Substance Abuse among Rural and Very Rural Drug Users at Treatment Entry*

Substance Abuse among Rural and Very RuralDrug Users at Treatment Entry*

Marlies L. Schoeneberger, Carl G. Leukefeld, Matthew L. Hiller,

and Ted GodlaskiCenter on Drug & Alcohol Research, University of Kentucky

Medical Center, Lexington, Kentucky, USA

Abstract: Historically, researchers and policymakers concerned with the pro-blems of substance abuse have focused their attention almost exclusively on urbanAmerica. However, this focus now includes the special needs of rural areas. In thepast, rural areas were more sheltered from the problems of mainstream America,but now, mass communication has decreased the isolation of rural areas. The pur-pose of this study is to examine rural and very rural drug users; including:

(a) demographic and other selected background characteristics of rural and veryrural drug users admitted to substance abuse treatment,

(b) lifetime drug use patterns,

(c) current drug use,

(d) age of first drug use,

(e) recognition of drug abuse as a problem, and

(f) correlates of drug use behavior to identify potential predisposing factors.

It was hypothesized that drug users from very rural areas would be more shelteredand the severity of their involvement with drug use would be less than that of drugusers from rural areas. Between November 15, 1999 and January 31, 2001, face-to-face interviews were conducted in three geographic regions of Kentucky(N¼604); eastern Kentucky (n¼206), south central Kentucky (n¼165)—both

�The points of view in this article are those of the authors.This project was supported by the Center for Substance Absue Treatment

(CSAT) and Substance Abuse and Mental Health Services Administration Coop-erative Agreement (SAMHSA), Grant number: 6 UR1 TI11613-01-1.Address correspondence to Marlies L. Schoeneberger, National Development

and Research Institutes, Inc., Center for Integration of Research and Practice,3600 Havana Street, Denver, CO 80239. E-mail: [email protected]

The American Journal of Drug and Alcohol Abuse, 32: 87–110, 2006

Copyright Q Taylor & Francis, Inc.

ISSN: 0095-2990 print/1097-9891 online

DOI: 10.1080/00952990500328687

87

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

frequently are considered rural (n¼371)—and Western Kentucky, which isconsidered more urban (N¼233). Over half of the participants were admittedto outpatient treatment (57%). The majority of participants were male (75%),White (94%), had no religious preference (51%), had been referred to by thecriminal justice system (64%), and had a history of prior substance abuse treat-ment (55%). Findings indicate that being from very rural areas may be somewhatprotective with lower current drug use and older age at drug use initiation. Whilealcohol remains the most used substance in terms of lifetime and current use,cocaine use was three times higher in rural areas than in very rural areas. Implica-tions are discussed and recommendations are presented for substance abuse treat-ment providers and policymakers.

Keywords: Drug use, rural, very rural, problem recognition

INTRODUCTION

Historically, researchers and policymakers concerned with substanceabuse problems focused their attention almost exclusively on urban Amer-ica. More recently, however, this focus shifted to include the special needsof people in rural areas. For example, chronic economic disadvantages inrural areas have been highlighted as one of the main reasons for theincreased health risks of persons in rural areas, which include emotional,behavioral, and substance use problems (1). In the past, rural areas mayhave been more sheltered from the problems of mainstream America,but now with mass communication rural areas are no longer isolated.

There also is increasing evidence of a symbiotic relationship betweenurban and rural drug traffickers, with urban drug dealers finding newdrug markets in rural areas and rural drug manufacturers selling in urbanareas (2). This symbiosis is visible in the substance abuse prevalence ratesin urban and rural areas. Examination of epidemiological trends showthat substance abuse in urban and rural areas of the United States havegrown similar (3), although, there appears to be some variation in drugsof choice. For example, earlier studies comparing urban and rural treat-ment populations, showed that marijuana, amphetamines, and sedativeswere the preferred drugs among rural clients, while urban clients pre-ferred opiates (4, 5).

The rural environment has changed the fortunes in rural America.Although there are wealthy rural people, most of rural America experi-enced devastating economic setbacks in the farming, manufacturing,and mining industries (6) during the past two decades. As a result, themyth of rural America as a stress-free bastion of serenity and healthdispelled. Indeed, research has shown that rural Americans, like their

88 M. L. Schoeneberger et al.

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

urban counterparts, suffer socioeconomic disadvantages similar to personsliving in poor urban neighborhoods (7). It also has been shown thateconomic stress is associated with a greater risk for mental disordersand functional impairment in rural as well as urban settings (8–11).

Chronic economic stress has had a long history in many parts of therural South. Among the manifestations of economic stresses in rural areasis what Davidson (12) called ‘‘America’s Rural Ghetto,’’ which is made upof the most disadvantaged and elderly of the population—due to the out-migration of the most affluent, educated, and younger rural people—notunlike inner-city urban areas (13). It is important to point out that regard-less of whether such rural economic stresses are chronic or recent, econ-omic deprivation is a common characteristic of social environments thatincrease the risk for substance abuse (6).

In an effort to establish criteria for what they called ‘‘the underclass,’’O’Hare and Curry-White (7) used census data to define 4 common char-acteristics: they tend to (a) have not completed high school, (b) receivepublic assistance, (c) be single mothers, or (d) be long-term unemployedmales. O’Hare and Curry-White (7) also found that, in 1990, 3.4% ofinner-city residents could be categorized as underclass when comparedto 2.4% of rural residents; and only 1.1% of urban residents not in theinner-city met the underclass criteria. These findings show that thereare differences within urban populations. If there are these kinds of dis-tinctions in urban populations, it could be hypothesized that there will besimilar distinctions in rural populations. If inner- and not-inner-citypopulations are different, then perhaps rural populations can be com-pared by examining rural areas and very rural areas.

