Rosa M. Crum, MD, MHS CHAPTER
2

CHAPTER OUTLINE

This chapter is organized to cover several areas. First, a few epidemiologic terms and types of epidemiologic studies are discussed. Second, some of the literature regarding prevalence, incidence, and trends of alcohol and drug use disorders is reviewed. The remainder of the chapter is devoted to discussing some of the correlates and risk factors associated with substance use disorders.

SOME EPIDEMIOLOGIC PRINCIPLES

Epidemiology has been defined in several different ways but may be considered the study of how diseases are distributed in populations as well as the study of the determinants of disease and health (1–3). Some basic terms used in epidemiology deserve attention in this chapter, because they are helpful to understanding the literature and some of the studies reported here. Prevalence generally is taken to represent the ratio of the total number of cases of a particular disease, divided by the total number of individuals in a particular population at a specific time. Incidence refers to the occurrence of new cases of a disease, divided by the total number at risk for the disorder during a specified period (4). Prevalence takes into account both the incidence and duration of a disease, because it depends not only on the rate of newly developed cases over time but also on the length of time the disease exists in the population. In turn, the duration of the disorder is affected by the degree of recovery and death from the disease. Incidence generally is taken to represent the risk of disease, whereas prevalence is an indicator of the public health burden the disease imposes on the community (4).

The strength of association between a particular characteristic and the development of disease generally is represented by the relative risk. The relative risk measures the incidence of disease among those with a particular characteristic (such as family history of alcohol addiction), divided by the incidence of disease among those without exposure to that characteristic. If there is no difference in the incidence among those with and without the characteristic, the ratio is equal to one. The odds ratio is also a measure of the strength of association such as between a characteristic and disease. A relative risk or odds ratio greater than one indicates a positive association of disease with a given characteristic. A relative risk or odds ratio less than one signifies a negative association, which may indicate a protective effect associated with the characteristic.

Excellent detailed discussions of epidemiologic study designs can be found elsewhere (1,2,4,5). For the purposes of this chapter, epidemiologic studies can be divided into two types: (i) observational or (ii) experimental. Observational studies may include cross-sectional, case–control, or cohort studies. In cross-sectional studies or surveys, individuals are evaluated (e.g., by interview or physical examination) at a particular point in time (4). Analytic studies usually are classified as case–control (retrospective) or cohort (longitudinal, prospective). Analytic studies generally test a hypothesis of a suspected association between a particular exposure (risk factor) and a disease or other outcome. In all observational studies, the investigator observes the study participants and gathers information for analysis (4). In contrast, experimental studies, such as randomized clinical trials, are designed by the investigator, study groups are selected, and often an intervention (such as a new type of treatment) is given to one group of participants. The study participants are followed, and the outcomes of each group are measured and compared.

