Prediction of Treatment Response and Diagnosis in PsychiatryVolume
12 Issue 2 Article 10
June 1995
Prediction of Treatment Response and Diagnosis in Psychiatry
Prediction of Treatment Response and Diagnosis in Psychiatry
Eduardo Dunayevich, MD University of Cincinnati, College of
Medicine, Cincinnati Ohio
Follow this and additional works at:
https://jdc.jefferson.edu/jeffjpsychiatry
Part of the Psychiatry Commons
Let us know how access to this document benefits you
Recommended Citation Recommended Citation Dunayevich, MD, Eduardo
(1995) "Prediction of Treatment Response and Diagnosis in
Psychiatry," Jefferson Journal of Psychiatry: Vol. 12 : Iss. 2 ,
Article 10. DOI: https://doi.org/10.29046/JJP.012.2.006 Available
at: https://jdc.jefferson.edu/jeffjpsychiatry/vol12/iss2/10
This Article is brought to you for free and open access by the
Jefferson Digital Commons. The Jefferson Digital Commons is a
service of Thomas Jefferson University's Center for Teaching and
Learning (CTL). The Commons is a showcase for Jefferson books and
journals, peer-reviewed scholarly publications, unique historical
collections from the University archives, and teaching tools. The
Jefferson Digital Commons allows researchers and interested readers
anywhere in the world to learn about and keep up to date with
Jefferson scholarship. This article has been accepted for inclusion
in Jefferson Journal of Psychiatry by an authorized administrator
of the Jefferson Digital Commons. For more information, please
contact:
[email protected].
Eduardo Dunayevich, M.D.
The lack oj diagnostic gold standards and the multiplicity oj
vantage points used to conceptualize mental illness renders
psychiatric diagnostic and therapeutic approaches susceptible to
individual and cultural biases. Developments in therapeutic
approacheshave oftenprovidedvantage points from which to
conceptualize psychopathology. The identification ojpredictors oj
treatment response may be a useful element for clinicians to select
between multiple treatment tools in an expanding therapeutic
armamentariumiohilefostering refinements in current diagnostic
systems.
A distinct problem of diagn osis in psychi a t ry, as compared to
other branch es of me dicine, is the lack of discr et e anatomical
or physiological correlates of th e various diagn ost ic const
ructs. Diagnosis in psychiat ry is esse n t ially based on
observable ph en om en a (I) and th e sta ndard for d iagnosis in
psychiatry has been two clinicians ag ree ing on a pa rticul a r
diagnosis for a given pa t ien t. Therefore, classifications of
mental illn ess have tended to be influ enced by soci ety's beliefs
regarding th e mind and its illn esses a t th e time th ese
classifications were cre a te d .
An on going dilemma in th e underst anding and classi ficatio n of
psychopathology is the debate about whether ent ities emerge fr om
a continuum with normal psycho logical processes or whether th ey
are discr et e en tit ies that have a d istinct pathophysi ology (
1,2) . This dich ot om y mirrors ot her classic d ich otomies, such
as nature vs. nurture or biological vs. psychosocial (3) . Not
surprising ly, clas sification in psychi a t ry has at times
favored one approach, at others the opposite one . Unfortunately,
th e relat ive weight of biological , psych ological a nd social
variables a nd th erefore th e relevance a tt ribut ed to each of
them in the treatment of mentally ill patients, is far fro m well
dem arca ted , a nd clinicia ns have to rely on the ir own
perceptions (a nd biases) to ge nera te a model for und erst anding
a nd diagnosing. T hus , it is com mo n to find th at th e adhe re
nce to a com pre he nsive biopsych osocial model of psychopathol
ogy is more t heo re t ica l than pract ical. In th is pap er, it
is hypot hesized tha t th e ident ification of indi cat ors of trea
tment respon se ca n enhance current diagnostic systems a nd adva
nce our underst anding of psychopathology.
A major ten et of this essay is th at what appears to be mainly a
th eoretical d iscussion, such as th e one that confronts sup
porters of psychodynamic a nd biological
Eduardo Du nayevich, M.D . is a 3rd year residen t a t th e
Universit y of Cincinnat i Co lleg e of Med icin e in th e Dep
artment of Psychiat ry.
53
54 J EFFERSON JOURNAL OF PSYCHIATRY
models, is actually derived from developm ents in th erapeutics.
