- Email: [email protected]
Factors associated with coercion in entering treatment for alcohol problems
Drug and Alcohol Dependence 54 (1999) 63 – 68
Factors associated with coercion in entering treatment for alcohol problems Douglas L. Polcin *, Constance Weisner Alcohol Research Group, Uni6ersity of California, Berkeley, 2000 Hearst A6enue, Ste. 300, Berkeley, CA 94709, USA Received 2 February 1998; accepted 11 August 1998
Abstract Although the importance of coercion in entry to treatment for alcohol problems is recognized, few studies have focused on different types and levels of coercion among heterogeneous groups of clients entering treatment agencies. This paper describes demographic and problem characteristics associated with various sources and levels of coercion. More than 40% (n =377) of individuals entering a representative sample of a county’s HMO, public, and private indemnity-based non-DUI alcohol treatment services (n=927) indicated they received an ultimatum to enter treatment from at least one person. The most common source of an ultimatum to enter treatment was from family members (n =222), followed by the legal system (n= 78), and healthcare professionals (n=55). Respondents experiencing pressure to enter treatment reported that ultimatums from more than one source were common. Individuals entering treatment who were most likely to report being coerced were white, young adults (age 18–39), and married or living with a partner. When controlling for demographic characteristics and problem severity, family problem severity and legal problem severity predicted having received an ultimatum to enter treatment. Alcohol and drug problem severity were not related to receiving a treatment ultimatum. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Addiction Severity Index; Coercion; Pressure to enter treatment; Ultimatum
1. Introduction Despite the high prevalence of substance abuse in the United States and its enormous effects on criminal justice, health care, and welfare systems, most people with alcohol problems do not receive professional treatment or join self help groups such as Alcoholics Anonymous (Regier et al., 1993). It is therefore important to understand the pathways by which individuals enter alcohol programs in order to develop strategies that facilitate entry into treatment. In this regard, literature in the alcohol field increasingly documents the important role of coercion in the treatment entry process. This research has found that few individuals seek help voluntarily to address a problem with alcohol (Weisner, 1990b, 1993; George and * Corresponding author.
Tucker, 1996). People are far more likely to enter treatment because of external pressure resulting from psychosocial, legal, medical, or occupational problems associated with alcohol abuse (Gerstein and Harwood, 1990; Weisner, 1990a). In this regard, recent studies have documented the importance of social consequences as compared to severity of dependence in entering treatment (Weisner, 1992; Kaskutas et al., 1997b). Examining coercion in treatment entry is particularly important because strategies oriented toward emphasizing coercion are increasingly considered in health and social policy strategies (Schmidt and Weisner, 1993). Studies of coercion have primarily focused on legal or workplace pressure, and are usually limited to specific populations. Types and levels of coercion and their association with different groups have not been contrasted (Weisner, 1990a). For example, few studies have
0376-8716/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S0376-8716(98)00143-4
64
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
examined a heterogeneous group of treatment programs and clients. It is unclear how characteristics of individuals entering a cross-section of programs may vary by sources and level of pressure to enter treatment. Further, individuals coerced from different sources (e.g., criminal justice, welfare department, workplace, family, friends, health professionals, etc.) may differ in regard to demographic characteristics, severity of their alcohol or drug problems, family and social functioning, legal status, or psychiatric symptoms. Studies have most often approached the study of coercion as an all or none phenomenon. However, recent research has argued for studying it as a continuum or as multidimensional (Weisner, 1990a; WellsParker, 1994). For example, individuals in the workplace may experience coercion ranging from mild suggestions from peers or supervisors to the threat of job loss, and they may or may not also be receiving pressure from family at the same time. Wells-Parker (1994) demonstrated that amount of coercion to treatment exerted on drinking drivers and how they perceive it may also vary substantially. Similarly, differences in receiving pressure to enter treatment which may differ by population characteristics such as gender and ethnicity have not been studied. Finally, whether such pressure is related to more serious levels of alcohol severity or whether other factors, such as demographic characteristics, social consequences, or psychiatric status play a role in coercive referrals is unknown. The purpose of this study is to describe factors associated with coercion among individuals entering a single county’s public, private, and HMO alcohol treatment programs. The study documents levels of pressure to enter treatment from criminal justice and welfare departments, healthcare workers, family, friends, and the workplace. It also examines sociodemographic factors, severity of drug and alcohol problems, and alcohol related social problems and consequences in relation to such pressures to enter treatment.
2. Methods
2.1. Sample The study was conducted in a Northern California County whose population is about 750000. This county is the site of the Alcohol Research Group’s Community Epidemiology Laboratory and was selected on the basis of its diversity on characteristics ethnicity, mix of rural, and urban areas and socioeconomic status (Weisner and Schmidt, 1995). The sample was representative of the county’s public, private and HMO treatment programs (n = 927) and included consecutive admissions of individuals entering public (n=298), HMO (n =334), and private (n = 295)
alcohol treatment programs. Clients were sampled from the full spectrum of care, which included detox, inpatient, residential, day treatment, and outpatient programs. Therapeutic approaches in these settings varied and included social, medical, and psychological models of treatment (Kaskutas et al., 1997a). Data collection took place over a nine month period from February 1995 to November 1995. Ten treatment programs took part in the study. They included all programs in the county that met the following selection criteria: (1) At least one new intake per week, (2) a majority of clients having primarily alcohol rather than solely drug problems, (3) programs that were not limited to treating DUI clients, (4) first line treatment entry (i.e., excluded programs that were limited to aftercare). Thus, our sample does not include programs which focus solely on drug abuse such as methadone maintenance or specialty driving under the influence (DUI) alcohol programs. The gender distribution in the sample was 65% male (n= 608) and 35% female (n= 320). Thirty percent were married or living together as married. Most clients were between the ages of 25 and 44, with 30% between 25–34 and 36% between 35–44. Thirty two percent of the sample reported an income over US $35000; 27% US $10000–35000; and 41% less than US $10000. Over one-half of the sample was educated at the high school level or above (51%), 27% had at least some college or technical training, and 22% had less than a high school education. The distribution by ethnicity included 54% who identified themselves as white, 33% as AfricanAmerican, and 5% as Hispanic. The response rate was 80%. Data were weighted to account for differences in field work duration across agencies and non-response differences (age, sex, ethnicity) within agencies (Kaskutas et al., 1997a).
