Open Journal of Psychiatry, 2012, 2, 272-280 OJPsych Published Online October 2012 (
Discontinuation of antidepressant therapy among patients
with major depressive disorder
Shreya Davé1*, Peter Classi2, Trong Kim Le2, Andrew Maguire1, Susan Ball2,3
1Epidemiology and Database Analytics, United BioSource Corporation, London, UK
2Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, USA
3Indiana University School of Medicine, Indianapolis, USA
Email: *
Received 27 June 2012; revised 29 July 2012; accepted 9 August 2012
Objective: Patients with major depressive disorder
(MDD) often discontinue antidepressant therapy pre-
maturely risking relapse, despite United Kingdom
(UK) guidelines recommending therapy for up to at
least six months after remission. More information is
needed on the patterns of antidepressant discontinua-
tion in UK primary care. Objectives of the study were
to assess the patterns, incidence and predictors of
therapy discontinuation among MDD patients initi-
ating treatment with selective serotonin reuptake in-
hibitors (SSRIs). Methods: This was a retrospective
cohort study using general practices registered with
the General Practice Research Database (GPRD).
15,274 patients with MDD receiving a first ever pre-
scription (index) for an SSRI between 2006-2008 were
identified in GPRD. Discontinuation (including tem-
porary gaps) and cessation of antidepressant therapy
were examined over follow-up. Predictors of inci-
dence of discontinuation in the six months after index
were assessed. Results: Incidence of discontinuation
of antidepressant therapy over follow-up was 80.05
per 100 person years (95% CI 78.94 - 81.17). At six
months after index 42% of patients had discontinued
and 33% had ceased therapy altogether. Lower dis-
continuation of index SSRI therapy in the first six
months after initiation was associated with higher age,
higher body mass index (BMI), and comorbid irrita-
ble bowel syndrome. Higher discontinuation was as-
sociated with paroxetine or fluoxetine at index, and a
more recent index calendar year. Conclusions: There
is a significant risk of discontinuation of antidepres-
sant therapy in the 6 months after initiation of treat-
ment for MDD. This finding requires awareness by
the general practitioner (GP) to ensure implementa-
tion of optimal treatment regimens, and minimization
of therap y non-complian ce among MDD patients .
Keywords: Anti-Depressive Medications; Serotonin
Uptake Inhibitors; Major Depressive Disorder;
Medication Adherence; General Practice
Approximately 13% of general practice attendees in the
United Kingdom (UK) suffer major depressive disorder
(MDD) [1]. Selective serotonin reuptake inhibitors (SSRIs)
are recommended as a first-line treatment option due
their relatively low cost, good efficacy and safety [2].
Therapy is recommended until depressive symptoms
have been managed, and the patient is in remission.
However, continuation of therapy is recommended for at
least six months after remission to reduce risk of relapse
[2], and can be up to two years maintenance treatment if
the patient has chronic depression, severe functional im-
pairment, or is at risk of severe consequences such as
suicide [2]. Premature discontinuation of antidepressant
treatment is associated with a high risk of relapse and
adverse consequences [3] including rapid re-onset of
future depressive episodes, that can be more chronic
and severe than if complete remission had been achieved,
and moreover this risk is enhanced with untreated re-
sidual symptoms following a depressive episode [4,5].
Hence, appropriate clinical management and reliable
follow-up of patients suffering a depressive episode are
Discontinuation of antidepressant treatment can occur
for several reasons [6]. Patients in remission from de-
pression may discontinue therapy because they believe
they can proceed with a non-depressed life without tak-
ing drugs. Patients may also discontinue due to an inade-
quate response to their initial treatment [7], without im-
plementation of appropriate alternative second-step treat-
ment strategies [8]. Furthermore, patients may be non-
compliant with their medication or discontinue because
of adverse side effects [9,10]. If we are to assess the po-
tential unmet need in antidepressant treatment, it is per-
*Corresponding author.
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280 273
tinent to examine patterns of therapy discontinuation and
associated factors. The aim of this study was to assess
incidence of discontinuation (including temporary treat-
ment gaps) and cessation (discontinuation without sub-
sequent therapy resumption) of antidepressant treatment
among new initiators of SSRI treatment in UK primary
care, and examine predictors of discontinuation in the six
months after therapy initiation.
2.1. Study Design and Data Source
This was a retrospective cohort study using data from the
General Practice Research Database (GPRD)
( GPRD is a large primary care database
containing anonymised patient level medical and demo-
graphic information on approximately 8% of the UK
population from 630 general practices. Containing ap-
proximately 11 million patients and 67 million patient
years of data, the GPRD is a good resource for longitu-
dinal epidemiological research, and has been widely used
for studies on depression [11-14]. The database contains
information on diagnoses, symptoms, prescriptions, spe-
cialist referrals, test results, demographics, and adminis-
trative information.