Psych oa nalyt ic th eory conce pt ua lized mental illn ess as
forming a con t inu u m with normal psycho log ical functioning (4)
and its etiology often linked to a nomalies of the psychosocial
environ m en t, es pec ia lly during psychological d evelopment in
child hood (5) . This th eory evolved at a time in whi ch th e main
tool available to treat th ese disorders was the person of th e ca
re give r in cha rge, a nd treatment was based on the conduction of
a therapeutic relationship in which th e vicissitudes of this ea
rly psych ologica l deve lop m ent would be replayed and correct
ed. Biological th eori es, a co unterpart to psycho a na lyt ic th
eories , a lso have been spurred by advances in th e rapeut ics
(6,7) . The dis covery of neuroleptics and their specific effec ts
ove r psych otic symptoms was the corne rs tone of th e dopamine
hypothesis in schizophre nia. The use of lithiu m salts in
psychiatry st im ula te d dis cu ssion over th e differential di
agn osis of psych ot ic agitation (8) . The development of sp
ecific treatments for lik ely biological cond it ions, added to th
e limitations of " the ore tica lly assumed psychodynamic fact ors"
(9) were significant influences in favor of psychiatry being thrust
back into th e med ical a nd empirical tradition , whi ch em
phasizes the recognition of di scr ete clinica l en t it ies,
differen ti al di agnosis a nd differential therapeu ti cs.
Specific cr ite r ia we re develop ed to iden tify a nd va lida te
clinical e nt it ies (10 ,11). The remarkable changes between th e
second a nd third versions of th e APA 's DSM are an expression of
th e pro found changes th at took place in th e field of
classification and di agn osis a t that t ime ( 12, 13) .
This ren ewed m edi cal pa radigm, despit e being founded on met
hodologically m ore solid grounds than th e psych osocial one ( I I
) a lso showed significant lim ita t ions. The cons t ruc ts of
schizophrenia a nd m anic depressive illn ess generated by research
di agnostic cr ite ri a tend to se lec t a fairly homogen eou s
group of pa tie nts (for the purpose of obtaining sa m ples th a t
most psych iatrist s will agree re presen t th e illn ess in
question , and that will all ow m eaningful research ) ( 10, 14)
but a lso leave a sign ifica nt population outside th ei r
boundaries. Despi te clear-cut diagnost ic cr iteria, the
differentiation between m anic depression a nd sc hizo p hrenia has
cont inued to present problems to the clinician, eve n wh en fam
ily history of psychia t r ic illn ess , cou rse, and resp on se to
treatment a re conside re d (15) . The ex iste nce of patients who
share phenomenolo gical characte ris t ics of both di agn ostic
groups regarding pr esen tat ion a nd cou rse of illn ess illu st
rates th e limitation s of men tal illness ca tegorizat ion ( 15,
16). If the boundaries between m anic d epressive illn ess a nd
schizophren ia, two of the most robust , di stinct a nd researc hed
co ns t ruc ts in psychi at ry, can still be p ro ble mat ic in so
me cases, it becom es appa re n t th at th e bou ndaries between
other Axis I di sorders ca n become eve n more challenging (2, 15,
I 7, 18). Axis II diso rd ers a re eve n more probl emat ic to d
ifferen ti a te from a ca tegorica l standpoint (19 ,20) as well as
th eir boundaries with Axis I conditio ns (2 1,22) . Already ea rly
psyc hiatrists su ch as Kra ep elin (23) had postulat ed a co n
tinu um between person ali ty disorders a nd th e m ajor psych oses
.
Acknowled gment of the ongoing difficulties in the appropriat e ca
tegor ica l d iagnosis of di fferent clinica l en tities and the
seeming " res istance" of patients to be ca tegorized accord ing to
s ta ndard crite r ia has spawned int e rest in promoting new ways
of approaching and underst a nding clinica l phe no mena. An "
affect ive sp ectrum
PREDICTION OF TREATMENT RESPONSE AND DIAGNO SIS IN PSYCHIATRY
55
disorder" (24) has been postu lat ed to account for th e ex iste
nce of seemingl y dist inct cond itions, which at times are related
through comorbidity or ph en om enology, that appear at higher
frequen cies in th e famili es of individuals affecte d by on e of
th ese cond itions, and tha t share response to ce r tain th erap
eutic int ervent ion s. T he exis tence of disabling conditions
tha t do not fit th e full cr ite ria for DSM disord ers but that
st ill demand atten tion an d treatment has fost ered interest in
the so ca lled "subsyndromal disorders" (8, 17,25) and th eir rel
ationship to th e syndromal on es , with which many times th ey
share re sponsivity to som e th erapeutic man euvers, a lso
pointing to the "soft " boundaries of th e diagnostic
categories.
