Diagnosis and Treatment: Basic Principles
In the last thirty years, we in the mental health field have witnessed a veritable explosion of new information that has shifted the emphasis from the more psychoanalytically based models and treatments that dominated psychiatry and psychology for the preceding thirty years to descriptive and biological ones. The new approach is based on a number of assumptions, three of which are relevant to this book. The first is that psychiatric disorders can be reliably classified according to diagnostic methods used in medicine before the introduction of laboratory tests. The second is that pharmacological treatments -- medications -- are effective in treating a variety of psychiatric disorders. The third is that the efficacy of psychiatric therapies can be evaluated by empirical studies. Not surprisingly, the bulk of these treatment studies have involved medications. In this chapter, these three assumptions -- that psychopathology can be usefully described by symptom-based terms, that medications effectively treat psychiatric disorders, and that scientific methods can be applied to evaluate treatments -- will be discussed. The goals of medication treatment, some general principles of psychopharmacological treatment, and criteria for the selection of appropriate patients for psychopharmacological evaluation will then be presented.
DESCRIPTIVE PSYCHIATRY AND DSM-IV
Although the introduction of a descriptive, diagnosis-based approach to psychopathology was a radical shift from the etiologically based language of psychoanalysis, it was far from new. During the late nineteenth and early twentieth centuries, descriptive approaches dominated European psychiatry and resulted in the first diagnostic distinction between manic-depressive illness and schizophrenia (then called dementia praecox) on the basis of their differing clinical pictures. However, the descriptive approach fell into disfavor in this country for many decades, resurfacing only with the emergence of the Diagnostic and Statistical Manual of Mental Disorders, Third Edition (DSM-III) in 1980, the revised edition (DSM-III-R) in 1987, and DSM-IV in 1994. The goal of DSM-IV, as for any other diagnostic classification scheme is "to provide a helpful guide to clinical practice,...to improve communication among clinicians and researchers...,for improving the collection of clinical information and as an educational tool for teaching psychopathology" (DSM-IV, p. xv). By defining disorders consistently, both clinicians and researchers may better agree on what is meant by terms such as depression, thereby allowing better prediction of important clinical variables such as prognosis and treatment response.
Because the first half of this book is organized according to the DSM's classifications, it is worth reviewing their essential components.
The most important question in the descriptive approach used by DSM-IV is "what," in contrast to the centrality of "why" in psychoanalysis. It attempts (with only varying degrees of success) to be atheoretical with regard to etiology. Thus, when a 37-year-old man goes through a period of depressed mood with alterations in sleep, appetite, energy, and concentrating ability, the reason why is irrelevant for making the diagnosis of depression. Whether this patient's depression is best understood by his poor introject of a maternal object, depressogenic assumptions (using a cognitive framework), or by an alteration in the regulation of norepinephrine or serotonin in certain parts of the brain does not alter the diagnosis.
Among the objections to the descriptive approach is that patients are pigeonholed into diagnostic boxes and not understood for the unique constellation of intrapsychic, historical, and environmental variables that set each one apart from others. This objection is valid if clinicians view the patient's identity as synonymous with his diagnosis. Is the patient viewed as a person with manic-depressive disorder or is he a manic-depressive? The difference is far from semantic. Although the descriptive approach can be misused to objectify and distance patients, that is neither its purpose nor its proper use. DSM-IV classifies disorders, not individuals.
The second essential feature of the DSMs is their mulfiaxial approach. Patients are rated along five axes simultaneously, each of which utilizes different types of information. Psychiatric disorders are listed on Axes I and II while Axes III, IV, and V describe associated medical conditions, psychosocial and environmental stressors, and level of functioning. Axis I disorders comprise all clinical syndromes except for personality disorders and developmental disorders arising in childhood, which are listed on Axis II. Thus, Axis II disorders are, in general, more stable and long lasting. When a clinician describes a patient as having an Axis I disorder, he is referring to the presence of a symptombased disorder, such as depression, phobias, or schizophrenia. Describing an adult patient as having an Axis II disorder is equivalent to saying he has a personality disorder. Among the goals of separating Axis I from Axis II is to encourage clinicians to conceptualize coexistent disorders. Instead of deciding whether the patient suffers from major depression or narcissistic personality disorder, the clinician can diagnose both disorders. This approach makes evaluation more difficult, but also more accurate. In this way, either/or formulations can be replaced by richer, more complex models.
