Q J Med 2003; 96: 125-132
© 2003 Association of Physicians
Changes in paracetamol, antidepressants and opioid poisoning in Scotland during the 1990s
From the Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh 1 Information & Statistics Division of the Common Services Agency, Edinburgh, and 2 Public Health, Lothian Health, Edinburgh, UK
Received 12 February 2002 and in revised form 22 November 2002
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Overdose is one of the commonest causes of medical admissions to UK hospitals. In Scotland (pop. 5.1 million), all NHS hospital discharge data is uniquely linked to enable identification of individuals re-presenting with the same diagnosis.
Aim: To examine trends in discharges for poisoning, in particular paracetamol, antidepressants and opioids from 1990–99.
Design: Retrospective analysis.
Methods: Discharge data from the Scottish Morbidity Record (SMR01) and mortality data from the General Register Office for Scotland (GROS) were analysed for 1990–99 by age and gender for the relevant codes.
Results: Overall discharge rates increased until 1997, after which they fell. This pattern was seen in paracetamol-related discharges, but not for antidepressants or for opioids. Overdose was more common in females, except for opioids. Discharges related to opioids increased in an exponential manner over the decade, five-fold in women and six-fold in men in 10 years.
Discussion: Increases in opioid-related presentations are of major concern. Changes in paracetamol pack-size have been associated with reduced discharge rates. In Scotland the age group with the highest rate of discharge (15–24 years) with paracetamol overdose is not the one with the highest mortality.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Overdose is one of the most frequent indications for patients to be admitted to medical wards in the UK. In the last decade, three changes have occurred which might influence self poisoning. Firstly, a change in available paracetamol pack size. Secondly, the introduction of new antidepressant drugs some of which, in particular the SSRI group, are perceived as being less toxic in overdose, has resulted in a more than two-fold increase in prescribing (ISD data on file, 2000). Thirdly an increasing use of drugs of abuse, specifically opiates,1 which is itself associated with an increase in self-harm and suicide.2
In Scotland, all data on hospital admissions are collected centrally, and it is one of the few countries in the world where hospital episodes are routinely linked together.3 It is therefore possible to study changes in the epidemiology of poisoning within a relatively large and stable population of 5.1 million. We report discharge statistics for the diagnosis of poisoning for Scotland for the years 1990 to 1999, highlighting those relating to paracetamol, antidepressants and opioids. We also report mortality data for all poisoning and paracetamol for the same period.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The Information and Statistics Division (ISD) of the NHS in Scotland routinely collects information on all hospital discharges through the Scottish Morbidity Record 01 (SMR01). The SMR01 includes the patient's demographic details, and information about the episode of care, including the diagnoses and any operations performed. This dataset has been used to examine discharge rates, by age-specific population groups, and for specific overdose diagnostic categories, within the database for the ten-year period 1990 to 1999. The diagnostic codes used in this analysis were ICD9 until the end of March 1996, when ICD10 codes were introduced. The relevant codes are: All poisonings ICD9 960–980, ICD10 T36–T51; Paracetamol ICD9 965.4, ICD10 T39.1; Opioid poisoning ICD9 965.0, ICD10 T40.0–T40.4; Opioid misuse ICD9 304.0, 305.5, ICD10F11; Antidepressants ICD9 969.0, ICD10 T43.0–T43.2. Age groups examined were determined by age at presentation: below 15 years of age and then in 10-year cohorts (e.g. 15–24yr) until 84 years. Population was estimated from national census data by the General Register Office for Scotland (GROS).
We also examined the mortality figures for Scotland available to us from GROS for diagnostic codes which include poisoning. In addition, we specifically examined deaths in which paracetamol was judged contributory.
We have shown our results as totals, and rates per 100 000 population. We include 95%CIs on data where this is informative, though with such a large dataset, these intervals are close to the mean rates.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Total overdoses
In the ten-year period there were 182 322 discharges recorded, involving a total of 119 044 patients. Total discharges for females exceeded those for males (100 112 vs. 82 210 respectively). Similarly, more individual female patients were discharged than males (66 293 vs. 52 751). The overall discharge rate per 100 000 population per annum was 379.3 in women and 331.5 in men. Discharges were most frequent in those aged 15–24 years, the annual average rate per 100 000 for the 10-year period in this age group being 805.5 in females and 601.8 in males (Table 1
).
|
Between 1990 and 1997, annual discharge rates per 100 000 population for overdose increased by 47.5% in women and 53% in men, from 300.7 to 443.6, and 248.9 to 381.8, respectively (Figure 1
). After 1997, overall rates fell in all age groups except those under 15 years; the rates of discharge in this group showed least variation over the decade. The under-15s were the only age group in which rates of discharge in 1999 were lower than in 1990.
