HOW OFTEN DOES THE ADULTERATION/ DILUTION(1) OF HEROIN ACTUALLY OCCUR?
AN ANALYSIS OF 228 'STREET' HEROIN SAMPLES ACROSS THE UK (1995-1996) AND DISCUSSION OFMONITORING POLICY
Ross Coomber, University of Greenwich, London
Recent research has suggested both that illicit heroin in the UK may not be adulterated1diluted ('cut') with dangerous substances and also that it is actually adulterated1diluted far less than is often believed (Coomber, 1997a, 1997b, 1997c). Forensic evidence does not report on the proportion of samples where no cutting agents are found but merely presents evidence of when they are and what they are, making judgements about how often the cutting of drugs takes place reliant upon other less definite data. Moreover, from the presentation of existing forensic data on cutting agents the impression can be formed that the adulteration/dilution of heroin is the norm. A specialty arranged analysis of 228 'street' seizures to test this assumption, however, suggests that nearly half of the heroin which was sold for use in the UK in 1995-96 may not have been adulterated at all. Comparison of this new data with previously published data on 'cutting agents' suggests that much less adulteration has probably also been common in previous years. When compared with other new data from Customs seizures, the argument put forward by Coomber (I 997a) that most adulteration/dilution is professionally managed (as opposed to the work of a strung-out 'junkie') prior to importation and that only a relatively small proportion of heroin sold is adulterated/diluted by those believed to do so- the 'street' dealer -is strengthened. Current monitoring procedures are unsystematic and insufficient. At the very least, effective monitoring of the makeup of illicit drugs would improve understanding of trafficking and production trends as well as provide evidence of what happens to drugs after they reach the street. On this basis, policy recommendations are made for improved strategic recording of future forensic data for, in the first instance, the UK, and then for the possibility of a Europe-wide approach to such monitoring.
LESS ADULTERATION/DILUTION: BACKGROUND INFORMATION
After reviewing the forensic evidence in Coomber (1997a), it was speculated that less adulteration/ dilution of street drugs actually takes place than is commonly perceived. Less adulteration/dilution was suggested to occur in two distinct senses: first, that the actual amount or percentage of adulterant/diluent generally found in heroin was lower than expected; second, that this was a likely consequence of less adulteration/dilution actually taking place clown through the chain of distribution than previously thought. In relation to heroin it was noted that when 'street' seizures were compared with Customs seizures that there is often less difference in purity levels between the two than might be expected.-This relative lack of disparity had also been noted by Lewis et al. (1995) (also UK) and Kaa (1994) in Denmark over a 12 -year period, and recently by De la Fuente et al. (1996) in Spain. Respectively, it was noted that average differences were found of around 8-14% (Coomber, 1996a); 15-25 % (Lewis et al., 1995) and 9% (Kaa, 1994). Coomber 0 99 7b) noted that information gleaned from 3 1 drug sellers at varying points in the chain of distribution supported t6;vroposition that adulteration/dilution is not a predictable outcome of various drugs' working their way through the chain of distribution. In relation to those who supplied/dealt in heroin, 65% (11) said that they never adulterated/diluted it at all. Only one heroin dealer (dealing 4 to 5 ounces a month) said he always diluted the heroin (glucose, by around 10-20%). Four others adulterated/diluted only 'sometimes'. No direct relationship appeared to exist with the level of involvement, i.e. how much they sold, how long they had been selling for or what proportion of their incomes depended on drug sales. Data from 80 drug dealers from 14 different countries, responding to research mediated through a questionnaire on the Internet,' and partially replicating the research of Coomber (1997b), indicated strongly that those findings can be applied, albeit with proper caution, internationally (Coomber, 1997c). Evidence was also submitted to suggest that less adulteration occurs (i.e. the number of times each sample of drugs is adulterated/diluted) than is commonly thought to take place with all 'street' drugs.
A further issue which was raised by Coomber (1997a) related to the fact that 'purity' is often not what it seems when considering issues around adulteration/dilution. Analysis reveals that a heroin sample, apparently only 65% pure, may in fact have no adulterants/diluents present (Gough, 1991; HM Customs & Excise, 1995). Depending on country of origin, and thus on method of manufacture, the production of the heroin itself produces a more or less 'pure' product. In some cases, various other opiate alkaloids, such as noscapine and papaverine and acetylcodeine (which is a by-product of heroin manufacture), may account for the bulk of the other 35%. In the reporting of drug purities this important fact almost invariably remains unstated, inadvertently giving the impression that the other 35% in fact comprises adulterants/diluents Put there by those who sell them.
