Journal of Rural Community Psychology, Vol. E3(1), 2000


Rural-Urban Differences in
Service Use by At-Risk Drinkers
 
 

Kathryn Rost, Ph.D.
JoAnn Kirchner, M.D.
John C. Fortney, Ph.D.
Brenda M. Booth, Ph.D.
 
 
 

 Abstract

Goals and Research Question: The purpose of the study was:  (1) to determine the health care settings in which modestly elevated treatment rates for rural at-risk drinkers may occur, and (2) to identify those sectors which should be targeted for improved detection and management of at-risk drinkers.

Data and Methods: This cross-sectional study used a brief telephone screening interview administered to over 12,000 respondents in six Southern states to identify a sample of 960 at-risk drinkers, 733 (76.4%) of whom reported on retrospective use of services for alcohol and other mental health problems (SA/MH) during the previous 6-months.

Results: Rural at-risk drinkers were significantly more likely to get SA/MH treatment from a general medical provider than their urban counterparts (6.5% versus 3.0%, p=.04).  There was slight evidence that rural at-risk drinkers were more likely than their urban counterparts to get SA/MH treatment in an emergency room (2.4% versus 0.5%, p=.09).  Specialty care SA/MH use was comparable for rural and urban at-risk drinkers, 4.3% versus 4.7% respectively (p=.60).   Rural and urban at-risk drinkers also visited general medical providers (42.9% versus 48.1% respectively) and emergency rooms (14.4% versus 13.5% respectively) for physical problems at comparable rates.

Conclusions:   Rural drinkers make general medical visits for alcohol and mental health problems more often than their urban counterparts. Because almost half of at-risk drinkers from both rural and urban communities make office visits to general medical providers over a 6 month period, focused attention should be given to improving both the detection and management of alcohol-related problems in primary care.
 
 

 Introduction

Recent community studies report that 10-11% of community residents meeting DSM criteria for one-year alcohol disorder receive some treatment for alcohol problems over one year (Grant, 1996; Booth, Kirchner, Fortney, Ross, & Rost, 1999; Bailey, Wilson, Weiss, & Windsor, 1994).  Contrary to expectation, rural drinkers appear more likely to receive any alcohol treatment than their urban counterparts (Booth, Kirchner, Fortney, Ross, & Rost, 1999), although low use rates reduce power to demonstrate that three-fold differences in use of a particular sector are statistically significant.  Because an additional 6-7% of community residents meeting DSM criteria for one-year alcohol disorder receive treatment for mental health problems other than alcohol (Regier, Narrow, Rae, Manderscheid, Locke, & Goodwin, 1993) which potentially impact outcomes, we chose to explore sector differences by analyzing rural-urban differences in at-risk drinkers’ use of care for mental health problems including but not limited to alcohol (SA/MH) treatment.

We hypothesized that rural residents would be significantly more likely to receive SA/MH treatment in general medical settings than their urban counterparts (Rost, Owen, Smith, & Smith, Jr., 1998), but that rates of SA/MH treatment in specialty care settings would be comparable (Rost, Owen, Smith, & Smith, Jr., 1998; Rost, Zhang, Fortney, Smith, & Smith, Jr., 1998; Booth, Kirchner, Fortney, Ross, & Rost, 1999).  We also examined at-risk drinkers’ use of services for physical problems to determine sectors which should be targeted to improve detection of at-risk drinking.
 
 

 Methods

We conducted brief screening telephone interviews with a stratified random sample of community-residing individuals to identify a cohort of at-risk drinkers who were recruited into a longitudinal study extensively described in earlier publications (Booth, Ross, & Rost, 1999; Booth, Kirchner, Fortney, Ross, & Rost, 1999).  A probability based random sample of phone numbers in Arkansas, Alabama, Georgia, Louisiana, Mississippi, and Tennessee designed to oversample rural residents was identified using a list-assisted random-digit dialing design.

Once an adult in the household was contacted, the designated respondent for the screening interview was identified as the household member with the most recent birthday in the preceding 12 months. When contacted, the designated respondent was asked to participate in a brief health survey being conducted by the University of Arkansas. Of 25,135 eligible respondents, 12,348 (49%) completed screening.  Another 4,766 eligible respondents (19%) could not be screened because the eligible respondent could not be contacted.  The remaining 8,021 (32%) refused to participate in screening.  An intensive follow-up with a sample of 100 screening refusals found no differences in at-risk drinking between individuals who completed and initially refused screening.