In 1990, according to the Census Bureau (14), about 62 millionAmericans (24.8% of the total population) lived in rural areas; the other75.2% resided in places defined as urban. Implicit in the above compa-rison is that the definitions of rural and urban are not completely clear.For example, the definition of ‘‘rural’’ has been applied to places withpopulations of less than 20,000 as well as to populations of less than2,500 (6, 15, 16, 17); and places with 50,000 or more inhabitants are oftenreferred to as urban (18–21); according to these definitions, rural areasare areas not considered as being urban. The problem with this dichot-omy is that it leaves a large gray area often not captured when this defi-nition is used for individuals living in areas with populations with morethan 2,500 but with fewer than 50,000 inhabitants. This omission isimportant since rural residents represent about one-fourth (14) of thetotal U.S. population and one-third of the country’s poor (22). Abouta decade ago, in an effort to be more inclusive of rural populations,Leukefeld and his colleagues (23) applied a very conservative definitionto the term rural, which they termed ‘‘very rural’’ (p. 103). In their

Rural and Very Rural Drug Users 89

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

research on rural drug use based on the 1985 National Household Surveyon Drug Use, rural was defined as ‘‘unincorporated areas of under 2,500residents’’ (p. 103). More recently, Warner and Leukefeld (21) examineddifferences in substance use and treatment utilization among prisonerswithin rural populations, defining ‘‘rural’’ as places with a 1990 popu-lation of 2,500–49,999 not within an urbanized area. They defined ‘‘veryrural’’ as places with populations of less than 2,500 persons outside of acensus-defined urbanized area (p. 269).

Like the rest of the United States, rural America is multiethnic. How-ever, while heterogeneous in ethnic composition, rural Americans sharecommon characteristics, such as individualism (associated with feelingsof self-sufficiency and strong ties to family and clan), isolation (associa-ted with limited access to role models for individual and group behaviorand meeting community needs), intense religiosity (providing rigid normsand the potential for normative conflict), conservatism (a form of con-servatism that is tolerant of bizarre behavior which fits into the com-munity framework), distrust toward newcomers (anyone who mightchallenge their way of life by introducing new ideas or new technologies),and economic deprivation (a consequence of labor market trends andtechnological advances) (24). Although these characteristics representresearch barriers in rural areas, they are common to these populations.Within the context of substance abuse, issues such as isolation (as a func-tion of location) and limited employment opportunities (due to accessi-bility) may contribute to problems like mental health disorders andincreased poverty (25). To be more specific, people living in rural areasrepresent a multitude of ethnic and cultural traditions ranging fromNative Americans to Hmong tribespeople from Southeast Asia in ruralIowa to African Americans in the rural South (6). As a function of thisdiverse background, rural Amrica is characterized by varying historiesof discrimination, disadvantage, and cultural practices, which, in turn,can affect attitudes toward the use of illegal substances and alcohol.

Using data from the 1985 National Household Survey on DrugAbuse, Leukefeld and colleagues (23) pointed out that, across all agegroups, marijuana and cocaine use in ‘‘truly’’ rural areas was at lowerlevels than in large metro, small metro, or nonmetro areas. Although,they found that rural drug use was extensive, and that alcohol use in ruralareas was similar to ruban areas, they indicated that while rural residentsreported lower illegal drug use than their urban counterparts, it was at anunacceptable level (23).

Other studies have examined drug use preferences among urban andrural substance abuse treatment populations, which revealed that thedrugs of choice for clients from rural treatment programs were mari-juana, amphetamines, and sedatives, and, for urban clients, opiates (4, 5).


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Recent studies by Leukefeld and colleagues (19, 23) showed that mari-juana and sedative use was more frequently reported by rural respon-dents, and cocaine use by urban respondents. Based on these studies,alcohol use was clearly identified as the substance of choice.

The current study contributes to the limited knowledge on rural druguse with data from drug users entering treatment. Specifically, we com-pare drug use patterns between rural and very rural drug users enteringsubstance abuse treatment in Kentucky, a predominantly rural state.Besides providing a more comprehensive understanding of rural druguse in general, differentiating between rural and very rural clients couldhave important implications for identifying risk factors associated withsubstance abuse as well as for the success of substance abuse treatmentprograms in rural locations. The purpose of this study, therefore, is to:

(a) examine demographic and other selected background characteristicsbetween rural and very rural drug users admitted to substance abusetreatment;

(b) examine lifetime drug use patterns;(c) examine current drug use;(d) examine age of first drug use;(e) examine whether drug use is recognized as a problem; and(f) examine correlates of drug use behavior in order to identify potential

predisposing factors. It is hypothesized that drug users from veryrural areas would be more sheltered and the severity of involvementwith drug use would be lower than drug users from rural areas.

METHOD

Participants for this study were recruited into a Center for SubstanceAbuse Treatment (CSAT)-sponsored cooperative agreement projectentitled Treatment Outcomes and Performance Pilot Studies Enhancement(TOPPS II), from drug users admitted to publicly-funded treatment pro-grams in three regions in Kentucky. Between November 15, 1999 andJanuary 31, 2001, data were collected in face-to-face interviews by traineddata collectors using a structured questionnaire in these geographicregions of Kentucky (N ¼ 604); eastern Kentucky (n ¼ 206), south cen-tral Kentucky (n ¼ 165)—both considered rural (N ¼ 371)—and westernKentucky, which is considered more urban (N ¼ 233). Although themajority of subjects from the urban programs were admitted to residen-tial treatment (N ¼ 199, 85%), the majority of participants from the ruralsites were admitted to outpatient treatment facilities (N ¼ 273, 74%).


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

The current study includes drug users who were admitted to one ofthe 3 participating treatment centers and who agreed to participate inthe TOPPS II study. Eligibility criteria included: a) having been admittedto substance abuse treatment, b) being at least 18 years of age, c) notbeing admitted for education purposes only (e.g. DUI), and d) not beingadmitted for mental health or mental retardation treatment only. Dualdiagnosis with substance abuse was an acceptable criterion for eligibility.

