A Life Course Approach to Understanding Racial/Ethnic Differences in Transitions Into and Out of Alcohol Problems

A Life Course Approach to Understanding Racial/Ethnic Differences in Transitions Into and Out of... Abstract Aims Alcohol problems are most prevalent in young adulthood and decrease thereafter, but some studies find that racial/ethnic minorities have elevated alcohol risk beyond the 20s. This study examines racial/ethnic differences in the transitions into and out of alcohol problems, and whether these are explained by heavy drinking (HD), socioeconomic disadvantages and adult role transitions from the 20s to 30s. Short summary Racial/ethnic groups had similar risks for earlier onset and recurrence/persistence of alcohol problems, but Blacks were at significantly greater risk than Whites for later onset in the 30s. Cumulative poverty and heavy drinking explained away this disparity, and were risk factors for recurring/persistent problems. Methods Using data from the US National Longitudinal Survey of Youth 1979–1994 waves (n = 6098), past-year alcohol problems were measured in 1989 (mean age = 28) and in 1994 (mean age = 33) among drinkers. Patterns of alcohol problems were categorized as no problems, earlier onset in 20s/offset in 30s, later onset in 30s, and recurrence or persistence (at both time points). Multinomial regression models adjusted for demographics, cumulative poverty, HD and timing of social role transitions (marital, parental). Results Compared to Whites, Blacks and Hispanics had similar risks for earlier alcohol problems but greater risk for developing problems in their 30s (AORs = 1.69 and 1.27, respectively, for later onset versus no problems); however, only the Black-White disparity was statistically significant. This was eliminated after taking into account cumulative poverty and lifecourse HD. There were no racial/ethnic differences in risk for recurring/persistent alcohol problems, which were associated with greater cumulative poverty and HD. Conclusions While Whites appear to ‘age out’ of alcohol problems in their 30s, Blacks are at greater risk after young adulthood. These findings signal a need for interventions that target racial/ethnic minorities beyond young adulthood. INTRODUCTION Epidemiological studies consistently show that heavy drinking (HD) and alcohol problems are highest in young adulthood, and then decline with age (Dawson et al., 2005; Hasin et al., 2007; Williams et al., 2018). This decline is a positive finding, given the significant social and economic costs associated with alcohol-related problems (Centers for Disease Control and Prevention, 2011; Sacks et al., 2015). However, some research finds that the timing of the peak and decline of HD (Evans-Polce et al., 2015) and progression to alcohol problems appear to differ across racial and ethnic groups (Alvanzo et al., 2011; Lopez-Quintero et al., 2011). Non-Hispanic Whites and Hispanics reach their highest levels of HD in the 20s while non-Hispanic Blacks show comparatively lower (Muthén and Muthén, 2000) or similar (Grant et al., 2012) levels. Importantly, some studies suggest that Hispanics and Blacks have more enduring HD patterns than Whites (Caetano and Kaskutas, 1995; Mulia et al., 2017), and that racial/ethnic disparities in alcohol problems become pronounced after young adulthood (Muthén and Muthén, 2000; Mulia et al., 2017). Compared to HD, few studies have examined disparities in alcohol problems across the life course and distinguished onset from recurrence (Grant et al., 2012). Cross-sectional research has suggested less remission of alcohol problems among minorities (Dawson et al., 2005), consistent with results from one-time follow-up studies to the US National Alcohol Survey (NAS) and National Epidemiologic Survey of Alcohol and Related Conditions (NESARC). In the latter, disparities in alcohol problems varied by age group and according to onset versus recurrence, but show a pattern of greater minority risk at older ages (Caetano and Kaskutas, 1996; Grant et al., 2012). Longitudinal studies are needed to confirm these findings and to attempt to explain them. If racial/ethnic differences in alcohol risk change over the life course, such that the lower or similar risk of minorities at younger ages reverses and gives way to greater risk at older ages, then a pressing question to address is why the racial/ethnic patterning of alcohol problems during this developmental period changes. Conceptual framework To understand racial/ethnic differences in alcohol problems, this study applies a life course framework and cumulative disadvantage theory (Dannefer, 2003; Elder et al., 2003). A key life course principle concerns the existence and timing of turning points which can alter behavioral trajectories (Elder et al., 2003). Prior studies have documented an ‘aging out’ of HD and alcohol problems during young adulthood, coinciding with the transition into adult social roles of marriage and parenthood (Bachman et al., 1997b; Dawson et al., 2006). However, the change in alcohol risk may depend on the timing of this role transition. Early transitions into adulthood have been associated with lower socioeconomic status (SES) (Furstenberg, 2008; Lui et al., 2014) and an increased risk for alcohol problems. While marriage is often associated with reduced risk of alcohol problems, possibly reflecting both selection of non-problem drinkers into marriage as well as a reduction of problem drinking after marriage (Miller-Tutzauer et al., 1991; Bachman et al., 1997b), getting married early can interfere with educational attainment and result in marital distress and subsequent divorce that increases alcohol risk (Kretsch and Harden, 2014). Similarly, while becoming a parent generally has been shown to reduce HD, early parenthood has been associated with social and economic stresses (Pearlin et al., 2005) which may lead to increased alcohol use (Yang et al., 2007). Surprisingly, studies of ‘aging out’ tend to overlook racial/ethnic differences in the transition to adult social roles, even though Blacks have lower marriage rates and earlier onset of parenthood than Whites (Raley and Sweeney, 2009; Kim and Raley, 2015), and Hispanics often transition to marriage and parenthood earlier than Whites (Vega, 1990; Oropesa and Landale, 2004). Given these timing differences, adult role transitions could have varied effects on alcohol problems across racial/ethnic groups and might help to account for alcohol-related disparities. A second life course principle relates to the ‘long arm of childhood’ and its influence on social and health outcomes (Elder et al., 2003). Early life experiences and the accumulation of disadvantages can have a profound effect on health behaviors and outcomes later in the life course (Dannefer, 2003), including heavier drinking in adulthood (Cerdá et al., 2011), and escalating risk for alcohol problems beyond young adulthood (Muthén and Muthén, 2000). Importantly, the accumulation of disadvantage across the life course and even across generations is greater for Blacks and some Hispanics than Whites (Darity and Nicholson, 2005; Shuey and Willson, 2008; Kochhar et al., 2011). Inequalities between racial/ethnic minorities and Whites increase with age such that greater wealth gaps are apparent at older ages (Brown, 2016). Thus, there is a question of whether the cumulative effects of disadvantage place racial/ethnic minorities at greater risk for alcohol problems than Whites as age increases, and even whether cumulative disadvantage trumps the more proximate effects of adult social role transitions. In this study, we examine the racial/ethnic differences in the onset, offset and recurrence or persistence of alcohol problems during the 20s and 30s. We hypothesize that racial/ethnic differences may be partly explained by differences in cumulative exposure to socioeconomic disadvantages from adolescence to the 30s, HD patterns and transitions in adult social roles. Specifically, we expect that racial/ethnic minorities will experience later onset and greater recurrence of alcohol problems than Whites, and that cumulative disadvantage will be more influential for racial/ethnic minorities than adult social roles in explaining the later onset of problems. METHOD Data Data are from the National Longitudinal Survey of Youth 1979 (NLSY79) cohort, an ongoing study by the US Department of Labor. The NLSY79 used a stratified, clustered design to select a nationally representative sample of individuals born in 1957–1964. With a 90% response rate in 1979, the initial sample size included 11,406 civilian participants aged 14–21. Interviews were conducted annually from 1979 to 1994, and survey retention up to 1994 was ~90% (U.S. Bureau of Labor Statistics, 2012). The sample was restricted to past-year drinkers who participated in the 1989 and 1994 interviews (when the dependent variable was measured, N = 6098). By 1994, participants were ages 30–37. Supplemental analysis (not shown here) indicated some differences between the analytic sample and those excluded from this study. Blacks, Hispanics, older respondents and those with lower SES were less likely to be included in the analytic sample than Whites, younger respondents and persons with higher SES. Measures Dependent variable Alcohol problems In 1989 and 1994, respondents were asked about past-year alcohol dependence symptoms including tolerance, withdrawal, drinking longer/larger quantities, trying to quit/control drinking, cutting down on activities to drink, spending a lot of time drinking/getting over drinking and continuing to use despite health/emotional problems. For sufficient power, we used a cutpoint of 2+ symptoms, which is similar to DSM-IV criteria for subclinical alcohol dependence (American Psychiatric Association, 1994), and consistent with DSM-V’s threshold used to define mild alcohol use disorder (American Psychiatric Association, 2013). The 1989 and 1994 measures were combined to capture transitions into and out of alcohol problems: (a) No problems in both time points, (b) 1989 onset and 1994 offset (hereafter referred to as earlier onset/offset from the 20s to 30s), (c) onset in 1994 (hereafter referred to as later onset in the 30s) and (d) recurrence or persistence in 1989 and 1994 (hereafter referred to as recurrence/persistence). Because alcohol problems were only captured at two time points, our measure overlooks those who may have experienced symptoms prior to 1989. We could not distinguish between recurring problems (separated by periods of remission) from chronic or persistent problems (with no remission between time points), and so we label the last group as recurrence/persistence. Primary independent variable Race/Ethnicity Respondents’ self-reported racial/ethnic origin (1979 survey) was categorized into (a) non-Hispanic White, (b) non-Hispanic Black, (c) Hispanic and (d) Other groups including Asian, Hawaiian/Pacific Islander, American Indian and unspecified. Explanatory factors Demographics/Socioeconomic status Demographics include sex and age. We created three measures of cumulative poverty (proportion of time spent in poverty based on yearly NLSY poverty measures defined under US federal poverty guidelines) to capture early poverty (1979–1984), early through young adult poverty (1979–1988), and cumulative poverty (1979–1994). Educational attainment by age 25 was categorized into less than high school, high school