Overall, 8% of screening participants (960/12,348) met criteria for at-risk drinking and were requested to participate in the study using enrollment procedures approved by the Human Research Advisory Committee of the University of Arkansas for Medical Sciences.   Seven hundred thirty-three of 960 (76%) at-risk drinkers agreed to participate in the study.   Participants reported a greater likelihood of making an SA/MH visit in the past 6 months than non-participants (8% versus 4%, p=.04) and lower emotional role functioning (80 versus 86, p=.01) than non-participants.  Participants did not differ from non-participants in rural residence, gender, education, age, race, income, marital status, health insurance, current alcohol consumption, physical functioning, or recent general medical visits. Participants were remunerated $50 for completing the study interview.

Operational Definitions of Major Constructs in the Study

At-Risk Drinking:  Individuals were defined as at-risk if they reported at least one of the lifetime DSM-IV criteria for alcohol abuse or dependence AND at least one of the following within the past 12 months: (a) any abuse or dependence criteria; (b) significant binge drinking defined as at least 12 drinks on an occasion for men (reduced to 8 drinks for men reporting 2 or more lifetime DSM-IV criteria) or 8 drinks for women;  or (c) frequent heavy drinking defined as at least 5 drinks for men or at least 3 drinks for women in a typical drinking day at least once a week in the past 12 months or on at least 21 out of the previous 28 days.  At-risk drinkers had a 90% probability of meeting CIDI-SAM (Cottler, Robins, & Helzer, 1989) criteria for lifetime alcohol use disorder and a 66% probability for one year alcohol use disorder (Booth, Ross, & Rost, 1999).

Independent Variable

Rurality – Rurality was defined as current residence in a non-metropolitan statistical area (MSA) as defined by the 1994 census.  This definition was chosen as a proxy indicator of distance to SA/MH care because non-MSAs have significantly fewer mental health specialists per capita than MSAs.  While non-MSAs also have fewer general medical physicians per capita than MSAs, the differences are generally not as great (Holzer, Goldsmith, & Ciarlo, 1998).

Dependent Variable

Service Use – We used three questions in the baseline interview adapted from the Epidemiological Catchment Area (ECA) study to measure use of services for problems with emotions, nerves, drugs, alcohol, or mental health from: (a) a general medical professional, (b) an emergency room, and  (c) a mental health professional. We used parallel questions to measure use of services for physical problems from: (d) a general medical professional and (e) an emergency room.

Covariates

In order to rule out whether observed rural-urban differences reflect sociodemographic or need differences between rural and urban populations, we tested a comprehensive group of covariates from previous research. Sociodemographic covariates included age, minority status, gender, education (high school graduate versus not), marital status (currently married versus not), annual family income, employment status (employed full-time versus not), and health insurance (any versus none).  Need covariates investigated as predictors of use in previous studies (Booth, Kirchner, Fortney, Ross, & Rost, 1999; Hingson, Mangione, Meyers, & Scotch, 1982; Weisner, 1993; Room, 1989; Kaskutas, Weisner, & Caetano, 1997; Weisner, 1990; Booth, Yates, Petty, & Brown, 1991; Hasin & Grant, 1995; Grant, 1996; Burton & Williamson, 1995) included alcohol problems, comorbid mental health disorders, comorbid physical disorders, social support, and life events.

Alcohol problems. Alcohol problems were defined as: (1) meeting DSM-IV criteria for diagnosis of alcohol abuse or dependence during the past 6 months on the CIDI-SAM (Cottler, Robins, & Helzer, 1989) and (2) the number of social problems the respondent attributed to their drinking during the past 6 months (Rost, Ross, Humphrey, Frank, Smith, & Smith, 1996).  Previous alcohol treatment was defined by respondent report of one or more episodes of alcohol treatment occurring one year or more before the interview to identify subjects who had treatment previous to any current treatment.

Comorbid mental health conditions. Major depression, generalized anxiety, and panic and drug disorders during the past 6 months were measured by the Diagnostic Interview Schedule, DSM-III-R version (Robins, Helzer, Croughan, & Ratcliff, 1981) because the DSM-IV version was not available at the time the study was fielded.

Comorbid physical conditions. This construct was measured by the sum of 18  subject-reported chronic conditions including diabetes, kidney, stomach, and liver disease.

Social support. This construct was measured by a 19-item scale of current social support originally developed for the Medical Outcomes Study (Sherbourne & Stewart, 1991), which demonstrated internal consistency of 0.95 in these subjects.