Baseline data were collected using face-to-face structured interviewslasting an average of 30 minutes, with a range between 10 and 67 minutes.The ASI lite (modified) was used as the baseline questionnaire. The ASIlite was modified to meet the needs of the CSAT cooperative agreementand included measures from the Addicition Severity Index (ASI) (26, 27),the TOPPS II Core Data Items, as well as the Treatment Event Data Set(TEDS) Items. Besides demographic and other relevant backgroundinformation, data were collected on the following 6 domains: medicalstatus, employment=support status, alcohol=drug use, legal status, family=social status, and psychiatric status. Demographic locator data, includingthe city and county participants considered their ‘‘home base,’’ also wereobtained to allow for a series of follow-up interviews after their treatmentepisodes had ended.

MEASURES

Rural–Very Rural

The definition of urban and rural is somewhat ambiguous. Althoughsome define rural simply as nonmetropolitan areas, others define ruralas places with a population of less than 20,000 people, and still othersdescribe rural as areas of less than 2,500 people (14, 16, 23, 28). Recently,a case has been made for using an urban-rural continuum instead of therather crude urban-rural dichotomy, which does not seem to adequatelycapture urban-rural differences (6, 17, 18). A recognized research practicefor defining such a continuum is the application of rural-urban con-tinuum codes (29, 30), a classification scheme that distinguishes metro-politan counties by size, and nonmetropolitan counties by degree ofurbanization and proximity to metro areas. The standardized Office ofManagement and Budget (OMB) metro and nonmetro categories havebeen subdivided into 4 metro and 6 nonmetro categories, resulting in a10-part county codification (20). As Leukefeld and colleagues (19)pointed out, this differentiation has limited utility for practitioners andpolicymakers. Therefore, the intent of this study was to further definethe rural dimension. For this purpose, we applied Beale codes to county


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

of residence, which was identified by participants as their ‘‘home base,’’captured in the question: ‘‘What town, county, and state do you consideryour home base or permanent residence?’’

Because the focus of this study was to examine and describe differ-ences in substance abuse as a function of rurality, the total sample(N¼604) was reduced to those participants with a home county thatwas classified by rural-urban continuum codes (29), ranging from 6 to9 (N¼478). The selection criteria for the rural group (N¼310) werehome county codes of 6 and 7, representing a population between2,500 and 19,999; the selection criteria for the very rural group(N¼168) were home county codes of 8 and 9, representing a completelyrural population of less than 2,500. Thus, drug users at substance abusetreatment entry from rural areas (population�2,500 but <20,000) werecompared with those from very rural areas (population <2,500).

Sample

Demographic and other relevant background characteristics of the studyparticipants are summarized in Table 1, which shows that over half of theparticipants were admitted to outpatient treatment (57%), the majoritywere male (75%), White (94%), had no religious preference (51%), hadbeen referred to by the criminal justice system (64%), and had a historyof prior substance abuse treatment (55%). Fifty-four percent had either ahigh school education or above and were not employed (61%). Themedian age of the study population was 32 years of age (ranging fromage 18 to 74), and the median current nonemployment income was$280 per month (ranging from $0 to $10,000 per month).

Dependent Variables

Because most models of treatment seeking suggest that the severity of theproblem is an important outcome predictor (31–33), the number of yearsany drugs had been used, drug use in the 30 days prior to treatment entry,the age at which drug use was initiated, and problem recognition were thedependent measures used in this study.

Lifetime Drug Use

To examine lifetime drug use, participants were asked how many yearsthey had ever used one or more of 11 substances and substance categoriesas well as about how many years they drank alcohol to the point


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Table

1.Characteristics

ofaruralpopulationenteringsubstance

abuse

treatm

ent

Variables

Rural

(n¼

310)

Veryrural

(n¼

168)

Total(N

¼478)

df

v2orF

Sig

Program

modality

11.534

.216

Residential

45%

39%

43%

Outpatient

55%

61%

57%

Averageage(range:

18–74)

(medianage)

32.69�

(32)�

34.60�

(33)�

33.36�

(32)�

1�

4.130�

.043�

Male

75%

77%

75%

1.302

.583

Race=ethnicity

2�

7.053�

.029�

White

94%

94%

94%

Black

5%

2%

4%

Other

1%

4%

2%

Religiouspreference

24.807

.090

Protestant=Catholic

33%

24%

30%

Other

19%

21%

19%

None

48%

55%

51%

94

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Maritalstatus

2.308

.857

Married=remarried

27%

29%

27%

Widowed,separated,divorced

40%

40%

40%

Never

married

33%

31%

33%

Highschoolorabove

57%

49%

54%

12.178

.140

Notem

ployed

57%

�70%

�61%

�1�

7.722�

.005�

Averagecurrentnonem

ploymentincome

(range:

$0–10,000)

$440

($230)

$551

($324)

$479

($280)

11.732

.189

(median)

Admissionsuggestedbycrim

inal

justicesystem

65%

63%

64%

1.102

.750

Hadpriorsubstance

abuse

treatm

ent

58%

49%

55%

12.954

.086

Prescriptionmedicationfor

psychological=em

otionalproblems:lifetime

45%

39%

43%

11.372

.242

Haschronic

medicalproblems

31%

32%

32%

1.037

.848

� Significantatthe.05level.

95

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

of intoxication (key for intoxication: females, �2 drinks; males, �4 drinks).Lifetime drug use assessed with the question: ‘‘In your lifetime, how manyyears did you use [specific drug] at least 3 times a week?’’ This questionwas repeated for: alcohol, heroin, methadone, other opiates=analgesics,barbiturates, other sedatives=hypnotics=tranquilizers, cocaine=crack,amphetamines, marijuana=hashish=THC, hallucinogens, and inhalants.

Current Drug Use

To examine current drug use, participants were asked about their use ofthe 11 substances and substance categories listed above during the pre-vious 30 days and whether they had used multiple substances per day,including alcohol. Current drug use was assessed with the question: ‘‘Inthe past 30 days, on how many days did you use [drug]?’’