2.2. Study Population
We extracted a cohort of adult patients who had received
a first-time prescription for an SSRI between 01/06/2006
and 31/12/2008; this was the enrolment period and the
date of first SSRI prescription was the index date. Pa-
tients were followed-up until 25/10/2010, where data were
Patients had to have registered with the practice at
least 12 months before the index date to avoid collecting
data from patients recently transferred into a practice
who were likely to have historical diagnoses updated to
their medical record [15]. Patients with prior prescrip-
tions for antidepressants were excluded to ensure only
new initiators of MDD therapy were included. In addi-
tion, inclusion was limited to patients with an index date
after the practice became “up-to standard” (i.e. had good
quality computerised data) (, and those
who were deemed “acceptable” according to GPRD’s da-
ta quality criteria.
Our final study cohort comprised of patients with a
specific Read code [16] for MDD or a positive score entry
for MDD on a recommended [2] depression screening ques-
tionnaire (Patient Health Questionnaire (PHQ)—cut-off
of 10 [17], Hospital Anxiety and Depression Scale
(HADS)—cut-off of 11 [18], or the Beck Depression
Inventory-II (BDI-II)—cut-off of 20 [19]) in the period
from six months before to three months after the index
date. This window allowed inclusion of patients experi-
encing a time lag between diagnosis and treatment with
SSRIs, and also for general practitioners (GPs) who en-
tered a diagnosis after treatment had commenced. The
PHQ is the most widely used measure and has been
validated in UK primary care, showing high sensitivity
(91.7%) and specificity (78.3%) for detection of MDD
when compared with a gold-standard diagnostic clinical
interview [20].
2.3. Treatment Outcomes
The MDD treatment trajectory (using relevant drug code
lists) was tracked among patients from the index date to
the end of follow-up. Discontinuation was defined as
discontinuation of an antidepressant prescription without
receiving another within 60 days of exhausting the sup-
ply of the prior prescription. If there were any other ad-
jacent (<60 days) therapy outcomes such as a drug
switch, dose increase or adjunctive therapy then this was
not classified as a discontinuation of the previous medi-
cation. Thus, discontinuation represented a gap in the
treatment trajectory if the patient subsequently resumed
MDD therapy, or was a cessation of therapy if no further
antidepressant prescriptions were received up to the end
of patient follow-up (to determine this accurately patients
were required to have at least six months of follow-up
after discontinuing with no prescriptions for antidepres-
sants during that period).
Persistence on therapy was defined as remaining on
antidepressants (not discontinuing as above) in the 12
months after index [21], with allowable gaps of <60 days
in treatment.
2.4. Covariates
Covariates included patient age at index, gender, and
several baseline variables including depression severity,
anxiety, alcohol and substance misuse, and chronic pain
conditions (including irritable bowel syndrome, diabetic
neuropathy, postherpetic neuropathy, trigeminal neural-
gia, lumbar radiculopathy, cervical radiculopathy, com-
plex regional pain syndrome, spinal cord injury, fibro-
myalgia, osteoarthritis, rheumatoid arthritis, migraine,
chronic low back pain, ankylosing spondylitis, psoriatic
arthropathy, and interstitial cystitis). These comorbidities
were indentified using relevant Read code lists. We cal-
culated the Charlson Comorbidity Index (CCI) [22], an
established measure of general disease burden that has
been adapted for use with UK primary care databases
[23]. These covariates were selected due to evidence of
their association with depression and/or its treatment. For
example, the concurrent treatment of pain conditions
with non-steroidal anti-inflammatory drugs (NSAIDs)
can reduce the efficiency of SSRIs [24].
Copyright © 2012 SciRes. OPEN ACCESS
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280
2.5. Statistics
Descriptive statistics, including frequency distributions,
percentages, means (SD) and medians (10th to 90th per-
centiles) were used to describe baseline patient charac-
teristics. Among patients with at least a year of follow-up
data we calculated: 1) Persistence [21] at 12 months on
the index SSRI, any SSRI, and any antidepressant; 2)
Frequency of patients by duration of index medication;
and 3) Median length of index therapy in the year after
Among all patients, Kaplan Meier methods were used
to describe the cumulative failure pattern for: 1) Index
SSRI therapy discontinuation; and 2) cessation of any
antidepressant therapy. This was assessed from the index
date up to the end of follow-up. Incidence of discontin-
uation of the index SSRI, any SSRI, and any antide-
pressant was calculated using total person-time at risk as
a denominator.
