition 21, Organon Inc., PamLab, LLC, Pfizer Inc,
PharmaStar, Pharmavite, Roche, Sanofi/Synthelabo, Sepracor,
Solvay Pharmaceuticals Inc., Somerset Pharmaceuticals and
Wyeth-Ayerst Laboratories. He has been on speaking bureaus for
Astra-Zeneca, Bristol-Myers Squibb Company, Cephalon, Forest
Pharmaceuticals Inc., GlaxoSmithKline, Eli Lilly & Company,
Novartis, Organon Inc., Pfizer Inc., PharmaStar and Wyeth-
Ayerst Laboratories. M.F. has equity holdings (excluding mutual
funds/blinded trusts) with Compellis and MedAvante.
A.J. Rush, MD has received research support from the Nation-
al Institute of Mental Health, the Robert Wood Johnson Founda-
tion and the Stanley Medical Research Institute. He has been on
the advisory boards and/or consultant for Advanced Neuromod-
ulation Systems Inc., AstraZeneca, Best Practice Project Manage-
ment Inc., Bristol-Myers Squibb Company, Cyberonics Inc., Eli
Lilly & Company, Gerson Lehman Group, GlaxoSmithKline, Jazz
Pharmaceuticals, Magellan Health Services, Merck & Co. Inc.,
Neuronetics, Ono Pharmaceutical, Organon USA Inc., Otsuka
Pharmaceuticals, Pam Lab, Personality Disorder Research Corp.,
Pfizer Inc., The Urban Institute and Wyeth-Ayerst Laboratories
Inc. He has been on the speaker’s bureau for Cyberonics Inc., For-
est Pharmaceuticals Inc. and GlaxoSmithKline, has equity hold-
ings (excluding mutual funds/blinded trusts) in Pfizer Inc. and
has royalty income affiliations with Guilford Publications and
Healthcare Technology Systems Inc.
G.K. Balasubramani, PhD, Kathy Shores-Wilson, PhD, and
Diane Stegman, RN have no disclosures to report.
References
1 Bockting CL, ten Doesschate MC, Spijker J,
Spinhoven P, Koeter MW, Schene AH: Con-
tinuation and maintenance use of antide-
pressants in recurrent depression. Psycho-
ther Psychosom 2008;
77: 17–26.
2 Katon W, Von Korff M, Lin E, Bush T, Or -
mel J: Adequacy and duration of antidepres-
sant treatment in primary care. Med Care
1992;
30: 67–76.
3 Maddox JC, Levi M, Thompson C: The com-
pliance with antidepressants in general prac-
tice. J Psychopharmacol 1994;
8: 48–53.
4 Lin EH, Von Korff M, Katon W, Bush T, Si-
mon GE, Walker E, Robinson P: The role of
the primary care physician in patients’ ad-
herence to antidepressant therapy. Med Care
1995;
33: 6774.
5 Machado M, Iskedjian M, Ruiz I, Einarson
TR: Remission, dropouts, and adverse drug
reaction rates in major depressive disorder: a
meta-analysis of head-to-head trials. Curr
Med Res Opin 2006;
22: 1825–1837.
6 Bull SA, Hu XH, Hunkeler EM, Lee JY, Ming
EE, Markson LE, Fireman B: Discontinua-
tion of use and switching of antidepressants:
influence of patient-physician communica-
tion. JAMA 2002;
288: 1403–1409.
7 Melfi CA, Chawla AJ, Croghan TW, Hanna
MP, Kennedy S, Sredl K: The effects of adher-
ence to antidepressant treatment guidelines
on relapse and recurrence of depression.
Arch Gen Psychiatry 1998;
55: 1128–1132.
8 Demyttenaere K, Adelin A, Patrick M, Wal-
there D, Katrien DB, Michele S: Six-month
compliance with antidepressant medication
in the treatment of major depressive disor-
der. Int Clin Psychopharmacol 2008;
23: 36–
42.
