High Quality
Open the downloaded document, and select print from the file menu (PDF reader required).
Whatever happened to symptom substitution?
Warren W. Tryon
⁎
Fordham University, Department of Psychology, 441 East Fordham Road, Bronx, NY 10458-9993, United States
Received 23 November 2007; received in revised form 28 January 2008; accepted 7 February 2008
Abstract
Symptom substitution is a theoretical consequence of the psychodynamic model of psychopathology and symptom formationthat contrasts markedly with behavioral models. Symptom substitution was a major scientific and clinical question about a half century ago that was abandoned rather than answered partly because it did not seem to occur and partly because perceivedmethodological problems impeded empirical research. The contemporary relevance of symptom substitution is that the psychodynamic model of psychopathology and symptom formation from which the prediction of symptom substitution stemscontinues to be widely taught and to broadly inform clinical practice. This article revisits the psychodynamic model of symptomformation and substitution, restates perceived methodological limitations to its empirical study, demonstrates that symptomsubstitution is an empirically testable prediction, reviews the relevant empirical literature, and discusses implications for the scienceand practice of clinical psychology including a proposed research design for certifying bona fide psychoanalytic symptoms.© 2008 Elsevier Ltd. All rights reserved.
Keywords:
Symptom formation; Symptom substitution; Psychoanalytic symptoms; Behavior therapy; Cognitive therapy; Psychoanalysis
Contents
1. Empirical evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9652. Conclusions and implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 967References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 968
Symptom substitution was once a core clinical and scientific issue for clinical psychology that centered on a crucialdebate concerning the development and treatment of psychopathology
1
“
In the 1950s and 1960s, symptom substitution was raised as a crucial issue that clearly delineated psychodynamic and behavior therapies
”
(p. 349). The major clinical concern was that substituting one symptom for another was not therapeutic and could harm clients if the substituted symptom was worse than the original one. The
Available online at www.sciencedirect.com
Clinical Psychology Review 28 (2008) 963
–
968
⁎
Tel.: +1 718 817 3787 (University office), +1 914 941 0632 (Home Office); fax: +1 718 817 3785 (University fax).
E-mail address:
1
Symptom substitution following medical treatment has also been discussed but is outside the current scope of interest (cf.Browning &Houseworth, 1953).0272-7358/$ - see front matter © 2008 Elsevier Ltd. All rights reserved.doi:10.1016/j.cpr.2008.02.003
major scientific issue was that symptom substitution provided a critical test of the validity of the psychodynamic modelof symptom formation.Grünbaum (1984)concisely presented the theoretical basis for predicting symptom
2
sub-stitution as follows:
“
A neurotic symptom is held to be a compromise
formed
in response to an unresolved conflict between aforbidden unconscious impulse and the ego's defense against it. The symptom is held to be
sustained
at anytime given time by a coexisting, ongoing unconscious conflict, which
—
as claimed by NCT
3
—
does not resolveitself without psychoanalytic intervention. Hence, if the repression of the unconscious wish is not lifted psychoanalytically, the underlying neurosis will persist, even if behavior therapy or hypnosis, for exampleextinguishes the particular symptom that only
manifests
the neurosis at the time. As long as the neurotic conflict does persist, the patient's psyche will call for the defensive service previously rendered by the banishedsymptom. Hence, typically and especially in severe cases, the unresolved conflict ought to engender a
new
symptom. And incidentally, this expectation qualifies as a
“
risky
”
prediction in Popper's sense, since such rivalextant theories as behavior intervention disavow just that expectation
”
(p. 162; italics is original).The stage was set to test a core theoretical prediction that carried important implications for mental health servicedelivery in addition to important scientific implications for psychopathology and symptom formation. A half centurylatter we continue to wait for an answer. So whatever happened to symptom substitution? The evidence suggests that this question was abandoned rather than answered; it simply faded from theoretical discourse and empirical research.Clinical concerns about harm gradually diminished as evidence of successful behavioral treatments accumulated.Empirical research was probably impeded by what was perceived to be the following six methodological problemsidentified byKazdin (2000): 1) The symptoms and causes of psychological disorders do not separate as easily as withmedical illnesses where an underlying pathogen can be readily distinguished from the symptoms it causes. Hence, aclear definition of what constitutes a psychological symptom is unavailable thus making it difficult or impossible toclearly determine if a new symptom has been substituted for a treated one. 2) If a new problem presents after the first one is treated does it automatically and always constitute symptom substitution? Uncertainty here is partlyrelated to the previously mentioned problem of symptom definition. 3) The time course for symptom substitution is unclear. Must symptom substitution occur immediately or can it be delayed for months or years? At what point can one terminatefollow-up study with confidence that if symptom substitution has not yet occurred then it will likely not occur in a particular case? 4) Problems emerge with a base rate probability in untreated populations. If a new problem emergesafter treatment of another problem, how shall we decide if this new problem is a substitute problem or one that wouldhave emerged at this time anyway? 5) If another symptom seems to occur after treatment is it really new or was it present all along and is just more noticeable now that the first problem has been treated? Comorbidity seems to be theclinical rule rather than the exception. 6) Demonstrating that symptom substitution does not occur under someconditions does not prove that it cannot occur under others.Kazdin (2000)concluded
“
These salient issues have ledmany authors to reject symptom substitution as a verifiable concept
”
(p. 351) and by implication constitute reasons whythe question of symptom substitution was abandoned rather than answered. Nevertheless, the connection betweensymptom substitution and the psychodynamic model of symptom formation remains and the validity of the former carries implications for the validity of the latter.The contemporary relevance of the symptom substitution question is that the psychodynamic perspective continuesto be widely taught in both Psy.D. and Ph.D. programs. Freud introduced his psychodynamic model of psycho- pathology, symptom formation, and psychotherapy to America in 1909 during his invited Clark University lectures.This was the only psychological model of mental disorder and it rapidly spread throughout America as the sole psychological basis for clinical training. Thirty-nine years later, the American Psychological Association beganaccrediting clinical training programs in 1948. Programs had to demonstrate that their students were receiving trainingin psychopathology which meant that their training was based on the psychodynamic model of psychopathology and
2
Shevrin, Bond, Brakel, Yertel, and Williams (1996)noted that while dreams and parapraxes (Freudian slips of the tongue) also putatively arisefrom unconscious conflicts,
“…
symptoms are, by definition, maladaptive
”
(p 24).
