Nearly two-thirds of American women aged 18 or older (66.5%) report having consumed alcohol at least once during the preceding year, with nearly two-thirds of those women (i.e., 49.7% of all adult women) reporting having had a drink during the previous month [
Research examining the factors associated with and/or underlying alcohol abuse among African American women in community samples, as opposed to clinical or criminal justice populations, has been limited. The available research has identified several possible influences on the drinking practices and problems of members of this population. For example, based on their study of African American women living in a rural area, Boyd et al. [
In the present paper, which is based on a community sample of African American adult women residing in the South, we address the following research questions: (1) How prevalent are alcohol use and drinking-related problems in this population? (2) What factors are associated with incurring a larger number of alcohol-related problems? (3) Is there evidence of potentiating, or syndemic, effects among the key predictors of drinking problems in this population?
Regarding the latter question, for readers who may be unfamiliar with the construct of syndemic effects, we encourage consultation with Singer’s [
Conceptual model.
In other studies, syndemics theory has been applied to study a wide array of health behaviors among a variety of populations within the United States [
Data for this study were collected as part of People and Places, a cross-sectional study of people’s perceptions of how their neighborhood impacted their daily lives and actions and vice versa. Data were collected in Atlanta, Georgia. Participants (
The census block groups chosen for inclusion in the study were selected based on neighborhood structural characteristics as reported in the 2000 US Census Data and based on data from the Atlanta Police Department. Consistent with the study’s conceptual model and previous research findings [
In order to be considered eligible for participation, respondents had to self-identify as African American or Black, be at least 18 years of age, and have lived in the same neighborhood or census block group continuously for at least one year. People were considered ineligible for the study if they (1) were in an institutional setting at the time of recruitment, (2) were homeless at any time during the preceding year (because this would have run counter to the stable residency eligibility requirement, just described), (3) were intoxicated at the time of consent or interview, or (4) displayed signs of cognitive impairment at the time of consent or interview.
Computer-assisted structured interviews (CASI) were conducted with eligible persons in a private office that was located at a field site centrally located in one of the catchment areas. The survey covered information about people’s demographic characteristics, psychological and psychosocial characteristics, substance use history, sexual activity, criminal justice involvement, support network, and neighborhood perceptions. On average, interviews lasted approximately 90 minutes. The Emory University Institutional Review Board approved the study protocol.
The main outcome variable used in this paper is a scale assessing the extent to which alcohol-related problems were experienced during the preceding year. The scale was comprised of 10 items, each indicating whether the experience in question occurred “never,” “a few times,” “quite a lot,” or “often” during the preceding year. For the purpose of determining the number of drinking-related problems, each item was recoded to indicate having “never happened” or “ever happened” during the previous year. Constituent items included (1) problems with one’s family because of one’s alcohol use, (2), problems with one’s friends because of one’s drinking, (3) legal problems resulting from alcohol (ab)use, (4) physical fights as a result of one’s drinking, (5) having wanted to quit or cut down on one’s drinking, (6) trying to hide one’s drinking from others, (7) receiving complaints from others as a result of one’s alcohol (ab)use, (8) having experienced cravings for alcohol, (9) having experienced withdrawal symptoms when unable to drink, and (10) having consumed more alcohol than one wanted to. The scale was found to be reliable (Kuder-Richardson20 = 0.81).
For the analyses associated with Research Question #2 (i.e., the factors associated with incurring a larger number of alcohol-related problems) and Research Question #3 (i.e., evidence of potentiating, or syndemic, effects among the key predictors of drinking problems), independent variables were included from each of the four domains shown in Figure
Five scales, all of which were developed by the present authors specifically for use in the People and Places study based on the preimplementation formative research, were used to assess
Three measures were used to assess
Research Question #1, pertaining to the prevalence of alcohol use and drinking-related problems, was addressed with the use of descriptive statistics.
