Ecstasy's lethality has increased

When the Advisory Council on the Misuse of Drugs (ACMD) published its 2009 report on Ecstasy, lay and professional press coverage did not refer to the basic data on drugs-related deaths (DRDs) nor to estimates of user-prevalence qualified by their standard errors.
 
For England and Wales, the publicly-conveyed message in 2009 was of 33 DRDs annually with “any mention of Ecstasy”, which included 17 with Ecstasy as “sole drug”. But this was misleading for the 21st century use of the drug.  
 
As detailed below, there have been major changes during 1996-2007, the period covered by ACMD’s report, in both numbers of Ecstasy-related deaths and the prevalence of past-year users as reported by the British Crime Survey.
 
The Table provides the basic data for England and Wales, taken from Health Statistics Quarterly.  For Scottish data, see Drug Misuse Statistics, Scotland 2008 and the Scottish Crime and Victimisation Survey.
 
Major changes have occurred that it was unwise to average across without precautionary comment. From 2001, the Office for National Statistics coded deaths according to the 10th International Classification of Diseases (ICD10). The same year, 2001, marks a sharp increase in the recorded number of DRDs with ‘any mention of Ecstasy’. Are the two related, or is there a more sinister explanation?
 
The ACMD report did not pay sufficient heed to standard errors because, as the Table shows, there were for important calendar periods changes in the past-year use of both Ecstasy and cocaine by 16-59 year old respondents in the British Crime Survey (BCS). The BCS enjoys high response rates and, since its 2004/05 wave, has had nearly 30,000 respondents per annum. Past-year use of Ecstasy in England and Wales was very significantly lower, by a fifth, at 1.67 per cent (se 0.05 per cent) during the most recent four BCS waves than in the previous three.
 
Piecing the jig-saw together. During the five years of 1996-2000, when the prevalence of past-year use of Ecstasy was 1.7 per cent, there were 97 DRDs with “any mention Ecstasy”, or 19.4 per-annum. For comparative purposes, we can express that as deaths per 1 per cent past-year users per annum. In that period, annual deaths were were 11 per 1 per cent past-year use of Ecstasy (where 11 = 19.4/1.7; with approximate 95 per cent CI: 9 to 14).
 
During 2001-03, when the prevalence of past-year use of Ecstasy was 2.1 per cent, there were 161 deaths with “any mention of Ecstasy”, or 53.7 per-annum, so that per-annum Ecstasy-related deaths were 26 per 1 per cent past-year use of Ecstasy (CI: 21 to 30).
 
During the four years of 2004-07, there were 29 Ecstasy-related DRDs per-annum per 1 per cent past-year use of Ecstasy (approximate CI: 25 to 34).
 
It is clear that Ecstasy’s implied lethality was higher from 2001 than it was in 1996-2000.
 
Adopting the same approach for cocaine-related deaths in England and Walessuggests that the lethality of cocaine was also significantly higher in 2004-07 per 1 per cent past-year users than in 2001-03. In 2004-07, there were 78 cocaine-related deaths per 1 per cent of past-year use (CI: 72 to 84, based on 716 cocaine-related DRDs in 2004-07). In 2001-03, cocaine-related DRDs were 54 per 1 per cent of past-year use (CI: 48 to 60, based on 354 cocaine-related DRDs in 2001-03). What accounts for this difference?
 
In Part 4 of this series on drug-related deaths, we saw that the increased C-related death risk applied especially to older users whose per-annum risk had increased, apparently by 40 per cent, between 2001-03 and 2004-07. Notice that cocaine’s lethality measured in this way was at least double Ecstasy’s in 2004-07.
 
Ecstasy’s death rate can also be calculated more conventionally, as a rate per 100,000 users. In 2007/08, the BCS estimated 2.3 per cent of respondents had used Ecstasy in the past year, so that there were about 470,000 users in England and Wales (with lower and upper limits of 419,000 to 526,000). Thus, in 2004-07 (when there were 196 Ecstasy-related deaths over four years), we can infer 10 deaths per 100,000 users of Ecstasy per year (approximate CI: 8 to 12 per 100,000 past-year users).
 
Why has Ecstasy’s lethality apparently increased since 2001? Explanations range from adulteration; more tablets per episode of use; change in the conduct, or reporting, of toxicology on DRDs; to change in the coding of DRDs.
 
Beyond question is that a notable increase had occurred, for which an explanation is needed. Whatever the explanation, it would be unlikely to affect ACMD’s published advice - which takes other harms than lethality into account – that, on an evidential basis, Ecstasy does not warrant its A-classification.
 
In 2004-07, cocaine-related DRD rate per 100,000 past-year cocaine-users was more than double Ecstasy’s but, as we have shown in Parts 3 & 4, men and older cocaine-users are at disproportionately high C-related DRD risk.
 
The devil is in the detail of fatalities associated with specific drugs, whether Ecstasy or cocaine. In public health terms, the siren message remains that heroin is fatally in a class of its own, as we showed in Party 5 of this series.
 
Table Drugs-related deaths (DRDs) and drug-specific mentions, as reported by ONS for England & Wales, 1996-2007; and pooled estimates for past-year of specific drugs, as reported by 16-59 year old respondents in England & Wales to the British Crime Survey.
 
 
              
 
Conflicts of interest: SMB serves on, and was inaugural chair of, the Surveys, Design and Statistics Subcommittee (SDSSC) of Home Office’s Scientific Advisory Committee. In 2008, SDSSC reported on 21st Century Drugs and Statistical Science. SMB has research interests in the epidemiology and prevention of drugs-related deaths; holds GSK shares; and has MRC-funding as part of the NIQUAD cluster on quantifying drugs harms.