ALCOHOL USE DISORDERS

Prevalence

A number of major surveys in the United States and internationally have assessed the prevalence of addiction. Comparison of these studies sometimes is difficult because they employ different measures and definitions of addiction. Some surveys have used structured interviews according to criteria that have become universally recognized, such as the Diagnostic and Statistical Manual of Mental Disorders (6), most recently in its fifth edition (DSM-5), and the International Classification of Diseases, now in its 10th revision (7). Throughout the text when we use the term substance use disorder, we are referring to substance abuse and/ or dependence. One of the earliest surveys to assess the epidemiology of substance use disorders in the United States using a structured psychiatric interview was the National Institute of Mental Health’s Epidemiologic Catchment Area (ECA) study (8–10). Baseline interviews for this study were conducted between 1980 and 1984, when collaborators in the ECA assessed a probability sample of more than 20,000 adult participants in five metropolitan areas of the United States. Using the Diagnostic Interview Schedule (11), diagnoses of substance abuse and dependence were assessed according to criteria from the Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition (DSM-III) (12). Other major surveys that have provided information on substance use disorders in the United States include the National Comorbidity Survey (NCS), which was first administered between 1990 and 1992 and assessed prevalence of abuse and dependence with a modified version of the Composite International Diagnostic Interview (13) based on DSM-III-R criteria (14). More recently, data from the 2001–2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), a nationally representative survey sponsored by the National Institute on Alcohol Abuse and Alcoholism, has provided lifetime and 12-month estimates of substance use disorders based on DSM-IV criteria (15,16) using the Alcohol Use Disorder and Associated Disabilities Interview Schedule (17,18). Twelve-month or prior year prevalence of alcohol abuse from the NESARC was found to be 6.9% among men and 2.6% among women. For alcohol dependence, the 12-month estimates were 5.4% for men and 2.3% for women. Lifetime prevalence of DSM-IV alcohol abuse and dependence was 17.8% and 12.5%, respectively (16). Most studies have found that the prevalence of alcohol use disorders is highest among young adults. For example, prevalence of 12-month alcohol use disorder was highest among the 18- to 29-year-old age group in the NESARC baseline survey, and prevalence generally decreased among older age groups (16). The most current prior year estimates of alcohol use disorders in a population-based sample of the United States come from the annual National Survey on Drug Use and Health (NSDUH). Since 2000, the NSDUH has gathered information on prior year prevalence of substance use disorders based on DSM-IV criteria. Data from the most recently available survey, the 2011 NSDUH, indicate that 6.5% of survey participants met criteria for alcohol abuse or dependence in the prior year. The highest prevalence for past year alcohol abuse or dependence was found for young adults aged 18 to 25 years (14.4%) (19).

Differences in estimates across surveys may be due to variations in the diagnostic instrumentation, the version of the DSM that prevailed at the time the survey was completed, the size of the survey sample, and the locale of the survey participants (nationally representative samples vs. individual communities), as well as specific characteristics of the populations surveyed, including the age range of study participants. For example, the NCS included a relatively younger population (persons ages 15 to 54 years) than some other surveys (20). In addition, specific methods used during data gathering (e.g., self-administered computerized vs. face-to-face interviews; use of identifiers for follow-up assessment vs. anonymity) may relate to differences in survey findings (21). Narrow et al. (22) have provided revised prevalence estimates, based on data from both the ECA and NCS, by focusing on clinical significance criteria among cases from both surveys that met diagnostic criteria. When the clinical significance criteria were used, disparities between the two surveys were attenuated. The revised 1-year prevalence of alcohol use disorders among adults aged 18 to 54 was estimated to be 6.5% once the clinical significance criteria were applied (as compared with 9.9% without the clinical significance criteria) (22). Differences in survey findings also may occur from use of “gated” procedures (23). For example, to maximize efficiency, in some surveys, only individuals who screen positive for abuse are assessed for dependence. This approach was assessed in several analyses that reported relatively small differences in estimated prevalence for ascertainment of some types of substance dependence (i.e., cannabis, cocaine) when “gated” as compared with “ungated” protocols are in place (23,24) but possibly more appreciable differences in assessment for other substances (i.e., alcohol) (25). As a consequence, some individuals with dependence may not be identified among population-based surveys when gated procedures are used (25,26). In some assessments, these cases appear to be less likely to have received treatment and therefore may be less likely to be clinically apparent cases but may be more likely to occur among specific subgroups of the population, such as among women and racial–ethnic minorities (26,27).

Incidence

Compared with information from cross-sectional surveys, prospective data gathered over time are less available; consequently, there is less information on incidence rates for substance use disorders in the general population. Early data from the Swedish Lundby study provide one of the few estimates of incidence of alcohol abuse and dependence over a prolonged follow-up period (28). The Lundby community was interviewed for the first time in 1947, reinterviewed in 1957 (with 1% lost to follow-up), and then examined again in 1972 (28–30). Of the 2,550 participants in the original survey, 98% of those still alive in 1972 were reinterviewed. The investigators found that, among males, the overall age-adjusted annual incidence of alcohol abuse or dependence was 0.3%. They further found a general decline in incidence with age, with a sharp drop in incidence of alcohol use disorders among men, beginning in their thirties (28). The highest annual incidence for an alcohol use disorder among men (0.67%) was found among the youngest age group: those 10 to 19 years of age (28). Of the 925 women examined in the Lundby community in 1972, only three were identified as having an alcohol use disorder (29).

