Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 15.
Published in final edited form as: Subst Use Misuse. 2016 Jan 15;51(1):1–14. doi: 10.3109/10826084.2015.1073321

The Effect of Childhood Supervisory Neglect on Emerging Adults’ Drinking

Susan M Snyder a, Darcey H Merritt b
PMCID: PMC4832597  NIHMSID: NIHMS774795  PMID: 26771736

Abstract

This study investigated the effect of childhood supervisory neglect on emerging adults’ drinking. Child supervisory neglect is the most common form of child maltreatment in the United States, but few studies explore supervisory neglect separate from other forms of maltreatment among emerging adults, 18–25 years old. The study sample included (n = 11,117) emerging adults, 18–25 years old who participated in Waves I and III of the National Longitudinal Study of Adolescent Health (Add Health). We conducted separate analyses for male and female emerging adults, because they have different rates of alcohol consumption and alcohol risk behaviors. Our study used latent class analysis to understand how patterns of alcohol risk behaviors clustered together. For males, we found the following four classes: (1) multiple-risk drinkers, (2) moderate-risk drinkers, (3) binge-drinkers, and (4) low-risk drinkers or abstainers. For females, we found the following three classes: (1) multiple-risk drinkers, (2) moderate-risk drinkers, and (3) low-risk drinkers or abstainers. For both males and females, supervisory neglect increased the odds of membership in the multiple-risk drinkers’ class compared to the low-risk drinkers or abstainers’ class. Single males who did not live with their parents, and who were white had increased odds of being in the multiple-risk drinkers. For females, being more educated, or in a serious romantic relationship increased the odds of membership in the multiple-risk drinkers’ class. Practitioners should ask about histories of supervisory neglect among emerging adults who engage in alcohol risk behaviors.

Keywords: Child supervisory neglect, alcohol, latent class analysis, emerging adults, National Longitudinal Study of Adolescent Health, gender differences, depression, parental drinking, Add Health

Introduction

Child supervisory neglect is a significant and under-studied problem that may contribute to distinct patterns of alcohol risk behaviors among emerging adults (18–25 years old). The absence of literature focusing on the long-term effects of supervisory neglect is surprising because supervisory neglect occurs more often than any other type of child maltreatment (Coohey, 2003; Mennen, Kim, Sang, & Trickett, 2010). The American Professional Society on the Abuse of Children (APSAC, 2008) defines supervisory neglect as adult supervision of a child that is inadequate to meet the child’s needs. Supervisory neglect consists of discrete events that can have immediate and sometimes catastrophic consequences. Although APSAC (2008) recommends classifying supervisory neglect as distinct from other forms of neglect, few studies have done so. This gap in the literature is significant because failing to examine supervisory neglect separately obscures its specific effects, making it difficult to effectively target interventions (Merritt & Klein, 2015). While research has linked child neglect generally with later heavy drinking (Shin, Edwards, & Heeren, 2009) and alcohol dependence (Mullings, Hartley, & Marquart, 2004), it is unclear how much of this effect can be attributed to supervisory neglect, and whether emerging adults who are most likely to engage in alcohol risk behaviors are more likely to report having experienced child supervisory neglect. This paper will begin to fill this gap in the literature.

A significant portion of alcohol-related public health costs are attributable to alcohol risk behaviors among emerging adults. Alcohol risk behaviors include binge drinking, experiencing hang-overs, throwing up after drinking, driving drunk, relationship problems with friends and dating, physical fights, problems at work or school, and sexual situations that an individual regrets (Khan, Cleland, Scheidell, & Berger, 2014; Kopak, Chia-Chen Chen, Haas, & Gillmore, 2012). Binge drinking has been linked to the following adverse health and social outcomes: violence, suicide, sexually transmitted infections, unintended pregnancies, hypertension, and unintentional injuries, such as motor vehicle crashes (Kanny, Liu, Brewer, & Lu, 2013). Bouchery, Harwood, Sacks, Simon, and Brewer (2011) estimate that in 2006 binge drinking yielded an economic cost of $223.5 billion in the United States. Compared to all other age groups, emerging adults have the highest rates of alcohol use (59.6%) and binge drinking (i.e., drinking five or more drinks on a single occasion; 37.9%). Male emerging adults have higher rates of alcohol use (62.3% vs. 56.9%) and binge drinking (44.4% vs. 31.4%) than female emerging adults (SAMHSA, 2014). In light of these notable differences, this paper examines the effect of supervisory neglect on alcohol risk behaviors separately for males and females. We will first present our theoretical framework. Then, we will provide justification for the analytic approach our study uses. This section will conclude by detailing the current study.

Social Development Model

This study applies the Social Development Model, which blends key elements of social learning, social control, and differential association theories to explain how risk and protective factors impact alcohol risk behaviors. Within the Social Development Model, the relationships surrounding an individual, such as family members, or romantic partners, may affect alcohol use consumption and alcohol related risk behaviors (Catalano & Hawkins, 1996; Catalano, Kosterman, Hawkins, Newcomb, & Abbott, 1996). On the one hand, individuals who are bonded to prosocial influences (e.g., parents or partners who do not drink heavily or engage in alcohol related risk behaviors) also tend to consume less alcohol and are less likely to engage in alcohol related risk behaviors (Catalano et al., 1996; Fleming et al., 2008; Snyder, Gwaltney, & Landeck, 2015). Conversely, emerging adults with parents or partners who have histories of heavy alcohol consumption are at greater risk of heavy alcohol consumption themselves (White & Jackson, 2004).

Supervisory neglect

According to the Social Development Model, children’s early life experiences of parental family management can reinforce later behavioral decisions. In particular, when parents closely supervise their children, children are more likely to perceive reinforcement for prosocial behaviors, including lower alcohol consumption and fewer alcohol risk behaviors (Catalano & Hawkins, 1996; Williams et al., 2007). Thus, the inverse would also be true; children who have experienced supervisory neglect may be at greater risk of higher alcohol consumption and more alcohol risk behaviors. Widom and colleagues have written extensively on the relationship between experiencing either child abuse or neglect and later alcohol use (cf. Horwitz, Widom, McLaughlin, & White, 2001; Widom & Hiller-Sturmhofel, 2001; Widom, Ireland, & Glynn, 1995; Widom, White, Czaja, & Marmorstein, 2007), but their work does not elucidate the specific effects of neglect, and it lacks a discussion of supervisory neglect. While some other studies have explored the relationship between neglect and alcohol risk behaviors, researchers have yet to focus on child supervisory neglect.

We found two studies of adults that investigated the relationship between alcohol use and earlier experiences of general neglect. In the first, Patock-Peckham and Morgan-Lopez (2010) conducted a two-group SEM path model with data from 404 university students. This study found that parent neglectfulness among the parent of the same-gender was associated with alcohol-related problems (e.g., used alcohol “to lose social and emotional inhibitions”; males 0.209, p < 0.001; females 0.152, p < 0.05). In the second, Mullings et al. (2004) interviewed female prisoners in Texas to explore the relationship between child maltreatment and alcohol dependency. They found that the women who were alcohol dependent were more likely to have retrospectively reported neglect compared to women who were not alcohol dependent. While each study has enhanced our understanding of the relationship between general child neglect and alcohol use, each also has some significant limitations. Neither study included questions about the frequency of neglect experiences, nor did either specify an age by which the experiences took place. Furthermore, these studies do not contribute to our understanding of supervisory neglect’s unique contribution to alcohol risk behaviors.

