substance abuse treatment effectiveness of publicly ...

SUBSTANCE ABUSE TREATMENT EFFECTIVENESS OF PUBLICLY FUNDED CLIENTS IN TENNESSEE Satish Kedia, Ph.D. and Stephanie W. Perry, M.D. The Tennessee Outcomes for Alcohol and Drug Services (TOADS) in collaboration with the Bureau of Alcohol and Drug Abuse Services at the Tennessee Department of Health evaluated the effectiveness of publicly funded substance abuse treatment programs in Tennessee by collecting and analyzing data from clients treated between 1998 and 2000. Using a structured questionnaire, TOADS staff conducted telephone interviews with clients 6 months after their admission to treatment facilities. The sample populations for these follow-up interviews ranged from 1,150 to 1,350 clients over the 3 years, and each year, post-treatment abstinence rates were around 60%, which suggests that treatment in Tennessee has been successful in reducing substance abuse. In addition, the follow-up interview data suggest that treatment also helped drastically reduce both unemployment and arrests among clients. These findings in Tennessee are comparable to treatment outcomes in other states. In addition to the positive effects that treatment has on clients, treatment is also cost-effective for state budgets since treatment reduces many of the burdens substance abuse places on the criminal justice system, the healthcare system, and other state-supported services. ( Natl Med Assoc. 2003;95:270-277.)

Key words: substance abuse * treatment effectiveness * Tennessee

INTRODUCTION In a time of increasing accountability for publicly funded social service programs, it is critical to assess the effectiveness of substance abuse treatment. Recent studies in various © 2003. Tennessee Outcomes for Alcohol and Drug Services, The University of Memphis and Bureau of Alcohol and Drug Abuse Services, Tennessee Department of Health. Dr. Satish Kedia is the Director of the Tennessee Outcomes for Alcohol and Drug Services and Assistant Professor of Medical Anthropology at The University of Memphis, and Dr. Stephanie W. Perry is the Assistant Commissioner at the Bureau of Alcohol and Drug Abuse Services, Tennessee Department of Health. Corresponding Author Satish Kedia, 316 Manning Hall, The University of Memphis, Memphis, TN 38152, USA, phone: 901-6781433, fax: 901-678-0707, email: [email protected] JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

parts of the United States have indicated that treatment has been effective in reducing substance abuse.' In addition to reducing substance abuse, treatment can be effective in increasing employment and productivity, reducing illegal activity and arrests, and improving the mental health and living circumstances of clients and their relationships with others2-4. This article describes the effectiveness of substance abuse treatment in Tennessee from 1998 to 2000. In collaboration with the Bureau of Alcohol and Drug Abuse Services at the Tennessee Department of Health, the Tennessee Outcomes for Alcohol and Drug Services (TOADS) at The University of Memphis evaluates all alcohol and drug abuse treatment facilities funded by the Substance Abuse Prevention and Treatment (SAPT) Block Grant. The fedVOL. 95, NO. 4, APRIL 2003

270

SUBSTANCE ABUSE TREATMENT EFFECTIVENESS

Table 1. Study Population 1998-1999

Population characteristics Unduplicated client admissions Clients who consented to participate in follow-up study Clients who provided correct phone numbers Clients with whom interviews were completed Interview completion rate (%)

eral government distributes these grants to states for planning, implementing, and evaluating alcohol and drug abuse treatment and prevention programs.

MATERIALS AND METHODS Data collection For this longitudinal study, TOADS used a pre- and post-test design. The pre-test client data were collected during admission by treatment facility staff; the post-test information was collected by TOADS interviewers via telephone 6 months after admission. For the follow-up interviews, TOADS staff used a structured questionnaire that adheres to the specifications of the Government Performance and Results Act (GPRA) and incorporates the Interstate Core Data Items and Performance Indicators of the Treatment Outcomes and Performance Pilot Studies II (TOPPS II) Enhancement, which maintains national standards for alcohol and drug treatment outcome evaluation. This questionnaire covered several topics: the client's demographics and economic situation, characteristics of the client's treatment, changes in the client's substance use patterns, and information about the client's employment status, living situation, and arrests incurred.