9 Warden D, Rush AJ, Wisniewski SR, Lesser
IM, Kornstein SG, Balasubramani GK, Thase
ME, Preskorn SH, Nierenberg AA, Young
EA, Shores-Wilson K, Trivedi MH: What
predicts attrition in second step medication
treatments for depression? A STAR * D re-
port. Int J Neuropsychopharmacol 2009;12:
459–473.
10 Hirschfeld RM, Dunner DL, Keitner G,
Klein DN, Koran LM, Kornstein SG, Mar-
kowitz JC, Miller I, Nemeroff CB, Ninan PT,
Rush AJ, Schatzberg AF, Thase ME, Trivedi
MH, Borian FE, Crits-Christoph P, Keller
MB: Does psychosocial functioning improve
independent of depressive symptoms? A
comparison of nefazodone, psychotherapy,
and their combination. Biol Psychiatry 2002;
51: 123–133.
11 Rush AJ, Trivedi MH, Wisniewski SR, Nie-
renberg AA, Stewart JW, Warden D, Nie-
derehe G, Thase ME, Lavori PW, Lebowitz
BD, McGrath PJ, Rosenbaum JF, Sackeim
HA, Kupfer DJ, Luther J, Fava M: Acute and
longer-term outcomes in depressed outpa-
tients requiring one or several treatment
steps: a STAR * D report. Am J Psychiatry
2006;
163: 1905–1917.
12 Insel TR, Quirion R: Psychiatry as a clinical
neuroscience discipline. http://www.nimh.
nih.gov/about/director/publications/psy-
chiatry-as-a-neuroscience-discipline.shtml
(last updated: 6–172008, accessed: 6–13–
2008).
13 Mant D: Can randomised trials inform clin-
ical decisions about individual patients?
Lancet 1999;
353: 743–746.
14 Aikens JE, Kroenke K, Swindle RW, Eckert
GJ: Nine-month predictors and outcomes of
SSRI antidepressant continuation in prima-
ry care. Gen Hosp Psychiatry 2005;
27: 229–
236.
15 Bultman DC, Svarstad BL: Effects of physi-
cian communication style on client medica-
tion beliefs and adherence with antidepres-
sant treatment. Patient Educ Couns 2000;
40:
173–185.
Identifying Risk for Attrition
Psychother Psychosom 2009;78:372–379
379
16 Hunot VM, Horne R, Leese MN, Churchill
RC: A cohort study of adherence to antide-
pressants in primary care: the influence of
antidepressant concerns and treatment pref-
erences. Prim Care Companion J Clin Psy-
chiatry 2007;
9: 91–99.
17 Lin EH, Von Korff M, Ludman EJ, Rutter C,
Bush TM, Simon GE, Unutzer J, Walker E,
Katon WJ: Enhancing adherence to prevent
depression relapse in primary care. Gen
Hosp Psychiatry 2003;
25: 303–310.
18 Leon AC, Demirtas H, Hedeker D: Bias re-
duction with an adjustment for participants’
intent to dropout of a randomized controlled
clinical trial. Clin Trials 2007;
4: 540–547.
19 Demirtas H, Schafer JL: On the perfor-
mance of random-coefficient pattern-mix-
ture models for non-ignorable drop-out. Stat
Med 2003;
22: 2553–2575.
20 Vis PM, van Baardewijk M, Einarson TR:
Duloxetine and venlafaxine-XR in the treat-
ment of major depressive disorder: a meta-
analysis of randomized clinical trials. Ann
Pharmacother 2005;
39: 1798–1807.
21 Zimmerman M, Mattia JI: A self-report scale
to help make psychiatric diagnoses: the Psy-
chiatric Diagnostic Screening Question-
naire. Arch Gen Psychiatry 2001;
58: 787–
794.
22 Zimmerman M, Mattia JI: The Psychiatric
Diagnostic Screening Questionnaire: devel-
opment, reliability and validity. Compr Psy-
chiatry 2001;
42: 175–189.