Research Question #2, which examined factors underlying alcohol problems in this population, was analyzed in a few successive steps. Initially, bivariate analyses were performed to determine which of the independent variables were associated with the number of alcohol problems experienced. Whenever the independent variable was dichotomous (e.g., sexual orientation, educational attainment), Student’s
For Research Question #3, which focused on whether or not the data provided specific evidence of syndemic effects, a variety of two-way and three-way interaction effects were examined for the key independent variables. To keep our analysis focused here, we elected to examine three specific independent variables for evidence of possible syndemic effects: age (comparisons were made between people aged 18–29 and those aged 30+), childhood maltreatment status (comparisons were based on persons experiencing no or very low levels of childhood maltreatment (i.e., mean scores ranging from 0 to 0.49) versus those experiencing higher levels of maltreatment (i.e., mean scores ranging from 0.5 to 4.0)), and criminal involvement of one’s friends (comparisons were based on people who reported having no friends who were criminally involved versus those who said that at least one of their friends was criminally involved). These three measures were chosen for these particular analyses because the distribution of the data was such that these classifications were likely to be the most robust statistically and because the multivariate data suggested that if syndemic-type effects were occurring, these three measures were the most likely candidates to illustrate the presence of such effects. Analysis of variance was used to compare the number of drinking-related problems based on all possible two-way interactions and the possible three-way interaction of these variables.
Throughout all of these analyses, results are reported as being statistically significant whenever
Table
Sample characteristics.
No alcohol during past-year |
Some alcohol during past-year | |
---|---|---|
|
||
18–29 | 48.3 | 35.9 |
30–39 | 26.1 | 18.7 |
40–49 | 18.9 | 28.1 |
50 and older | 16.7 | 17.3 |
|
||
|
||
Less than high school graduation | 31.0 | 40.6 |
High school graduation or equivalent | 52.3 | 34.7 |
At least some college | 16.7 | 24.7 |
|
||
|
||
Unemployed | 74.1 | 68.6 |
Employed, part-time | 9.2 | 14.2 |
Employed, full-time | 8.1 | 8.1 |
|
||
|
||
Not married or “involved” | 35.6 | 40.0 |
Married or “involved” | 64.4 | 60.0 |
|
||
|
||
Heterosexual | 93.1 | 89.4 |
Gay, lesbian, or bisexual | 6.9 | 10.6 |
Please note that statistical differences between past-year alcohol users and nonusers are as follows:
The large majority of the women who had ever consumed an alcoholic beverage (89.7%) reported having consumed at least one drink of alcohol during the preceding year. Almost all of these women (94.2%) indicated also having consumed at least one alcoholic drink during the previous 90 days. Throughout all of the remainder of the Results, past-year abstainers are excluded, so that the focus is on alcohol-using women.
Experiencing drinking-related problems was the norm, not the exception, among those reporting recent alcohol use. About one-third (32.8%) of those reporting alcohol consumption during the preceding year reported incurring no alcohol-related problems. Approximately one-third of the past-year alcohol users (32.7%) said that they experienced 1 or 2 of the 10 drinking-related problems that they were asked about. The mean number of drinking-related problems experienced was 3.3 (SD = 2.3). As Table
Prevalence of alcohol-related problems.
Alcohol-related problem | Prevalence (%) |
---|---|
Wanted to quit or cut down on one’s drinking | 49.6 |
Consumed more alcohol than one wanted to | 31.0 |
Experienced cravings for alcohol | 27.8 |
Receiving complaints from others as a result of one’s alcohol (ab)use | 24.6 |
Problems with one’s family because of one’s alcohol use | 21.0 |
Physical fights as a result of one’s drinking | 20.1 |
Problems with one’s friends because of one’s drinking | 18.5 |
Trying to hide one’s drinking from others | 9.5 |
Legal problems resulting from alcohol (ab)use | 9.2 |
Experiencing withdrawal symptoms when unable to drink | 8.4 |
Of the seven
All six of the
Four of the five
Finally, two of the three
When the statistically significant preceding items were entered into a multivariate equation together to determine which ones contributed significantly to the overall model when the influence of the others was taken into account, seven items were found to contribute uniquely and significantly to the overall prediction of the number of drinking-related problems women experienced (see Table
Factors associated with the number of drinking problems women experienced.