The bivariate association of each covariate with inci-
dence of discontinuation of the index SSRI in the six
months after therapy initiation was tested individually
using Poisson regression. Associations significant at the
p = 0.10 level in the bivariate analysis were tested in an
adjusted Poisson regression model to assess independent
predictors of incidence of discontinuation of the index
SSRI in the six months after therapy initiation. Wald
tests aided modeling.
This study was reviewed and approved by the Inde-
pendent Scientific Advisory Committee (protocol num-
ber 10_138).
94,932 eligible patients received a first-time SSRI pre-
scription in the enrolment period. Of these, 2059 had a
recent specific MDD Read code entry, and an additional
13,215 scored above the cut-off for MDD on the PHQ,
HADs and/or BDI-II, resulting in a final study cohort of
15,274 patients. 13,927/15,274 (91.2%) patients had at
least a year of follow-up data after index. Median length
of follow-up was 33.8 months (10th-90th percentiles 14.0 -
48.2), median age was 38.0 years (10th-90th percentiles
21.0 - 63.0), and 59.6% were female (Table 1). Further
characteristics are described in Table 1.
Persistence (at 12 months) for the index SSRI, for any
SSRI and for any antidepressant was 2075/13,927
(14.9%), 3251/13,927 (23.3%) and 3516/13,927 (25.3%)
respectively. Median duration of index SSRI therapy in
the year after index was 61 days (10th to 90th percentiles
28 to 365, N = 13,927). Figure 1 shows 4744/13,927
(34%) patients remained on the index SSRI for 28 - 30
Figure 2 shows Kaplan Meier failure curves for dis-
continuation of the index medication, and cessation of all
Table 1. Baseline patient characteristics (N = 15,274).
Variable Numbers of patients (%)a
Follow-up: >1 year 13,927 (91.2%)
Median follow-up
(percentiles: 10th, 90th) (m onths) 33.8 (14.0 - 48.2)
Median age at index
(percentiles: 10th, 90th) (years) 38.0 (21.0 - 63.0)
Age group at index (years)
18 - 29 4526 (29.6%)
30 - 39 3512 (23.0%)
40 - 49 3087 (20.2%)
50 - 59 2097 (13.7%)
60 - 69 1023 (6.7%)
70 - 79 643 (4.2%)
80+ 386 (2.5%)
Gender: Female 9100 (59.6%)
Index SSRI
Citalopram 7468 (48.9%)
Escitalopram 555 (3.6%)
Fluoxetine 6477 (42.4%)
Paroxetine 161 (1.1%)
Sertraline 612 (4.0%)
Mean body mass index
(index value) 27.7 (SD 6.6)
History of depressive disorder 2256 (14.8%)
Anxiety disorder in past 12 months 2144 (14.0%)
Alcohol and substance abuse in
past 12 months 2017 (13.2%)
Chronic pain conditions 3278 (21.5%)
Irritable bowel syndrome 1040 (6.8%)
Charlson comorbidity index:
0 10,758 (70.4%)
1 3271 (21.4%)
2 644 (4.2%)
3 345 (2.3%)
4 130 (0.9%)
5 62 (0.4%)
6+ 64 (0.4%)
Median CCI (percentiles:
10th, 90th) (CCI value) 0.0 (0.0 - 1.0)
aUnless otherwise specified; Note: Comorbidities included in the table are
those associated with discontinuation of antidepressant therapy in the
bivariate analysis.
antidepressant therapy. At three months after index 25%
of patients had discontinued their index drug; at six
months this was 42%, at 12 months 57%, at 24 months
64% and by 48 months 67% had discontinued their index
medication. Twenty-two percent of patients had treat-
Copyright © 2012 SciRes. OPEN ACCESS
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280
Copyright © 2012 SciRes.
05001000 1500 2000 2500 3000 3500
Numbers of patients
028 56 84 112140 168 196 224 252 280308 336 364
Dur a t ion of inde x t h er apy ( da y s )
Figure 1. Frequency of patients by duration of index therapy in first year after index (N = 13,927).
Proportion failing
0 10 20 30 40 50
Time from index (months)
index_discon cessation_the
Figure 2. Kaplan-Meier failure curves (hazards) for first event of discontinuation and cessation of
antidepressant therapy after index.
ment cessation at three months after index, 33% at 6
months, 46% at 12 months, 59% at 24 months, and 69%
at 48 months after index.
