by David Cohen, Ph.D.*

* Researcher, Health and Prevention Social Research Group, and Associate Professor, School of Social Work, University of Montreal, Canada. [e-mail:]

Prepared at the request of the Connecticut Legal Rights Project, Inc., and the Office of the Probate Court Administrator (Connecticut), July, 1997


This document provides a synopsis of current information about the main classes of commonly used psychiatric drugs, focusing on their indications, effects, and effectiveness.

For thousands of years, humans in all known societies have used psychotropic drugs (substances that act on the central nervous system [CNS] and affect mood, thinking, and behavior). Roughly speaking, all psychotropics may be classified as CNS depressants (e.g., alcohol, opium, Valium), CNS stimulants (e.g., coffee, cocaine, Ritalin), or hallucinogens (e.g., LSD). Psychotropics either tranquillize or sedate, awake or stimulate, or impair perception. Occasionally, shades of these effects are produced by a single drug.

Over the last 45 years, with progress in synthesizing chemicals and with the rapid acceptance of medical approaches to understand and treat human distress, many psychotropics (except, since the 1960s, hallucinogens) have been prescribed to treat emotional and behavioral problems. These drugs are often called "psychiatric drugs," although primary care physicians write the bulk of prescriptions. Today, about 85 psychiatric drugs are on the U.S. market and over 70 other substances are in various phases of development as future psychiatric drugs.

Psychiatric drugs may be divided into 5 major classes, each typically named according to its main indication in psychiatry. However, these names mean little since most drugs are actually used for most indications: stimulants may be prescribed to calm children, antidepressants to relieve anxiety; anticonvulsants to control mania, etc. Classifying drugs on the basis of known chemical structure or action might be more accurate but would create numerous categories. Given the absence of pathophysiological findings underlying psychiatric diagnoses, current classification of psychiatric drugs remains provisional.


Neuroleptic (NLP) drugs were introduced in 1953; 18 exist on the U.S. market today. They triggered the "pharmacological revolution" in mental health and helped empty psychiatric institutions beginning in the mid-1960s. "Typical" or "first generation" NLPs include: haloperidol (Haldol), thioridazine (Mellaril), thiothixene (Navane), fluphenazine (Prolixin), chlorpromazine (Thorazine). NLPs are indicated to control symptoms of psychosis (defined as loss of contact with reality), typical of diagnoses such as schizophrenia, bipolar disorder, brief reactive psychosis, etc. (said to affect about 1.5% of the population). NLPs are the drugs of choice for any disorder with strong elements of agitation or delusion. They are also widely prescribed to demented elderly persons in nursing homes and persons with developmental disabilities in institutions. NLPs are usually effective in initially calming almost any type of psychomotor agitation or emotional outburst ("positive" symptoms), in humans and animals. Depending on their mode of administration-intravenous or intramuscular injection, liquid, pill-they may do this in hours, days, or weeks. This short-term effect is visible in about 65% of cases, though most acutely psychotic patients are also concomitantly prescribed several other CNS depressants. NLPs are less effective for, and often worsen, symptoms of social withdrawal ("negative" symptoms).

After the initial psychotic episode, lower doses of NLPs are prescribed on a continuation and then on a maintenance basis, in order to delay expected relapses or recurrences of psychosis. For this, they may be effective in about 50% of cases. Since we cannot identify who would so benefit, NLPs are prescribed to all patients after the initial crisis, roughly as follows: after one episode, for 1-2 years; after multiple episodes, from 5 years to indefinitely.

Most patients find NLPs very unpleasant ("zombie effect," "chemical straitjacket," etc.). NLPs affect virtually every cell and organ system, but their most worrisome effects are neurological. These include abnormal movements termed extrapyramidal symptoms (EPS), all of which may become irreversible. Four types of early-onset (acute) and late-onset (tardive) EPS affect 40-90% of patients: parkinsonism (rigid muscles, loss of facial expression, unsteady gait, drooling); akathisia (inner distress, rocking, pacing, agitation); dystonia (sudden, bizarre muscle spasms); dyskinesia (rythmic movements of face, mouth and tongue, sometimes extremities). Tardive dyskinesia cannot be predicted in advance nor treated effectively and may be masked by the NLP itself so that a dose reduction is necessary to detect it. It will affect 20-35% of adult patients (40-70% of elderly) within the first 5 years of treatment and persist in most cases unless NLPs are withdrawn. Controversies surround NLPs, notably concerning the possibility of their inducing dementias after long-term use and psychoses upon abrupt withdrawal. In 1987, Congress enacted OBRA-87, mandating nursing homes receiving federal funding to implement systematic NLP dose reduction and withdrawal programs to reduce the incidence of adverse effects.

