- Email: [email protected]
Pretrial release performance of addict defendants: Examination of court non-appearance and rearrest rates
Drug and Alcohol Dependence, 5 (1980) 311 - 319 0 Elsevier Sequoia S.A., Lausanne - Printed in the Netherlands
311
PRETRIAL RELEASE PERFORMANCE OF ADDICT DEFENDANTS: EXAMINATION OF COURT NON-APPEARANCE AND REARREST RATES*
JAMES C. WEISSMAN Department
of Health and Hospitals, Denver, Colorado (U.S.A.)
SAMUEL W. MARR and ELAINE LAWRIE Boulder County Jail, Boulder,
Colorado (U.S.A.)
(Received August 15, 1979)
Summary The relationship between addiction and pretrial performance has attracted increasing attention in the diversion and bail study literature. Many jurisdictions have imposed special restrictive bail conditions upon the bail agreements of addict defendants on the heretofore unproven assumption that appropriate remedial conditions of release (for example, supervised drug abuse treatment) will improve the pretrial release performance of addict defendants. To determine the validity of these assumptions, the pretrial release performance of three groups (addict defendants without treatment supervision, addict defendants involved in a supervised drug abuse treatment program, and non-addict, unsupervised felony defendants) was examined. The results indicate that, measured in terms of appearance at scheduled court proceedings, the unsupervised, untreated addict defendants are poorer bond risks than the non-addict felony defendants, but that the court appearance performance of the supervised, treated addicts was equivalent to the performance of the non-addicts. Rearrest data showed virtually no difference between the groups for non-drug offenses, although the addict groups displayed significantly higher overall rearrest rates.
Introduction “If the Federal or State court assumes that addicts are less likely to appear at trial than non-addicts, it is only logical to condition release upon *This study was conducted under the auspices of the Denver TASC (Treatment Alternatives to Street Crime) project which is supported by the U.S. Department of Justice, Law Enforcement Assistance Administration, Denver High Impact Anti-Crime Program 72-IC-0002~(1)95.
312
successful participation in TASC. However, upon challenge, this assumption may be difficult to support in the absence of empirical data showing addicts are less likely to appear at trial.” [l] . In recent years most jurisdictions have enacted comprehensive bail statutes, setting forth both the purposes of bail and the procedures to be employed in the determination of the amount of bail, type of bond, and conditions of release [ 21. Legal scholars attribute various purposes to bail: protection of the community, prevention of punishment absent a finding of guilt, therapeutic goals, and the assuring of a defendant’s appearance in court [ 1, 21. The design and use of a bail system is, however, restricted by the Eighth Amendment to the United States Constitution which states that “excessive bail shall not be required”. The United States Supreme Court has interpreted that phrase to mean that the only permissible function of bail is to assure the defendant’s future appearances in court *. Therefore, the only constitutionally proper inquiry relates to determining the amount of bond (dollars), type of bond (cash or personal recognizance, for example), and conditions of release (for example, travel restrictions) which will assure appearance of the defendant at future court hearings. Traditional bail proceedings have involved only a determination of the amount and type of bond to be required by the court. Magistrates now, however, frequently add “special conditions” of release to the standard bond agreement. Modern bail statutes often explicitly encourage the use of “special conditions” considered by the magistrates to be reasonably related to the purpose of bail. Third-party supervision, restrictions of travel and association, and participation in a pretrial intervention program are common examples of “special conditions” [ 31. Upon challenge of such “special conditions” as an interference with the Eighth Amendment guarantee against excessive bail, the courts have upheld the use of the conditions only to the extent that they constitute “the least restrictive condition assuring future court appearance” [ 3]* *. *The Eighth Amendment prohibition against excessive bail is one of the least understood guarantees of the Bill of Rights. It has not yet been incorporated to the States, and there are few definitive federal appellate cases concerning the Eighth Amendment bail provision. The only major United States Supreme Court case, Stack u. Boyle 342 U.S. 1 (1951), while upholding the right of the legislature to provide a statutory scheme denying bail for capital offense defendants, held that bail may not be set “at a figure higher than an amount reasonably calculated to assure the defendant’s appearance”. This holding, along with various other federal appellate decisions, has led legal commentators to conclude universally that the only constitutionally justifiable function of bail is to assure a defendant’s appearance at trial (“TASC Legal Analysis”, 1973:4). **The ABA’s National Pretrial Intervention Service Center’s interpretation [ 31 of Stack u. Boyle and the other applicable federal appellate cases is that the test of the constitutionality of a special bail condition is “the relevance of that condition to assuring future appearances and whether the condition is excessive, since the least restrictive condition assuring future appearances is the one to be imposed. ” The authors concur in this interpretation of the case law.