When considering the entire period of follow-up, over-
all incidence of discontinuation of the index SSRI was
51.23 per 100 person years (95% CI 50.23 - 52.25); any
SSRI was 77.45 per 100 person years (95% CI 76.32 -
78.60); and any antidepressant was 80.05 per 100 per-
son years (95% CI 78.94 - 81.17). Incidence of cessation
of all antidepressant therapy over the follow-up period
was 41.83 per 100 person years (95% CI 40.99 - 42.70).
Table 2 shows a multivariable analysis of incidence of
index therapy discontinuation in the first six months.
Patients prescribed fluoxetine at index were 19% more
likely (p 0.001) to discontinue compared with those
starting on citalopram. Patients prescribed paroxetine at
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280
Ta bl e 2 . Multivariable analysis of incidence of discontinuation of first SSRI therapy in the six months after therapy initiation (N =
Variable Events, No Person years at riskUnadjusted incidence rate
ratio (95% CI) Adjusted incidence rate ratio
(95% CI)
Index SSRI
Citalopram 1086 4602.30
Escitalopram 74 367.75 0.85 (0.67 - 1.08) 0.90 (0.71 - 1.14)
Fluoxetine 1103 3999.69 1.17 (1.08 - 1.27)*** 1.19 (1.09 - 1.30)***
Sertraline 24 88.26 1.15 (0.77 - 1.73) 1.06 (0.71 - 1.59)
Paroxetine 111 403.59 1.17 (0.96 - 1.42) 1.25 (1.03 - 1.52)*
Age at index (years) 2398 9461.58 0.99 (0.98 - 0.99)*** 0.99 (0.99 - 0.99)***
Enrolment year
2006 634 3102.05
2007 980 3907.02 1.23 (1.11 - 1.36)*** 1.21 (1.10 - 1.34)***
2008 784 2452.51 1.56 (1.41 - 1.74)*** 1.53 (1.38 - 1.70)***
Body mass index (index value) 2398 9461.58 0.99 (0.98 - 0.99)*** 0.99 (0.98 - 0.99)***
Irritable bowel syndrome (yes) 166 757.81 0.85 (0.73 - 1.00)* 0.85 (0.72 - 0.99)*
***p 0.001; **p 0.01; *p 0.05.
index were 25% (p = 0.03) more likely to discontinue
compared with patients prescribed citalopram at index.
Likelihood of discontinuation declined with increased
age (p 0.001). Discontinuation was more likely among
those initiating therapy in more recent years with a 53%
(p 0.001) and 21% (p 0.001) higher likelihood of
discontinuation respectively among those enrolled in
2008 and 2007 compared with those enrolled in 2006. A
lower likelihood of discontinuation was associated with a
higher BMI (p 0.001), and presence of IBS (p = 0.04).
This study provides much needed UK data on discon-
tinuation of antidepressant therapy among patients with
MDD who were new initiators of treatment with SSRIs.
Discontinuation was assessed as both temporary gaps in
the depression treatment trajectory as well as true cessa-
tion of therapy, since gaps in treatment can indicate in-
adequate or under-treatment of depression. There are
varying definitions of discontinuation of antidepressant
therapy in the literature representing cessation (i.e. no
further antidepressant prescription refills), and discon-
tinuation defined using allowable gaps in prescriptions
of <30 days [25,26], <90 days [27], and <180 days [28].
In this study we opted for a 60 day allowable gap in an-
tidepressant prescribing after which the patient was
deemed to have discontinued.
4.1. Summary of Main Findings
Only a quarter of patients persisted on antidepressant
medication up to at least one year after initiating MDD
treatment, and 15% continued on their index SSRI medi-
cation at least up to this time. A high proportion of pa-
tients had only approximately a month’s duration for
their index therapy, suggesting many patients received a
one-off antidepressant prescription. At six months after
treatment initiation 42% of patients had discontinued
their index drug, and 33% had cessation of therapy with-
out subsequent resumption. Discontinuation of the index
medication showed an increasing trend in more recent
years. Higher discontinuation of the index treatment was
associated with an index medication of fluoxetine or
paroxetine compared with citalopram. Lower disconti-
nuation of index medication was associated with older
age, higher BMI, and IBS.
4.2. Study Cohort
Only 2059 and 13,215 of the 94,932 patients on SSRIs
had a diagnostic Read code entry, and positive depress-
sion screening questionnaire score for MDD respectively.
Since antidepressants are often used to treat several other
disorders (e.g. Anxiety disorders, and eating disorders)
this may account for some of the remaining patients who
were not identified as MDD patients. The relatively
small proportion of the final study cohort defined by a
Read code entry for MDD likely reflects evidence of a
trend in GPs recording symptom rather than diagnostic
depression codes in the UK [29]. However, use of re-
commended [2] validated depression screening question-
naire scores allowed us to identify a population of MDD
patients with high sensitivity and specificity [20].