PRESENTATION OF FORENSIC EVIDENCE: CUTTING AGENTS
As stated above, normal reporting of heroin purity profiles may be inadvertently misleading by giving the impression that 65% purity means the sample contains 35 impurities'. Again this will not necessarily be assumed by the forensic science community but others who consume these figures may not be as well informed. In fact, a senior forensic scientist related to me that he has constantly to remind even his own staff of this point (King L, Head of Druijs Intelligence Laboratory, Forensic Science Service, UK, personal communication, 1996). The public reporting of drug purities certainly gives no indication that 65% purity may in fact also refer to drugs with no adulterants or diluents found.
A second problem lies in the way 'cutting agents' found in illicit drugs are commonly presented. For example, in 1993 the substances listed in Table 1 were presented to be the cutting agents reported in heroin (Drug Abuse Trends, 1993, p. 19) which also contributed to public information on drug purity.
In both the UK and USA reporting of adulterant/diluent content of illicit drugs, as in the example as above, data are given only where substances are actually found. No data are given on the proportion of samples where no adulterant or diluent was found.' It is not uncommon for an adulterated/diluted sample to contain more than one such substance. Given this, and the way that the data are presented, it is not entirely implausible for Table 1 to be interpreted as showing 4 1 % of the samples to have one or more adulterant/diluents and the remaining 59% to be 'pure' heroin. As it is, the percentages, which add up to 100%, certainly do not tell us that 100% of the samples had an adulterant/diluent added - although it may be argued that, perhaps unintentionally, they suggest that a very high percentage did.
ANALYSIS OF THE 228 HEROIN SAMPLES FROM ACROSS THE UK
To date it has not been normal practice for the disparate forensic services in the UK to collate information on drug purity, drug adulterants/diluents and impurities in any nationally coordinated or even consistent manner as regards type or method of analysis for the purpose of monitoring. As such, many of the data which inform bodies such as the National Criminal Intelligence Service (NCIS) and the Home Office on purity and cutting agents result from a mixture of procedures and approaches. Little comprehensive profiling of drugs is undertaken and none is done as a matter of considered policy in the interest of monitoring. Over any given year, for different reasons, a number of heroin samples will be analysed for purity and profiled for adulterants/diluents. This information will then be collated (as a rough guide to trends) for publication in Home Office Bulletins and other such public (and restricted) publications.
In collaboration with the Forensic Science Service at Aldermaston in England, it was arranged for information on all heroin samples that had been testd for adulterants using gas chromatography and mass ectrometry in UK laboratories (1995-96) to be colred and brought together at Aldermaston. A total of 28 samples made up the newly collated information. 11 samples were police seizures and were designated 'street' seizures by the Forensic Science Service.
Because of the nature of drug seizures, and of the ay that such profiling is carried out, the samples analysed represent a fairly random selection of heroin in the UK during this period.
Confirmation of less adulteration
The most important finding revealed by the analysis is that in 44% (100) of the samples no adulterants were found. Whilst we might expect a percentage of dlicit drugs to be adulterated prior to importation this finding appears to support the findings of Coomber 0997b) where only a minority of those who sold heroin stated that they adulterated/diluted the heroin they sold. Thus, of 228 random street seizures of heroin nearly half contained neither paracetamol nor caffeine (the predominating adulterants of heroin) or any other such agent. As we can see from Table 2, these two substances still predominate, as in Table 1, but dissimilarly here we can see that a maximum of 56% of the 228 samples contained cutting agents.
Perceived differences between the 1993 representation of the proportion of cutting agents and the present sample
If we now restructure Table 2 and, like the table from Drug Abuse Trends, exclude those samples where no cutting agent was found, we see that the percentage figures change in a potentially significant way. Percentages for the 1995-96 samples (taking the 186' maximum which did contain a cutting agent) now refer to the proportionality of appearances of cutting agents in samples where cutting agenCs appeared. Once this recalculation has been carried out we can see that the figures begin to look remarkably like those presented in Drug Abuse Trends (1993). Respective percentages thus become those given in Table 3.