313
Both the bail and pretrial intervention literature ascribe unique consideration for addict defendants [3 - 61. Despite a total absence of empirical data, commentators have widely held that addicts present poor bail risks as compared to non-addicts [ 1, 3, 41. A vivid example of this attitude is evidenced in the statement of Chief Judge Harold H. Greene [4] of the District of Columbia Superior Cqurt that “. . . the addiction of an individual bears directly upon the likelihood of his appearance for trial. Because of their personalities and because of the constant pressures of the craving for narcotics, addicts as a group are less stable and less reliable than other citizens.” The general legal rule holds that “special conditions” of bail are permissible only to the extent that they constitute the “least restrictive condition necessary to assure future court appearance”. If the addict is as poor a bond risk as Judge Greene and others believe, the imposition of “special conditions” (for example, supervised drug abuse treatment) reasonably calculated to assure future court appearance of the addict defendant would seem entirely proper. Previous studies, however, do not supply empirical data pertaining to the pretrial release performance record of the addict defendant. Research questions The beliefs that unsupervised, untreated addicts are poorer bond risks compared to non-addicts and that participation in a supervised drug abuse treatment program improves the court performance of addict defendants are clearly unanswered research questions. We have undertaken the present study to address the following questions concerning addiction and pretrial release performance: (1) Are unsupervised, untreated addicts poorer bond risks as compared to non-addict defendants with regard both to court appearance and rearrest? (2) Does involvement in supervised drug abuse treatment affect the pretrial release performance of addicts? Method In an attempt to answer these questions, the authors compared the court appearance activities and rearrest data of three groups of defendants released on bond. Group 1 consisted of addicts not involved in a court-supervised drug abuse treatment program during the pretrial release period; group 2 comprised addicts participating in the Denver TASC program during the pretrial release period* ; and group 3 consisted of non-addict felony defendants not involved in any program of supervision during the pretrial release period. *TASC (Treatment Alternatives to Street Crime) is a federally sponsored addict diversion program. The methodology of TASC involves the identification, screening, and treatment referral of addicts entering municipal and county jails. Over fifty cities have participated in the National TASC Program at a cost in excess of $30 million,
314
The group 1 subjects (non-TASC addicts) were identified as current regular opiate users during the TASC screening at the Denver City Jail. They were qualified to participate in TASC-monitored drug abuse evaluation and treatment, but for various reasons never became actively involved in the TASC process*. Group 2 (TASC clients) was also composed of addicts screened by TASC at Denver City Jail, but these subjects became successfully enrolled in TASC drug abuse evaluation and treatment. The subjects of these addict groups were chosen serially; i.e. the first in number of individuals meeting the criteria specified for inclusion in each respective group. Group 3 (felons), however, comprised subjects randomly selected from all criminal defendants against whom felony information was filed in the first three months of 1974. Each group comprised 50 subjects. With respect to demographic factors, the groups were generally comparable. The addict groups (groups 1 and 2) contained 26 per cent and 24 per cent females, respectively, compared to only 16.0 per cent of females in group 3. Group 2 had the least number of Whites (16.0 per cent), and group 3 had the largest concentration of Whites (40.0 per cent) and the smallest concentration of Chicanos (24.0 per cent). Group 1 showed a relative balance between the three racial groupings.
Results The type of bond received and the stage at which bonded show similar distributions among the three groups (Table 1). The most frequently occurring pattern is that of an individual receiving either a personal recognizance or professional surety bond immediately upon arrest or at the time of the rights advisement. One noticeable difference is the number of group 2 subjects receiving personal recognizance bonds; approximately twice as many group 2 subjects received personal recognizance bonds as group 1 and group 3 subjects. That apparent dissimilarity must be considered, however, within the context of the bail system operative in the study locale. Personal recognizance bonds are granted, for the most part, upon motions for reduced bail. Since 72.0 per cent of the group 1 (non-TASC addicts) subjects had been bonded immediately after arrest, only 28.0 per cent remained eligible for the bond reduction process which produced personal recognizance bond agreements.
*The following reasons, in order of decreasing frequency, were given for not participating in the TASC alternative: unsure about the desirability of TASC, not interested, and no reason given.
315 TABLE 1 Type and stage of bond Group 1 (non-TASC addicts)
Group 2 (TASC clients)
Group 3 (felons)
n
%
n
%
n
%
Type of bond* None required Personal recognizance Professional surety Personal cash/Property/Other Total
5 10 30 5 50
10.0 20.0 60.0 10.0 100.0
1 20 25 4 50
2.0 40.0 50.0 8.0 100.0
11 34 5 50
-
Stage bonded Immediately after arrest Rights advisement Post rights advisement Total
36 12 2 50
72.0 24.0 4.0 100.0
30 13 7 50
60.0 26.0 14.0 100.0
30 15 5 50
60.0 30.0 10.0 100.0
22.0 68.0 10.0 100.0
n = 150, d.f. = 4, x2 = 1.57, N.S. *Chisquare
test precluded due to presence of empty cell.
