Editorial
Personalized psychiatry: many questions, fewer answers
Martin Alda, MD
Department of Psychiatry, Dalhousie University, Halifax, NS, Canada
Clinical practice in many areas of medicine is shifting toward designed for purposes other than response prediction and
personalized treatment. In other words, clinicians aim to treat rely on variables that could be viewed as opportunistic.
patients based on their individual characteristics, including Nevertheless, there are examples of psychiatric treatments
clinical presentation and/or biological markers. Biomarkers that can be selected rationally without resorting to a trial and
can help clinicians select the most effective treatments or re- error approach and that can make a great difference for pa-
duce the risk of side effects by avoiding certain treatments in tients, their families and society.
susceptible patients. In addition, new treatments are being So where in psychiatry can we do better than chance? There
developed for specific patient populations based on better are some promising examples, but they seem to be lost in the
understanding of disease processes associated with identifi- vast sea of DSM 5 diagnostic categories. Disorders that offer the
able markers.1 Advantages of an individualized approach are most promise with respect to effective treatment selection
obvious — patients should receive treatments that are effect- might be periodic catatonia, bipolar disorder responsive to lith-
ive and better tolerated. And in theory, personalization of ium and melancholic depression. These are also among condi-
treatment should lead to lower health care costs. Medical tions where treatment can make a substantial difference with
spending continues to increase in most countries (more or less respect to disability; on the other hand, benefits of treatments
in parallel with increases in life expectancy).2 Personalization for cognitive (neurodegenerative) disorders or substance abuse
should optimize health care spending by selecting treatments are quite limited.3 So it is those relatively few selected treat-
that are effective and avoiding those that are unnecessary. ments indicated in specific patient populations that can pro-
Psychiatry is following the same trends, but its situation is duce excellent results (if applied properly), but that can also be
seemingly more difficult. Psychiatric disorders represent a overlooked easily, perhaps for one of the following reasons.
mix of diagnostic categories of variable validity and predict-
ive value. Practically all medications used to treat psychiatric Low expectations and stigma
disorders have been discovered in a nonsystematic, seren-
dipitous way or have been derived from existing prototypical A factor that needs to be taken into account is the often low ex-
drugs. Most psychiatric conditions are probably heterogen- pectation that some clinicians may have with respect to out-
eous, which makes the search for biomarkers even more im- comes of psychiatric disorders. Such expectations are not infre-
perative, as practically no treatments are universally effective quently based on observations of patients who could do better,
in patients whose conditions are diagnosed within broad and but are “well enough.” I wonder to what extent this is also a
nonspecific categories. At present, we do not have any clinic- consequence of medical and postgraduate training. Residents
ally applicable biomarkers in a narrow sense (laboratory mostly see patients in emergency departments, inpatient units
test), yet there exist clinical findings that might allow person- or chronic care settings, but rarely encounter patients who
alization of treatment in patients with certain psychiatric con- have fully recovered and have more or less resumed normal
ditions. Clues to individualized treatments are hardly rec- lives. In a similar fashion, many clinical trials commonly target
ognized in current treatment guidelines, which are often an arbitrarily defined change on a rating scale, such as 50% im-
based on levels of evidence derived from clinical trials in un- provement over a relatively short period of time, instead of full
differentiated patient samples. Arguably, in most areas of recovery. Such studies may have an heuristic value in un-
psychiatry, predictors of treatment response or side effects covering some basic aspects of treatment response, but their
are simply unavailable. Studies have attempted to find such usefulness in recommending a treatment that would be fully
predictors; however, many of them were derived in studies effective in the long term is low.
Correspondence to: M. Alda, Department of Psychiatry, Dalhousie University, 5909 Veterans’ Memorial Lane, Halifax NS B3H 2E2;
J Psychiatry Neurosci 2013;38(6):363-5.
