Co-proxamol and suicide: a study of national mortality statistics and local non-fatal self poisonings

BMJ 2003; 326 doi: (Published 10 May 2003) Cite this as: BMJ 2003;326:1006
  1. Keith Hawton (keith.hawton{at}, professor of psychiatrya,
  2. Sue Simkin, researcher and coordinatora,
  3. Jonathan Deeks, senior medical statisticianb
  1. a Centre for Suicide Research, University of Oxford Department of Psychiatry, Warneford Hospital, Headington, Oxford OX3 7JX
  2. b Centre for Statistics in Medicine, Institute of Health Sciences, Headington, Oxford OX3 7LF
  1. Correspondence to: K Hawton
  • Accepted 5 March 2003


Objectives: To examine the incidence of suicides due to co-proxamol compared with tricyclic antidepressants and paracetamol, and to compare fatality rates for self poisonings with these drugs.

Design: Analysis of routinely collected national and local data on suicides and self poisonings.

Setting: Records of suicides in England and Wales 1997-9; non-fatal self poisonings in Oxford District 1997-9.

Data sources: Office for National Statistics and Oxford monitoring system for attempted suicide.

Main outcome measures: Incidence of suicides with co-proxamol or tricyclic antidepressants or paracetamol. Ratios of fatal to non-fatal self poisonings.

Results: Co-proxamol alone accounted for 5% of all suicides. Of 4162 drug related suicides, 18% (766) involved co-proxamol alone, 22% (927) tricyclic antidepressants alone, and 9% (368) paracetamol alone. A higher proportion of suicides in the 10-24 year age group were due to co-proxamol than in the other age groups. The odds of dying after overdose with co-proxamol was 2.3 times (95% confidence interval 2.1 to 2.5) that for tricyclic antidepressants and 28.1 times (24.9 to 32.9) that for paracetamol.

Conclusions: Self poisoning with co-proxamol is particularly dangerous and contributes substantially to drug related suicides. Restricting availability of co-proxamol could have an important role in suicide prevention.

What is already known on this topic

What is already known on this topic Co-proxamol is dangerous in overdose

Restricting availability of specific means of suicide can reduce deaths

What this study adds

What this study adds Fatal overdoses due to co-proxamol are the second most frequent means of suicide with prescribed drugs in England and Wales

The risk of death associated with co-proxamol overdose seems to be higher than for either tricyclic antidepressants or paracetamol


Restriction of availability of means for suicide is a key strategy for prevention of suicide. 1 2 This approach has been shown to be effective by the reduction in deaths after recent UK legislation that reduced pack sizes of analgesics3 and previous restriction of prescribing toxic sedatives.4

Co-proxamol is a prescription only analgesic that combines paracetamol and dextropropoxyphene. Respiratory depression and consequent death may occur with overdose due to ingestion of excessive dextropropoxyphene. 5 6 Concern about the number of such deaths was expressed in the BMJ as long ago as 1980.7

We compared the numbers of suicides from poisoning with co-proxamol, paracetamol, and tricyclic antidepressants (these being relatively common methods in poisoning suicides) in England and Wales. We also compared fatal and non-fatal self poisonings to estimate the relative fatality of overdoses with these three drugs.


Mortality data—We obtained data from the Office for National Statistics on deaths in people aged 10 years and over in England and Wales for 1997-9. We considered those deaths that involved drugs and medicines in which a verdict of suicide (international classification of diseases, ninth revision, codes E950.0-E959.5) or undetermined cause (codes E980.0-E989.5) was recorded. Deaths involving co-proxamol, paracetamol, or tricyclic antidepressants were identified by searching for all variants of description of these drugs in the textual fields for cause of death.

Non-fatal self poisonings—Non-fatal self poisonings with co-proxamol, paracetamol or tricyclic antidepressants alone for 1997-9 were identified through the Oxford monitoring system for attempted suicide.8 This records all presentations to the general hospital in Oxford of deliberate self harm. The comprehensiveness and reliability of the data has previously been shown.9 The pattern of drugs used for self poisoning in the Oxford area is similar to that seen elsewhere.10

Statistical analyses—We used Poisson regression to compute estimates, confidence intervals, and comparison of death rates and presentation rates according to drug, age, and sex. Odds ratios for the relative lethality of different drugs were calculated by computing ratios of relative death rates to relative non-fatal presentation rates. Confidence intervals for relative lethality were calculated with Monte Carlo methods. We used Stata release 7.0 (StataCorp, College Station, TX, USA) for the analyses.


In England and Wales in 1997-9, 15 299 deaths were recorded as suicides or open verdicts. Of these, 4162 (27%) were drug related (such as self poisoning). There were more drug related deaths in men than in women, although a higher proportion of women who killed themselves did so by self poisoning (1835/3762 (49%) v 2327/11 537 (20%) in men).