23 American Psychiatric Association: Diagnos-
tic and Statistical Manual of Mental Disor-
ders, ed 4, rev. Washington, American Psy-
chiatric Press, 2000.
24 Trivedi MH, Rush AJ, Wisniewski SR, Nie-
renberg AA, Warden D, Ritz L, Norquist G,
Howland RH, Lebowitz B, McGrath PJ,
Shores-Wilson K, Biggs MM, Balasubramani
GK, Fava M: Evaluation of outcomes with
citalopram for depression using measure-
ment-based care in STAR * D: implications
for clinical practice. Am J Psychiatry 2006;
163: 28–40.
25 Trivedi MH, Rush AJ, Gaynes BN, Stewart
JW, Wisniewski SR, Warden D, Ritz L, Lu-
ther JF, Stegman D, DeVeaugh-Geiss J, How-
land R: Maximizing the adequacy of medica-
tion treatment in controlled trials and
clinical practice: STAR * D measurement-
based care. Neuropsychopharmacology
2007;
32: 2479–2489.
26 Trivedi MH, Daly EJ: Measurement-based
care for refractory depression: a clinical de-
cision support model for clinical research
and practice. Drug Alcohol Depend 2007;
88(suppl 2):S61–S71.
27 Rush AJ, Carmody TJ, Reimitz PE: The In-
ventory of Depressive Symptomatology
(IDS): clinician (IDS-C) and self-report
(IDS-SR) ratings of depressive symptoms.
Int J Methods Psychiatr Res 2000;
9: 45–59.
28 Rush AJ, Trivedi MH, Ibrahim HM, Carmo-
dy TJ, Arnow B, Klein DN, Markowitz JC,
Ninan PT, Kornstein S, Manber R, Thase
ME, Kocsis JH, Keller MB: The 16-Item
Quick Inventory of Depressive Symptom-
atology (QIDS), clinician rating (QIDS-C),
and self-report (QIDS-SR): a psychometric
evaluation in patients with chronic major
depression. Biol Psychiatry 2003;
54: 573–583
(erratum p 585).
29 Trivedi MH, Rush AJ, Ibrahim HM, Carmo-
dy TJ, Biggs MM, Suppes T, Crismon ML,
Shores-Wilson K, Toprac MG, Dennehy EB,
Witte B, Kashner TM: The Inventory of De-
pressive Symptomatology, Clinician Rating
(IDS-C) and Self-Report (IDS-SR), and the
Quick Inventory of Depressive Symptom-
atology, Clinician Rating (QIDS-C) and Self-
Report (QIDS-SR) in public sector patients
with mood disorders: a psychometric evalu-
ation. Psychol Med 2004;
34: 73–82.
30 Levine J, Schooler NR: SAFTEE: a technique
for the systematic assessment of side effects
in clinical trials. Psychopharmacol Bull
1986;
22: 343–381.
31 Wisniewski SR, Rush AJ, Balasubramani
GK, Trivedi MH, Nierenberg AA: Self-rated
global measure of the frequency, intensity,
and burden of side effects. J Psychiatr Pract
2006;
12: 71–79.
32 Fava M, Rush AJ, Trivedi MH, Nierenberg
AA, Thase ME, Sackeim HA, Quitkin FM,
Wisniewski S, Lavori PW, Rosenbaum JF,
Kupfer DJ: Background and rationale for the
Sequenced Treatment Alternatives to Relieve
Depression (STAR * D) study. Psychiatr Clin
North Am 2003;
26: 457–494, x.
33 Rush AJ, Fava M, Wisniewski SR, Lavori PW,
Trivedi MH, Sackeim HA, Thase ME, Nie-
renberg AA, Quitkin FM, Kashner TM, Kup-
fer DJ, Rosenbaum JF, Alpert J, Stewart JW,
McGrath PJ, Biggs MM, Shores-Wilson K,
Lebowitz BD, Ritz L, Niederehe G: Sequenced
Treatment Alternatives to Relieve Depres-
sion (STAR * D): rationale and design. Con-
trol Clin Trials 2004;
25: 119–142.