Stressful life events. This construct was measured by summing respondent reports on 20 life events during the past 6 months (Williams, Ware, & Donald, 1981).

Data Analysis

All analyses were conducted using weights developed to adjust for the oversampling of rural residents and to adjust for non-response so that the geographic distribution of the sample matched the population distributions in each of the six states in age, gender, race, and rurality.  Numerical data reported here in tables and the text are unweighted data but the associated inferential analyses were conducted with the appropriate weights.  The initial analyses compared rural-urban differences in service use without controlling for sociodemographic and clinical covariates.  The subsequent analyses compared rural-urban differences in service use controlling for the sociodemographic and need covariates listed above using PROC LOGISTIC in SAS.

Results

The 733 individuals completing the interview had an average age of 31.9 years (SD=10.6), were 50.3% rural, 32.6% female, 20.0% minority, 38.7% currently married, 85.2% high school graduated, 65.2% full-time employed, and 71.8% health insured with a mean income of $33,355 (SD=$16,969).  During the past 6 months, 46.4% of respondents met criteria for alcohol disorder with 40.1% attributing one or more social problems to their drinking; 11.3% met criteria for current drug disorder, 12.7% met criteria for major depression, 2.0% met criteria for generalized anxiety disorder, and 3.4% met criteria for panic disorder.

Respondents reported an average of 1.1 (SD=1.4) chronic physical problems; 21.3% of respondents reported treatment for alcohol problems one year or more ago.

Respondents reported an average of 4.3 (SD=2.9) life events during the past 6 months; social support averaged 4.1 (SD=0.9) on a five point scale.   Compared to their urban counterparts, rural respondents were less likely to be high school graduates (80.5% versus 90.1%, p<.05) and more likely to meet criteria for alcohol disorder during the previous 6 months (50.4% versus 42.3%, p<.05); rural respondents also reported more chronic physical problems (1.3 versus 1.0, p<.05) and lower incomes ($30,537 versus $36,236, p<.05).

Although there were no differences in seeing a general medical provider for physical problems, Table 1 demonstrates that rural respondents were significantly more likely to get SA/MH treatment from a general medical provider than their urban counterparts (6.5% versus 3.0%, p=.04).

Rural respondents were only somewhat more likely to get SA/MH treatment in an emergency room (2.4% versus 0.5%, p=.09), although rural and urban patients visited emergency rooms for physical problems at comparable rates.

Specialty care SA/MH use was comparable for rural and urban respondents, 4.3% versus 4.7% respectively.  Rural and urban at-risk drinkers also visited general medical providers (42.9% versus 48.1% respectively) and emergency rooms (14.4% versus 13.5% respectively) for physical problems at comparable rates.

Table 1.  Rural-Urban Differences in At-Risk Drinkers’ Use of Services for Alcohol and Other Mental Health
               Problems (SA/MH) During the Past Six Months
 
Reason for Vist Rural (n = 369) Urban (n = 364)
Any General Medical Visits 
For Physical Problems
42.9% 48.1%
Any General Medical Visits 
For SA/MH *
 6.5%  3.0%
Any Emergency Room Visits 
For Physical Problems
14.4% 13.5%
Any Emergency Room Visits 
For SA/MH +
 2.4%  0.5%
Any Specialty Care Visits 
For SA/MH
 4.3%  04.7%

Estimate unweighted for stratified sampling and non-participation.  Rural-urban comparisons based on weighted data.
+ p < .10
*p < .05

In multivariate analyses, rural and urban residents used general medical settings, emergency rooms, and specialty care for SA/MH problems at statistically comparable rates. The only significant predictor of getting help from a general medical provider for SA/MH problems was the number of chronic physical problems the subject reported (OR=1.5, p=.005).

Similarly, the number of chronic physical problems significantly predicted whether subjects got help for SA/MH problems from an emergency room (OR=2.1, p=.04);  current panic disorder was an unusually strong predictor of SA/MH services from an emergency room as well (OR=312.1, p=.001).

Respondents with current depression (OR=9.0, p=.0001) and with current panic (OR=7.2, p=.01) disorder were more likely to get SA/MH help from a specialty care setting, as were minorities (OR=2.9, p=.02) and the unemployed (OR=3.2, p=.01).