Protective relationships

Relationships can function not only as risk factors, but can also protect against risky behaviors. In particular, studies have found that emerging adults who do not live with their parents are at greater risk of heavy episodic drinking than emerging adults who live with their parents (Gfroerer, Greenblatt, & Wright, 1997; White et al., 2006). Remaining at home may signify a close emotional bond between parent and child, and it may provide parents opportunities to monitor the behaviors of their emerging adult children. Relationships with romantic partners may affect alcohol use in a similar way. When emerging adults are involved in serious romantic relationships (i.e., cohabitation or marriage) their risk of heavy alcohol consumption and/or engaging in alcohol-related risk behaviors is lower than their single counterparts (Fleming, White, & Catalano, 2010; Snyder & Merritt, 2015; Snyder & Rubenstein,2014).

Individual characteristics

In addition to accounting for the relationships surrounding an individual, the Social Development Model also considers the role individual characteristics play in either promoting or deterring heavy alcohol consumption or alcohol risk behaviors (Catalano & Hawkins, 1996). Each relevant individual characteristic is discussed below.

Gender

Gender constitutes one of the most robust correlates of alcohol consumption and alcohol risk behaviors. Specifically, emerging adult males tend to consume more alcohol than females (Chartier, Hesselbrock, & Hesselbrock, 2011; Delucchi, Matzger, & Weisner, 2008; SAMHSA, 2013). Gender differences are important to examine because males and females experience different psychological, social, and physical or physiological effects of alcohol consumption (Wilsnack & Wilsnack, 2013). One explanation of physiological differences comes from the National Institute on Alcohol Abuse and Alcoholism (NIAAA,2013), which explains that women typically weigh less and have less water in their bodies compared to men, so women tend to have higher blood alcohol concentrations after drinking the same amount of alcohol as men. For this reason NIAAA has defined binge drinking for women as four drinks and binge drinking for men as five drinks (NIAAA, n.d.). Regarding risk behaviors, research has found that males are more likely to engage in risk behaviors, such as binge drinking and drunk driving (Wilsnack & Wilsnack, 2013).

Age, ethnicity or race, and education

Age, ethnicity or race, and education are also important correlates to take into account (Merritt, 2009). The Social Development Model takes into account the role development plays in contributing to alcohol risk behaviors. Alcohol use and related risk behaviors are more prevalent during emerging adulthood (SAMHSA, 2014), but then decreases dramatically (Arnett, 2005). An additional factor is race. Emerging adults who are White or Hispanic are at greater risk of alcohol risk behaviors (Chen, Yi, Williams, & Faden, 2009). Regarding education, emerging adults in college engage in more binge drinking and drunk driving than emerging adults who are not attending college (Hingson & White, 2010).

Emotional and behavioral factors

The Social Development Model also accounts for emotional and behavioral factors that have been associated with alcohol consumption. Several studies have linked depression (cf. Dixit & Crum, 2000; Lee, Kosterman, McCarty, Hill, & Hawkins, 2012; Weitzman, 2004) and adolescent delinquency (cf. Guo, Hawkins, Hill, & Abbott, 2001; Hunter, Miles, Pedersen, Ewing, & D’Amico, 2014; Williams et al., 2007) to alcohol use during emerging adulthood. The effects of delinquency and depression may differ by gender because males are more likely to have engaged in adolescent delinquency (Connell, Cook, Aklin, Vanderploeg, & Brex, 2011; Farrington et al., 2010; Schwartz et al., 2010) and females experience higher rates of depression (Blanco et al., 2010; Kessler et al., 2003; Lee et al., 2012). Lee et al. (2012) used data from the Seattle Social Development Project (SSDP) to investigate gender differences in how major depressive disorder prevalence fluctuates by patterns of alcohol use disorder symptoms over time. Lee et al. found that the females with the highest depression symptoms were in the group with the most severe alcohol risk behaviors, which they termed the “chronic alcohol disorder symptom subgroup.” However, males had fairly consistent rates of depression between the more moderate group, which they termed the “decreaser alcohol disorder symptom subgroup” and the “chronic alcohol disorder symptom subgroup.”

Latent class analysis studies

A significant body of literature has indicated that a single measure of alcohol use is inadequate to capture the heterogeneity of alcohol-related behaviors among emerging adults. While some individuals consume little to no alcohol, others binge drink and engage in a range of alcohol risk behaviors. To address these differences in alcohol use studies have applied latent class analysis (LCA) (cf. Beseler, Taylor, Kraemer, & Leeman, 2012; Cleveland, Mallett, White, Turrisi, & Favero, 2013). Much of the LCA alcohol literature regarding emerging adults has focused on the frequency and quantity of use, with fewer studies exploring alcohol risk behaviors. Below, we discuss two studies that focus exclusively on emerging adult drinking and risk behaviors. Cleveland et al. (2013) used data from 264 emerging adults who were not attending college to explore patterns of drinking behaviors. This study found the following four classes: (1) daily drinkers (5%), (2) weekend risky drinkers (23%), (3) weekend light drinkers (38%), and (4) current nondrinkers (34%). Using these classes Cleveland et al. then examined the probability of risk behaviors such as having hangovers, being sick, getting into fights, regretting sex, missing work, and driving drunk. Daily drinkers were most likely to get into fights, drive drunk and miss work, while weekend risky drinkers were most likely to get sick and regret sex. Beseler et al. (2012) examined DSM-IV alcohol use criteria and binge drinking among undergraduate students. This study found that a 3-class solution had the best fit. (1) Class 1 members (60.11%, n = 217) mainly endorsed tolerance (18.4%); none were alcohol dependent. (2) Class 2 members (31.58%, n = 114) mainly endorsed tolerance (81.6%) and drinking more than intended (74.6%); 34.2% met criteria for dependence. (3) Class 3 members (8.31%, n = 30) endorsed all dependence criteria (30%–100%); all met criteria for dependence. We were not able to find studies that have formed classes of alcohol risk behaviors among emerging adults.

The current study

This study employs the following research questions: Do male and female emerging adults have distinct patterns of alcohol use? Do child supervisory neglect, depression, and parental drinking correlate with different latent classes of alcohol use during emerging adulthood? Are being in a serious committed relationship and living with parents protective factors against problematic alcohol use? By focusing on gender differences among emerging adults, using a national dataset, and studying covariates that have not previously been studied together our study will begin to fill gaps in the literature. Our study includes the following three hypotheses:

  • Hypothesis 1. Both male and female emerging adults will form at least three distinct classes of alcohol-use, including a group with low alcohol use or abstinence and at least one group of alcohol users and who engage in several risk behaviors associated with alcohol use.