2.2. Procedures Trained interviewers who were independent of the treatment agencies administered structured in-person questionnaires to all participants by the end of their third day of residential treatment or third outpatient visit. Each interview took approximately one hour and fifteen minutes. Informed consent was obtained and participation was independent of receiving agency services.
2.3. Measurement Level of coercion is a count of respondent responses to a question asking whether any of the following had given an ultimatum to enter treatment: mothers, fathers, sisters, brothers, sons, daughters, friends, welfare, police or courts, workplace colleagues, supervisors, physicians or other healthcare workers, or clergy.
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
Alcohol and drug problems were assessed through two different sets of questions. First, DSM IV diagnoses of alcohol and drug dependence were made through administration of questions based on the Diagnostic and Statistical Manual of the American Psychiatric Association (American Psychiatric Association, 1994). Level of alcohol and drug severity was also assessed by the composite scores from the Addiction Severity Index (ASI) (McLellan et al., 1992). We also used the ASI to assess the severity of employment, medical, psychiatric, family/social, and legal problems.
2.4. Data analysis
65
Table 1 Number of sources of ultimatums to enter treatment by demographic characteristics (in percent) 0 (550)
1
2+
(269)
(109)
Gender Men (608) Women (320)
67 32
64 36
59 41
Ethnicity* White (498) African-American (301) Hispanic (50) Other (77)
51 36 4 8
56 30 6 8
61 21 9 9
Marital status** Married/living together (281) Separated/divorced (313) Widowed (29) Never married (302)
26 36 3 35
40 30 4 26
28 31 4 37
Income (US$) B3000 (164) 3000–9999 (209) 10000–19999 (121) 20000–34999 (124) \35000 (285)
19 25 12 14 30
16 23 14 13 34
19 16 19 12 34
Chi-square tests were used to assess significant differences for categorical variables, and analysis of variance F-tests were conducted to assess mean differences for the ASI. For bivariate analysis, coercion was coded as the number of ultimatums that respondents reported (0, 1 or 2+ ). We used independent tests of proportions to assess differences in level of alcohol severity among different sources of coercion. The proportion of clients reporting each source of pressure who obtained scores above the median on the ASI Alcohol Severity subscale were contrasted using pressure from family as the reference group. We used logistic regression to assess the relationship of demographic measures and problem severity to predict having received one or more ultimatums (= 1) versus not (=0). For the logistic regressions, dichotomous and categorical variables used the following reference categories: female, age 40 – 49, white, less than a high school education, married, US $20000 – 34,000 income level, and ASI scores below the mean.
Age*** 18–24 (67) 25–34 (278) 35–44 (340) 45–54 (178) 55+ (60)
7 26 38 19 9
5 35 35 23 3
15 37 36 10 3
Education*** BHigh School (207) High School (469) 2 yr College/Tech. School (148) College grad (101)
23 47 15 14
19 56 20 6
25 55 15 6
3. Results
* pB0.05, ** pB.001, *** pB0.001. Chi-square tests of significance. Controlling for multiple comparisons: 0.05/6= pB0.008. Values of N in parentheses.
Table 1 examines demographic characteristics of respondents reporting ultimatums from 0, 1, or E 2 sources. Over 40% of the sample reported receiving an ultimatum to enter treatment from at least one person. When controlling for multiple comparisons, only age
and education were significant. Larger proportions of those reporting E 2 sources were in the youngest age groups than those with no ultimatums. The reverse was the case for the two oldest groups and for those who were college graduates.
Table 2 Sources of ultimatums and alcohol severity (in percent) Source of ultima- Percent of tum sample
At least one additional source of ultimatum
Percentage above the median on ASI alcohol severity scale
Family (222) Friend (51) Work (51) Healthcare (55) Legal (78) Welfare (19)
27 61 43 43 37 58
51 48 46 63 35* 20**
24 6 6 6 8 2
* pB0.05, ** pB0.01. Independent tests of proportions with family as the reference category. Controlling for multiple comparisons: 0.05/5= 0.01. Values of N are in parentheses.
66
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
Table 3 Number of sources of ultimatums by Addiction Severity Index composite scores ( 9 SD)
Alcohol severity Drug severity Psychiatric severity Medical severity Employment severity Family or social severity Legal problem severity
N
0
883 886 908 917 900 913 877
0.379 0.123 0.371 0.276 0.733 0.227 0.111
1 (0.32) (0.13) (0.25) (0.37) (0.33) (0.28) (0.19)
0.377 0.127 0.403 0.311 0.707 0.332 0.156
2+ (0.33) (0.13) (0.24) (0.37) (0.34) (0.32) (0.20)
0.314 0.161 0.463 0.238 0.784 0.414 0.240
(0.33) (0.14)* (0.23)** (0.37) (0.31) (0.30)*** (0.25)***
* pB0.05, ** pB0.01, *** pB0.001. Analysis of variance, F-tests of mean differences.