Fillmore examined longitudinal data from population-based U.S. samples and also found that incidence of drinking problems generally was lower for women than for men and that incidence for both men and women declined with age (31,32). Drinking problems were identified using scales that included measures of binge drinking, drinking to cope, loss of control of alcohol use, belligerence, and drinking-related problems involving one’s spouse, friends, job, finances, and the law. Fillmore’s findings also indicated different patterns of drinking by gender. For example, women tended to develop problems associated with drinking later in life than did men, and women were found to have higher rates of remission across all age groups than did men (32). Data from the 1-year follow-up of the ECA surveys are consistent with other prospective studies (33). Among men, the estimated annual incidence of alcohol use disorders in the 1-year follow-up of the ECA surveys was highest for the youngest age group (those aged 18 to 29 years: 5.8 per 100 person-years) and decreased with age, with an overall annual incidence among men of 3.7 per 100 person-years. For women, incidence also decreased with age, with an overall incidence of 0.6 per 100 person-years. The peak incidence in women also was among those 18 to 29 years of age (1.1 per 100 person-years) (33). Analyses using the extended follow-up from the Baltimore site of the ECA, between 1981 and 1996, indicate similar trends for the development of alcohol dependence (34). The more recent prospective data from the NESARC have provided annual incidence rates for DSM-IV alcohol abuse (1.0 per 100 person-years), and alcohol dependence (1.7 per 100 person-years), and also indicate that the greatest risk for alcohol use disorders occurs during young adulthood (35).

DRUG USE DISORDERS

Prevalence

Several major surveys estimate the prevalence of drug use in the United States (e.g., 36,37). Analyses from the ECA surveys provided early estimates of prevalence of illicit drug abuse and dependence (including prescription drug use disorders) (38). As discussed with regard to alcohol disorders, men generally were found to have a higher lifetime prevalence of drug use and drug use disorders overall in the study population (38,39). However, among drug users, lifetime prevalence differed little by gender (in the ECA for male users, lifetime prevalence was 21%, whereas for female users, it was 19%) (38). The 2001–02 NESARC survey has provided more recent data on the prevalence of 12-month and lifetime DSM-IV drug abuse and drug dependence. Lifetime prevalence of drug abuse and dependence in the 2001–2002 NESARC survey were found to be 7.7% and 2.6%, respectively, and the corresponding estimates for 12-month prevalence were 1.4% and 0.6%, with the overall prevalence of drug use disorders being consistently higher for males (40). From the most recent available NSDUH survey in 2011, prevalence of illicit drug use disorders in the year before the survey was reported to be 2.5% overall, with the largest prevalence found among young adults aged 18 to 25 years (7.5%) (19). As indicated previously in the discussion on alcohol use disorders, differences in data collection methodology as well as diagnostic instrumentation may account for some of the variations found in prevalence estimates reported for these large national surveys (21).

From the 2011 NSDUH, we know that 26.5% of the population reported current use of tobacco products, and 22.2% reported smoking tobacco cigarettes (19). Data from the 2001–2002 NESARC indicate that 12.8% of the population met DSM-IV criteria for nicotine dependence in the prior year (41). A greater proportion of males (14.1%) than females (11.1%) had prior year nicotine dependence. The vast majority of the individuals with nicotine dependence use tobacco cigarettes (93.7%) (41).

