Journal of Rural Community Psychology Volume E6 Number 2 Fall 2003
The Relationship Between Depression and
Alcohol / Substance Use in the Rural Midwestern United States
Jason M. Wise, Roselyn E. Miller, & Donald. W. Preussler
The purpose of this study was to determine if there is a relationship between depression and alcohol/substance use in a rural population in the Midwestern United States. Participants included 1,191 adults (age 18-65) assessed and treated by the Farm Resource Center (FRC) between 1995-2000. For purposes of this study, demographic and assessment data collected by FRC was analyzed. The participants included farmers, miners, and other rural individuals and families served by FRC. A reliable depression scale was formed from the participants’ pre-treatment assessment symptom checklist responses. In a case control design, those who reported alcohol use or alcohol use with other substance use were found to score significantly higher on the depression scale than those who reported no alcohol or substance use, although the effect sizes were small. It was found that males and females who reported alcohol use only did not differ significantly on the depression scale.
The relationship between depressive disorders and substance abuse/dependence has been documented, both in the general population and in the rural United States. The high comorbidity rates are especially documented in studies of the general population. For example, Grant (1995) detailed the patterns of comorbidity drug use disorders (DUDs) and major depression in the U.S. Data were derived from the Longitudinal Alcohol Epidemiologic Survey consisting of 42,862 respondents, 18-years-old and older, residing in the contiguous U.S. Analyses of the data showed that comorbidity of a variety of DUDs and major depression was pervasive in the general population. The association between drug dependence and major depression was greater than the association between abuse and major depression. Compared with men, the abuse-depression association was consistently greater for women with regard to prescription drugs, sedatives, tranquilizers, and amphetamines, but not with regard to cocaine or hallucinogens. The study revealed that men generally demonstrated stronger relations between drug dependence and major depression. Swedsen and Merikangas (2000) point out that depression and substance use disorders are highly prevalent in the general population and often co-occur within the same individual. This association is most commonly explained either by a causal relationship or a shared etiologic factor underlying both disorders.
Manley (1992) argued that deinstitutionalization and national patterns of illicit drug use have contributed to the comorbidity of mental health and addictive disorders in the U.S. A study by the National Institute of Mental Health found that among those with a history of drug abuse, 53.1% also had a mental disorder (four times the risk than in the general population) and 47.3% had an alcohol abuse disorder (seven times the risk than in general population). Of all individuals with a lifetime diagnosis of schizophrenia (1.5% of the U.S. population), 47.3% met criteria for some form of substance abuse. Rates of substance abuse among individuals with bipolar disorder are several times higher than in those with unipolar depression. Manley suggests that mental disorders must be addressed as a central part of drug abuse prevention efforts.
Harka et al. (1991) examined the relationship between alcohol consumption per typical occasion and depressive symptomatology in a meta-analysis of 8 longitudinal studies from the U.S., Canada, and the U.K. The aim was to determine whether co-occurrence of heavy drinking and depressive symptoms in a clinical population is consistently present in the general population. Their findings were supportive of a relationship between drinking and depression. Age significantly and consistently predicted quantity for both sexes combined, depression significantly predicted quantity over longer intervals for females, and quantity significantly predicted depression over shorter intervals for females.
Klerman et al. (1996) examined whether the recent increase in rates of major depressive disorder (MDD) could be attributed to comorbid alcohol and drug abuse. Using data derived from 1,996 relatives of probands with an affective disorder and 13,177 adults in the general population, elevated rates of MDD occurred among those with comorbid drug and alcohol abuse in both samples.
Depression and substance-related disorders in the rural U.S.
In rural populations, most of the literature supports a relationship between depressive disorders and substance abuse/dependence. For instance, greater alcohol and drug abuse has been found to be positively associated with more negative affect (i.e. anxiety, depression, and hostility) in rural outpatient populations (Pasion-Gonzalez, 1996). The comorbidity rates may be especially high for rural pre-adolescents and adolescents. Price-Sharps (2000) found that Attention-Deficit Disorder was associated with substance abuse, depression, and paranoia in a non-clinical sample of adolescents in a rural mountain area.
In another study, Blau et al. (1988) investigated the relationship between self-esteem, depression, anxiety, and drug abuse in 143 rural 5th and 118 rural 7th graders, using the Revised Children’s Manifest Anxiety Scale, depression and self-appraisal inventories, a drug attitude scale, and a drug use scale. The authors found that self-esteem, depression, and anxiety were significantly correlated with variables associated with potential drug use (attitudes toward drugs and willingness to use drugs). Findings suggest that low self-esteem and high anxiety and depression can be predispositions to drug use in this age group.