Table 2 presents sources of ultimatums. The first column identifies the proportion of those who reported receiving an ultimatum from each source. The second column indicates the percent of those from each source who reported receiving an ultimatum from at least one additional source. The third column presents the proportion who scored above the median of the ASI alcohol composite score for those receiving each source of ultimatum. The largest proportion of respondents reported an ultimatum from family members. Although more of those married received an ultimatum than those unmarried, (Table 1), 70% of the respondents reporting ultimatums from family members were not married. Ultimatums from welfare departments were relatively rare in comparison to other sources. If any ultimatum was received, ultimatums from more than one source were common, particularly so for individuals who reported ultimatums from friends and welfare. Around 60% of those who reported receiving such ultimatums also reported ultimatums from at least one other source. Smaller proportions of respondents who received ultimatums from family members reported receiving them from another source. On the whole, source of ultimatum was not associated with high levels of alcohol severity; about half of those reporting ultimatums from family, friends, and the workplace were close to the median of the alcohol severity composite score, and 63% of those with a health care ultimatum were above the median. Respondents who received ultimatums from legal and welfare systems had significantly lower levels of alcohol severity than clients who received ultimatums from family members. Table 3 presents ASI composite scores (mean9SD) measuring problem severity for individuals reporting 0, 1, or 2+ sources of ultimatums. There was a significant linear relationship for level of severity of drug, psychiatric, family/social, and legal problems associated with increasing numbers of sources of ultimatums to enter treatment. Similar relationships for medical and employment status were found, but not for alcohol status.
Consistent with these findings, the relationship of alcohol dependence with reporting an ultimatum to enter treatment was not significant (x 2 = 2.3, df =2, p= 0.32), (not shown). However, the relationship of drug dependence for particular drugs to reporting ultimatums differed from that of the overall ASI drug composite scores. A higher proportion of respondents who met DSM-IV criteria for amphetamine dependence reported receiving ultimatums than other respondents. Of the 151 respondents who met criteria for amphetamine dependence, 36% reported an ultimatum from one source while 22% reported two or more sources (x 2 = 25.12, df=2 pB 0.001). However, cocaine dependence was inversely related to reporting an ultimatum (x 2 = 6.45, df= 2, pB 0.05. We used logistic regression analysis (Table 4) to assess the importance of demographic characteristics and ASI problem severity in predicting whether respondents had received an ultimatum to enter treatment or not. Individuals over age 60 were only about one fourth as likely as those aged 40–49 to report receiving an ultimatum (pB 0.05), and African-Americans were about two thirds as likely (pB 0.05) as whites. Respondents who were not married or living with a partner were about two-thirds as likely as others to report an ultimatum (pB 0.01). In regard to problem severity, clients above the mean on legal severity were over twice as likely to report an ultimatum (pB 0.001), and those above the mean on family severity were one and onehalf times as likely to report an ultimatum (pB0.05). Those above the mean on the psychiatric severity score were more than a third more likely (p= 0.07). Those above the mean on the alcohol severity score were less likely (p=0.09) than their counterparts to report receiving an ultimatum.
4. Discussion In examining the role of coercion among a sample of individuals entering a representative sample of a county’s public, private and HMO alcohol treatment programs, we examined how sociodemographic charac-
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
teristics, problem severity, and type of dependence were associated with coercion in general, as well as with different types and levels of coercion. Our bivariate analysis found that treatment clients who were younger and had less than a college education were more likely to report receiving ultimatums. These are characteristics that may expose individuals to pressure, or to varying types of pressure to enter treatment. Young age may be related to lifestyles that involve the use of substances in public social settings. It is important to interpret these findings in the context of our sample. It included the county’s system of public, private and managed care alcohol treatment agencies, whereas previous studies on alcohol treatment entry have focused on particular treatment sectors, primarily public programs or private programs, which may overrepresent criminal justice or workplace referrals, respectively. The role of coercion generally, as well as different types of pressure, may differ across treatment sectors. It must also be acknowledged that our sample was limited to clients entering non-DUI programs. The factors associated with coercion in programs specializing in drug problems, such as methadone maintenance, may differ. Table 4 Logistic regression of demographic characteristics and Addiction Severity Index mean scores on receiving at least one ultimatum (versus none)
Male Age 18–29 30–39 50–59 60+ Ethnicity African-American Other Education High School Grad. \High School Grad. Not married or living together Income (US$) B3000 3000–9999 10000–19999 \35000 Employment Medical severity Employment severity Alcohol severity Drug severity Family severity Legal severity Psychiatric severity * pB0.05, ** pB0.01, *** pB0.001.
Odds ratio
95% C.I.