34 Hamilton M: A rating scale for depression. J
Neurol Neurosurg Psychiatry 1960;
23: 56
62.
35 Hamilton M: Development of a rating scale
for primary depressive illness. Br J Soc Clin
Psychol 1967;
6: 278–296.
36 Sangha O, Stucki G, Liang MH, Fossel AH,
Katz JN: The Self-Administered Comorbid-
ity Questionnaire: a new method to assess
comorbidity for clinical and health services
research. Arthritis Rheum 2003;
49: 156
163.
37 Demyttenaere K, Albert A, Mesters P, Dewe
W, De Bruyckere K, Sangeleer M: What hap-
pens with adverse events during 6 months of
treatment with selective serotonin reuptake
inhibitors? J Clin Psychiatry 2005;
66: 859–
863.
38 Sotsk y S, Glass D, Shea M, Pil konis P, Collins
J, Elkin I, Watkins JT, Imber SD, Leber WR,
Moyer J, Oliveri ME: Patient predictors of re-
sponse to psychotherapy and pharmacother-
apy: findings in the NIMH Treatment of De-
pression Collaborative Research Program.
Am J Psychiatry 1991;
148: 9971008.
39 Krell HV, Leuchter AF, Morgan MM, Cook
IA, Abrams M: Subject expectations of treat-
ment effectiveness and outcome of treatment
with an experimental antidepressant. J Clin
Psychiatry 2004;
65: 1174–1179.
40 Imber SD, Pilkonis PA, Sotsky SM, Elkin I,
Watkins JT, Collins JF, Shea MT, Leber WR,
Glass DR: Mode-specific effects among three
treatments for depression. J Consult Clin
Psychol 1990;
58: 352–359.
41 Enck P, Benedetti F, Schedlowski M: New in-
sights into the placebo and nocebo respons-
es. Neuron 2008;
59: 195–206.
42 Warden D, Rush AJ, Trivedi MH, Fava M,
Wisniewski SR: The STAR * D project results:
a comprehensive review of findings. Curr
Psychiatry Rep 2007;
9: 449–459.
43 Katon W, Von Korff M, Lin E, Walker E, Si-
mon GE, Bush T, Robinson P, Russo J: Col-
laborative management to achieve treatment
guidelines: impact on depression in primary
care. JAMA 1995;
273: 1026–1031.
44 Katon W, Robinson P, Von Korff M, Lin E,
Bush T, Ludman E, Simon G, Walker E: A
multifaceted intervention to improve treat-
ment of depression in primary care. Arch
Gen Psychiatry 1996;
53: 924–932.
45 Katon W, Von Korff M, Lin E, Simon G,
Walker E, Unutzer J, Bush T, Russo J, Lud-
man E: Stepped collaborative care for prima-
ry care patients with persistent symptoms of
depression: a randomized trial. Arch Gen
Psychiatry 1999;
56: 1109–1115.
46 Stiefel F, Zdrojewski C, Bel Hadj F, Boffa D,
Dorogi Y, So A, Ruiz J, de Jonge P: Effects of
a multifaceted psychiatric intervention tar-
geted for the complex medically ill: a ran-
domized controlled trial. Psychother Psy-
chosom 2008;
77: 247–256.