Age at First Drug Use

To examine the age of drug use initiation, participants were asked: ‘‘Howold were you when you first used [drug]?’’ This question was repeated foralcohol, opiates, tranquilizers=sedatives, cocaine=crack, stimulants, mari-juana, hallucinogens, and inhalants.

Problem Recognition

To examine problem recognition, participants were asked whether theyhad experienced alcohol and drug problems, whether these problemstroubled or bothered them, whether they thought they needed treatmentfor these problems, and how often had they recently attended self-helpgroups. Problem recognition was assessed by asking the questions:‘‘How many days in the past 30 days have you experienced a) alcohol,b) drug problems?’’; ‘‘How troubled or bothered have you been inthe past 30 days by these a) alcohol, b) drug problems?’’ (0¼not atall, 1¼slightly, 2¼moderately, 3¼considerably, 4¼extremely); ‘‘Howimportant to you now is treatment for these a) alcohol, b) drug pro-blems?’’ (0¼not at all, 1¼ slightly, 2¼moderately, 3¼considerably,4¼extremely); and ‘‘How many days have you attended self-help groups(AA=NA=CA) in the past 30 days?’’ The effects of rurality were examinedby entering a dummy variable ‘‘very rural’’ (0¼rural, 1¼very rural).Internal consistency reliability for this problem recognition index wasgood (Cronbach’s alpha¼ .77).


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Other Variables

To gain a better understanding of life circumstances that may have eitherprotective or facilitating effects on substance abuse among rural and veryrural drug users, variables related to demographic characteristics, econ-omic stresses, substance abuse treatment availability and utilization,health, and criminal justice system involvement were analyzed. Toexamine demographic characteristics variables such as age—mediansplit—(0¼younger than age 32, 1¼age 32 or older), gender (0¼ female,female, 1¼male), race (0¼White, 1¼Black), religion (0¼has no religiousaffiliation, 1¼has religious affiliation), marital status (0¼not single,1¼single, never married), and education (0¼ less than high school,1¼high school above) were included. Economic stresses includedvariables such as employment (0¼not employed, 1¼ full-or part-timeemployed) and nonemployment income. Examination of substance abusetreatment availability and utilization included variables such as treatmentmodality (0¼residential, 1¼outpatient) and substance abuse treatmenthistory (0¼had no prior substance abuse treatment, 1¼had prior sub-stance abuse treatment). Variables to examine health included chronicmedical problems (0¼no, 1¼yes) and prescribed medications regularlyfor physical and=or psychological problems lifetime (0¼no, 1¼yes).Admission suggested by criminal justice system (0¼no, 1¼yes) wasincluded to examine criminal justice system involvement. To identifythe effects of being either from rural or very rural areas, a dummy vari-able was included (rural¼0, very rural¼1).

Analytic Plan

Differences in demographic and other background characteristicsbetween rural and very rural drug users were identified using chi-squareanalyses for categorical variables and analysis of variance (ANOVA)for continuous variables. Ordinary least squares (OLS) regressions wereperformed to further explore the impact of rurality on a) substanceabuse problem recognition and b) on drug use. Dependent variableswere determined by conducting independent t-tests on all relevant sub-stance abuse variables comparing rural and very rural areas. Variablesthat produced significant differences were then entered as dependentvariables in the regression model. This was the case for methadoneuse, lifetime; other opiate use, lifetime; sedatives use, lifetime; cocai-ne=crack use, lifetime; marijuana use, lifetime; multiple drug use, life-time; treatment for drug abuse, lifetime; and attendance of self-helpgroups, last 30 days.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

RESULTS

Demographic Characteristics

Examination of demographic and other background characteristics ofparticipants from rural and very rural areas showed they were similarin terms of the proportions of program modality, gender, marital status,religious preference, education, admission suggested by the criminal jus-tice system, substance abuse treatment history, taking any prescriptionmedication for psychological=emotional problems, and chronic medicalproblems. However, African American subjects were significantly morelikely to be from rural areas (5 vs. 2%) and those of other racial=ethnicbackground from very rural areas (1 vs. 4%). Also, significantly moresubjects who were unemployed (57 vs. 70%) were from the very ruralareas. On average, subjects from very rural areas tended to be signifi-cantly older (34.60 years) than those from rural areas (32.69 years).

Substance Abuse among Rural and Very Rural Drug Users

Tables 2 and 3 summarize differences in drug use prevalence (expressed inpercent) of lifetime as well as of current (last 30 days prior to the inter-view) drug use between rural and very rural drug areas. Table 2 showsthat very rural lifetime drug users were significantly less likely to reportusing other opiates (39 vs. 27%), cocaine=crack (34 vs. 20%), marijuana(61 vs. 46%), and multiple drugs (57 vs. 42%). Furthermore, very rurallifetime drug users were marginally less likely to report using methadone(6 vs. 2%), barbiturates (9 vs. 4%), and amphetamines (23 vs. 16%).

Table 3 shows that for current drug use, participants from very ruralareas were significantly less likely to be using cocaine=crack (10 vs. 3%).Although there were no statistically significant differences between thetwo groups in the use of other types of drugs examined, it is interestingthat current alcohol, barbiturate, and inhalant use were marginally higherin very rural areas.

Differences in Age of First Drug Use among Rural and Very

Rural Drug Users

Examination of the age of first drug use, as seen in Table 4 shows that,except for inhalants, participants from very rural areas were older whenthey first used drugs. This difference in drug use initiation age was signifi-cant for those who used tranquilizers or sedatives (rural: age 20.66 vs.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

very rural: age 22.95) and for those who used cocaine=crack (rural: age20.63 vs. very rural: age 22.36).