Diagnost ic systems, like many other th eoretical const ruc ts,
serve the fun ct ion of providing a fr ame, a referen ce that
allows th e understanding and structuring of com plex ph enomena in
an orderly fashion . T hey gen erate a star t ing point from which
further inquiry, dis cussion and eventually act ion, ca n be undert
ak en. T he clini cal situation in psychiatry, with its lack of
"gold standards" ( 1,3) such as reliab le diagnostic tes ts and
pathological findings, and in which th e reality of the encoun te r
between pa tient and physician is dependent on how it is st ruc tu
red by the par tici pants, is on e in whi ch such fram es are high
ly rel ied upon to ge ne ra te "se nse" out of its inherent subj
ect ivity . Fo r some, this subj ectivity is th e raw mat eri al
that allows th e diagnostic and therapeutic act ivity to proceed
(26), even though it may pr even t (at least to some ex te n t) th
e nonparticip ants from full y gras ping th e ongoing process. For
others, subj ectivity is "noise" that interferes with th e task s
of diagnosing and treating an a priori well defined and recognizab
le syndromal enti ty. In the end, it appe a rs that th e diagnost
ic process is heavily influ en ced by th e clinician 's pr e
existing th eoretical and th erapeutic fr amework. T o some ex te n
t , th e th eoreti cal (and th erapeutic) stance of th e individual
clini cian is det ermining th e outco me of the di agnostic process
.
In a ll br an ch es of medicine, a diagn osis is th e sho r tha nd
by which the curre nt understanding of spe cific ph enomena a re
codified a nd how th ey a re to be approach ed (10 ,27). It may
convey dismay, stigma, relief. It conveys what is to be exp ect ed
a nd defin es what psychiatrist s, as agents of soc ie ty, a re
accounta ble for. Dia gnosis is ce n t ra l to medicine as a
scientific ac t ivity. It defin es what th e field of appl ication
of suc h scie n tific knowled ge will be. Sti ll, it see ms th at
in th e field of mental illn ess , th e process of reaching a di
agnosis not only ca n be held sus pect of biases that vary from
practition er to pr actition er, but it has also been shown th at
practitioners th at abide by a give n se t of cr ite r ia for
diagnosis (such as DSM -III -R) do not necessarily use such crite r
ia appropria tely (11 ,27) . And it is also un clear whether the
fas hion in which di agnostic en t it ies a re distingui sh ed from
one anot her is a va lid on e, as shown by unresolved issu es rega
rding diagn ost ic bounda ries.
Psychi atry's ability to ge ne ra te va lid , reliabl e d iagnosis
is a t th e core of its viability as a medi cal spec ia lty . It is
likely th at adva nces in neuroim aging and neuroch emistry will
greatly improve our underst anding a nd classification of ment al
illn ess. But so me of th e answers to qu estion s in diagn osis
see m to lie in the ori gin s of diagnostic syste ms. In a com plex
field such as psychi atry, trying to find a single point of
observation or conce ptua liza t ion to enco m pass a ll of its mul
tidi mensional features
56 JEFFERSON JOURNAL OF PSYCHIATR Y
is probably illusory, as is trying to compare different vant age
points in terms of some objective truth value . As stated by Millon
(I):
Clinical processes a nd events have been described in terms of cond
it ioned habits, reaction formations, cognit ive expectancies or
neu rochemica l dysfunctions . These domains ca nnot be arranged in
a hierarchy, with one level viewed as reducibl e to another.. . .