Incidence

There is a relative paucity of information regarding the incidence of drug use disorders as a group, with less information available for specific drugs. Early findings from the 1-year prospective ECA data showed that the incidence of illicit drug use disorders as a group is greater for men than for women across the entire life span (33). The estimated annual incidence was 1.09 per 100 person-years of risk (33). For men, the estimated annual incidence for drug abuse/dependence was 1.66 per 100 person-years of risk, whereas for women, the estimated annual incidence was 0.66 per 100 person-years of risk. As was the case for alcohol-related disorders, the highest incidence for both men and women was found in the 18- to 29-year-old age group; incidence dropped sharply after young adulthood. The incidence of drug use disorders was zero among those 65 years of age and older (33). More recent analyses of the 3-year prospective NESARC data have provided data for annual incidence of drug use disorders (35). Annual incidence of drug abuse (0.28 per 100 person-years) and drug dependence (0.32 per 100 person-years) was similar. As discussed previously, variations in incidence across longitudinal studies may reflect differences in methodology, survey design, and study population (35).

RECENT TRENDS OF ALCOHOL, TOBACCO, AND ILLICIT DRUG USE

Trends of alcohol, tobacco, and illicit drug use over the past decade can be examined with estimates provided by annual surveys such as the NSDUH. Based on the most recent available survey completed in 2011, approximately half (51.8%) of the U.S. population report current drinking (having at least one drink in the past month) (19). Over the past decade, the proportion of current drinkers remains essentially unchanged. Estimates from the 2002 survey indicate that at that time, 51.0% reported current drinking. In 2002, a total of 22.9% of the population described binge drinking (five or more drinks on at least one occasion in the past month), and 6.7% were reported to be heavy drinkers (binge drinking 5 days or more in the past month). These findings are similar to reports from the 2011 NSDUH: 22.6% reported binge alcohol use and 6.2% reported heavy drinking. However, among young adults aged 18 to 25 years, overall binge drinking has decreased slightly over the past decade from 40.9% in 2002 to 36.8% in 2011 (19,42–50).

Use of tobacco products in general also has declined over the prior decade based on data from the NSDUH from a high of 30.4% prevalence in the population in 2002 to the current low of 26.5%. Cigarette smoking specifically has declined from 26.0% in 2002 to a past month prevalence of 22.1% in 2011. The decline is more pronounced among adolescents aged 12 to 17 years, where reports of past month use have dropped from 13.0% in 2002 to 7.8% in 2011. Although absolute values of prevalence estimates for cigarette smoking differ from other sources of adolescence use such as the Monitoring the Future (MTF) surveys (36,51) sponsored by the National Institute on Drug Abuse, both the NSDUH and MTF surveys indicate a similar pattern of overall decline for cigarette smoking through the last decade (19).

Illicit drug use as identified by the NSDUH includes reported use of marijuana or hashish, cocaine (as well as crack), heroin, hallucinogens, inhalants, and/or nonmedical use of prescription-type psychotherapeutics (19). In 2002, 8.3% of the population reported use of one or more of these substances in the prior month. Data from the most current survey indicates that in 2011, 8.7% reported past month illicit substance use. The most commonly used illicit substance is marijuana, which had an estimated prior month prevalence of 6.2% in 2002, decreased to a low of 5.8% in 2007, and in 2011 has significantly risen to 7.0%. Currently, 2.4% of the population reports nonmedical use of psychotherapeutic substances, composed primarily of nonmedical use of pain relievers.

New trends are continually being assessed by ongoing surveys and provide important information regarding trends of use over time as well as surveillance of newer substances. Since 2011, information on synthetic marijuana is now being gathered as part of the MTF, which annually completes a school-based sample of 8th, 9th, and 12th graders in the United States. Prior year prevalence estimates in the 2011 survey for synthetic marijuana among 12th graders was reported to be 11.4% with similar estimates reported in 2012 (11.3%). Information on synthetic marijuana was first assessed among 8th and 9th graders for the MTF in 2012. Among 8th graders, 4.4% reported using synthetic marijuana in the prior year, and among 9th graders, the prior year prevalence was 8.8% (51).

