International Journal of the Addictions
ISSN: 0020-773X (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/isum19
The Effect of Prior Treatment on Treatment Success Eduardo N. Siguel & William H. Spillane To cite this article: Eduardo N. Siguel & William H. Spillane (1978) The Effect of Prior Treatment on Treatment Success, International Journal of the Addictions, 13:5, 797-805, DOI: 10.3109/10826087809039303 To link to this article: http://dx.doi.org/10.3109/10826087809039303
Published online: 03 Jul 2009.
Submit your article to this journal
Article views: 1
View related articles
Citing articles: 3 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=isum19 Download by: [RMIT University Library]
Date: 30 March 2016, At: 11:16
The International Journal of the Addictions, 13(5), 797-805, 1978
Downloaded by [RMIT University Library] at 11:16 30 March 2016
The Effect of Prior Treatment on Treatment Success* Eduardo N. Siguel, Ph.D. William H. Spillane, Ph.D. Division of Scientific and Program Information National Institute on Drug Abuse (NIDA) Rockville, Maryland 20652
Abstract
The characteristics of clients discharged from federally funded treatment facilities reporting to CODAP are analyzed in terms of several variables. These are the reasons for discharge (treatment completed, transfer or referral, or treatment not completed), the primary drug of abuse at admission (opiates and nonopiates), and the number of prior treatment experiences. Two independent replications using data for two consecutive quarters produced the same results, finding that the likelihood of completing treatment decreases as the number of prior treatment experiences increases. This relationship was found to apply to both opiate and nonopiate users. Although large differences exist in completion rates between opiate and nonopiate users without prior treatment experience, only small differences exist between opiate and non*This article was written by Drs. Siguel and Spillane in their private capacity. N o official support or endorsement by NIDA is intended or should be inferred. 797 Copyright @ 1978 by Marcel Dekker, Inc. All Rights Reserved. Neither this work nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher.
SIGUEL AND SPILLANE
79s
opiate users who have had prior treatment. The largest difference in treatment outcome occurs between those with no prior treatment and those who have been treated before. Having one or more prior treatment experiences does not seem to have a major impact on the probability of completing treatment.
Downloaded by [RMIT University Library] at 11:16 30 March 2016
INTRO D U CTlO N It is well known by clinicians (associated with treatment) that clients in treatment for drug abuse often drop out of treatment programs, to be subsequently readmitted. However, little statistical evidence is available concerning the ways in which prior treatment experiences influence the likelihood of an individual’s completing treatment. Staff experience seems to have proposed t w o conflicting hypotheses. The first of these is that the probability of successfully completing treatment increases each time a client is readmitted to a treatment program. This hypothesis argues that individuals with prior treatment experiences reenter a program with a greater willingness to complete treatment. The alternative hypothesis proffers the opposite view: the probability of success decreases with an increase in the number of prior treatment experiences. The number of prior treatment experiences is seen as an indicator of the severity of the drug abuse problem of the individual. According to this viewpoint, individuals with more prior treatment experiences must have a more severe problem and therefore should be more difficult to treat. The resolution of this conflict has significant implications for the management of drug abuse programs. If the former hypothesis is true, then treatment facilities should pursue an aggressive policy of encouraging clients to return t o treatment even if they have been treated unsuccessfully in the past. If, however, the latter hypothesis is valid, maximum rehabilitation efforts should be concentrated on the first treatment experience, and treatment should be modified to accommodate those with prior experiences. The first hypothesis, that of probability increase, was fostered by the observation that individuals who were readmitted to treatment on several occasions eventually did complete treatment. The fallacy in this type of reasoning, in the absence of statistical evidence, is that the two events occurring simultaneously does not necessarily preclude any definite relationship between them. If the number of readmissions does influence the completion of treatment, this may not be an increase in probability, but rather a decrease or perhaps some other pattern. In opposition to these three possibilities, a study by the Engineer Studies
Downloaded by [RMIT University Library] at 11:16 30 March 2016
EFFECT O F PRIOR TREATMENT O N TREATMENT SUCCESS
799
Group (1974) found that the probability of treatment success does not depend on the number of prior treatment experiences. However, it is difficult to form generalizations using this study as a basis, as it is curtailed by a small sample size and contains limitations in the measures of prior treatment and reasons for discharge from a program. By contrast, the recent White Paper on Drug Abuse (1975) suggests that the probability of treatment success increases as the number of prior treatment experiences increases. The White Paper maintains that “Experience shows that individual addicts who return to treatment exhibit more progress the second time; more again the third, and so on.” However, this statement is based on personal observations of the treatment staff and is not a result of statistical analysis. Recently, adequate data have been made available for the first time through The Client Oriented Data Acquisition process (CODAP) which collects data on all clients admitted to and discharged from federally supported drug treatment programs. Data from this collection procedure for the first two quarters of 1975 have been analyzed and the results are presented in this paper. For an extensive description of CODAP, the reader is referred to Siguel and Spillane (1975).