It is in the evaluation and diagnosis of personality disorders that the descriptive approach of DSM-IV is most problematic. Inherently, personality features are difficult to describe using the language of symptoms and signs. As an example, criteria used to diagnose personality disorders such as lack of empathy or persistent identity disturbance simply do not fit well into a classification system that is defined as atheoretical and descriptive. Compounding the problem of classifying personality disorders in the DSMs is the use of a categorical system in which patients either meet criteria for a diagnosis (a category) or they don't. Another type of diagnostic system describes patients along a number of dimensions without specific cutoffs demarcating normal from abnormal (or having a disorder vs. not having it). It is especially in the diagnosis and description of personality disorders that a dimensional system has the strongest and most persuasive proponents -- and that the categorical system has the most difficulties (see chapter 7 for more details about dimensional approaches to personality).
It is rather easy to criticize, with merit, the entire DSM system. Three different manuals -- DSM-III, III-R, and IV -- have been published over the short span of fourteen years, an insufficient time to gather enough new data to warrant many of the diagnostic definitions and changes in the new editions (Zimmerman, 1990). Disorders are defined and undefined in successive DSM editions without obvious justification. As an example, a manic episode precipitated by an antidepressant was diagnosed as bipolar disorder in DSM-III-R. In DSM-IV, that same episode would be called substance-induced mood disorder and specifically prohibited from contributing to a diagnosis of bipolar disorder, despite a lack of published data supporting or refuting either definition in the last six years. Furthermore, it continues to be difficult to use DSM-IV to describe patients with milder disorders, such as those with low self-esteem or heightened rejection sensitivity without significant depressive symptoms.
Additionally, the DSMs have done a better job at defining disorders reliably (can three clinicians agree that a certain patient has these specific symptoms and therefore meets specified criteria for a diagnosis?) than in demonstrating validity (do these diagnostic criteria have practical value with regard to prognosis, family history, treatment responsiveness, and so on?). Since for clinicians reliability is far less important than validity, the utility of the diagnostic system is not always apparent.
Despite these valid criticisms, the DSMs have accomplished a great deal and fill a vital need for mental health professionals. They have forced us to become more precise in our terminology, fostering both clearer thinking and clearer communication between ourselves and with those governing the finances of health care. Each succeeding DSM edition has been increasingly based on data and changes have often been made in response to feedback from the clinical community. If mental health professionals are to be taken seriously in our chaotic, rapidly changing system of health care (whatever its ultimate form), it is mandatory that we have clear definitions of disorders -- even if those definitions change somewhat over time -- and that we be able to perform the large-scale studies demonstrating both the prevalence and morbidity of these disorders. The DSMs foster these goals in ways that other systems might not.
Another major stumbling block for the acceptance of the descriptive approach as used in DSM-IV has been an understandable concern that this new model would replace and discount all other ways of understanding psychopathology. Descriptive models, however, should never preclude other ways of understanding psychological phenomena. For any clinical disorder, for any individual patient, different models will each have advantages and disadvantages in explaining the psychopathology. The clinical phenomenon of acute mania may be best viewed using the descriptive model, while adjustment disorders or narcissistic personalities will be better understood by an interpersonal or psychoanalytic perspective. A patient with a mild to moderate depression triggered by a loss, however, might be best understood using both descriptive and psychological concepts, with each model clarifying only a piece of the puzzle. It would be redundant and disruptive to point out continually in this text that other ways of understanding patients are helpful and, at times, mandatory. The use of multiple conceptual models should be considered a basic prerequisite for the full understanding of patients.