degree or equivalent, some college, and college degree or higher. Life course drinking Early onset of drinking at least one to two times a month before age 15 was used as a marker of higher risk for alcohol problems (Muthén and Muthén, 2000; Hingson et al., 2006). In the 1982–1984, 1988–1989 and 1994 surveys, respondents reported past-month frequency of drinking 6+ drinks on same occasion. Frequent HD was defined as 6+ drinking at least four to five times in the past month. Two variables were created to capture the number of years of frequent HD at two different periods: early HD (1982–1984) and later HD (1988–89). We created a separate dichotomous variable to capture frequent HD in 1994, current HD (yes/no). Marriage and parenthood Transitions into adult social roles were created according to (a) early onset, (b) young adulthood and (c) concurrent transitions. Early marriage was defined as being married before the age of 20 (Kretsch and Harden, 2014). Early parenthood was defined as having a first child before the age of 23 based on a past study that found significant effects on binge drinking (Wolfe, 2009). In young adulthood, marital and parenting statuses were created from 1988 survey (status before the outcome was measured). Marital status included (a) Never married, (b) Married and (c) Separated/Divorced/Widowed. Parenting status was based on having a child. Finally, role transitions that occurred between 1989 and 1994 (risk period for onset/offset of alcohol problems) were included. The 1989–1994 marital transitions were coded as (a) None, (b) Marital formation and (c) Marital dissolution (separated/divorced/widowed). Multiple marriage transitions were possible and evident within the 5-year period, in which case we selected the transition that occurred most proximal to 1994. Results could differ if we accounted for multiple transitions in our analysis. Parenting transition was captured as having a first child between 1989 and 1994. Analysis Descriptive analysis compared racial/ethnic differences in alcohol problems and the explanatory variables. This provided information on differential exposure of explanatory factors across racial/ethnic groups, and by sex. Next, a series of regression models were conducted. Multinomial logistic regression was used to compare the four categories of alcohol problems: no problems (reference), earlier onset/offset, later onset and recurrence/persistence. For analyses of recurrence/persistence, earlier onset/offset was the referent rather than no problems in order to identify potential factors for understanding why some continue to experience problems while others do not. Model 1 served as the based model with race/ethnicity and demographics (i.e. sex and age). Model 2 accounts for differential exposures early in the life course from adolescence (1979; mean age 17.6) to early 20 s (1984; mean age 22.6). Model 3 considers factors occurring in the mid-to-late 20s by 1988 (mean age 27.1). Model 4 focuses on the most proximate influences that may explain higher risk of later onset or recurrence/persistence (1994, mean age 33.0). Building on model 4, the final model 5 includes concurrent effects of HD (1994). To further examine the role of explanatory factors (specifically, cumulative disadvantage, HD and adult social role transitions), we tested interactions of race/ethnicity with these factors when race/ethnicity was a significant predictor of alcohol problems. All statistical analyses were conducted with Stata v.13.0 (StataCorp., 2013). The Stata survey command, -svy-, was applied to account for stratification and clustering, and sample weights were used to correct for non-probability sampling, attrition over time and population generalizability. RESULTS Descriptive statistics Fifteen percent of the total sample experienced two or more alcohol problems in the past year either in their 20s or 30s. Among this sample, 6.2% transitioned into and out of alcohol problems (i.e. earlier onset/offset), 5.2% had a later onset in the 30s, and 3.2% had recurring/persistent problems in the 20s and 30s. As shown in Table 1, Whites and Hispanics had higher rates of young adult onset/offset than later onset; by contrast, Blacks had higher rates of later onset than earlier onset/offset, and compared to Whites and Hispanics. There were no significant racial/ethnic differences in alcohol problems among women (not shown). But among men, there were significant differences in later onset for Blacks compared to Whites (10.7 versus 6.0%; P < 0.01), and a marginal Hispanic-White difference (9.1%; P = 0.052). Table 1. Descriptive statistics by race/ethnicity, NLSY 1979, n = 6098 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 aBlack-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. bHispanic-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. cProportion of time spent in poverty. Table 1. Descriptive statistics by race/ethnicity, NLSY 1979, n = 6098 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 aBlack-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. bHispanic-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. cProportion of time spent in poverty. Socioeconomic status As expected, differential life course exposure to poverty was stark, with Blacks and Hispanics spending more time in poverty than Whites (Table 1). By age 25, Blacks and Hispanics were less likely to have a high school or college degree than Whites. Life course drinking Whites were less likely than Hispanics to begin drinking regularly before age 15. However, Whites had more years of frequent HD in their early 20s than either Blacks or Hispanics. These racial/ethnic differences in the number of years of HD disappeared in the late 20s, and by the 30s, Blacks and Hispanics were much more likely than Whites to engage in current HD. Adult roles We found racial/ethnic differences in the timing of marriage and parenthood. Marriage, whether early or during the 20s, was more common among Hispanics and Whites than Blacks. A higher proportion of Blacks experienced marital dissolution compared to Whites. As expected, early parenthood was greater among Blacks and Hispanics than Whites, a trend that continued into the mid-20s. Whites were more likely to become a parent later. These racial/ethnic trends were seen in both sexes, but compared to men, women were more likely to be a parent in the 20s and less likely to transition to parenthood at a later age. Findings from multivariate models Earlier onset of alcohol problems Table 2 presents multinomial regression coefficients showing the relative risk of earlier onset of problems in the mid-to-late 20s versus no problems. Blacks and Hispanics were not significantly different from Whites in their relative risk of earlier onset/offset. Both early and later HD were associated with increased risk for earlier onset/offset, as was having lower educational attainment. Table 2. Multinomial logistic regression model: early onset/offset of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Table 2. Multinomial logistic regression model: early onset/offset of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Later onset of alcohol problems Table 3 presents multinomial regression coefficients showing the risk of later onset of alcohol problems in the 30s compared to no problems. In Model 1, Blacks have higher relative risks than Whites for experiencing later onset compared to no problems. A statistically significant difference was not found for Hispanics versus Whites. After accounting for early poverty, HD and adult role transitions in Model 2, Blacks remained at higher risk for alcohol problems; however, this disparity disappeared when adjusting for these factors in young adulthood (mid-to-late 20s) in Model 3 and more proximal factors (during 30s) in Model 4. In the final model, cumulative poverty across all study years, lower educational attainment and current HD were significant risk factors for later onset. In addition, being married (versus single) by the mid-20s was significantly associated with 44% lower likelihood of later onset. However, none of the role transitions from the 20s to the 30s were significant. Table 3. Multinomial Logistic Regression Model: later onset in 30s of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Table 3. Multinomial Logistic Regression Model: later onset in 30s of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. We tested whether the effects of our explanatory factors on later onset differed by racial/ethnic groups by examining interactions. The effects of SES and HD on later onset were similar across racial/ethnic groups (i.e. non-significant interaction terms). The only significant interaction was found for marital status in the 20s, specifically for Blacks and Hispanics who were separated/divorced/widowed (PBlack = 0.012 and PHispanic = 0.035), indicating that marital dissolution was associated with a lower risk of alcohol problems in Blacks and Hispanics compared to those who were married or single. Recurrence/persistence of alcohol problems Table 4 presents the findings for the small subset who experienced recurring/persistent problems in their 20s and 30s compared to those who transitioned out of alcohol problems in their 30s. Controlling for age and sex, Blacks and Hispanics had higher relative risks than Whites for recurrence/persistence (versus transitioning out) of alcohol problems but these relationships were not significant. Proximal factors were important predictors of recurrence/persistence (Models 4 and 5), in particular current HD and cumulative poverty. Neither marriage or parenthood, nor transitions into these roles had a significant effect on recurrence/persistence. Table 4. Multinomial Logistic Regression Model: recurrence/persistence of alcohol problems (versus early onset/offset), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Table 4. Multinomial Logistic Regression Model: recurrence/persistence of alcohol problems (versus early onset/offset), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. DISCUSSION Few studies have examined racial/ethnic differences in the transitions in and out of alcohol problems, and even fewer have investigated explanatory factors that might account for group differences. Using longitudinal data from a US general population sample, our study revealed a Black-White disparity in later onset of alcohol problems in the 30s, but not in young adult onset/offset nor in recurrence/persistence of alcohol problems in the 30s. Guided by a life course framework and cumulative disadvantage theory, our results further indicated that chronic poverty is an important predictor of alcohol problems above and beyond transitions into adult social roles of marriage and parenthood. We discuss these key findings and their implications for alcohol-related disparities below. A significant portion of the racial/ethnic differences was found in the exposure to factors expected to relate to the risk for alcohol problems. Blacks and Hispanics spent more time in poverty, had lower education, and had different patterns in the acquisition and timing of adult social roles than Whites. These findings are not new. Demographic trends consistently show greater disadvantages in racial/ethnic groups (more poverty among Blacks, and lower education for Blacks and Hispanics), and marital differences (lower rates among Blacks versus Whites and Hispanics) and parenting transitions (Blacks and Hispanics