METHOD CODAP data for all clients discharged during the first two quarters of 1975 were analyzed with respect to reasons for discharge, primary drug of abuse at admission, and number of prior treatment experiences. 1. Completed treatment: Client completed treatment successfully 2. Transfer/referral: Client transferred or referred to another treatment facility 3 . Treatment not completed: Client left before completing treatment or was discharged due to noncompliance with program rules
Clients who were incarcerated or died were not considered. The percent of clients who completed treatment or were transferred/referred was calculated for both the first and second calendar quarters of 1975 on the basis of the three classifications described above. These percentages were further subdivided into categories in relation to the number of prior treatment experiences and the primary drug of abuse, which was grouped according to opiates and nonopiates. These categories were derived from the classification of drug types on the CODAP admission report. Opiates
SIGUEL A N D SPILLANE
Downloaded by [RMIT University Library] at 11:16 30 March 2016
800
include heroin, illegal methadone, and other opiates and synthetics with morphine-like effects, while nonopiates are all other drugs. Admission and discharge reports are completed for each client upon admission to and discharge from a treatment facility (also referred to as a clinic). As a result of errors in data collection, not all admission and discharge reports could be matched. However, the data presented in this study are based only on those reports which could be matched. More reports were matched during the second quarter than during the first quarter, because more clients were discharged during the second quarter and data collection procedures improved. The matched reports can be considered a random sample of the total number of possible discharges because the errors that prevent the admission and discharge reports from being matched occur randomly (such as keypunching errors in the data items used to match reports). The data for the two quarters were analyzed to determine whether the results are consistent or a statistical artifact, and both sets of data can be considered independent replications of the same study.
RESULTS There are two important aspects of the CODAP data which should be discussed. The first is the large number of cases that have been used in preparing these tables. Over 26,000 cases were utilized in the first quarter and nearly 40,000 in the second. The similarity in the percentages for each quarter is also significant. When dealing with a large’ sample size, statistical techniques used to test differences are likely to show highly significant results. Even when the difference between two proportions (or rates) is as low as 0.01 (lx),that small difference will be detected as a statistically significant difference (i.e., a difference not due to chance) when enough cases are considered. However, a statistically significant difference may not be large enough to be important for policy making or be constant or consistent over time. If two separate replications (i.e., independent sets of data) produce the same results, the validity of the findings will be reinforced. Table 1 shows that for both the opiate and nonopiate categories there is a noticeable drop in the percent of completions between those clients with no prior treatment experiences and those with one or more prior experiences. It should be noted that the “Treatment not completed” group is not inluded in this table. The table also shows no major change in percent of completions among clients with one, two, three. and over three prior treatment experiences, suggesting that the
-
0 1 2 3 24 Total
16,321
Total
17.1
23,263
25.4
23.0 25.8 27.3 28.8 27.9
%
% 18.3 16.2 17.2 15.4 16.6
Transfer/ referral,
26.7
Treatment completed,
Opiate
17.5
21.6 27.8 30.6 31.5 35.6
%
% 18.9 16.2 15.9 18.1 15.7
Transfer/ referral,
Treatment completed,
9,178 6,602 3,507 1,859 2,117
Total discharged 2nd quarter 1975, number
6,587 4,767 2,333 1,229 1,426
Total discharged 1st quarter 1975, number
0 1 2 3 24
Number of prior treatment experiences
Opiate
16,835
12,160 2,871 937 386 48 1
Total discharged 2nd quarter 1975, number
27.0
10,280
27.2
29.9 21.4 20.1 14.0 16.0
%
Treatment completed,
Nonopiate
30.4 19.3 16.0 18.5 13.5
%
Treatment completed,
7,467 1,743 562 233 275
Total discharged 1st quarter 1975, number
Nonopiate
20.2
18.2 22.5 27.6 31.9 31.4
%
referral,
Transfer /
21.2
18.8 26.6 28.8 21.0 38.2
%
Transfer/ referral,
Reason for Discharge from Treatment by Number of Prior Treatment Experiences for Opiate and Nonopiate Users
Table 1
Downloaded by [RMIT University Library] at 11:16 30 March 2016
SIGUEL AND SPILLANE
Downloaded by [RMIT University Library] at 11:16 30 March 2016
XO:!