The groups differed substantially with regard to the various appearance variables. The group 2 subjects have the greatest number of appearances required*, with an average number of 5.08 appearances required per individual. The group 1 subjects average 4.18 appearances required, and the group 3 subjects average only 3.70 appearances required. The number of appearances required statistic merely reflects the total court involvement in the various cases and has no discernible intrinsic criminal justice significance**. The primary factor under study is the relative success of the groups with regard to actual attendance at required court appearances. Three different methods can be used to measure appearance performance. First, the frequency of failures can be shown to describe each group’s performance. Secondly, failure rates can be computed for each individual. Comparison of the failure rates per individual by t-test will establish which groups’ members exhibited the least failures. The third method of analysis is the failure rate per scheduled appearance which will eliminate any differences attributable to the disparate number of required appearances. Each group’s frequency of failure is displayed in Table 2. Group 1 (nonTASC addicts) shows a frequency of failure markedly greater than that of the other two groups. Nearly one half (46.0 per cent) of the group failed to ap*This phrase refers to a scheduled court appearance at which the subject’s attendance is mandatory. **No correlation was discovered between the number of appearances required variable and the outcome variables under study; i.e. court appearance success, final disposition, and rearrest.
316 TABLE
2
Appearance
Proportion
performance
of group
No failures One or more n = 150,
variables
addicts)
Group (TASC
2 clients)
Group 3 (felons)
n
%
n
%
n
%
27 23
54 46
36 14
12 28
38 12
16 24
failing
failures
d.f. = 2, x2 = 6.55,
Frequency
Group 1 (non-TAX
p < 0.05
of failure
1 time 2 times 3 times
12 8 2
24 16 4
9 3 2
18 6 4
10 2
20 4 _
4 times Subtotal
1 23
2 46
14
28
12
24
Group 1 (non-TAX
addicts)
Group (TASC
TABLE
_
3
Comparison
of appearance
measures
Failure rate per individual Total number of failures Total number of individuals Mean failure rate per individual Standard deviation Rate comparison by Group 1 us group Group 2 us group Group 1 us group
t-test 2 3 3
Failure rate per appointment Total number of failures Total number of required appearances Mean failure rate per appointment Standard deviation Rate comparison by Group 1 us group Group 2 us group Group 1 US group
t-test 2 3 3
38 50 0.76 0.96
t = 1.96 t = 1.06 t = 3.12
2 clients)
21 50
Group 3 (felons)
14 50
0.42 0.17
0.28 0.52
d.f. = 98 d.f. = 98 d.f. = 98
p < 0.025 N.S. p < 0.005
38
21
14
209
253
185
0.182 0.58
t = 2.21 t = 0.211 t = 2.26
0.083 0.37
d.f. = 98 d.f. = 98 d.f. = 98
0.076 0.30
p < 0.025 N.S. p < 0.025
317
pear on at least one occasion; 22.0 per cent of the group failed to appear on two or more occasions. The members of group 2 (TASC clients) were very similar to the members of group 3 (felons) with respect to frequency of failures. Upon application of the Chi-square test, these differences achieve statistical significance @ < 0.05). The difference between the two addict groups (1 and 2) is further evidenced by the statistical analysis of failure rate per individual. As shown in Table 3, there is a statistically significant difference (p < 0.025) between the mean failure rate per person of the members of group 1 (non-TASC addicts) and members of group 2 (TASC clients). The suggestion that TASC involvement is associated with an improvement of the appearance performance of addicts is strengthened by the finding that there is no statistically significant difference between the mean rates of group 2 (TASC clients) and group 3 (felons). However, the difference in mean failure rate per individual between the non-TASC addicts and the felons is statistically significant (p < 0.005). Computation of failure rate per scheduled appearance purports to examine failure rates while eliminating the effects of the differing number of required appearances among the groups. Analysis by this technique supports the indication of the prior two computations. Again, group 1 has the highest rate of failure; group 2 shows a rate comparable to the non-addict group 3. The difference in mean failure rate per appointment is statistically significant between group 1 and group 2 (p < 0.025) as well as with group 3 (p < 0.025). There is no significant difference between TASC clients and felons. Measured in terms of rearrest, the pretrial performances of the three groups differ substantially. The rearrest data are divided into three categories: drug arrests, non-drug arrests, and total arrests. These data are examined in frequency distribution format (none/one or more arrests) and group rearrest rates per month of exposure are compared (Table 4). Both analyses indicate a similar pattern. Little difference is revealed with regard to rearrests among the addict groups. The non-addict group, however, displays a substantially reduced rearrest pattern. Notwithstanding the fact that the group 3 subjects (felons) show the lowest overall rearrest performance, a basic equivalency of the three groups is obscured by the impact of drug arrests of groups 1 and 2; the non-drug rearrest measures are comparable among the three groups. The differences in rearrest frequencies and rates between the addict groups and non-addict groups achieved a level of statistical significance. Statistical significance, however, was not achieved when rearrest measures between the two addict groups were compared.