The yea rn ing among taxonomists for a neat pa ckage of et iologic
attributes sim ply ca nno t be reconciled with th e com plex
philosophical a nd methodological issu es a nd th e d ifficult to
dis entangle networks th at sha pe our mental di sorders. It a lso
ma kes understandable th e decision of th e DSM-III T ask Force to
se t e tiolog ica l a nd course variables aside as clini cal grist
for it s tax oni c mill s. T urn ing from th e anteced ent to th e
conseq ue n t side of th e clinica l course, logic a rgues th at th
e nature of a mental disorder must be a t least parti ally revealed
by its response to treatment. The dat a ava ila ble on this matt er
, however, provides little that goes beyond broad ge nera liza t
ions. T his con t ras ts with medi cin e at large, in whi ch a
variet y of intervent ions are spec ific to particul ar
disorders.
With this view in mind, it ca n be posit ed th at one of th e major
reason s th a t th e adven t of lithium generated so mu ch excite
me n t was becau se it a llowed th e identifi ca t ion of a ce r
ta in gro up of severely impaired individual s who could have a n
imp roved treatment a nd better ou tco me . Even th ough lithium
did not becom e the ac id test for diagn osis of manic depression ,
it is a n exam ple of how th e responsivity to a certain th erap
eutic a pproac h (or th e lack of it ) ca n impact th e who le
process of diagn osis and treatment. It has been said that " the re
are no sicknesses, only sick peopl e" but even if th e di agn osti
c ent it ies of DSM-III-R sho uld be considered as 'pro to typica
l' (2), and th e pursuit of more clear-cut di scr et e ca tegorica
l ent ities in psychia try wou ld reach a stands t ill, th ere will
still be finit e ways of approachin g th em from a th erap eutic s
tand poin t. It mak es as mu ch (or even more) se nse, to look at
psychopathology from th e va n tage point of finit e th erapeutic
tools, th an from th e see ming ly in finit e va ria tions that th
e clini cal pr esentation may di spl ay. Wh et her there is such a
thi ng as a n a ffec t ive spec t rum or not may have grea t th
eoretical importance, but for th e clinicia n it is most important
becau se it identifies clinical ent ities th a t respond to a ce
rtain treatm ent. And for th e resea rche r, it provides indirect
evide nce tha t ca n lead to direct, confirma tory work (6,29). As
in th e mani c depressive pa radigm, as in Freud 's neurotic
patients, a significan t eleme n t is th e likely respo nse to a
particul ar th erap eutic mode. In this con te x t, it follows th
at a n im portan t challe nge fac ing psychi atry a t pr esent
would be identi fyin g ele me nts th at will pred ict respon se to
particul ar treatments.
The st re ng th of a science is not only mani fest ed by its
explana tory power , but particul arly by it s pr edi ctive power.
In this era in which psychiat ry is being viewed less a nd less as
prima ry ca re a nd more of a specia lty tha t is to be accessed
whe n less costly a nd more readily ava ilable mean s have been
exha uste d, a nd recruitment of medical s tu den ts into t he fie
ld dwindles, part of psychiatry's ability to remain a viab le
PREDICTIO N OF TR EATM ENT RESPO NSE AND DIAGNOSIS IN PSYCHIATRY
57
specia lty may lie in it s ability to ge nera te pred ict ion s as
to what th erap eutic tools fro m our ever br oad ening a r mame
ntarium sho uld be institu ted init ia lly, a nd what ste ps a re
to be followed , to ens ure th e most effective trea tmen t. This
applies not on ly for biological , but a lso for psychological a nd
social t reatments. A substant ia l a mou nt of research in this
field has already ge ne ra te d sup port for validating di agnostic
conside ra tions on t he basis of t reatme n t respon se, as shown
by th e research in a typica l sym pto ms in depression (29,30),
neurop sych ological deficits a nd negat ive sym ptoms in schizo
phre nia (3 1,32) a nd cycling cha racteristics in bipolar disorder
(33) between ot he rs . Research in non- biological t reatment s is
also yielding promising results for identifyin g markers of
treatment respon se (34,35). Aside from th eir pr agm atic interest
, markers of treatment re sponse ca n a lso provide an underst and
ing of th e pathophysiology of mental illn ess th at could step
beyond th e curre nt ca tegoriza t ion of psychi atric illn ess
(24,36) . There a re significant obstacles facin g thi s typ e of
research , su ch as identifyin g likely markers a nd designing
adequate pr osp ec tive stud ies that ca n suppo r t or rul e ou t
a putative marker. Mor eover, treatm ent respon se is subject to
multiple variables th at ca n be difficult to con t rol for.