Copyright © 2012 SciRes. OPEN ACCESS
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280 277
4.3. Comparison with Existing Literature
Other studies have also shown significant early discon-
tinuation of first-time antidepressant therapy, with be-
tween 30% - 53% of MDD patients discontinuing treat-
ment during the six months after initiation [28,30,31];
these percentages are comparable to those observed in
our study. A similar extent of discontinuation (37% -
56%) was observed among those treated for a new epi-
sode of depression, but who were not necessarily new
initiators of therapy [32-34].
The reasons for discontinuation are manifold. The
most common reason for discontinuation given by pa-
tients in one study was “feeling better” [30]. Another
study showed compliance with antidepressant treatment
was twice as high among patients with high levels of
polypharmacy compared with those with lower levels of
polypharmacy [33]. Therefore, implementation of alter-
native strategies for inadequate responders to antide-
pressant therapy may enhance compliance and reduce the
likelihood of discontinuation. Aikens et al. (2005) [35]
found patient skepticism towards antidepressant therapy
was a strong predictor of discontinuation. Medication
side-effects, the most common of which are headache
and nausea [9], have also been shown to be associated
with premature discontinuation of antidepressants [9,
Our multivariable analysis assessed factors associated
with discontinuation of index therapy in the six months
after therapy initiation. We chose this time window since
it represented early discontinuers, and patients clearly
falling short of the recommended six months of antide-
pressant maintenance treatment post remission [2].
As in our study, other studies have also found older
age to be associated with higher compliance [33] and
longer-term [36] antidepressant therapy. Patient beliefs
about medication are an important predictor of therapy
adherence [37]. Another study by Aikens et al. (2008) [38]
on patient beliefs about antidepressants found older pa-
tients were more likely to perceive a need for treatment
and hence were less likely to discontinue therapy. Fur-
thermore, patients with a relatively low comorbidity bur-
den were likely to report an attitude of perceived harm-
fulness related to antidepressant therapy [38], which may
explain our findings of an association between presence
of a comorbid condition (IBS) and a lower likelihood of
discontinuation. Moreover, a higher comorbidity burden
is likely to be associated with more contact with the GP,
which may also increase the opportunity for treatment
for depression.
Higher BMI was associated with a lower likelihood of
discontinuation. This may be explained by the fact that
patients remaining on SSRI treatment can be likely to
suffer weight gain as a side effect [39]. However, there is
also evidence that weight gain is associated with lower
satisfaction and lower adherence to medication [40] and
may indeed be a long-term predictor of discontinuation
beyond the six-month window we examined.
The results also suggest an increasing trend in discon-
tinuation of antidepressant therapy with more recent
study enrolment years; this points to the need for imple-
mentation of better adherence strategies and improved
patient-doctor communication. Another possible expla-
nation could be a negative perception about antidepres-
sant efficacy resulting from research showing limited
difference between antidepressants and placebo [41,42].
Patients prescribed paroxetine or fluoxetine at index
were more likely to discontinue in the first six months
compared with patients prescribed citalopram. This may
suggest relatively higher intolerability and adverse events
associated with paroxetine and fluoxetine compared with
citalopram [43,44], however there is little substantive
evidence for differences in overall intolerability between
SSRIs among non-select patients groups [45]. Further-
more, since patients were not randomly assigned to treat-
ment, differences in unmeasured characteristics between
patients prescribed citalopram, fluoxetine and paroxetine
may explain the differing discontinuation rates.
4.4. Study Strengths and Limitations
1) It is possible that MDD patients without a diagnos-
tic Read code entry or a depression screening question-
naire score entry indicating MDD were missed;
2) We were not able to identify possible antidepressant
prescriptions received from mental health specialists.
However, we would expect a relatively small proportion
of patients, mainly those who were treatment-resistant, to
have been likely to be under the sole care of a specialist
rather than their GP;
3) Reasons for discontinuation were not available in
GPRD. Future studies using GPRD may utilize addi-
tional free text information which may provide such in-
4) GPRD contains information on drugs prescribed
and does not verify whether the medication was dis-
pensed or taken by the patient;
5) The results cannot be generalized to MDD patients
who are not new initiators of antidepressant therapy,
since their treatment course may differ from those on
first time antidepressant therapy;
6) It was beyond the scope of the study to examine
psychological therapies; future studies may consider this
to determine whether patients who discontinue pharma-
cological treatment early are likely to be receiving psy-
chological therapies instead.