Substance Abuse Problem Recognition among Rural and

Very Rural Drug Users

Table 5 examined substance abuse problem recognition among drug usersfrom rural and very rural areas. There was a significant relationship(b¼ .520, p< .001) between the number of days participants reported cur-rently having alcohol problems and the degree to which they weretroubled by them. The relationship was also significant (b ¼ .641,p< .001) between the number of days participants reported currently hav-ing drug problems and the degree to which they were troubled by them.However, the model revealed no significant relationships in the degreethat participants thought treatment for alcohol or drugs was important,

Table 2. Lifetime drug use: rural vs. very rural

Total(N ¼ 478)

Variables

Rural(n ¼ 310)Percentage

Veryrural

(n ¼ 168)Percentage Percentage df v2 Sig

Fisher’sexact

(2-sided)

Alcohol 81 86 83 1 1.711 .191Alcohol tointoxication

73 78 75 1 1.409 .235

Heroin 4 2 3 1 1.952 .162Methadonea 6 2 4 1 3.717 .054Other opiates 39� 27� 35� 1� 6.507� .011�

Barbituratesa 9 4 7 1 3.403 .065Other sedatives 30 23 28 1 3.235 .072Cocaine=crack 34� 20� 29� 1� 9.819� .002�

Amphetaminesa 23 16 20 1 3.426 .064Marijuana 61� 46� 56� 1� 8.922� .003�

Hallucinogens 14 9 12 1 2.496 .114Inhalantsb 3 3 3 1 .002 .964 1.000Multiple

drug use

57� 42� 52� 1� 9.450� .002�

�Significant at the .05 level.aBorderline significance.bOne cell (25%) has expected count less than 5. The minimum expected count is4.92.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Table

3.Currentdruguse:ruralvs.veryrural

Variables

Total(N

¼478)

Rural(n

¼310)

Percentage

Veryrural(n

¼168)

Percentage

Percentage

df

v2Sig

Fisher’sexact

(2-sided)

Alcohol

31

35

32

1.631

.427

Alcoholto

intoxication

24

27

25

1.649

.420

Heroin

00

0—

——

Methadonea

1.6

11

.509

.476

.661

Other

opiatesy

21

14

19

13.498

.061

Barbituratesb

24

31

.710

.399

1.000

Other

sedatives

15

12

14

1.788

.375

Cocaine=crack

10�

3�

7�

1�

7.210�

.007�

Amphetamines

54

51

.112

.738

Marijuana

27

22

25

11.142

.285

Hallucinogensc

.6.6

.61

.004

.947

1.000

Inhalantsd

0.6

.21

1.849

.174

.351

Multiple

druguse

22

19

21

1.464

.496

� Significantatthe.05level.

y Borderlinesignificance.

aTwocells(50%)haveexpectedcountless

than5.Theminim

um

expectedcountis1.76.

bOnecell(25%)hasexpectedcountless

than5.Theminim

um


c Twocells(50%

)haveexpectedcountless

than5.Theminim

um


dTwocells(50%)haveexpectedcountless

than5.Theminim

um


100

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

their recent attendance of self-help groups, or rurality. Overall, in termsof recognition of their substance abuse problems, the variation for alco-hol (27%) and drug (41%) problems was explained by the degree towhich participants were troubled by them.

Correlates of Substance Abuse among Rural and Very Rural Drug

Users at Treatment Entry

To gain a more comprehensive understanding of the impact of rurality onsubstance abuse, the variables age, gender, religion, marital status,education, employment, income, substance abuse treatment history,admission by criminal justice system, taking prescription drugs, andhaving chronic medical problems were included in a model to examinedemographic characteristics, economic stresses, substance abuse treat-ment availability and utilization, health, and criminal justice systeminvolvement. As Table 6 shows, for these rural and very rural drug users,admission to substance abuse treatment suggested by the criminal justicesystem (b¼ �.104, p< .05) and being from very rural areas (b¼ �.092,p< .05) were negative correlates for any methadone use. Being employedfull- or part-time (b¼ �.094, p < .05) and admission to substance abusetreatment suggested by the criminal justice system (b¼ � .143, p < .01)were negative correlates for any opiate use, while taking prescribed med-ications regularly for physical and=or psychological problems (b¼ .128,p < .01) was a positive correlate. Having had any prior substance abusetreatment (b¼ .117, p < .05) and taking prescribed medications regularlyfor physical and=or psychological problems (b¼ .139, p < .01) were

Table 4. Age of first drug use: rural vs. very rural

Rural (n ¼ 310) Very rural (n ¼ 168)Total (N ¼ 478)

Variables Mean Mean Mean df F Sig

Alcohol 15.34 15.76 15.49 1 1.096 .296Opiates 22.74 23.36 22.95 1 .340 .560Tranquilizers,

sedatives

20.66� 22.95� 21.45� 1� 5.647� .018�

Cocaine=crack 20.63� 22.36� 21.16� 1� 4.810� .029�

Stimulants 19.78 19.88 19.81 1 .018 .892Marijuana 15.56 15.98 15.70 1 .938 .333Hallucinogens 18.46 19.56 18.77 1 2.509 .115Inhalants 14.50 14.00 14.38 1 .170 .681

�Significant at the .05 level.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

positive correlates for any use of sedatives, and being from very ruralareas (b¼ � .094, p < .05) was a negative correlate. Having had anyprior substance abuse treatment (b¼ .186, p < .001) was a positive corre-late for any cocaine=crack use, while being from very rural areas(b¼ � .119, p < .01) was a negative correlate. Being treated as an out-patient (b¼ � .133, p < .01) was a negative correlate for any marijuanause, and being male (b¼ .123, p < .01) and having had any prior sub-stance abuse treatment (b¼ .158, p < .001) were positive correlates. Beingtreated as an outpatient (b¼�.220, p < .001) was a negative correlate forany multiple drug use, while being age 32 or older (b¼ .109, p < .05), hav-ing had any prior substance abuse treatment (b¼ .158, p < .001) and tak-ing prescribed medications regularly for physical and=or psychologicalproblems (b¼ .096, p < .05) were positive correlates. Being employedfull- or part-time (b¼�.106, p < .05), admission to substance abusetreatment suggested by the criminal justice system (b¼�.115, p < .01),