Neverthe less, new light ca n be she d into how th e boundaries
bet ween dis eases are cha rac te r ized accord ing to clini cal
cha racte r istics th at differentia te samples of pat ients who
respond to a pa rticular th erap eu tic mane uve r as opposed to th
e on es who do not. In th is con tex t , th e incr easin g use of
sca les a nd q ues tio nna ires in clini cal pr acti ce, borrowed
many t im es from research applica tions, can serve as attempts to
obtai n st ruct ur ed information and span the gap bet ween the
subjective realit y of th e clini cal inte rview, with its weal th
of in te rpersonal da ta , an d th e possibl e obj ectiv e dist ur
bances of br ai n structu re a nd fun ction. T hey ca n a lso be
th e means through wh ich prospective data ca n be gat hered to
characte rize pr ed ict ors of treat ment respon se.
T he determi nat ion of the cha rac ter ist ics of subgroups of
patients respon sive to specific the ra peu tic approaches might
provide psychiatry with more data to help generate cons istent an d
meaningfu l d iagnost ic syst ems th at will spa n th e dichot omy
between ca te gorica l an d cont inuous conceptualizations of
mental illn ess, incr easin g psychi at ry's ex planatory a nd
predictive powers. It may a lso help psychi at ry maint ain quality
care in an era of managed reso urces.
REFER ENC ES
I. Millon T : C lassification in psychopat ho logy : Ra t ional e,
a lternatives, a nd sta nda rds . J our nal of Abnormal Psychology
1991; 100:245- 261
2. Carson RC : Dile m mas in the pa thway of the DSM -IV. J ourn a
l of Abnorm al Psychology 1991; 100:302-307
3. Blashfie1d RK, Lives leyJ W: Meta phorical analysis of psychiat
ric c1assificat ion as a psycho logica l t esr. Tourna l of Abnorm
a l Psych ology 1991; 100:262- 270
4. Fr eud S: The psychopathology of eve ryday life. Standard
Edition 1960; 6:1- 279, Lond on, Hoga rt h Press
5. Gabbard GO: Psychodyna m ic Psych ia t ry in Clinica l Pr acti
ce 1990. Am eri can Psychiat ric Pr ess, Inc.
58 JEFFERSON JO URNAL OF PSYCHIATRY
6. Barr LC , Goodman WK, Pri ce LH, et a l: The se ro tonin
hypothesis of obs essive com pulsive dis order: Implications of ph
armacologic cha lle nges. J C lin Psych ia try; 53 suppl: 17-
28
7. De Lecuona JM, Joseph KS, Iqb al N, e t a l: The dopamine
hypothesis of schi zoph reni a revisit ed . Psychi atric Annals
1993; 23: 179-1 85
8. Wing JK, Coope r JE, Sartorius N: The m ea surem ent a nd
classificat ion of psych ia tri c symptoms. Cambridge, Cambrid ge
Univers ity Pr ess, 1974
9. Morey LC : C lass ifica t ion of mental disorder as a co llec t
ion of hypothetica l cons tr uc ts. Journal of Abnormal Psychology
1991; 100:289-293
10. FeighnerJP, Robbins E, Guze SB, e t al: Diagn ostic cr iter ia
for use in psych iatric research . Ar ch G en Psychi atry 1972;
26:57-63
II. Widiger TA , Fr an ces AJ, Pin cu s HA , et a l: T oward an
empirical classification for th e DSM-IV.Journal of Abnorma l
Psychology 1991; 100: 280- 288
12. Am erican Psychiatric Association : Diagn osti c a nd sta t ist
ica l ma nual of mental disorders, second edit ion. Washingt on, DC
, 1968
13. Am eri can Psychiat ric Association : Diagn osti c and statis t
ica l ma nual of mental disorders , th ird edit ion . Washingt on,
DC , 1980
14. BoydJH, BurkeJD, G rue nbe rg E, e t al: Excl us ion criteria
of DSM-III. Arch Gen Psychiatry 1984; 4 1:983-989
15. Blacker D , T su an g MT: Co n tes te d boundaries of Bipolar
Disord er and th e lim it s of ca tegorica l diagn osis in psychi