become parents earlier than Whites) (Aughinbaugh et al., 2013; Maralani, 2013; Kim and Raley, 2015). Our study suggests that these differences in exposures and timing of risk and protective factors appear to contribute to racial/ethnic differences in transitions into and out of alcohol problems. Much of the alcohol research overlooks how ‘aging out’ effects of marriage or parenthood, or cumulative poverty differ across racial/ethnic groups. These disadvantages and earlier timing of developmental turning points underscore the need to recognize these different lived experiences and the implications for alcohol problems among minorities. The key disparity finding of Black-White differences in later onset of alcohol problems disappeared after accounting for explanatory factors. Cumulative poverty and lower education were consistent and strong predictors of later onset. This later onset might be related to the accumulation of disadvantage that builds up across the life course. Notably, the transition from the 20s to 30s is a critical period for achieving SES, and the 30s is often a time when SES stabilizes (Furstenberg, 2008). If minorities have trouble achieving socioeconomic stability in the 30s, compounded with early disadvantages, the stresses and realities associated with this may place them at higher risk for HD and alcohol problems. Future studies should take a longitudinal perspective on changes in SES to see if this influences continued HD and later onset of alcohol problems among racial/ethnic groups. Traditional markers of adulthood, such as getting married and having children, have been shown to reduce the risk of alcohol problems (Miller-Tutzauer et al., 1991; Bachman et al., 1997a; Dawson et al., 2006). While early parenting has been associated with reduced alcohol consumption, this protective effect wanes over time (Wolfe, 2009). Moreover, taking into account Blacks’ and Hispanics’ greater cumulative disadvantage, earlier transitions to marriage and parenthood might result in greater economic stress and set the stage for more HD and later alcohol problems. We found that Blacks and Hispanics have less HD in the 20s compared to Whites, but significantly greater HD in their 30s. This finding is similar to other studies showing a crossover in risk, whereby racial/ethnic minorities are at lower risk for adverse health behaviors in adolescence but higher risk in early adulthood compared to Whites (Brown et al., 2007; Lawrence et al., 2014). Differential exposures also operated in unexpected ways to explain the later onset of alcohol problems for racial/ethnic minorities. In our multivariate models, being married in the mid-20s, but not marital transitions in the 30s, was a consistent and significant protective factor for later onset across all groups. However, the significant interaction found between race and marital dissolution indicates that marital dissolution may have less of an adverse effect for minorities. While Blacks and Hispanics were more likely to be separated/divorced/widowed in the 20s than Whites, this status afforded some protection against later onset of alcohol problems compared to being never married or even married. Marital dissolution has been associated with harmful effects (Kretsch and Harden, 2014), which was consistent for Whites in this study, but could getting out of a marriage may be less harmful for racial/ethnic minorities? This finding warrants further research to examine the meaning of marriage and marital dissolution in relation to alcohol problems in Blacks, Hispanics, and Whites. We did not find racial/ethnic disparities in earlier onset nor recurrence/persistence of alcohol problems. These findings conflict with NESARC data showing Whites are more likely to experience past-year alcohol dependence in the 20s (Grant et al., 2004) and minorities are more likely to experience recurrence (Grant et al., 2012). The discrepant finding may be related to measurement (capturing DSM-IV alcohol dependence versus alcohol problems) or cohort effects, given that the young adults who were part of the larger NESARC sample of US adults 18+ were born in the late 1970–1980s while this study sampled young adults born 1957–1964. Future research is needed to examine racial/ethnic disparities in alcohol dependence, based on DSM-criteria, longitudinally over time. These study findings should be considered in light of the following limitations. First, findings are only generalizable to the U.S. population born between 1957 and 1964, reflective of the post-baby boomer cohort. Second, the outcome measure was derived from data that are more than 20 years old. However, national longitudinal data permitting a life course perspective and analysis of alcohol problems are rare; most US national data are cross-sectional requiring retrospective reports or if prospective, are limited to several years of data (versus NLSY’s 16 years of data up to 1994). Third, the measures for alcohol problems, two symptoms experienced in the same 12-month period consistent with DSM-IV criteria, are likely to capture mild dependence. Furthermore, additional time points to capture recurrence (repeated problems with remission) versus persistence (continued symptoms without remission) of problems are needed and would offer insight into clinical and public health interventions. Fourth, the NLSY79 surveys assessed past month’s frequency of 6+ drinks on occasion, which significantly exceeds that of current NIAAA HD guidelines using sex-specific 5+/4+ drinks on occasion (National Institute on Alcohol Abuse and Alcoholism, 2005); and thus risks under-identifying HD in the sample. Also, while we found that HD was a significant predictor of alcohol problems, the observed relationship may be stronger than had we been able to use the 5+/4+ HD criteria. Fifth, due to the smaller sample sizes of Blacks and Hispanics compared to Whites, the model estimates of alcohol problems that were more rare and, in particular, interactions of race by explanatory factors may be less reliable for Blacks and Hispanics and should be interpreted with caution. Also, while we include other racial/ethnic groups in our analyses, we do not interpret the findings given the heterogeneity of this group. Finally, we did not have sufficient power to conduct sex-stratified analyses. While we controlled for sex, past studies have shown significant sex differences in alcohol problems, with men at higher risk (Grant et al., 2012). Poverty and adult social roles can have differential effects on alcohol problems across sex. Women have more risks associated with poverty experiences, and men have greater benefits from marital formation than women while women have less risk from marital dissolution (Zick and Smith, 1991; Raley and Sweeney, 2009). Future studies should consider sex in racial/ethnic differences of alcohol problems. Overall, alcohol problems are more common in young adulthood and decline over time. However, our study shows that Blacks are more likely than Whites to experience a later onset of alcohol problems in the 30s, due in part to differences in the accumulation of disadvantage and HD patterns, both of which have a stronger effect on risk for later onset and recurrence/persistence of alcohol problems than do social role transitions. These study findings suggest the benefits of interventions to alleviate social disadvantages across the life course (e.g. policy efforts to reduce poverty and increase educational opportunities) and of interventions targeting racial/ethnic minorities during their late 20s and 30s to prevent HD and alcohol problems. Universal prevention strategies to reduce HD and alcohol-related problems are abundant for young adults. However, if Blacks are transitioning into alcohol problems while Whites are transitioning out of alcohol problems, this points to a clear need for targeted prevention strategies beyond young adulthood to address alcohol-related disparities. ACKNOWLEDGEMENTS The authors would like to thank the study team, Ms Edwina Williams, Drs Katherine Karriker-Jaffe, Jane Witbrodt and Sarah Zemore at the Alcohol Research Group, and Dr Margaret Ensminger at the Johns Hopkins Bloomberg School of Public Health for their feedback on study conceptualization and results. FUNDING This work was supported by the National Institute on Alcohol Abuse and Alcoholism at the National Institutes of Health [Grant number R01AA022668 to N.M.]. CONFLICT OF INTEREST The authors declare that they have no conflict of interest. REFERENCES Alvanzo AA , Storr CL , La Flair L , et al. . ( 2011 ) Race/ethnicity and sex differences in progression from drinking initiation to the development of alcohol dependence . Drug Alcohol Depend 118 : 375 – 82 . Google Scholar CrossRef Search ADS PubMed American Psychiatric Association . ( 1994 ) DSM-IV: Diagnostic & Statistical Manual of Mental Disorders , 4th edn . Washington, DC : American Psychiatric Association . American Psychiatric Association . ( 2013 ) Diagnostic and Statistical Manual of Mental Disorders , 5th edn . Arlington, VA : DSM-5 . Aughinbaugh A , Robles O , Sun H . ( 2013 ) Marriage and Divorce: Patterns by Gender, Race, and Educational Attainment [Online]. http://www.bls.gov/opub/mlr/2013/article/pdf/marriage-and-divorce-patterns-by-gender-race-and-educational-attainment.pdf. Bachman JG , Wadsworth KN , O’Malley PM , et al. . ( 1997 a) Smoking, Drinking, and Drug Use in Young Adulthood: The Impacts of New Freedoms and New Responsibilities . Mahwah, NJ : Erlbaum . Bachman JG , Wadsworth KN , O’Malley PM , et al. . ( 1997 b) Marriage, divorce, and parenthood during the transition to young adulthood: impacts on drug use and abuse. In Schulenberg J , Maggs JL , Hurrelmann K (eds) . Health Risks and Developmental Transitions during Adolescence . New York : Cambridge University Press , 246 – 79 . Brown TH . ( 2016 ) Diverging fortunes: racial/ethnic inequality in wealth trajectories in middle and late life . Race Soc Prob 8 : 29 – 41 . Google Scholar CrossRef Search ADS Brown JS , Meadows SO , Elder GH . ( 2007 ) Race-ethnic inequality and psychological distress: depressive symptoms from adolescence to young adulthood . Dev Psychol 43 : 1295 – 311 . Google Scholar CrossRef Search ADS PubMed Bureau of Labor Statistics and U.S. Department of Labor. ( 2012 ) National Longitudinal Survey of Youth 1979 Cohort, 1979–2010 (Rounds 1–24). Archived by WebCite® at http://www.webcitation.org/6egSq0L6G] (20 January 2016, date last accessed). Columbus, OH. [Online]. http://www.bls.gov/nls/nlsy79.htm. Caetano R , Kaskutas LA . ( 1995 ) Changes in drinking patterns among Whites, Blacks and Hispanics: 1984–1992 . J Stud Alcohol 56 : 558 – 65 . Google Scholar CrossRef Search ADS PubMed Caetano R , Kaskutas LA . ( 1996 ) Changes in drinking problems among Whites, Blacks and Hispanics: 1984–1992 . Subst Use Misuse 31 : 1547 – 71 . Google Scholar CrossRef Search ADS PubMed Centers for Disease Control and Prevention (CDC). ( 2011 ) CDC Reports Excessive Alcohol Consumption Costs the U.S. $224 Billion in 2006. Archived by WebCite® at http://www.webcitation.org/6ED3xmU2H] (5 February 2013, date last accessed). Atlanta, GA. [Online]. http://www.cdc.gov/media/releases/2011/p1017_alcohol_consumption.html. Cerdá M , Johnson-Lawrence VD , Galea S . ( 2011 ) Lifetime income patterns and alcohol consumption: investigating the association between long- and short-term income trajectories and drinking . Soc Sci Med 73 : 