Table 2.  Predictors of At-Risk Drinkers’ Use of Services for Alcohol and Other Mental Health Problems (SA/MH) During the Past Six Months (n = 733)
 
 

Discussion

This study found that rural drinkers were significantly more likely to use general medical services for alcohol and other mental health problems and somewhat more likely to use emergency room services for these problems than their urban counterparts; however, these differences disappear in models controlling for sociodemographic and clinical covariates, particularly the greater number of physical comorbidities rural drinkers report.  Rural and urban at-risk drinkers report a comparable likelihood of specialty care for alcohol and other mental health problems in bivariate and multivariate models.

Primary care and emergency room physicians may be more successful ‘detecting’ the drinking problems in rural patients presenting with physical problems because the dense social networks in rural areas make it hard for a patient to keep drinking problems ‘private’.  Previous studies have found that bipolar disordered rural residents are also more likely than their urban counterparts to use general medical services for mental health despite having comparable rates of specialty care use (Rost, Owen, Smith, & Smith, Jr., 1998).  In contrast, rural and urban depressed patients report comparable rates of depression treatment in both general medical and specialty care settings (Rost, Zhang, Fortney, Smith, & Smith, Jr., 1998).

We found that an estimated 47% of rural at-risk drinkers visit a general medical provider over a 6-month period, including 6% who currently receive some services for alcohol or other mental health problems.  This suggests that interested rural policy analysts should focus their efforts at improving the detection and management of over 40% of rural at-risk drinkers whose alcohol consumption is not addressed during their medical visit, as well as improving the management of the 6% of rural drinkers who currently get help for some emotional problem in this sector.

Because drinkers are unlikely to seek help for alcohol problems before they become severe (Booth, Kirchner, Fortney, Ross, & Rost, 1999), brief interventions shown to improve alcohol outcomes (Fleming, Barry, Manwell, Johnson, & London, 1997; WHO Brief Intervention Study Group, 1996; Wallace, Cutler, & Haines, 1988) need to be actively disseminated to rural primary care settings.

Because many at-risk drinkers have co-occurring physical problems, it may be difficult for rural primary care providers who see even more patients per day than their urban counterparts (Rost, Humphrey, & Kelleher, 1994) to adequately address both physical and SA/MH problems at the same visit (Rost, Nutting, Smith, Coyne, Rubenstein, & Cooper-Patrick, 1999).  Thus, the feasibility and ultimate effectiveness of incorporating brief interventions in rural primary care settings may depend in large measure on using office nurses in combination with closely scheduled repeat visits to successfully intervene with physically ill patients whose drinking also needs to be addressed.   Similar approaches to improve the detection and management of major depression have proved particularly successful for rural primary care practices (Rost, Nutting, Smith, Werner, & Duan, in review; Smith, Rost, Nutting, Elliott, & Duan, in review).   Policymakers should also consider the potential of improving detection and successful referral of alcohol problems in rural emergency room settings where 15% of rural drinkers make visits over a 6-month period.   While these interventions will no doubt be complicated by acute physical problems, drinkers seeking emergency room services for alcohol-related injuries may have increased motivation to reduce their drinking (Cherpitel, 1995).
While the overall low rate of SA/MH service use by at-risk drinkers is of concern, it is encouraging that clinical factors (comorbid physical and psychiatric illness) predict use of SA/MH services to a greater extent than sociodemographic factors.  This suggests that scarce treatment resources are being distributed more on the basis of clinical need than on social factors. The one exception to this pattern is that minorities and the unemployed are more likely to receive SA/MH services in the specialty care setting, which may reflect the special targeting of publicly funded treatment programs to these generally disadvantaged groups.