REMISSION FROM SUBSTANCE USE DISORDERS

Remission from alcohol use disorders such as alcohol dependence in community samples varies by individual characteristics (52,53) and subtype of alcohol dependence (based on age of onset, family history, other substance use, psychiatric disorders, and specific alcohol abuse and dependence criteria (54)). Maintaining remission is also dependent on consumption patterns while in remission. Using data from the NESARC, Dawson et al. (55) assessed occurrence of relapse among individuals who were in remission from alcohol dependence at the baseline survey and found that at the time of the follow-up survey approximately 3 years later, maintenance of successful recovery from dependence was greatest for abstainers as compared with other groups in remission that continued to drink.

Only a minority of individuals with substance abuse or dependence report using treatment services (including self-help groups, employee assistance programs, inpatient and outpatient facilities, rehabilitation centers, crisis centers, health care professionals, halfway houses, detoxification units, as well as others) (16,40). Based on data from the NESARC, Compton et al. (40) found that only 8.1% of those with lifetime drug abuse and 37.9% of those with lifetime drug dependence report receiving treatment. Among those with alcohol use disorders, the proportion receiving treatment is even lower (7.0% among those with lifetime alcohol abuse and 24.1% for those with lifetime alcohol dependence) (16). However, among those with illicit drug use disorders, treatment is positively associated with having psychiatric disorder comorbidity (40).

CORRELATES AND SUSPECTED RISK FACTORS

Many correlates and suspected risk factors for alcohol and drug use disorders have been examined. Some of these are discussed in this section for both alcohol and drugs, because the suspected risk factors are similar, and there are many findings in common. This section is restricted to a discussion of a small selection of personal or individual characteristics that have been found to be associated with drug and alcohol addiction. The discussion is by no means exhaustive and reviews only a fraction of the investigations in this area.

Gender

As discussed previously, alcohol disorders and alcohol-related problems are more common among men than among women. This consistent finding has been shown in a number of cross-sectional surveys (13,16,19,20,29,39,56,57), as well as in prospective studies (32,33,35,58). In the 2011 NSDUH (19), the occurrence of past-year alcohol abuse or dependence was approximately twofold greater for males (8.4% and 4.7% for males and females, respectively), and binge drinking among males (30%) was twice as high as that for females (15.8%); heavy drinking was three times higher for males than females (9.4% and 3.2%, respectively) (19). Differences in prevalence and incidence between men and women have been attributed to a number of factors. Cultural norms, societal standards, body size, and differences in the metabolism of alcohol all may contribute to the finding that women appear to use less alcohol and to have lower rates of alcohol addiction. However, as discussed in the systematic review by Keyes et al. (59), survey data in the United States over the past 20 years provide evidence that the gender gap for prevalence of alcohol use disorders is narrowing, perhaps as the result of changes in drinking patterns (59–61). Yet these patterns may differ in other areas of the world (59). Some hypothesize that changes in patterns of drinking among women may be a result of deviations from traditional female social roles, or related to changes brought about by the increased number of women in the labor force, as well as the combined input of home and work environments (62,63). Many characteristics (e.g., marital status, children in the home, full-time employment, ethnicity, age, occupation, educational level), as well as the occurrence of life events, and the presence of other psychopathology (such as depression) may play a role in gender variability with respect to alcohol consumption and the development of alcohol disorders (64–67). In addition, there may be gender differences in drinking as a response to stress and specific stressors (68). Assessments across cultural groups have reported on the impact of gender equity as well as social roles in explaining gender differences in drinking patterns (69,70). In some subpopulations, there may be strong associations between physical or sexual violence and alcohol use initiation (71), and the occurrence of problem drinking or alcohol use disorders (72). A history of childhood abuse has been found to be a potential predictor of women’s risk for alcohol and drug use disorders (73) and may impact progression through stages of alcohol involvement (74). However, the relationship of childhood abuse to alcohol disorders among women is complex and may involve a number of other family characteristics (75).