Independent variable |
|
|
|
---|---|---|---|
Age = under 30 | –1.27 | 0.25 | <.001 |
Health insurance = none in previous year | 0.54 | 0.11 | .002 |
Educational attainment = less than high school | 0.50 | 0.10 | .004 |
Sexual orientation = lesbian or bisexual | 0.75 | 0.09 | .008 |
Amount of childhood maltreatment | 0.52 | 0.16 | <.001 |
Perceived level of neighborhood safety | –0.95 | 0.09 | .007 |
Extent of friends’ criminal involvement | 1.00 | 0.24 | <.001 |
Considerable evidence was found in support of the notion that syndemic-type effects among key independent variables influenced the extent to which women experienced drinking-related problems. Highly statistically significant main effects (
Figure
Number of drinking problems, by age and friends’ criminality.
Number of drinking problems, by age group, friends’ criminality, and childhood maltreatment.
Figure
The authors would like to acknowledge a few potential limitations of this research. First, it was conducted with a sample of African Americans residing in a major metropolitan area. Persons living in other environments that are less densely populated may not share the same life and community experiences as those living in urban areas such as the one where the present study was conducted. Additionally, this research was conducted in the American South. African Americans living in other parts of the country may have different socioecological experiences compared to the persons who participated in the People and Places study. The extent to which these geographic factors affected the present study’s findings is not known and cannot be assessed with the available data.
Second, all data used in the People and Places study were based on uncorroborated self-reports. Therefore, the extent to which respondents underreported or overreported their involvement in various behaviors, such as alcohol consumption or drinking-related problems, is unknown. In all likelihood, the self-reported data can be trusted, as numerous authors have noted that persons in their research studies (which, like the present study, have included fairly large numbers of substance abusers and/or persons at risk for contracting or transmitting HIV) have provided accurate information about their behaviors [
A third possible limitation pertains to recall bias. Respondents were asked to report about their beliefs, attitudes, and behaviors during the past 30 days, the past 90 days, and the past-year, depending upon the measure in question. These time frames were chosen specifically (1) to incorporate a large enough amount of time in the risk behavior questions’ time frames so as to facilitate meaningful variability from person to person and (2) to minimize recall bias. The exact extent to which recall bias affected the data cannot be assessed although other researchers collecting data similar to that captured in this study have reported that recall bias is sufficiently minimal that its impact upon study findings is likely to be small [
A fourth potential limitation pertains to the recruitment methods used for identifying study participants for the People and Places study. Both passive and active recruitment strategies were used in order to identify eligible respondents and to gain their involvement in the study. Passive recruitment approaches in particular are subject to self-selection bias because they rely upon people who are interested in the subject matter of the research to volunteer to participate in the study. As a result, certain individuals who find the research subject matter less salient to their lives may choose not to participate in studies such as People and Places. As with any research study relying upon voluntary participation (which means almost all research studies involving human subjects), the extent to which this type of bias occurred and affected the findings obtained in this study simply cannot be assessed. A comprehensive discussion of the various ways in which self-selection bias may and may not affect the quality of information obtained in community-based studies of alcohol abuse has been provided by Connors and Volk [
Despite these potential limitations, we believe that the present study has much to offer in terms of its insights into the issues surrounding alcohol abuse in the population in question. In this community-based sample of urban African American women in the southern United States, alcohol use was highly prevalent and, for most women, so too were drinking-related problems. More than two-thirds (67.2%) of the alcohol-using women in the study said that, during the preceding year, they had experienced at least one problem as a result of their drinking; and more than one-third (34.5%) of the alcohol-using women reported three or more of such problems during that time period. With the two most common problems of these drinking-related ones being a desire to quit drinking or, in the alternative, to cut down on one’s drinking and experiencing alcohol-related cravings, there are strong indications of prevalent symptoms of alcohol dependence in this population. This suggests a need for alcohol(ism) treatment facilities that can offer recovery services to and be effective with African American women who reside in urban areas. Other authors have addressed the need to develop effective ways of facilitating substance abuse recovery among low-income, urban African American women [