Kahn (1986) reviewed the literature regarding psychosocial disorders among Aborigines of Australia and among American Indians and Eskimos of the US. The author found high rates of psychosocial disorders among both populations. Alcohol abuse, the most visible disorder, was associated with other high-prevalence disorders, including depression, suicide, family instability, delinquency, and accidental death.
The link between depression and substance abuse/dependence is not as clear in rural individuals seeking substance-related treatment. For example, Steffenhagen and Steffenhagen (1985), in examining 61 alcoholic detoxification clients, found that rural residence and level of self-esteem predicted depression, which predicted alcoholism. However, Booth et al. (1991) found that psychiatric comorbidity was not associated with the rate and time of alcohol-related inpatient readmissions for a group of 255 patients discharged from alcoholism treatment at a Midwestern rural medical center. In this study, a structured interview obtained information regarding psychiatric disorders including depression, antisocial personality disorder, and polysubstance abuse, as well as alcohol history and sociodemographics. Ninety-eight participants were readmitted for alcoholism-related diagnoses within 15 months of discharge. Participants with a long history of heavy drinking, high daily alcohol consumption, and history of previous alcoholism treatment were most likely to be readmitted with an alcoholism-related primary diagnosis. Once these variables were controlled for, other major psychiatric disorders, polysubstance abuse, and sociodemographic variables did not appear to predict time to readmission.
Authors have offered hypotheses regarding the roles alcohol consumption play in handling life stressors. Neff and Husaini (1982) suggested that alcohol consumption played a stress-buffering role. The authors administered a life-events checklist to and interviewed 713 rural adult participants. Results showed that higher levels of depressive symptomatology were found among those experiencing more life events, although there was no significant effect of alcohol use. The number of reported life events was strongly related to depression for abstainers and heavy drinkers but not for moderate drinkers. Financial and “calamitous” life events and depressive symptomatology were less strongly related among moderate and heavy drinkers than among abstainers, whereas health and “relational” (involving the family) life events were more strongly related to depression among heavy drinkers and abstainers than among moderate drinkers. It was concluded that a typical pattern of alcohol consumption affects the relationship between life events and the occurrence of depressive symptoms. Neff (1985) also suggested that occasional drinking served as a buffer for depression symptoms, while moderate drinking had a less pronounced effect in a sample of 1,270 rural White 18-60 year-old individuals. The author concluded that the buffer effects from alcohol vary with event type and symptom dimension.
Prescott, Aggen, and Kendler (2000) found that major depression was found to be more common in women and alcohol disorders more prevalent in men. However, substance abuse and dependence are also problematic for rural women. Behnke et al. (1997) compared 154 rural pregnant cocaine users and 154 rural pregnant controls and found that the cocaine users were more likely than non-users to be older, to use other drugs, to begin their drug use at an earlier age, to have more depressive symptoms, to haven an external locus of control, to have lower self-esteem, to have a more simplistic understanding of child development, and to have higher positive life event impact scores.
Whether substance abuse and dependence are related to depression in rural women has implications for primary care providers who provide much of the mental health screening services to women living in rural areas. Van Hook (1996) examined how women use rural primary care providers (PCPRs) for their depression, some of the barriers to providing optimal care for women with depression within primary care, and the implications for social workers in primary and mental health care and other community services. Over a two-week period in 1993, 297 18-78 year-old women seeking help from rural PCPRs in Iowa and Michigan completed a questionnaire containing measures of depression, somatic problems, physical violence, sexual abuse, and alcohol abuse. Results showed that many women typically do not discuss their depression (major depression or other types) with their PCPRs, although those who did found it very helpful. Major barriers to discussing depression with PCPRs include stigma about depression, lack of time, and perceptions that the PCPRs were not interested in these concerns. Depressed women tend to seek help for somatic problems like headaches, backaches, muscle pains, sleep problems, feeling tired, and abdominal pains. The findings of this study suggest that it may be important to screen for depression in indirect ways, such as checking for clinical correlates of depression.
