Canadian heroin indicators: the effect of delay between arrest and conviction on official statistics

Sections

Abstract
Introduction
Method
Findings
Discussion
Conclusions
Bibliography
Acknowledgement

Details

Author: Eleanor THOMAS,
Pages: 33 to 38
Creation Date: 1977/01/01

Canadian heroin indicators: the effect of delay between arrest and conviction on official statistics

M.A. Eleanor THOMAS,
Epidemiology and Social Research Division, Non-Medical Use of Drugs Directorate, Health and Welfare Canada, Ottawa, Ontario.

Abstract

The value of using Bureau of Dangerous Drugs (B.D.D.) statistics on heroin convictions and heroin users as indicators of trends in heroin use in Canada has been questioned, on the grounds that in recent years court backlogs and court procedures have introduced long delays between arrest and conviction. These delays may be sufficiently long to render B.D.D. figures unrepresentative of the current situation. The present study sampled heroin convictions under the Narcotic Control Act that were reported to the Bureau from 1973 to 1975, in an attempt to estimate the average arrest-conviction interval over-all, to compare provinces, and to look at changes over time. Results showed that for simple heroin possession, the interval averaged 4.8 months, with little variation from province to province and no significant variation over time. Trafficking and possession for the purpose of trafficking resulted in longer intervals in all provinces, averaging 10.5 months over-all, but with the interval increasing over time. It was concluded that court backlogs have not rendered B.D.D. statistics on convictions for simple possession invalid as indicators of heroin use, but that long delays and changes in these delays over time between arrest and conviction for trafficking and possession for the purpose of trafficking have distorted B.D.D. figures somewhat. Therefore, trends seen in the latter data should be interpreted with caution.

Introduction

The Bureau of Dangerous Drugs (B.D.D.) annually publishes two sets of statistics on narcotics use in Canada, (1) conviction statistics and (2) user statistics. The conviction statistics report the number of convictions under the Narcotic Control Act (N.C.A.) recorded by the Bureau for the preceding calendar year. 1 The user statistics report the number of narcotics users known to the Bureau from all their information sources. In the case of heroin users, their major sources include treatment centre personnel and pharmacists who are required to report all methadone prescriptions and dispensing to the Bureau, and law enforcement sources, consisting mainly of police officers who forward reports of heroin-related convictions. 2 The conviction and user statistics are regularly used by researchers as indicators of trends in narcotics use in Canada under the assumption that numbers of convictions and numbers of reported users increase and decrease in parallel with numbers of users (LeDain, Campbell, Lehmann, Stein and Bertrand 1973, Rootman and Richman 1975, Thomas 1976). Several criticisms have been levelled at this assumption (Thomas 1975), including a concern about the effect of delays in the court process on B.D.D. conviction and user figures. Reports have noted that the backlog of cases in courts handling drug cases, especially in Vancouver, have caused delays of up to two years or more between the arrest and the conviction of some narcotic users, and that the average delay has been increasing (Tomalty 1977, C.L.E.U. 1977).

1These figures reflect convictions recorded by the Bureau for the year up to the time of tabulation, but do not include their inevitable backlog of unrecorded cases. Because tables are not updated by the Bureau once the backlog is cleared, the use of their conviction statistics as an indicator of drug use over time in Canada requires the assumption that this backlog remains constant in relative size and in the distribution of the users it includes.

The present study was undertaken to investigate several questions. (1) Is the arrest-conviction lag sufficiently long to make conviction (and user) statistics for any one year irrelevant in a discussion of current narcotics issues? (2) Do backlogs of court cases and subsequent delays in reporting of arrests to the B.D.D. differ sufficiently from province to province to invalidate comparisons? (3) Has the arrest-conviction lag changed sufficiently over time to render conviction (and user) statistics not comparable from one year to the next?

Method

An attempt was made to estimate the arrest-conviction (and notification) interval for individuals convicted on heroin related charges since 1973. Until April 1976, 3 the B.D.D. maintained a ledger for recording all convictions under the N.C.A. involving heroin. To estimate average lag a random sample was drawn from this ledger of 10 per cent of all recorded heroin convictions under the Act during 1973, 1974, and 1975. Sampling procedures involved the use of random numbers to select one case in ten. The arrest and conviction dates were recorded for the sample cases, and the arrest-conviction intervals were calculated. Mean intervals were tabulated for each province, year, and section of the Act.

