Transcript Course and outcome of Depressive illness

Course and outcome of
Depressive illness
Rajini Ramana
Illness phases
REMISSION- the point at which improvement of sufficient
magnitude is observed, implying that no further increase in
treatment levels is required.
RECOVERY- a period of remission of an episode lasting for a
sufficient period, such that further treatment can be
discontinued or aimed at preventing a subsequent episode.
RELAPSE-return of symptoms meeting full syndrome criteria
during a period of remission ( part of the same episode).
RECURRENCE- appearance of a new episode following
recovery from a previous episode.
Treatment phases of depression
Kupfer DJ. J Clin Psychiatry 1991;52 (5, Suppl): 28–34.
Course of Unipolar Depression




Historically considered to be an episodic illness,
following a relapsing- remitting course or a
chronic unremitting course.
Evidence from a variety of studies show that this
is not the case in a significant proportion of
sufferers.
Many have sub-syndromal depression in between
episodes or a chronic dysthymic course
with/without full blown recurrences.
This level of morbidity does contribute to longer
term morbidity and mortality as well as disability .
Key periods in outcome research



Kraepelin felt that the longer term outcome was
favourable, despite the episodic nature.
Pre treatment era studies focussed on in-patients in
asylums, most methodologically flawed due to
variability in diagnostic criteria . Poor global
outcomes, but probably due to sample population (
Rennie 1942, Poort 1945, Stenstedt 1952).
Post- treatment era outcome ought to be better- but
is it?
Key early post-treatment studies
( pre- maintenance phase era)



Series of papers from the NIMH Psychobiology of
depression project over the last two decades,
highlighting chronicity, high rates of recurrence
and low levels of treatment in well designed
prospective studies ( Keller et 1984, Keller et al
1992).
Sydney ( Kiloh et al,1988) and Maudsley ( Lee &
Murray 1988) and Edinburgh (Surtees and
Barkley 1994).
All studied very long term outcome ( 12-18 years)
and assessed symptom remission and outcomes as
well as disability.
Why were these studies taken
more seriously?

Good methodology:
–
–
–
–
Clear and standardised diagnostic assessments
Comprehensive follow-up
Standardised definitions for outcome
Use of appropriate statistical tools ( eg:survival analysis
to account for attrition of initial sample)
– More representative sample, as far as was possible.
– Better follow- up techniques for both retrospective and
prospective studies.

Longer periods of follow- up.
What are residual symptoms?




Most research on treatment outcomes use
dichotomous measures of acute treatment outcome
( HDRS, BDI,MADRS).
Almost invariably some symptoms persist after
‘response’ to treatment.
The ‘subclinical nature of persistent symptoms
may not warrant a Dx of a new episode.
However, often cause significant subjective
distress and are often risk factors for a subsequent
episode. ( Judd et al 1998, Paykel et al 1995)
What happens after the first severe episode of
treated depression?




Judd et al ( 2000), in the NIMH Collaborative program on
the psychobiology of depression, studied 122 patients in
their first ever episode of major depression for 9 years.
All white, IQ>70, spoke English and had no co-morbid
conditions or substance misuse.
26 had incomplete recovery and 70 had full recovery.
Group with incomplete recovery had
– significantly more severe and chronic course of illness
– more chronic major depressive episodes (more than 2
years), shorter well intervals, and far fewer weeks free
of depressive symptoms over a 12 year follow- up
period.
– median weeks to new episode 33 vs 184.
What happens to patients with residual
symptoms after their first episode?
Judd et al 2000

Recovery with residual symptoms is a strong ,
reliable clinical marker of rapid and frequent
relapse to depressive episodes:
– More severe and chronic illness course
– More episodes of MDD and chronic depressive
episodes
– Shorter periods of inter-episode recovery
– Far fewer weeks free of depressive symptoms

Symptom variability during maintenance Rx is
associated with a higher risk of recurrence ( Karp
et al 2004).
Cambridge Cohort study



70 patients ( 76% inpatients) with RDC MDD,
consecutive, no co-morbid conditions.
Interviewed at 3 monthly intervals for 15 months
to assess relapse rates and for 15 months thereafter
to assess early relapse rates. ( Ramana et al 1995,
Paykel et al 1995).
Followed up at 8-11 years- data included
longitudinal information on course of depression
and other psychiatric disorders, pharmacological
and psychological treatment and social
functioning. ( Kennedy et al 2003).
Cambridge Cohort study 1




70% remitted in 6 months, only 6% failed to do so by 15
months.
40% relapsed over the subsequent 15 months, with all
relapses occurring in the first 10 months. Greater severity
of index episode predicted relapse.
Residual symptoms (RS) reaching HDRS(17) of 8 or more
were present in 32% of those who remitted.
RS were very strong predictors of early relapse ( 76% of
those with RS relapsed compared to 25% of those without
RS).
Cambridge Cohort Study 2







Over the long term 99% of patients were followed up, and 61 of 66
living subjects were interviewed, detailed follow- up data obtained in
65 subjects.
60 of 65 subjects recovered during follow-up, but two- thirds had a
recurrence.
11 ( 17%) had a chronic episode during follow-up and 18% never
achieved asymptomatic status.
Greater severity of initial episode and number of previous episodes
were the most consistent predictors of poor outcome.
Female gender predicted chronicity and a longer time spent at full
criteria for depression.
Social functioning was good and there were high levels of
pharmacological and psychological treatment.
Times to recovery and recurrence and recovery and recurrence rates
were similar to previous studies.
Comparisons with premaintenance phase studies




