to the editor: I read with interest the article “Generalized Anxiety Disorder.”¹ It is a much needed general summary of a disease that affects a great percentage of our patients. I was distressed to see, however, the emphasis on benzodiazepines as treatment. After a short discussion of psychologic treatment, Dr. Gliatto devotes more than one and one-half pages to benzodiazepines. Finally, there is a small “other medications” section to cover the remainder of pharmacologic treatments, including herbal therapies.

While the section on benzodiazepines was well-written, it overshadows the other safe and effective treatments that exist. Benzodiazepines should not be first-line therapy for generalized anxiety disorder. As Dr. Gliatto describes, drugs such as buspirone (BuSpar) may be impaired in their effectiveness if given after treatment with benzodiazepines. In addition, many patients who have experienced the rapid benefits of benzodiazepines seem less willing to accept the slower benefit of other medications and nonpharmacologic therapy. In addition, there are no studies documenting the benefit of long-term benzodiazepine therapy, suggesting it would not be a good drug for use as treatment for a long-term disease. Finally, the concept that “benzodiazepines in therapeutic dosages do not lead to abuse” is frankly incorrect. Benzodiazepines can be abused in standard therapeutic dosages resulting in significant medical complications. Benzodiazepines, while valuable for rapid treatment of symptoms of anxiety, are not a cure for generalized anxiety and should be considered second-line treatment to other therapies with proven efficacy.

to the editor: I appreciated the editorial¹ that accompanied the article² by Dr. Gliatto on generalized anxiety; the editorial was a valuable counterpoint. Adding to the editorial comments on the use of benzodiazepines (BZDPs), I respectfully disagree with the following citation from the article: "Use of benzodiazepines in therapeutic dosages does not lead to abuse, and addiction is rare."³ I teach my residents that treating patients with anxiety with BZDPs is akin to treating pain with morphine: nothing makes your patient feel that good, and efforts to wean the patient to another agent are very frustrating to the patient and the physician. I had many patients clamor for more BZDPs—and while they were not addicted, their pattern of drug-seeking resembled abuse. A Navy veteran patient recognized this in saying, "Doctor, I took myself off that stuff. It made me feel too good, and I was getting to where I was thinking too much about my next dose!"

In my opinion, cognitive, behavioral, and non-benzodiazepine therapies are distinctly preferable in the treatment of patients with generalized anxiety disorder.

in reply: The comments of Drs. Knudson and Smith attest to the ambivalence physicians, including myself, feel in prescribing benzodiazepines for anxiety. However, several points are in order. In my article,¹ I did not intend to minimize the importance of psychotherapy as a treatment modality. My experience teaching medical students and nonpsychiatric colleagues is that they are interested in psychotherapy, but because of the nature of their practices they are more interested in medical management; hence, I wrote more about medications than psychotherapy. However, as I said in the article, patients have complicated life histories and social problems, and physicians must use both psychotherapeutic techniques (usually supportive and family therapy techniques) and medications to treat patients who are anxious. Also, access to quality psychotherapists may not be available in particular sections of the country. I wrote more about benzodiazepines than other medications because they are effective, safe, and many doctors have questions about their use, especially concerning the potential for abuse.

I think Dr. Smith's comments comparing the use of morphine and benzodiazepines are not apt. If a patient has a recent fracture, does he advocate withholding opiates because "nothing makes your patient feel that good?" When a patient is anxious, one must do a through assessment, and patients with histories of substance abuse and personality disorder^2,3 should not be given benzodiazepines. I grant that assessment of personality disorders can be difficult, but one should be wary of patients who see multiple doctors and have had multiple prescriptions for sedatives. According to an American Psychiatric Association Task Force about benzodiazepines,4 prescribing benzodiazepines in appropriate patients rarely leads to abuse and the number of patients who use benzodiazepines responsibly is far greater than patients who do not.^3,4

I agree with Dr. Knudson that there are no long-term studies that support the use of benzodiazepines on a chronic basis, and efforts should be made to wean patients off these medications. The Task Force also found that chronic users of benzodiazepines tend to decrease the dose over time.⁴

Although I did not in the article, I should emphasize that benzodiazepines are riskier to use in elderly patients because of the propensity for falls and increased confusion.3 In these patients, other agents should be used first; physicians must use clinical judgment and prescribe what is best for the patient.