1178 – 85 . Google Scholar CrossRef Search ADS PubMed Dannefer D . ( 2003 ) Cumulative advantage/disadvantage and the life course: cross-fertilizing age and social science theory . J Gerontol B Psychol Sci Soc Sci 58B : S327 – 37 . Google Scholar CrossRef Search ADS Darity W Jr , Nicholson MJ . ( 2005 ) Racial wealth inequality and the Black family . Afr Am Fam Life 78 : 85 . Dawson DA , Grant BF , Stinson FS , et al. . ( 2005 ) Recovery from DSM-IV alcohol dependence: United States, 2001–2002 . Addiction 100 : 281 – 92 . Google Scholar CrossRef Search ADS PubMed Dawson DA , Grant BF , Stinson FS , et al. . ( 2006 ) Maturing out of alcohol dependence: the impact of transitional life events . J Stud Alcohol 67 : 195 – 203 . Google Scholar CrossRef Search ADS PubMed Elder GH Jr , Johnson MK , Crosnoe R . ( 2003 ) The emergence and development of life course theory. In Mortimer JT , Shanahan MJ (eds) . Handbook of the Life Course . New York : Springer , 3 – 19 . Google Scholar CrossRef Search ADS Evans-Polce RJ , Vasilenko SA , Lanza ST . ( 2015 ) Changes in gender and racial/ethnic disparities in rates of cigarette use, regular heavy episodic drinking, and marijuana use: ages 14 to 32 . Addict Behav 41 : 218 – 22 . Google Scholar CrossRef Search ADS PubMed Furstenberg FF . ( 2008 ) The intersections of social class and the transition to adulthood. In Mortimer JT (ed) . Social Class and Transitions to Adulthood. New Directions for Child and Adolescent Development, 119 . San Francisco : Jossey-Bass , 1 – 10 . Grant BF , Dawson DA , Stinson FS , et al. . ( 2004 ) The 12-month prevalence and trends in DSM-IV alcohol abuse and dependence: United States, 1991–1992 and 2001–2002 . Drug Alcohol Depend 74 : 223 – 34 . Google Scholar CrossRef Search ADS PubMed Grant JD , Vergés A , Jackson KM , et al. . ( 2012 ) Age and ethnic differences in the onset, persistence and recurrence of alcohol use disorder . Addiction 107 : 756 – 65 . Google Scholar CrossRef Search ADS PubMed Hasin DS , Stinson FS , Ogburn E , et al. . ( 2007 ) Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions . Arch Gen Psychiatry 64 : 830 – 42 . Google Scholar CrossRef Search ADS PubMed Hingson RW , Heeren T , Winter MR . ( 2006 ) Age at drinking onset and alcohol dependence: age at onset, duration, and severity . Arch Pediatr Adolesc Med 160 : 739 – 46 . Google Scholar CrossRef Search ADS PubMed Kim Y , Raley RK . ( 2015 ) Race-ethnic differences in the non-marital fertility rates in 2006–2010 . Popul Res Policy Rev 34 : 141 – 59 . Google Scholar CrossRef Search ADS PubMed Kochhar R , Fry R , Taylor P. ( 2011 ) Wealth Gaps Rise to Record Highs Between Whites, Blacks and Hispanics. Archived by WebCite® at http://www.webcitation.org/6WDdIhIwD] (9 February 2015, date last accessed). Washington, DC. [Online]. http://www.pewsocialtrends.org/files/2011/07/SDT-Wealth-Report_7-26-11_FINAL.pdf. Kretsch N , Harden KP . ( 2014 ) Marriage, divorce, and alcohol use in young adulthood: a longitudinal sibling-comparison study . Emerg Adulthood 2 : 138 – 49 . Google Scholar CrossRef Search ADS Lawrence EM , Pampel FC , Mollborn S . ( 2014 ) Life course transitions and racial and ethnic differences in smoking prevalence . Adv Life Course Res 22 : 27 – 40 . Google Scholar CrossRef Search ADS PubMed Lopez-Quintero C , Pérez de los Cabos J , Hasin DS , et al. . ( 2011 ) Probability and predictors of transition from first use to dependence on nicotine, alcohol, cannabis, and cocaine: results of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) . Drug Alcohol Depend 115 : 120 – 30 . Google Scholar CrossRef Search ADS PubMed Lui CK , Chung PJ , Wallace SP , et al. . ( 2014 ) Social status attainment during the transition to adulthood . J Youth Adolesc 43 : 1134 – 50 . Google Scholar CrossRef Search ADS PubMed Maralani V . ( 2013 ) The demography of social mobility: Black-White differences in the process of educational reproduction . Am J Sociol 118 : 1509 – 58 . Google Scholar CrossRef Search ADS Miller-Tutzauer C , Leonard KE , Windle M . ( 1991 ) Marriage and alcohol use: a longitudinal study of maturing out . J Stud Alcohol 52 : 434 – 40 . Google Scholar CrossRef Search ADS PubMed Mulia N , Karriker-Jaffe KJ , Witbrodt J , et al. . ( 2017 a) Racial/ethnic differences in 30-year trajectories of heavy drinking in a nationally representative U.S. sample . Drug Alcohol Depend 170 : 133 – 41 . Google Scholar CrossRef Search ADS PubMed Mulia N , Tam T , Bond J , et al. . ( 2017 b) Racial/ethnic differences in life-course heavy drinking from adolescence to midlife . J Ethn Subst Abuse 17 : 167 – 86 . Google Scholar CrossRef Search ADS PubMed Muthén BO , Muthén LK . ( 2000 ) The development of heavy drinking and alcohol-related problems from ages 18 to 37 in a U.S. national sample . J Stud Alcohol 61 : 290 – 300 . Google Scholar CrossRef Search ADS PubMed National Institute on Alcohol Abuse and Alcoholism ( 2005 ) Helping patients who drink too much: a clinician’s guide. Updated 2005 Edition, revised October 2015 (NIH Publication No. 07–3769). Archived by WebCite® at http://www.webcitation.org/6jiIsTNiY] (12 August 2016, date last accessed). Rockville, MD. [Online]. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf. Oropesa RS , Landale NS . ( 2004 ) The future of marriage and hispanics . J Marriage Fam 66 : 901 – 20 . Google Scholar CrossRef Search ADS Pearlin LI , Schieman S , Fazio EM , et al. . ( 2005 ) Stress, health, and the life course: some conceptual perspectives . J Health Soc Behav 46 : 205 – 19 . Google Scholar CrossRef Search ADS PubMed Raley RK , Sweeney MM . ( 2009 ) Explaining race and ethnic variation in marriage: directions for future research . Race Soc Problems 1 : 132 – 42 . Google Scholar CrossRef Search ADS Sacks JJ , Gonzales KR , Bouchery EE , et al. . ( 2015 ) 2010 National and state costs of excessive alcohol consumption . Am J Prev Med 49 : e73 – 9 . Google Scholar CrossRef Search ADS PubMed Shuey KM , Willson AE . ( 2008 ) Cumulative disadvantage and Black-White disparities in life-course health trajectories . Res Aging 30 : 200 – 25 . Google Scholar CrossRef Search ADS StataCorp . ( 2013 ) Stata Statistical Software: Release 13.0 [Computer program]. Stata Corporation. (Accessed). Vega WA . ( 1990 ) Hispanic families in the 1980s: a decade of research . J Marriage Fam 52 : 1015 – 24 . Google Scholar CrossRef Search ADS Williams E , Mulia N , Karriker-Jaffe KJ , et al. . ( 2018 ) Changing racial/ethnic disparities in heavy drinking trajectories through young adulthood: a comparative cohort study . Alcoholism 42 : 135 – 43 . Google Scholar CrossRef Search ADS PubMed Wolfe JD . ( 2009 ) Age at first birth and alcohol use . J Health Soc Behav 50 : 395 – 409 . Google Scholar CrossRef Search ADS PubMed Yang S , Lynch JW , Raghunathan TE , et al. . ( 2007 ) Socioeconomic and psychosocial exposures across the life course and binge drinking in adulthood: population-based study . Am J Epidemiol 165 : 184 – 93 . Google Scholar CrossRef Search ADS PubMed Zick CD , Smith KR . ( 1991 ) Marital transitions, poverty, and gender differences in mortality . J Marriage Fam 53 : 327 – 36 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Alcohol and Alcoholism Oxford University Press

A Life Course Approach to Understanding Racial/Ethnic Differences in Transitions Into and Out of Alcohol Problems

Alcohol and Alcoholism , Volume Advance Article (4) – Mar 13, 2018

Loading next page...
 
/lp/ou_press/a-life-course-approach-to-understanding-racial-ethnic-differences-in-I0gYpx5YeU
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved.
ISSN
0735-0414
eISSN
1464-3502
D.O.I.
10.1093/alcalc/agy015
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims Alcohol problems are most prevalent in young adulthood and decrease thereafter, but some studies find that racial/ethnic minorities have elevated alcohol risk beyond the 20s. This study examines racial/ethnic differences in the transitions into and out of alcohol problems, and whether these are explained by heavy drinking (HD), socioeconomic disadvantages and adult role transitions from the 20s to 30s. Short summary Racial/ethnic groups had similar risks for earlier onset and recurrence/persistence of alcohol problems, but Blacks were at significantly greater risk than Whites for later onset in the 30s. Cumulative poverty and heavy drinking explained away this disparity, and were risk factors for recurring/persistent problems. Methods Using data from the US National Longitudinal Survey of Youth 1979–1994 waves (n = 6098), past-year alcohol problems were measured in 1989 (mean age = 28) and in 1994 (mean age = 33) among drinkers. Patterns of alcohol problems were categorized as no problems, earlier onset in 20s/offset in 30s, later onset in 30s, and recurrence or persistence (at both time points). Multinomial regression models adjusted for demographics, cumulative poverty, HD and timing of social role transitions (marital, parental). Results Compared to Whites, Blacks and Hispanics had similar risks for earlier alcohol problems but greater risk for developing problems in their 30s (AORs = 1.69 and 1.27, respectively, for later onset versus no problems); however, only the Black-White disparity was statistically significant. This was eliminated after taking into account cumulative poverty and lifecourse HD. There were no racial/ethnic differences in risk for recurring/persistent alcohol problems, which were associated with greater cumulative poverty and HD. Conclusions While Whites appear to ‘age out’ of alcohol problems in their 30s, Blacks are at greater risk after young adulthood. These findings signal a need for interventions that target racial/ethnic minorities beyond young adulthood. INTRODUCTION Epidemiological studies consistently show that heavy drinking (HD) and alcohol problems are highest in young adulthood, and then decline with age (Dawson et al., 2005; Hasin et al., 2007; Williams et al., 2018). This decline is a positive finding, given the significant social and economic costs associated with alcohol-related problems (Centers for Disease Control and Prevention, 2011; Sacks et al., 2015). However, some research finds that the timing of the peak and decline of HD (Evans-Polce et al., 2015) and progression to alcohol problems appear to differ across racial and ethnic groups (Alvanzo et al., 2011; Lopez-Quintero et al., 2011). Non-Hispanic Whites and Hispanics reach their highest levels of HD in the 20s while non-Hispanic Blacks show comparatively lower (Muthén and Muthén, 2000) or similar (Grant et al., 2012) levels. Importantly, some studies suggest that Hispanics and Blacks have more enduring HD patterns than Whites (Caetano and Kaskutas, 1995; Mulia et al., 2017), and that racial/ethnic disparities in alcohol problems become pronounced after young adulthood (Muthén and Muthén, 2000; Mulia et al., 2017). Compared to HD, few studies have examined disparities in alcohol problems across the life course and distinguished onset from recurrence (Grant et al., 2012). Cross-sectional research has suggested less remission of alcohol problems among minorities (Dawson et al., 2005), consistent with results from one-time follow-up studies to the US National Alcohol Survey (NAS) and National Epidemiologic Survey of Alcohol and Related Conditions (NESARC). In the latter, disparities in alcohol problems varied by age group and according to onset versus recurrence, but show a pattern of greater minority