Other factors such as utilizing community or religious support may influence whether there is a relationship between alcohol use and depression in rural populations. For instance, Musick, Blazer, and Hays (2000) examined whether breaking the norms of a religious group can have deleterious consequences for individual mental health and whether this effect is exacerbated by frequent service attendance. To test these ideas, the authors used data collected from 1,897 older adults (age 65 and older) living in central North Carolina. Using only sample members who reported an affiliation with a Baptist denomination, the authors tested whether attending services more often and living in rural areas were associated with a smaller likelihood of alcohol use. The authors further tested whether the use of alcohol in this sample was associated with higher counts of depressive symptoms. The authors found that those who attend services more often were less likely to use alcohol. It was also found that alcohol use had no effect on depressive symptoms. One exception to this latter finding was that using alcohol was associated with more depressive symptoms among rural Baptists who rarely attended religious services.
The purpose of this article is to determine if a relationship between depressive symptoms and substance or alcohol use exists in adults from the rural Midwestern United States. Whether a relationship between depression and alcohol/substance use exists could provide useful information for rural health and mental health care providers. Given Van Hook’s (1996) findings, primary care providers may screen for depression in more indirect ways (i.e. the presence of clinical correlates such as substance or alcohol use, abuse, or dependence).
The Farm Resource Center
The data for this study was collected by the Farm Resource Center (FRC), a non-profit, non-partisan, non-sectarian corporation established in 1985 to serve farm families in need of mental health crisis. Farmers, coal miners, and their families are assessed and treated based on a community intervention model. Thousands of data points have been collected, including demographic information such as marital status, employment status, initial presenting problem, occupation, age, gender, and race. FRC also prides itself in providing treatment that works. Included in each client’s intervention plan are follow-up treatment, well-being assessments, and customer satisfaction ratings. FRC is unique in its approach to assessment and treatment. Clients come to FRC from a variety of sources. FRC utilizes referral networks in potential consumers’ communities. Outreach Workers and volunteers from local environments are utilized in providing services for FRC. Consumers are assessed on a variety of measures, including symptom checklists, risk for suicide or domestic violence, medical problems, overall well-being, pre-post GAF, and progress notes. Each consumer is given a treatment plan with goals based on his or her needs. As the treatment plan is implemented, each consumer is monitored closely to ensure that goals are met. Cases are closed in a formal manner. Referrals are provided throughout treatment, linking consumers to community resources. Consumers of FRC’s services are linked to the services by a number of methods including but not limited to local referrals, a toll-free Crisis Intervention Response line, local outreach workers, volunteers, and community seminars. The services FRC provides are comprehensive, including assessment, crisis intervention, referral, consultation, and educational services. FRC’s unique method of data collection and number of cases collected to date (over 2,000) allow researchers to explore differences in trends in assessment data between different groups of consumers, thus identifying and meeting the needs of target populations.
The participants for this study included individuals and families who received services from FRC between 1995-2000. Only adult participants age 18-65 and were assessed pre-treatment were included as participants in this study. The total number of participants who met these criteria was 1,191.
The median age of the participants in this study was 42.0, with the middle 50% ranging from 33.0-51.0. Of the 1,191 participants, 56.3% (n = 670) were female, and 43.7% (n = 521) were male. More than half (51.6%) of the participants reported being married (n = 614). The next largest percentage of participants (22.2%) were divorced (n = 264), while 13.9% reported never being married (n = 166), 8.9% reported being separated (n = 106), and 3.4% were widowed (n = 41).
A majority (96.3%) of the sample was Caucasian (n = 1,147), with African-Americans comprising the next largest percentage of participants (2.2%, n = 26). Two-thirds of the sample reported unemployment (n = 788). The most reported presenting problems were personal/emotional (62.3%, n = 742) and financial (22.8%, n = 271) in nature. The remaining presenting problems (employment, family, health, marital) each comprise less than 5% of the sample.
This sample of participants is representative of individuals and families seeking and obtaining community services in a rural setting in the Midwestern United States. The sample includes farmers, coal miners, family members of farmers and coal miners, and other members of the rural community. All of FRC’s services are confidential and are provided at no cost to the participants.
This case control study examined the relationship between depression and alcohol and / or substance use. Participants were not assigned to treatment or control conditions. They were naturally assigned to the demographic characteristics and assessment characteristics they met. Participants were classified according to their alcohol/substance use pattern. They were classified into four groups: no reported use of alcohol or any substance excluding tobacco (NONE), alcohol use only (ALC), another substance only (SUB), or both alcohol and another substance (BOTH). There were no control groups in this study. Every participant received FRC standard protocol treatment. For purposes of this study, only demographic and assessment data were used. This study is not aimed at measuring the effectiveness of treatment nor is it to examine which groups improve significantly with treatment.