1.11
0.80–1.56
0.98 0.94 0.89 0.28*
0.61–1.57 0.65–1.37 0.50–1.57 0.09–0.82
0.68* 1.03
0.47–0.99 0.64–1.65
1.23 0.87 0.60**
0.83–1.83 0.54–1.39 0.43–0.86
1.09 1.06 1.38 1.04 0.75 1.05 1.11 0.76 1.09 1.51* 2.10*** 1.36
0.60–2.01 0.60–1.87 0.75–2.53 0.62–1.73 0.46–1.22 0.77–1.44 0.75–1.64 0.55–1.03 0.78–1.51 1.08–2.10 1.53–2.87 0.98–1.90
67
We found that substantially more ultimatums to enter treatment came from family members, with those from legal sources second. However, family ultimatums were not related to marital status. While married clients received more ultimatums than those unmarried, ultimatums from family members were not related to being married. Overall, receiving a treatment ultimatum appeared to be independent from severity of alcohol problems; most sources of ultimatums were not associated with high levels of alcohol severity. The notable exception was receiving an ultimatum from a health professional where severity levels were highest. However, even in this case, only about two thirds of those individuals had severity scores above the median on the ASI. Alcohol problem and medical severity are real exceptions to other measures of severity in our findings; increases in severity are not related to increasing sources of ultimatums. These findings suggest that for clients entering treatment, severity of alcohol problems per se is not what results in family, friends, or representatives from community institutions to become concerned and pressure them to treatment. Rather, it is the type of alcohol-related problem that makes a difference. This is consistent with other studies where alcohol-related social consequences have been more important in predicting treatment entry than have alcohol dependence symptoms (Weisner, 1992; Weisner et al., 1995; Kaskutas et al., 1997b). In our study, individuals who reported ultimatums from legal and welfare systems had significantly lower levels of alcohol severity in comparison to those receiving them from family. Further, although some areas of severity, in particular, legal, were not common, they were important in predicting receiving an ultimatum to enter treatment. The findings from our treatment sample suggest that professionals in medical settings compared with professionals in other community institutions may be less likely to respond to early signs in intervening in getting someone to treatment. (The rates for health professionals ‘suggesting’ treatment as opposed to giving an ultimatum were also lower in the same direction.) A higher proportion of respondents reporting pressure to enter treatment from a health professional had alcohol severity levels above the median ASI composite score. This does not reflect an early case finding approach in settings where there are large numbers of individuals with such problems (Cherpitel, 1994; Weisner and Schmidt, 1995; National Institute on Alcohol Abuse and Alcoholism, 1993; 1997). Although previous studies have documented high rates of alcohol problems in criminal justice and welfare systems (Tam et al., 1996), pressure to enter treatment from these sources was not common. This may be due to the characteristics of our sample, which attempted to equally represent the county in regard to public and
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
68
private programs. However, as discussed previously, respondents who reported such pressure were those with lower levels of alcohol severity, reflecting screening on the basis of alcohol-related involvement in criminal offenses rather than on the basis of alcohol problems. In general, in our treatment sample, factors other than alcohol severity, that is, family and social, legal, drug, and psychiatric problems, were related to receiving an ultimatum to enter treatment. These findings suggest that treatment professionals in alcohol programs need to assess a variety of problem areas in addition to alcohol dependence when treating patients who are referred from other institutional sources. Pressure to enter treatment, even when measured in the extreme form of ultimatums, cannot be assumed to indicate higher levels of alcohol severity, but rather, may indicate the existence of family, social, drug, legal or psychiatric problems needing to be addressed in treatment.
Acknowledgements This study was funded by NIAAA RO1 AA09750.
References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. American Psychiatric Association, Washington DC, USA. Cherpitel, C., 1994. Alcohol use among primary care patients: Comparing an HMO with county clinics and the general population. Drug Alcohol Depend. 36, 167–173. George, A.A., Tucker, J.E., 1996. Help-seeking for alcohol-related problems: Social contexts surrounding entry into alcoholism treatment or Alcoholics Anonymous. J Stud. Alcohol 57, 449– 457. Gerstein, D.R., Harwood, H.J. (Eds.), 1990. Treating Drug Problems, Vol 1. A Study of the Evolution, Effectiveness, and Financing of Public and Private Drug Treatment Systems. National Academy Press, Washington, DC.
.
Kaskutas, L., Russell, G., Dinish, M., Tam, T., 1997. Technical Report on the Alcohol Utilization Study in Public and Private Sectors. Alcohol Research Group, Berkeley, CA. Kaskutas, L., Weisner, C., Caetano, R., 1997b. Predictors of help seeking among a longitudinal sample of the general population, 1984 – 1992. J. Stud. Alcoholism 58, 155 – 161. McLellan, A.T., Cacciola, J., Kushner, H., Peters, F., Smith, I., Pettinati, H., 1992. The fifth edition of the Addiction Severity Index: Cautions, additions, and normative data. J. Substance Abuse Treatment 9, 199 – 213. National Institute on Alcohol Abuse and Alcoholism (1993). Alcohol and Health: Eighth Special Report to the US Congress. US Department of Health and Human Services, Washington, DC. National Institute on Alcohol Abuse and Alcoholism (1997). Alcohol and Health: Ninth Special Report to the US Congress. US Department of Health and Human Services, Washington, DC. Regier, D.A., Narrow, W.E., Rae, D.S., Manderscheid, R.W., Locke, B.Z., Goodwin, F.K., 1993. The de facto U.S. mental and addictive disorders services system: Epidemiological catchment area prospective 1-year prevalence rates of disorders and services. Arch. Gen. Psychiatry 50, 85 – 94. Schmidt, L., Weisner, C., 1993). Developments in alcoholism treatment. In: Galanter M. (Ed.), Recent Developments in Alcoholism, Vol. 11. New York, Plenum. Tam, T.W., Schmidt, L., Weisner, C., 1996. Patterns in the institutional encounters of problem drinkers in a community human services network. Addiction 91 (5), 657 – 669. Weisner, C., 1990a. Coercion in alcohol treatment. In: Broadening the Base of Treatment for Alcohol Problems. Institute of Medicine, National Academy Press, Washington, DC. Weisner, C., 1990b. The alcohol treatment-seeking process from a problems perspective: Response to events. Br. J. Addictions 85, 561 – 569. Weisner, C., 1992. A comparison of alcohol and drug treatment clients: Are they from the same population? Am. J. Drug Alcohol Abuse 18 (4), 429 – 444. Weisner, C., 1993. Toward an alcohol treatment entry model: A comparison of problem drinkers in the general population and in treatment. Alcoholism Clin. Exp. Res. 17, 746 – 752. Weisner, C., Schmidt, L., 1995. The community epidemiology laboratory: Studying alcohol problems in community and agency-based populations. Addiction 90, 329 – 341. Weisner, C., Greenfield, T., Room, R., 1995. Trends in the treatment of alcohol problems in the U.S. general population, 1979 through 1990. Am. J. Publ. Health 85 (1), 55 – 60. Wells-Parker, E., 1994. Mandated treatment: Lessons from research with drinking and driving offenders. Alcohol Health Res. World 18 (4), 302 – 306.