Discussion Several methodological shortcomings limit the conclusions to be drawn and require mention at the outset of this discussion section. First, the study groups were not particularly large, and the serial selection nature of the two
318 TABLE
4
Rearrest
data
A. Frequency
comparison
Drug arrests None One or more Chi-square
analysis
precluded
Total arrests None One or more n = 150,
d.f. = 2, x2 = 14.75,~
B. Rearrest
rates per month
addicts)
Group 2 (TASC clients)
Group 3 (felons)
n
%
n
%
n
%
32 18
64 36
33 17
66 34
50 0
100 0
due to empty
Non-drug arrests None One or more
n = 150, d.f. = 2, x2 = 2.22,
Group 1 (non-TASC
40 10
80 20
36 14
72 28
42 8
84 16
27 23
54 46
25 25
50 50
42 8
84 16
1 us
Group
N.S.
< 0.001
of exposure Group group
Drug arrests Non-drug Total
*d.f.
arrests
arrests
cell
comparisons* 1 us 2
t = 0.481,
Group group
N.S.
3
group _
(no group
3 drug arrests) N.S.
t = 0.871,
N.S.
t = 0.448,
t = 0.176,
N.S.
t = 3.33, p < 0.01
2 us 3
t = 1.35,
N.S.
t = 3.04, p < 0.01
= 98.
addict groups was not as desirable as the classic random technique. Second, the available data concerning the non-TASC addicts do not adequately address the important dynamics of the group members’ failure to become involved with the TASC process. Without discerning the nature of this phenomenon, the assumption of comparability between the TASC clients and nonTASC addicts is arguable. Finally, the study lacks cleanness. Although the authors were familiar with area treatment program rosters, confidentiality barred the rigorous examination of treatment records to determine unknown treatment participation by the non-TASC addicts. Also, there is a lack of uniformity concerning the circumstances under which the TASC clients became involved in TASC supervision during the pretrial release period. Some were ordered to participate as a formal condition of personal recognizance
319
bond release, but most subjects participated as a result of judicial exhortation, not ordering. Nonetheless, given these practical research constraints, the study data show that unsupervised, untreated addicts do perform substantially more poorly as compared to non-addict defendants with respect to required court appearances while on bond. However, involvement in supervised drug abuse treatment is associated with improved court appearance performance of addict defendants, resulting in the achievement of parity with non-addict defendants. With regard to rearrest, the data display virtually no difference between the two addict groups. Both addict groups were rearrested at a high percentage. The non-addict subjects showed much lower overall rearrest figures. Comparison with non-drug rearrest figures, however, failed to demonstrate significant differences between the addict groups and the non-addict group. The data thus lend support to the proponents of TASC-type addict diversion programs. Although the results do not provide a definitive answer to the legal question of whether or not TASC bail conditions represent “the least restrictive conditions necessary to guarantee future court appearance” for addict defendants, they do provide support for that position by furnishing highly relevant empirically based information heretofore unavailable. Future investigators should refine the group selection techniques and expand the size and geographic scope of sampling to test the significance of these findings; the external validity of this study is limited by the size and geographic restrictions of the sample design.
References 1 TASC Legal Analysis, Special Action Office for Drug Abuse Prevention, Washington, D.C., 1973, I, iii. 2 P. B. Wice, Bail and Its Reform: A National Survey, National Institute of Law Enforcement and Criminal Justice, Law Enforcement Assistance Administration, U.S. Department of Justice, Washington, D.C., 1973. 3 Legal Issues and Characteristics of Pretrial Intervention Programs, National Pretrial Intervention Service Center of the American Bar Association Commission on Correctional Facilities and Services, Washington, D.C., 1974. 4 H. H. Greene, Proceedings of the First National Conference on Treatment Alternatives to Street Crime, Special Action Office for Drug Abuse Prevention, Washington, D.C., 1973, pp. 1 - 16. 5 M. P. Pines, Answer to the problem of bail: A proposal in need of empirical confirmation. Columbia J. Law Sot. Probl., 9 (1973) 394 - 441. 6 D. L. Skoler, Protecting the rights of defendants in pretrial intervention programs. Criminal Law Bull., 10 (1974) 473 - 492.