We wish to acknowledge several limitations of the findings we report here.  First,  although there is evidence that respondents can accurately report care they received during the previous 6 months (Golding, Gongla, & Brownell, 1988), self-report of service use does introduce measurement error, particularly for hospitalization (Fortney, Rost, & Warren, 1999);  however, there is no evidence that this measurement error biases rural-urban comparisons.   Second, because the widely-used service use items we analyzed in this manuscript do not differentiate between treatment for alcohol and treatment for other substances or mental health problems, it is not possible to know precisely how many at-risk drinkers received help for their drinking.  While we agree that at-risk drinking needs to be addressed, we note that effective treatment for co-occurring mental health problems may help respondents reduce their drinking, particularly those that are drinking to self-medicate untreated mental health problems.  Third, metropolitan/nonmetropolitan comparisons of service use rates ignore the fact that residents of nonmetropolitan counties adjacent to metropolitan areas have a greater availability of service providers than more remote residents.  More refined analyses are needed to determine how travel distance to general medical and specialty providers predicts service use (Fortney, Rost, & Warren, 1999).  Fourth, the generalizability of our conclusions to rural America is admittedly limited because (1) the study focused on six Southern states, (2) 50% of residents refused to participate in initial telephone screening (although non-screened individuals were NOT more likely to meet criteria for at-risk drinking), (3) our methodology precluded the recruitment of individuals residing in an estimated 8% of households without telephones, presumably a more disadvantaged population, and (4) participants were twice as likely to have received recent SA/MH treatment than non-participants.  These limitations are balanced however by a high participation rate of identified eligibles from a socioeconomically and ethnically diverse group of at-risk community drinkers.  Furthermore, the large sample size gives the study power to find even small differences; for example, the sample size provides 82% power to find a significant rural-urban difference in specialty care use if the ‘true’ rural rate was 4% and the ‘true’ urban rate was 10%.
In summary, rural drinkers make general medical visits for alcohol and mental health problems more often than their urban counterparts. Because almost half of at-risk drinkers from both rural and urban communities make office visits to general medical providers over a 6-month period, focused attention should be given to improving both the detection and management of alcohol-related problems in primary care.
 
 

References

   Bailey, W.C., Wilson, S.R., Weiss, K.B., & Windsor, R.A. (1994).  Measures for use in asthma clinical research - Overview of the NIH workshop.  American Journal of Respiratory Critical Care Medicine,  149, S1-S8.

   Booth, B.M., Kirchner, J., Fortney, J., Ross, R., & Rost, K. (1999).  Rural at-risk drinkers:  Correlates and one-year use of alcohol services.  Journal of Studies on Alcohol, in press.

   Booth, B.M., Ross, R.L., & Rost, K. (1999).  Rural and urban problem drinkers in six southern states.  Substance Use and Misuse, 34, 471-491.

   Booth, B.M., Yates, W.R., Petty, F., & Brown, K. (1991).  Patient factors predicting early alcohol-related readmissions for alcoholics: Role of alcoholism severity and psychiatric co- morbidity.  Journal of Studies on Alcohol,  52, 37-43.

   Burton, T.L., & Williamson, D.L. (1995).  Harmful effects of drinking and the use and perceived effectiveness of treatment. Journal of Studies on Alcohol,  56, 611-615.

   Cherpitel, C.J. (1995).  Analysis of cut points for screening instruments for alcohol problems in the emergency room. Journal of Studies on Alcohol,  56, 695-700.

   Cottler, L.B., Robins, L.N., & Helzer, J.E. (1989).  The reliability of the CIDI-SAM: A comprehensive substance abuse interview. British Journal of Addiction,  84, 801-814.

   Fleming, M.F., Barry, K.L., Manwell, L.B., Johnson, K., & London, R. (1997).  Brief physician advice for problem alcohol drinkers: A randomized controlled trial in community-based primary care practices.  Journal of the American Medical Asociation, 277, 1039-1045.

   Fortney, J., Rost, K., & Warren, J. (1999).  Comparing alternative methods of measuring geographic access to health services.  Health Services and Outcomes Research Methodology, in press.

   Golding, J.M., Gongla, P., & Brownell, A. (1988).  Feasibility of validating survey self-reports of mental health service use.  American Journal of Community Psychology,  16, 39-51.

   Grant, B.F. (1996).  Toward an alcohol treatment model: A comparison of treated and untreated respondents with DSM-IV alcohol use disorders in the general population.  Alcoholism: Clinical and Experimental Research,  20, 372-378.

Hasin, D.S., & Grant, B.F. (1995).  AA and other helpseeking for alcohol problems:  Former drinkers in the U.S. general population. Journal of Substance Abuse,  7, 281-292.

   Hingson, R., Mangione, T., Meyers, A., & Scotch, N. (1982).  Seeking help for drinking problems: A study in the Boston metropolitan area.  Journal of Studies on Alcohol,  43, 273-288.

   Holzer, C.E., Goldsmith, H.F., & Ciarlo, J.A. (1998). The availability of health and mental health providers by population density: Letter to the field no. 11.  Frontier Mental Health Services Resource Network.

   Kaskutas, L.A., Weisner, C., & Caetano, R. (1997).  Predictors of help seeking among a longitudinal sample of the general population, 1984-1992.  Journal of Studies on Alcohol,  58, 155-161.