For most patients, no NLP is considered more efficacious than another. Prescription depends upon side effect profile, patient's history of response, and cost. Four "atypical" or "second generation" NLPs, most introduced within the last three years, appear to provoke lower rates of early-onset EPS, at least at lower doses (clozapine [Clozaril], risperidone [Risperdal], sertindole, and olanzapine [Zyprexa]). Although more expensive than conventional drugs, these NLPs are understandably gaining great popularity and replacing conventional NLPs. However, long-term studies of atypical NLPs' toxicity and effectiveness are lacking.


Currently, 21 such drugs are included in 3 main categories:
1) 2 "monoamine oxidase inhibitors" (MAOIs: e.g., phenelzine [Nardil]), little used today because of their toxicity;
2) 11 "tricyclic antidepressants" (TCAs: e.g., amitriptyline [Elavil], imipramine [Tofranil]), mainstay of antidepressant treatment since the 1950s, less used today because of their many adverse effects and their cardiotoxicity in overdose;
3) 4 "selective serotonin reuptake inhibitors" (SSRIs: fluoxetine [Prozac], fluvoxamine [Luvox], paroxetine [Paxil], sertraline [Zoloft]), introduced during the last decade and now first-line drugs because of a more tolerable adverse effect profile, lower cardiotoxicity in overdose, and intense marketing. Four other atypical antidepressants complete the class (e.g., trazodone [Desyrel], venlafaxine [Effexor]).

No antidepressant is more efficacious than another and prescription depends mostly upon the side effect profile. TCAs are typically sedating. Frequent adverse effects are dry mouth, sudden drops in blood pressure, weakness, blurred vision, constipation, confusion, irregular heartbeat, sexual dysfunctions. Abrupt withdrawal of TCAs may produce several unpleasant symptoms. SSRIs, on the other hand, typically are stimulating. Their frequent adverse effects are nausea and headaches, nervousness, fatigue, weight loss, sexual dysfunctions. Abrupt withdrawal of SSRIs may produce several unpleasant symptoms.

A course of treatment with antidepressants usually lasts about 4 months; 3 weeks are considered necessary before deciding that drugs are not useful. Increasingly, antidepressants are prescribed on a "maintenance" basis, for several months or years. Because of earlier media interest in first-person accounts of SSRI effectiveness, and because of SSRIs' rapid use in many other conditions other than depression (anxiety and eating disorders, premenstrual syndrome, etc.), it is rarely acknowledged that antidepressants often rate just slightly superior than placebo. Most depressions (said to affect about 5-10% of the population) are temporary, self-limiting situations, which also respond well to counselling or changes in the environment.


In the U.S., this class of drugs mostly includes 14 benzodiazepines (BZDs), first introduced in 1961. Valium, a BZD, was the most prescribed drug in the world in 1970. Other BZDs are alprazolam [Xanax), chlordiazepoxide [Librium], clonazepam [Kionopin], flurazepam [Dalmane], lorazepam [Ativan], oxazepam [Serax], temazepam [Restoril], triazolam [Halcion].

The BZDs, currently DEA Schedule IV controlled substances, all have varying degrees of muscle relaxant, tranquillizing, anticonvulsant, and sleep-inducing properties. They are prescribed to relieve anxiety. including panic and phobias (said to affect about 5% of the population) and insomnia (said to affect about 10% of younger adults, 35-50% of elderly). Some BZDs are marketed as anti-anxiety drugs and others as hypnotics for insomnia, but a given BZD can serve as either, depending on whether dosage is decreased or increased. Ninety percent of drugs prescribed for insomnia and sleep problems in the U.S. are BZDs. Other frequently used hypnotics include antihistamines (e.g., diphenhydramine [Benadryl]). BZDs are heavily prescribed in nursing homes as tranquillizers and hypnotics.

For very short-term uses, BZDs are effective for their main indication, and the quick onset of their effects has made them popular and desired. However, BZDs are much less effective after 3 to 6 weeks, and about a third of users will experience withdrawal symptoms (including "rebound" anxiety and insomnia). As a result, about a sixth of initial users remain on the drugs for months or years, despite repeated efforts at withdrawal. To dissuade physicians from prescribing BZDs, prescriptions in the U.S. may not be refilled more than 6 months after the date issued or be refilled more than 5 times. Withdrawal difficulties are well documented and evidence of very long-term use abounds, especially among the elderly. Nevertheless, some experts and many general practitioners still believe the risk of BZD dependence to be small.

The other main adverse effects of BZDs are: drowsiness, impaired coordination, confusion, memory loss. They contribute to falls among the elderly and may be a hidden cause of motor vehicle accidents. BZDs are also reported to produce "paradoxical" agitation, disinhibition, and hostility. BZDs resemble alcohol: in their widespread, socially approved occasional usage; in their short term relief of anxiety and tension; in their typical adverse effects; in their tendency to produce dependence after chronic use; in their withdrawal symptoms; and in their occasional agitating effects.


