Being Careful With Benzodiazepines

I frequently see patients in the hospital who receive benzodiazepines from their doctors in outpatient clinics, both psychiatric and primary care. Many of them are diverting their prescriptions, overusing the benzodiazepines, and getting them for conditions which are not indications for this class of drug.

I guess my perspective on benzodiazepines probably speaks to the different patient population I treat. In my role as a consulting psychiatrist in a general hospital, benzodiazepines are frequently administered together with opioids. The result is often delirium, and when I’m called and I check the medication list as well as the medical problems associated with delirium, I recommend stopping the benzodiazepine and moderating the use of anticholinergic drugs, including opioids if at all possible.

I also see many patients who’ve overdosed on benzodiazepines in the ICUs, often with other drugs to be sure, but they’re pretty sick despite the often repeated declaration that they’re safer in overdose. They are safer than barbiturates. Lorazepam has been identified as being an independent predictor of delirium in the ICU, (Pandharipande P, Shintani A, Peterson J, Pun BT, Wilkinson GR, Dittus RS, et al. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anesthesiology. 2006; 104(1):21-6).

When I’m called to evaluate alcohol withdrawal delirium, I sometimes find benzodiazepine intoxication delirium from overcalling the withdrawal. It happens, despite using symptom-triggered withdrawal protocols.

I am often called to see patients who struggle with alcoholism and other substance abuse including benzodiazepine abuse, almost always active addictions. Some of them ask for benzodiazepines for their “anxiety”, which few can describe with any clarity— as long as I don’t coach them by describing the typical syndromes myself.

On the other hand, occasionally I see catatonia in the general hospital and the critical care unit. I’ve seen sedative-hypnotic, anxiolytic withdrawal-induced catatonia (Amos JJ, M.D. Lorazepam withdrawal-induced catatonia. Annals of Clinical Psychiatry. 2012; 24(2):170-1). Of course, the treatment for both initially begins with giving the patients benzodiazepine, which wakes them up and sometimes looks like a miracle to med-surg nurses, medical residents, and medical students.

And the latest Cochrane Database Survey on benzodiazepines for aggression or agitation in the setting of psychosis:

Information sourced from The Cochrane Library:

Benzodiazepines for psychosis-induced aggression or agitation



Acute psychotic illness, especially when associated with agitated or violent behaviour, can require urgent pharmacological tranquillisation or sedation. In several countries, clinicians often use benzodiazepines (either alone or in combination with antipsychotics) for this outcome.


To estimate the effects of benzodiazepines, alone or in combination with antipsychotics, when compared with placebo or antipsychotics, alone or in combination with antihistamines, to control disturbed behaviour and reduce psychotic symptoms.

Search methods

We searched the Cochrane Schizophrenia Group’s register (January 2012), inspected reference lists of included and excluded studies and contacted authors of relevant studies.

Selection criteria

We included all randomised clinical trials (RCTs) comparing benzodiazepines alone or in combination with any antipsychotics, versus antipsychotics alone or in combination with any other antipsychotics, benzodiazepines or antihistamines, for people with acute psychotic illnesses.

Data collection and analysis

We reliably selected studies, quality assessed them and extracted data. For binary outcomes, we calculated standard estimates of relative risk (RR) and their 95% confidence intervals (CI) using a fixed-effect model. For continuous outcomes, we calculated the mean difference (MD) between groups. If heterogeneity was identified, this was explored using a random-effects model.

Main results

We included 21 trials with a total of n = 1968 participants. There was no significant difference for most outcomes in the one trial that compared benzodiazepines with placebo, although there was a higher risk of no improvement in people receiving placebo in the medium term (one to 48 hours) (n = 102, 1 RCT, RR 0.62, 95% CI 0.40 to 0.97, very low quality evidence).

There was no difference in the number of participants who had not improved in the medium term when benzodiazepines were compared with antipsychotics (n = 308, 5 RCTs, RR 1.10, 95% CI 0.85 to 1.42, low quality evidence); however, people receiving benzodiazepines were less likely to experience extrapyramidal effects (EPS) in the medium term (n = 536, 8 RCTs, RR 0.15, 95% CI 0.06 to 0.39, moderate quality of evidence). Data comparing combined benzodiazepines and antipsychotics versus benzodiazepines alone did not yield any significant results.