There is a significant risk of discontinuation of antide-
Copyright © 2012 SciRes. OPEN ACCESS
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280
pressant therapy in the six months after initiation of
treatment for MDD. This requires awareness by the cli-
nician both in terms of ensuring an optimal treatment
regimen is implemented, and also in helping minimize
non-compliance from MDD patients in persisting with
their prescribed therapy [31]. GPs should monitor their
MDD patients initiating pharmacotherapy for presence of
depressive symptoms, and adhere to recommended
guidelines by continuing antidepressant therapy for up to
at least six months after remission to avoid risk of re-
lapse of depression. Moreover, the importance of con-
tinuing treatment even if their depressed mood has im-
proved should be appropriately communicated to the
patient. GPs should aim to minimise discontinuation and
optimise adherence to antidepressant therapy by active
follow-up of patients suffering potential inadequate or
non-response to pharmacotherapy, and those experienc-
ing adverse side effects from their antidepressant medi-
cation. If such problems are identified, GPs should im-
plement alternative treatment strategies such as a dose
increase, therapy switch, augmentation or referral to a
[1] King, M., Nazareth, I., Levy, G., Walker, C., Morris, R.,
Weich, S., Bellon-Saameno, J.A., Moreno, B., Svab, I.,
Rotar, D., Rifel, J., Maaroos, H.I., Aluoja, A., Kalda, R.,
Neeleman, J., Geerlings, M.I., Xavier, M., de Almeida,
M.C., Correa, B. and Torres-Gonzalez, F. (2008) Preva-
lence of common mental disorders in general practice at-
tendees across Europe. The British Journal of Psychiatry,
192, 362-367. doi:10.1192/bjp.bp.107.039966
[2] NICE (2009) Depression: The treatment and management
of depression in adults (Clinical Guideline 90). National
Institute for Health and Clinical Excellence, London.
[3] Claxton, A.J., Li, Z. and McKendrick, J. (2000) Selective
serotonin reuptake inhibitor treatment in the UK: Risk of
relapse or recurrence of depression. The British Journal
of Psychiatry, 177, 163-168. doi:10.1192/bjp.177.2.163
[4] Judd, L.L., Paulus, M.J., Schettler, P.J., Akiskal, H.S.,
Endicott, J., Leon, A.C., Maser, J.D., Mueller, T., Solo-
mon, D.A. and Keller, M.B. (2000) Does incomplete re-
covery from first lifetime major depressive episode herald
a chronic course of illness? American Journal of Psy-
chiatry, 157, 1501-1504. doi:10.1176/appi.ajp.157.9.1501
[5] Judd, L.L., Akiskal, H.S., Maser, J.D., Zeller, P.J., Endi-
cott, J., Coryell, W., Paulus, M.P., Kunovac, J.L., Leon,
A.C., Mueller, T.I., Rice, J.A. and Keller, M.B. (1998)
Major depressive disorder: A prospective study of resid-
ual subthreshold depressive symptoms as predictor of
rapid relapse. Journal of Affective Disorders, 50, 97-108.
[6] Mitchell, A.J. and Selmes, T. (2007) Why don’t patients
take their medicine? Reasons and solutions in psychiatry.
Advances in Psychiatric Treatment, 13, 336-346.
[7] Fortney, J.C., Pyne, J.M., Edlund, M.J., Stecker, T., Mit-
tal, D., Robinson, D.E. and Henderson, K.L. (2011) Rea-
sons for antidepressant nonadherence among veterans
treated in primary care clinics. Journal of Clinical Psy-
chiatry, 72, 827-834. doi:10.4088/JCP.09m05528blu
[8] Rush, A.J., Trivedi, M.H., Wisniewski, S.R., Nierenberg,
A.A., Stewart, J.W., Warden, D., Niederehe, G., Thase,
M.E., Lavori, P.W., Lebowitz, B.D., McGrath, P.J.,
Rosenbaum, J.F., Sackeim, H.A., Kupfer, D.J., Luther, J.
and Fava, M. (2006) Acute and longer-term outcomes in
depressed outpatients requiring one or several treatment
steps: A STAR*D report. American Journal of Psychiatry,
163, 1905-1917. doi:10.1176/appi.ajp.163.11.1905
[9] Anderson, H.D., Pace, W.D., Libby, A.M., West, D.R.
and Valuck, R.J. (2012) Rates of 5 common antidepres-
sant side effects among new adult and adolescent cases of
depression: A retrospective US claims study. Clinical
Therapeutics, 34, 113-123.