risk at older ages (Caetano and Kaskutas, 1996; Grant et al., 2012). Longitudinal studies are needed to confirm these findings and to attempt to explain them. If racial/ethnic differences in alcohol risk change over the life course, such that the lower or similar risk of minorities at younger ages reverses and gives way to greater risk at older ages, then a pressing question to address is why the racial/ethnic patterning of alcohol problems during this developmental period changes. Conceptual framework To understand racial/ethnic differences in alcohol problems, this study applies a life course framework and cumulative disadvantage theory (Dannefer, 2003; Elder et al., 2003). A key life course principle concerns the existence and timing of turning points which can alter behavioral trajectories (Elder et al., 2003). Prior studies have documented an ‘aging out’ of HD and alcohol problems during young adulthood, coinciding with the transition into adult social roles of marriage and parenthood (Bachman et al., 1997b; Dawson et al., 2006). However, the change in alcohol risk may depend on the timing of this role transition. Early transitions into adulthood have been associated with lower socioeconomic status (SES) (Furstenberg, 2008; Lui et al., 2014) and an increased risk for alcohol problems. While marriage is often associated with reduced risk of alcohol problems, possibly reflecting both selection of non-problem drinkers into marriage as well as a reduction of problem drinking after marriage (Miller-Tutzauer et al., 1991; Bachman et al., 1997b), getting married early can interfere with educational attainment and result in marital distress and subsequent divorce that increases alcohol risk (Kretsch and Harden, 2014). Similarly, while becoming a parent generally has been shown to reduce HD, early parenthood has been associated with social and economic stresses (Pearlin et al., 2005) which may lead to increased alcohol use (Yang et al., 2007). Surprisingly, studies of ‘aging out’ tend to overlook racial/ethnic differences in the transition to adult social roles, even though Blacks have lower marriage rates and earlier onset of parenthood than Whites (Raley and Sweeney, 2009; Kim and Raley, 2015), and Hispanics often transition to marriage and parenthood earlier than Whites (Vega, 1990; Oropesa and Landale, 2004). Given these timing differences, adult role transitions could have varied effects on alcohol problems across racial/ethnic groups and might help to account for alcohol-related disparities. A second life course principle relates to the ‘long arm of childhood’ and its influence on social and health outcomes (Elder et al., 2003). Early life experiences and the accumulation of disadvantages can have a profound effect on health behaviors and outcomes later in the life course (Dannefer, 2003), including heavier drinking in adulthood (Cerdá et al., 2011), and escalating risk for alcohol problems beyond young adulthood (Muthén and Muthén, 2000). Importantly, the accumulation of disadvantage across the life course and even across generations is greater for Blacks and some Hispanics than Whites (Darity and Nicholson, 2005; Shuey and Willson, 2008; Kochhar et al., 2011). Inequalities between racial/ethnic minorities and Whites increase with age such that greater wealth gaps are apparent at older ages (Brown, 2016). Thus, there is a question of whether the cumulative effects of disadvantage place racial/ethnic minorities at greater risk for alcohol problems than Whites as age increases, and even whether cumulative disadvantage trumps the more proximate effects of adult social role transitions. In this study, we examine the racial/ethnic differences in the onset, offset and recurrence or persistence of alcohol problems during the 20s and 30s. We hypothesize that racial/ethnic differences may be partly explained by differences in cumulative exposure to socioeconomic disadvantages from adolescence to the 30s, HD patterns and transitions in adult social roles. Specifically, we expect that racial/ethnic minorities will experience later onset and greater recurrence of alcohol problems than Whites, and that cumulative disadvantage will be more influential for racial/ethnic minorities than adult social roles in explaining the later onset of problems. METHOD Data Data are from the National Longitudinal Survey of Youth 1979 (NLSY79) cohort, an ongoing study by the US Department of Labor. The NLSY79 used a stratified, clustered design to select a nationally representative sample of individuals born in 1957–1964. With a 90% response rate in 1979, the initial sample size included 11,406 civilian participants aged 14–21. Interviews were conducted annually from 1979 to 1994, and survey retention up to 1994 was ~90% (U.S. Bureau of Labor Statistics, 2012). The sample was restricted to past-year drinkers who participated in the 1989 and 1994 interviews (when the dependent variable was measured, N = 6098). By 1994, participants were ages 30–37. Supplemental analysis (not shown here) indicated some differences between the analytic sample and those excluded from this study. Blacks, Hispanics, older respondents and those with lower SES were less likely to be included in the analytic sample than Whites, younger respondents and persons with higher SES. Measures Dependent variable Alcohol problems In 1989 and 1994, respondents were asked about past-year alcohol dependence symptoms including tolerance, withdrawal, drinking longer/larger quantities, trying to quit/control drinking, cutting down on activities to drink, spending a lot of time drinking/getting over drinking and continuing to use despite health/emotional problems. For sufficient power, we used a cutpoint of 2+ symptoms, which is similar to DSM-IV criteria for subclinical alcohol dependence (American Psychiatric Association, 1994), and consistent with DSM-V’s threshold used to define mild alcohol use disorder (American Psychiatric Association, 2013). The 1989 and 1994 measures were combined to capture transitions into and out of alcohol problems: (a) No problems in both time points, (b) 1989 onset and 1994 offset (hereafter referred to as earlier onset/offset from the 20s to 30s), (c) onset in 1994 (hereafter referred to as later onset in the 30s) and (d) recurrence or persistence in 1989 and 1994 (hereafter referred to as recurrence/persistence). Because alcohol problems were only captured at two time points, our measure overlooks those who may have experienced symptoms prior to 1989. We could not distinguish between recurring problems (separated by periods of remission) from chronic or persistent problems (with no remission between time points), and so we label the last group as recurrence/persistence. Primary independent variable Race/Ethnicity Respondents’ self-reported racial/ethnic origin (1979 survey) was categorized into (a) non-Hispanic White, (b) non-Hispanic Black, (c) Hispanic and (d) Other groups including Asian, Hawaiian/Pacific Islander, American Indian and unspecified. Explanatory factors Demographics/Socioeconomic status Demographics include sex and age. We created three measures of cumulative poverty (proportion of time spent in poverty based on yearly NLSY poverty measures defined under US federal poverty guidelines) to capture early poverty (1979–1984), early through young adult poverty (1979–1988), and cumulative poverty (1979–1994). Educational attainment by age 25 was categorized into less than high school, high school degree or equivalent, some college, and college degree or higher. Life course drinking Early onset of drinking at least one to two times a month before age 15 was used as a marker of higher risk for alcohol problems (Muthén and Muthén, 2000; Hingson et al., 2006). In the 1982–1984, 1988–1989 and 1994 surveys, respondents reported past-month frequency of drinking 6+ drinks on same occasion. Frequent HD was defined as 6+ drinking at least four to five times in the past month. Two variables were created to capture the number of years of frequent HD at two different periods: early HD (1982–1984) and later HD (1988–89). We created a separate dichotomous variable to capture frequent HD in 1994, current HD (yes/no). Marriage and parenthood Transitions into adult social roles were created according to (a) early onset, (b) young adulthood and (c) concurrent transitions. Early marriage was defined as being married before the age of 20 (Kretsch and Harden, 2014). Early parenthood was defined as having a first child before the age of 23 based on a past study that found significant effects on binge drinking (Wolfe, 2009). In young adulthood, marital and parenting statuses were created from 1988 survey (status before the outcome was measured). Marital status included (a) Never married, (b) Married and (c) Separated/Divorced/Widowed. Parenting status was based on having a child. Finally, role transitions that occurred between 1989 and 1994 (risk period for onset/offset of alcohol problems) were included. The 1989–1994 marital transitions were coded as (a) None, (b) Marital formation and (c) Marital dissolution (separated/divorced/widowed). Multiple marriage transitions were possible and evident within the 5-year period, in which case we selected the transition that occurred most proximal to 1994. Results could differ if we accounted for multiple transitions in our analysis. Parenting transition was captured as having a first child between 1989 and 1994. Analysis Descriptive analysis compared racial/ethnic differences in alcohol problems and the explanatory variables. This provided information on differential exposure of explanatory factors across racial/ethnic groups, and by sex. Next, a series of regression models were conducted. Multinomial logistic regression was used to compare the four categories of alcohol problems: no problems (reference), earlier onset/offset, later onset and recurrence/persistence. For analyses of recurrence/persistence, earlier onset/offset was the referent rather than no problems in order to identify potential factors for understanding why some continue to experience problems while others do not. Model 1 served as the based model with race/ethnicity and demographics (i.e. sex and age). Model 2 accounts for differential exposures early in the life course from adolescence (1979; mean age 17.6) to early 20 s (1984; mean age 22.6). Model 3 considers factors occurring in the mid-to-late 20s by 1988 (mean age 27.1). Model 4 focuses on the most proximate influences that may explain higher risk of later onset or recurrence/persistence (1994, mean age 33.0). Building on model 4, the final model 5 includes concurrent effects of HD (1994). To further examine the role of explanatory factors (specifically, cumulative disadvantage, HD and adult social role transitions), we tested interactions of race/ethnicity with these factors when race/ethnicity was a significant predictor of alcohol problems. All statistical analyses were conducted with Stata v.13.0 (StataCorp., 2013). The Stata survey command, -svy-, was applied to account for stratification and clustering, and sample weights were used to correct for non-probability sampling, attrition over time and population generalizability. RESULTS Descriptive statistics Fifteen percent of the total sample experienced two or more alcohol problems in the past year either in their 20s or 30s. Among this sample, 6.2% transitioned into and out of alcohol problems (i.e. earlier onset/offset), 