  • Hypothesis 2. The following risk factors will increase the likelihood of alcohol risk behaviors among both male and female emerging adults: retrospectively reported child supervisory neglect (Wave III), depression (Wave III), adolescent delinquency (Wave 1), parent-reported parental drinking (Wave I), and being single (Wave III).

  • Hypothesis 3. Living with a parent or parents and being Black are protective factors that will reduce the likelihood of heavy alcohol consumption and related problems.

Method

Study design

We used data from Waves I and III of the Add Health dataset. The Add Health website provides a more comprehensive discussion of the Add Health study design (Harris et al., 2009). Add Health data collection began in 1994 when 90,118 youth completed in-school surveys. After stratifying the school sample by gender and grade, an in-home sample was drawn from the school sample. Wave I in-home interviews were conducted between 1994 and 1995, with 20,745 youth in grades 7 through 12 (11–21 years). Wave II interviews were completed between April and August 1996, with 14,738 youth (12–21 years old) from Wave I completed in-home interviews (Wave II). Between August 2001 and April 2002, 15,197 young adults 18–28 years old completed in-home interviews for Wave III. The institutional review board at the University of North Carolina at Chapel Hill approved all of the original Add Health study protocols and consent was obtained from all study participants.

Weighted design

All of our analyses included Wave III sampling weights and survey analysis techniques to adjust for the unequal probability of selection, clustered sampling design, and participant attrition from prior waves. The selection probability was unequal because of nonresponse, poor frame coverage, and randomization in the sample selection. Additional information regarding the sampling design Add Health is available from Chantala and Tabor (2010) or http://www.cpc.unc.edu.libproxy.lib.unc.edu/projects/addhealth/.

Study sample

Of the 15,197 cases available at Wave III, 991 cases were lost because our analyses required the inclusion of cases with weights, stratification and cluster variables (Chantala, 2006). We also omitted the 125 cases that were not 18–25 years old. In addition, we left out 2,957 cases due to missing covariate information and seven cases due to nonresponse on questions regarding alcohol consumption or related risk behaviors. Mplus employs listwise deletion for missing covariates. As a result our analytic sample included 11,117 emerging adult participants from Wave III of Add Health. Only participants with complete data were used in the present study.

Measures

Alcohol use

During Wave III, eight dichotomous (Yes = 1) items were used to assess emerging adults’ alcohol use and risk behaviors. Unless otherwise specified the items refer to experiences over the past 12 months. To account for the physiological differences in how males and females physically process alcohol, male emerging adults were asked whether they had drank 5 or more drinks during the past 2 weeks, while female emerging adults were whether they had drank 4 or more drinks during the past 2 weeks. To assess for other alcohol risk behaviors which occurred over the past 12 months we used five items from Add Health’s “Alcohol-related problems scale” (Harris et al.,2009): (1) “You had problems at school or with school work because you had been drinking,” (2) “Did you get into a sexual situation that you later regretted because you had been drinking,” (3) “You got into trouble with your parents because you had been drinking,” (4) “You were hung over,” and (5) “Did you get into a physical fight because you had been drinking.” These items serve as a proxy for alcohol abuse as defined in the DSM-IV (American Psychiatric Association, 1994). We did not include the following two items from the original scale because they were highly correlated with other items: “You had problems with your friends because you had been drinking,” and “You were sick to your stomach or threw up after drinking.” The last two items we included asked if they had (1) been drunk at school or work; or (2) “driven drunk during the past 7 years.”

Demographics

The following demographic characteristics were captured during Wave 1: gender, race, Hispanic origin, and date of birth. For our analyses we coded male gender as 1. In the Add Health dataset Hispanic is considered an ethnicity and is handled separately from race. A dichotomous question asked, “Are you of Hispanic or Latino origin?” which we coded “yes” as 1 with the referent (0) being those who reported not being Hispanic. We used the following categories for race: Black, White, Asian/Pacific Islander, and other. Because we expected that alcohol consumption would be highest among whites we coded white as 1. Thus, the three other categories for race (i.e., blacks, Asian/Pacific Islander, and other) served as the referent (were equal to zero). We calculated respondents’ ages at Wave III using the dates of birth provided by respondents at Wave I. Lastly, during Wave III respondents provided their highest level of education, which ranged from not completing high school to beyond college. For our analyses we standardized education and ages (M = 0; SD = 1).

Risk and protective factors

Child supervisory neglect

During Wave III respondents were asked to retrospectively answer a question regarding experiences prior to the 6th grade: “How often had your parents or other adult care-givers left you home alone when an adult should have been with you?” The items were originally scaled so that 1 = 1 time, 2 = 2 times, 3 = 3–5 times, 4 = 6–10 times, and 5 = more than 10 times. Following the approach used by Currie and Tekin (2006), we dichotomized supervisory neglect to include instances when individuals experienced at least 10 experiences of supervisory neglect.

Depression

The 10-item modified version of the Center for Epidemiologic Studies–Depression (CES-D) scale was used to determine whether participants were depressed. The internal consistency of the scale was 0.80. For our analyses, we standardized the summed scores for this measure (M = 0; SD = 1).

Parental drinking

During Wave I, the parent survey asked parents whether they had consumed 5 or more alcoholic drinks on one occasion in the past month (1 = yes). The inclusion of heavy parental drinking as a covariate controls for both genetic and social influences of a parent who drinks heavily (Wilson & Widom, 2010). The vast majority of parents surveyed were biological mothers (86.6%). The next largest groups were biological fathers (4.1%), adoptive mothers (2.8%), and grandmothers (1.8%). In addition to these relationships, other “parents” who completed the survey (4.7%) included adoptive fathers, step-mothers, step-fathers, other relatives, and foster care providers.

Live with parents

When respondents specified that they resided with their parents during Wave III we coded the response (yes = 1).

Relationship status

At Wave III, respondents who answered that they had been married at least one time and answered yes to the question, “Are you still married?” were considered to be married. Respondents were cohabitating if they answered that they had ever lived in a “marriage-like relationshipfor at least one month and if they were “still living together.” Respondents who were not married or cohabitating were designated as single. Because prior studies have found that being in a serious romantic relationship reduces substance use, single functioned as the referent (Fleming et al., 2010; Snyder & Rubenstein, 2014).

Statistical analysis

Descriptive statistics were calculated using Stata 13.1 and LCA were conducted in Mplus 7.11. We took into account data stratification, clustering, and sampling weights for all of these analyses. To compute the F-statistic, degrees of freedom and p value for the comparisons across gender for categorical variables, we used the svy: tab command; and for continuous variables, we used the svy: mean command followed by the test command.