Factors associated with coercion in entering treatment for alcohol problems Douglas L. Polcin *, Constance Weisner Alcohol Research Group, Uni6ersity of California, Berkeley, 2000 Hearst A6enue, Ste. 300, Berkeley, CA 94709, USA Received 2 February 1998; accepted 11 August 1998
Abstract Although the importance of coercion in entry to treatment for alcohol problems is recognized, few studies have focused on different types and levels of coercion among heterogeneous groups of clients entering treatment agencies. This paper describes demographic and problem characteristics associated with various sources and levels of coercion. More than 40% (n =377) of individuals entering a representative sample of a county’s HMO, public, and private indemnity-based non-DUI alcohol treatment services (n=927) indicated they received an ultimatum to enter treatment from at least one person. The most common source of an ultimatum to enter treatment was from family members (n =222), followed by the legal system (n= 78), and healthcare professionals (n=55). Respondents experiencing pressure to enter treatment reported that ultimatums from more than one source were common. Individuals entering treatment who were most likely to report being coerced were white, young adults (age 18–39), and married or living with a partner. When controlling for demographic characteristics and problem severity, family problem severity and legal problem severity predicted having received an ultimatum to enter treatment. Alcohol and drug problem severity were not related to receiving a treatment ultimatum. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Addiction Severity Index; Coercion; Pressure to enter treatment; Ultimatum
1. Introduction Despite the high prevalence of substance abuse in the United States and its enormous effects on criminal justice, health care, and welfare systems, most people with alcohol problems do not receive professional treatment or join self help groups such as Alcoholics Anonymous (Regier et al., 1993). It is therefore important to understand the pathways by which individuals enter alcohol programs in order to develop strategies that facilitate entry into treatment. In this regard, literature in the alcohol field increasingly documents the important role of coercion in the treatment entry process. This research has found that few individuals seek help voluntarily to address a problem with alcohol (Weisner, 1990b, 1993; George and * Corresponding author.
Tucker, 1996). People are far more likely to enter treatment because of external pressure resulting from psychosocial, legal, medical, or occupational problems associated with alcohol abuse (Gerstein and Harwood, 1990; Weisner, 1990a). In this regard, recent studies have documented the importance of social consequences as compared to severity of dependence in entering treatment (Weisner, 1992; Kaskutas et al., 1997b). Examining coercion in treatment entry is particularly important because strategies oriented toward emphasizing coercion are increasingly considered in health and social policy strategies (Schmidt and Weisner, 1993). Studies of coercion have primarily focused on legal or workplace pressure, and are usually limited to specific populations. Types and levels of coercion and their association with different groups have not been contrasted (Weisner, 1990a). For example, few studies have
0376-8716/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S0376-8716(98)00143-4
64
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
examined a heterogeneous group of treatment programs and clients. It is unclear how characteristics of individuals entering a cross-section of programs may vary by sources and level of pressure to enter treatment. Further, individuals coerced from different sources (e.g., criminal justice, welfare department, workplace, family, friends, health professionals, etc.) may differ in regard to demographic characteristics, severity of their alcohol or drug problems, family and social functioning, legal status, or psychiatric symptoms. Studies have most often approached the study of coercion as an all or none phenomenon. However, recent research has argued for studying it as a continuum or as multidimensional (Weisner, 1990a; WellsParker, 1994). For example, individuals in the workplace may experience coercion ranging from mild suggestions from peers or supervisors to the threat of job loss, and they may or may not also be receiving pressure from family at the same time. Wells-Parker (1994) demonstrated that amount of coercion to treatment exerted on drinking drivers and how they perceive it may also vary substantially. Similarly, differences in receiving pressure to enter treatment which may differ by population characteristics such as gender and ethnicity have not been studied. Finally, whether such pressure is related to more serious levels of alcohol severity or whether other factors, such as demographic characteristics, social consequences, or psychiatric status play a role in coercive referrals is unknown. The purpose of this study is to describe factors associated with coercion among individuals entering a single county’s public, private, and HMO alcohol treatment programs. The study documents levels of pressure to enter treatment from criminal justice and welfare departments, healthcare workers, family, friends, and the workplace. It also examines sociodemographic factors, severity of drug and alcohol problems, and alcohol related social problems and consequences in relation to such pressures to enter treatment.
2. Methods
2.1. Sample The study was conducted in a Northern California County whose population is about 750000. This county is the site of the Alcohol Research Group’s Community Epidemiology Laboratory and was selected on the basis of its diversity on characteristics ethnicity, mix of rural, and urban areas and socioeconomic status (Weisner and Schmidt, 1995). The sample was representative of the county’s public, private and HMO treatment programs (n = 927) and included consecutive admissions of individuals entering public (n=298), HMO (n =334), and private (n = 295)
alcohol treatment programs. Clients were sampled from the full spectrum of care, which included detox, inpatient, residential, day treatment, and outpatient programs. Therapeutic approaches in these settings varied and included social, medical, and psychological models of treatment (Kaskutas et al., 1997a). Data collection took place over a nine month period from February 1995 to November 1995. Ten treatment programs took part in the study. They included all programs in the county that met the following selection criteria: (1) At least one new intake per week, (2) a majority of clients having primarily alcohol rather than solely drug problems, (3) programs that were not limited to treating DUI clients, (4) first line treatment entry (i.e., excluded programs that were limited to aftercare). Thus, our sample does not include programs which focus solely on drug abuse such as methadone maintenance or specialty driving under the influence (DUI) alcohol programs. The gender distribution in the sample was 65% male (n= 608) and 35% female (n= 320). Thirty percent were married or living together as married. Most clients were between the ages of 25 and 44, with 30% between 25–34 and 36% between 35–44. Thirty two percent of the sample reported an income over US $35000; 27% US $10000–35000; and 41% less than US $10000. Over one-half of the sample was educated at the high school level or above (51%), 27% had at least some college or technical training, and 22% had less than a high school education. The distribution by ethnicity included 54% who identified themselves as white, 33% as AfricanAmerican, and 5% as Hispanic. The response rate was 80%. Data were weighted to account for differences in field work duration across agencies and non-response differences (age, sex, ethnicity) within agencies (Kaskutas et al., 1997a).