311
PRETRIAL RELEASE PERFORMANCE OF ADDICT DEFENDANTS: EXAMINATION OF COURT NON-APPEARANCE AND REARREST RATES*
JAMES C. WEISSMAN Department
of Health and Hospitals, Denver, Colorado (U.S.A.)
SAMUEL W. MARR and ELAINE LAWRIE Boulder County Jail, Boulder,
Colorado (U.S.A.)
(Received August 15, 1979)
Summary The relationship between addiction and pretrial performance has attracted increasing attention in the diversion and bail study literature. Many jurisdictions have imposed special restrictive bail conditions upon the bail agreements of addict defendants on the heretofore unproven assumption that appropriate remedial conditions of release (for example, supervised drug abuse treatment) will improve the pretrial release performance of addict defendants. To determine the validity of these assumptions, the pretrial release performance of three groups (addict defendants without treatment supervision, addict defendants involved in a supervised drug abuse treatment program, and non-addict, unsupervised felony defendants) was examined. The results indicate that, measured in terms of appearance at scheduled court proceedings, the unsupervised, untreated addict defendants are poorer bond risks than the non-addict felony defendants, but that the court appearance performance of the supervised, treated addicts was equivalent to the performance of the non-addicts. Rearrest data showed virtually no difference between the groups for non-drug offenses, although the addict groups displayed significantly higher overall rearrest rates.
Introduction “If the Federal or State court assumes that addicts are less likely to appear at trial than non-addicts, it is only logical to condition release upon *This study was conducted under the auspices of the Denver TASC (Treatment Alternatives to Street Crime) project which is supported by the U.S. Department of Justice, Law Enforcement Assistance Administration, Denver High Impact Anti-Crime Program 72-IC-0002~(1)95.
312
successful participation in TASC. However, upon challenge, this assumption may be difficult to support in the absence of empirical data showing addicts are less likely to appear at trial.” [l] . In recent years most jurisdictions have enacted comprehensive bail statutes, setting forth both the purposes of bail and the procedures to be employed in the determination of the amount of bail, type of bond, and conditions of release [ 21. Legal scholars attribute various purposes to bail: protection of the community, prevention of punishment absent a finding of guilt, therapeutic goals, and the assuring of a defendant’s appearance in court [ 1, 21. The design and use of a bail system is, however, restricted by the Eighth Amendment to the United States Constitution which states that “excessive bail shall not be required”. The United States Supreme Court has interpreted that phrase to mean that the only permissible function of bail is to assure the defendant’s future appearances in court *. Therefore, the only constitutionally proper inquiry relates to determining the amount of bond (dollars), type of bond (cash or personal recognizance, for example), and conditions of release (for example, travel restrictions) which will assure appearance of the defendant at future court hearings. Traditional bail proceedings have involved only a determination of the amount and type of bond to be required by the court. Magistrates now, however, frequently add “special conditions” of release to the standard bond agreement. Modern bail statutes often explicitly encourage the use of “special conditions” considered by the magistrates to be reasonably related to the purpose of bail. Third-party supervision, restrictions of travel and association, and participation in a pretrial intervention program are common examples of “special conditions” [ 31. Upon challenge of such “special conditions” as an interference with the Eighth Amendment guarantee against excessive bail, the courts have upheld the use of the conditions only to the extent that they constitute “the least restrictive condition assuring future court appearance” [ 3]* *. *The Eighth Amendment prohibition against excessive bail is one of the least understood guarantees of the Bill of Rights. It has not yet been incorporated to the States, and there are few definitive federal appellate cases concerning the Eighth Amendment bail provision. The only major United States Supreme Court case, Stack u. Boyle 342 U.S. 1 (1951), while upholding the right of the legislature to provide a statutory scheme denying bail for capital offense defendants, held that bail may not be set “at a figure higher than an amount reasonably calculated to assure the defendant’s appearance”. This holding, along with various other federal appellate decisions, has led legal commentators to conclude universally that the only constitutionally justifiable function of bail is to assure a defendant’s appearance at trial (“TASC Legal Analysis”, 1973:4). **The ABA’s National Pretrial Intervention Service Center’s interpretation [ 31 of Stack u. Boyle and the other applicable federal appellate cases is that the test of the constitutionality of a special bail condition is “the relevance of that condition to assuring future appearances and whether the condition is excessive, since the least restrictive condition assuring future appearances is the one to be imposed. ” The authors concur in this interpretation of the case law.