Study population The TOADS interviewers were able to complete 6-month follow-up interviews with 1,258, 1,150, and 1,350 clients respectively during the 1998-1999, 1999-2000, and 2000-2001 fiscal years. The number of unduplicated clients admitted to the treatment facilities, clients who consented to participate in the follow-up study, and clients who had correct phone numbers 271

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

6,403 5,292 2,438 1,258 51.6

1999-2000 6,910 4,504 2,1 15 1 ,150 54.4

2000-2001 7,611 5,239 2,366 1 ,350 57.1

for the interviewers to be able to contact them are included in Table 1. Out of those clients who consented to participate in the follow-up study and provided a correct phone number, the average interview completion rate over the 3 years was about 54% (see Table 1). Around two thirds of clients each year were male. The proportion of clients who were White varied between 59.6% and 63.4% for the 3 years, but the percentage of African Americans varied between 30.9% and 38%. In any given year, less than 3% of clients were of other ethnic backgrounds. At least nine tenths of the clients each year were adults (i.e., at least 18 years of age). Education level had the most variation among clients across the 3 years. The majority (52.9% to 70.1%) of clients had either finished high school or had some high school education. The proportion of clients who did not study beyond middle school varied between 7.6% and 16.5%, while those who had attended college composed between 16.1 % and 21% of clients. Less than 0.5% of clients in any year had no schooling (see Table 2). With regard to treatment modality, the residential population composed roughly six tenths (61.4% and 60.5%) of the population in 1998-1999 and 1999-2000 and seven tenths (71.1 %) in 2000-2001. Clients treated in a halfway house composed about one tenth of clients, with a gradual increase over the 3 years from 8.4% the first year to 11.3% the third. Representation of outpatients decreased in the sample, going from 30.1 % the first year to 29.3% the second year, with a sharp drop to 17.6% the third year (see Table 3). The majority (54.2% to 66.5%) of clients participated in Alcoholics Anonymous or Narcotics AnonyVOL. 95, NO. 4, APRIL 2003


Table 2. Demographic Characteristics of Clients at the Time of Admission 1999-2000 1998-1999 (N = 1,150) (N = 1,258) Demographics Gender* Male Female Ethnicity* White African American Other No answer Age category* Youth (0-17 years) Adult (18+ years) No answer Education level* Middle school High school**

College No schooling Noanswer

%

%

2000-2001 (N = 1,350) %

67.1 32.9

62.1 37.9

66.7 33.3

62.7 35.7 1.1 0.5

63.4 30.9 1.2 4.5

59.6 38.0 2.4 0.0

8.2 90.2 1.6

6.4 93.6 0.0

6.7 93.3 0.0

16.5 61.8 20.6 0.2 0.9

16.5 52.9 16.1 0.3 14.2

7.6 70.1 21.0 0.0 1.3

*The distribution of the categories within each variable is statistically significant at the p < .05 level using the Chi-square test. **Have either completed high school or have some high school education. mous each year, whereas participation in aftercare programs ranged from 27.9% to 38.3% (see Table 3).

Limitations of the study Similar to other substance abuse studies, this study has several limitations. The study population was not homogeneous; rather, it represented clients of different ethnicity, gender, ages, and geographic regions who had different kinds of substance abuse problems and received treatment in one or more of the three modalities (residential, halfway house, or outpatient) for varying lengths of time. As a result, these analyses can indicate only general trends for treatment effectiveness. Since TOADS is required to generate facility-level outcome performance reports, the study population included all those clients who consented to participate in the study, rather than random sampling, in order to maximize the number of clients for each of the participating agencies. The admission data are collected by the treatJOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

Table 3. Treatment Features 1998-1999 1999-2000 2000-2001 (N = 1,258) (N = 1,150) (N = 1,350) % % Features % Treatment modality* 71.1 61.4 60.5 Residential 11.3 9.5 8.4 Halfway house 29.3 17.6 30.1 Outpatient 0.0 0.7 0.1 No answer Substance use* 63.3 61.3 57.0 Abstinence 36.7 38.7 43.0 Recidivism Participation in AA/NA* 66.5 64.2 54.2 Yes 33.5 35.8 45.8 No Participation in aftercare* 27.9 34.1 38.3 Yes 65.9 72.1 61.7 No *The distribution of the categories within each variable is statistically significant at the p < .05 level using the Chisquare test. VOL. 95, NO. 4, APRIL 2003