PHARMACOTHERAPY AND ITS IMPLICATIONS FOR OTHER THERAPIES
The second assumption of the medical model in psychiatry, that some psychiatric disorders are effectively treated by medications, has been established through the astonishing amount of research over the last forty years devoted to the discovery, development, and documentation of psychopharmacologically active drugs. Initially used for severe depressions and psychotic disorders, medications have now been demonstrated to be useful for at least some patients with a wide variety of disorders. The simple existence of two editions of this book is testimony to the extent to which medications have been established as a treatment modality for psychiatric disorders. Despite the dramatic effect of medications in reducing or preventing psychopathology, however, their limitations have tempered some of the early, unrealistic hopes of the more biologically oriented clinicians. In a variety of disorders, medications are profoundly effective, yet still leave untouched some core aspects of the disorder which must be treated with other modalities. These limitations are most obvious in schizophrenia (see chapter 5), but are apparent also in panic disorder, bipolar disorder, and others.
Since 1987, with the release of fluoxetine (Prozac) as the first of the powerfully serotonergic antidepressants with fewer side effects than the older antidepressants, an increasing number of individuals with relatively mild psychiatric disorders and difficulties have taken one of these new medications. As an example, through early 1995, approximately 16 million patients had taken fiuoxetine alone (Dista, 1995). The rapid proliferation of these medications in psychiatric treatment has led many therapists and interested laypeople to be concerned that prescribed drugs would soon become the quick fix for all problems, that a patient suffering distress would be given a medication to feel better at the risk of ignoring psychological and psychosocial factors. ("I just want to be more assertive with my girlfriend" or "I want to be bolder and more creative.") The explosion of media interest in these new medications has only exacerbated the problem, as exemplified by one Newsweek cover that pictured a Prozac capsule (labeled by its trade name and not its generic name) and by the controversy surrounding Peter Kramer's book Listening to Prozac (1993). (See chapters 3 and 7 for more detailed discussions of the serotonergic antidepressants for mild depressions and personality disorders.)
It is likely that some individuals have been prescribed one of the serotonergic antidepressants with little clinical justification. However, a great number of people with mild psychiatric disorders, many of whom have worked hard in psychotherapy but are still symptomatic, have greatly benefited from one of these newer medications. Moreover, in the treatment of the more serious psychiatric disorders, extraordinary numbers of patients continue to go untreated. In the most recent epidemiological study, only 42 percent of those with a psychiatric disorder had ever sought treatment for that disorder (Kessler et al., 1994). Of those with an active psychiatric disorder within the last year, less than 30 percent had received any type of treatment (Regier et al., 1993; Kessler et al., 1994). There is even evidence that only one quarter of chronically anxious patients use tranquilizers (Uhlenhuth, Balter, Mellinger, Cisin, and Clinthorne, 1983). Finally, of those patients with a severe mental illness as defined by psychosis or marked functional impairment, almost 40 percent sought no treatment within a one-year period (National Advisory Mental Health Council, 1993). Together, these studies suggest that, even now, we continue to be more of an undertreated than an overtreated society.
Another concern of psychotherapists about the advent of medications was their implication for the etiology of psychiatric disorders. There is a natural assumption that if medications are helpful, they must be correcting some biochemical abnormality which would then be viewed as the sole cause of the disorder. With the simplistic information promulgated in the press, patients come to their primary care physicians claiming that they know they have a serotonin deficiency and that they want a medication to fix this problem! Despite a remarkable amount of research over the last twenty-five years, however, there is still no definitive biological explanation for any psychiatric disorder (see chapter 2). Furthermore, even if a biological cause might be found for one or a number of disorders, it would not, by itself, imply the proper or effective methods of treatment. For example, coronary artery disease culminating in heart attacks has genetic and biological causes. Yet its course and outcome can be altered by life-style changes, such as diet, smoking, exercise, and the like. Similarly, even if depression or rejection sensitivity were shown to be caused by a specific neurotransmitter abnormality, this would have no necessary implication for the efficacy of psychotherapy in treating it.