The predominating cutting agent, paracetamol, at 40% is almost identical to that in the table from Drug Abuse Trends (1993). The percentage with caffeine is slightly higher but still well within reasonable comparable limits and the rest tail off in almost the same proportions. Slight differences no doubt reflect minor changes in either trends of production or in origin of samples. We might reasonably speculate, then, that in 1993 a similar percentage of street drugs did not contain any cutting agents. As we shall see below this proportion varies owing to different contingencies but, as we have seen in relation to 199596, much heroin that is sold is not adulterated at all and there is little reason to believe this figure would differ dramatically for other years. For example, and as we shall see below, only 36% of those samples seized at importation in 1993 were adulterated. This figure may under-represent the actual percentage of drugs imported in 1993 that found their way on to the streets which had adulterants, owing to the vagaries of drug seizures. We can see that the gap between 36% and that speculated as possible (in and around 50%) could easily be closed, with the need for little extra adulteration to occur once in the country. As will be discussed below, it is not suspected that too much adulteration does occur after entry.
At the very least, it would be useful for monitoring purposes if future reporting on cutting agents were also to include reference to those samples where none was found. Although quantitative data were not available it should also be noted that in some cases cutting agents were present in trace (minuiscule) amounts only.
Findings from samples seized by HM Customs and Excise 1990-1993: new material
Between 1990 and 1993 HM Customs and Excise commissioned a relatively extensive analysis of most of the imported illicit heroin seized by Customs. Being a privately commissioned report the Data have not been available for public scrutiny. The information in the report does, however, once compared with the findings from the 228 heroin samples which are of 'street' seizures, provide valuable insight into certain aspects of how much adulteration/dilution takes place, when and where. Furthermore, it also illustrates that levels of adulteration/dilution and thus purity probably reflect less any trend in drug dealing than they do in predominating sources of production/supply.
The analysis and subsequent profiling of all heroin seizures by Customs and Excise between 1990 and 1993 (inclusive) provides data on a range of important trends in drug trafficking. By carrying out a profile analysis of a sample it can be determined (with reasonable accuracy) the source region from which the heroin originated, e.g. Turkey, Pakistan, Nigeria, as each source country or region tends to produce a particular configuration, or profile, of heroin.
One sample configuration of Turkish heroin for example ( 1992) looked as set out in Table 4.
The proportions of the various alkaloids may vary but the profile is relatively distinctive. The absence of methaqualone, caffeine and paracetamol, phenobarbitone and other miscellaneous substances is notable. Other alkaloids, particularly narcotine, are present insignificant amounts.
If we look at an example of heroin of Kenyan origin (1993) (Table 5) we can see an immediate contrast, particularly regarding the alkaloids.
In recent years in the UK, heroin from Turkey, has been prevalent, followed by that from Pakistan and India. When it comes to adulterants we find that Turkish heroin rarely contains any, for example in 1990 there were no adulterants found in 103 samples, in 1991 none was found out of 12 samples, in 1992 only four of 92 samples contained an adulterant (phenobarbitone in each case, with a range of 1.7% to 2.1%) and in 1993 only four from 50 samples had low levels of adulterant (three included paracetamol, one griseofulvin). Thus, of the 257 samples tested only 3% contained (small) quantities of adulterant.
Other source areas such as Africa, which fluctuates in terms of how important it is as a supplier to the UK, supply drugs with greater levels of adulteration. Of the 48 analysed over the four-year period, 40 (83%) contained adulterants. This was also reflected in a comparatively lower average heroin purity of around 34% compared with the average of 66%.
Table 6 shows the proportions of samples from all sources analysed for Customs and Excise which were und to have adulterants present.
In 1990 Turkish heroin (almost no adulteration) accounted for 37% of the seizures and consequently the figure of 17% reflects this. In 1991, where the percentage of samples found to be adulterated was 41 %, Turkish heroin accounted for only 6% of the overall samples analysed that year. In 1992, when the number found to contain adulterants was back down to 20%, Turkish heroin was again the predominant supply constituting 5 5 % of samples. In 1993 when the proportion of samples found with adulterants rose to 36%, Turkish heroin did constitute a healthy 49% of he samples but Pakistani and Indian heroin constiuted 30% as opposed to 24% in 1992; Pakistani heron samples in particular exhibited a rise from 42% in 1992 to 71 % where adulterants were recorded.