5.2% had a later onset in the 30s, and 3.2% had recurring/persistent problems in the 20s and 30s. As shown in Table 1, Whites and Hispanics had higher rates of young adult onset/offset than later onset; by contrast, Blacks had higher rates of later onset than earlier onset/offset, and compared to Whites and Hispanics. There were no significant racial/ethnic differences in alcohol problems among women (not shown). But among men, there were significant differences in later onset for Blacks compared to Whites (10.7 versus 6.0%; P < 0.01), and a marginal Hispanic-White difference (9.1%; P = 0.052). Table 1. Descriptive statistics by race/ethnicity, NLSY 1979, n = 6098 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 aBlack-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. bHispanic-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. cProportion of time spent in poverty. Table 1. Descriptive statistics by race/ethnicity, NLSY 1979, n = 6098 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 Survey year(s) (age range) White Black Hispanic Other P-value Weighted % (Unweighted n) 65.4 (2759) 11.8 (1573) 5.4 (993) 17.5 (773) Dependent variable Alcohol problems (%) 1989 and 1994 (25–37)  No alcohol problems 85.3 83.2 81.8b 88.5 0.001  Early onset/offset 6.8 5.4 8.0 3.9  Later onset in 30s 4.8 7.9a 6.0 4.5  Recurrence/persistence 3.1 3.5 4.1 3.1 Demographics  Sex (%) 1979 (14–21)   Female 49.4 48.9 47.8 51.9 0.323   Male 50.6 51.1 52.2 48.1  Age (mean; in years) 1988 (23–31) 27.10 27.17 26.91 26.95 0.070 Socioeconomic status  Early povertyc 1979–1984 (14–27) 0.09 0.35a 0.22b 0.11 <0.001  Early to young adult povertyc 1979–1988 (14–31) 0.07 0.30a 0.16b 0.07 <0.001  Cumulative povertyc 1979–1994 (14–37) 0.09 0.33a 0.20b 0.09 <0.001  Educational attainment (%) 1982–1994 (25)   Less than high school 9.3 19.1a 24.6b 9.8 <0.001   High school degree or GED 43.8 44.1 39.4 42.6   Some college 21.2 25.7a 24.7 23.3   College degree or higher 25.7 11.1a 11.3b 24.4 Alcohol consumption  Early onset regular drinking (%) 1983 (<15) 17.2 18.96 23.2b 20.9 0.010  Early heavy drinking (# years) 1982–1984 (18–27) 0.62 0.35a 0.53 0.58 <0.001  Later heavy drinking (# years) 1988–1989 (24–32) 0.30 0.28 0.33 0.28 0.282  Current heavy drinking (%) 1994 (30–37) 11.3 14.3a 15.5b 10.2 0.021 Marriage  Early onset of marriage (%) 1979–1994 (<20) 23.6 12.3a 30.5b 25.4 <0.001  Marital status (%) 1988 (24–31)   Never 34.7 57.8a 38.0 33.0 <0.001   Married 55.1 28.3a 48.1b 55.2   Separated/divorced/widowed 10.2 13.9a 13.8 11.8  Transition in marital status (%) 1989–1994 (25–37)   No change 66.1 71.0a 68.1 64.3 0.001   Marital formation 25.9 18.5a 23.5 25.9   Marital dissolution 8.0 10.5a 8.4 9.7 Parenthood  Early onset of parenthood (%) 1979–1994 (<23) 23.1 48.6a 39.7b 23.3 <0.001  Parent status (%) 1988 (24–31) 45.1 63.9a 60.0b 46.7 <0.001  Transition into parenthood (%) 1989–1994 (25–37) 23.0 13.4a 15.0b 22.8 <0.001 aBlack-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. bHispanic-White pairwise difference for covariates at P < 0.01 and for outcome at P < 0.05. cProportion of time spent in poverty. Socioeconomic status As expected, differential life course exposure to poverty was stark, with Blacks and Hispanics spending more time in poverty than Whites (Table 1). By age 25, Blacks and Hispanics were less likely to have a high school or college degree than Whites. Life course drinking Whites were less likely than Hispanics to begin drinking regularly before age 15. However, Whites had more years of frequent HD in their early 20s than either Blacks or Hispanics. These racial/ethnic differences in the number of years of HD disappeared in the late 20s, and by the 30s, Blacks and Hispanics were much more likely than Whites to engage in current HD. Adult roles We found racial/ethnic differences in the timing of marriage and parenthood. Marriage, whether early or during the 20s, was more common among Hispanics and Whites than Blacks. A higher proportion of Blacks experienced marital dissolution compared to Whites. As expected, early parenthood was greater among Blacks and Hispanics than Whites, a trend that continued into the mid-20s. Whites were more likely to become a parent later. These racial/ethnic trends were seen in both sexes, but compared to men, women were more likely to be a parent in the 20s and less likely to transition to parenthood at a later age. Findings from multivariate models Earlier onset of alcohol problems Table 2 presents multinomial regression coefficients showing the relative risk of earlier onset of problems in the mid-to-late 20s versus no problems. Blacks and Hispanics were not significantly different from Whites in their relative risk of earlier onset/offset. Both early and later HD were associated with increased risk for earlier onset/offset, as was having lower educational attainment. Table 2. Multinomial logistic regression model: early onset/offset of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Table 2. Multinomial logistic regression model: early onset/offset of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/ethnicity  White Ref Ref Ref Ref Ref  Black 0.82 [0.59, 1.14] 0.98 [0.69, 1.39] 0.71 [0.48, 1.04] 0.69 [0.47, 1.00] 0.69 [0.47, 1.01]  Hispanic/Latino 1.21 [0.86, 1.71] 1.27 [0.90, 1.80] 1.09 [0.76, 1.56] 1.10 [0.77, 1.58] 1.10 [0.77, 1.59]  Other 0.56 ** [0.37, 0.83] 0.56 ** [0.37, 0.84] 0.53 ** [0.35, 0.82] 0.52 ** [0.34, 0.80] 0.51 ** [0.34, 0.78] Demographics  Male (versus female) 2.04 *** [1.56, 2.67] 1.50 ** [1.13, 2.00] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40] 1.03 [0.76, 1.40]  Age 0.91 *** [0.86, 0.96] 0.90 *** [0.85, 0.95] 0.94 [0.88, 1.01] 0.95 [0.89, 1.01] 0.94 [0.88, 1.01] Poverty over the life course  Early poverty 0.96 [0.56, 1.62]  Early to young adult poverty 0.86 [0.48, 1.53]  Cumulative poverty 0.90 [0.48, 1.53] 0.85 [0.43, 1.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.07 [0.73, 1.56] 1.10 [0.75, 1.62] 1.08 [0.74, 1.60] 1.07 [0.72, 1.58]  Early heavy drinking 1.83 *** [1.62, 2.08] 1.23 ** [1.05, 1.43] 1.22 ** [1.05, 1.43] 1.23 ** [1.06, 1.44]  Early marriage 1.00 [0.69, 1.44]  Early parenting 1.37 [0.99, 1.89] Young adulthood  Education   Less than high school 2.42 ** [1.34, 4.40] 2.31 ** [1.26, 4.25] 2.34 ** [1.27, 4.31]   High school graduate 1.90 ** [1.20, 3.01] 1.87 ** [1.20, 2.98] 1.89 ** [1.19, 3.00]   Some college 1.39 [0.83, 2.35] 1.40 [0.83, 2.36] 1.41 [0.83, 2.38]   College grad or higher Ref Ref Ref  Later heavy drinking 3.47 *** [2.81, 4.30] 3.46 *** [2.80, 4.28] 3.49 *** [2.82, 4.31]  Marital status   Never married Ref   Married 0.45 *** [0.31, 0.65] 0.43 *** [0.29, 0.65] 0.44 *** [0.29, 0.65]   Sep/div/widowed 0.73 [0.45, 1.18] 0.70 [0.43, 1.14] 0.71 [0.44, 1.14]  Parenting 1.24 [0.90, 1.71] 1.21 [0.85, 1.73] 1.19 [0.83, 1.69] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.95 [0.70, 1.30] 0.93 [0.68, 1.28]   Marital dissolution 1.64 * [1.13, 2.39] 1.63 * [1.12, 2.38]  Transition into parenthood 0.97 [0.63, 1.47] 0.94 [0.62, 1.43] Current heavy drinking 0.90 [0.62, 1.32] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Later onset of alcohol problems Table 3 presents multinomial regression coefficients showing the risk of later onset of alcohol problems in the 30s compared to no problems. In Model 1, Blacks have higher relative risks than Whites for experiencing later onset compared to no problems. A statistically significant difference was not found for Hispanics versus Whites. After accounting for early poverty, HD and adult role transitions in Model 2, Blacks remained at higher risk for alcohol problems; however, this disparity disappeared when adjusting for these factors in young adulthood (mid-to-late 20s) in Model 3 and more proximal factors (during 30s) in Model 4. In the final model, cumulative poverty across all study years, lower educational attainment and current HD were significant risk factors for later onset. In addition, being married (versus single) by the mid-20s was significantly associated with 44% lower likelihood of later onset. However, none of the role transitions from the 20s to the 30s were significant. Table 3. Multinomial Logistic Regression Model: later onset in 30s of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Table 3. Multinomial Logistic Regression Model: later onset in 30s of alcohol problems (versus no problems), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.69 ** [1.25, 2.28] 1.56 * [1.07, 2.26] 1.09 [0.74, 1.61] 0.97 [0.64, 1.47] 0.94 [0.61, 1.45]  Hispanic/Latino 1.27 [0.88, 1.82] 1.11 [0.75, 1.63] 0.89 [0.60, 1.32] 0.87 [0.59, 1.30] 0.80 [0.52, 1.23]  Other 0.91 [0.59, 1.39] 0.89 [0.58, 1.35] 0.86 [0.56, 1.33] 0.87 [0.56, 1.34] 0.92 [0.59, 1.41] Demographics  Male (versus female) 2.40 *** [1.86, 3.11] 1.94 *** [1.42, 2.65] 1.60 ** [1.18, 2.18] 1.65 ** [1.20, 2.28] 1.48 * [1.06, 2.05]  Age 0.93 * [0.88, 0.99] 0.93 * [0.88, 0.99] 0.96 [0.90, 1.02] 0.96 [0.90, 1.02] 0.96 [0.90, 1.03] Poverty over the life course  Early poverty 2.68 *** [1.70, 4.21]  Early to young adult poverty 2.20 ** [1.32, 3.69]  Cumulative poverty 3.20 *** [1.71, 6.00] 2.73 ** [1.42, 5.24] Transition from adolescence to adulthood  Early onset of regular drinking 1.00 [0.71, 1.40] 1.01 [0.71, 1.43] 0.98 [0.68, 1.39] 0.98 [0.68, 1.41]  Early heavy drinking 1.87 *** [1.66, 2.11] 1.52 *** [1.33, 1.74] 1.52 *** [1.33, 1.74] 1.33 *** [1.16, 1.53]  Early marriage 1.45 [0.98, 2.15]  Early parenting 1.27 [0.91, 1.77] Young adulthood  Education   Less than high school 5.49 *** [3.13, 9.62] 4.74 *** [2.74, 8.17] 4.00 *** [2.26, 7.06]   High school graduate 2.95 *** [1.85, 4.70] 2.72 *** [1.72, 4.30] 2.29 ** [1.42, 3.69]   Some college 2.13 ** [1.27, 3.56] 2.00 ** [1.20, 3.35] 1.91 * [1.08, 3.35]   College graduate or higher Ref  Later heavy drinking 1.75 *** [1.47, 2.09] 1.75 *** [1.46, 2.09] 1.30 * [1.07, 1.58]  Marital status   Never married Ref   Married 0.56 ** [0.40, 0.78] 0.56 ** [0.38, 0.84] 0.58 ** [0.39, 0.86]   Sep/div/widowed 0.78 [0.52, 1.17] 0.79 [0.51, 1.21] 0.79 [0.50, 1.24]  Parenting 1.09 [0.81, 1.48] 0.91 [0.64, 1.30] 0.91 [0.62, 1.33] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.56 * [0.47, 0.97] 0.73 [0.51, 1.05]   Marital dissolution 0.79 [0.90, 2.15] 1.27 [0.82, 1.97]  Transition into parenthood 0.72 [0.44, 1.15] 0.78 [0.48, 1.27] Current heavy drinking 6.31 *** [4.44, 8.95] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. We tested whether the effects of our explanatory factors on later onset differed by racial/ethnic groups by examining interactions. The effects of SES and HD on later onset were similar across racial/ethnic groups (i.e. non-significant interaction terms). The only significant interaction was found for marital status in the 20s, specifically for Blacks and Hispanics who were separated/divorced/widowed (PBlack = 0.012 and PHispanic = 0.035), indicating that marital dissolution was associated with a lower risk of alcohol problems in Blacks and Hispanics compared to those who were married or single. Recurrence/persistence of alcohol problems Table 4 presents the findings for the small subset who experienced recurring/persistent problems in their 20s and 30s compared to those who transitioned out of alcohol problems in their 30s. Controlling for age and sex, Blacks and Hispanics had higher relative risks than Whites for recurrence/persistence (versus transitioning out) of alcohol problems but these relationships were not significant. Proximal