In summary, just as descriptive models of psychopathology, as exemplified by DSM-IV, must be supplemented by other models to best understand our patients' problems, pharmacotherapy never precludes other methods of treatment. For some disorders, such as mild to moderate depression or obsessive compulsive disorder, there may be a variety of different, valid therapeutic approaches. As discussed in more detail in chapter 14, even with those disorders for which medications are the most effective treatments available, such as bipolar disorder, psychotherapy is likely to enhance the treatment and help patients in ways not measured in research studies. Moreover, since there is no evidence that, when utilized together, medication and psychotherapy interfere with the efficacy of each other, combination treatment should always be considered. Because this book focuses on pharmacological therapies, it will often not discuss the use of other valid treatments. Nonetheless, it should be assumed that other approaches do exist and may at times be preferable to medications for specific patients with certain disorders.
EVALUATING TREATMENTS: THE MEANING OF THE WORD EFFECTIVE
The application of scientific methods of evaluating treatment is the third important assumption inherent in the medical model approach to psychiatric disorders. An in-depth discussion of statistics is hardly necessary or relevant for this book. What is important, however, is a clarification of the use of the word "effective," since throughout the book statements will be made referring to a medication's effectiveness in treating a psychiatric disorder. At first glance, the meaning of the word "effective" seems clear -- that the medication is useful in diminishing the manifestations of the disorder being treated. However, a number of questions about the use of this word must be addressed.
First and most important, how does this treatment compare to others? As already noted, the effectiveness of a medication does not bear on that of another type of treatment. For instance, the efficacy of certain antidepressants in diminishing the symptoms of obsessive compulsive disorder (see chapter 4) does not, in any way, negate the well-documented efficacy of behavior therapy for the same disorder. Similarly, the effectiveness of a medication does not bear on the potential value of other treatments administered simultaneously. In the treatment of schizophrenia, for instance, antipsychotics, although vital, are rarely sufficient for maximal response. A combination of medication with psychotherapy is likely to be the best treatment.
Another important question is that of assessment: how is effectiveness evaluated? During the twentieth century, and increasingly over the last thirty years, the hallmark of efficacy is that the medication has been shown to be effective in research studies using a double-blind, placebo-controlled design. In these studies, patients are randomly assigned to receive either the medication or an identical looking placebo. Patients and investigators are unaware of (i.e., blind to) the identity of the treatment actually received. In this way, the enthusiasm of the investigator as well as that of the patient ("I'm receiving a new pill that will make me all better") is similar for both the real medication and the placebo. If a drug is consistently associated with more improvement than a placebo in these studies, it is considered an effective treatment. For new drugs to be released in the United States, efficacy in double-blind studies (along with numerous other requirements) must be demonstrated.
A third important question is that of relativity: what is meant when one active treatment is described as more effective than another? As a generalization, this means that a larger percentage of patients responded (or improved to a greater degree) to one treatment compared to another. The magnitude of this difference may be relatively small (e.g., 60 percent vs. 50 percent) or large (70 percent vs. 30 percent). In both examples, however, some patients responded to the less effective treatment. (The inclusion of a placebo group would help evaluate whether the less effective treatment was superior to a placebo.) In a number of clinical situations, it is entirely appropriate to use the less effective treatment. If the more effective of two treatments has significantly more side effects, the less effective medication might be preferable. Furthermore, statements of comparative efficacy do not predict the response of an individual patient to either treatment since studies refer to groups, not individuals. Thus, a number of individual patients may respond better to the less effective treatment. The comparison statement simply states that a greater number of patients will respond to one medication than another.
Furthermore, even if a methodologically careful study demonstrates the effectiveness of a medication (compared to a placebo or to another drug), it should never be accepted as conclusive or proven. The recent history of medicine in general, and psychiatry specifically, is filled with carefully executed initial studies, the results of which are never replicated. There are a variety of explanations for this phenomenon, including subtle biases in the study design or selection of unusual patients (for instance, depressed patients seen at a university medical school may not be representative of the types of patients seen in the community and might respond to treatment differently). Despite the occasional story in the press, fraud is rarely responsible for contradictory findings in research. In general, any finding that is valid will be demonstrated in a number of different studies. Unfortunately, in their zeal for "hot" news, television and newspapers quote single studies as if they discovered the revealed truth. Patients who read these reports often look for this new magic. Therapists as well as psychopharmacologists need to help keep our patients from being swayed by these distortions.