It seems clear that if Turkish heroin were to monopolise the market then adulteration/dilution of heroin prior to importation would be negligible. As it is, the UK market is serviced by numerous source countries. In 1993 for example, around half of the seizures analysed by Customs and Excise came from sources outside Turkey, of which 70% were found to have adulterants present. Those which were designated as illicit heroin from Europe (meaning predominantly Holland, France and Belgium) were found to have adulterants in 100% (10 out of 10) of samples, sometimes (and unusually compared with other sources) in quite significant amounts.
In the 228 samples from 1995-96 we know that 6% had one or more cutting agent(s) present. Figures for 1995 Customs seizures are not yet available but if we hypothesise a similar proportion of adulteration (and non adulteration) to that of 1993, as was previously speculated might not be unreasonable (see Table 3), we would need to explain a difference between 36% found at the point of import and the speculated band (around the figure of 56%) once on the street. Clearly, seizures at import (in terms of overall profiles/proportions) are not going to reflect exactly the drugs which reach the street. In fact a large Customs seizure from one source country may allow street heroin source prevalence to be shifted. Given this, we need only recognise a broadish percentage band within which we would need to approximate extra cutting to occur after import. The 20% difference is therefore potentially quite a bit lower once this has been fully considered, and "hen we consider that the 56% may in fact be less as well the gap may not be so great in reality. In fact, as we shall see, the difference may in fact be negligible. We need, then, to consider how much adulteration/dilution results from the practice of drug dealers once the heroin is in the country.
In Coomber (1997b) around 35 % of the 17 heroin sellers indicated that they sometimes (4); rarely (1) - once in two years; and always (1) adulterated/diluted the heroin they sold. If this is any way representative of UK heroin sellers then most (almost all) heroin passes through their hands without having any extra cut added. Some of those who do add adulterate/dilute the heroin they sell will of course be adding a cut to a sample which has already been adulterated/diluted rather than simply contributing to the overall percentage of those samples which have a cut in them. None of those who adulterated/diluted the heroin they sold as noted by Coomber (1997b) did so in excess of 25% of the sample. Most did so with a range of 10-20%. Although only 35% admitted to cutting the heroin they sold, the vast majority (94%) either never cut it, did so rarely or did so sometimes. If this is in any way representative of how much cutting occurs to heroin once it is in the UK then we can see that there should be little difference (in a perfect comparative world) between the percentage found to 'have been adulterated by Customs and the police and/or the purities of these respective samples. Without exception, those who did adulterate/dilute claimed to use sugars (glucose or lactose) to dilute the heroin they sold. We might thus expect that most dilution which occurs to heroin once it enters the UK is carried out with sugars.
Sugars as a common cutting agent in heroin
The table which appeared in Drug Abuse Trends (1993) does not mention any sugars as being found as cutting agents in heroin. In cocaine, the same publication lists glucose as appearing in 21 % of samples analysed, mannitol in 16%, lactose in 7%. For amphetamine, glucose was found in 10% and lactose in 3%. Although the 228 heroin samples being considered here were not analysed for sugars it was the general opinion of the various forensic laboratories involved that when heroin in the UK is tested for sugars these are rarely found. When they are, the most common substances are the sugars mannitol and glucose. If, as stated earlier, those reported by Coomber (1997b) as selling heroin and who also admitted cutting the drugs they sold all said they used a sugar (glucose or lactose) to cut the heroin with, then we would expect to find sugars at about the rate of cutting that occurs. As stated, sugars are not commonly found in heroin, possibly supporting the evidence that the cutting of heroin by drug sellers, whilst occurring, is neither prevalent nor probable.