The present study is also informative in its findings pertaining to the factors underlying drinking problems among southern, urban African American women. Seven such factors were identified in our multivariate analysis, and each merits a brief discussion. First, our analyses revealed that women aged 30 and older experienced significantly more problems as a result of their alcohol use compared to women aged 18 to 29. Various factors are likely to be relevant here. One of them is the simple matter of progression—that is, drinking problems and alcohol dependence progress and worsen as people continue to drink regularly, heavily, and/or abusively over time. Another factor that probably contributes to this finding, at least to some extent, is that the women at the lowest end of the adult age spectrum—namely, those aged 18 to 20—are unable to purchase alcohol legally or to consume it legally. This probably prevents some of them from having easy access to alcohol, which in turn leads to reduced alcohol-related problems among these younger women. Regardless of the underlying cause, the older women were more at risk for experiencing drinking-related problems than their younger counterparts, thereby leaving us with the recommendation of considering these older women to be a group in need of targeted intervention, education, and/or treatment with regard to alcohol-related matters. Other researchers have spoken about the need to offer age-targeted alcohol intervention programs [
Second, our multivariate analysis revealed that women who had not had any health insurance during the previous year experienced approximately 50% more alcohol problems than their counterparts who were insured for part or all of that time. As of 2013 with the passage of the Affordable Care Act, all health insurance plans offer coverage for mental health counseling and/or substance abuse treatment. But not all adults have signed up for health insurance [
Third, in this study, we learned that women who had not graduated from high school or earned a general equivalency diploma (GED) experienced approximately one-third more drinking-related problems than their better-educated counterparts. Based on this finding, it would be wise for alcohol education, alcohol intervention, and substance abuse treatment programs to target African American women with low levels of education. When preparing written materials for their target audience, educational and intervention initiatives may have to be particularly sensitive to the literacy levels of these women. Effective skills-building components of these programs may find it beneficial to incorporate information about the importance of earning a GED and to encourage women with lower levels of educational attainment to bolster their schooling.
Fourth, women who self-identified as lesbian or bisexual experienced more than 1.5 times as many alcohol problems as their heterosexual counterparts. Many studies have shown an increased risk for substance use and abuse among lesbians and bisexual women [
Fifth, the more childhood maltreatment that women experienced, the more alcohol-related problems that they incurred. This finding is consistent with numerous studies, which have shown an association between sexual, physical, and/or emotional abuse during the formative years and substance (ab)use problems in adulthood [
Sixth, the more unsafe women perceived their neighborhood to be, the more alcohol problems they tended to experience. This finding is consistent with the main tenets of social disorganization theory (consult [
Seventh, drinking problems were more prevalent among people whose friends were criminally involved than they were among people whose friends were law abiding. Similar to the discussion (above) regarding social-ecological factors, here we have a finding that also pertains to environmental influences on individuals’ drinking behaviors. In this instance, however, those influences are more direct and more personal, because they come in the form of friendships and how those friendships shape alcohol (ab)use practices. If the principles of social-ecological and social disorganization theories are applicable to the “feeling unsafe in one’s neighborhood” finding discussed above, then the principles of social control theory, containment theory, and social bond theory (again, consult [
Finally, we would like to discuss our findings pertaining to the matter of whether or not the present research yielded evidence to support the notion of syndemic effects being operative with regard to the factors underlying drinking problems among urban African American women. It did. All of the two-way interaction analyses that we conducted not only demonstrated significant main effects but also demonstrated large intergroup differences when the combined effects of age and childhood maltreatment, or age and having criminally involved friends, or childhood maltreatment and having criminally involved friends were examined. The same was true for our analysis of the three-way interplay of age, childhood maltreatment experiences, and associating with criminally involved friends. These findings offer strong and consistent evidence of syndemic-type effects underlying the alcohol-related problems experienced by the southern, urban African American women we studied. As we mentioned earlier, although it has not been uncommon for syndemics-focused studies to incorporate substance use/abuse measures into their models [
Our findings, therefore, suggest two very important things: (1) the principles of syndemics theory may prove to be useful in the development of educational efforts and prevention/intervention initiatives striving to reduce the adverse impact of alcohol abuse. (2) Syndemics theory may be able to offer important and practical insights into the myriad factors underlying the health problems faced by African American women. On this basis, we recommend that future health educators, interventionists, and substance abuse treatment specialists working with African American women consider incorporating syndemics theory-based components into their alcohol education, intervention, and treatment endeavors. Our findings suggest that this may very well prove to be an effective approach for them to take.
The authors declare that they have no competing interests.
This research was supported by a grant from the National Institute on Drug Abuse (R01DA25607).