3
Grünbaum's (1984)referred to the conjunction of the following two causally necessary conditions as Freud's Necessary Condition Thesis(NCT):
“
(1) only the psychoanalytic method of interpretation and treatment can yield or mediate to the patient correct insight into the unconscious pathogens of his psychoneurosis, and (2) the analysand's correct insight into the etiology of his affliction and into the unconscious dynamics of hischaracter is, in turn, causally necessary for the therapeutic conquest of his neurosis (pp. 139-140, italics is original).964
W.W. Tryon / Clinical Psychology Review 28 (2008) 963
–
968
symptom formation. Since then, other models have been introduced but it should not be surprising that manycontemporary clinical psychology training programs continue to articulate the psychodynamic perspective to a greater or lesser degree. The psychodynamic view summarized above, which maintains that underlying psychopathology givesrise to symptoms, continues as a core component of contemporary psychodynamic clinical case formulations. Thisview compels the corollary conviction that effective psychotherapies must address underlying psychological issues inorder to produce lasting results. While the clarity of this clinical conviction once considered behavioral treatments to border on malpractice, concern has lessened considerably for empirical rather than theoretical reasons. However, thisdecreased concern has not altered the psychodynamic model of symptom formation which continues to requiresymptom substitution when core conflicts are not treated. This article intends to inform contemporary discussionregarding the empirical basis of symptom substitution and its consequences regarding the psychodynamic theory of psychopathology and symptom formation.A review of the scientific empirical literature on symptom substitution is relevant if, and only if, symptomsubstitution is an empirically testable hypothesis that can be falsified. Kazdin's six methodological concerns arerevisited with the intention of demonstrating that the prediction of symptom substitution can be empirically tested; i.e.,is falsifiable. With regard to Kazdin's first concern that a clear definition of psychological symptoms is not possible, Inote that symptoms can be clearly separated from hypothesized underlying psychodynamic conflicts and that the
Diagnostic and Statistical Manual of Mental Disorders
(APA, 2000) contains and codifies many psychologicalsymptoms. Accepted methods used to certify psychological symptoms for diagnostic purposes apply with equal forceand validity to certifying new psychological symptoms that might emerge subsequent to the application of behavior therapies. Kazdin's second and fifth points raise concerns over how to interpret the
presence
of new symptoms that might emerge after therapy has begun or finished. On the contrary, it is the
absence
of any new symptoms that is of equal or greater interest here. The absence of new symptoms constitutes definitive negative evidence capable of falsifying the psychodynamic claim regarding symptom formation. Kazdin's third and fourth points concern the timingof the emergence of new symptoms. Short term effects are of more immediate consequence and therefore greater concern than are long delayed and potentially diluted effects. The implied danger to the patient of symptomaticinterventions is based on the assumption that symptom substitution will occur during the time the patient is under thetherapist's care thus rendering the therapist potentially liable for adverse outcomes. Hence, symptom substitution isexpected within a time frame that would generally enable the therapist to know about it. Kazdin's sixth point is that
absence
under one condition does not prove absence under all conditions. On the contrary, the psychodynamic theoryof symptom formation requires that the
presence
of symptom substitution occur about as often as symptomatictreatment is applied, especially in cases of moderate or greater severity. Differential rates of application of symptomoriented therapy and reports of symptom substitution, if sufficiently large, can falsify the psychodynamic hypothesisregarding symptom formation absent explanation of buffering effects which have not yet been provided. I thereforesubmit that symptom substitution is a reasonably testable consequence of the psychodynamic theory of psychopathology and symptom formation. The next section reviews the empirical evidence bearing on the frequencyof symptom substitution in line with these new empirical criteria and considers implications regarding the psychodynamic model of symptom formation.
1. Empirical evidence
The cumulative body of empirical evidence available today that has focused on symptom substitution is small,uneven, was published at irregular intervals, and is not current. The many thousands of behavioral studies that did not make symptom substitution a focus of their investigation are not reviewed because a good way of not finding a phenomenon is not to look for it. Hence, studies not designed to detect symptom substitution are unlikely to find it andif found are unlikely to report it. Nurnberger and Hingtgen (1973)conducted an extensive literature review looking for evidence of symptomsubstitution resulting from: systematic desensitization of phobias and anxiety states, behavioral treatments of nocturnalenuresis, stuttering, tics, and modification of
“
bad habits
”
(p. 230). These authors generously concluded that symptomsubstitution rarely occurs. Theirgenerosity consisted of counting cases where anxiety returned as a result of an accident 8months post treatment and where relapse occurred followed by successful treatment as casesof symptom substitution.They also included some cases of surgical treatment of ulcer patients as evidence for symptom substitution. Otherwisethey could not find evidence of symptom substitution. They examined 2.5 years of follow-up data for the behavioral
965
W.W. Tryon / Clinical Psychology Review 28 (2008) 963
–
968
Add a Comment