There also are gender differences with respect to illicit drug disorders. Males (boys and men) generally are more likely to use illicit drugs and may have a higher prevalence (19,38) and incidence (35,76) of drug use and disorders than do females. In the 2011 NSDUH (19), the overall proportion of illicit drug use disorder in the past year (abuse or dependence) among participants 12 years of age or older was approximately twice as large for males (3.4%) than females (1.7%). The social or cultural restrictions that are possible explanations for the reduced prevalence of alcohol use among girls and women also may apply to some types of illicit drug use. However, gender differences vary by the specific substance and the age of use. For example, in the 2012 MTF, annual prevalence of some substances including inhalant, tranquilizer, and amphetamine use was higher for 8th grade girls than boys (36). Similar findings are reported for inhalant use in the 2011 NSDUH with a prior year prevalence of 3.1% among boys aged 12 to 17 years, whereas for girls in the same age group, the prevalence is 3.5%. In addition, some (77) but not all data (78) show that among drug users, the proportion of males and females who develop dependence is similar. Using data from the NCS Replication, Wagner and Anthony (79) found that males had a higher risk of progression from first use to dependence for cannabis, but relatively small sex differences in progression risk for alcohol and cocaine.

Age

Prevalence of alcohol use disorders is generally lower among older adults (16,57,80,81). This may occur for a number of reasons. Because the measure of prevalence depends on the incidence as well as the duration of the disease (1,2,4), alcohol use disorders may be less prevalent among the elderly because the incidence decreases over the life span, the duration of the disorder is reduced, or some combination of the two factors is in effect. If the duration of the disorder is reduced, it may be a result of an increase in remission with age, or a reduction in survival. In other words, with age, prevalence may be reduced because fewer individuals develop the disease, because the addiction problems have resolved, or because addicted individuals die earlier. Explanations for a decreased prevalence with age also may include a reduced tolerance to alcohol with age (82), poorer recall among older adults, or a cohort effect (57). Further, the means by which alcohol disorders are identified in young adults may not be relevant to the elderly (83–87), with the result that alcohol problems and disorders may be under-recognized in older adults (85,87,88). Surveys that include only household participants may miss many with alcohol disorders who reside in nursing homes; also, older community residents with alcohol disorders may be less willing to participate in household surveys. Findings from prospective studies show that incidence of alcohol disorders generally decline with age (34,35). The hazard rate for alcohol abuse and dependence is reported to be highest at approximately age 19, with a steady reduction in hazard with increasing age (16). However, problems related to alcohol use among the elderly may occur at lower levels of consumption than in younger adults, and older adults with alcohol use disorders may be at greater risk for comorbid problems (89–91).

The age of onset of alcohol use has been investigated as a predictor of subsequent alcohol abuse and dependence (92–94). In general, the earlier the age of first use, the greater the estimated risk associated with the subsequent development of an alcohol use disorder. In addition, early drinking onset is associated with severity of alcohol dependence symptoms (95), elevated risk of alcohol-related injuries (96), motor vehicle accidents (97), physical violence after drinking (98), and the level of drinking in response to stressors (99). Moreover, early-onset drug use is associated with an increased risk for the development of drug use disorders (100,101), as well as alcohol dependence (101). Smoking at an early age has been identified as a predictor of drinking, increases the risk for transition to alcohol abuse and dependence, and is associated with greater severity of alcohol use disorders if they do develop (102). Earlier age of onset among women observed for more recent birth cohorts may also explain the rise in prevalence of alcohol dependence among women (103).