A more recent consideration in clinical psychopharmacology distinguishes between efficacy and effectiveness, two terms that are typically used synonymously. For this distinction, efficacy refers to the percent of patients responding to a medication in a controlled study. Although controlled studies are vital for comparing treatments and for describing the percent of patients who respond under optimal conditions, patients in these studies tend to be highly motivated and physically healthy and typically do not suffer from any of a host of other concomitant psychiatric disorders common in the community. Thus, efficacy in controlled studies bears an uncertain relationship to the usefulness of a medication for the majority of more complicated patients seen in everyday clinical practice. The more clinically relevant variable is effectiveness, which takes into account ease of administration, side effects, and rates of noncompliance, and examines the utility of a medication for real patients who frequently suffer from multiple disorders, both psychiatric and medical. Using these definitions, it is effectiveness, not efficacy, that ultimately dictates which medications are prescribed in the community. As recently acknowledged (Klein, 1993b), almost no research studies have explored the types of practical questions confronted in everyday practice. Thus, from research studies we know very little on topics such as how quickly to raise the dose of a medication, what dose schedule should be used (once daily vs. twice daily), how to treat side effects, how long to keep patients on medications once they are better, and the optimal time to discontinue medications. By necessity, there is much accumulated experience and wisdom in these matters, but surprisingly little data. Therefore, much of what will be presented in the rest of the book will rely heavily on this wisdom, with the results of studies quoted when available.
Finally, the relationship between starting an individual patient on a medication (or any new treatment) and the resulting clinical response is not always as clear as it may seem. A patient's improvement may be due to one of three variables. First, the medication itself may have a pharmacological effect. Second, a placebo response may occur, defined here as improvement from any and all aspects of treatment that have no specific value for the condition being treated. Thus, a patient who is given a medication may improve because of increased hope, the magic of taking a pill administered by a societally sanctioned healer, or positive transference to a parental figure. This would be described as a placebo response insofar as the clinical change was unrelated to the pharmacological effects of the specific medication. A third variable is spontaneous remission. A variety of psychiatric disorders for which medications are prescribed are self-limited by nature, with or without treatment. As an example, major depressive disorder lasts an average of six to eight months. Thus, if a medication is started during the eighth month, it might be difficult to know whether the improvement seen was due to the treatment or the lifting of the depression that would have occurred anyway at that time.
GOALS OF PHARMACOTHERAPY
Pharmacological treatments can be considered to have more than one goal. Specifically, medications may be prescribed to (1) treat an acute disorder, (2) prevent a relapse soon after clinical improvement, or (3) prevent future episodes of the disorder. These three goals or phases are termed acute, continuation, and maintenance treatment.
Acute treatment is used to alleviate the symptoms of an actively occurring disorder. When most people think of treatment as necessary, they are referring to acute treatment. A depressed patient is given antidepressants to alleviate the active symptoms of the disorder; lithium is prescribed to diminish the symptoms of an acute manic episode.
The goal of continuation treatment is to prevent a relapse into the same episode for which treatment was begun. As an example, the depressed patient who is given an antidepressant may improve over four weeks. Once the symptoms of the disorder have remitted, acute treatment ends and continuation treatment begins. If the antidepressant is stopped at this point, when the patient has only recently become asymptomatic, the risk of relapse is high. The analogy in general medicine is the standard recommendation to continue antibiotics after the cough of a respiratory infection has stopped. The cough may remit after three to four days, but the antibiotics are typically prescribed for an additional seven to ten days as a continuation treatment. For psychiatric disorders, continuation treatment is typically extended for many months. Recommendations as to the length of continuation treatment are slightly different for each disorder. These will be covered in the chapters on the individual disorders. Unfortunately, there is an astonishing paucity of research regarding the appropriate length of continuation treatment. Thus, as with many of the practical issues noted above, the recommendations given will reflect clinical wisdom more than validated research findings.