2For a detailed description of B.D.D. sources, see Thomas 1975. Prior to April 1972, the Bureau of Dangerous Drugs received reports from police forces across Canada about most drug-related arrests. This information came to the Bureau on completed police report forms. When reports of drug-related arrests were received, names of arrestees were checked against existing files by the Bureau. If an individual was previously unknown, his or her name was added to the user file as a newly reported case, and was subsequently included in narcotics use statistics published by the Bureau. Later, when the case was tried or otherwise disposed of, the B.D.D. was notified of the disposition, and convictions were recorded in the conviction statistics ledger used for tabulating their annual published conviction statistics. Beginning in April 1972, reports of drug-related arrests were no longer received regularly by the B.D.D., although some police personnel did continue to send in the old forms. Rather, a new reporting procedure was instituted whereby arrested narcotics users were reported to the B.D.D. only upon conviction on a drug charge under the N.C.A. This change in procedure should have had no effect on conviction statistics, but necessarily introduced a new delay between the police-user contact and the recording of that contact in the Bureau's user statistics. Furthermore it would have altered somewhat the population being sampled by B.D.D. procedures in that narcotics-related charges that were later disposed of by other than a conviction would no longer be systematically reported to and recorded by the Bureau. Therefore, user statistics before and after 1972 are not strictly comparable, to the extent that they are based on law enforcement sources of information. However, since that date, assuming that these and other procedures remained more or less unchanged, user statistics should be comparable from year to year. For a discussion of the Bureau's other major source of user information, methadone treatment facilities, and potential problems introduced by changes in that sphere since 1972, see McKim, DaSilva, Frederick, Henderson, LeCavalier, Thomas and Wilson 1976.

3Information received after April 1976 has been compiled in machine-readable form, and should be available for analysis in 1977.

Findings

SAMPLE

A detailed breakdown of the sample by year and section appears in table 1. The sample included 132 heroin convictions in 1973, 81 in 1974, and 42 in 1975. These figures are consistent with the total heroin conviction statistics reported by the B.D.D. in their annual reports for these years: 1,290 in 1973, 798 in 1974, and 511 in 1975. All cases in the sample occurred in one of four provinces (B.C., Alta, Ont., or Que.), reflecting the distribution of narcotics cases reported by the B.D.D. (Thomas 1976).

RESULTS

Table 2 gives the mean interval between arrest and conviction for all years combined, broken down by province and section of the N.C.A. This table shows that the interval between arrest and conviction for possession of heroin (section 3(1), N.C.A.) averaged 4.8 months (standard deviation = 4.8, n=131) for the sample, compared with 11.5 months (standard deviation = 7.5, n=64) for trafficking and 9.3 months (standard deviation = 7.5, n=59) for possession for the purpose of trafficking (section 4(1) and 4(2), N.C.A.). The pattern was generally the same for all provinces, though it was less clear in Quebec where numbers were low for convictions under sections 3(1) and 4(2), and one extreme case of a 37-month interval distorted the average for possession convictions. Without that extreme case, the Quebec 3(1) mean interval of 11.5 (standard deviation = 17.5, n=4) would have been 3.0 months.

TABLE 1

Random sample a of heroin related convictions under the Narcotic Control Act recorded by the B.D.D. 1973-75, by year and section of the Act, b and total figures reported by the B.D.D. for those years

 

Year of conviction

 
 

Section

1973

1974

1975

All

3 (1)
Sample n
75 37 19 131
 
B.D.D. total n
762 445 250 1 457
 
Sample/total
0.10 0.08 0.08 0.09
4 (1)
Sample n
31 19 14 64
 
B.D.D. total n
289 185 127 601
 
Sample/total
0.11 0.10 0.11 0.11
4 (2)
Sample n
25 25 9 59
 
B.D.D. total n
235 167 132 534
 
Sample/total
0.11 0.15 0.07 0.11
5 (1)
Sample n
1 0 0 1
 
B.D.D. total n
4 1 2 7
 
Sample/total
0.25
-
-
0.14
All
Sample n
132 81 42 255
 
B.D.D. total n
1 290 798 511 2 599
 
Sample/total
0.10 0.10 0.08 0.10

aSampling technique described in text.

bN.C.A. Section: 3(1) possession; 4(1) trafficking; 4(2) possession for the purpose of trafficking; 5(1) importing.