272


Variables Gender* Male

Table 4. Substance Abuse Treatment Effectiveness by Gender and Ethnicity 1998-1999 (N = 1,258) % 1999-2000 (N = 1,150) % 2000-2001 (N = 1,350) % Abstinence Recidivism Abstinence Recidivism Abstinence Recidivism (N = 717) (N = 541) (N = 705) (N = 445) (N = 495) (N = 855)

57.8 42.2 60.2 39.8 62.0 38.0 Female 55.2 44.8 63.1 36.9 65.9 34.1 Ethnicity* White 58.9 41.1 60.6 39.4 61.0 39.0** African American 54.3 45.7 65.1 34.9 32.2 67.8 Other 35.7 64.3 42.9 57.1 50.0 50.0 No Answer 66.7 33.3 50.0 50.0 0.0 0.0 *Percentages for abstinence and recidivism for each year will add up to 100% along the rows. * *The distribution of the categories within each variable is statistically significant at the p < .05 level using the Chi-square

test.

ment facilities, whereas the follow-up data are collected by the TOADS project. Furthermore, both the admission and the follow-up data are based on the clients' self-reports, which may lead to some bias in reporting treatment effectiveness. Despite these limitations, this longitudinal evaluation provides sufficient evidence regarding the overall substance abuse treatment effectiveness in the State of Tennessee.

Overview of treatment programs Using Addiction Severity Index (ASI) and American Society of Addiction Medicine (ASAM) criteria, each client was assessed at the time of admission to a treatment facility. Based on the client's reported substance abuse severity level, he/she was recommended broadly to one of the three primary levels of care: residential, halfway house, or outpatient treatment. Severely addicted clients were referred to residential rehabilitation treatment programs in which they received treatment in either a hospital or residential setting. In this context, they received at least five counseling contacts and five lectures/seminars per week. The next level of treatment was the halfway house, a structured residential treatment program in which clients had at least one counseling contact and one educational lecture/seminar per week. The third level of treatment was through out273


patient services, consisting of a wide range of nonresidential services that allowed clients to remain in their community and required no more than one outpatient service per day. Services at these modalities included any combination of individual, group, and/or family therapy or counseling, alcohol and drug education, skill training, and support group/aftercare participation. The overall goals of these treatment services were to help clients abstain or reduce alcohol or drug abuse, be more productive, reduce unlawful activities, develop better parenting skills and family functioning, improve psychological and physical health, and expand life management skills needed to build a satis-

fying lifestyle. RESULTS Abstinence Each year, the majority of clients (57% to 63.3%) reported being abstinent between admission and the 6-month follow-up interview (see Table 3). Additional analysis by gender reveals that, while a slightly higher percentage of males remained abstinent during 1998-1999, in the subsequent 2 years female clients fared better in achieving abstinence. Between the White and African American clients, in general, African American clients performed much better than their White counterparts during VOL. 95, NO. 4, APRIL 2003


Table 5. Substance Abuse Treatment Effectiveness Performance Indicators 1998-1999 (N = 1,258) % Follow-up Admission

1999-2000 (N = 1,150) % 2000-2001 (N = 1,350) % Follow-up Admission Follow-up Admission

Indicators Substance use* 26.8 61.7 22.4 60.4 21.5 57.4 Alcohol 8.4 43.7 3.7 41.9 46.4 8.3 Cocaine 6.4 31.3 4.0 35.0 28.8 4.8 Marijuana/hashish 7.4 1.3 1.9 6.9 0.3 6.8 Opiates/narcotics 0.1 5.9 0.4 6.8 1.3 4.5 Sedatives/hypnotics 2.1 0.0 2.2 0.1 0.1 0.6 Stimulants/amphetamines 0.1 0.3 0.1 0.9 0.1 0.0 Inhalants 0.3 0.0 1.2 0.2 1.4 0.0 Hallucinogens 1.9 1.9 0.9 1.7 0.2 0.6 Other Arrest record 13.0** 52.4 53.9 13.7** 53.5 19.2 Yes 87.0 47.6 86.3 46.1 46.5 80.8 No Employment situation 57.6** 18.0 56.1** 15.4 37.0 53.7* * Employed 32.8** 68.0 25.0** 24.0* * 60.0 41.0 Unemployed 9.6** 14.0 18.9** 24.6 22.0 22.3 Other*** Living arrangement 14.4 17.0 13.6 11.7 15.7 15.6 Alone 82.2** 73.8 86.0** 74.5 80.6** 69.8 Family & relatives 3.4** 9.2 11.9 2.3** 3.8** 14.5 Other living situations**** *Percentages will add up to more than 100% because some clients gave multiple responses. * *The chan e from the time of admission to the time of follow-up for this category is statistically significant at the p < .05 level using t e McNemar test. ***lncludes student, retired, and missing values. ****Includes foster care, homeless, anJ other kinds of living arrangements.