2.2. Procedures Trained interviewers who were independent of the treatment agencies administered structured in-person questionnaires to all participants by the end of their third day of residential treatment or third outpatient visit. Each interview took approximately one hour and fifteen minutes. Informed consent was obtained and participation was independent of receiving agency services.
2.3. Measurement Level of coercion is a count of respondent responses to a question asking whether any of the following had given an ultimatum to enter treatment: mothers, fathers, sisters, brothers, sons, daughters, friends, welfare, police or courts, workplace colleagues, supervisors, physicians or other healthcare workers, or clergy.
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
Alcohol and drug problems were assessed through two different sets of questions. First, DSM IV diagnoses of alcohol and drug dependence were made through administration of questions based on the Diagnostic and Statistical Manual of the American Psychiatric Association (American Psychiatric Association, 1994). Level of alcohol and drug severity was also assessed by the composite scores from the Addiction Severity Index (ASI) (McLellan et al., 1992). We also used the ASI to assess the severity of employment, medical, psychiatric, family/social, and legal problems.
2.4. Data analysis
65
Table 1 Number of sources of ultimatums to enter treatment by demographic characteristics (in percent) 0 (550)
1
2+
(269)
(109)
Gender Men (608) Women (320)
67 32
64 36
59 41
Ethnicity* White (498) African-American (301) Hispanic (50) Other (77)
51 36 4 8
56 30 6 8
61 21 9 9
Marital status** Married/living together (281) Separated/divorced (313) Widowed (29) Never married (302)
26 36 3 35
40 30 4 26
28 31 4 37
Income (US$) B3000 (164) 3000–9999 (209) 10000–19999 (121) 20000–34999 (124) \35000 (285)
19 25 12 14 30
16 23 14 13 34
19 16 19 12 34
Chi-square tests were used to assess significant differences for categorical variables, and analysis of variance F-tests were conducted to assess mean differences for the ASI. For bivariate analysis, coercion was coded as the number of ultimatums that respondents reported (0, 1 or 2+ ). We used independent tests of proportions to assess differences in level of alcohol severity among different sources of coercion. The proportion of clients reporting each source of pressure who obtained scores above the median on the ASI Alcohol Severity subscale were contrasted using pressure from family as the reference group. We used logistic regression to assess the relationship of demographic measures and problem severity to predict having received one or more ultimatums (= 1) versus not (=0). For the logistic regressions, dichotomous and categorical variables used the following reference categories: female, age 40 – 49, white, less than a high school education, married, US $20000 – 34,000 income level, and ASI scores below the mean.
Age*** 18–24 (67) 25–34 (278) 35–44 (340) 45–54 (178) 55+ (60)
7 26 38 19 9
5 35 35 23 3
15 37 36 10 3
Education*** BHigh School (207) High School (469) 2 yr College/Tech. School (148) College grad (101)
23 47 15 14
19 56 20 6
25 55 15 6
3. Results
* pB0.05, ** pB.001, *** pB0.001. Chi-square tests of significance. Controlling for multiple comparisons: 0.05/6= pB0.008. Values of N in parentheses.
Table 1 examines demographic characteristics of respondents reporting ultimatums from 0, 1, or E 2 sources. Over 40% of the sample reported receiving an ultimatum to enter treatment from at least one person. When controlling for multiple comparisons, only age
and education were significant. Larger proportions of those reporting E 2 sources were in the youngest age groups than those with no ultimatums. The reverse was the case for the two oldest groups and for those who were college graduates.
Table 2 Sources of ultimatums and alcohol severity (in percent) Source of ultima- Percent of tum sample
At least one additional source of ultimatum
Percentage above the median on ASI alcohol severity scale
Family (222) Friend (51) Work (51) Healthcare (55) Legal (78) Welfare (19)
27 61 43 43 37 58
51 48 46 63 35* 20**
24 6 6 6 8 2
* pB0.05, ** pB0.01. Independent tests of proportions with family as the reference category. Controlling for multiple comparisons: 0.05/5= 0.01. Values of N are in parentheses.
66
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
Table 3 Number of sources of ultimatums by Addiction Severity Index composite scores ( 9 SD)
Alcohol severity Drug severity Psychiatric severity Medical severity Employment severity Family or social severity Legal problem severity
N
0
883 886 908 917 900 913 877
0.379 0.123 0.371 0.276 0.733 0.227 0.111
1 (0.32) (0.13) (0.25) (0.37) (0.33) (0.28) (0.19)
0.377 0.127 0.403 0.311 0.707 0.332 0.156
2+ (0.33) (0.13) (0.24) (0.37) (0.34) (0.32) (0.20)
0.314 0.161 0.463 0.238 0.784 0.414 0.240
(0.33) (0.14)* (0.23)** (0.37) (0.31) (0.30)*** (0.25)***
* pB0.05, ** pB0.01, *** pB0.001. Analysis of variance, F-tests of mean differences.