TABLE 2

Average interval in months between arrest and conviction under the Narcotic Control Act for heroin-related offences for a sample of all such conviction cases recorded by the B.D.D., 1973-75, by province and section of the Act a

 

Section of the N.C.A.

Province

3(1)

4(l)

4(2)

4(1) + 4(2)

5(1)

British Columbia
         
Mean interval (mo.)
4.5 10.7 9.1 9.8
-
Number
107 26 31 57
-
(Standard deviation)
(4.0) (7.5) (6.9) (7.1)  
Alta
         
Mean interval
3.6 7.1 8.3 7.4 10.0
Number
9 17 8 25 1
(Standard deviation)
(3.4) (2.5) (5.3) (3.5)
(-)
Ontario
         
Mean interval
6.5 12.8 10.5 11.4
-
Number
11 12 19 31
-
(Standard deviation)
(3.1) (5.1) (9.2) (7.8)  
Quebec
         
Mean interval
11.5 20.7 1.0 18.7
-
Number
4 9 1 10
-
(Standard deviation)
(17.5) (9.0)
( - )
(10.5)  
All
         
Mean interval
4.8 11.5 9.3 10.5 10.1
Number
131 64 59 123 1
(Standard deviation)
(4.8) (7.5) (7.5) (7.5)
( - )

aN.C.A, Section: 3(1) possession; 4(1) trafficking; 4(2) possession for the purpose of trafficking; 5(1) importing.

Table 3 presents heroin possession convictions, (section 3(1), N.C.A.) by province and year, from 1973 to 1975. In British Columbia, which was the only province with numbers sufficiently large to be meaningful for all three years, the average interval was relatively constant: 4.0 months (standard deviation = 3.5, n=61) in 1973, 5.7 months (standard deviation = 4.6, n=30) in 1974, and 4.4 months (standard deviation = 4.0, n=16) in 1975. Total figures for Canada reflect this same pattern.

TABLE 3

Average interval in months between arrest and conviction under Section 3(1) of the Narcotic Control Act, for possession of heroin, for a sample of all such conviction cases recorded by the B.D.D., by province and year

 

Year of conviction

 

Province

1973

1974

1975

All

British Columbia
       
Mean interval
4.0 5.7 4.4 4.5
Number
61 30 16 107
Standard deviation
(3.5) (4.6) (4.0) (4.0)
Alta
       
Mean interval
2.7 9.0 4.0 3.6
Number
7 1 1 9
Standard deviation
(3.1)
( - )
( - )
(3.4)
Ontario
       
Mean interval
8.4 5.0 5.0 6.5
Number
5 4 2 11
Standard deviation
(2.2) (3.6) (2.8) (3.1)
Quebec
       
Mean interval
<1
23.0 0 11.5
Number
2 2 0 4
Standard deviation
(<1)
(19.8)
( - )
(17.5)
All
       
Mean interval
4.1 6.6 4.4 4.8
Number
75 37 19 131
Standard deviation
(3.6) (6.7) (3.7) (4.8)

Table 4 presents data on the arrest-conviction interval for convictions on charges of trafficking and possession for the purpose of trafficking (section 4(1) and 4(2), N.C.A.) in heroin. The mean interval increased for Canada and for all provinces except B.C. In B.C. an increase from 7.4 months (standard deviation = 5.0, n=28) in 1973 to 12.5 months (standard deviation = 8.3, n=19) in 1974 was followed by no change to 1975 when the mean was 11.6 months (standard deviation = 8.0, n=10).

Discussion

Sample numbers were small in some of the cells in the tables. Extreme scores, therefore, distorted findings somewhat, but the general picture was one of a relatively short mean arrest-conviction interval for simple heroin possession charges, combined with a relatively long interval for trafficking charges, with that interval increasing over the three year period.

These findings suggest that B.D.D. statistics on heroin possession convictions reflect fairly well current law enforcement activity in that sphere, and that declines seen in conviction statistics are not an artefact of increasing delays in court procedures. Therefore, B.D.D. figures on possession convictions are comparable, from year to year, at least since 1973. However, because sample numbers were fairly small in all provinces except B.C., reliable interprovincial comparisons cannot be made.