Scholars use LCA to investigate patterns of alcohol use during one point in time. By carefully combing through a dataset, LCA locates and groups together individuals with similar patterns of survey question responses. The resulting groups that are found are referred to as classes. Because individual persons’ similarities are used to form the classes, LCA is referred to as person-centered (Barnes, Boutwell, Morris, & Armstrong, 2012; Snyder & Merritt,2015). We fit the eight alcohol use items with a one-class model, and increased classes until we had evaluated seven models. The model fit statistics used to assess the models were log likelihood, Bayesian Information Criteria (BIC),and Aikaike Information Criteria (AIC). The lowest possible values are preferred for the log likelihood, BIC, and AIC statistics. McCutcheon (2002) explains that while models with more parameters technically “fit” the data best, an ideal solution is the most parsimonious model that has an acceptable fit to the observed data. As a result, the objective of determining model fit does not necessitate that fit statistics bottom out. Instead a model should be selected when it is interpretable or substantively meaningful and parsimonious (Cleveland et al., 2013; Snyder & Smith, in press). Thus, the findings from prior literature helped with the process of determining the best model (Collins & Lanza, 2010).

We also assessed the models using entropy statistics, which are model usefulness statistics that should be as close to one as possible. In addition to evaluating fit statistics, we examined plots of each class’ results to determine which number of classes intuitively makes sense (Nylund, Bellmore, Nishina, & Graham, 2007; Snyder & Monroe,2013; Snyder & Smith, 2014). Lastly, we followed Khan et al.’s (2014) approach of examining class prevalence and preferring classes that accounted for 5% or more of the sample because this improved the reliability of estimates.

Attrition analysis

We used chi-square tests and t tests on the unweighted sample of 14,081 emerging adults with weights, stratification, and cluster variables to conduct an attrition analysis comparing the analytic sample (11,117) and the attrited cases (2,964). We found that the attrited cases contained more males (χ2(1, 14,079) = 7.39, p < 0.01); were more likely to be Black (χ2(1, 14,056) = 7.39, p < 0.001); and were more depressed (t = 3.52, p < 0.01). Regarding alcohol use and related problems, the analytic sample included more individuals who had drank five or more drinks during the past two weeks (χ2(1, 14,072) = 12.33, p < 0.001); who had experienced problems with dating because of drinking (χ2(1, 10,061) = 7.32, p < 0.01); and who had experienced a hangover (χ2(1, 10,038) = 8.40, p < 0.01). The attrited cases included more individuals who had driven drunk (χ2(1, 14,072) = 26.66, p < 0.001).

Results

Table 1 provides the sample characteristics and indicates some noteworthy differences between genders. Compared to males, females were more educated, more likely to be in a serious romantic relationship, more likely to live outside of their parents’ home, and they reported more depression symptoms. Males were more likely to have engaged in adolescent delinquent behaviors.

Table 1.

Weighted sample characteristics for male and female respondents.

Female
Male
F df p value
Unweighted n % or mean (SE) Unweighted n % or mean (SE)
Actual sample size* 5,949 49.99 5,168 50.01
Race
 Asian/Pac. Island. 291 2.52 296 2.81 0.98 1,942 > 0.05
 Black 1,228 14.57 934 13.61
 White 3,653 72.74 3,241 72.51
 Other 777 10.17 697 11.07
Hispanic
 Yes 885 10.81 831 11.66 1.04 1,942 > 0.05
Education
 < High school 440 8.54 522 11.93 12.15 1,942 < 0.001
 High school/GED 4,212 71.49 3,818 73.48
 Some college 457 6.95 338 5.56
 College 788 12.09 459 8.43
 Beyond college 52 0.92 31 0.59
Relationship status
 Single 3,824 62.68 3,756 72.81 41.22 1,942 < 0.001
 Cohabitation 1,181 19.96 704 12.74
 Married 944 17.36 708 14.45
Living situation
 Parents 2,242 36.85 2,314 44.96 13.83 1,942 < 0.001
 Another’s home 322 4.94 283 5.39
 Own place 3,054 52.63 2,274 44.18
 Group quarters 298 5.17 253 4.52
 Other 33 0.42 44 0.95
Supervisory neglect**
 Yes 449 7.49 450 7.78 0.21 1,942 > 0.05
Parental drinking ***
 Yes 714 12.46 627 14.10 3.31 1,942 > 0.05
Mean Age 5,949 21.61 (0.04) 5,168 21.72 (0.04) 6.35 1,942 < 0.05
Mean delinquency**** 5,949 0.23 (0.01) 5,168 0.32 (0.01) 105.10 1,942 < 0.001
Mean depression***** 5,949 6.11 (0.09) 5,168 4.70 (0.08) 143.11 1,942 < 0.001

Note. Above are frequencies for categorical variables and means and standard errors for continuous variables. Some numbers may not sum to 100 due to rounding. Also, to compute the F statistic, degrees of freedom, and p value for the comparisons across gender for categorical variables, we used the svy: tab command; and for continuous variables, we used the svy: mean command followed by the test command.

*

The weighted sample consists of 8,734,994 females and 8,739,895 males.

**

During Wave III respondents were asked to retrospectively answer one question regarding experiences prior to the 6th grade: “How often had your parents or other adult caregivers left you home alone when an adult should have been with you?” Ten or more experiences of neglect were valued as 1.

***

Parental drinking is a dichotomous measure capturing whether parents had consumed five or more alcoholic drinks on one occasion in the past month (1 = yes) during Wave I.

****

During Wave I respondents answered 15 questions regarding delinquent behaviors with ranges from 0 = never to 3 = five or more times; then scores were averaged, then standardized (M = 0; SD = 1).

*****

The CES-D is scored as follows: <10 indicates no depression; 10–14 indicates mild depression; and >14 indicates severe depressive symptoms. Scores ranged from 0 to 28. The values presented for depression are mean values.

Table 2 provides the prevalence of each alcohol use behavior used to form the latent class structure for both genders, and the total sample. Because of physiological differences binge drinking was measured as four drinks for females and five drinks for males. Males were significantly more likely to engage in each of the alcohol risk behaviors. For males the most frequent behavior was drinking five or more drinks in the past 2 weeks (69.62%). For females the most frequent behavior was experiencing a hangover (56.76%). The least frequent behavior for females was being drunk at school or work (3.65%), while the least frequent behavior for males was experiencing problems at school or work (6.60%).

Table 2.

Weighted percentage of respondents indicating engagement in alcohol-related behaviors.

Unweighted n Female % Unweighted n Male % F df p value
Drank ≥ 4 drinks
 past 2 weeks
3,131 53.70
Drank ≥ 5 drinks
 past 2 weeks
3,534 69.62
Problems at school
 or work because
 of drinking past 12
 months
263 6.60 425 11.21 29.20 1,712 <0.001
Problems with
 dating because of
 drinking past 12
 months
535 13.45 622 16.76 10.09 1,712 <0.01
Hung over past 12
 months
2,334 56.76 2,453 63.98 22.96 1,710 <0.001
Got into a sexual
 situation regretted
 because of
 drinking
547 20.04 660 27.01 22.08 1,512 <0.001
Got into a physical
 fight because of
 drinking
203 5.00 652 18.31 231.59 1,703 <0.001
Drunk at
 school/work past
 12 months
154 3.65 429 11.69 106.31 1,713 < 0.001
Driven drunk past
 7 years
2,702 46.38 2,987 58.80 89.64 1,942 < 0.001

Note. Unless otherwise indicated values are based on the weighted sample. Unless otherwise indicated the behavior took place during the past 12 months. There are not comparisons between the first two items because binge drinking for females is four or more drinks, while binge drinking for males is five or more drinks.