1999-2000 and 2001-2001, with White clients reporting higher abstinence rates in 1998-1999 (see Table 4). Alcohol, cocaine, and marijuana/hashish were the most frequently abused substances, both at the time of admission and at the 6-month follow-up. During the 3 years, the majority (57.4% to 61.7%) of clients were abusing alcohol upon admission, 41.9% to 46.4% were abusing cocaine, and 28.8% to 35.0% were abusing marijuana/ hashish. However, use of all three substances dramatically decreased by the time of the 6-month follow-up. Although 21.5% to 26.8% of clients were still using alcohol each year, usage rates had been cut in half. Even greater reductions occurred in cocaine and marijuana/hashish use, with only 3.7% to 8.4% of clients still using cocaine and 4.0% to 6.4% using marijuana/hashish (see Table 5). JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

The other classes of drugs used by more than 5% of clients at admission were opiates/narcotics and sedatives/hypnotics. Opiates/narcotics use at the time of admission hovered around 7% all 3 years and was reduced each year to less than 2% 6 months later. Sedatives/hypnotics use was 5.9% to 6.8% at admission and reduced to less than 0.5% in 1999-2000 and 20002001; however, the proportion of clients using sedatives/hypnotics in 1998-1999 increased from 1.3% to 4.5% between admission and follow-up. This increase in the use of sedatives/ hypnotics is the only increase in any class of substances over the 3 years. Stimulants/amphetamines were used by around 2% of clients at admission. Stimulants/amphetamines use decreased all 3 years, from 2% (less than 0.6% in 1998-1999) at admission to 0.1% or less at follow-up. All other classes of substances, such VOL. 95, NO. 4, APRIL 2003

274


as inhalants and hallucinogens, were used by less than 2% of the population in all 3 years. These rates indicate that treatment in Tennessee has been quite effective in reducing substance abuse and is comparable to the treatment outcomes in other states'. Encouragingly, research has suggested that abstinence shortly after treatment (within 3 to 6 months) is indicative of abstinence a year or more later5-7, suggesting that abstinence found in the short run should carry over to the long run.

Arrest record According to the California Drug and Alcohol Treatment Assessment (CALDATA) General Report, substance abuse treatment decreases criminal activity by developing new moral and ethical standards to replace those used to justify criminal activities in the past and by reducing the need to procure money to buy drugs or alcohol8. Treatment also reduces crimes committed while a person's judgment is impaired from substance use. The TOADS data support these findings. During each year of this study, the number of clients with arrest records dramatically dropped after treatment compared to arrest records before treatment. In all 3 years, close to 53% of clients had arrest records prior to admission; however, less than 20% in 1998-1999 and around 13% in the other 2 years were arrested between admission and the follow-up interviews (see Table 5).

Employment situation Clients' employment substantially increased

after treatment. In 1998-1999, full- and parttime employment among clients increased from 37% at admission to 53.7% at follow-up, while unemployment decreased from 41% to 24% in the same period. The change in employment was even more dramatic in 19992000, with unemployment decreasing from 60% to 25%. The most substantial decrease in unemployment was in 2000-2001, when it dropped from 68% to 32.8% of clients (see Table 5). A majority of the treatment programs have skills training as well as motivational ses275


sions to prepare clients to find employment. Increases in the employment situation of these clients definitely reflect the success of these programs and indicate that treatment helps clients become economically productive citizens of society.

Living arrangement Each year, more clients (80.6% to 86.0%) lived with family or relatives at the time of the follow-up than at the time of admission (69.8% to 74.5%), while the proportion of clients living in other situations or alone decreased (see Table 5). This shift in living arrangements suggests that clients were able to get their lives together and improve their relationships with their family members and relatives as a result of rehabilitation.