Arguably, sugars would have been expected to occur more often than they actually do, not only in heroin, but also in cocaine and amphetamine. That it is paracetamol (41 %) and caffeine (33%) that are predominant in heroin, caffeine (24%) in cocaine, and caffeine again (75%) which greatly predominates in amphetamine (the next closest being glucose at 10%) suggests that the bulk of the cuts are added high up the chain of distribution and that less cutting goes on once the drugs work their way through the chain. This may be presupposed for two reasons. First, it is widely suspected by forensic agencies that the paracetamol found in heroin is not commercial paracetamol but a brownish illicitly manufactured one, and caffeine in powder form is not easily available in bulk and costs more than substances like glucose. Second, and consistent with the findings in Coomber (1997b), those who did cut the, heroin they sold without exception said they used sugars such as glucose and lactose. In fact the existence of mannitol, as with caffeine, is likely to indicate higher-level cutting of the drug as opposed to street-level cutting. Both mannitol and caffeine, when compared with glucose and lactose, are relatively expensive products but provide better 'quality' cuts than the latter. The general absence of sugars in heroin therefore probably correlates with the amount of cutting which takes place after importation, i.e. comparatively little.
Some forensic scientists believe that a timelagged histogram of the purity of Customs seizures and police 'street') seizures would in fact match, indicating that little or no adulteration/ dilution takes place once the drugs are in the UK.
Considerations such as these would be easily resolved if more profiling was undertaken of both Customs and street seizures.
'Purity' as an indicator
If we compare the purity of Customs seizures with that of street seizures during this period we will see that there is a spread of between 2% (1989) and 20% difference between average Customs seizure purity figures (always higher) and street seizure purity figures (always lower) (see Table 7).
The spread may differ widely due t4 a number of factors, not least the fact that single (or* a number of single) large seizures of either high- or low-purity heroin from a particular source in any one year may distort the overall picture. It may indeed be the case that for the years 1988 and 1989 the 2% and 39/0 difference is a result of either a good match between Customs seizures and police (street) seizures and thus reflects that little adulteration/dilution took place at all during this period, or a coincidence which was corrected in subsequent years. Improved monitoring strategies and recording of information would enable such pictures to be more transparent.
Knowledge about how often adulteration/dilution of illicit heroin takes place and when is currently hampered by the absence of a strategic and coordinated approach to drugs analysis. The foregoing discussion, however, suggests that a more strategic approach could offer new and important information on drug purity and the constituents of illicit drugs. Current presentation may be inadvertently misleading. The analysis of the 228 UK heroin samples for 1995-96 revealed, importantly, that nearly 50% of the drugs seized by police did not contain any adulterants. This is probably a figure which is at variance with most common perceptions of heroin sold in the UK. Moreover, it appears that most adulteration, where it has occurred, takes place prior to importation and is less the result of a haphazard throwing in of anything which comes to hand but rather a reasoned and not particularly unsafe process. When heroin is cut after importation it appears that this is predominantly with sugars (Coomber, 1997b). Further evidence presented by Coomber (199 7b) also suggests that to cut the heroin they sell is not a normal practice of heroin sellers. This is further supported by the fact that few heroin samples are found to contain sugars under forensic analysis. It is hoped that the newly formed UK-wide Forensic Science Service, which has amalgamated all of the previously disparate services, may be able to provide the coordination and overall strategy for the monitoring of illegal drugs by forensic analysis in the UK. But at the time of writing no such plans are evident.
It can be recognised that there is a relative dearth of consistent and reliable information on trends regarding drug purity, drug composition, variations and prevalence of drug origins, trafficking and adulteration/dilution practices. Recommendations are made below which would help ameliorate this situation in the UK. It may of course be preferable to see systematic and comparable monitoring on a Europewide scale facilitated by a body such as the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) or some similar body. This would provide a broader dataset, able to track drug movements within Europe as well as revealing similarities and differences of distribution practices within and across boundaries. If centrally funded via Europe the problem of such monitoring (as is now the case) being subject to 'local' political and economic priorities would be largely resolved.
Recommendations for future recording of adulterant/diluent data
The following steps are suggested as a refinement to how existing data regarding adulteration/dilution and drug composition should be collected and reported:
(1) Random samples of heroin and other street drugs t from police seizures should be analysed regularly for purity and composition (proportional), t including adulterants, diluents and substances resultant from the production process. Samples t should include both street seizures and Customs seizures for comparison. This would allow the monitoring of a range of trends and potential hazards.
(2)Similar to the practice adopted by the Drug Enforcement Administration's Domestic Monitor Program in the USA, a retail- (street-)level purchase programme should be set up to provide samples to compare with street seizures analysed.
(3)The percentage of samples where adulterants/ diluents are not found should be recorded.