313
Both the bail and pretrial intervention literature ascribe unique consideration for addict defendants [3 - 61. Despite a total absence of empirical data, commentators have widely held that addicts present poor bail risks as compared to non-addicts [ 1, 3, 41. A vivid example of this attitude is evidenced in the statement of Chief Judge Harold H. Greene [4] of the District of Columbia Superior Cqurt that “. . . the addiction of an individual bears directly upon the likelihood of his appearance for trial. Because of their personalities and because of the constant pressures of the craving for narcotics, addicts as a group are less stable and less reliable than other citizens.” The general legal rule holds that “special conditions” of bail are permissible only to the extent that they constitute the “least restrictive condition necessary to assure future court appearance”. If the addict is as poor a bond risk as Judge Greene and others believe, the imposition of “special conditions” (for example, supervised drug abuse treatment) reasonably calculated to assure future court appearance of the addict defendant would seem entirely proper. Previous studies, however, do not supply empirical data pertaining to the pretrial release performance record of the addict defendant. Research questions The beliefs that unsupervised, untreated addicts are poorer bond risks compared to non-addicts and that participation in a supervised drug abuse treatment program improves the court performance of addict defendants are clearly unanswered research questions. We have undertaken the present study to address the following questions concerning addiction and pretrial release performance: (1) Are unsupervised, untreated addicts poorer bond risks as compared to non-addict defendants with regard both to court appearance and rearrest? (2) Does involvement in supervised drug abuse treatment affect the pretrial release performance of addicts? Method In an attempt to answer these questions, the authors compared the court appearance activities and rearrest data of three groups of defendants released on bond. Group 1 consisted of addicts not involved in a court-supervised drug abuse treatment program during the pretrial release period; group 2 comprised addicts participating in the Denver TASC program during the pretrial release period* ; and group 3 consisted of non-addict felony defendants not involved in any program of supervision during the pretrial release period. *TASC (Treatment Alternatives to Street Crime) is a federally sponsored addict diversion program. The methodology of TASC involves the identification, screening, and treatment referral of addicts entering municipal and county jails. Over fifty cities have participated in the National TASC Program at a cost in excess of $30 million,
314
The group 1 subjects (non-TASC addicts) were identified as current regular opiate users during the TASC screening at the Denver City Jail. They were qualified to participate in TASC-monitored drug abuse evaluation and treatment, but for various reasons never became actively involved in the TASC process*. Group 2 (TASC clients) was also composed of addicts screened by TASC at Denver City Jail, but these subjects became successfully enrolled in TASC drug abuse evaluation and treatment. The subjects of these addict groups were chosen serially; i.e. the first in number of individuals meeting the criteria specified for inclusion in each respective group. Group 3 (felons), however, comprised subjects randomly selected from all criminal defendants against whom felony information was filed in the first three months of 1974. Each group comprised 50 subjects. With respect to demographic factors, the groups were generally comparable. The addict groups (groups 1 and 2) contained 26 per cent and 24 per cent females, respectively, compared to only 16.0 per cent of females in group 3. Group 2 had the least number of Whites (16.0 per cent), and group 3 had the largest concentration of Whites (40.0 per cent) and the smallest concentration of Chicanos (24.0 per cent). Group 1 showed a relative balance between the three racial groupings.
Results The type of bond received and the stage at which bonded show similar distributions among the three groups (Table 1). The most frequently occurring pattern is that of an individual receiving either a personal recognizance or professional surety bond immediately upon arrest or at the time of the rights advisement. One noticeable difference is the number of group 2 subjects receiving personal recognizance bonds; approximately twice as many group 2 subjects received personal recognizance bonds as group 1 and group 3 subjects. That apparent dissimilarity must be considered, however, within the context of the bail system operative in the study locale. Personal recognizance bonds are granted, for the most part, upon motions for reduced bail. Since 72.0 per cent of the group 1 (non-TASC addicts) subjects had been bonded immediately after arrest, only 28.0 per cent remained eligible for the bond reduction process which produced personal recognizance bond agreements.
*The following reasons, in order of decreasing frequency, were given for not participating in the TASC alternative: unsure about the desirability of TASC, not interested, and no reason given.
315 TABLE 1 Type and stage of bond Group 1 (non-TASC addicts)
Group 2 (TASC clients)
Group 3 (felons)
n
%
n
%
n
%
Type of bond* None required Personal recognizance Professional surety Personal cash/Property/Other Total
5 10 30 5 50
10.0 20.0 60.0 10.0 100.0
1 20 25 4 50
2.0 40.0 50.0 8.0 100.0
11 34 5 50
-
Stage bonded Immediately after arrest Rights advisement Post rights advisement Total
36 12 2 50
72.0 24.0 4.0 100.0
30 13 7 50
60.0 26.0 14.0 100.0
30 15 5 50
60.0 30.0 10.0 100.0
22.0 68.0 10.0 100.0
n = 150, d.f. = 4, x2 = 1.57, N.S. *Chisquare
test precluded due to presence of empty cell.