For both males and females, the results of the LCA for each of the seven classes is provided in Table 3. The top row provides the fit indices used to evaluate the models and the distributions across the classes. The left-hand side provides the number of classes in the model. For females we chose the three-class solution and for males we chose the 4-class solution. The decisions for the class solutions were based on the best fit, interpretability, and parsimony.

Table 3.

Indicators of fit with one through seven latent classes by sex.

Females
Males
LL BIC AIC Entropy Class n Class % LL BIC AIC Entropy Class n Class %
1 −16,932.37 33,934.26 33,880.73 5,949 100.00 1 −17,176.99 34,422.39 34,369.99 5,168 100.00
2 −14,663.96 29,475.66 29,361.91 0.76 2,793 46.95 2 −15,346.21 30,837.78 30,726.42 0.67 2,173 42.05
3,156 53.05 2,995 57.95
3 −14,518.47 29,262.91 29,088.95 0.61 303 5.09 3 −15,149.79 30,521.88 30,351.57 0.57 567 10.96
2,444 41.08 1,376 26.63
3,203 53.84 3,225 62.40
4 −14,450.30 29,204.78 28,970.59 0.65 233 3.92 4 −15,104.34 30,507.93 30,278.68 0.52 524 10.15
311 5.23 642 12.43
2,386 40.10 1,389 26.87
3,019 50.75 2,613 50.56
5 −14,425.93 29,234.26 28,939.86 0.62 11 0.18 5 −15,066.29 30508.78 30,220.57 0.49 279 5.40
297 5.00 526 10.18
364 6.12 609 11.78
2,393 40.23 1,397 27.04
2,883 48.47 2,357 45.61
6 −14,411.90 29,284.43 28,929.81 0.64 10 0.17 6 −15,036.50 30,526.16 30,179.00 0.51 314 6.07
18 0.31 352 6.81
353 5.93 373 7.21
1,185 19.93 452 8.74
1,952 32.81 1,402 27.12
2,430 40.86 2,276 44.04
7 −14,400.10 29,339.04 28,924.20 0.64 11 0.18 7 −15,020.09 30,570.29 30,164.18 0.57 160 5.69
30 0.51 189 7.34
188 3.16 202 8.28
277 4.66 255 9.26
332 5.58 672 10.69
2,414 40.58 1,357 13.75
2,697 45.33 2,333 44.99

Note. LL = Log Likelihood; AIC = Aikaike Information Criteria; BIC = Bayesian Information Criteria; BIC SSA = sample-size-adjusted BIC LRT = Lo-Mendell Rubin Test.

The three class model is boldfaced to indicate that it was the model chosen for females and the four class model is boldfaced to indicate the model chosen for males.

Figures 1 and 2 visually depict the item-response probabilities for each of the alcohol-use behaviors for males and females respectively. For males we found the following four classes: (1) multiple-risk drinkers, (2) moderate-risk drinkers, (3) binge-drinkers, and (4) low-risk drinkers or abstainers. In the first class, respondents reported binge drinking within the past 2 weeks and a range of associated risks, including problems at school or work, problems with friends, problems dating, regretting a sexual situation, and driving drunk. Participants in the second class have experienced problems at school or work because of drinking, have had a hangover, have regretted a sexual situation, and have driven drunk. In the third class members had drank five or more drinks, had experienced a hangover, and had driven drunk. The fourth class includes members who consumed small amounts of alcohol or abstained from drinking.

Figure 1.

Figure 1

Weighted profile plot of alcohol risk behaviors classes for females.

Figure 2.

Figure 2

Weighted profile plot of alcohol risk behaviors classes for females.

For females, we found the following three classes: (1) multiple-risk drinkers, (2) moderate-risk drinkers, and (3) low-risk drinkers or abstainers. Members of the first class had the highest probabilities of all risk behaviors. Members of the second class had high probabilities of binge drinking, hangovers, regretting a sexual situation, and drunk driving. The probabilities of risk behaviors among the third group were the lowest for all measures.

The results of regressing the covariates onto the classes using the three step method are presented as odds ratios (ORs) and 95% confidence intervals in Table 4 for males and Table 5 for females. For both males and females the low-risk drinkers or abstainers latent class functions as the reference.

Table 4.

Males odds ratios comparing multiple-risk drinkers, moderate-risk drinkers, and binge-drinkers to the low-risk drinkers or abstainers.

Multiple-risk drinkers vs. low-risk drinkers
or abstainers
Moderate-risk drinkers vs. low-risk drinkers
or abstainers
Binge drinkers vs. low-risk drinkers
or abstainers
OR 95% CI OR 95% CI OR 95% CI
Hispanic 0.53 [021–1.34] 1.38 [0.81–2.35] 0.72 [0.42–1.24]
White 2 81*** [1.42–5.57] 1.60*** [1.02–2.53] 0.67* [0.46–0.97]
Wave 3 age 0.88 [0.71–1.09] 0.88 [0.70–1.10] 1.08 [0.91–1.29]
Wave 3 education 1.02 [0.83–1.26] 1.02 [0.83–1.26] 0.70** [0.56–0.88]
Wave 3 single 5.56*** [2.86–10.81] 1.72* [1.01–2.94] 0.81 [0.55–1.19]
Wave 3 live with
 parent(s)
0.46*** [0.30–0.69] 1.08 [0.72–1.64] 1.48* [1.01–2.16]
Wave 3 child
 supervisory
 neglect
2.16* [1.11–4.21] 0.93 [0.44–1.99] 1.22 [0.68–2.20]
Wave 1 parental
 drinking
0.74 [0.41–1.33] 1.61* [1.02–2.54] 1.30 [0.77–2.19]
Wave 3 depression 1.36** [1.11–1.68] 1.44*** [1.18–1.76] 1.19 [0.99–1.43]
Wave 1
 delinquency
1 92*** [1.53–2.41] 1.34** [1.10–1.63] 0.79 [0.58–1.07]
*

p < 0.05,

**

p < 0.01,

***

p <0.001.

Note. Low-risk drinkers or abstainers are the referent for multiple-risk drinkers, moderate-risk drinkers, and binge-drinkers. Hispanic was equal to 1 for individuals who responded yes to a question asking if they were of Hispanic decent and zero for those who responded no. Thus, the referent group for Hispanics was non-Hispanics. Because of their low substance use during emerging adulthood, white was given a value of 1 and the referent group contained individuals who identified as Asian/Pacific Islander, black and other.

Table 5.

Females’ odds ratios comparing multiple-risk drinkers and binge-drinkers to the low-risk drinkers or abstainers.