classifications of one or more prior experiences could be combined. Pursuant to this, the totals for one to more than four prior treatment experiences are summed for each opiate and nonopiate group in each time period. The percentage of completions and transfer/referral is then recalculated for the different opiate and nonopiate groups on the basis of the combined categories. Table 2 shows the percent of individuals who completed treatment in relation to the drug of abuse and the new prior treatment classifications. It was found that the percent of individuals completing treatment with one or more prior experiences was significantly lower than those with no prior experience in both drug categories within both replications. A significant difference was also found between opiates and nonopiates in both prior treatment classifications in both replications. The exception to this was in the first quarter for the one or more prior experiences classification. It is also interesting to note that while there are large differences in the percentages of completion between those clients in the two drug categories with no prior treatment experience, the differences are minor in both quarters for those with one or more prior experiences. Table 3 reports the frequencies and percentages of clients in the opiate and nonopiate categories as they are distributed over the number of prior treatment experiences. It is seen that 40"/;, of opiate clients versus 72y0 of Table 2 Percent of Clients W h o Complete Treatment According to Whether or Not They Had Prior Treatment Experience" ~~
Primary drug problem at admission ~
Opiate Number of prior treatment experiences
Total discharged, number
~~~~
Nonopiate
Completed treatment,
%
Total discharged, number
Completed treatment,
%
Data for the period January 1 to March 31, 1975 6,587 9,734
18.9 0.9 16.3 i 0.7
7,467 2,813
30.4 & 1.0 18.0 i 1.4
Data for the period April 1 to June 30, 1975 0 21 a
9,178 14,085
*
18.3 0.8 16.4 5 0.6
12,160 4,675
29.9 20.0
* 0.8 * 1.1
The Iindicates the range of the 95% confidence interval. Example: 30.4 :t 1.0 indicates that the 95% confidence interval is (30.4 - 1.0, 30.4 4 1.0) = (29.4, 31.4).
803
EFFECT OF PRIOR TREATMENT ON TREATMENT SUCCESS
Table 3 Percent in Prior Treatment Categories for Opiate and Nonopiate Users Number of prior treatment experiences
Total discharged Opiate
%
Number
Nonopiate
%
Number
72.6 17 5.5 2.3 2.7 100
7,467 1,743 562 233 275 10,280
72.2 17.1 5.6 2.3 2.9 100
12,160 2,871 937 386 48 1 16,835
Downloaded by [RMIT University Library] at 11:16 30 March 2016
Period January 1 to March 31, 1975 0 1 2 3 24 Total
40.4 29.1 14.3 7.5 8.7 100
6,587 4,746 2,333 1,229 1,426 16,321
Period April 1 to June 30, 1975
0 1 2 3 24 Total
39.5 28.4 15.1 8.0 9.1 100
9,178 6,602 3,507 1,859 2,117 23,263
nonopiates had no prior treatment experience. The reverse situation is seen if the categories of two, three, and four or more prior treatment experiences are combined. The combination of these categories contains about 30% of the opiate-using clients versus less than 11% of the nonopiate-using clients.