Table 2 presents sources of ultimatums. The first column identifies the proportion of those who reported receiving an ultimatum from each source. The second column indicates the percent of those from each source who reported receiving an ultimatum from at least one additional source. The third column presents the proportion who scored above the median of the ASI alcohol composite score for those receiving each source of ultimatum. The largest proportion of respondents reported an ultimatum from family members. Although more of those married received an ultimatum than those unmarried, (Table 1), 70% of the respondents reporting ultimatums from family members were not married. Ultimatums from welfare departments were relatively rare in comparison to other sources. If any ultimatum was received, ultimatums from more than one source were common, particularly so for individuals who reported ultimatums from friends and welfare. Around 60% of those who reported receiving such ultimatums also reported ultimatums from at least one other source. Smaller proportions of respondents who received ultimatums from family members reported receiving them from another source. On the whole, source of ultimatum was not associated with high levels of alcohol severity; about half of those reporting ultimatums from family, friends, and the workplace were close to the median of the alcohol severity composite score, and 63% of those with a health care ultimatum were above the median. Respondents who received ultimatums from legal and welfare systems had significantly lower levels of alcohol severity than clients who received ultimatums from family members. Table 3 presents ASI composite scores (mean9SD) measuring problem severity for individuals reporting 0, 1, or 2+ sources of ultimatums. There was a significant linear relationship for level of severity of drug, psychiatric, family/social, and legal problems associated with increasing numbers of sources of ultimatums to enter treatment. Similar relationships for medical and employment status were found, but not for alcohol status.
Consistent with these findings, the relationship of alcohol dependence with reporting an ultimatum to enter treatment was not significant (x 2 = 2.3, df =2, p= 0.32), (not shown). However, the relationship of drug dependence for particular drugs to reporting ultimatums differed from that of the overall ASI drug composite scores. A higher proportion of respondents who met DSM-IV criteria for amphetamine dependence reported receiving ultimatums than other respondents. Of the 151 respondents who met criteria for amphetamine dependence, 36% reported an ultimatum from one source while 22% reported two or more sources (x 2 = 25.12, df=2 pB 0.001). However, cocaine dependence was inversely related to reporting an ultimatum (x 2 = 6.45, df= 2, pB 0.05. We used logistic regression analysis (Table 4) to assess the importance of demographic characteristics and ASI problem severity in predicting whether respondents had received an ultimatum to enter treatment or not. Individuals over age 60 were only about one fourth as likely as those aged 40–49 to report receiving an ultimatum (pB 0.05), and African-Americans were about two thirds as likely (pB 0.05) as whites. Respondents who were not married or living with a partner were about two-thirds as likely as others to report an ultimatum (pB 0.01). In regard to problem severity, clients above the mean on legal severity were over twice as likely to report an ultimatum (pB 0.001), and those above the mean on family severity were one and onehalf times as likely to report an ultimatum (pB0.05). Those above the mean on the psychiatric severity score were more than a third more likely (p= 0.07). Those above the mean on the alcohol severity score were less likely (p=0.09) than their counterparts to report receiving an ultimatum.
4. Discussion In examining the role of coercion among a sample of individuals entering a representative sample of a county’s public, private and HMO alcohol treatment programs, we examined how sociodemographic charac-
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
teristics, problem severity, and type of dependence were associated with coercion in general, as well as with different types and levels of coercion. Our bivariate analysis found that treatment clients who were younger and had less than a college education were more likely to report receiving ultimatums. These are characteristics that may expose individuals to pressure, or to varying types of pressure to enter treatment. Young age may be related to lifestyles that involve the use of substances in public social settings. It is important to interpret these findings in the context of our sample. It included the county’s system of public, private and managed care alcohol treatment agencies, whereas previous studies on alcohol treatment entry have focused on particular treatment sectors, primarily public programs or private programs, which may overrepresent criminal justice or workplace referrals, respectively. The role of coercion generally, as well as different types of pressure, may differ across treatment sectors. It must also be acknowledged that our sample was limited to clients entering non-DUI programs. The factors associated with coercion in programs specializing in drug problems, such as methadone maintenance, may differ. Table 4 Logistic regression of demographic characteristics and Addiction Severity Index mean scores on receiving at least one ultimatum (versus none)
Male Age 18–29 30–39 50–59 60+ Ethnicity African-American Other Education High School Grad. \High School Grad. Not married or living together Income (US$) B3000 3000–9999 10000–19999 \35000 Employment Medical severity Employment severity Alcohol severity Drug severity Family severity Legal severity Psychiatric severity * pB0.05, ** pB0.01, *** pB0.001.
Odds ratio
95% C.I.