DISCUSSION It is important to note that the effectiveness of substance abuse treatment in Tennessee is comparable to the effectiveness of substance abuse treatment in other states. For example, in Iowa in 1999-2000, a random sample of 396 adult clients 6 months after discharge indicated declines in the primary use of alcohol (from 58.5% of clients at admission to 43.9% at follow-up), marijuana/hashish (from 19.7% to

7.2%), cocaine/crack (from 10.8%

to

3.1%),

and methamphetamines (from 9.6% to 4.1%)9. In Illinois, among a sample of 1,890 adult clients recruited from 40 treatment programs from 1998 to 2000, substance use in the 30 days prior to the 6-month follow-up interview decreased substantially: alcohol from 59% of all clients to 30%, cocaine from 37% to 6%, marijuana from 30% to 6%, and heroin from 24% to 6%10. In a California study consisting of a random sample of 3,000 clients, a substantial decrease in substance use was found when comparing pre-treatment data 12 months before admission to follow-up interview data 9-24 months after discharge, with an average of 15 months. The percentage of clients using alcohol declined from 70.9% prior to admission to 49.9% afterwards, crack from 29.1% to 18.5%, VOL. 95, NO. 4, APRIL 2003


cocaine from 40.4% to 21.6%, heroin from 42.4% to 32.7%, and amphetamines from 21.8% to 11.2%8. Likewise, in a 1997 study conducted in Hawaii, 65.8% of 1,041 adult clients had not used any substances during the 30 days prior to the follow-up interview". Some differences emerge in the type of information used when comparing information about Tennessee to that of other states. This Tennessee study included approximately 54% of all the clients who consented to participate in the follow-up study and provided correct phone numbers, whereas in several other states, much smaller samples (2% to 5%) were used. In addition, different states used different lengths of time for the follow-up interviews, making comparisons of results more difficult. Compared to studies in some states, Tennessee used a more stringent criteriow for determining treatment effectiveness: complete abstinence during the 6 months after the date of admission. Abstinence rates since treatment tended to be lower than abstinence rates during the 30 days prior to follow-up. For example, in 2000-2001, the TOADS data indicated that while 63.3% of clients abstained since admission, 82.1% abstained during the 30 days prior to the follow-up interview12. Some clients temporarily relapsed after treatment for a variety of reasons (including pressure from friends, stress, spousal issues, emotional and mental health, death of a loved one, and by choice) but were able to stop using the substances again. Therefore, abstinence rate during the 30 days prior to follow-up is generally higher than the overall abstinence rate during the 6 months after admission. The substantial decrease in arrest records or criminal activity after treatment found in this study of Tennessee clients is similar to other statewide studies, including California8, Iowa9, and Hawaii". Another similarity between the findings of this Tennessee study and findings from studies conducted in other states was the improved level of post-treatment employment. Increases in full- and part-time employment accompanied by decreases in unemployment JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

were found in Iowa9 and Illinois'0. Like Tennessee, Iowa's evaluation found an increase of clients living with family, including one's significant other, children, or both, although there was a decrease of clients living with their parents9. It is important to remember that the description of outcomes in this paper and for other states reflects a general trend for clients of different ethnicity, gender, age, and geographic region collapsed together and does not reveal specific successes or shortfalls within the diverse populations of Tennessee and other states. In 1998, substance abuse-related spending by the State of Tennessee was reported to be $92 million not including substantial federal contributions. The largest portion (37%) of that amount was spent in the criminal justice system. Almost 10 years ago, it was estimated that 1 year of incarceration cost society $39,600 and 1 year of untreated addiction cost an estimated $43,00013. Furthermore, because healthcare systems treat not only the substance abuse condition but also other conditions that are exacerbated by substance abuse (e.g., injuries caused by substance-related car crashes and other accidents), it is estimated that states spend about 25% of their healthcare budgets treating conditions related to substance abuse'4. Tennessee spends approximately $2,700 per year on the treatment of each client. According to the data collected by TOADS, a post-treatment abstinence rate of close to 60% is estimated for substance abuse clients, reducing substance abuse, decreasing arrest records, and increasing employment and improving living situations. Considering the high costs of untreated substance abuse, treatment is far less expensive than the likely consequences. In addition, the clients' productivity and employment are increased with treatment, expanding the tax base, saving employers money by having more productive employees, and reducing the state's public assistance rolls. Given the climate of fiscal responsibility, substance abuse treatment is cost-effective and worth the investment. VOL. 95, NO. 4, APRIL 2003

276


ACKNOWLEDGMENTS The authors would like to thank the Bureau of Alcohol and Drug Abuse Services at the Tennessee Department of Health for funding this project and for providing admission data collected by treatment facilities in Tennessee. The interviewers at the Tennessee Outcomes for Alcohol and Drug Services, The University of Memphis, deserve special thanks for their tenacity in conducting follow-up telephone interviews. We would like to thank statistician Ling Zhu for her assistance with the analyses on this research project. We also appreciate Margaret Kedia, Glenn Sanders, Linda Sadler, and the anonymous reviewers for their comments on earlier drafts of this manuscript.