atry. AmJ Psychi atry 1992; 149:1473- 1483
16. Bro ckingt on IF, Melt zer HY : The nosology of schizoaffect
ive psychosis . Psychi atric Devel opme nts 1983; 1:317- 338
17. Liebowitz MR , Hollander E, Schneier F, e t a l: Anxiet y and
depression: Discre te diagnostic ent it ies?J C lin Psych opharm
1990; 10 suppl:6 1S-66S
18. Boul enger JP, Lavall ee YJ: Mix ed anxie ty and dep ression:
Diagnost ic issu es. J C lin Psychi atry 1993; 54 suppl:3-8
19. Fr ances A: The DSM-III person al ity d isord ers sect ion : A
com me nta ry. Am J Psychi atry 1980; 137:1050-1054
20. Livesley JW, Sch ro ed er ML, J ackson DN, et a l: Categorica l
distinct ions in th e study of personality disord er : Im plica t
ions for classification.J ourn a l of Abnorm al Psychol ogy 1994;
103:6-1 7
21. Akis ka l H , H irsch feld RMA, Yereva nian BI: T he relat ions
hip of personality to affec tive disordes . Arch Gen Psychi atry
1983; 40:80 1-810
22. Wid iger TA, Shea T : Differenti a tion of axis I a nd axis II
disorders. Journal of Abn orm al Psychology 1991; 100:399-406
23. Kraepeli n E: C linica l Psychia t ry. New Yor k, Macmill an,
1907 24. Hu dsonJI, Pope H G: Affective spectrum disorder: Does
antidepressant response iden tify a
fami ly of disord ers with a com mon pathophysiology? AmJ Psychi
atry 1990; 147:552- 564 25. Aronson TA : Atypica l anx iety d isor
der: A des criptive study. Compr Psychi at ry 1990;
3 1:152-161 26. Ornst ein A, Ornstein PH: Empathy and th e th
erapeutic di a logu e. Exp anded a nd updat ed
version of a paper first pr esented at th e Fifth Annu al Psych
otherapy Symposiu m on "Psychotherapy: The Therapeutic Dialogu e."
Harvard University, The Cambridge Hos pi ta l, Bost on , Massachu
setts,June 28-30,1984
27. First MB: Com pu ter-assis te d assessmen t of DSM-lII-R
diagnoses. Psychiatric Annals 1994; 24:25-29
28. American Psych ia t ric Associat ion : Diagnost ic a nd stat
ist ica l manua l of men tal disorde rs , third edition-r evised .
Washington, D.C ., 1987
PREDICTION OF TREATMENT RESPO NSE AND DIAGNOSIS IN PSYCHIATRY
59
29. Liebowitz MR, Quitkin FM , St ewartJW, e t a l: Psych opharm
acologic va lidat ion of atypical depression.J C lin Psychiat ry
1984; 45 (7 Pt 2) :22- 25
30. Quitkin FM, H arrison W, Liebowitz M, e t a l: Defining th e
boundaries of atypical depress ion .J C lin Psychiatry 1984; 45 (7
Pt. 2): 19-21
3 1. Convit A, VolavkaJ, Czobor P, e t aI: Effect of subtle
neurological dysfu nction on response to ha lop eridol treatment of
schizophrenia. AmJ Psychi atry 1994; 151:49- 56
32. Goldman RS, Ax elrod BN, T andon R, e t a l: Neuropsychological
pr edi ct ion of treatment efficacy and on e-year outcome in
schizophreni a . Psychopathology 1993; 26: 122-1 26
33 . Keck PE, McE lro y SL: Current persp ectives on treatment of
bipolar d isorder with lit hiu m. Psychiatric Annals 1993;
23:64-69
34. Luborsky L, Crits-Christoph P, MintzJ, e t a l: Who will ben
efit from psychotherapy. New York, Basic Books In c: 269-286,
1988
35. Moras K, Strupp H : Pretherapy int erpersonal rel ations,
patient s' a llia nce , and outcome in Bri ef Ther apy. Ar ch Gen
Psych iatry 1982; 39:405-409
36. McE lroy SL, Hudson J I, Ph illips KA, e t a l: Clinical a nd
th eor etical im plica tions of a possible link between
obsessive-com pulsive and impulse con tro l di sorders. Depression
1993; 1:121-1 32
Prediction of Treatment Response and Diagnosis in Psychiatry
Let us know how access to this document benefits you
Recommended Citation