factors were important predictors of recurrence/persistence (Models 4 and 5), in particular current HD and cumulative poverty. Neither marriage or parenthood, nor transitions into these roles had a significant effect on recurrence/persistence. Table 4. Multinomial Logistic Regression Model: recurrence/persistence of alcohol problems (versus early onset/offset), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. Table 4. Multinomial Logistic Regression Model: recurrence/persistence of alcohol problems (versus early onset/offset), NLSY 1979 data Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Model 1 Model 2 Model 3 Model 4 Model 5 RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI RRR P 95% CI Race/Ethnicity  White Ref Ref Ref Ref Ref  Black 1.40 [0.90, 2.16] 1.05 [0.60, 1.86] 1.15 [0.63, 2.08] 1.08 [0.59, 1.97] 1.04 [0.56, 1.94]  Hispanic/Latino 1.12 [0.71, 1.76] 0.99 [0.60, 1.63] 0.93 [0.56, 1.54] 0.89 [0.54, 1.49] 0.84 [0.48, 1.44]  Other 1.74 [0.98, 3.10] 1.70 [0.96, 3.00] 1.66 [0.91, 3.04] 1.75 [0.96, 3.20] 2.04 * [1.09, 3.82] Demographics  Male (versus female) 1.63 * [1.02, 2.62] 1.44 [0.85, 2.44] 1.43 [0.80, 2.58] 1.49 [0.83, 2.66] 1.30 [0.71, 2.39]  Age 1.02 [0.94, 1.10] 1.02 [0.94, 1.12] 1.01 [0.92, 1.11] 1.00 [0.91, 1.10] 1.02 [0.92, 1.12] Poverty over the life course  Early poverty 2.43 * [1.05, 5.65]  Early to young adult poverty 2.32 [0.88, 6.12]  Cumulative poverty 3.38 * [1.17, 9.72] 3.30 * [1.12, 9.69] Transition from adolescence to adulthood  Early onset of regular drinking 1.08 [0.62, 1.87] 1.01 [0.58, 1.77] 1.01 [0.58, 1.76] 1.09 [0.61, 1.95]  Early heavy drinking 1.21 * [1.00, 1.46] 1.16 [0.93, 1.45] 1.18 [0.94, 1.47] 0.99 [0.79, 1.24]  Early marriage 0.65 [0.33, 1.27]  Early parenting 1.28 [0.71, 2.31] Young adulthood  Education   Less than high school 2.23 [0.87, 5.73] 2.19 [0.83, 5.77] 1.71 [0.67, 4.36]   High school graduate 1.97 [0.89, 4.35] 1.96 [0.88, 4.36] 1.56 [0.71, 3.39]   Some college 1.86 [0.77, 4.49] 1.78 [0.74, 4.27] 1.61 [0.67, 3.89]   College grad or higher Ref Ref Ref  Later heavy drinking 1.21 [0.88, 1.68] 1.20 [0.871.65] 0.86 [0.62, 1.20]  Marital status   Never married Ref   Married 1.04 [0.56, 1.96] 1.21 [0.63, 2.33] 1.17 [0.61, 2.24]   Sep/div/widowed 1.33 [0.68, 2.61] 1.50 [0.75, 2.97] 1.46 [0.72, 2.96]  Parenting 0.97 [0.56, 1.69] 0.68 [0.37, 1.27] 0.72 [0.39, 1.36] Social role transitions  Transition in marriage   No change Ref Ref   Marital formation 0.72 [0.46, 1.12] 0.86 [0.54, 1.38]   Marital dissolution 0.80 [0.47, 1.38] 0.74 [0.42, 1.29]  Transition into parenthood 0.43 * [0.21, 0.85] 0.53 [0.27, 1.05] Current heavy drinking 9.42 *** [5.62, 15.79] Notes: Relative risk ratios using weighted data. ***P < 0.001; **P < 0.01; *P < 0.05. DISCUSSION Few studies have examined racial/ethnic differences in the transitions in and out of alcohol problems, and even fewer have investigated explanatory factors that might account for group differences. Using longitudinal data from a US general population sample, our study revealed a Black-White disparity in later onset of alcohol problems in the 30s, but not in young adult onset/offset nor in recurrence/persistence of alcohol problems in the 30s. Guided by a life course framework and cumulative disadvantage theory, our results further indicated that chronic poverty is an important predictor of alcohol problems above and beyond transitions into adult social roles of marriage and parenthood. We discuss these key findings and their implications for alcohol-related disparities below. A significant portion of the racial/ethnic differences was found in the exposure to factors expected to relate to the risk for alcohol problems. Blacks and Hispanics spent more time in poverty, had lower education, and had different patterns in the acquisition and timing of adult social roles than Whites. These findings are not new. Demographic trends consistently show greater disadvantages in racial/ethnic groups (more poverty among Blacks, and lower education for Blacks and Hispanics), and marital differences (lower rates among Blacks versus Whites and Hispanics) and parenting transitions (Blacks and Hispanics become parents earlier than Whites) (Aughinbaugh et al., 2013; Maralani, 2013; Kim and Raley, 2015). Our study suggests that these differences in exposures and timing of risk and protective factors appear to contribute to racial/ethnic differences in transitions into and out of alcohol problems. Much of the alcohol research overlooks how ‘aging out’ effects of marriage or parenthood, or cumulative poverty differ across racial/ethnic groups. These disadvantages and earlier timing of developmental turning points underscore the need to recognize these different lived experiences and the implications for alcohol problems among minorities. The key disparity finding of Black-White differences in later onset of alcohol problems disappeared after accounting for explanatory factors. Cumulative poverty and lower education were consistent and strong predictors of later onset. This later onset might be related to the accumulation of disadvantage that builds up across the life course. Notably, the transition from the 20s to 30s is a critical period for achieving SES, and the 30s is often a time when SES stabilizes (Furstenberg, 2008). If minorities have trouble achieving socioeconomic stability in the 30s, compounded with early disadvantages, the stresses and realities associated with this may place them at higher risk for HD and alcohol problems. Future studies should take a longitudinal perspective on changes in SES to see if this influences continued HD and later onset of alcohol problems among racial/ethnic groups. Traditional markers of adulthood, such as getting married and having children, have been shown to reduce the risk of alcohol problems (Miller-Tutzauer et al., 1991; Bachman et al., 1997a; Dawson et al., 2006). While early parenting has been associated with reduced alcohol consumption, this protective effect wanes over time (Wolfe, 2009). Moreover, taking into account Blacks’ and Hispanics’ greater cumulative disadvantage, earlier transitions to marriage and parenthood might result in greater economic stress and set the stage for more HD and later alcohol problems. We found that Blacks and Hispanics have less HD in the 20s compared to Whites, but significantly greater HD in their 30s. This finding is similar to other studies showing a crossover in risk, whereby racial/ethnic minorities are at lower risk for adverse health behaviors in adolescence but higher risk in early adulthood compared to Whites (Brown et al., 2007; Lawrence et al., 2014). Differential exposures also operated in unexpected ways to explain the later onset of alcohol problems for racial/ethnic minorities. In our multivariate models, being married in the mid-20s, but not marital transitions in the 30s, was a consistent and significant protective factor for later onset across all groups. However, the significant interaction found between race and marital dissolution indicates that marital dissolution may have less of an adverse effect for minorities. While Blacks and Hispanics were more likely to be separated/divorced/widowed in the 20s than Whites, this status afforded some protection against later onset of alcohol problems compared to being never married or even married. Marital dissolution has been associated with harmful effects (Kretsch and Harden, 2014), which was consistent for Whites in this study, but could getting out of a marriage may be less harmful for racial/ethnic minorities? This finding warrants further research to examine the meaning of marriage and marital dissolution in relation to alcohol problems in Blacks, Hispanics, and Whites. We did not find racial/ethnic disparities in earlier onset nor recurrence/persistence of alcohol problems. These findings conflict with NESARC data showing Whites are more likely to experience past-year alcohol dependence in the 20s (Grant et al., 2004) and minorities are more likely to experience recurrence (Grant et al., 2012). The discrepant finding may be related to measurement (capturing DSM-IV alcohol dependence versus alcohol problems) or cohort effects, given that the young adults who were part of the larger NESARC sample of US adults 18+ were born in the late 1970–1980s while this study sampled young adults born 1957–1964. Future research is needed to examine racial/ethnic disparities in alcohol dependence, based on DSM-criteria, longitudinally over time. These study findings should be considered in light of the following limitations. First, findings are only generalizable to the U.S. population born between 1957 and 1964, reflective of the post-baby boomer cohort. Second, the outcome measure was derived from data that are more than 20 years old. However, national longitudinal data permitting a life course perspective and analysis of alcohol problems are rare; most US national data are cross-sectional requiring retrospective reports or if prospective, are limited to several years of data (versus NLSY’s 16 years of data up to 1994). Third, the measures for alcohol problems, two symptoms experienced in the same 12-month period consistent with DSM-IV criteria, are likely to capture mild dependence. Furthermore, additional time points to capture recurrence (repeated problems with remission) versus persistence (continued symptoms without remission) of problems are needed and would offer insight into clinical and public health interventions. Fourth, the NLSY79 surveys assessed past month’s frequency of 6+ drinks on occasion, which significantly exceeds that of current NIAAA HD guidelines using sex-specific 5+/4+ drinks on occasion (National Institute on Alcohol Abuse and Alcoholism, 2005); and thus risks under-identifying HD in the sample. Also, while we found that HD was a significant predictor of alcohol problems, the observed relationship may be stronger than had we been able to use the 5+/4+ HD criteria. Fifth, due to the smaller sample sizes of Blacks and Hispanics compared to Whites, the model estimates of alcohol problems that were more rare and, in particular, interactions of race by explanatory factors may be less reliable for Blacks and Hispanics and should be interpreted with caution. Also, while we include other racial/ethnic groups in our analyses, we do not interpret the findings given the heterogeneity of this group. Finally, we did not have sufficient power to conduct sex-stratified analyses. While we controlled for sex, past studies have shown significant sex differences in alcohol problems, with men at higher risk (Grant et al., 2012). Poverty and adult social roles can have differential effects on alcohol problems across sex. Women have more risks associated with poverty experiences, and men have greater benefits from marital formation than women while women have less risk from marital dissolution (Zick and Smith, 1991; Raley and Sweeney, 2009). Future studies should consider sex in racial/ethnic differences of alcohol problems. Overall, alcohol problems are more common in young adulthood and decline over time. However, our study shows that Blacks are more likely than Whites to experience a later onset of alcohol problems in the 30s, due in part to differences in the accumulation of disadvantage and HD patterns, both of which have a stronger effect on risk for later onset and recurrence/persistence of alcohol problems than do social role transitions. These study findings suggest the benefits of interventions to alleviate social disadvantages across the life course (e.g. policy efforts to reduce poverty and increase educational opportunities) and of interventions targeting racial/ethnic minorities during their late 20s and 30s to prevent HD and alcohol problems. Universal prevention strategies to reduce HD and alcohol-related problems are abundant for young adults. However, if Blacks are transitioning into alcohol problems while Whites are transitioning out of alcohol problems, this points to a clear need for targeted prevention strategies beyond young adulthood to address alcohol-related disparities. ACKNOWLEDGEMENTS The authors would like to thank the study team, Ms Edwina Williams, Drs Katherine Karriker-Jaffe, Jane Witbrodt and Sarah Zemore at the Alcohol Research Group, and Dr Margaret Ensminger at the Johns Hopkins Bloomberg School of Public Health for their feedback on study conceptualization and results. FUNDING This work was supported by the National Institute on Alcohol Abuse and Alcoholism at the National Institutes of Health [Grant number R01AA022668 to N.M.]