Relative toxicity of benzodiazepines in overdose

BMJ 1995; 310 doi: (Published 28 January 1995)
Cite this as: BMJ 1995;310:219
  1. N A Buckley, lecturera,
  2. A H Dawson, staff specialista,
  3. I M Whyte, staff specialista,
  4. D L O'Connell, senior lecturer in biostaticsb
  1. a Departments of Clinical Toxicology and Pharmacology, University of Newcastle and Newcastle Mater Misericordiae Hospital, Newcastle, New South Wales 2298, Australia
  2. b Centre for Clinical Epidemiology and Biostatistics, University of Newcastle, Newcastle, New South Wales, Australia
  1. Correspondence to: Dr Buckley.
  • Accepted 23 November 1994


Objective: To assess the sedative effects in overdose of temazepam and oxazepam compared with other benzodiazepines to determine if this explains reported differences in fatal toxicity

Design: Cohort study of patients admitted with benzodiazepine poisoning

Setting: Newcastle, Australia

Subjects: 303 patients who had ingested benzodiazepine alone or in combination with alcohol and presented to a general hospital which served a well defined geographical area

Main outcome measures: Degree of sedation: Glasgow coma score, McCarron Score, and whether patients were stuporose or comatose

Results: Oxazepam produced less and temazepam more sedation than other benzodiazepines. Unadjusted odds ratios for coma with oxazepam and temazepam compared with other benzodiazepines were 0.0 (95% confidence interval 0.0 to 0.85) and 1.86 (0.68 to 4.77) respectively, x2=7.08, 2df, P=0.03. After adjustment for potentially confounding effects of age, dose ingested, and coingestion of alcohol, the odds ratios were 0.22 (0.0 to 1.43) for oxazepam and 1.94 (0.57 to 6.23) for temazepam. Similar results were obtained for other measures of sedation

Conclusions: These results were in accordance with fatal toxicity indices derived from coroners' data on mortality and rates of prescription. The relative safety of benzodiazepines in overdose should be a consideration when they are prescribed

Key messages

  • Key messages

  • In a recent study of coronial data large differences were noted in the number of fatal poisonings caused by different benzodiazepines (after correction for market share)

  • This study provides evidence that there are differences, both before and after adjustment for a number of potential confounders, between benzodiazepines in the incidence with which they cause major sedation in overdose

  • These clinical differences provide explanation and support for the coronial data

  • Benzodiazepines cannot be assumed to be safe in patients at risk of self poisoning, but some benzodiazepines seem to be safer than others


Benzodiazepines are generally thought to be safe in overdose.1 2 Death after admission is rare and due to respiratory depression with aspiration of gastric contents.2 Over 10 years in the United Kingdom, however, 1512 fatal poisonings have been attributed to benzodiazepines with or without alcohol.3 These were compared with prescription data to establish a fatal toxicity index (deaths per million prescriptions) for each benzodiazepine. Similar indices have been derived for antidepressants4 and barbiturates.5 There were clear differences between benzodiazepines. Of drugs frequently prescribed, temazepam had the highest number of deaths per million prescriptions at 11.9 (95% confidence interval 10.9 to 12.8); above that of some tricyclic antidepressants.3 4 In contrast, oxazepam had an index of 2.3 (1.2 to 3.4), and the index for all benzodiazepines combined was 5.7.

Although there are potential sources of error in these studies,6 a bias that would lead to differences between compounds was not identified.3 Clinical studies can adjust for potential confounders which studies that use coronial data are unable to take into account. If differences between the benzodiazepines are supported by data from clinical studies this also adds credence to the fatal toxicity index which first noted these findings.

Our aim was therefore to determine if temazepam caused more sedation and oxazepam less sedation than other benzodiazepines when taken in overdose.


This was a follow up study of consecutive presentations to hospital after self poisoning with benzodiazepines between January 1991 and January 1994. The department has a regional responsibility for all poisonings in the lower Hunter Valley (population about 350000). The data, collected prospectively by casualty doctors and subsequently verified by the clinical toxicology team, included patient's characteristics (age, sex), all drugs and dose ingested, coingested substances, regular medication, history of abuse of drugs or alcohol, or both, details of management, and complications of poisoning. The state of intoxication was determined by three different but overlapping methods. These were identical with those described by McCarron et al for assessing the severity of barbiturate intoxication.7 A 7 point scale of conscious state (alert, drowsy, stuporous, coma 1–4) was used. Deeper levels of coma indicate loss of response to painful stimuli, inadequate respiration, and hypotension. The Glasgow coma score and McCarron score (a modified Glasgow coma score which includes scores based on vital signs)7 were also calculated.


