Methodological considerations regarding Joyce et al. 2016
Baethge et al. Int J Bipolar Disord
Methodological considerations regarding Joyce et al. 2016
Christopher Baethge cbaethge@uni‑koeln.de 0 2
Leonardo Tondo 1
Ross J. Baldessarini 1
0 Department of Psychiatry and Psychotherapy, University of Köln , Cologne , Germany
1 Department of Psychiatry, Harvard Medical School , Boston, MA , USA
2 Department of Psychiatry and Psychotherapy, University of Köln , Cologne , Germany
The recent review by Joyce et al. (2016) concludes that
early intervention and treatment of bipolar disorder (BD)
patients were consistently superior in 10 studies selected
for analysis, largely involving short-term responses
following an index episode of illness. We agree that there
are compelling clinical reasons to encourage timely
identification and treatment of such patients in efforts to limit
long-term morbidity. However, we reported previously,
based on 39 published reports plus original data from
a large international sample, that neither latency from
illness-onset nor the number of illness episodes before
initiating treatment was related to long-term morbidity
(percent of time ill or episodes/year) in BD subjects
during 3.8–5.4 years of treatment (Baethge et al. 2003a; Bratti
et al. 2003). Most of these studies were not included in
the review by Joyce et al. (2016). We considered all
subjects treated at various times or following various
numbers of recurrences, and found that long-term morbidity
following initial treatment did not vary significantly with
treatment-delay or episode-count.
One possible explanation for the differences in
conclusions arising in these studies may be caused by the lack
of search terms more specific to the pertinent
literature, such as “delay” (Baethge et al. 2003b) or “latency”
(Baldessarini et al. 2003) in the overview by Joyce and
Also important is a methodological problem in
several studies referred to by Joyce et al.: comparing
treatment response following early versus late interventions
is likely to compare clinically dissimilar sub-populations.
Early intervention groups include patients with a range of
prognoses and illness severities, whereas later
intervention involves subjects who have experienced illness
recurrences and may have been exposed to various periods
of active, failed, and discontinued treatment. That is,
it seems likely that samples with a range of prognoses
were compared to generally less favorable samples; such
comparisons would be expected to yield more favorable
initial treatment responses among early intervention
subjects, as was found. As examples, this confounding
factor pertains to studies by Franchini et al. (1999) and Keck
et al. (1995).
Other methodological problems include use of
multiple outcome measures not adjusted for multiple
comparisons (e.g., Colom et al. 2010), as well as use of cut-off
points selected retrospectively with risk of false-positive
findings (e.g., Swann et al. 1999). The fact that
studies used different treatments and that some investigated
maintenance treatment whereas others focused on acute
treatment introduces substantial heterogeneity and adds
to the considerable difficulties in interpreting the results.
A powerful potential motivating factor favoring early
intervention in BD would be a “progressive” course, in
which more-or-less euthymic intervals usually become
shorter with a growing number of illness recurrences.
This hypothesis, while plausible at first sight, has been
considered repeatedly over the past century. However,
evidence supporting it has been inconsistent and
generally unfavorable; moreover, potential sampling bias has
been noted when subjects unmatched for recurrence
counts are compared (Oepen et al. 2004; Baldessarini
et al. 2012).
In conclusion, we note that patients treated early in
the course of BD are likely to include a proportion with
relatively favorable responses to treatment, and fewer
favorable cases among those with a longer history of
illness and variable responses to treatment. Our previous
findings also support the impression that some morbidity
in BD remains during long-term treatment but that it is
not worse after longer delay of treatment or more
previous illness recurrences. This conclusion runs counter to
current expectations that early therapeutic intervention
in BD might lead to long-term reductions of morbidity
© The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license,
and indicate if changes were made.
and disability superior to what can be achieved by later
interventions. Again, this hypothesis is attractive, but it
is far from undisputed (Duffy et al. 2016). On clinical and
ethical grounds, we do strongly encourage timely
recognition and appropriate treatment of BD patients in efforts
to limit risks of morbidity, disability, and mortality which
can arise from prolonged and inadequately treated
illness. However, the hypotheses that BD is generally
progressive and that early treatment intervention can modify
its long-term course should be considered plausible but
unproved at this point.
Baethge C , Baldessarini RJ , Bratti IM , Tondo L. Prophylaxis‑latency and outcome in bipolar disorders . Can J Psychiatry. 2003a; 48 ( 7 ): 449 - 57 .
Baethge C , Smolka MN , Gruschka P , Berghöfer A , Schlattmann P , Bauer M , et al. Does prophylaxis‑ delay in bipolar disorder influence outcome? Results from a long‑term study of 147 patients . Acta Psychitrica Scand. 2003b; 107 : 260 - 7 .
Baldessarini RJ , Tondo L , Hennen J. Treatment‑latency and previous episodes: relationships to pretreatment morbidity and response to maintenance treatment in bipolar I and II disorders . Bipolar Disord . 2003 ; 5 : 169 - 79 .
Baldessarini RJ , Salvatore P , Imaz‑Etxeberria H , Khalsa H‑MK , Gonzalés‑Pinto A , Tohen M. Episode cycles with increasing recurrences in first‑ episode bipolar‑I disorder patients . J Affect Disord . 2012 ; 136 : 149 - 54 .
Bratti IM , Baldessarini RJ , Baethge C , Tondo L. Pretreatment episode count and response to lithium treatment in manic‑ depressive illness . Harv Rev Psychiatry . 2003 ; 11 : 245 - 56 .
Colom F , Reinares M , Pacchiarotti I , Popovic D , Mazzarini L , Martinez‑Aran A , et al. Has number of previous episodes any effect on response to group psychoeducation in bipolar patients? A 5‑ year follow‑up post hoc analysis . Acta Neuropsychiatr . 2010 ; 22 : 50 - 3 .
Duffy A , Mahli GS , Grof P. Do the trajectories of bipolar disorder and schizophrenia follow a universal staging model? Can J Psychiatry 2016 . [ePub ahead of print (4 May)] .
Franchini L , Zanardi R , Smeraldi E , Gasperini M. Early onset of lithium prophylaxis as a predictor of good long‑term outcome . Eur Arch Psychiatry Clin Neurosci . 1999 ; 249 : 227 - 30 .
Joyce K , Thompson A , Marwaha S. Is treatment for bipolar disorder more effective earlier in illness course? A comprehensive literature review . Int J Bipolar Disord . 2016 ; 4 : 19 - 28 .
Keck PE , McElroy SL , Strakowski SM , West SA , Hawkins JM , Huber TJ , et al. Outcome and comorbidity in first‑ compared with multiple‑ episode mania . J Nerv Ment Dis . 1995 ; 183 : 320 - 4 .
Oepen G , Salvatore P , Baldessarini RJ . On the periodicity of manic‑ depressive insanity by Eliot Slater ( 1938 ) : translated excerpts and commentary . J Affect Disord . 2004 ; 78 : 1 - 9 .
Swann AE , Bowden CL , Calabrese JR , Dilsaver SC , Morris DD . Differential effect of number of previous episodes on response to lithium or divalproex in acute mania . Am J Psychiatry . 1999 ; 156 : 1264 - 6 .