As was discussed for alcohol disorders, age correlates with the occurrence of drug use disorders. The highest prevalence and incidence rates for illicit drug disorders are found among individuals in late adolescence and young adulthood (33,38). As with alcohol use disorders, incidence of drug abuse and dependence decreases with age (33,35). Onset of drug use disorders is the highest at approximately age 19, with sharp declines thereafter, so that hazard rates after age 25 are relatively low (40). In addition, early onset of drug use is associated with elevated risk for subsequent substance abuse and dependence (93,101,104,105). Although incidence is low among older adults, and survival may be decreased for individuals with drug disorders as they age, other factors also may be involved. Prevalence may be lower among older adults because of a cohort effect. Exposure and availability of illicit drugs differ by birth cohort. For example, the current cohort of older adults had no access to crack cocaine in their youth. When evaluating changes in the frequency of a disorder (in this case, drug and alcohol addiction), distinctions need to be made between changes that uniformly occur for all age groups during a particular historic period (period effect), changes that occur with age as the individual matures (age effect), and a cohort effect that reflects differences in disease rate for individuals born in different years (106,107).

Some analyses show that patterns of addictions have changed and show a greater prevalence of alcohol and illicit drug dependence among cohorts born since World War II (59,77,78). Similarly, the risk of nicotine dependence has been reported to be greatest among smokers in the more recent birth cohorts (108). Evidence examining birth cohort associations with initiation of use for specific substances also indicates that more recent birth cohorts have been more likely to initiate drug use in childhood and early adolescence, particularly for cannabis, cocaine, and extra-medical drugs (76), including nonmedical use of analgesics (109,110). These patterns also have been reported in other global areas (59,111–113). However, in some analyses, there is evidence of a broader period effect across all age cohorts, which may explain the recent rise of marijuana use in the United States (114).

Race and Ethnicity

Information on the relationship between alcohol and illicit drug disorders and racial and ethnic background is complex and sometimes conflicting. Some of the inconsistent findings result from the relative paucity of data involving ethnic and racial subgroups, the classifications used to group ethnic minorities, variations in the social acceptability of drinking and drug use patterns within groups, and the relationship of socioeconomic status and the availability of health care to ethnic minority populations.

Findings from many studies indicate that drinking patterns among African Americans differ from Whites (16,115,116). African American youth begin drinking at older ages (117,118) and generally report lower levels of alcohol use, intoxication, and episodes of heavy drinking (19,50). Prevalence estimates for alcohol abuse and dependence are also reduced relative to Whites (57,115,116). Although the odds of alcohol dependence are lower for blacks compared with whites, African Americans tend to suffer more medical and social consequences from drinking, including higher mortality and psychiatric comorbidity (119–123). This may relate to differences in socioeconomic status, access to health care, health service utilization, and differing social and cultural environments (119,124–129). For example, neighborhood poverty has a greater effect on alcohol problems among black relative to white or Hispanic men (130). When socioeconomic factors are taken into account, race differences are sometimes attenuated (131–133).

In many reports, prevalence of alcohol abuse and dependence is lower for Hispanics relative to whites (16,134). Yet alcohol dependence among Hispanics may be more persistent after it develops (135). Furthermore, mortality from liver cirrhosis differs by race as well as ethnicity (121,136), with higher rates for white Hispanic males than white non-Hispanic males (121). Drinking patterns and alcohol-related problems vary among Hispanic subgroups in the United States (136–139), and variations may reflect factors such as social and cultural environment, degree of acculturation, country of national origin, generational status, nativity, and time since immigration (140–146). Acculturation has been found to relate to time of first alcohol use, drinking level, characteristics that may relate to drinking such as suicide attempts, and self-reported health status (144,147,148), and the associations may differ by gender (145,149). However, it is not clear if acculturation influences alcohol use patterns among those seeking treatment (150), or alters treatment response (151).