TABLE 4

Average interval in months between arrest and conviction under section 4(1) and 4(2) combined of the Narcotic Control Act, for trafficking and possession for the purpose of trafficking in heroin, for a sample of all such conviction cases recorded by the B.D.D. by province and year

 

Year of conviction

 

Province

1973

1974

1975

All

British Columbia
       
Mean interval
7.4 12.5 11.6 9.8
Number
28 19 10 57
Standard deviation
(5.0) (8.3) (8.0) (7.1)
Alta
       
Mean interval
6.1 7.8 9.3 7.4
Number
9 12 4 25
Standard deviation
(2.7) (2.9) (6.2) (3.5)
Ontario
       
Mean interval
10.9 9.5 20.7 11.4
Number
18 10 3 31
Standard deviation
(5.7) (5.3) (19.1) (7.8)
Quebec
       
Mean interval
1.0 16.0 23.0 18.7
Number
1 3 6 10
Standard deviation
( - )
(13.2) (6.3) (10.5)
All
       
Mean interval
8.2 10.8 15.3 10.5
Number
56 44 23 123
Standard deviation
(5.3) (7.2) (10.3) (7.5)

The findings of this study cast a different light on statistics regarding convictions for trafficking and possession for the purpose of trafficking in heroin, however. The data indicated that this interval was longer than that for possession of heroin in all provinces over most of the three-year period. Futhermore, the arrest-conviction interval for these charges increased over time. These findings suggest that B.D.D. statistics on heroin trafficking convictions, (section 4(1) and 4(2) of the N.C.A.) do not reflect current police activity. Declines in the number of persons convicted for these offences may be due, in part, to the length of this arrest-conviction interval, possibly caused by delays in the legal process. Therefore, B.D.D. statistics on trafficking convictions may not be comparable from year to year, and may increasingly under-represent enforcement activity against traffickers since 1973.

B.D.D. user statistics are based in part on their conviction statistics, including not only possessional but also trafficking convictions. Because the relative contribution of reported trafficking convictions to B.D.D. annual user statistics is unknown, the extent of an artefactual decline, if any, in these statistics is unknown.

Conclusions

B.D.D. conviction statistics and user statistics are two major indicators of heroin use trends in Canada. Their use as estimators involves many assumptions, including consistency of reporting procedures. The present study suggests that the number of convictions under the N.C.A. for simple possession of heroin (section 3(1), N.C.A.) can be used as an indicator with some confidence that the backlog of court cases has not caused an unreasonable delay between the initial police contact at the time of arrest and the final report to the Bureau of the conviction, and that this delay has been fairly constant since 1973. The longer average arrest-conviction delays on trafficking charges (section 4(1) and 4(2)), and the fact that this lag has increased since 1973, suggests that caution should be used in interpreting trafficking conviction trends. Futhermore, if user statistics that are based on these conviction statistics are used as indicators of heroin use, conclusions should be carefully qualified.

Bibliography

C.L.E.U. (Co-ordinated Law Enforcement Unit), A Proposal for Reducing Drug Trafficking and Abuse in British Columbia, Ministry of Attorney-General of British Columbia, March 1977.

LeDain G., J. L. Campbell, H. E. Lehmann, J. P. Stein and M. A. Bertrand. Final Report of the Commission of Inquiry into the Non-Medical Use of Drugs, Information Canada, Ottawa 1973.

Rootman I. and A. Richman. Trends in reported illegal narcotic use in Canada: 1956-1973. Bulletin on Narcotics, XXVII:4, 27-40, 1975.

McKim T. R., T. DaSilva, G. L. Frederick, I. Henderson, J. S. LeCavalier, E. M. Thomas and E. Wilson. Trends in Methadone Use in the Treatment of Opiate Dependence in Canada, Report of a special committee established by the Health Protection Branch, Non-Medical Use of Drugs Directorate, Health and Welfare Canada, Ottawa 1976.

Thomas E. M. The epidemiology of narcotic-related problems in Canada: use of the B.D.D. narcotic users file. In Proceedings, Epidemiology of Drug-Related Problems in Canada: A Pilot Workshop, Non-Medical Use of Drugs Directorate, Health and Welfare Canada, Hawkesbury, Ontario, March 1975.

Thomas E. M. Illicit use of Narcotic Control Act Drugs in Canada: Study of Users First Identified in 1973 and 1974, Non-Medical Use of Drugs Directorate, Health and Welfare Canada, Ottawa 1976.

Tomalty, G. L. (Officer in Charge, R.C.M. Police Drug Enforcement Branch, Ottawa), Personal communication, January 11, 1977.

Acknowledgement

The assistance of Bruce MacLean in the compilation and tabulation of the data is acknowledged.