Multiple-risk drinkers vs. low-risk drinkers or abstainers
Binge drinkers vs. low-risk drinkers or abstainers
OR 95% CI OR 95% CI
Hispanic 0.57 [020–1.62] 1.18 [0.87–1.61]
White 13.68*** [4.23–44.26] 1 85*** [1.44–2.37]
Wave 3 age 1.68*** [1.20–2.36] 1.40** [1.21–1.63]
Wave 3 education 6.30*** [2.71–14.65] 1.13 [0.83–1.52]
Wave 3 single 0.57* [0.34–0.98] 0.95 [0.74–1.22]
Wave 3 live with parent(s) 1.07 [0.33–3.45] 1.27 [0.76–2.12]
Wave 3 child supervisory
 neglect
2.41* [1.22–4.77] 1.03 [0.71–1.49]
Wave 1 parental drinking 0.95 [0.68–1.34] 0.97 [0.85–1.11]
Wave 3 depression 1.41* [1.06–1.89] 1.04 [0.92–1.18]
Wave l delinquency 2.01*** [1.50–2.70] 1.13 [0.96–1.34]
*

p < 0.05,

**

p < 0.01,

***

p < 0.001.

Note. Low-risk drinkers or abstainers are the referent for both multiple-risk drinkers and binge-drinkers. Hispanic was equal to 1 for individuals who responded yes to a question asking if they were of Hispanic decent and zero for those who responded no. Thus, the referent group for Hispanics was non-Hispanics. Because of their low substance use during emerging adulthood, white was given a value of 1 and the referent group contained individuals who identified as Asian/ Pacific Islander, black, and other.

Male emerging adults who were white, single, and did not live with their parents had significantly higher odds of being in the multiple-risk behaviors class compared to being in the low-risk drinkers or abstainers’ class. The odds were significantly higher of being in the multiple-risk behaviors class compared to being in the low-risk drinkers or abstainers’ class among emerging adult males who were depressed (OR = 1.36, p <0.01, 95% CI [1.11–1.68]), who had experienced child supervisory neglect (OR = 2.16, p <0.05, 95% CI [1.11–4.21]), and who had engaged in adolescent delinquency (OR = 1.92, p <0.001, 95% CI [1.53–2.41]). Males who were single, with a parent who reported drinking 5 or more drinks on a single occasion, or had engaged in adolescent delinquency had significantly higher odds of being in the moderate-risk behaviors class compared to the referent group. Being less educated and living with parents also increased the odds of membership in the binge-drinkers class compared to the referent group. Additionally, compared to the referent group, males had lower odds of being members of the binge-drinkers class if they were white.

For female emerging adults, being white, older, more educated, or in a serious romantic relationship increased the odds of membership in the multiple-risk behaviors class compared to being in the low-risk drinkers or abstainers’ class. Likewise, females who were depressed (OR = 1.41, p <0.05, 95% CI [1.06–1.89]), had experienced child supervisory neglect (OR = 2.41, p <0.05, 95% CI [1.22–4.77]), or had engaged in adolescent delinquency (OR = 2.01, p <0.001, 95% CI [1.50–2.70]) had higher odds of being in the multiple-risk behaviors class compared to the referent group.

Discussion

Applying the social development model, this study used data from Wave III of Add Health to investigate how the effect of child supervisory neglect on alcohol use during emerging adulthood differed based on gender. Our study controlled for risk and protective factors that have been linked with alcohol use, such as depression (Dixit & Crum, 2000; Weitzman, 2004), engaging in adolescent delinquency (cf. Guo et al., 2001; Hunter et al., 2014; Williams et al., 2007), being single (Fleming et al., 2010; Snyder & Rubenstein, 2014), parental drinking (White & Jackson, 2004; Wilson & Widom, 2010), living with parents (Gfroerer et al., 1997; White et al., 2006), and identifying as white (Chen et al., 2009). Our study is the first to examine how child supervisory neglect affects patterns of alcohol use differentially among male and female emerging adults.

Our first hypothesis is supported because both the male and female models demonstrated similar results to prior studies (cf. Beseler et al., 2012; Cleveland et al.,2013). The four class model we identified for males consisted of (1) multiple-risk drinkers, (2) moderate-risk drinkers, (3) binge-drinkers, and (4) low-risk drinkers or abstainers. The three class model identified for females consisted of (1) multiple-risk drinkers, (2) moderate-risk drinkers, and (3) low-risk drinkers or abstainers.

We found partial support for our second hypothesis. Consistent with the literature on general neglect (Patock-Peckham and Morgan-Lopez, 2010; Mullings et al., 2004), for both males and females, child supervisory neglect more than doubled the odds of membership in the multiple-risk drinkers class compared to low-risk drinkers or abstainers class. However, supervisory neglect did not increase the odds of membership in other classes for either male or female emerging adults. This finding suggests experiencing supervisory neglect is associated with a greater likelihood of engaging in the most deleterious combination of behaviors as opposed to regulating drinking behaviors. Future research should consider assessing the nuances of neglectful supervision that leads to an inability to moderate alcohol use as an emerging adult.

Similar to the findings of prior studies (Dixit & Crum,2000; Harrell & Karim, 2008; Weitzman, 2004), depression increased the odds of membership in the multiple-risk drinkers class compared to the referent class for both males and females. Depression raised the odds of moderate-risk drinkers for male emerging adults, but not for females. Dixit and Crum (2000) found that a history of depressive disorder more than doubled the risk of heavy drinking in women. So it may be that depressed women generally do not moderate their drinking. Alternatively, it could be that similar to the findings of Lee et al. (2012), that males’ alcohol risk behaviors do not vary with depression as dramatically as females’ risk behaviors do. The results pertaining to depression should be interpreted with caution because a single depression measure was used at one point in time, which does not capture past or future episodes of depression.

Also consistent with the literature (cf. Guo et al., 2001; Hunter et al., 2014; Williams et al., 2007), for both males and females, adolescent delinquency increased the odds of membership in the multiple-risk drinkers class compared to low-risk drinkers or abstainers class. For males only adolescent delinquency also increased the odds of membership in the moderate-risk drinkers’ class. One explanation for this gender difference is that males are more likely to engage in adolescent delinquent behaviors (Connell et al., 2011; Farrington et al., 2010; Schwartz et al., 2010).

Interestingly, parent-reported parental heavy drinking at Wave I only increased the odds of membership in the multiple-risk drinking class for males, but did not have a relationship to class membership for females. Perhaps many emerging adults who have seen a parent abuse alcohol are less likely to abuse alcohol themselves. Another explanation for these results contrary to prior studies (White & Jackson, 2004) is that parents may not have answered the question regarding their drinking behavior accurately. These results should also be interpreted with caution because the measure captured responses from mothers, grandmothers and others who may be less likely to drink five or more drinks at a time. The fathers who were largely absent from this measure have been linked to female heavy drinking in Dixit and Crum’s (2000) study.