|
Paracetamol overdose
In the 10 years of the study, 54 204 discharges were recorded (F 31 546, M 22 658) in which paracetamol was recorded as ingested in 38 767 patients (F 23 024, M 15 743). Discharge rates per 100 000 population were 119.5 and 91.4 in women and men respectively, with individual patient involvement rates of 87.2 (F) and 63.5 (M) per 100 000 (Table 1
). Discharge rates for women exceeded those for men in younger age groups, and were greatest in those aged 15–24 years (F 343.2, M 188.6). The difference between women and men was also greatest in this age band. Annual discharge rates for paracetamol overdose increased between 1990 and 1997 by 103% in women, from 79.1 (95%CI 75.7–82.5) to 160.2 (155.4–165) per 100 000, and 110% in men from 56.5 (53.6–59.5) to 118.9 (114.6–123.2) per 100 000 (Table 2). By 1999, rates per 100 000 had fallen significantly to 124.4 (120.2–128.7) in women and 97.2 (93.3–101) in men (Figure 2). The fall was greatest in females under 15 years (-38%), and in both men and women aged 15–24 years (F–27%, M–30%) (Table 2).
|
|
Antidepressant overdose
There were a total of 25 686 discharges related to antidepressant overdose, of which the majority (15 611) were in women. The overall discharge rate per 100 000 population was 59.1 in women and 40.6 in men. For both male and female patients, the highest rate of discharges was for those aged 25–34 years, 113 per 100 000 for women and 76.6 per 100 000 population for men. Discharge rates for women exceeded those in men for all age categories apart from the over-85s (Table 1).
The annual discharge rate in men and women increased more than two-fold over the period of this study, from 36.2 to 82.6 per 100 000 (128%) in women, and from 22.5 to 60.4 per 100 000 (168%) in men. Unlike paracetamol, it showed no decline after 1997 (Table 3). The increase in discharge was greatest in those aged 25–34 years, 154% in women and 198 % in men (Table 3).
|
Opioid overdose and misuse
For opioid intoxication, we examined two diagnostic categories, overdose and misuse. Opioid overdose represented the major component (75%) of this patient group, but its contribution fell through the decade (81% in 1990; 57% in 1999). There were 18 300 discharges in the combined category in the decade, 11 689 being overdoses. In contrast to other types of poisoning, these diagnoses were more common in men (11 220) than in women (7080). The rate of discharge per 100 000 in the combined diagnostic category for the decade was 45.2 in men and 26.8 in women (27.3 and 18.6, respectively, for poisoning). Rates of discharge were highest in 25–34-year-old men (118.4 per 100 000) and 15–24-year-old women (63.2 per 100 000) (Table 1), for the combined category, but for 15–24-year-olds when opioid poisoning was considered alone (Table 4).
|
During the decade, the female discharge rate rose by more than five-fold from 10.4 (9.2–11.6) to 54.1 (51.3–57) per 100 000, and there was an even larger increase in males, from 15.2 (13.7–16.8) to 91.3 (87.5–95) per 100 000. For poisoning the changes were slightly less (females 8.3 to 34.2; males 11.1 to 48).
Mortality
Despite the increases in hospital discharge rates for poisoning, overall mortality decreased after 1993 (Table 5). In contrast to discharge data, mortality was higher in males than in females in all years. Of 3945 deaths recorded, 2620 were in men. Mortality in different age groups did not mirror discharge data. Highest numbers (and rates per 100 000) occurred in those aged 25–44 years (e.g. 1999 mortality per 100 000; mean for all ages, 6.7; aged 25–34 years, 11.4; aged 35–44 years,10.7).
|
For paracetamol-associated deaths mortality was also highest in 25–44-year-olds, age groups that are not those with highest rates of discharge. Overall deaths were similar in males and females. There was no significant trend in paracetamol mortality overall. The death rate per 100 000 in 1990 was 1.44 (0.98–1.9) in women and 1.38 (0.92–1.84) in men, in 1997 it was 2.39 (1.80–2.98) and 1.53 (1.04–2.02), and in 1999, 1.37 (0.92–1.81) and 1.45 (0.98–1.92), respectively.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
We have examined trends in poisons admissions overall, and for three diagnostic groups of interest for the last decade. One important change has occurred in that time: in 1998, legislation was introduced limiting the pack size of paracetamol available over-the-counter within the UK. Since then a number of small studies have produced conflicting results on the impact of this change,4–6 although a larger study has suggested a reduced severity of paracetamol poisoning in England.7
Previous reports using this data showed a decrease in self-poisoning in the early 1980s, with an upward trend at the end of this decade and into the early 1990s.8 At that time, overdose rates in Scotland were also increasing more in men than in women. Paracetamol overdose was increasing most rapidly and was a major concern in all parts of the UK.9,10
Our data show that the rate of discharge with a diagnosis of poisoning had steadily increased until 1997, but that the change in paracetamol pack size availability in 1998 was associated with a downturn in admissions associated with paracetamol in that year. This change does not seem to have resulted in a major change in paracetamol mortality. Other Scottish data on acute hepatic necrosis are in keeping with our cautious interpretation.11 Changes in the availability of paracetamol seem to have had most impact on discharges in younger adults, who tend to take impulsive overdoses. This group of patients is probably not the one at greatest risk of mortality, since our data indicate the majority of paracetamol deaths occur in an older population, a finding in keeping with previous work.12 Limitation of pack size is less likely to impact on these older patient groups, whose self harm is less often impulsive. Nevertheless there would appear to have been a reduction in paracetamol use as a drug involved in overdose in Scotland. This supports the suggestion that pack size change would reduce poisoning admissions,9 and endorses licensing action in this area. There remains a doubt as to the magnitude of effect on mortality in Scotland.