. CONFLICT OF INTEREST The authors declare that they have no conflict of interest. REFERENCES Alvanzo AA , Storr CL , La Flair L , et al. . ( 2011 ) Race/ethnicity and sex differences in progression from drinking initiation to the development of alcohol dependence . Drug Alcohol Depend 118 : 375 – 82 . Google Scholar CrossRef Search ADS PubMed American Psychiatric Association . ( 1994 ) DSM-IV: Diagnostic & Statistical Manual of Mental Disorders , 4th edn . Washington, DC : American Psychiatric Association . American Psychiatric Association . ( 2013 ) Diagnostic and Statistical Manual of Mental Disorders , 5th edn . Arlington, VA : DSM-5 . Aughinbaugh A , Robles O , Sun H . ( 2013 ) Marriage and Divorce: Patterns by Gender, Race, and Educational Attainment [Online]. http://www.bls.gov/opub/mlr/2013/article/pdf/marriage-and-divorce-patterns-by-gender-race-and-educational-attainment.pdf. Bachman JG , Wadsworth KN , O’Malley PM , et al. . ( 1997 a) Smoking, Drinking, and Drug Use in Young Adulthood: The Impacts of New Freedoms and New Responsibilities . Mahwah, NJ : Erlbaum . Bachman JG , Wadsworth KN , O’Malley PM , et al. . ( 1997 b) Marriage, divorce, and parenthood during the transition to young adulthood: impacts on drug use and abuse. In Schulenberg J , Maggs JL , Hurrelmann K (eds) . Health Risks and Developmental Transitions during Adolescence . New York : Cambridge University Press , 246 – 79 . Brown TH . ( 2016 ) Diverging fortunes: racial/ethnic inequality in wealth trajectories in middle and late life . Race Soc Prob 8 : 29 – 41 . Google Scholar CrossRef Search ADS Brown JS , Meadows SO , Elder GH . ( 2007 ) Race-ethnic inequality and psychological distress: depressive symptoms from adolescence to young adulthood . Dev Psychol 43 : 1295 – 311 . Google Scholar CrossRef Search ADS PubMed Bureau of Labor Statistics and U.S. Department of Labor. ( 2012 ) National Longitudinal Survey of Youth 1979 Cohort, 1979–2010 (Rounds 1–24). Archived by WebCite® at http://www.webcitation.org/6egSq0L6G] (20 January 2016, date last accessed). Columbus, OH. [Online]. http://www.bls.gov/nls/nlsy79.htm. Caetano R , Kaskutas LA . ( 1995 ) Changes in drinking patterns among Whites, Blacks and Hispanics: 1984–1992 . J Stud Alcohol 56 : 558 – 65 . Google Scholar CrossRef Search ADS PubMed Caetano R , Kaskutas LA . ( 1996 ) Changes in drinking problems among Whites, Blacks and Hispanics: 1984–1992 . Subst Use Misuse 31 : 1547 – 71 . Google Scholar CrossRef Search ADS PubMed Centers for Disease Control and Prevention (CDC). ( 2011 ) CDC Reports Excessive Alcohol Consumption Costs the U.S. $224 Billion in 2006. Archived by WebCite® at http://www.webcitation.org/6ED3xmU2H] (5 February 2013, date last accessed). Atlanta, GA. [Online]. http://www.cdc.gov/media/releases/2011/p1017_alcohol_consumption.html. Cerdá M , Johnson-Lawrence VD , Galea S . ( 2011 ) Lifetime income patterns and alcohol consumption: investigating the association between long- and short-term income trajectories and drinking . Soc Sci Med 73 : 1178 – 85 . Google Scholar CrossRef Search ADS PubMed Dannefer D . ( 2003 ) Cumulative advantage/disadvantage and the life course: cross-fertilizing age and social science theory . J Gerontol B Psychol Sci Soc Sci 58B : S327 – 37 . Google Scholar CrossRef Search ADS Darity W Jr , Nicholson MJ . ( 2005 ) Racial wealth inequality and the Black family . Afr Am Fam Life 78 : 85 . Dawson DA , Grant BF , Stinson FS , et al. . ( 2005 ) Recovery from DSM-IV alcohol dependence: United States, 2001–2002 . Addiction 100 : 281 – 92 . Google Scholar CrossRef Search ADS PubMed Dawson DA , Grant BF , Stinson FS , et al. . ( 2006 ) Maturing out of alcohol dependence: the impact of transitional life events . J Stud Alcohol 67 : 195 – 203 . Google Scholar CrossRef Search ADS PubMed Elder GH Jr , Johnson MK , Crosnoe R . ( 2003 ) The emergence and development of life course theory. In Mortimer JT , Shanahan MJ (eds) . Handbook of the Life Course . New York : Springer , 3 – 19 . Google Scholar CrossRef Search ADS Evans-Polce RJ , Vasilenko SA , Lanza ST . ( 2015 ) Changes in gender and racial/ethnic disparities in rates of cigarette use, regular heavy episodic drinking, and marijuana use: ages 14 to 32 . Addict Behav 41 : 218 – 22 . Google Scholar CrossRef Search ADS PubMed Furstenberg FF . ( 2008 ) The intersections of social class and the transition to adulthood. In Mortimer JT (ed) . Social Class and Transitions to Adulthood. New Directions for Child and Adolescent Development, 119 . San Francisco : Jossey-Bass , 1 – 10 . Grant BF , Dawson DA , Stinson FS , et al. . ( 2004 ) The 12-month prevalence and trends in DSM-IV alcohol abuse and dependence: United States, 1991–1992 and 2001–2002 . Drug Alcohol Depend 74 : 223 – 34 . Google Scholar CrossRef Search ADS PubMed Grant JD , Vergés A , Jackson KM , et al. . ( 2012 ) Age and ethnic differences in the onset, persistence and recurrence of alcohol use disorder . Addiction 107 : 756 – 65 . Google Scholar CrossRef Search ADS PubMed Hasin DS , Stinson FS , Ogburn E , et al. . ( 2007 ) Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions . Arch Gen Psychiatry 64 : 830 – 42 . Google Scholar CrossRef Search ADS PubMed Hingson RW , Heeren T , Winter MR . ( 2006 ) Age at drinking onset and alcohol dependence: age at onset, duration, and severity . Arch Pediatr Adolesc Med 160 : 739 – 46 . Google Scholar CrossRef Search ADS PubMed Kim Y , Raley RK . ( 2015 ) Race-ethnic differences in the non-marital fertility rates in 2006–2010 . Popul Res Policy Rev 34 : 141 – 59 . Google Scholar CrossRef Search ADS PubMed Kochhar R , Fry R , Taylor P. ( 2011 ) Wealth Gaps Rise to Record Highs Between Whites, Blacks and Hispanics. Archived by WebCite® at http://www.webcitation.org/6WDdIhIwD] (9 February 2015, date last accessed). Washington, DC. [Online]. http://www.pewsocialtrends.org/files/2011/07/SDT-Wealth-Report_7-26-11_FINAL.pdf. Kretsch N , Harden KP . ( 2014 ) Marriage, divorce, and alcohol use in young adulthood: a longitudinal sibling-comparison study . Emerg Adulthood 2 : 138 – 49 . Google Scholar CrossRef Search ADS Lawrence EM , Pampel FC , Mollborn S . ( 2014 ) Life course transitions and racial and ethnic differences in smoking prevalence . Adv Life Course Res 22 : 27 – 40 . Google Scholar CrossRef Search ADS PubMed Lopez-Quintero C , Pérez de los Cabos J , Hasin DS , et al. . ( 2011 ) Probability and predictors of transition from first use to dependence on nicotine, alcohol, cannabis, and cocaine: results of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) . Drug Alcohol Depend 115 : 120 – 30 . Google Scholar CrossRef Search ADS PubMed Lui CK , Chung PJ , Wallace SP , et al. . ( 2014 ) Social status attainment during the transition to adulthood . J Youth Adolesc 43 : 1134 – 50 . Google Scholar CrossRef Search ADS PubMed Maralani V . ( 2013 ) The demography of social mobility: Black-White differences in the process of educational reproduction . Am J Sociol 118 : 1509 – 58 . Google Scholar CrossRef Search ADS Miller-Tutzauer C , Leonard KE , Windle M . ( 1991 ) Marriage and alcohol use: a longitudinal study of maturing out . J Stud Alcohol 52 : 434 – 40 . Google Scholar CrossRef Search ADS PubMed Mulia N , Karriker-Jaffe KJ , Witbrodt J , et al. . ( 2017 a) Racial/ethnic differences in 30-year trajectories of heavy drinking in a nationally representative U.S. sample . Drug Alcohol Depend 170 : 133 – 41 . Google Scholar CrossRef Search ADS PubMed Mulia N , Tam T , Bond J , et al. . ( 2017 b) Racial/ethnic differences in life-course heavy drinking from adolescence to midlife . J Ethn Subst Abuse 17 : 167 – 86 . Google Scholar CrossRef Search ADS PubMed Muthén BO , Muthén LK . ( 2000 ) The development of heavy drinking and alcohol-related problems from ages 18 to 37 in a U.S. national sample . J Stud Alcohol 61 : 290 – 300 . Google Scholar CrossRef Search ADS PubMed National Institute on Alcohol Abuse and Alcoholism ( 2005 ) Helping patients who drink too much: a clinician’s guide. Updated 2005 Edition, revised October 2015 (NIH Publication No. 07–3769). Archived by WebCite® at http://www.webcitation.org/6jiIsTNiY] (12 August 2016, date last accessed). Rockville, MD. [Online]. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf. Oropesa RS , Landale NS . ( 2004 ) The future of marriage and hispanics . J Marriage Fam 66 : 901 – 20 . Google Scholar CrossRef Search ADS Pearlin LI , Schieman S , Fazio EM , et al. . ( 2005 ) Stress, health, and the life course: some conceptual perspectives . J Health Soc Behav 46 : 205 – 19 . Google Scholar CrossRef Search ADS PubMed Raley RK , Sweeney MM . ( 2009 ) Explaining race and ethnic variation in marriage: directions for future research . Race Soc Problems 1 : 132 – 42 . Google Scholar CrossRef Search ADS Sacks JJ , Gonzales KR , Bouchery EE , et al. . ( 2015 ) 2010 National and state costs of excessive alcohol consumption . Am J Prev Med 49 : e73 – 9 . Google Scholar CrossRef Search ADS PubMed Shuey KM , Willson AE . ( 2008 ) Cumulative disadvantage and Black-White disparities in life-course health trajectories . Res Aging 30 : 200 – 25 . Google Scholar CrossRef Search ADS StataCorp . ( 2013 ) Stata Statistical Software: Release 13.0 [Computer program]. Stata Corporation. (Accessed). Vega WA . ( 1990 ) Hispanic families in the 1980s: a decade of research . J Marriage Fam 52 : 1015 – 24 . Google Scholar CrossRef Search ADS Williams E , Mulia N , Karriker-Jaffe KJ , et al. . ( 2018 ) Changing racial/ethnic disparities in heavy drinking trajectories through young adulthood: a comparative cohort study . Alcoholism 42 : 135 – 43 . Google Scholar CrossRef Search ADS PubMed Wolfe JD . ( 2009 ) Age at first birth and alcohol use . J Health Soc Behav 50 : 395 – 409 . Google Scholar CrossRef Search ADS PubMed Yang S , Lynch JW , Raghunathan TE , et al. . ( 2007 ) Socioeconomic and psychosocial exposures across the life course and binge drinking in adulthood: population-based study . Am J Epidemiol 165 : 184 – 93 . Google Scholar CrossRef Search ADS PubMed Zick CD , Smith KR . ( 1991 ) Marital transitions, poverty, and gender differences in mortality . J Marriage Fam 53 : 327 – 36 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Alcohol and AlcoholismOxford University Press

Published: Mar 13, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off