Because of the large additive effect on sedation, patients who ingested more than one sedative drug were excluded from further analysis. The outcomes analysed were whether patients were stuporose or comatose on presentation and the mean Glasgow coma and McCarron scores. The differences in potency between benzodiazepines were adjusted for by converting the amount (mg) ingested to defined daily doses.8 To investigate the strength of the associations between temazepam and oxazepam and the main clinical outcomes we calculated the odds of outcome in those exposed and those not exposed. Odds ratios were adjusted for age, sex, coingestion of alcohol, chronic benzodiazepine use, and dose ingested by logistic regression by using maximum likelihood or an exact method.9


During 1991–3, 542 patients with benzodiazepine poisoning presented to this hospital, 239 of these patients, however, had ingested either more than one benzodiazepine or coingested other sedating drugs. The drugs ingested by the remainder were temazepam (64), oxazepam (45), diazepam (113), clonazepam (24), flunitrazepam (21), nitrazepam (18), others (18). Table I compares the characteristics of patients ingesting these drugs.

View this table:

Comparison of characteristics of patients ingesting different benzodiazepines

Oxazepam was less and temazepam was more sedating than other benzodiazepines with significance obtained for oxazepam in both the unadjusted and adjusted analyses (table II). (In the analysis of coma the odds ratios were adjusted for fewer variables as the others were not needed.)

View this table:

Details of coma scores and odds ratios (95% confidence intervals) for sedation for oxazepam and temazepam (compared with all other benzodiazepines)


Our results show that there are differences between temazepam, oxazepam, and other benzodiazepines in the degree of sedation they cause in overdose, and the observed differences are not due to confounding by age, sex, dose ingested, coingestion of alcohol, chronic benzodiazepine use, or history of drug or alcohol abuse. This provides a plausible explanation why temazepam and oxazepam have different fatal toxicity indices from other benzodiazepines.3

The sedation produced by benzodiazepines in therapeutic doses and overdose has a poor correlation with measured drug concentration but is increased with rapid absorption.1 10 11 12 Temazepam is more rapidly absorbed and oxazepam is more slowly absorbed than most other benzodiazepines.1 10 11 12 13 Further research is required to determine if the rate of absorption is different in overdose and is sufficient to explain the differences in sedation. Slowing the rate of absorption may reduce toxicity, but this would also reduce their sedative effect in therapeutic doses.11 Drug regulatory authorities should be aware that changes in formulation of benzodiazepines may affect toxicity in overdose.

Pharmacodynamic factors such as benzodiazepine receptor affinity and potency may also be important. Because of the wide variations in half life, adjustments for dose by conversion into defined daily doses or diazepam equivalents14 is imperfect. These are designed to compare use rather than potency. Though they correlate reasonably well with sizes of prescriptions and tablets, they may not account for potency pertablet taken in overdose. Differences in potency could also explain the results in both our study and that of Serfaty and Masterton.3

Flurazepam, now rarely prescribed. in the United Kingdom and Australia, had the highest fatal toxicity index of any benzodiazepine (15.0).3 It has also been found to be more sedating in overdose than diazepam and chlordiazepoxide.10 That study and our own support the differences between drugs noted in the fatal toxicity index.3 If the coroners' data on which this index is based are reliable enough to detect small differences in toxicity between drugs then it gives credence to other work based on these data, according to which, temazepam is the sixth most common cause of death attributable to a single drug in the United Kingdom.15 in contrast, there have been only a few deaths directly attributed to benzodiazepine poisoning reported in the medical literature.*RF1, 16, 17* Benzodiazepines are the most common prescription drugs taken in overdose.18 Although all these drugs are much safer than barbiturates and sedating tricyclic antidepressants,3 4 5 they should not be prescribed to patients at high risk of suicide. Differences between benzodiazepines may be relevant when prescriptions are dispensed for depressed or suicidal patients, and appropriate choices may lead to a considerable reduction in lives lost to suicide.

We are grateful to Peter McManus, secretary of the Australian Drug Utilization Subcommittee, for market share data.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.