DOI: 10.1503/jpn.130221
© 2013 Canadian Medical Association
J Psychiatry Neurosci 2013;38(6) 363
Personalized psychiatry: many questions, fewer answers
Martin Alda, MD
Department of Psychiatry, Dalhousie University, Halifax, NS, Canada
Clinical practice in many areas of medicine is shifting toward designed for purposes other than response prediction and
personalized treatment. In other words, clinicians aim to treat rely on variables that could be viewed as opportunistic.
patients based on their individual characteristics, including Nevertheless, there are examples of psychiatric treatments
clinical presentation and/or biological markers. Biomarkers that can be selected rationally without resorting to a trial and
can help clinicians select the most effective treatments or re- error approach and that can make a great difference for pa-
duce the risk of side effects by avoiding certain treatments in tients, their families and society.
susceptible patients. In addition, new treatments are being So where in psychiatry can we do better than chance? There
developed for specific patient populations based on better are some promising examples, but they seem to be lost in the
understanding of disease processes associated with identifi- vast sea of DSM 5 diagnostic categories. Disorders that offer the
able markers.1 Advantages of an individualized approach are most promise with respect to effective treatment selection
obvious — patients should receive treatments that are effect- might be periodic catatonia, bipolar disorder responsive to lith-
ive and better tolerated. And in theory, personalization of ium and melancholic depression. These are also among condi-
treatment should lead to lower health care costs. Medical tions where treatment can make a substantial difference with
spending continues to increase in most countries (more or less respect to disability; on the other hand, benefits of treatments
in parallel with increases in life expectancy).2 Personalization for cognitive (neurodegenerative) disorders or substance abuse
should optimize health care spending by selecting treatments are quite limited.3 So it is those relatively few selected treat-
that are effective and avoiding those that are unnecessary. ments indicated in specific patient populations that can pro-
Psychiatry is following the same trends, but its situation is duce excellent results (if applied properly), but that can also be
seemingly more difficult. Psychiatric disorders represent a overlooked easily, perhaps for one of the following reasons.
mix of diagnostic categories of variable validity and predict-
ive value. Practically all medications used to treat psychiatric Low expectations and stigma
disorders have been discovered in a nonsystematic, seren-
dipitous way or have been derived from existing prototypical A factor that needs to be taken into account is the often low ex-
drugs. Most psychiatric conditions are probably heterogen- pectation that some clinicians may have with respect to out-
eous, which makes the search for biomarkers even more im- comes of psychiatric disorders. Such expectations are not infre-
perative, as practically no treatments are universally effective quently based on observations of patients who could do better,
in patients whose conditions are diagnosed within broad and but are “well enough.” I wonder to what extent this is also a
nonspecific categories. At present, we do not have any clinic- consequence of medical and postgraduate training. Residents
ally applicable biomarkers in a narrow sense (laboratory mostly see patients in emergency departments, inpatient units
test), yet there exist clinical findings that might allow person- or chronic care settings, but rarely encounter patients who
alization of treatment in patients with certain psychiatric con- have fully recovered and have more or less resumed normal
ditions. Clues to individualized treatments are hardly rec- lives. In a similar fashion, many clinical trials commonly target
ognized in current treatment guidelines, which are often an arbitrarily defined change on a rating scale, such as 50% im-
based on levels of evidence derived from clinical trials in un- provement over a relatively short period of time, instead of full
differentiated patient samples. Arguably, in most areas of recovery. Such studies may have an heuristic value in un-
psychiatry, predictors of treatment response or side effects covering some basic aspects of treatment response, but their
are simply unavailable. Studies have attempted to find such usefulness in recommending a treatment that would be fully
predictors; however, many of them were derived in studies effective in the long term is low.
Correspondence to: M. Alda, Department of Psychiatry, Dalhousie University, 5909 Veterans’ Memorial Lane, Halifax NS B3H 2E2;
J Psychiatry Neurosci 2013;38(6):363-5.
DOI: 10.1503/jpn.130221
© 2013 Canadian Medical Association
J Psychiatry Neurosci 2013;38(6) 363