[10] Kikuchi, T., Suzuki, T., Uchida, H., Watanabe, K. and
Kashima, H. (2011) Subjective recognition of adverse
events with antidepressant in people with depression: A
prospective study. Journal of Affective Disorders, 135,
347-353. doi:10.1016/j.jad.2011.07.011
[11] Moore, M., Yuen, H.M., Dunn, N., Mullee, M.A., Mas-
kell, J. and Kendrick, T. (2009) Explaining the rise in an-
tidepressant prescribing: A descriptive study using the
general practice research database. British Medical Jour-
nal, 339, b3999. doi:10.1136/bmj.b3999
[12] Wade, A.G., Saragoussi, D., Despiegel, N., Francois, C.,
Guelfucci, F. and Toumi, M. (2010) Healthcare expendi-
ture in severely depressed patients treated with escitalo-
pram, generic SSRIs or venlafaxine in the UK. Current
Medical Research & Opinion, 26, 1161-1170.
[13] Kurd, S.K., Troxel, A.B., Crits-Christoph, P. and Gelfand,
J.M. (2010) The risk of depression, anxiety, and suicidal-
ity in patients with psoriasis: A population-based cohort
study. Archives of Dermatology, 146, 891-895.
[14] Becker, C., Brobert, G.P., Johansson, S., Jick, S.S. and
Meier, C.R. (2011) Risk of incident depression in patients
with Parkinson disease in the UK. European Journal of
Neurology, 18, 448-453.
[15] Lewis, J.D., Bilker, W.B., Weinstein, R.B. and Strom,
B.L. (2005) The relationship between time since registra-
tion and measured incidence rates in the General Practice
Research Database. Pharmacoepidemiology and Drug
Safety, 14, 443-451. doi:10.1002/pds.1115
[16] Chisholm, J. (1990) The Read clinical classification. Bri-
tish Medical Journal, 300, 1092.
[17] Kroenke, K., Spitzer, R.L. and Williams, J.B.W. (2001)
The PHQ-9: Validity of a brief depression severity meas-
ure. Journal of General Internal Medicine, 16, 606-613.
Copyright © 2012 SciRes. OPEN ACCESS
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280 279
[18] Zigmond, A.S. and Snaith, R.P. (1983) The Hospital An-
xiety and Depression Scale. Acta Psychiatrica Scandi-
navica, 67, 361-370.
[19] Beck, A.T., Steer, R.A. and Brown, G.K. (1996) Manual
for the beck depression inventory-II. Psychological Cor-
poration, San Antonio.
[20] Gilbody, S., Richards, D., Brealey, S. and Hewitt, C.
(2007) Screening for depression in medical settings with
the Patient Health Questionnaire (PHQ): A diagnostic
meta-analysis. Journal of General Internal Medicine, 22,
1596-1602. doi:10.1007/s11606-007-0333-y
[21] Elliott, W.J., Plauschinat, C.A., Skrepnek, G.H. and
Gause, D. (2007) Persistence, adherence, and risk of dis-
continuation associated with commonly prescribed anti-
hypertensive drug monotherapies. The Journal of the
American Board of Family Medicine, 20, 72-80.
[22] Charlson, M.E., Pompei, P., Ales, K.L. and MacKenzie,
C.R. (1987) A new method of classifying prognostic co-
morbidity in longitudinal studies: Development and vali-
dation. Journal of Chronic Diseases, 40, 373-383.
[23] Khan, N.F., Perera, R., Harper, S. and Rose, P.W. (2010) Ada-
ptation and validation of the Charlson Index for Read/
OXMIS coded databases. BMC Famil y Pract ice, 11, 1.
[24] Warner-Schmidt, J.L., Vanover, K.E., Chen, E.Y., Mar-
shall, J.J. and Greengard, P. (2011) Antidepressant effects
of selective serotonin reuptake inhibitors (SSRIs) are at-
tenuated by anti-inflammatory drugs in mice and humans.
Proceedings of the National Academy of Sciences of the
United States of America, 108, 9262-9267.
[25] Lee, Y.M. and Lee, K.U. (2011) Time to discontinuation
among the three second-generation antidepressants in a
naturalistic outpatient setting of depression. Psychiatry
and Clinical Neurosciences, 65, 630-637.
[26] Olfson, M., Marcus, S.C., Tedeschi, M. and Wan, G.J.
(2006) Continuity of antidepressant treatment for adults
with depression in the United States. American Journal of
Psychiatry, 163, 101-108. doi:10.1176/appi.ajp.163.1.101
[27] Simon, G.E., Von Korff, M., Wagner, E.H. and Barlow,
W. (1993) Patterns of antidepressant use in community
practice. General Hospital Psychiatry, 15, 399-408.