Maintenance treatment is synonymous with preventive treatment. Because many psychiatric disorders occur in episodes throughout a person's lifetime, a decision can be made whether to treat each episode only when it arises (acute treatment), or to prevent recurrences by the ongoing, maintenance use of a medication. The two most common examples of medication maintenance treatment in psychiatry are lithium for bipolar disorder and antipsychotics for schizophrenia. In both disorders, there is an overwhelming likelihood of repeated recurrences. The decision to institute a psychopharmacological maintenance treatment is based on a judgment that takes into account such factors as the length of time between episodes, the severity and destructiveness of the episodes, the ease of treating acute episodes, the rapidity with which the episodes begin, patients' capacity for insight into the beginning of an episode (that is, can they recognize the warning signs so that acute treatment can begin quickly?), the potential toxicity of the treatment, and alternative preventive therapies. Thus, for each patient and for each disorder, somewhat different considerations apply. Maintenance treatment is discussed further in the chapters covering individual disorders.
SOME GENERAL ISSUES IN PSYCHOPHARMACOLOGICAL TREATMENT
A number of issues pertinent to all psychopharmacological treatments are relevant for any mental health professional seeing patients taking medications.
FDA Approval and PDR Doses
One series of concerns sometimes expressed by both patients and mental health professionals surrounds the regulation of medications by the Food and Drug Administration (FDA). New medications are approved by the FDA and then released based on their safety and efficacy for a specific disorder (or disorders) as demonstrated by rigorous double-blind studies. The medication is then described as being indicated or approved for the treatment of that disorder. This information is listed in the Physicians' Desk Reference (PDR), along with the dosage range that was tested during the research trials. The specific language and information given in the PDR is negotiated and ultimately approved by the FDA. As such, the PDR should be considered as providing information for marketing purposes and medicolegal protection for the pharmaceutical firms. It specifies what pharmaceutical firms may claim for their products, and gives warnings and notes side effects in order to avoid a later charge of not informing physicians about their products.
Often, medications are prescribed for nonapproved uses -- that is, for disorders or uses other than those evaluated and accepted by the FDA. Similarly, medications may be prescribed at doses higher than those recommended in the PDR or in the package insert given at pharmacies. This is acceptable and appropriate clinical practice (assuming the doses used are not dangerous and the rationale for the medication's use is reasonable). Neither the FDA indications nor the recommended maximal doses should be considered legally or ethically binding, nor are they necessarily based on the best scientific information.
To achieve an FDA indication requires enormous time and money. If a medication is already available and clinical studies and experience show that it is beneficial for a disorder other than the original indication, it is generally not worth the money for the pharmaceutical firm to submit an application to the FDA to obtain the additional indication. The most obvious example of this is the use of tricyclic antidepressants and imipramine (Tofranil) in particular for successfully treating panic disorder. Despite the ample evidence that these medications are effective antipanic medications and have been used for that purpose for over twenty years, they are not FDA indicated for panic disorder.
Similarly, dosage guidelines in the PDR and package inserts are conservative estimates, documented by studies submitted to the FDA by the pharmaceutical firm typically before the release of the medications. If clinicians and clinical researchers discover that higher doses of a certain medication are both safe and effective, it will generally not result in a change in the package insert which, like adding a new indication, is usually prohibitively expensive. Because of these limitations in the system, all psychiatrists -- indeed, all physicians -- routinely prescribe medications for nonapproved uses and intermittently at doses outside the package insert recommendations. Prescription must simply be consistent with sound medical judgment (although prescribing outside FDA indications and PDR dosage ranges may be factors used in a malpractice suit if a bad outcome ensues).
A far more difficult dilemma surrounds the use of FDA nonapproved medications -- drugs that are unavailable in the United States but are available, for example, in Canada, Mexico, or Europe. Some of these medications have been prescribed for decades in other countries but their manufacturers never bothered to obtain FDA approval for release in the United States because of the extraordinary cost involved (currently estimated at many tens of millions of dollars). Clomipramine (Anafranil), the most well-documented treatment for obsessive compulsive disorder, was available in Europe for twenty years before its FDA approval in 1990. Only after epidemiological studies showed that obsessive compulsive disorder was common, and that therefore the potential market for the medication was large, was there a push for its release in this country. The FDA permits patients to carry nonapproved medications into the United States or to receive them by direct mail shipments, if they are for personal use (i.e., not for sale) and if the import is neither fraudulent nor dangerous (Kessler, 1989). Malpractice insurance companies are less consistently accepting of this practice and will at times refuse to cover physicians in their supervising the use of nonapproved medications. Thus, physicians who supervise use of nonapproved drugs find themselves in something of a limbo -- the practice is legally acceptable but not consistently covered medicolegally. Because of this, only some psychiatrists are willing to participate in these prescribing practices.