When comparing combined benzodiazepines/antipsychotics (all studies compared haloperidol) with the same antipsychotics alone (haloperidol), there was no difference between groups in improvement in the medium term (n = 155, 3 RCTs, RR 1.27, 95% CI 0.94 to 1.70, very low quality evidence) but sedation was more likely in people who received the combination therapy (n = 172, 3 RCTs, RR 1.75, 95% CI 1.14 to 2.67, very low quality evidence). However, more participants receiving combined benzodiazepines and haloperidol had not improved by medium term when compared to participants receiving olanzapine (n = 60, 1 RCT, RR 25.00, 95% CI 1.55 to 403.99, very low quality evidence) or ziprasidone (n = 60, 1 RCT, RR 4.00, 95% CI 1.25 to 12.75 very low quality evidence). When haloperidol and midazolam were compared with olanzapine, there was some evidence the combination was superior in terms of improvement, sedation and behaviour.

Authors’ conclusions

The evidence from trials for the use of benzodiazepines alone is not good. There were relatively little good data and most trials are too small to highlight differences in either positive or negative effects. Adding a benzodiazepine to other drugs does not seem to confer clear advantage and has potential for adding unnecessary adverse effects. Sole use of older antipsychotics unaccompanied by anticholinergic drugs seems difficult to justify. Much more high quality research is needed in this area.

Gillies D, Sampson S, Beck A, Rathbone J. Benzodiazepines for psychosis-induced aggression or agitation. Cochrane Database of Systematic Reviews 2013, Issue 9. Art. No.: CD003079. DOI: 10.1002/14651858.CD003079.pub4.

[Free full-text PDF] [Cochrane Library abstract]

Copyright © 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

The above message comes from The Cochrane Library, who is solely responsible for its content.

Doctors come from different training backgrounds and clinical settings differ. That doesn’t mean we’re right or wrong about the use of benzodiazepines, generally. It means we’re wise to remain aware of the importance of clinical context. Benzodiazepines are not bad in and of themselves. Physicians and their patients just need to have a conversation about why and how long they are going to use them. Benzodiazepines can be the right thing for some people at a particular time in their lives. On the other hand, sometimes we start off with the best of intentions with this drug, but run into all sorts of unintended consequences.



  1. Benzodiazepines are good drugs for catatonia, some acute movement disorders (akathisia), alcohol and benzodiazepine detox, some sleep disorders, terminating status epilepticus and treatment of acute agitation and aggression in combination with an antipsychotic. The Cochrane report is a good example of the limitations of an evidence based approach. Like may Cochrane studies the authors’ complain about small sample sizes, variable outcome measures (that do not appear to be equivalent) and treat some expected outcomes (less EPSE and more sedation in the benzodiazepine group).

    On the other hand benzodiazepines are not that useful for anxiety and don’t add anything when added into polypharmacy scenarios. Most importantly – there is no practical way for a clinician to predict who will have problems with benzodiazepines and who will not. There are still many clinicians who maintain that benzodiazepines can be safely prescribed to patients with benzodiazepine and alcohol use disorders but I remain very skeptical. I am skeptical that most people who are prescribed a drug for an anxiety disorder have no access to psychotherapy when even computer delivered psychotherapy is useful for many.

    When I first started running and acute inpatient service in the 1980s, acute agitation was being treated with rapidly escalating doses of typical antipsychotics. Doses of haloperidol in excess of 100 mg/day or other typicals like thiothixene and fluphenazine were being prescribed in similar high doses. I am not sure why that practice stopped, but it did provide widespread evidence that just using antipsychotic medications alone to treat agitation and aggression does not necessarily work. Adding a benzodiazepine works and prevents the patient from being exposed to excessive doses of antipsychotics. It also prevents injury to patients and staff in many of these situations.


  2. And the above is for those who are aware that benzodiazepines are still widely prescribed for patients with PTSD despite the evidence that they tend to interfere with recovery. I think the genetic and brain structural changes correlating with cognitive behavioral therapy (CBT) speak to the efficacy of psychotherapy, too often considered a non-biological treatment.


%d bloggers like this: