GAD Pharmacotherapy: Addressing Comorbidities and Special Populations

David N. Osser, MD

Associate Professor of Psychiatry
Harvard Medical School

Last updated: September 21, 2018


  • In elderly patients, two SSRIs are preferred: sertraline and escitalopram
  • In patients with comorbid GAD and PTSD:
    • Prazosin added to an antidepressant is recommended to address sleep difficulties, daytime hyperarousal, and irritability
    • Avoid benzodiazepines because of lack of effectiveness in PTSD and abuse potential
  • In patients with comorbid GAD and bipolar depression: consider using hydroxyzine, pregabalin, and benzodiazepines


Now, we turn to node 2 of the algorithm where we now do evaluate for comorbidity and other special patient populations and explain how there will be differences in the algorithm we’re about to show you if they have these different circumstances and comorbidities.

The first one we’re going to discuss is sleep disturbance or insomnia which is a common problem in people with GAD. There are multiple causes of insomnia. You can’t just assume it’s part of the GAD. Some of the treatments that we have actually worsen insomnia. So that may lead you to want to skip those options and consider other ones. For example, SSRIs and SNRIs generally are not effective for insomnia and may make it worse. Some 10% to 20% of patients given those products can have worsened or new onset of insomnia as a side effect.


We do have more sedating agents among our options for GAD. They include drugs like hydroxyzine and pregabalin. I should point out that pregabalin is not approved in the US for GAD nor is hydroxyzine. But these might be preferred over SSRIs or SNRIs when there is prominent insomnia. Another issue or option for dealing with insomnia due to the SSRIs could be to add trazodone. There are two controlled studies where they added trazodone or placebo to an SSRI for insomnia and found good efficacy. And another thing you could add would be benzodiazepines for the insomnia. The problem with using benzodiazepines right away is you’re stuck on having to keep the patient on them forever because most often they cause rebound effects the night after stopping them. If you give drugs like zolpidem and benzos for sleep, interesting studies have shown that if you give one group the benzo for one night, another group placebo for one night, then on the second night you give a placebo to both groups, the people who got better sleep the first night on the benzo, their sleep now on the second night on placebo is worse than the group that got placebo both nights. Remarkable. As soon as they miss one night of a benzo, their sleep is even worse than before. So what do patients conclude from that? They conclude I better keep taking this. So that’s one of the real negatives about using benzos as sleeping pills and zolpidem.


Another group, special patient population is elderly people. Now, we don’t have a strong difference in the algorithm for them but we do point out that all medications have side effects and they all tend to be worse in the elderly. So you have to monitor those side effects closely. SSRIs that we would prefer in the elderly would be sertraline and escitalopram. They have a little bit better safety profile.


There is a problem in the elderly of gait impairment so called postural sway problems with SSRIs that is greater than in younger people. And gastrointestinal bleeding is a concern in all age groups but it’s worse in the elderly. Same thing with bone loss, low sodium levels.


SNRIs, they can raise blood pressure, at least venlafaxine. That can be worse in the elderly. And pregabalin, if you use that, it causes sleepiness, dizziness and falls. Fractures have occurred in some of the controlled work with pregabalin in the elderly. Basically, you have to monitor them much more closely. Benzodiazepines, they are also more side-effect-inducing in the elderly. Falls are what you worry about. Decreased respiratory drive especially if they have sleep apnea that you either do or do not know about. Auto accidents are more frequent. Substance use sedation. So those are the issues with the benzos in the elderly. And quetiapine, metabolic risks and increase in QTc. You must monitor for those if you’re using quetiapine which we are very uninterested in using in GAD as you’ll be seeing a little bit later.


The next group of people for whom there are special considerations is women of childbearing potential and who may become pregnant during treatment. Generalized anxiety disorder is more common in women than men and it is particularly seen in younger women to have GAD.


So how will the algorithm be different for them? Well, benzodiazepines are going to drop down from what they would otherwise be. They are all category D rating for pregnancy due to risk of cleft palate. The one exception for that is clonazepam. C means no known abnormalities with the fetus have been proven but there have been isolated reports. D means there’s been consistent evidence of some problem, in this case, cleft palate. The SNRIs and SSRIs are all category C except paroxetine which is a category D. And the atrial septal defects are the concerns with paroxetine. There was a recent very large observational study by Reefhuis confirming that the category D is deserved by paroxetine.