CONCLUSION A N D DISCUSSION This paper shows that using the current interpretations of the discharge categories of successful completion and transfer/referral, the percentage of individuals completing treatment successfully is greater for clients with no prior treatment experiences than for those who have one or more. This result is found to be significant in both replications and in both opiate and nonopiate classifications of drug abuse, providing evidence of a decrease of the probability of success with increased number of treatment experiences. It seems that the probability of success as it is distributed over more than one prior treatment experience does not increase, but further investigation in this area is necessary.
Downloaded by [RMIT University Library] at 11:16 30 March 2016
so.+
SIGUEL AND SPILLANE
Several factors must be considered in order to interpret this data. It has been suggested (Rohrs et al., 1974) that many individuals who abandon treatment do not return to drug abuse. If this were the case, then noncompletion of treatment should not be regarded as a failure. There is a belief among some treatment specialists that because the process of drug abuse treatment is a complex one, many drug users require several attempts at treatment before they finally decide to give up drugs. Therefore, the fact that an individual is seen several times at a treatment facility (he drops out and comes back again) should not be seen as a discouraging sign or proof that he (she) cannot be treated. Even individuals who have had several prior treatment experiences have completed treatment successfully. Therefore, the fact that an individual has dropped out of treatment before does not mean that he (or she) will drop out again. Furthermore, for opiate users, there are no clear differences in the percent of successful completion rates between those with one prior treatment experience and those with more than one prior treatment experience. Additional research is suggested by looking at TBble 1, where the probability of transfer1 referral increases with the increase of prior treatment experiences. It is possible that individuals with prior treatment experiences are often found more difficult to treat with the approach supplied in a given clinic and are therefore referred to another facility where it is felt treatment could be more effective. One of the critical issues in evaluation of treatment outcome is the understanding that a person’s presence in a program is equivalent to his partaking in a treatment process which must be suited to both him and to the treatment staff. The client’s strengths and weaknesses must be taken into account, as well as the skills and limitations of the treatment staff and the expectations of the community. The results discussed in this paper suggest that staff should be trained to utilize information about an individual’s prior treatment experiences in order to better plan his current treatment. This type of information may help to identify the treatment modality and/or environment which is most likely to be successful during the current admission. It is probable that a greater proportion of those clients transferred to other treatment programs are likely to complete treatment than those who are first admissions. This hypothesis is to be tested in a further study. Therefore, the eventual rate of successful completion (after consideration for transfers) of clients in treatment is probably greater than indicated by the figures presented in this report. The results represented in this paper imply a striking difference between opiate and nonopiate users. Opiate users have a smaller probability of successful
EFFECT OF PRIOR TREATMENT O N TREATMENT SUCCESS
805
Downloaded by [RMIT University Library] at 11:16 30 March 2016
completion of treatment and have more prior treatment experiences than nonopiate users. Both findings combine to produce a substantially lower successful completion rate for opiate users than non-opiate users. Additional research is presently needed to identify other variables that are associated with treatment outcome. It is necessary to determine whether these results are consistent across all race and sex combinations, and to evaluate the effect of modality and environment on treatment outcome. REFERENCES ENGINEERS STUDY GROUP; THE DEPARTMENT OF THE ARMY. Bringing a Systems Approach to Drug Abuse Treatment and Rehabilitation Management-An Anthology (Produced under Contract NIMH-RA-MH-73-7). September 30, 1974. ROHRS, C.C., GOLDSMITH, B., and DENSEN-GERBER, J. Drug-free treatment farlerres: A preliminary report. Clin. Toxicol. 7(3): 279-280, 1974. SIGUEL, E., and SPILLANE, W. The Client Oriented Data Acquisition Process ( C O D A P ) . Washington, D.C.: National Institute on Drug Abuse, 1975. WHITE PAPER ON DRUG ABUSE. Domestic Council Drug Abuse Task Force, September 1975. For Sale by the Superintendent of Documents, U S . Government Printing Office, Washington, D.C. 20402, Price $1.55, Stock Number 000-010-8802714.
