Health Canada's Drug Analysis Service (DAS) is responsible for testing suspected illegal substances that are seized by Canadian police officers and custom agents. For each sample, the test results are entered into a Laboratory Information Management System (LIMS) which captures such information as the date of the seizure, the primary substance found in the sample and any other adulterants if analyzed and reported, and police detachment or customs office location. Further, under the Controlled Drugs and Substances Act, Health Canada is responsible for authorizing the destruction of all controlled substances seized in Canada. As a result, upon completion of all law enforcement cases, information on the suspected drug seized, the offence, information on the defendant, and the disposition of the charges is forwarded to Health Canada and retained in the Controlled Drugs and Substances Database (CDSD). While primarily developed for administrative purposes, these two systems – LIMS and CDSD – also have a significant strategic value, helping to identify and track emerging trends in illicit drugs in Canada.
Data Sources
Laboratory Information Management System (LIMS)
Health Canada's Drug Analysis Service (DAS) is responsible for analyzing suspected illegal substances that are seized by Canadian police officers and custom agents. Through their testing, they issue certificates of analysis that are used as evidence in Canadian Courts. Once the identity of the substance is confirmed, the results are entered into the national LIMS database. LIMS contains information on the date of the seizure, the primary substance found in the sample and any other adulterants, if analyzed and reported. Further, the police detachment or customs office responsible for the seizure is also recorded, and thus allows for identifying and comparing trends across the country. The amount seized is not recorded in LIMS.
Controlled Drugs and Substances Database (CDSD)
Whenever a police or customs officer seizes a drug in Canada, the information is collected on a paper form known as the Drug Offence and Disposition form (HC 3515). This form is used for the logging in, tracking, and disposition of the drug. Police or customs officers manually enter the information as the case progresses, and only upon completion of the case is the information forwarded to Health Canada. This form contains information on the suspected drug seized, the offence, information on the defendant, and the disposition of the charges. Upon completion of the case, the form is forwarded to Health Canada before any exhibit can be destroyed. This information contains a significant lag time between the date when the substance was seized to the time it is entered into the CDSD database, since it is sent to Health Canada only when a case is closed. However, it can serve as a complementary source of information on drug seizures in Canada since it identifies the amount of drugs seized in each case.
The LIMS and CDSD data systems differ in the number of samples contained in each dataset. This is because the LIMS data source contains information on all substances contained in a sample, whereas the CDSD database contains drug information based upon what was suspected by police or customs officers. As such, this system is not as comprehensive when identifying substances.
Data Indicators
The LIMS and CDSD contain over 800 and 430 different substances respectively. Given that the numbers of different substances within each database is quite large, the focus will be on the most prevalent substances as these are expected to have the most impact on society. The most common substances are: cannabis (marijuana, hashish, and hash oil), cocaine, heroin, methamphetamine, MDMA (ecstasy), psilocybin (PSBN), and lysergic acid diethylamide (LSD). The indicators below will be examined for each of these drug types. For the purpose of this feasibility report, examples of data results for cocaine for the province of British Columbia will be highlighted for each of the indicators below. However, the full scope of results with the most prevalent substances will be covered in an upcoming report.
Through examinations of the samples analyzed by DAS, we are able to explore the prevalence levels of all substances seized that are being prosecuted in Canadian Courts. This provides an indication of what is occurring at the street level, what types of drugs are being seized through custom agents, and how these differ from what is being seized by police officers. Both the LIMS and CDSD datasets will provide reasonable estimates on the number of illicit drugs seized across Canada. The LIMS dataset contains historical information for British Columbia on a total of 272,448 cases dating from January 1, 1988, while the CDSD dataset contains reliable data entries since January 1, 1986 on 373,202 British Columbian cases. Each sample is date-stamped to allow for accurate assessments to be drawn as to when the sample entered police custody, and further, each sample can be identified as to which police department seized the drug. This allows for trend analyses to be computed for Canada, each province, and regional districts within each province. By exploring the changing prevalence in drugs seized over time, an indication in changes in the illicit drug market can be derived.