Table 5. Substance abuse problem recognition among rural and very rural drugusers

Dependent variables

Had alcoholproblems # days(last 30 days)

Had drugproblems # days(last 30 days)

Independent variables Beta Beta

Troubled by recentalcohol problemsa

.520�� .077

Troubled by recentdrug problemsa

.210 .641��

Importance of treatment forrecent alcohol problemsa

.177 �.096

Importance of treatment forrecent drug problemsa

.148 �.192

Recent attendance of self-helpgroups # of days last 30 days

�.040 .217

Very rural .067 �.017Overall model statisticsdf 1 1F 17.078 32.092Sig .000 .000R2 .271 .411

�p < .05.��p < .01.��p < .001.aCronbach’s alpha ¼ .7732.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Table6.OLSregressionstepwise:correlatesofsubstance

abuse

inasampleofruralandveryruraldrugusers

admittedto

substance

abuse

treatm

ent(N

¼461)

Variables

Methadone

use,

lifetime

Beta

Other

opiate

use,

lifetime

Beta

Sedative

use,

lifetime

Beta

Cocaine=crack

use,lifetime

Beta

Marijuana

use,

lifetime

Beta

Multiple

drug

use,

lifetime

Beta

Treatm

ent

fordrug

abuse,

lifetime

Beta

Attendance

ofself-help

groups,last

30days

Beta

Outpatient

�.133��

�.220��

�

Age32orolder

.109�

Male

.123�

Black

Hasareligious

affiliation

.105�

Single

Highschoolorabove

Isfull-orpart-tim

eem

ployed

�.094�

�.106�

Nonem

ployment

income

Admissionsuggested

bycrim

inaljustice

system

�.104�

�.143��

�.115��

HadpriorSA

treatm

ent

.117�

.186��

�.158��

�.158��

�.300��

�.146��

(Continued)

103

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Table

6.

Continued

Variables

Methadone

use,

lifetime

Beta

Other

opiate

use,

lifetime

Beta

Sedative

use,

lifetime

Beta

Cocaine=crack

use,lifetime

Beta

Marijuana

use,

lifetime

Beta

Multiple

drug

use,

lifetime

Beta

Treatm

ent

fordrug

abuse,

lifetime

Beta

Attendance

ofself-help

groups,last

30days

Beta

Takes

prescribed

medicationsregularly

forphysicaland=or

psychologicalproblems

.128��

.139��

.096�

Chronic

medical

problems

.109�

.118��

Veryrural

�.092�

�.094�

�.119��

�.114��

Overallmodel

statistics

df

23

32

34

53

F4.444

9.047

7.787

12.514

11.153

13.650

17.861

9.074

Sig

.012

.000

.000

.000

.000

.000

.000

.000

R2

.019

.056

.049

.052

.068

.108

.164

.057

� p<

.05.

��p<

.01.

�� p

<.001.

104

Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

and being from very rural areas (b¼�.114, p < .05) were negative corre-lates for having been in any treatment for drug abuse, and having hadprior substance abuse treatment (b¼ .300, p < .001) and having chronicmedical problems (b¼ .109, p < .05) were positive correlates. Having areligious affiliation (b¼ .105, p < .05), having had prior substance abusetreatment (b¼ .146, p < .01), and having chronic medical problems(b¼ .118, p < .05) were positive correlates for recent attendance of self-help groups. However, there was no relationship between race=ethnicity,icity, marital status, education, and nonemployment income and any ofthe examined drug use variables.

DISCUSSION

Study participants from the very rural areas were more likely to be olderand unemployed supports other findings (8–11, 13, 25), which refer to theeffects of economic stresses on rural populations. Overall, however, drugusers in this study from rural and very rural areas were more alike thandifferent in their sociodemographic background. Their demographiccharacteristics are similar to those described by Davidson (12) as thosefor the ‘‘new rural ghetto.’’ He maintains that since the farm crisis inthe 1980s, many rural areas have been confronted with similar problemsas inner-city areas. Citing the 1971 work of Bender, Green, and Camp-bell, he adds that an initial economic crisis, like the farm crisis of the1980s, sets a process in motion that is influenced by some of the samesocial forces that affect the inner-city, such as intergenerational poverty,class-selective migration—out-migration of more prosperous andyounger residents, leaving behind aging communities characterized bymore concentrated poverty—and an accelerated downward spiral resul-ting in ‘‘ghetto-like’’ conditions.

In Kentucky, the effect of the farm crisis was magnified because ‘‘itcame on the heels of the mining crisis’’ (21, p. 276). With farming andmining being among the main income sources in rural Kentucky, thisstudy shows some pronounced consequences of these two economicsetbacks among persons admitted to substance abuse treatment in veryrural areas. On average, these participants were older and, as might beexpected their unemployment rate was significantly higher than amongparticipants in rural areas.

In the context of overall drug use, the findings from this study aresimilar to an earlier observation by Calahan and colleagues (34) that indi-cates drug abuse in rural areas is a relatively new problem compared toalcohol use, which is older but growing. We found that any alcoholuse, including use to the point of intoxication, was higher in the very rural


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

areas. This finding could be an artifact of social norms—alcohol is a legalsubstance and readily available, while illegal drugs are relatively expens-ive (6). Illegal drug use in this study was higher in the rural areas with theexception of current barbiturate and inhalant use. Drug use in rural areasalso is reflected by age at drug use initiation with the exception of inhal-ant use, which was higher in very rural areas.

Participants were significantly troubled or bothered by their currentalcohol and drug abuse problems. Although not significant, alcohol userswere more likely to view treatment for their recent alcohol and drug pro-blems as more important than drug users. While alcohol users were morelikely to have attended fewer self-help groups than drug users, neitherrural nor very rural areas were significantly correlated with recent alcoholor drug use.