Deaths due to self poisoning with co-proxamol

There were 766 deaths due to poisoning with co-proxamol alone during the three year period (255 per year, 95% confidence interval 238 to 274). These comprised 18% of all drug related deaths (table 1) and 5% of all suicides. After tricyclic antidepressants co-proxamol was the second most common prescribed drug used for suicide. More co-proxamol poisoning deaths occurred in men than in women (P=0.01). Although the total numbers of deaths increased with age, the proportion due to co-proxamol poisoning was significantly higher (24%) in 10-24 year-olds (P=0.01). In addition, there were 171 deaths in which co-proxamol was used with another drug.

Table 1

Details of suicides by co-proxamol, tricyclic antidepressants, and paracetamol, England and Wales, 1997-9. Figures are numbers of suicides (percentage of all drug related suicides in age group), unless stated otherwise

View this table:

Comparison with deaths due to tricyclic antidepressants and paracetamol

There were an average 309 (289 to 330) deaths per year due to tricyclic antidepressants alone. These deaths comprised 22% of all drug related suicides, significantly more than for co-proxamol alone (P<0.001). Compared with men, women were more likely to use tricyclics than co-proxamol (P=0.02). Use of tricyclics was similar across age groups (P=0.07). In 270 deaths tricyclics were taken with other drugs.

An average 123 (110 to 136) deaths per year were due to paracetamol alone, and this comprised 9% of all drug related deaths (table 1). Compared with men, women were more likely to use paracetamol that co-proxamol (P=0.02), and its use increased more clearly with age (P<0.001). In 128 deaths paracetamol was taken with other drugs.

Fatal and non-fatal self poisonings

Table 2 gives details of annual numbers of non-fatal self poisonings. Non-fatal tricyclic antidepressant and paracetamol overdoses presented 2.7 (2.1 to 3.6) times and 13.5 (10.7 to 17.0) times more frequently than non-fatal co-proxamol overdoses, respectively.

Table 2

Comparison of numbers (95% confidence intervals) of drug related suicides and undetermined deaths in England and Wales with non-fatal self poisoning in Oxford, 1997-9, for co-proxamol, paracetamol, and tricyclic antidepressants (used alone). Odds ratios shown with 95% confidence intervals

View this table:

Comparison of ratios of death and non-fatal presentations suggests that the odds that an overdose will be fatal with co-proxamol is 2.3 (2.1 to 2.5) times higher than for tricyclic overdoses and 28.1 (24.9 to 32.9) times higher than for paracetamol overdoses.


In England and Wales co-proxamol is the second most common prescribed drug that people use to commit suicide and is used as the sole method in 18% of all drug related suicides for 1997-9 and 5% of all total suicides. Comparison with non-fatal poisonings indicates that overdoses of co-proxamol are more likely to result in death than overdoses with tricyclic antidepressant or paracetamol. Death can result from an overdose with relatively few tablets, especially when alcohol is also taken. 5 6 Given earlier concerns about deaths from poisoning with to co-proxamol 5 7 the absence of specific initiatives to try to reduce them is surprising and should now be addressed.1

Factors to be considered in tackling this problem are that co-proxamol is a prescription only drug, patients with pain already have an increased risk of suicide, 11 12 the risk of self poisoning is not restricted to the person for whom the drug is prescribed,13 and, as with paracetamol,14 availability within a household may be an influential factor, especially in impulsive overdoses. Also, while co-proxamol is regarded as an important analgesic, 15 16 a systematic review has shown that it is no more effective than paracetamol for short term relief of pain.17

Reducing the availability of drugs used for suicide can result in a reductions in deaths, 3 4 and availability of co-proxamol should be restricted. Clinicians must be informed about the risks of overdose of co-proxamol, both for their patients and others in the household, and large quantities should not be prescribed without good reason. Patients should be instructed to dispose of unwanted supplies. Finally, clinicians should consider whether there are other equally effective but less dangerous methods of pain relief, such as combining a safer analgesic with an opiate while maximal relief is required.


We thank Alan Baker of the Office for National Statistics, and Liz Bale, Alison Bond, and the staff of the Department of Psychological Medicine, John Radcliffe Hospital.

Contributors: KH initiated the study, SS coordinated the data collection, and JD conducted the statistical analyses. All authors contributed to the study design, interpretation of the results, and preparation of the report. KH is guarantor for the study.


  • Funding South East Region NHSE Research and Development Committee. KH is also supported by Oxfordshire Mental Healthcare NHS Trust. The guarantor accepts full responsibility for the conduct of the study, had access to the data, and controlled the decision to publish.

  • Competing interests None declared.


  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.
View Abstract