REFERENCES 1. Schneider Institute for Health Policy, Brandies University. Substance abuse: the nation's number one health problem. Princeton, NJ: Robert Wood Johnson Foundation; 2001. 2. Appel PW, Joseph H, Kott A, Nottingham W, Tasiny E, Habel E. Selected in-treatment outcomes of long-term methadone maintenance treatment patients in NewYork State. Mt Sinai JMed. 2001;68(1):55-61. 3. Maxwell JC, Spence RT, Nelson R, Madry KG, Kavinsky J, LeekaJ. Substance abuse treatment in Texas, 1975-1991: as reported on CODAP, the Client Oriented Data Acquisition Process, executive summary; 2001. Retrieved October 25, 2001, from www.tcada.state.tx.us/research/treatment. 4. Mueller MD, Wyman JR. Study sheds new light on the state of drug abuse treatment nationwide. NIDA Notes Online. 1997;12(5). Retrieved March 29, 2002, from www.nida.nih.gov/

NIDA_Notes/NNVoll2N5/Study.html. 5. Latimere WW, Newcomb M, Winters KC, Stinchfield RD. Adolescent substance abuse treatment outcome: the role of substance abuse problem severity, psychosocial, and treatment factors. J Consult Clin PsychoL 2000;68 (4) :684- 696. 6. Hu T, Hunkeler EM, Weisner C, Li E, Grayson DK, WestphalJ, McLellan AT. Treatment participation and outcome among problem drinkers in a managed care alcohol outpatient treatment program. J Mental Health Admin. 1997;24:23-34. 7. Powell BJ, LandonJF, Cantrell PJ, Penick EC, Nickel EJ, Liskow BI, Coddington TM, Campbell JL, Dale TM, Vance MD, Rice AS. Prediction of drinking outcomes for male alcoholics after 10 to 14 years. Alcohol Clin Exp Res. 1998;22:559-566. 8. California Department of Alcohol and Drug Programs. Evaluating recovery services: the California Drug and Alcohol Treatment Assessment (CALDATA) General Report. Sacra-

277


mento, CA: California Department of Alcohol and Drug Programs; 1994. 9. Iowa Consortium for Substance Abuse Research and Evaluation. Outcomes monitoring system Iowa project: year two report. Iowa City, IA: Iowa Consortium for Substance Abuse Research and Evaluation; 2000. 10. Illinois Department of Human Services. The effectiveness of substance abuse treatment in Illinois: results of the Illinois statewide treatment outcomes project. Chicago, IL: Office of Alcoholism and Substance Abuse; 2001. 11. Hawaii Department of Human Services. Public health resources: alcohol and drug abuse. Kapolei, HI: Alcohol and Drug Abuse Division; 2001. Retrieved March 28, 2002, from

www.state.hi.us/health/resource/adad/adtrtwo.htm. 12. Kedia, S. Substance abuse treatment effectiveness in Tennessee: 2000-2001 statewide treatment outcome evaluation. Memphis, TN: Tennessee Outcomes for Alcohol and Drug Services; 2002. 13. National Institute on Drug Abuse (NIDA), Department of Health and Human Services. Drug abuse treatment: an economical approach to addressing the drug problem in America. Rockville, MD: U.S. Department of Health and Human Services, Public Health Services; 1991. 14. Booth A, Goichman S. States spend $81 billion to "shovel up" impact of substance abuse. Columbia News: The Public Affairs and Record Home Page; 2001. Retrieved January 15, 2002, from www.Columbia.edu/cu/news/01/02/CASA.html.

We Welcome Your Comments Journal of the National Medical Association welcomes your Letters to the Editor about articles that appear in the JNMA or issues relevant to minority health care. Address correspondence to Editor-in-Chief, ]NMA, 1012 Tenth St, NW, Washington, DC 20001; fax (202) 371-1162; or ktaylor

@nmanet.org.

VOL. 95, NO. 4, APRIL 2003