Antimanics are drugs used to treat various phases of bipolar disorder (manic-depressive psychosis, said to affect about 1% of the population). They include lithium (Eskalith, Lithobid, etc.); anticonvulsants used to treat epilepsy (carbamazepine [Tegretol], valproate [Depakene, Depakote]); as well as benzodiazepines (BZDs) and neuroleptics (NLPs).

The calming properties of lithium were discovered in 1949 but it was not until 1971 that the drug was marketed for mania. Since then, it has also been used to boost antidepressant drugs, to calm aggressive patients, to tranquillize alcoholics, etc. Because lithium's CNS depressant properties appear after 1-2 weeks, acutely manic individuals are usually prescribed NLPs at least until the lithium takes effect. A trial of lithium in acute mania lasts about 3 weeks, usually followed by 4-6 months of continuation treatment. Lifelong lithium has been recommended for those with multiple episodes.

However, hailed for two decades as a specific anti-manic drug, lithium's reputation has taken a heavy blow lately as most long-term studies indicate that manic-depressives using it faithfully relapse as often as those not taking it. For this reason, other drugs, notably anticonvulsants and BZDs, are replacing lithium or added to it. This polytherapy is now the norm though it is not well-studied nor judged more efficacious. The use of anticonvulsants is justified on the basis of speculative analogies between bipolar disorder and epilepsy.

Lithium's most frequent adverse effects are slowing of intellectual function, slurred speech, headaches, apathy, mild hand tremor, thirst, excessive urination, memory loss, metallic taste in mouth, decreased libido. Renal function may be impaired after many years of use. Lithium is highly toxic and its toxic dose is very close to its therapeutic dose: blood samples must be checked regularly to ensure serum lithium levels are within range and to avoid grave intoxication. Because of lithium's vast range of effects on the human body, complete medical workups before and during treatment are essential.


Stimulants (mostly comprised of amphetamines, introduced in the 1930s), used to be prescribed as often by physicians 35 years ago as antidepressants are today. Stimulants have long been used by those needing concentration or energy for arduous tasks (drivers, pilots, athletes, students, etc.) and by those interested in the drugs' appetite-suppressing effects. Cocaine is the best-known illicit stimulant.

Just as their addictive properties became well known in the 1960s, amphetamines ("speed") began to be used recreationally, leading to more stringent controls. However, since the mid-i 970s, 3 stimulants (methyiphenidate [Ritalin], dextroamphetamine [Dexedrine], pemoline [Cylert]) are back as treatments for hyperactivity and inattention in children (also known as ADHD, said to affect 2-5% of children, mostly boys). Although it is a DEA Schedule II controlled substance, Ritalin is today the most frequently prescribed psychotropic drug to school-age children. This has led both to increased prescribing to adolescents and young adults as well as diversion to recreational use.

At small initial doses, across all age groups and diagnoses, stimulants improve concentration and alertness, induce a sense of well-being, and reduce fatigue. About 70% of children on stimulants become less distractible and calmer. Larger doses may inverse these effects, and at high doses taken orally or injected, especially in illicit circumstances and for long periods, anxiety, agitation, suspiciousness, hallucinations and aggressive behavior are likely to develop. Less frequently, a full-blown "amphetamine psychosis" is provoked. Stimulants' main adverse effects are sleeplessness, weight loss, abdominal cramps, nervousness, headaches, fatigue, nervous tics. An apparently reversible growth suppression is well-documented in children. Because of the initial well-being produced by stimulants, chronic users often become psychologically addicted. In humans, Ritalin and cocaine, for example, compete for identical brain sites and produce their "high" in the same amount of time. Abrupt withdrawal from stimulants may produce a bout of depression ("crash").


Every emotional and behavioral problem for which psychiatric drugs are prescribed responds to nonpharmacological interventions, especially individual and group therapy. Moreover, despite claims of superiority from proponents of some treatments, all methods evaluated, including drugs, are associated with equivalent outcomes. The now-famous Consumer Reports survey published in November 1995 provides the most recent illustration. In the largest study of its kind, over 4000 individuals replied to a detailed questionnaire about their experiences with mental health treatment. Physicians, psychologists, and social workers received identical approval ratings, whatever the problem they were consulted for and the length of the therapy. There was no difference between psychotherapy alone and psychotherapy plus medication. Long-term therapy produced more improvement than short-term therapy.