An examination of correlates of drug use and related issues, such astreatment for drug abuse and recent self-help group attendance, revealedthat drug users from very rural areas were less likely than those fromrural areas to use any methadone, sedatives, and cocaine=crack, and alsowere less likely to have received any drug abuse treatment. These negativecorrelates (see Table 2) may be an indication that being from very ruralareas is somewhat protective for certain drug use. However, the negativerelationship between very rural and treatment for drug abuse is open andwarrants additional research on the availability of treatment programs invery rural areas, their accessibility, referrals, and drug users’ problem rec-ognition. With respect to substance abuse treatment, sedative, cocaine=crack, marijuana, and multiple drug users were likely to have a substanceabuse history with repeated treatment episodes, which evokes the imageof a revolving door and begs the questions ‘‘How successful are currentsubstance abuse treatment approaches?’’ and ‘‘How can we change thestatus quo?’’ Specifically, participants who were employed, referred bythe criminal justice system, and came from very rural areas were morelikely to have had fewer treatment episodes, and those with chronic medi-cal problems were likely to have had more treatment episodes. Withrespect to any multiple drug use, we found evidence that older subjects(32 years or older), those who had a substance abuse treatment history,and took prescription drugs regularly were more likely to have used mul-tiple drugs. On the other hand, subjects who were in outpatient treatmentwere less likely to have used multiple drugs. With respect to any mari-juana use, male subjects and those who had a substance abuse treatmenthistory were more likely to have used marijuana. Similar to our findingsfor multiple drug use, participants who were in outpatient treatment wereless likely to have used marijuana.

These findings are of value to substance abuse treatment providers inrural and very rural areas because providers can use them for assessing


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

specific substance abuse treatment needs in rural locations. For example,a better understanding of the effects of ‘‘rurality,’’ not only on drug usepatterns and on substance abuse problem recognition but also on the het-erogeneity of rural populations (16, 35), will allow providers and policy-makers to improve currently existing substance abuse treatment programsand=or to develop new programs. Our study shows that further research isneeded to define specific needs so theymay be implemented in new treatmentapproaches. Particularly, ‘‘successful treatment’’ may need to be redefined.Completionof repeated treatment episodes doesnot fit the bill. Policymakersneed to take notice and be willing to act on the fact that one treatmentapproach may be too narrow when funding dollars are awarded.

There are limitations associated with this study. Because the sampleof this study was not randomly selected and data were collected only inone state—Kentucky—generalizability is limited. However, firm adher-ence to the research protocol greatly reduced, if not eliminated, anychance for bias, and our findings can be applied to other rural places.The data presented here were self-reported; notwithstanding the fact thatsubjects consented to the study and were assured strictest confidentiality,we were not able to validate the self-reported data. As a group, drug usersadmitted to substance abuse treatment in rural facilities may not be rep-resentative of all substance abuse treatment seekers.

In conclusion, our findings indicate that being from very rural areasmay have some sheltering properties, which is indicated by lower currentdrug use and the generally older age at drug use initiation.

Although there were no differences in alcohol use among subjectsfrom both rural and very rural areas, alcohol is the most common life-time and currently used substance. Cocaine use, on the other hand,was three times higher in rural areas than in very rural areas. However,it is important to point out that, while not markedly different, illegaldrug use remains at unacceptable high levels in both areas. More thana decade ago, Leukefeld and colleagues (23) reported similar findingsand little has changed since. However, participants in this study recog-nized they had substance abuse problems and were considerably both-ered by them. Interestingly, participants were more bothered by theirdrug-related problems than by their alcohol-related problems, which isanother indication that alcohol, despite that it is the most used substancein both rural areas, is thought to be less problematic. These findings areimportant because they contribute to the very limited body of literaturethat has examined substance abuse differences among rural populations.With a better understanding of the effects of ‘‘rurality’’ on different druguse patterns and on substance abuse problem recognition, providers andpolicymakers will be able to use the data to improve currently existingsubstance abuse treatment programs and=or to develop new programs.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

REFERENCES

1. Conger RD, Elder GH Jr. Families in Troubled Times: Adapting to Changein Rural America. New York: Aldine de Gruyter, 1994.

2. O’Dea PJ, Murphy B, Balzer C. Traffic and illegal production of drugs inrural America. In Rural Substance Abuse: State of Knowledge and Issues.Robertson EB, Sloboda Z, Boyd GM, Beatty L, Kozel NJ, eds. NIDAMonograph 168; Rockville, MD: U.S. Department of Health and HumanServices, National Institute on Drug Abuse, 1997; 79–89.

3. Wagenfeld MO, Murray JD, Mohatt DF, DeBryn JD. Mental Health andRural America: 1980–1993. An Overview and Annotated Bibliography;NIH Pub. No. 94-3500. Washington, DC: Public Health Service, 1994.

4. Brown BS, Voskuhl TC, Lehman PE. Comparison of drug abuse clients inurban and rural settings. Am J Drug Alcohol Abuse 1977; 4(4):445–454.

5. NASADAD Task Force on Federal=State Planning. Priorities and Opport-unities for Research Related to the Treatment and Prevention of Alcoholand Other Drug Abuse and Dependency. Washington, DC: National Associ-ation of State Alcohol and Drug Abuse Directors, September 1991.

6. CongerRD.The special natureof ruralAmerica. InRural SubstanceAbuse: Stateof Knowledge and Issues. Robertson EB, Sloboda Z, Boyd GM, Beatty L,Kozel NJ, eds. NIDA Monograph 168; Rockville, MD: U.S. Departmentof Health and Human Services, National Institute on Drug Abuse, 1997; 37–52.

7. O’Hare WP, Curry-White B. The Rural Underclass: Examination ofMultiple-Problem Populations in Urban and Rural Settings. PopulationReference Bureau. Louisville, KY: University of Louisville, 1992.

8. Brody GH, Stoneman Z, Flor D, McCrary C, Hastings L, Conyers O. Finan-cial resources, parent psychological functioning, parent co-parenting, andearly adolescent competence in rural two-parent African-American families.Child Dev 1994; 65:590–605.