ISSN: 0020-773X (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/isum19
The Effect of Prior Treatment on Treatment Success Eduardo N. Siguel & William H. Spillane To cite this article: Eduardo N. Siguel & William H. Spillane (1978) The Effect of Prior Treatment on Treatment Success, International Journal of the Addictions, 13:5, 797-805, DOI: 10.3109/10826087809039303 To link to this article: http://dx.doi.org/10.3109/10826087809039303
Published online: 03 Jul 2009.
Submit your article to this journal
Article views: 1
View related articles
Citing articles: 3 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=isum19 Download by: [RMIT University Library]
Date: 30 March 2016, At: 11:16
The International Journal of the Addictions, 13(5), 797-805, 1978
Downloaded by [RMIT University Library] at 11:16 30 March 2016
The Effect of Prior Treatment on Treatment Success* Eduardo N. Siguel, Ph.D. William H. Spillane, Ph.D. Division of Scientific and Program Information National Institute on Drug Abuse (NIDA) Rockville, Maryland 20652
Abstract
The characteristics of clients discharged from federally funded treatment facilities reporting to CODAP are analyzed in terms of several variables. These are the reasons for discharge (treatment completed, transfer or referral, or treatment not completed), the primary drug of abuse at admission (opiates and nonopiates), and the number of prior treatment experiences. Two independent replications using data for two consecutive quarters produced the same results, finding that the likelihood of completing treatment decreases as the number of prior treatment experiences increases. This relationship was found to apply to both opiate and nonopiate users. Although large differences exist in completion rates between opiate and nonopiate users without prior treatment experience, only small differences exist between opiate and non*This article was written by Drs. Siguel and Spillane in their private capacity. N o official support or endorsement by NIDA is intended or should be inferred. 797 Copyright @ 1978 by Marcel Dekker, Inc. All Rights Reserved. Neither this work nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher.
SIGUEL AND SPILLANE
79s
opiate users who have had prior treatment. The largest difference in treatment outcome occurs between those with no prior treatment and those who have been treated before. Having one or more prior treatment experiences does not seem to have a major impact on the probability of completing treatment.
Downloaded by [RMIT University Library] at 11:16 30 March 2016
INTRO D U CTlO N It is well known by clinicians (associated with treatment) that clients in treatment for drug abuse often drop out of treatment programs, to be subsequently readmitted. However, little statistical evidence is available concerning the ways in which prior treatment experiences influence the likelihood of an individual’s completing treatment. Staff experience seems to have proposed t w o conflicting hypotheses. The first of these is that the probability of successfully completing treatment increases each time a client is readmitted to a treatment program. This hypothesis argues that individuals with prior treatment experiences reenter a program with a greater willingness to complete treatment. The alternative hypothesis proffers the opposite view: the probability of success decreases with an increase in the number of prior treatment experiences. The number of prior treatment experiences is seen as an indicator of the severity of the drug abuse problem of the individual. According to this viewpoint, individuals with more prior treatment experiences must have a more severe problem and therefore should be more difficult to treat. The resolution of this conflict has significant implications for the management of drug abuse programs. If the former hypothesis is true, then treatment facilities should pursue an aggressive policy of encouraging clients to return t o treatment even if they have been treated unsuccessfully in the past. If, however, the latter hypothesis is valid, maximum rehabilitation efforts should be concentrated on the first treatment experience, and treatment should be modified to accommodate those with prior experiences. The first hypothesis, that of probability increase, was fostered by the observation that individuals who were readmitted to treatment on several occasions eventually did complete treatment. The fallacy in this type of reasoning, in the absence of statistical evidence, is that the two events occurring simultaneously does not necessarily preclude any definite relationship between them. If the number of readmissions does influence the completion of treatment, this may not be an increase in probability, but rather a decrease or perhaps some other pattern. In opposition to these three possibilities, a study by the Engineer Studies
Downloaded by [RMIT University Library] at 11:16 30 March 2016
EFFECT O F PRIOR TREATMENT O N TREATMENT SUCCESS
799
Group (1974) found that the probability of treatment success does not depend on the number of prior treatment experiences. However, it is difficult to form generalizations using this study as a basis, as it is curtailed by a small sample size and contains limitations in the measures of prior treatment and reasons for discharge from a program. By contrast, the recent White Paper on Drug Abuse (1975) suggests that the probability of treatment success increases as the number of prior treatment experiences increases. The White Paper maintains that “Experience shows that individual addicts who return to treatment exhibit more progress the second time; more again the third, and so on.” However, this statement is based on personal observations of the treatment staff and is not a result of statistical analysis. Recently, adequate data have been made available for the first time through The Client Oriented Data Acquisition process (CODAP) which collects data on all clients admitted to and discharged from federally supported drug treatment programs. Data from this collection procedure for the first two quarters of 1975 have been analyzed and the results are presented in this paper. For an extensive description of CODAP, the reader is referred to Siguel and Spillane (1975).