Quantity
Each exhibit that is entered into the CDSD database contains information on the total quantity of the seizure. Depending on the substance, a common quantity unit will be computed. For example, the dataset may contain cocaine samples that are entered as either millilitres or kilograms, while ecstasy samples may be entered in terms of weight (e.g. grams) or in what form they were seized (tablet or capsule). Whenever possible, units will be recalculated so that a common unit is generated. In terms of sample form (e.g. tablet, capsule, blotter, plant), quantities will be categorized based upon the unit form. Ranges will be calculated for each unit, and graphical displays will be designed. Further, linear regression analyses will be performed to determine if the quantity of each substance per seizure has changed over time.
Purity and Adulterations
Another important source of information pertains to the purity of the samples analyzed by DAS. Analyses of contents and purity of samples can give important indications of major emerging trends, especially when used in conjunction with interviews with drug users and key informants. However, an important limitation to these analyses is that the data do not represent a random sample as not all samples analyzed are tested for purity. Sometimes, these tests are only performed when the substance is contested in Court or when the prosecutor requests a purity analysis to be conducted as part of the evidence for Court. Although these samples are not a true representation of purity of the drugs seized in Canada, they do allow for trend analyses to be conducted. In samples in which purity levels have been analyzed, descriptive statistics such as means, standard deviations, or 95% confidence intervals will be computed. Moreover, linear regression modeling will be explored to determine if any changing patterns were observed over time in the purity levels of the most frequently analyzed substances.
Another important result that can be captured is the amount of drug adulterations observed in samples. Drugs that are analyzed by DAS often contain multiple substances. Although the system does not identify the percentage of each contaminant within each sample, having indications that samples contain more than one substance provides strategic information regarding the illicit drug market. These contaminants can also be tracked over time and by location in order to determine if patterns are emerging or changing. Drug adulterations have important implications from a health promotion perspective, from a drug user perspective, and also from a law enforcement perspective. Cross-tabulations will be utilized to explore these patterns.
Suspected versus Actual
Another key element that can be obtained from the substances analyzed by the Drug Analysis Service is the comparison of what was suspected by police officers and border agents when a sample was seized versus what was actually found in the sample. This has implications for police/border officials in their being prepared as to what is happening at the street level and being able to identify changes in the presentation of illicit substances. These comparisons will be examined through cross-tabulations.
Case Disposition and User Information
Using the CDSD database, information will be examined on individuals who have been arrested, charged, and/or convicted under the Criminal Code of Canada, Controlled Drug and Substances Act (CDSA), the Food and Drug Act (FDA), Food and Drug Regulations (FDR), and the Narcotic Control Act (NCA). Demographic information includes: gender, age at time of arrest, province of residence, and citizenship. Furthermore, case information such as arresting police detachment, date of seizure, arrest date, charges filed, convictions, and length of sentence will be explored if applicable. These variables will be examined using cross-tabulations, analysis of variance, and descriptive statistics.
Regional Comparisons
Utilizing seizure data from the LIMS system, comparisons can be made inter-provincially based upon Statistics Canada’s census data. Based upon the level of granularity that will ensure privacy protection of the police or customs office, comparisons will be made on several regional distinctions, such as census sub-divisions or urban area coding. When plotted by year, geographical variations can be identified for each drug type. Rates will be computed using 2006 census data, and graphs will represent the number of seizures per 100,000 residents.
Statistical Analyses
Quantitative data will be analyzed using Stata 9.2. The most frequently seized substances will be coded. Cross-tabulations, regression analysis, and analysis of variance techniques will be computed to obtain descriptive statistics of the LIMS and CDSD data platforms.
Temporal and Geographic Information
All analyses will be conducted for the province of British Columbia, with comparisons to national figures when appropriate and possible. Furthermore, regional comparisons will be examined when available. Data extracted from LIMS will cover the period from January 1, 1988 to December 31, 2006, while data from the CDSD database will cover the period from January 1, 1986 to December 31, 2005. Data from the CDSD for 2006 will not be representative as it is an underestimation due to the lag in time from when the case is completed to the time it is entered into the database. As a result, this year will be excluded from all analyses using CDSD data.