(4)The geographical location of where each sample was seized or bought should be recorded. This will, as in the DEAs Domestic Monitor Program, provide improved monitoring of drug use in key cities and surrounding locales. In the USA for example, certain cities consistently have heroin of very high purity and others heroin of medium and low purity.
(5)The monitoring should be strategically managed and coordinated by a centralised body or authority with responsibility for collating and presenting the data. Such arrangements could be managed nationally, and would result in improved monitoring. However, a preferable move would be for central funding from the European Community to facilitate and coordinate analysis in a consistent and strategic way.
Concerns around how comprehensive analysis of heroin and other drugs might affect a prosecution to which the sample is related (normally individuals are charged with supplying or possessing a single drug; the finding of other illicit substances could mean higher sentences) could be managed by samples for he strategic review being anonymised and the normal checks (e.g. simply to confirm the presence of the drug) can be carried out separately as at present.
Ross Coomber, Principal Lecturer in Sociology, School of Social Sciences, University of Greenwich, Eltham, London SE9 2HB, UK
1. The term adulterant is used in this paper to refer to substances added to illicit drugs in the process of selling and distribution. Adulterants proper are in fact other psychoactive drugs (like caffeine, or paracetamol) which are much cheaper than the main substance, have a similar or complimentary effect when mixed with it, and therefore help hide the fact that the substance has been diluted. Substances which are not psychoactive, such as glucose and lactose, are more formally known as 'diluents'. These are added to a drug to increase the amount of drug available to be sold. It should be noted, however, that some substances which are found in street drugs will be the result of the particular manufacturing process used to make the drug. In this sense those substances might be more properly referred to as impurities'. 'Excipients' found in drugs (primarily pills/tablets) are the products used to bind the drug together. Common excipients are starch, gelatin or other gums (ISDD, 1994).
2 Whilst it is true that forensic scientists might not expect the difference to be great, the general rationale for how street drugs become adulterated is heavily bound up in the mythology of the dope-fiend which suggests that most adulteration/dilution is carried out by the 'street' dealers themselves (see Coomber 1997a for elaboration). Moreover, the reporting of these relatively narrow differences as being important as an indicator of drug distribution practices is not something which has been of concern to forensic scientists.
3. The point to be made here is that whilst drugs such as amphetamine are heavily diluted it appears that this is normally the result of a very large initial 'cut' down to a low purity. That seizures rarely find gradations of purity (e.g. 60% pure, 40% pure, 20% pure) suggests this is true. If amphetamine were diluted down through the hierarchy of distribution (as in the classical model) then it is likely that such gradations would be found.
4. See Coomber (19970 for further information on this research and the methodology which enabled it to take place.
5. In the USA information on the percentage of analysed street samples which contain no adulterants/diluents is available through the Drug Enforcement Agency's Domestic Monitor and Heroin Signature Programmes. It is not, however, contained within the information which is released for public perusal or even within the DEA itself.
Coomber R (1997a). Vim in the veins - fantasy or fact: the adulteration of illicit drugs. Addiction Research 5: 195-212.
Coomber R (1997b). The adulteration of drugs - what dealers do, what dealers think'. Addiction Research 5: 297-306.
CoomberR (I 997c). Dangerous drug adulteration -an international survey of drug dealers using the Internet and the WorldWide web (WWW). InternationaljournalofDrugPolicy 8: 18-28.
Coomber R (1997d). Using the Internet for survey research. Sociological Research Online 2(2): .
Drug Abuse Trends (19193) No. 102 (October- December). London: Home Office.
De la Fuente L, Saavedra P, Barrio G, Royuela L, Vincente J (1996). Temporal and geographic variations in the characteristics of heroin seized in Spain and their relation with the route of administration. Drug and Alcohol Dependence 40: 185-194.
Gough TA (1991). The examination of drugs in smuggling offences. In Gough TA (Ed.) The Analysis of Drugs of Abuse. London: Wiley.
HM Customs and Excise (1995). A Report on Illicit Heroin Products Seized by HM Cus to?ns and Excise from 1990to 1993.London: HM Customs.
ISSD 0 994). What's in a drug-Druglink Factsheet 10. Druglink Nov/Dec.
KaaE(1994). Impurities, adulterants and diluent of illicit heroin: changes during a 12 year period. Forensic Science International64:171-179.