The last point under women of childbearing potential is that SSRIs may affect risks of postpartum hemorrhage, premature delivery and other postnatal complications. These are statistical findings but they may be due to confounding by indication which is when you have uncontrolled studies where you find an increase in something but you don’t know if it’s because you gave them an antidepressant or whether the thing you were treating with your antidepressant is what actually causes those complications. This remains unclear.


The next comorbidity situation to discuss would be people who are actively abusing substances. Well, first of all, it’s a pretty common comorbidity. Some 15% of people with GAD are actively using some substance. So our recommendations for them that would be different from our standard would first of all benzos are going to go way down in the options list. Pregabalin which was one of our options we’re going to be considering drops way down because that’s a schedule IV. It is prone to abuse. Gabapentin probably has comparable abuse potential to pregabalin. They’re very similar compounds in many respects, structurally and how they work, where they work, the indications they have. Gabapentin somehow got by without getting labeled as a schedule IV but many reports suggest there is misuse going on. We’ll be discussing more about the evidence base on gabapentin and pregabalin for GAD later.


The next comorbidity is major depression which as I suggested is quite common. There is some overlap of the criteria for GAD in many aspects of these patients who are worrying. But studies have shown that you can clearly delineate who has GAD, who has depression and who has both if you work carefully with the DSM-5 criteria. You might be able to, in a matter of speaking, kill two problems with one drug. But unfortunately, that’s not what the evidence supports. The evidence actually seems to show that when you have major depression with comorbid anxiety the antidepressants work much less well than when they don’t have the comorbid anxiety.


So what does work well? We don’t have great studies showing us what to use for major depression and GAD together. But we would speculate that there is a role for benzos here as an augmentation of antidepressants. Another option is picking among the more sedating antianxiety agents such as hydroxyzine and pregabalin. For the anxiety part of it, they are not antidepressants though. You would still need to be using an antidepressant.


The next disorder that is an issue with comorbidity and could change your algorithm is bipolar depression. We just discussed major depression. But what if it’s bipolar depression? And in fact, the rates of GAD are higher in bipolar depression compared to unipolar depression. So you’re going to see this a lot. So what to do?


Antidepressants are normally effective for GAD as we will be explaining such as SSRIs. But we do not recommend them for bipolar depression especially if they’re rapid cycling or mixed rapid cycling being more than four episodes a year. Do not use antidepressants, in our opinion, for GAD in patients with bipolar depression or if they have bipolar mixed states even if they’re not rapid cycling or mixed. There are evidences that you are going to produce greater mood destabilization if you add antidepressants. Please read those references I have listed especially Pacchiarotti. So there was 80% agreement on all the recommendations in there. And it is totally consistent with what I’ve just told you. I make a stress on this because I know there are a lot of you out there using antidepressants for bipolar depression. We urge you not to do so.


What we do recommend for bipolar depression is shown in the next clip here. The four most effective, most evidenced treatments for bipolar depression are lithium, quetiapine, lamotrigine and lurasidone, not necessarily in that order. For the GAD that’s comorbid with this bipolar depression, we would consider skipping to where we recommend hydroxyzine, pregabalin and benzodiazepines. We would skip the antidepressants and go to those options. Now, it is kind of a surprise because – why didn’t I mention quetiapine?


Quetiapine is a treatment that is one of the effective treatments for bipolar depression. It also has had positive studies for GAD as a monotherapy for generalized anxiety disorder, two studies. So how about if the patient has both? So surprise. A study by Keming Gao in 2014 sponsored by quetiapine found no efficacy for the depression or the GAD. So I guess it seems like it wouldn’t be your first choice for someone who has that combination of problems.


And the last comorbidity I’m going to talk about is posttraumatic stress disorder. There are FDA-approved SSRIs for PTSD but we don’t recommend them. We recommend instead prazosin as your add-on for insomnia, nightmares, disturbed awakenings supported by four placebo-controlled studies. I refer you back to our complete algorithm for PTSD which is already in the series of talks along with this one. But also what’s not often fully understood about prazosin is that it also worked very well for the daytime symptoms of PTSD, the hyperarousal, the hypervigilance, the irritability. They gave up to 5 mg typically at 10 o’clock in the morning in the fourth of those four studies by Raskind and colleagues in 2013. They ramped them up to that. And they had excellent results on those other symptoms. What about benzodiazepines though? They are not thought effective for PTSD. They have high abuse potential on PTSD. We do not recommend them.