METHOD CODAP data for all clients discharged during the first two quarters of 1975 were analyzed with respect to reasons for discharge, primary drug of abuse at admission, and number of prior treatment experiences. 1. Completed treatment: Client completed treatment successfully 2. Transfer/referral: Client transferred or referred to another treatment facility 3 . Treatment not completed: Client left before completing treatment or was discharged due to noncompliance with program rules
Clients who were incarcerated or died were not considered. The percent of clients who completed treatment or were transferred/referred was calculated for both the first and second calendar quarters of 1975 on the basis of the three classifications described above. These percentages were further subdivided into categories in relation to the number of prior treatment experiences and the primary drug of abuse, which was grouped according to opiates and nonopiates. These categories were derived from the classification of drug types on the CODAP admission report. Opiates
SIGUEL A N D SPILLANE
Downloaded by [RMIT University Library] at 11:16 30 March 2016
800
include heroin, illegal methadone, and other opiates and synthetics with morphine-like effects, while nonopiates are all other drugs. Admission and discharge reports are completed for each client upon admission to and discharge from a treatment facility (also referred to as a clinic). As a result of errors in data collection, not all admission and discharge reports could be matched. However, the data presented in this study are based only on those reports which could be matched. More reports were matched during the second quarter than during the first quarter, because more clients were discharged during the second quarter and data collection procedures improved. The matched reports can be considered a random sample of the total number of possible discharges because the errors that prevent the admission and discharge reports from being matched occur randomly (such as keypunching errors in the data items used to match reports). The data for the two quarters were analyzed to determine whether the results are consistent or a statistical artifact, and both sets of data can be considered independent replications of the same study.
RESULTS There are two important aspects of the CODAP data which should be discussed. The first is the large number of cases that have been used in preparing these tables. Over 26,000 cases were utilized in the first quarter and nearly 40,000 in the second. The similarity in the percentages for each quarter is also significant. When dealing with a large’ sample size, statistical techniques used to test differences are likely to show highly significant results. Even when the difference between two proportions (or rates) is as low as 0.01 (lx),that small difference will be detected as a statistically significant difference (i.e., a difference not due to chance) when enough cases are considered. However, a statistically significant difference may not be large enough to be important for policy making or be constant or consistent over time. If two separate replications (i.e., independent sets of data) produce the same results, the validity of the findings will be reinforced. Table 1 shows that for both the opiate and nonopiate categories there is a noticeable drop in the percent of completions between those clients with no prior treatment experiences and those with one or more prior experiences. It should be noted that the “Treatment not completed” group is not inluded in this table. The table also shows no major change in percent of completions among clients with one, two, three. and over three prior treatment experiences, suggesting that the
-
0 1 2 3 24 Total
16,321
Total
17.1
23,263
25.4
23.0 25.8 27.3 28.8 27.9
%
% 18.3 16.2 17.2 15.4 16.6
Transfer/ referral,
26.7
Treatment completed,
Opiate
17.5
21.6 27.8 30.6 31.5 35.6
%
% 18.9 16.2 15.9 18.1 15.7
Transfer/ referral,
Treatment completed,
9,178 6,602 3,507 1,859 2,117
Total discharged 2nd quarter 1975, number
6,587 4,767 2,333 1,229 1,426
Total discharged 1st quarter 1975, number
0 1 2 3 24
Number of prior treatment experiences
Opiate
16,835
12,160 2,871 937 386 48 1
Total discharged 2nd quarter 1975, number
27.0
10,280
27.2
29.9 21.4 20.1 14.0 16.0
%
Treatment completed,
Nonopiate
30.4 19.3 16.0 18.5 13.5
%
Treatment completed,
7,467 1,743 562 233 275
Total discharged 1st quarter 1975, number
Nonopiate
20.2
18.2 22.5 27.6 31.9 31.4
%
referral,
Transfer /
21.2
18.8 26.6 28.8 21.0 38.2
%
Transfer/ referral,
Reason for Discharge from Treatment by Number of Prior Treatment Experiences for Opiate and Nonopiate Users
Table 1
Downloaded by [RMIT University Library] at 11:16 30 March 2016
SIGUEL AND SPILLANE
Downloaded by [RMIT University Library] at 11:16 30 March 2016
XO:!