Predictably, psychotic disorders are most resistant to brief psychosocial interventions. However, for first-episode schizophrenia, controlled studies have shown that intensive psychosocial counselling (of 6 months' duration or more) within small, residential settings provides results equivalent to drugs and hospitalization, with superior improvement in social integration. Because these interventions require major changes in philosophies, methods, and staffing, they remain marginal. In the larger economic and organizational context of mental health care today, clinicians, scientists and funding bodies have few incentives to promote or even evaluate them carefully. Peer-run alternatives (often part of the self-help movement) also suggest that informal but intensive support is essential for formerly hospitalized persons to regain normalcy.

Disadvantages of drug treatment include adverse effects, interference with learning, possibility of developing psychological or physical dependence, withdrawal difficulties, and the implicit message that the solution to the problem is beyond one's control or responsibility. Some advantages are quick onset of action, no need to question or change lifestyle, inexpensive in the short-term, easily available.

Disadvantages of psychological treatment are higher cost compared to drugs, hit-or-miss process of finding a proper therapist, requirement of active client resolve and verbal skills. Advantages include non-invasiveness, social interactiveness, longer lasting effects, educational nature (clients actively learn how to prevent future recurrences), and empowering impact.


In conclusion, it may be useful to consider the following points when trying to reach a judgment about the value of psychotropic drugs in the treatment of emotional or behavioral problems:

1. Because we do not know the biological causes of any routinely encountered behavior disorder, drugs are only expected to relieve symptoms, never to completely resolve a condition. Furthermore, we do not know precisely how any psychiatric drug acts to produce the effects we seek, nor can we reliably predict how a given drug will affect a given individual.

2. Response to an "active" drug (or an "inert" placebo) is a complex phenomenon, involving ill-understood social, psychological, and physiological processes. A consistent research finding across drug classes is that, on average and in the short term, 2 persons in 3 respond positively, including 1 in 3 responding to placebo. Therefore, "net" positive drug effect is 1 person in 3.

3. No drug has objectively defined "main" and "side effects": all drugs have many different effects, and users, prescribers, and interested parties may differ on how these effects should be categorized. How subtle and persistent undesirable effects of psychotropic drugs impact patients' quality of life remains generally unknown.

4. The public in general and patients in particular remain poorly informed with respect to the efficacy of specific drugs and general drug classes, adverse and withdrawal effects, and alternatives to drugs.

5. Most psychotropics in society follow this cycle: first, enthusiastic acceptance; second, balanced appraisal; third, limited use or even outright rejection. For this last reason, many "medications" become "dangerous drugs," and vice-versa, or both at once. The current place of a psychotropic in society has less to do with its known chemical properties than with popular or medical approval.



Gorman, J.M. (1990). The Essential Guide to Psychiatric Drugs. New York: St. Martin's Press.


Baldessarini, R.J., & Viguera, AC. (1995). Neuroleptic withdrawal in schizophrenic patients. Archives of General Psychiatry, vol. 52, no. 6, pp.189-192.

Hegarty, J., Baldessarini, R.J., Tohen, M., Waternaux, C., & Oepen, G. (1994). One hundered years of schizophrenia: A meta-analysis of the outcome literature. American Journal of Psychiatry, vol. 151, no. 11, 1409-1416.

Lehman, A.F. (1996). EValuating outcomes of treatments for persons with psychotic disorders. Journal of Clinical Psychiatry, Vol. 57 (supplement 11), pp. 61-67.


Liberman, J., & Kane, J.M. (Eds.). (1993). Adverse Effects of Psychotropic Drugs. New York: Guilford.

Keshavan, M.S., & Kennedy, J.S. (1993). Drug-Induced Dysfunction in Psychiatry. New York: Hemisphere.


Fisher, S., & Greenberg, R.P. (Eds.). (1989). The Limits of Biological Treatments for Psychological Distress: Comparisons with Psychotherapy and Placebo. Hillsdale, NJ: LEA.

Seligman, M.E.P. (1995). The effectiveness of psychotherapy: The Consumer Reports study. American Psychologist, vol. 50, no. 12, pp. 965-974.


Ciompi, L., Dauwalder, H.P., Maier, C., et al. (1992). The pilot project 'Soteria Berne': Clinical experiences and results. British Journal of Psychiatry, vol. 161 (supplement 18), pp. 145-153.

Birchwood, M., & Shepherd, G. (1992). Controversies and growing points in cognitive-behavioral interventions for people with schizophrenia. Behavioural Psychotherapy, vol. 20, pp. 305-342.


Breggin, P.R. (1997). Brain Disabling Treatments in Psychiatry: Drugs, Electroshock, and the FDA. New York: Springer. Fisher, S., & Greenberg, R.P. (Eds). (1997). From Placebo to Panacea: Putting Psychotropic Drugs to the Test. New York: John Wiley.