The participants received standard treatment in their natural setting. Consumers are assessed by Outreach Workers on a variety of measures, including symptom checklists, risk for suicide or domestic violence, medical problems, overall well-being, pre-post GAF, and progress notes. For purposes of this study, a “depression” scale was artificially derived from the FRC symptom checklist. Symptoms of depression included fatigue, insomnia, poor appetite, depressed, bored, urge to cry, negative thoughts/attitude, lonely, suicidal thoughts, helplessness/hopelessness, inability to concentrate, decreased energy, inappropriate crying, over eating, under eating, lowered sex drive, decline in work effectiveness, and isolation / withdrawal from others. Only status variables and assessment information were used for this study. In addition to the depression variables listed above, the remaining variables of interest were whether the participants reported using alcohol (alcohol use) or substances (substance use). Alcohol/substance abuse and dependence are not assessed with the FRC symptom checklist.
In order to address whether there were differences between alcohol/substance users and non-users on the depression scale, reliability analysis was first conducted on the depression scale. Using Cronbach’s alpha, a value of 0.83 was computed as an estimate of internal consistency of the depression scale. (The item “over eating” was dropped to increase reliability.)
In order to address whether the four groups differed on the depression scale, an analysis of variance (ANOVA) was conducted. Although a Kolmogorov-Smirnov test for normality revealed that depression scores in this sample were not normally distributed (K-S = .11, p < .001), analysis proceeded with the ANOVA given the sensitivity of the test for normality and given that no other assumptions required by ANOVA were violated. The homogeneity of variance assumption required when conducting an ANOVA was not found to be violated (Levene's F = 2.32, p = .07). The groups being compared can be considered independent because participants belonged to one of four groups. The measure for depression is the artificially derived depression scale of the FRC symptom checklist, which is scale-level.
The overall ANOVA was found to be significant [F(3,1187) = 7.24, p < .001]. Using Tukey’s HSD, post-hoc analysis revealed that two relationships accounted for the overall significance. Those in the NONE group scored significantly lower on the depression scale compared to those in the ALC group (p = .005). Those in the NONE group also scored significantly lower than those in the BOTH group (p = .02). The magnitude of the differences for both comparisons are small (r = .10 and r = .09, respectively). There were no significant differences between the ALC, SUB, and BOTH groups. Table 1 presents mean comparisons for all for groups.
Mean comparisons for substance use groups
Substance Use Group
It is important to note that most participants (n = 1,107) did not report using any substance. For those who did report at least using alcohol and/or another substance (n = 84), most reported using only alcohol (n = 57), followed by using both (n = 16), and using another substance without alcohol (n = 11). Of those using at least one substance, almost 60% were male (n = 50) and most were Caucasian (n = 82, 97.6%). The median age of these participants was 37.0. Almost two-thirds were presenting with a personal or emotional problem as their primary presenting problem (n = 54). Roughly one-third were divorced or married (n = 27 for each group), one-fifth were never married (n = 18), and the remainder were separated (n = 10) or widowed (n = 2). Implications of the finding that few participants reported alcohol or substance use will be presented in the Discussion section.
Since very few participants reported using a substance without alcohol or using a substance with alcohol, analysis by gender is provided for only those reporting alcohol use without any other substances. In order to address whether those who report alcohol use only (ALC) differed by gender on depression, an independent samples t-tests was conducted. Although a Kolmogorov-Smirnov test for normality revealed that depression scores in this sample were not normally distributed, as described above, analysis proceeded with the independent samples t-test since no other assumptions were violated.
The homogeneity of variance assumption required when conducting t-tests was not found to be violated (Levene's F = 0.21, p =.65). The independent samples t-test revealed that, on average, depression scores for females in the ALC group (`x = 8.29, s = 4.60, n = 24) were not significantly different than depression scores for males in the ALC group (`x = 8.15, s = 4.24, n = 33), [t(55) = 0.12, p = .91]. There is a 95% likelihood that the true difference between these two groups lies between -2.22 and 2.50. Given that this confidence interval includes "0," there appears to be a no differences in depression levels between females and males in the ALC group. The effect size for the result is very small (r = .02).
The findings indicate that those who report using only alcohol or alcohol and another substance report higher levels of depression than those who report using no substances. Those who use alcohol or alcohol and another substance have slightly higher scores depression scores compared to those who report no substance use. No differences between any of the other comparisons were found in this study. Overall, alcohol use by gender did not appear to be related to depression in the sample of participants from the rural Midwestern United States. Depression scores were relatively equal among males and females.