classifications of one or more prior experiences could be combined. Pursuant to this, the totals for one to more than four prior treatment experiences are summed for each opiate and nonopiate group in each time period. The percentage of completions and transfer/referral is then recalculated for the different opiate and nonopiate groups on the basis of the combined categories. Table 2 shows the percent of individuals who completed treatment in relation to the drug of abuse and the new prior treatment classifications. It was found that the percent of individuals completing treatment with one or more prior experiences was significantly lower than those with no prior experience in both drug categories within both replications. A significant difference was also found between opiates and nonopiates in both prior treatment classifications in both replications. The exception to this was in the first quarter for the one or more prior experiences classification. It is also interesting to note that while there are large differences in the percentages of completion between those clients in the two drug categories with no prior treatment experience, the differences are minor in both quarters for those with one or more prior experiences. Table 3 reports the frequencies and percentages of clients in the opiate and nonopiate categories as they are distributed over the number of prior treatment experiences. It is seen that 40"/;, of opiate clients versus 72y0 of Table 2 Percent of Clients W h o Complete Treatment According to Whether or Not They Had Prior Treatment Experience" ~~
Primary drug problem at admission ~
Opiate Number of prior treatment experiences
Total discharged, number
~~~~
Nonopiate
Completed treatment,
%
Total discharged, number
Completed treatment,
%
Data for the period January 1 to March 31, 1975 6,587 9,734
18.9 0.9 16.3 i 0.7
7,467 2,813
30.4 & 1.0 18.0 i 1.4
Data for the period April 1 to June 30, 1975 0 21 a
9,178 14,085
*
18.3 0.8 16.4 5 0.6
12,160 4,675
29.9 20.0
* 0.8 * 1.1
The Iindicates the range of the 95% confidence interval. Example: 30.4 :t 1.0 indicates that the 95% confidence interval is (30.4 - 1.0, 30.4 4 1.0) = (29.4, 31.4).
803
EFFECT OF PRIOR TREATMENT ON TREATMENT SUCCESS
Table 3 Percent in Prior Treatment Categories for Opiate and Nonopiate Users Number of prior treatment experiences
Total discharged Opiate
%
Number
Nonopiate
%
Number
72.6 17 5.5 2.3 2.7 100
7,467 1,743 562 233 275 10,280
72.2 17.1 5.6 2.3 2.9 100
12,160 2,871 937 386 48 1 16,835
Downloaded by [RMIT University Library] at 11:16 30 March 2016
Period January 1 to March 31, 1975 0 1 2 3 24 Total
40.4 29.1 14.3 7.5 8.7 100
6,587 4,746 2,333 1,229 1,426 16,321
Period April 1 to June 30, 1975
0 1 2 3 24 Total
39.5 28.4 15.1 8.0 9.1 100
9,178 6,602 3,507 1,859 2,117 23,263
nonopiates had no prior treatment experience. The reverse situation is seen if the categories of two, three, and four or more prior treatment experiences are combined. The combination of these categories contains about 30% of the opiate-using clients versus less than 11% of the nonopiate-using clients.