9. Conger RD, Ge X, Elder GH, Lorenz FO, Simmons RL. Economic stress,coercive family process and developmental problems of adolescents. ChildDev 1994; 65:541–561.

10. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshelman S,Wittchen H, Kendler KS. Lifetime and 12-month prevalence of DSM-III-Rpsychiatric disorders in the United States: results from the national comor-bidity study. Arch Gen Psychiatry 1994; 51:8–19.

11. McLoyd VC, Jayaratne TE, Ceballo R, Borquez J. Unemployment and workinterruption among African American single mothers: effects on parentingand adolescent socioemotional functioning. Child Dev 1994; 65:562–589.

12. Davidson OG. Broken Heartland: The Rise of America’s Rural Ghetto. IowaCity: University of Iowa Press, 1996.

13. Wilson WJ. The Truly Disadvantaged: The Inner City, the Underclass, andPublic Policy. Chicago: University of Chicago Press, 1996.

14. Bureau of the Census. Statistical Abstract of the United States, 1993: TheNational Data Book.Washington, DC: U.S. Department of Commerce, 1993.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

15. Dahmann DC, Dacquel LT. Residents of Farms and Rural Areas: 1990.Current Population Reports, Population Characteristics, Series P-20, No.457, U.S. Department of Agriculture, Economic Research Service and U.S.Department of Commerce, Bureau of the Census, Washington, DC, 1992.

16. Edwards RW. Drug and alcohol use among youth in rural communities.In Rural Substance Abuse: State of Knowledge and Issues. Robertson EB,Sloboda Z, Boyd GM, Beatty L, Kozel NJ, eds. NIDA Monograph 168;Rockville, MD: U.S. Department of Health and Human Services, NationalInstitute on Drug Abuse, 1997; 52–78.

17. Patton H. Setting the rural health services research agenda: the congressionalperspective. Health Serv Res 1989; 23:1005–1013.

18. Hewit M. Defining ‘rural’ areas: impact on health care policy and research.Unpublished staff paper prepared for the Congressional Office of Tech-nology Assessment. Washington, DC, 1989.

19. Leukefeld CG, Narevic E, Hiller ML, Gillespie W, Webster JM, Staton K,Shaw K, Purvis R. Alcohol and drug use among rural and urban incarceratedsubstance abusers. Int J Offender Therapy Comp Crim 2002; (46):715–728.

20. Rickets TC, Johnson-Webb KD, Taylor P. Definitions of Rural: A Hand-book for Health Policy Makers and Researchers. Bethesda, MD: FederalOffice of Rural Health Policy, 1998.

21. Warner BD, Leukefeld CG. Rural-urban differences in substance use andtreatment utilization among prisoners. Am J Drug Alcohol Abuse 2001;27(2):265–280.

22. Rural Clearinghouse. National adult literacy survey overlooks rural illit-eracy. Rural Clearinghouse Digest on Rural Literacy 1993; 1(1):1–8.

23. Leukefeld CG, Clayton RR, Myers JA. Rural drug and alcohol treatment.Drugs Soc 1992; 7(1–2):95–116.

24. Beltrame TF. Meeting the special needs of Appalachian alcoholics. HospCommunity Psychiatry 1978; 29(12):792–794.

25. Wagenfeld MO, Wagenfeld JK. Values, culture and the delivery of mentalhealth services in rural areas. In Perspectives on Rural Mental Health(New Directions for Mental Health Services Series, No. 9). WagenfeldMO, ed. San Francisco: Jossey-Bass, 1981; 1–12.

26. McLellan AT, Luborsky L, O’Brien CP, Woody GE. An improved diagnos-tic instrument for substance abuse patients: the Addiction Severity Index. JNerv Ment Dis 1980; 168:26–33.

27. McLellan AT, Kushner H, Metzger D, Peters F, Smith I, Grissom G, Petti-nati H, Argeriou M. The fifth edition of the Addiction Severity Index. J SubstAbuse Treat 1992; 9:199–213.

28. Robertson EB, Donnermeyer JF. Illegal drug use among rural adults: mentalhealth consequences and treatment utilization. Am J Drug Alcohol Abuse1997; 23(3):467–484.

29. Beale C, Johnson K. Non-metropolitan populations continues post-1990rebound. Rural Cond Trends 1995; 6:2.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

30. Butler MA, Beale CA. Rural-urban continuum codes for metropolitan andfor nonmetropolitan counties, 1993. Agriculture and Rural EconomyDivision, Economic Research Service, U.S. Department of Agriculture,1994. Available online: http://www.ers.usda.gov/Briefing/Rurality/RuralUrbCon/

31. Andersen RM. Revisiting the behavioral model and access to medical care:Does it matter? J. Health Soc Behav 1995; 36:1–10.

32. Anderson RM, Newman JF. Societal and individual determinants of medicalcare utilization in the United States. Milbank Mem Fund Q Health Soc 1973;51:95–124.

33. Hartnoll R. Research and help-seeking process. Br J Addict 1992; 87:429–437.

34. Calahan D, Cisin RI, Crossley HM. American Drinking Practices. New Haven,CT: College and University Press, 1969.

35. Oetting ER, Edwards RW, Kelly K, Beauvais F. Risk and Protective Factorsfor Drug Use Among Rural American Youth. In Rural Substance Abuse:State of Knowledge and Issues. Robertson EB, Sloboda Z, Boyd GM, BeattyL, Kozel NJ, eds. NIDA Monograph 168; Rockville, Maryland. U.S.Department of Health and Human Services, National Institute on DrugAbuse. 1997; 90–130.


Am

J D

rug

Alc

ohol

Abu

se D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/04/

14Fo

r pe

rson

al u

se o

nly.

Substance Abuse among Rural and Very Rural Drug Users at Treatment Entry*

Documents

Transcript of Substance Abuse among Rural and Very Rural Drug Users at Treatment Entry*