1.11
0.80–1.56
0.98 0.94 0.89 0.28*
0.61–1.57 0.65–1.37 0.50–1.57 0.09–0.82
0.68* 1.03
0.47–0.99 0.64–1.65
1.23 0.87 0.60**
0.83–1.83 0.54–1.39 0.43–0.86
1.09 1.06 1.38 1.04 0.75 1.05 1.11 0.76 1.09 1.51* 2.10*** 1.36
0.60–2.01 0.60–1.87 0.75–2.53 0.62–1.73 0.46–1.22 0.77–1.44 0.75–1.64 0.55–1.03 0.78–1.51 1.08–2.10 1.53–2.87 0.98–1.90
67
We found that substantially more ultimatums to enter treatment came from family members, with those from legal sources second. However, family ultimatums were not related to marital status. While married clients received more ultimatums than those unmarried, ultimatums from family members were not related to being married. Overall, receiving a treatment ultimatum appeared to be independent from severity of alcohol problems; most sources of ultimatums were not associated with high levels of alcohol severity. The notable exception was receiving an ultimatum from a health professional where severity levels were highest. However, even in this case, only about two thirds of those individuals had severity scores above the median on the ASI. Alcohol problem and medical severity are real exceptions to other measures of severity in our findings; increases in severity are not related to increasing sources of ultimatums. These findings suggest that for clients entering treatment, severity of alcohol problems per se is not what results in family, friends, or representatives from community institutions to become concerned and pressure them to treatment. Rather, it is the type of alcohol-related problem that makes a difference. This is consistent with other studies where alcohol-related social consequences have been more important in predicting treatment entry than have alcohol dependence symptoms (Weisner, 1992; Weisner et al., 1995; Kaskutas et al., 1997b). In our study, individuals who reported ultimatums from legal and welfare systems had significantly lower levels of alcohol severity in comparison to those receiving them from family. Further, although some areas of severity, in particular, legal, were not common, they were important in predicting receiving an ultimatum to enter treatment. The findings from our treatment sample suggest that professionals in medical settings compared with professionals in other community institutions may be less likely to respond to early signs in intervening in getting someone to treatment. (The rates for health professionals ‘suggesting’ treatment as opposed to giving an ultimatum were also lower in the same direction.) A higher proportion of respondents reporting pressure to enter treatment from a health professional had alcohol severity levels above the median ASI composite score. This does not reflect an early case finding approach in settings where there are large numbers of individuals with such problems (Cherpitel, 1994; Weisner and Schmidt, 1995; National Institute on Alcohol Abuse and Alcoholism, 1993; 1997). Although previous studies have documented high rates of alcohol problems in criminal justice and welfare systems (Tam et al., 1996), pressure to enter treatment from these sources was not common. This may be due to the characteristics of our sample, which attempted to equally represent the county in regard to public and
D.L. Polcin, C. Weisner / Drug and Alcohol Dependence 54 (1999) 63–68
68
private programs. However, as discussed previously, respondents who reported such pressure were those with lower levels of alcohol severity, reflecting screening on the basis of alcohol-related involvement in criminal offenses rather than on the basis of alcohol problems. In general, in our treatment sample, factors other than alcohol severity, that is, family and social, legal, drug, and psychiatric problems, were related to receiving an ultimatum to enter treatment. These findings suggest that treatment professionals in alcohol programs need to assess a variety of problem areas in addition to alcohol dependence when treating patients who are referred from other institutional sources. Pressure to enter treatment, even when measured in the extreme form of ultimatums, cannot be assumed to indicate higher levels of alcohol severity, but rather, may indicate the existence of family, social, drug, legal or psychiatric problems needing to be addressed in treatment.
Acknowledgements This study was funded by NIAAA RO1 AA09750.
References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. American Psychiatric Association, Washington DC, USA. Cherpitel, C., 1994. Alcohol use among primary care patients: Comparing an HMO with county clinics and the general population. Drug Alcohol Depend. 36, 167–173. George, A.A., Tucker, J.E., 1996. Help-seeking for alcohol-related problems: Social contexts surrounding entry into alcoholism treatment or Alcoholics Anonymous. J Stud. Alcohol 57, 449– 457. Gerstein, D.R., Harwood, H.J. (Eds.), 1990. Treating Drug Problems, Vol 1. A Study of the Evolution, Effectiveness, and Financing of Public and Private Drug Treatment Systems. National Academy Press, Washington, DC.
.
Kaskutas, L., Russell, G., Dinish, M., Tam, T., 1997. Technical Report on the Alcohol Utilization Study in Public and Private Sectors. Alcohol Research Group, Berkeley, CA. Kaskutas, L., Weisner, C., Caetano, R., 1997b. Predictors of help seeking among a longitudinal sample of the general population, 1984 – 1992. J. Stud. Alcoholism 58, 155 – 161. McLellan, A.T., Cacciola, J., Kushner, H., Peters, F., Smith, I., Pettinati, H., 1992. The fifth edition of the Addiction Severity Index: Cautions, additions, and normative data. J. Substance Abuse Treatment 9, 199 – 213. National Institute on Alcohol Abuse and Alcoholism (1993). Alcohol and Health: Eighth Special Report to the US Congress. US Department of Health and Human Services, Washington, DC. National Institute on Alcohol Abuse and Alcoholism (1997). Alcohol and Health: Ninth Special Report to the US Congress. US Department of Health and Human Services, Washington, DC. Regier, D.A., Narrow, W.E., Rae, D.S., Manderscheid, R.W., Locke, B.Z., Goodwin, F.K., 1993. The de facto U.S. mental and addictive disorders services system: Epidemiological catchment area prospective 1-year prevalence rates of disorders and services. Arch. Gen. Psychiatry 50, 85 – 94. Schmidt, L., Weisner, C., 1993). Developments in alcoholism treatment. In: Galanter M. (Ed.), Recent Developments in Alcoholism, Vol. 11. New York, Plenum. Tam, T.W., Schmidt, L., Weisner, C., 1996. Patterns in the institutional encounters of problem drinkers in a community human services network. Addiction 91 (5), 657 – 669. Weisner, C., 1990a. Coercion in alcohol treatment. In: Broadening the Base of Treatment for Alcohol Problems. Institute of Medicine, National Academy Press, Washington, DC. Weisner, C., 1990b. The alcohol treatment-seeking process from a problems perspective: Response to events. Br. J. Addictions 85, 561 – 569. Weisner, C., 1992. A comparison of alcohol and drug treatment clients: Are they from the same population? Am. J. Drug Alcohol Abuse 18 (4), 429 – 444. Weisner, C., 1993. Toward an alcohol treatment entry model: A comparison of problem drinkers in the general population and in treatment. Alcoholism Clin. Exp. Res. 17, 746 – 752. Weisner, C., Schmidt, L., 1995. The community epidemiology laboratory: Studying alcohol problems in community and agency-based populations. Addiction 90, 329 – 341. Weisner, C., Greenfield, T., Room, R., 1995. Trends in the treatment of alcohol problems in the U.S. general population, 1979 through 1990. Am. J. Publ. Health 85 (1), 55 – 60. Wells-Parker, E., 1994. Mandated treatment: Lessons from research with drinking and driving offenders. Alcohol Health Res. World 18 (4), 302 – 306.