CONCLUSION A N D DISCUSSION This paper shows that using the current interpretations of the discharge categories of successful completion and transfer/referral, the percentage of individuals completing treatment successfully is greater for clients with no prior treatment experiences than for those who have one or more. This result is found to be significant in both replications and in both opiate and nonopiate classifications of drug abuse, providing evidence of a decrease of the probability of success with increased number of treatment experiences. It seems that the probability of success as it is distributed over more than one prior treatment experience does not increase, but further investigation in this area is necessary.
Downloaded by [RMIT University Library] at 11:16 30 March 2016
so.+
SIGUEL AND SPILLANE
Several factors must be considered in order to interpret this data. It has been suggested (Rohrs et al., 1974) that many individuals who abandon treatment do not return to drug abuse. If this were the case, then noncompletion of treatment should not be regarded as a failure. There is a belief among some treatment specialists that because the process of drug abuse treatment is a complex one, many drug users require several attempts at treatment before they finally decide to give up drugs. Therefore, the fact that an individual is seen several times at a treatment facility (he drops out and comes back again) should not be seen as a discouraging sign or proof that he (she) cannot be treated. Even individuals who have had several prior treatment experiences have completed treatment successfully. Therefore, the fact that an individual has dropped out of treatment before does not mean that he (or she) will drop out again. Furthermore, for opiate users, there are no clear differences in the percent of successful completion rates between those with one prior treatment experience and those with more than one prior treatment experience. Additional research is suggested by looking at TBble 1, where the probability of transfer1 referral increases with the increase of prior treatment experiences. It is possible that individuals with prior treatment experiences are often found more difficult to treat with the approach supplied in a given clinic and are therefore referred to another facility where it is felt treatment could be more effective. One of the critical issues in evaluation of treatment outcome is the understanding that a person’s presence in a program is equivalent to his partaking in a treatment process which must be suited to both him and to the treatment staff. The client’s strengths and weaknesses must be taken into account, as well as the skills and limitations of the treatment staff and the expectations of the community. The results discussed in this paper suggest that staff should be trained to utilize information about an individual’s prior treatment experiences in order to better plan his current treatment. This type of information may help to identify the treatment modality and/or environment which is most likely to be successful during the current admission. It is probable that a greater proportion of those clients transferred to other treatment programs are likely to complete treatment than those who are first admissions. This hypothesis is to be tested in a further study. Therefore, the eventual rate of successful completion (after consideration for transfers) of clients in treatment is probably greater than indicated by the figures presented in this report. The results represented in this paper imply a striking difference between opiate and nonopiate users. Opiate users have a smaller probability of successful
EFFECT OF PRIOR TREATMENT O N TREATMENT SUCCESS
805
Downloaded by [RMIT University Library] at 11:16 30 March 2016
completion of treatment and have more prior treatment experiences than nonopiate users. Both findings combine to produce a substantially lower successful completion rate for opiate users than non-opiate users. Additional research is presently needed to identify other variables that are associated with treatment outcome. It is necessary to determine whether these results are consistent across all race and sex combinations, and to evaluate the effect of modality and environment on treatment outcome. REFERENCES ENGINEERS STUDY GROUP; THE DEPARTMENT OF THE ARMY. Bringing a Systems Approach to Drug Abuse Treatment and Rehabilitation Management-An Anthology (Produced under Contract NIMH-RA-MH-73-7). September 30, 1974. ROHRS, C.C., GOLDSMITH, B., and DENSEN-GERBER, J. Drug-free treatment farlerres: A preliminary report. Clin. Toxicol. 7(3): 279-280, 1974. SIGUEL, E., and SPILLANE, W. The Client Oriented Data Acquisition Process ( C O D A P ) . Washington, D.C.: National Institute on Drug Abuse, 1975. WHITE PAPER ON DRUG ABUSE. Domestic Council Drug Abuse Task Force, September 1975. For Sale by the Superintendent of Documents, U S . Government Printing Office, Washington, D.C. 20402, Price $1.55, Stock Number 000-010-8802714.