Similar to prior studies (Fleming et al., 2010; Snyder & Merritt, 2015; Snyder & Rubenstein, 2014), for males being single raised the odds of multiple-risk drinkers’ class membership and moderate-risk drinkers’ class membership. These findings suggest that serious romantic relationships are especially protective for males. However, for females being single reduced the odds of being in the multiple-risk drinkers class. The interpretation of these findings is complicated by the lack of information on the quality of the relationship (see Fleming et al., 2010). It may be that the female respondents were unhappy in their relationships and used alcohol to self-medicate. In addition, we do not have information regarding partners’ substance use behavior. So it may be that males reduce their alcohol risk behavior because they enter a relationship with a female who engages in few alcohol risk behaviors, while females could increase their alcohol risk behaviors in response to having a partner who engages in multiple risk behaviors.

Consistent with the findings of White et al. (2006) and Gfroerer, Greenblatt, and Wright (1997), living with a parent was a protective factor for males and decreased the odds of multiple-risk drinkers’ class membership. However, it was not protective for other male classes or for any female classes. It is likely that this finding only applies to males who did not have a parent who reported binge drinking. Males living at home had increased odds of being in the binge drinking class compared to the referent class. On the one hand, it may be that males who reside with their parents may engage in less risky drinking behaviors because of parental oversight. Alternatively, the protective nature of living with parents may not have been as strong for alcohol as it may be for other substances since the use of alcohol is more socially acceptable.

For both males and females, consistent with the literature, being white (Chen et al., 2009) was a risk factor that significantly increased the odds of membership in either the multiple-risk drinkers or the moderate-risk drinkers’ classes. Simultaneously, white males also had a lowered risk of membership in the binge drinking class. Perhaps binge-drinking for males is more of a normative behavior that transcends racial differences. Less educated males had increased odds of membership in the binge-drinking class. For females only, being older and being more educated increased the odds of membership in the multiple-risk drinkers’ class. This seems to suggest that risk behaviors associated with college-life or later stages of emergence to adulthood may be especially pronounced for females.

Implications

Although alcohol risk behaviors among emerging adults constitute a complicated issue, important steps can be taken to address the problem in the realms of research, practice, and policy. Starting with research, future studies should replicate this study to explore these relationships during adolescence, middle, and late-adulthood. Ideally researchers should conduct longitudinal studies that examine how these relationships change over time. It would also be informative if qualitative researchers would explore how and why child supervisory neglect influences alcohol use among male emerging adults, in addition to a nuanced assessment regarding the range of behaviors associated with neglect. Clinicians who are treating emerging adults for alcohol abuse should ask clients about child supervisory neglect. Finally, policies should be enacted that fund programs designed to address the mental health needs of victims of supervisory neglect prior to emerging adulthood.

While our study has several strengths, we did not examine how the relationships between alcohol use and child supervisory neglect change over time. Many of the findings presented in this study are cross-sectional so cause and effect cannot be established. Additionally, it would have been informative if there had been more questions regarding various forms of neglect similar to the Mullings et al. (2004) study, questions regarding the ages at which time neglect occurred, the frequency of neglect, and the features of the neglectful family. Another limitation is that recall bias could have affected the response rate and correctness of responses. Hopefully, future studies can build on the work of this study to further our understanding of the effects of the most prevalent form of maltreatment in the United States on risky alcohol use behaviors.

Glossary

Latent class analysis (LCA)

LCA is a person-centered analytic approach that uses maximum likelihood procedures to comb through the dataset and identify homogenous subgroups of individuals drawn from a larger heterogeneous sample or population, at a single point in time.

Supervisory neglect

Supervisory neglect constitutes situations when adult supervision of a child is inadequate to meet the child’s needs.

Biographies

Notes on contributors

graphic file with name nihms-774795-b0003.gif

Susan M. Snyder, Ph.D., is an Assistant Professor in the School of Social Work at the Georgia State University. Dr. Snyder’s work focuses on problem behavior, including substance use and illegal behaviors among system-involved youth and emerging adults. Her work contributes to the understanding of the correlates associated with typologies of cooccurring problem behaviors.

graphic file with name nihms-774795-b0004.gif

Darcey H. Merritt, Ph.D., M.S.W. is an Assistant Professor at the New York University Silver School of Social Work. Her research focuses on families involved with the public child welfare system, with particular attention to the perspectives of children and parents; distinguishing between specific types of maltreatment in an effort to highlight appropriate needed resources and services; and the impact of client perceived experiences on socio-developmental and behavioral outcomes. Her work contributes to developing a more refined understanding of the characteristics of abusive and neglectful parents, the neighborhood structural impact on parenting, and parental and child well-being outcomes for child welfare involved families.

Footnotes

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.