In general, most psychopharmacological texts emphasize that polypharmacy -- the use of multiple medications prescribed simultaneously -- should be avoided whenever possible. With single agent treatment, interactions between drugs are avoided and side effects more easily understood. Yet, over time, the limitations of single treatments have become more apparent and the rationales for multiple medications increasingly persuasive. Polypharmacy can be very useful, but because it clearly introduces complications and some risks, it should only be undertaken thoughtfully and by a practitioner familiar with drug interactions.
The most common reasons for rational polypharmacy are listed in Table 1-1. An example of a patient with more than one disorder would be one with both panic disorder and depression who might require a benzodiazepine tranquilizer along with an antidepressant. The use of adjunctive treatments -- second medications added to the first to augment effectiveness -- has arisen with the awareness that single agent treatment does not result in sufficient improvement in many patients. The most common augmenting agents are those prescribed for treating depression, such as lithium or T3 which, when added to an antidepressant, enhance the antidepressant effect (see chapter 3 for details). Combination treatment is the prescription of two independently effective medications together, with the possibility that, when used concurrently, they will be more effective than either medication if prescribed alone. Examples are the use of two different mood stabilizers -- such as lithium plus valproate -- for difficult-to-manage bipolar disorder or a tricyclic antidepressant plus a serotonergic antidepressant for treating depression. Especially in the beginning of treatment, a medication may be temporarily prescribed as an aid for symptomatic relief along with the primary medication until the latter takes effect. For instance, a depressed patient with anxiety and insomnia might benefit from a brief prescription of a tranquilizer in the beginning of treatment until the antidepressant becomes effective (which may take a few weeks). In addition, increasing use is being made of second medications to treat the side effects of the primary medication. As an example, trazodone, a sedating antidepressant, is often properly and successfully prescribed to help combat the insomnia caused by the serotonergic antidepressants. Finally, a few disorders are well documented as being optimally treated with medications of two different types. The most well known example is psychotic depression which requires a combination of an antidepressant with an antipsychotic for effective treatment.
Despite these clear guidelines, all psychopharmacologists have had experiences with particularly difficult patients who end up being treated with multiple (up to seven or eight) simultaneous medications. Even though this may at times represent sloppy clinical practice, it may also represent the limitations of pharmacotherapy and the goal of minimizing side effects. As an example, a brittle bipolar depressed patient might be treated with two mood stabilizers (if neither one was sufficiently effective), an antidepressant, a hypnotic for sleep (if the antidepressant is effective but causes insomnia), a beta-blocker to reduce the tremors caused by the mood stabilizer, and a medication to counteract the sexual side effect of the effective antidepressant. This may look like psychopharmacology gone wild, but there is a clear rationale for each of these treatments and together they make the treatment both more effective and more tolerable. When multiple medications are prescribed, it is incumbent on the physician to regularly review the treatment regimen and to eliminate those medications that are no longer needed.
Over the last two years, the American Psychiatric Association has published a series of practice guidelines for a variety of disorders such as eating disorders, major depression, bipolar disorder, and substance abuse. Future guidelines are planned for schizophrenia and other disorders. These documents contain consensus recommendations for suggested treatments (including medications) for specific indications, such as the proper role of antidepressants in bulimia nervosa or lab tests that should be checked during ongoing lithium use. The guidelines are not meant to define the standards of appropriate treatment to be followed in all cases, or specific treatment mandates for individual patients, but rather to provide overall direction (Zarin, Pincus, and McIntyre, 1993).
WHO SHOULD HAVE A MEDICATION CONSULTATION?