One threat to the validity of these findings has to do with sample characteristics. A large number of participants reported no alcohol or substance use. In fact 1,107 out of 1,191 reported not using alcohol or any other substance. Underreporting due to fear of stigmatization may be an issue. Characteristics of the Outreach Worker conducting the assessment may also be considered a threat to the validity of the findings in this study (i.e. how the questions are asked). Further, developing symptom items that address substance “abuse” and “dependence” as opposed to “use” may be considered in the future to assess abuse/dependence.
The preliminary findings of this study suggest that primary care providers should assess for alcohol and substance use in rural individuals. Given that many primary care physicians or nurses are the health care providers seeing patients with mental health disorders or psychosocial stressors that mask themselves as physical illness, they will be the provider to address mental health issues. Based on the Van Hook (1996) study, directly measuring depression may be difficult with rural populations. Rural individuals and families may be reluctant to disclose such information. Based on this study as well, rural individuals and families may be reluctant to discuss alcohol and/or substance use. Further, those providing mental health services may not be appropriately trained in psychiatric assessment. Obtaining clinical correlates through a good history is likely a viable option for mental health screening. The results of this study suggest that alcohol and substance use are at least slightly related to depression in individuals in the rural Midwestern United States.
Behnke, M., Eyler, F.D., Woods, N.S., Wobie, K., & Conlon, M. (1997). Rural
pregnant cocaine users: An in-depth sociodemographic comparison. Journal
of Drug Issues, 27(3), 501-524.
Blau, G.M., Gillespie, J.F., Felner, R.D., & Evans, E.G. (1988). Predisposition to
drug use in rural adolescents: Preliminary relationships and methodological
considerations. Journal of Drug Education, 18(1), 13-22.
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(1), 37-43.
Grant, B.F. (1995). Comorbidity between DSM-IV drug use disorders and major
depression: Results of a national survey of adults. Journal of Substance Abuse,
Harka, E., Johnstone, B.M., Leino, E.V., Motoyoshi, M.M., et al. (1991). A
meta-analysis of depressive symptomatology and alcohol consumption over time.
British Journal of Addiction, 86(10), 1283-1298.
Kahn, M.W. (1986). Psychosocial disorders of aboriginal people of the United States
and Australia. Journal of Rural Community Psychology, 17(1), 45-59.
Klerman, G.L., Leon, A.C., Wickramaratne, P., Warshaw, M.G., Mueller, T.I.,
Weissman, M.M., & Akiskal, H. (1996). The role of drug and alcohol abuse in recent
increases in depression in the U.S. Psychological Medicine, 26(2), 343-351.
Manley, A. (1992). Comordidity of mental and addictive disorders. Journal of Health
Care of the Poor & Underserved, 3(1), 60-72.
Musick, M.A., Blazer, D.G., & Hays, J.C. (2000). Religious activity, alcohol use, and
depression in a sample of elderly Baptists. Research on Aging, 22(2), 91-116.
Neff, J.A. (1985). Evaluating the stress-buffering role of alcohol-consumption: Variation
by type of event and type of symptom. Alcohol & Alcoholism, 20(4), 391-401.
Neff, J.A., & Husaini, B.A. (1982). Life events, drinking patterns and depressive
symptomatology: the stress-buffering role of alcohol consumption. Journal of Studies
on Alcohol, 43(3), 301-318.
Pasion-Gonzalez, P.R. (1996). Correlates of substance use among chronic mentally ill
outpatients in a rural county mental heath setting. Dissertation Abstracts
International, 56(10-B), 5778.
Prescott, C.A., Aggen, S.H., & Kendler, K.S. (2000). Sex-specific genetic influences on
the comorbidity of alcoholism and major depression in a population-based sample of
US twins. Archives of General Psychiatry, 57(8), 803-811.
Price-Sharps, J.L. (2000). Symptoms of attention deficit disorder, substance abuse, and
comorbid correlates in adolescents. Dissertation Abstracts International, 60(7-B),
Steffenhagen, L.A., & Steffenhagen, R.A. (1985). Self-esteem and primary demographic
characteristics of alcoholics in a rural state. Journal of Alcohol & Drug Education,
Swendsen, J.D., & Merikangas, K.R. (2000). The comorbidity of depression and
substance use disorders. Clinical Psychology Review, 20(2), 173-189.
Van Hook, M.P. (1996). Challenges to identifying and treating women with depression
in rural primary care. Social Work in Health Care, 23(3), 73-92.