The groups differed substantially with regard to the various appearance variables. The group 2 subjects have the greatest number of appearances required*, with an average number of 5.08 appearances required per individual. The group 1 subjects average 4.18 appearances required, and the group 3 subjects average only 3.70 appearances required. The number of appearances required statistic merely reflects the total court involvement in the various cases and has no discernible intrinsic criminal justice significance**. The primary factor under study is the relative success of the groups with regard to actual attendance at required court appearances. Three different methods can be used to measure appearance performance. First, the frequency of failures can be shown to describe each group’s performance. Secondly, failure rates can be computed for each individual. Comparison of the failure rates per individual by t-test will establish which groups’ members exhibited the least failures. The third method of analysis is the failure rate per scheduled appearance which will eliminate any differences attributable to the disparate number of required appearances. Each group’s frequency of failure is displayed in Table 2. Group 1 (nonTASC addicts) shows a frequency of failure markedly greater than that of the other two groups. Nearly one half (46.0 per cent) of the group failed to ap*This phrase refers to a scheduled court appearance at which the subject’s attendance is mandatory. **No correlation was discovered between the number of appearances required variable and the outcome variables under study; i.e. court appearance success, final disposition, and rearrest.
316 TABLE
2
Appearance
Proportion
performance
of group
No failures One or more n = 150,
variables
addicts)
Group (TASC
2 clients)
Group 3 (felons)
n
%
n
%
n
%
27 23
54 46
36 14
12 28
38 12
16 24
failing
failures
d.f. = 2, x2 = 6.55,
Frequency
Group 1 (non-TAX
p < 0.05
of failure
1 time 2 times 3 times
12 8 2
24 16 4
9 3 2
18 6 4
10 2
20 4 _
4 times Subtotal
1 23
2 46
14
28
12
24
Group 1 (non-TAX
addicts)
Group (TASC
TABLE
_
3
Comparison
of appearance
measures
Failure rate per individual Total number of failures Total number of individuals Mean failure rate per individual Standard deviation Rate comparison by Group 1 us group Group 2 us group Group 1 us group
t-test 2 3 3
Failure rate per appointment Total number of failures Total number of required appearances Mean failure rate per appointment Standard deviation Rate comparison by Group 1 us group Group 2 us group Group 1 US group
t-test 2 3 3
38 50 0.76 0.96
t = 1.96 t = 1.06 t = 3.12
2 clients)
21 50
Group 3 (felons)
14 50
0.42 0.17
0.28 0.52
d.f. = 98 d.f. = 98 d.f. = 98
p < 0.025 N.S. p < 0.005
38
21
14
209
253
185
0.182 0.58
t = 2.21 t = 0.211 t = 2.26
0.083 0.37
d.f. = 98 d.f. = 98 d.f. = 98
0.076 0.30
p < 0.025 N.S. p < 0.025
317
pear on at least one occasion; 22.0 per cent of the group failed to appear on two or more occasions. The members of group 2 (TASC clients) were very similar to the members of group 3 (felons) with respect to frequency of failures. Upon application of the Chi-square test, these differences achieve statistical significance @ < 0.05). The difference between the two addict groups (1 and 2) is further evidenced by the statistical analysis of failure rate per individual. As shown in Table 3, there is a statistically significant difference (p < 0.025) between the mean failure rate per person of the members of group 1 (non-TASC addicts) and members of group 2 (TASC clients). The suggestion that TASC involvement is associated with an improvement of the appearance performance of addicts is strengthened by the finding that there is no statistically significant difference between the mean rates of group 2 (TASC clients) and group 3 (felons). However, the difference in mean failure rate per individual between the non-TASC addicts and the felons is statistically significant (p < 0.005). Computation of failure rate per scheduled appearance purports to examine failure rates while eliminating the effects of the differing number of required appearances among the groups. Analysis by this technique supports the indication of the prior two computations. Again, group 1 has the highest rate of failure; group 2 shows a rate comparable to the non-addict group 3. The difference in mean failure rate per appointment is statistically significant between group 1 and group 2 (p < 0.025) as well as with group 3 (p < 0.025). There is no significant difference between TASC clients and felons. Measured in terms of rearrest, the pretrial performances of the three groups differ substantially. The rearrest data are divided into three categories: drug arrests, non-drug arrests, and total arrests. These data are examined in frequency distribution format (none/one or more arrests) and group rearrest rates per month of exposure are compared (Table 4). Both analyses indicate a similar pattern. Little difference is revealed with regard to rearrests among the addict groups. The non-addict group, however, displays a substantially reduced rearrest pattern. Notwithstanding the fact that the group 3 subjects (felons) show the lowest overall rearrest performance, a basic equivalency of the three groups is obscured by the impact of drug arrests of groups 1 and 2; the non-drug rearrest measures are comparable among the three groups. The differences in rearrest frequencies and rates between the addict groups and non-addict groups achieved a level of statistical significance. Statistical significance, however, was not achieved when rearrest measures between the two addict groups were compared.