    • "In a study of patients' mental representations of antidepressant medications , harm was the most common theme to emerge (Piguet et al., 2007 ). Patients who express early ambivalence are twice as likely to discontinue medications prematurely and three times more likely to stop medications prematurely in the context of side effects (Warden et al., 2009). Residents who understand that the deck is often stacked against medication adherence before the patient even begins treatment will be more equipped to recognize, explore, and ameliorate ambivalence about medications. "
    [Show abstract] [Hide abstract] ABSTRACT: Abstract A growing body of evidence suggests that psychiatric medication outcomes are shaped significantly by psychological and social factors surrounding the prescribing process. Little, however, is known about the extent to which psychiatry programs integrate this evidence base into residency training or the methods by which this is accomplished. Psychiatry residency program directors and chief residents participated in an exploratory online survey to establish how psychosocial factors known to impact medication outcomes are integrated into psychopharmacology education. While participants highly valued the importance of psychosocial factors in the prescribing process, there was limited emphasis of these factors in psychopharmacology training. Additionally, some teaching methods that could advance understanding of complex interactions in the psychopharmacology relationship were found to be underutilized. Given that medication outcomes are significantly influenced by psychosocial factors, psychiatric educators have a responsibility to teach residents about the evidence base available. Residents exposed to this evidence base will be better equipped to manage the complexities of the psychopharmacology role. The results of this study offer clues as to how psychosocial factors may be more fully integrated into residency psychopharmacology training.
    Article · Jun 2014
  • [Show abstract] [Hide abstract] ABSTRACT: Attitudes and expectations about treatment have been associated with symptomatic outcomes, adherence and utilization in patients with psychiatric disorders. No measure of patients' anticipated benefits of treatment on domains of everyday functioning has previously been available. The Anticipated Benefits of Care (ABC) is a new, 10-item questionnaire used to measure patient expectations about the impact of treatment on domains of everyday functioning. The ABC was collected at baseline in adult out-patients with major depressive disorder (MDD) (n=528), bipolar disorder (n=395) and schizophrenia (n=447) in the Texas Medication Algorithm Project (TMAP). Psychometric properties of the ABC were assessed, and the association of ABC scores with treatment response at 3 months was evaluated. Evaluation of the ABC's internal consistency yielded Cronbach's alpha of 0.90-0.92 for patients across disorders. Factor analysis showed that the ABC was unidimensional for all patients and for patients with each disorder. For patients with MDD, lower anticipated benefits of treatment was associated with less symptom improvement and lower odds of treatment response [odds ratio (OR) 0.72, 95% confidence interval (CI) 0.57-0.87, p=0.0011]. There was no association between ABC and symptom improvement or treatment response for patients with bipolar disorder or schizophrenia, possibly because these patients had modest benefits with treatment. The ABC is the first self-report that measures patient expectations about the benefits of treatment on everyday functioning, filling an important gap in available assessments of attitudes and expectations about treatment. The ABC is simple, easy to use, and has acceptable psychometric properties for use in research or clinical settings.
    Article · Sep 2009
  • [Show abstract] [Hide abstract] ABSTRACT: Meta-analyses have consistently concluded that a positive therapeutic alliance is associated with better clinical outcomes and progress. To date, however, very few studies have focused on sociodemographic or clinical patient characteristics as moderators of alliance. A multicenter longitudinal treatment outcome study was conducted to investigate the associations of patient and clinician perceptions of the therapeutic alliance with improvement in depression, and to investigate whether these associations were influenced by sociodemographic or clinical characteristics of the patient. Clinician-rated Montgomery Åsberg Depressive Rating Scale scores and both patient- and therapist-rated Helping Alliance Questionnaire (HAQ-I) scores were obtained from 567 outpatients with major depressive disorder who received 6 months of combined psycho- and pharmacotherapy. Multilevel repeated-measures analyses indicated that patient- and therapist-rated HAQ-I scores, 4 weeks after treatment began, positively predicted subsequent clinical change, controlling for the effect of early improvement and a range of patient characteristics. Next to alliance, early improvement, initial depressive symptom severity, a history of psychiatric disorders, and occupational status affected the rate of clinical improvement. Personality pathology comorbidity, marital and occupational status, and the atypical character of the major depressive episode (MDE) moderated the alliance-outcome relationship, depending on the informant (patient or therapist) of therapeutic alliance. The present findings suggest that therapist and patient ratings of therapeutic alliance predict therapeutic progress, and that this relation may be moderated by client characteristics, including personality pathology comorbidity, marital status, occupational status, and the atypical character of the MDE.
    Article · Sep 2010
Show more

  • undefined · undefined
  • undefined · undefined
  • undefined · undefined