The rate of increase of antidepressant poisoning appeared to slow before the change in paracetamol pack size, but still continues to rise (Figure 2). Admissions related to opioid poisoning and misuse have shown the most dramatic change, and continue to rise in what appears to be a linear trend.
A disturbing aspect of this study is the overall increase in admissions with poisoning in Scotland that occurred in the 1990s. This group of patients presents challenges to medical and nursing staff. Recommendations on the psychosocial assessment of this patient group13 has important implications for their management in general hospitals, and in particular their accident and emergency departments and acute medical units. One encouraging feature is the reduction in admissions for poisoning in those under 15 years. Since we are not able to measure accident and emergency attendance, as opposed to admissions that receive a hospital discharge coding, we cannot ascertain whether this change is the effect of a change in admission policy of children, or an effect of policies such as the wider use of child-resistant containers.
In conclusion, we have shown that the increase in discharges secondary to self-poisoning with paracetamol has reversed following the change in over the counter pack size. This single change seems insufficient on its own to account for the overall change in discharges with a diagnosis of poisoning. Discharges secondary to antidepressant poisoning increased at a slower rate in the second half of the decade. Discharges for opioid poisoning and misuse appear to be increasing at a worrying rate.
|
Acknowledgments |
|---|
We wish to acknowledge the help of Graham Jackson, Head of Vital Events & NHS Branch, GROS, for provision of mortality data, and James Boyd of ISD for statistical advice.
|
Notes |
|---|
Address correspondence to Dr D.N. Bateman, Scottish Poisons Information Bureau, 1 Lauriston Place, Edinburgh EH3 9YW. e-mail: spib{at}luht.scot.nhs.uk
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Jackson GWL, Cole SK. Drug-related deaths in Scotland 1998. Drug Misuse Statistics Scotland, 1999, Edinburgh 2000: Scottish Information and Statistics Division.
2. Oyefeso A, Ghodse H, Clancy C, Corkery JM. Suicide among drug addicts in the UK. Br J Psychiatr 1999; 175:277–82.
3. Kendrick S, Clark J. The Scottish Record Linkage System. Health Bull 1993; 54:72–9.
4. Turvill JL, Burroughs AK, Moore KP. Change in occurrence of paracetamol overdose in UK after introduction of blister packs. Lancet 2000; 355:2048–9.[CrossRef][Web of Science][Medline]
5. Prince MI, Thomas SHL, James OFW, Hudson M. Reduction in incidence of severe paracetamol poisoning. Lancet 2000; 355:2047–9.[CrossRef][Web of Science][Medline]
6. Robinson D, Smith AMJ, Johnston JD. Severity of overdose after restriction of paracetamol availability: retrospective study. Br Med J 2000; 321:926–7.
7. Hawton K, Townsend E, Deeks J, Appleby L, Gunnell D, Bennewith O, Cooper J. Effects of legislation restricting pack sizes of paracetamol and salicylate on self poisoning in the United Kingdom: before and after study. Br Med J 2001; 322:1–7.
8. McCloone P, Crombie IK. Hospitalisation for deliberate self-poisoning in Scotland from 1981 to 1993: trends in rates and types of drugs used. Br J Psychiatr 1996; 169:81–5.
9. Hawton K, Ware C, Mistry H, Hewitt J, Kingsbury S, Roberts D, Weitzel H. Paracetamol self-poisoning. Characteristics, prevention and harm reduction. Br J Psychiatr 1996; 168:43–8.
10. Thomas SHL, Horner JE, Chew K, Connolly J, Dorani B, Bevan L, Bhattacharyya S, Bramble MG, Han KH, Rodgers A, Sen B, Tesfayohannes B, Wynne H, Bateman DN. Paracetamol poisoning in the North East of England: presentation, early management and outcome. Hum Exp Toxicol 1997; 16:495–500.
11. Newsome PN, Bathgate AJ, Henderson N, et al. Referral patterns and social deprivation in paracetamol-induced liver injury in Scotland; has the legislation worked? Lancet 2001; 358:1612–13.[CrossRef][Web of Science][Medline]
12. Schmidt LE. Paracetamol poisoning among adolescents in a department of Hepatology. Int J Adolesc Med Health 2001; 13:327–34.
13. Isacsson G, Rich CL. Management of patients who deliberately harm themselves. Br Med J 2001; 322:213–15.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P Unverir, R Atilla, O Karcioglu, H Topacoglu, Y Demiral, and Y Tuncok A retrospective analysis of antidepressant poisonings in the emergency department: 11-year experience Human and Experimental Toxicology, October 1, 2006; 25(10): 605 - 612. [Abstract] [PDF]
|
|
|
|
 |
|
 S F J Clarke, P I Dargan, and A L Jones Naloxone in opioid poisoning: walking the tightrope Emerg. Med. J., September 1, 2005; 22(9): 612 - 616. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L E Schmidt Age and paracetamol self-poisoning Gut, May 1, 2005; 54(5): 686 - 690. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||