Key Points

The key points probably are as follows. That insomnia is a symptom. It’s not a disease usually. So you have to consider the differential diagnoses in your GAD patients and treat accordingly. A second key point, avoid benzodiazepines in patients with anxiety who are substance abusers. Next, avoid antidepressants in bipolar patients with anxiety. And the final bullet point then on summarizing this node 2 is to use prazosin for PTSD-related sleep problems and daytime hyperarousal and irritability added to the antidepressants they’re on for their GAD for this comorbidity.



  1. Holsboer-Trachsler E, Prieto R. Effects of pregabalin on sleep in generalized anxiety disorder. Int J europsychopharmacol 2013;16:925–36.
  2. (Kaynak 2004, Nierenberg 1994) (Pollack 2008, Walsh 2007)
  3. Benitez et al., Use of benzodiazepines and selective serotonin reuptake inhibitors in middle-aged and older adults with anxiety disorders: A longitudinal and prospective study. Am J Geriat Psychiatry 2008;16:5-13.
  4. Berger et al., Magnitude of potentially inappropriate prescribing in Germany among older patients with generalized anxiety disorder. BMC Geriatr 2009;9:31.
  5. Montgomery, S., Chatamra, K., Pauer, L., Whalen, E., & Baldinetti, F. (2008). Efficacy and safety of pregabalin in elderly people with generalised anxiety disorder. The British Journal of Psychiatry, 193(5), 389-394.
  6. Reefhuis J et al. Specific SSRIs and birth defects: Bayesian analysis to interpret new data in the context of previous reports. BMJ 2015:351;h3190
  7. Kessler RC et al., The epidemiology of co-occurring addictive and mental disorders: Implications for prevention and service utilization. Am J Orthopsychiatry 1996:66:17-31.
  8. Fava M et al. Difference in treatment outcome in outpatients with anxious versus non-anxious depression: A STAR*D report. Am J Psychiatry 2008;165:342-51.
  9. Abejuela, H.R. and Osser, D.N., 2016. The Psychopharmacology Algorithm Project at the Harvard South Shore Program: an algorithm for generalized anxiety disorder.
  10. Harvard review of psychiatry, 24(4), pp.243-256.
  11. Leverich GS, et al. Risk of switch in mood polarity to hypomania and mania in patients with bipolar depression during acute and continuation trials of venlafaxine, sertraline, and bupropion as adjuncts to mood stabilizers. Am J Psychiatry 2006;163:232-9.
  12. Pacchiarotti et al. The International Society for Bipolar Disorders (ISBD) task force report on antidepressant use in bipolar disorders. Am J Psychiatry 2013; 170:1249-62.
  13. Gao, K., Wu, R., Kemp, D.E., Chen, J., Karberg, E., Conroy, C., Chan, P., Ren, M., Serrano, M.B., Ganocy, S.J. and Calabrese, J.R., 2014. Efficacy and safety of quetiapine-XR as monotherapy or adjunctive therapy to a mood stabilizer in acute bipolar depression with generalized anxiety disorder and other comorbidities: a randomized, placebo-controlled trial. The Journal of clinical psychiatry, 75(10), pp.1062-1068.
  14. Bajor, L. A., Ticlea, A. N., & Osser, D. N. (2011). The Psychopharmacology Algorithm Project at the Harvard South Shore Program: an update on posttraumatic stress disorder. Harvard review of psychiatry, 19(5), 240-258.
  15. Raskind, M. A., Peterson, K., Williams, T., Hoff, D. J., Hart, K., Holmes, H., … & Millard, S. P. (2013). A trial of prazosin for combat trauma PTSD with nightmares in active-duty soldiers returned from Iraq and Afghanistan. American Journal of Psychiatry, 170(9), 1003-1010.

Also, you can access to the original version in Spanish:  “Farmacoterapia del TAG: Comorbilidades y Poblaciones Especiales”

Earn CME/SA credits

  • Up to 36 CME credits/year
  • Up to 16 SA credits/year
  • Download more than 70 videos
  • Exclusive updates

Free Course: “SSRIs: The Essentials”

Learn the essentials of SSRIs with our free online course.

One module per week:

  • The Mechanism of Action of SSRIs
  • Fluoxetine
  • Fluvoxamine
  • Paroxetine
  • Sertraline
  • Citalopram and Escitalopram

2019-20 Psychopharmacology Online Update

Earn Up to 36 CME Credits / Year

  • Access exclusive updates: new videos every month
  • Download our entire library: more than 70 videos
  • Earn up to 36 CME credits /year