Discussion Several methodological shortcomings limit the conclusions to be drawn and require mention at the outset of this discussion section. First, the study groups were not particularly large, and the serial selection nature of the two
318 TABLE
4
Rearrest
data
A. Frequency
comparison
Drug arrests None One or more Chi-square
analysis
precluded
Total arrests None One or more n = 150,
d.f. = 2, x2 = 14.75,~
B. Rearrest
rates per month
addicts)
Group 2 (TASC clients)
Group 3 (felons)
n
%
n
%
n
%
32 18
64 36
33 17
66 34
50 0
100 0
due to empty
Non-drug arrests None One or more
n = 150, d.f. = 2, x2 = 2.22,
Group 1 (non-TASC
40 10
80 20
36 14
72 28
42 8
84 16
27 23
54 46
25 25
50 50
42 8
84 16
1 us
Group
N.S.
< 0.001
of exposure Group group
Drug arrests Non-drug Total
*d.f.
arrests
arrests
cell
comparisons* 1 us 2
t = 0.481,
Group group
N.S.
3
group _
(no group
3 drug arrests) N.S.
t = 0.871,
N.S.
t = 0.448,
t = 0.176,
N.S.
t = 3.33, p < 0.01
2 us 3
t = 1.35,
N.S.
t = 3.04, p < 0.01
= 98.
addict groups was not as desirable as the classic random technique. Second, the available data concerning the non-TASC addicts do not adequately address the important dynamics of the group members’ failure to become involved with the TASC process. Without discerning the nature of this phenomenon, the assumption of comparability between the TASC clients and nonTASC addicts is arguable. Finally, the study lacks cleanness. Although the authors were familiar with area treatment program rosters, confidentiality barred the rigorous examination of treatment records to determine unknown treatment participation by the non-TASC addicts. Also, there is a lack of uniformity concerning the circumstances under which the TASC clients became involved in TASC supervision during the pretrial release period. Some were ordered to participate as a formal condition of personal recognizance
319
bond release, but most subjects participated as a result of judicial exhortation, not ordering. Nonetheless, given these practical research constraints, the study data show that unsupervised, untreated addicts do perform substantially more poorly as compared to non-addict defendants with respect to required court appearances while on bond. However, involvement in supervised drug abuse treatment is associated with improved court appearance performance of addict defendants, resulting in the achievement of parity with non-addict defendants. With regard to rearrest, the data display virtually no difference between the two addict groups. Both addict groups were rearrested at a high percentage. The non-addict subjects showed much lower overall rearrest figures. Comparison with non-drug rearrest figures, however, failed to demonstrate significant differences between the addict groups and the non-addict group. The data thus lend support to the proponents of TASC-type addict diversion programs. Although the results do not provide a definitive answer to the legal question of whether or not TASC bail conditions represent “the least restrictive conditions necessary to guarantee future court appearance” for addict defendants, they do provide support for that position by furnishing highly relevant empirically based information heretofore unavailable. Future investigators should refine the group selection techniques and expand the size and geographic scope of sampling to test the significance of these findings; the external validity of this study is limited by the size and geographic restrictions of the sample design.
References 1 TASC Legal Analysis, Special Action Office for Drug Abuse Prevention, Washington, D.C., 1973, I, iii. 2 P. B. Wice, Bail and Its Reform: A National Survey, National Institute of Law Enforcement and Criminal Justice, Law Enforcement Assistance Administration, U.S. Department of Justice, Washington, D.C., 1973. 3 Legal Issues and Characteristics of Pretrial Intervention Programs, National Pretrial Intervention Service Center of the American Bar Association Commission on Correctional Facilities and Services, Washington, D.C., 1974. 4 H. H. Greene, Proceedings of the First National Conference on Treatment Alternatives to Street Crime, Special Action Office for Drug Abuse Prevention, Washington, D.C., 1973, pp. 1 - 16. 5 M. P. Pines, Answer to the problem of bail: A proposal in need of empirical confirmation. Columbia J. Law Sot. Probl., 9 (1973) 394 - 441. 6 D. L. Skoler, Protecting the rights of defendants in pretrial intervention programs. Criminal Law Bull., 10 (1974) 473 - 492.