[28] Hansen, D.G., Vach, W., Rosholm, J.U., Sondergaard, J.,
Gram, L.F. and Kragstrup, J. (2004) Early discontinua-
tion of antidepressants in general practice: Association
with patient and prescriber characteristics. The Journal of
Family Practice, 21, 623-629.
[29] Rait, G., Walters, K., Griffin, M., Buszewicz, M., Peter-
sen, I. and Nazareth, I. (2009) Recent trends in the inci-
dence of recorded depression in primary care. The British
Journal of Psychiatry, 195, 520-524.
[30] Demyttenaere, K., Enzlin, P., Dewe, W., Boulanger, B.,
De, B.J., De, T.W. and Mesters, P. (2001) Compliance
with antidepressants in a primary care setting, 1: Beyond
lack of efficacy and adverse events. Journal of Clinical
Psychiatry, 62, 30-33.
[31] Lin, E.H., Von, K.M., Katon, W., Bush, T., Simon, G.E.,
Walker, E. and Robinson, P. (1995) The role of the pri-
mary care physician in patients’ adherence to antidepres-
sant therapy. Medical Care , 33, 67-74.
[32] Lewis, E., Marcus, S.C., Olfson, M., Druss, B.G. and Pin-
cus, H.A. (2004) Patients’ early discontinuation of anti-
depressant prescriptions. Psychiatric Services, 55, 494.
[33] Serna, M.C., Cruz, I., Real, J., Gasco, E. and Galvan, L.
(2010) Duration and adherence of antidepressant treat-
ment (2003-2007) based on prescription database. Euro-
pean Psychiatry, 25, 206-213.
[34] Mabotuwana, T., Warren, J., Orr, M., Kenealy, T. and
Harrison, J. (2011) Using primary care prescribing data to
improve GP awareness of antidepressant adherence issues.
Informatics in Primary Care, 19, 7-15.
[35] Aikens, J.E., Kroenke, K., Swindle, R.W. and Eckert, G.J.
(2005) Nine-month predictors and outcomes of SSRI an-
tidepressant continuation in primary care. General Hos-
pital Psychiatry, 27, 229-236.
[36] Meijer, W.E., Heerdink, E.R., Leufkens, H.G., Herings,
R.M., Egberts, A.C. and Nolen, W.A. (2004) Incidence
and determinants of long-term use of antidepressants.
European Journal of Clinical Pharmacology, 60, 57-61.
[37] Horne, R. (1999) Patients’ beliefs about treatment: The
hidden determinant of treatment outcome? Journal of Psy-
chosomatic Research, 47, 491-495.
[38] Aikens, J.E., Nease Jr., D.E. and Klinkman, M.S. (2008)
Explaining patients’ beliefs about the necessity and harm-
fulness of antidepressants. The Annals of Family Medi-
cine, 6, 23-29. doi:10.1370/afm.759
[39] Hirschfeld, R.M. (2003) Long-term side effects of SSRIs:
Sexual dysfunction and weight gain. Journal of Clinical
Psychiatry, 64, 20-24.
[40] Fakhoury, W.K., Wright, D. and Wallace, M. (2001) Pre-
valence and extent of distress of adverse effects of antip-
sychotics among callers to a United Kingdom National
Mental Health Helpline. International Clinical Psycho-
pharmacology, 16, 153-162.
[41] Kirsch, I., Deacon, B.J., Huedo-Medina, T.B., Scoboria,
A., Moore, T.J. and Johnson, B.T. (2008) Initial severity
and antidepressant benefits: A meta-analysis of data sub-
mitted to the Food and Drug Administration. PLOS Me-
dicine, 5, e45. doi:10.1371/journal.pmed.0050045
[42] Kirsch, I. (2009) Antidepressants and the placebo re-
sponse. Epidemiologia e psichiatria sociale, 18, 318-322.
[43] Thase, M.E., Ferguson, J.M., Lydiard, R.B. and Wilcox,
Copyright © 2012 SciRes. OPEN ACCESS
S. Davé et al. / Open Journal of Psychiatry 2 (2012) 272-280
Copyright © 2012 SciRes.
C.S. (2002) Citalopram treatment of paroxetine-intolerant
depressed patients. Depression and Anxiety, 16, 128-133.
[44] Rambelomanana, S., Depont, F., Forest, K., Hebert, G.,
Blazejewski, S., Fourrier-Reglat, A., Molimard, M. and
Moore, N. (2006) Antidepressants: General practitioners’
opinions and clinical practice. Acta Psychiatrica Scandi-
navica, 113, 460-467.
[45] Edwards, J.G. and Anderson, I. (1999) Systematic review
and guide to selection of selective serotonin reuptake in-
hibitors. Drugs, 57, 507-533.