In deciding whether a psychotherapy patient should be referred for medication consultation, an initial problem is to elicit the relevant information needed to make the decision. Typically, the interviewing style in psychotherapy, in initial sessions and even more during the course of the therapy, is open-ended and nondirective. What is easily missed using that technique is the presence of symptoms that the patient is either unaware of or whose significance the patient does not grasp. In the past, the mental status examination, in which the patient's behavior in the interview setting was observed and systematically evaluated, was emphasized. To a great degree, this has been replaced by a heightened emphasis on obtaining an accurate history. With the current focus on longitudinal data (what symptoms are present and for how long?), it is less important (though not unimportant) to accurately describe the patient's affect in the interview or to distinguish between flight of ideas or loose associations than to find out how long the patient has been unable to concentrate or felt depressed. Thus, in evaluating patients for pharmacological consultation, the therapist may need to shift into a more directive style of questioning. The timing of these questions -- whether in the first session or later -- will depend, in great part, on the therapist's suspicion about the presence of a pharmacologically treatable disorder.
A series of clinical clues that are nonspecific with regard to diagnosis but that suggest the presence of the type of Axis I disorder for which medication might be appropriate are listed in Table 1-2. These items are broad-based and do not substitute for specific questions that are needed to diagnose specific disorders. Sample questions that will help diagnose specific disorders such as depression, bipolar disorder, a variety of anxiety disorders, or schizophrenia are given in the individual chapters.
Most important among the general items that should suggest a consultation is the ability to describe the patient's difficulties using the language of symptoms, such as insomnia or fatigue, as opposed to psychological feelings or interpersonal interactions. Marital conflict, as an example, can usually be described only using interactive terms. Axis I disorders, for which medications are most commonly prescribed, are defined by these types of symptoms. Therefore, the more the patient's problems or complaints focus on sensations or bodily feelings, the more a consultation should be considered. This is especially true when the symptoms involve cognitive capacities that have changed. Common examples include a new-onset memory disturbance or a diminution in concentrating ability. Psychotic symptoms, that is, those that involve a gross impairment in reality testing, comprise another group of important symptoms that usually require pharmacological intervention.
The presence of medical symptoms or disorders is another clue for consultation. New or recent-onset medical symptoms, such as headaches, abdominal pain, or clumsiness may reflect a medical or a psychiatric disorder. If the patient has an ongoing medical problem that has not been recently reevaluated; or takes medication and has symptoms or physical complaints, an evaluation, either by a psychopharmacologist or an internist, should be done.
Patients who are significantly suicidal are also candidates for psychopharmacological consultation. The most important reason for this is that the majority of people who commit suicide suffer from types of Axis I disorders (the most common of which is depression) for which medications can often be helpful (Robins, 1986). Second, with the medicolegal climate as it currently exists, if a patient commits suicide without having been evaluated (but not necessarily treated) for medication, the therapist may be considered negligent and at higher risk for being sued.
As will be highlighted during the course of the book, the types of disorders for which medications are useful tend to run in families. It is usually impossible to tease apart early environmental variables from genetic ones, since the parent who may have transmitted the genetic vulnerability is usually the same one who raised the patient. Nonetheless, a patient who describes mood swings that are mild but whose mother and brother have clear-cut bipolar disorder is more likely to have a pharmacologically treatable disorder than is another patient with no history of mood disorders in the family.
Many patients generically described as moody or explosive have typically been conceptualized as having personality disturbances that were not amenable to pharmacotherapeutic treatment. Yet recent clinical experience suggests that many of these patients have pharmacologically responsive disorders or characteristics. (Chapter 7 discusses this issue in more detail.) Therefore, patients who are overly sensitive to rejection or get enraged in response to minor provocation, especially if they are not improving in their psychotherapy, should be considered for psychopharmacological consultation.
Finally, nonresponse to psychotherapy might also suggest a consultation. This often takes the form of patient and therapist acknowledging that the work of the therapy has gone well -- a therapeutic alliance has been established, the patient has gained significant insight into the source and context of his problems -- yet the depressed mood, or the chronic anxiety is unchanged. Certainly, an unsatisfactory response to psychotherapy does not by itself imply a pharmacologically treatable disorder, but it may be worth considering.
Copyright © 1990, 1996 by The Free Press