Lee & Murray 1988, London ( N=89).
Kiloh et al 1988, Sydney ( N =133).
Surtees & Barkley 1994, Edinburgh ( N=80).
NIMH Collaborative Study, US sites ( N=431)
(Keller et al 1992, Mueller et al 1999).
– Similar subjects.
– Higher levels of psychological and pharmacological
treatment and treatment concordance in Cambridge
study.
– Recovery and recurrence rates similar to older studies.
– Better psychosocial and global outcomes in the
Cambridge study.
Is the outcome better in less
severely ill cohorts?



Kanai et al 2003 studied an untreated Japanese
cohort, 70% of whom were in their first episode.
95/1843 patients screened from a variety of
participating centres ( GLAD project) ( Furukawa
2000).
Prospective follow up with a slightly high attrition
rate of a moderate to severely ill cohort.
Better outcomes with less chronicity and
recurrence and a median well interval of > 72
weeks.
Why do we need to take residual
symptoms seriously?

Burden of depression:
– Largest single cause of disability in the minority world
and fourth cause of worldwide disability, likely to
become second by 2020 (Murray et al, 1996).


High cost of untreated depression ( less than 10%
of this is direct cost of treatment).
Patients with residual sub-threshold symptoms
during recovery relapse more frequently ( Paykel
et al 1995) and about five times faster than
patients with full remission ( Judd et al 1998).
Or do we need to take them
seriously at all?



Mojtabai (2001) reported findings from the National Comorbidity Survey.
– nationally representative household sample
– Respondents categorised according to time since last
episode of major depression vs respondents without a
history of major depression.
Residual symptoms persist long after the episode resolves.
Impairment resolved shortly after the episode resolved
(only those still in the episode had more lost days than
cutback days or difficult days), and was similar to those
without previous episodes of depression within 6- 12
months or resolution of the episode.
Is the burden of depression avoidable by long
–term treatment strategies?




Van Os et al (2005) used a ‘modelling study’ to
assess the impact of Rx on depression.
Describe the epidemiology of depression and
current health service use in adults in 2000.
Evaluate the impact of effective treatments by
translating outcome measures from meta-analyses
of trials into a change in DALY.
Compare the amount of treatment experienced
under current and and expanded treatment options
vs the hypothetical disease burden in the absence
of treatment.
Is the burden of depression avoidable by long
–term treatment srategies?



Evaluated drug treatment for episodes of major
depression+ continuation phase + maintenance
phase for 5 years after remission; CBT treatment
of episodes and a maintenance variant of CBT
with booster sessions over a 5 year period.
Used DALY to assess the health benefit of these
interventions.
Translated Effect Sizes of each of the
interventions into a reduction in disability weights.
Is the burden of depression avoidable by long
–term treatment srategies?




Current episodic treatment averts 9% (95% uncertainty
interval, 6%-12%) of the disease burden of major
depression in Australian adults.
Optimal episodic treatment with CBT could avert 28%
(95% uncertainty interval, 19%-39%) of this disease
burden, and with drugs 24% (95% uncertainty interval,
19%-30%) could be averted.
During the 5 years after an episode of major
depression,current episodic treatment patterns would
avert 13% (95% uncertainty interval, 10%-17%) of DALY.
Maintenance drug treatment could avert 50% (95%
uncertainty interval, 40%-60%) and maintenance cognitive
behavioural therapy could avert 52% (95% uncertainty
interval, 42%-64%),even if adherence of around 60% is
taken into account.
Evidence about efficacy of longterm treatment






Geddes et al (2003) systematic overview of evidence from
RCTs of CT in patients who had responded to acute
treatment.
31 RCTs and 4410 patients.
CT with antidepressants reduced the odds of depressive
relapse by 2/3rds ( roughly halving the absolute risk).
Treatment effect was similar across different classes of
antidepressants.
Risk of relapse was similar across heterogeneous groups of
patients.
For patients still at appreciable risk of recurrence after 4–6
months of CT antidepressants, another year of CT will
approximately halve their risk.
Predictors of outcome in the
elderly

White matter hypersensitivities (WMH) indicate a
poorer prognosis.
– Subcortical structures and their frontal projections.



Ischaemic lesions of the L frontal lobe and caudate
head often lead to depression.
Executive dysfunction predicts chronicity as well
as relapse and recurrence.
Memory impairment does not predict chronicity or
recurrence, or indeed response to antidepressants.
To summarise



Good evidence of response to pharmacological
and psychological treatment in the short –term
BUT
High rates of chronicity (20%) in both recurrent
and non recurrent groups.
Effective/prophylactic treatment strategies only
postpone the risk of a recurrence, with the risk of
recurrence returning to pre-treatment levels within
a year of discontinuing treatment.
Best predictors of poor outcome are number of
previous episodes and residual symptoms.
Outcome of treatment in
depression.
“Failure to remit, delayed remission,
partial remission with considerable
residual symptoms, relapse and
recurrence are common outcomes of
the depressive illness.
Paykel (1994)