   Regier, D.A., Narrow, W.E., Rae, D.S., Manderscheid, R.W., Locke, B.Z., & Goodwin, F.K. (1993).  The de facto US mental and addictive disorders service system: Epidemiologic Catchment Area prospective 1-year prevalence rates of disorders and services.  Archives of General Psychiatry,  50, 85-94.

   Robins, L.N., Helzer, J.E., Croughan, J., & Ratcliff, K.S. (1981).  National Institute of Mental Health Diagnostic Interview Schedule.  Archives of General Psychiatry,  38, 381-389.

   Room, R. (1989).  The U.S. general population's experiences of responding to alcohol problems.  British Journal of Addiction,  84, 1291-1304.

   Rost, K., Humphrey, J., & Kelleher, K. (1994).  Physician management preferences and barriers to care for rural patients with depression.  Archives of Family Medicine,  3, 409-414.

   Rost, K.M., Nutting, P., Smith, J., Coyne, J., Rubenstein, L., & Cooper-Patrick, L. (1999).  The role of competing demands in the treatment provided primary care patients with major depression. Archives of Family Medicine, in press.

   Rost, K.M., Nutting, P., Smith, J., Werner, J., & Duan, N. (In review).  Primary care intervention improves depression outcomes: Results of a randomized trial.

   Rost, K.M., Owen, R.R., Smith, J., & Smith, G.R., Jr. (1998).  Rural-urban differences in service use and course of illness in bipolar disorder.  Journal of Rural Health,  14, 36-43.

   Rost, K.M., Ross, R.L., Humphrey, J., Frank, S., Smith, J., & Smith, G.R. (1996).  Does this treatment work: Validation of an outcomes module for alcohol dependence.  Medical Care,  34, 283-294.

   Rost, K.M., Zhang, M., Fortney, J., Smith, J., & Smith, G.R., Jr. (1998).  Rural-urban differences in depression treatment and suicidality.  Medical Care,  36, 1098-1107.

   Sherbourne, C.D., & Stewart, A.L. (1991).  The MOS social support survey.  Social Science and Medicine,  32, 705-714.

   Smith, J., Rost, K., Nutting, P., Elliott, C., & Duan, N. (In review).  Assessing a primary care intervention's impact on the process of care for depression in rural versus urban practice settings.

   Wallace, P., Cutler, S., & Haines, A. (1988).  Randomised controlled trial of general practitioner intervention in patients with excessive alcohol consumption.  British Medical Journal,  297, 663-668.

   Weisner, C. (1990).  The role of alcohol-related problematic events in treatment entry.  Drug and Alcohol Dependence,  26, 93-102.

   Weisner, C. (1993).  Toward an alcohol treatment entry model: A comparison of problem drinkers in the general population and in treatment.  Alcoholism: Clinical and Experimental Research,  17, 746-752.

   WHO Brief Intervention Study Group. (1996).  A cross-national trial of brief interventions with heavy drinkers.  American Journal of Public Health,  86, 948-955.

   Williams, A.W., Ware, J.E., & Donald, C.A. (1981).  A model of mental health, life events, and social supports applicable to general populations.  Journal of Health and Social Behavior,  22, 324-336.

All authors are investigators in the Centers listed above.  Dr. Rost is an investigator on the Rural Alcohol Study (RAS) and has previously examined rural-urban differences in use of services for depression and bipolar disorder.  Dr. Kirchner is Project Director on the RAS and has a special interest in depressed at-risk drinkers.  Dr. Kirchner is supported by a Department of Veterans Affairs Health Services Research Development Career Development Award (CD97-308.A).  Dr. Fortney is an investigator on the RAS and has previously used geographic information systems to examine how travel distance impacts use of services for depression in rural populations.  Dr. Booth is principal investigator of the RAS and has previously investigated how social support influences treatment for alcohol problems in a rural substance abuse treatment program.
 

Correspondence should be addressed to: Kathryn Rost, Ph.D., 5800 W. 10th St, Suite 605, Little Rock, AR 72204; phone (501) 660-7500, fax (501) 660-7542, e-mail rostkathrynm@exchange.uams.edu. This project was funded by grants AA10372, MH54444, and MH48197.

Acknowledgements

We thank Fred Licari at the Institute for Survey Research at Temple University; Stacy Kimbrel, Susan Moore, Carl Elliott, and Margaret White at the University of Arkansas for Medical Sciences; and the men and women who participated in this study.