Asian Americans generally have the lowest levels of alcohol consumption and lowest prevalence of alcohol use disorders in the United States (16,19,134). There is evidence that this in part may relate to the discomfort and sensitivity to alcohol due to the physiologic effects of the variants in alcohol-metabolizing genes, such as aldehyde dehydrogenase genotypes (152–155). There also is evidence that genetic heterogeneity may explain differences in rates of alcohol dependence among certain subgroups (156). However, as with Hispanic Americans, the Asian American population is composed of many subgroups with different backgrounds and cultural drinking patterns (157,158), and nativity, immigration status, and ethnic identity have been found to be associated with differential risk of some alcohol and drug use disorders (159–161). Native Americans historically have had the highest prevalence of heavy drinking and alcohol use disorders in the United States (16,19,134,162), with the highest drinking-related death rates (163–164). However, it is not accurate to generalize to all Native American populations, as drinking practices are varied across tribal groups (165–167), and cultural factors as well as socioeconomic factors play a role (162,168).

Patterns of drug use and drug use disorders also vary by racial-ethnic group (19,38,78,169). However, less information is available for drug addiction than for alcohol use disorders among different ethnic populations, particularly for specific substances. Data from the NCS show that African Americans and Hispanics are less likely to report lifetime drug use relative to whites but are more likely to have persistent dependence once the disorder develops (78). Reports using data from the 2011 NSDUH indicate that the occurrence of past year illicit drug abuse and dependence was highest among Native Americans (classified as American Indian or Alaskan Native in the NSDUH) and lowest for the Asian American subgroup (19). Similar relationships were documented in analyses of 12-month prevalence using the NESARC (134). However, there is limited evidence for differences in incidence of drug disorders by race–ethnicity as assessed with the prospective NESARC data (35). In assessments of specific nonmedical prescription drug use and disorders (sedative, tranquilizer, opioid, and amphetamine), relative to Whites, the Black, Asian, and Hispanic subgroups were less likely and Native Americans more likely to report use and to have lifetime abuse or dependence (170). As with alcohol disorders, prevalence also varies by nativity, immigration status, country of origin, and degree of acculturation (160,171–174), and these influences relate to substance use in the United States (160,171,172) as well as other nations globally (175). It also has been shown that the relationship of some risk factors among adolescents (such as low family pride, depressed mood, and low self-esteem) differs by ethnic group (147). Comorbid patterns of substance use disorders (134,176,177) and consequences of drug use, such as risk of arrest, also vary by race-ethnicity (178). Correlates and specific patterns of substance use may differ for specific race–ethnic groups (179). Furthermore, there are race–ethnicity differences in service utilization and criminal justice system involvement as well as racial disparities in treatment for substance use disorders (180–183). The evaluation of race and its association with addiction is complex. As indicated in the discussion on alcohol disorders, when examining illicit drug use patterns among different ethnic groups, consideration should be given to socioeconomic characteristics. For example, one study found that, although national survey data indicated a higher prevalence of crack cocaine smoking among some ethnic minorities, when area of neighbor-hood residence was taken into account, differences in the prevalence of drug use between racial groups were attenuated (131). Another found that racial–ethnic disparities in treatment for substance abuse were reduced after adjustment for socioeconomic status and criminal history (180). Addressing socioeconomic disparities and financial barriers to treatment, as well as cultural competency in treatment settings, should help to improve treatment outcomes across race–ethnicity groups (184–188).

Family History

Alcohol disorders cluster in families (189,190), and family history of dependence may predict the severity of the disorder in probands (191). Twin studies (192–195), adoption and cross-fostering studies (196–198) have attempted to answer the question of whether such familial relationships are the result of genetic transmission or a shared environment. Recent genome-wide association studies may provide evidence for specific genes that may be involved in alcohol dependence (199,200). There also have been genetic investigations of enzymes involved in alcohol metabolism (201–203) and studies of genes involved with neurotransmission as related to alcohol dependence (204–207). Although many studies have indicated a possible genetic relationship for alcohol dependence, the association is complex. Evidence indicates that approximately half of the risk may be attributed to genetic influences (192

Stay updated, free articles. Join our Telegram channel

Join