References

  1. American Psychiatric Association . Diagnostic and statistical manual of mental disorders (DSM III) Author; Washington, DC: 1994. [Google Scholar]
  2. APSAC . Practice guidelines: Challenges in the evaluation of child neglect. American Professional Society on the Abuse of Children (APSAC); Elmhurst, IL: 2008. [Google Scholar]
  3. Arnett JJ. The developmental context of substance use in emerging adulthood. Journal of Drug Issues. 2005;35(2):235–253. [Google Scholar]
  4. Barnes J, Boutwell BB, Morris RG, Armstrong TA. Explaining differential patterns of self-reported delinquency: Evidence from a latent class analysis of sibling pairs. Journal of Contemporary Criminal Justice. 2012;28(3):254–272. [Google Scholar]
  5. Beseler CL, Taylor LA, Kraemer DT, Leeman RF. A latent class analysis of DSM-IV alcohol use disorder criteria and binge drinking in undergraduates. Alcoholism: Clinical and Experimental Research. 2012;36(1):153–161. doi: 10.1111/j.1530-0277.2011.01595.x. doi:10.1111/j.1530-0277.2011.01595.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Blanco C, Okuda M, Markowitz JC, Liu SM, Grant BF, Hasin DS. The epidemiology of chronic major depressive disorder and dysthymic disorder: Results from the national epidemiologic survey on alcohol and related conditions. The Journal of Clinical Psychiatry. 2010;71(12):1645–1656. doi: 10.4088/JCP.09m05663gry. doi:10.4088/JCP.09m05663gry. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bouchery EE, Harwood HJ, Sacks JJ, Simon CJ, Brewer RD. Economic costs of excessive alcohol consumption in the US, 2006. American Journal of Preventive Medicine. 2011;41(5):516–524. doi: 10.1016/j.amepre.2011.06.045. [DOI] [PubMed] [Google Scholar]
  8. Catalano RF, Hawkins JD. The social development model: A theory of antisocial behavior. In: Hawkins JD, editor. Delinquency and crime: Current Theories. Cambridge University Press; New York: 1996. [Google Scholar]
  9. Catalano RF, Kosterman R, Hawkins JD, Newcomb MD, Abbott RD. Modeling the etiology of adolescent substance use: A test of the social development model. Journal of Drug Issues. 1996;26(2):429. doi: 10.1177/002204269602600207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chantala K. Guidelines for analyzing Add Health data. National Longitudinal Study of Adolescent Health; University of North Carolina at Chapel Hill: 2006. [Google Scholar]
  11. Chantala K, Tabor J. National Longitudinal Study of Adolescent Health: Strategies to perform a design-based analysis using the Add Health data. 2010 Retrieved from http://www.cpc.unc.edu/projects/addhealth/data/guides/weight1.pdf.
  12. Chartier KG, Hesselbrock MN, Hesselbrock VM. Alcohol problems in young adults transitioning from adolescence to adulthood: The association with race and gender. Addictive Behaviors. 2011;36(3):167–174. doi: 10.1016/j.addbeh.2010.10.007. doi:10.1016/j.addbeh.2010.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen CM, Yi H, Williams GD, Faden VB. Surveillance report #86: Trends in underage drinking in the United States, 1991–2007. National Institute on Alcohol Abuse and Alcoholism; Bethesda, MD: 2009. [Google Scholar]
  14. Cleveland MJ, Mallett KA, White HR, Turrisi R, Favero S. Patterns of alcohol use and related consequences in non-college-attending emerging adults. Journal of Studies on Alcohol and Drugs. 2013;74(1):84. doi: 10.15288/jsad.2013.74.84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Collins LM, Lanza ST. Latent class and latent transition analysis with applications in the social, behavioral, and health sciences. John Wiley & Sons; Hoboken, NJ: 2010. [Google Scholar]
  16. Connell CM, Cook EC, Aklin WM, Vanderploeg JJ, Brex RA. Risk and protective factors associated with patterns of antisocial behavior among non-metropolitan adolescents. Aggressive Behavior. 2011;37(1):98–106. doi: 10.1002/ab.20370. doi:10.1002/ab.20370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Coohey C. Defining and classifying supervisory neglect. Child Maltreatment. 2003;8(2):145–156. doi: 10.1177/1077559502250786. doi:10.1177/1077559502250786. [DOI] [PubMed] [Google Scholar]
  18. Currie J, Tekin E. Does child abuse cause crime? (NBER Working Paper No.12171) National Bureau of Economic Research; Cambridge, MA: 2006. Retrieved from http://www.nber.org/papers/w12171. [Google Scholar]
  19. Delucchi KL, Matzger H, Weisner C. Alcohol in emerging adulthood: 7-year study of problem and dependent drinkers. Addictive Behaviors. 2008;33(1):134–142. doi: 10.1016/j.addbeh.2007.04.027. doi: http://dx.doi.org/10.1016/j.addbeh.2007.04.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Dixit AR, Crum RM. Prospective study of depression and the risk of heavy alcohol use in women. The American Journal of Psychiatry. 2000;157(5):751–758. doi: 10.1176/appi.ajp.157.5.751. doi:10.1176/appi.ajp.157.5.751. [DOI] [PubMed] [Google Scholar]
  21. Farrington DP, Jolliffe D, Hawkins JD, Catalano RF, Hill KG, Kosterman R. Why are boys more likely to be referred to juvenile court? Gender differences in official and self-reported delinquency. Victims & Offenders. 2010;5(1):25–44. doi:10.1080/15564880903422963. [Google Scholar]
  22. Fleming CB, Catalano RF, Mazza JJ, Brown EC, Haggerty KP, Harachi TW. After-school activities, misbehavior in school, and delinquency from the end of elementary school through the beginning of high school. The Journal of Early Adolescence. 2008;28(2):277–303. doi:10.1177/0272431607313589. [Google Scholar]
  23. Fleming CB, White HR, Catalano RF. Romantic relationships and substance use in early adulthood. Journal of Health and Social Behavior. 2010;51(2):153–167. doi: 10.1177/0022146510368930. doi:10.1177/0022146510368930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Gfroerer JC, Greenblatt JC, Wright DA. Substance use in the US college-age population: differences according to educational status and living arrangement. American Journal of Public Health. 1997;87(1):62–65. doi: 10.2105/ajph.87.1.62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Guo J, Hawkins JD, Hill KG, Abbott RD. Childhood and adolescent predictors of alcohol abuse and dependence in young adulthood. Journal of Studies on Alcohol. 2001;62(6):754–762. doi: 10.15288/jsa.2001.62.754. doi:10.15288/jsa.2001.62.754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Harrell ZAT, Karim NM. Is gender relevant only for problem alcohol behaviors? An examination of correlates of alcohol use among college students. Addictive Behaviors. 2008;33(2):359–365. doi: 10.1016/j.addbeh.2007.09.014. doi:10.1016/j.addbeh.2007.09.014. [DOI] [PubMed] [Google Scholar]
  27. Harris KM, Halpern CT, Whitsel EA, Hussey JM, Tabor J, Entzel P, Udry J. The National Longitudinal Study of Adolescent Health: Research Design [WWW document] 2009 Retrieved September 15, 2010, from http://www.cpc.unc.edu/projects/addhealth/design. [Google Scholar]
  28. Hingson RW, White AM. Mental health care in the college community. John Wiley & Sons, Ltd; 2010. Magnitude and prevention of college alcohol and drug misuse: US College Students Aged 18–24; pp. 289–324. [Google Scholar]
  29. Horwitz AV, Widom CS, McLaughlin J, White HR. The impact of childhood abuse and neglect on adult mental health: A prospective study. Journal of Health and Social Behavior. 2001;42(2):184–201. doi:10.2307/3090177. [PubMed] [Google Scholar]
  30. Hunter SB, Miles JNV, Pedersen ER, Ewing BA, D’Amico EJ. Temporal associations between substance use and delinquency among youth with a first time offense. Addictive Behaviors. 2014;39(6):1081–1086. doi: 10.1016/j.addbeh.2014.03.002. doi:10.1016/j.addbeh.2014.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kanny D, Liu Y, Brewer RD, Lu H. Binge drinking-United States, 2011 CDC Health Disparities and Inequalities Report — United States, 2013. Vol. 62. Centers for Disease Control and Prevention; Washington, DC: 2013. pp. 77–80. [Google Scholar]
  32. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Rush AJ, National Comorbidity Survey Replication The epidemiology of major depressive disorder: Results from the national comorbidity survey replication (ncs-r) JAMA. 2003;289(23):3095–3105. doi: 10.1001/jama.289.23.3095. doi:10.1001/jama.289.23.3095. [DOI] [PubMed] [Google Scholar]
  33. Khan MR, Cleland CM, Scheidell JD, Berger AT. Gender and racial/ethnic differences in patterns of adolescent alcohol use and associations with adolescent and adult illicit drug use. The American Journal of Drug and Alcohol Abuse. 2014;40(3):213–224. doi: 10.3109/00952990.2014.892950. doi:10.3109/00952990.2014.892950. [DOI] [PubMed] [Google Scholar]
  34. Kopak AM, Chia-Chen Chen A, Haas SA, Gillmore MR. The importance of family factors to protect against substance use related problems among Mexican heritage and White youth. Drug and Alcohol Dependence. 2012;124(1-2):34–41. doi: 10.1016/j.drugalcdep.2011.12.004. doi:10.1016/j.drugalcdep.2011.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lee JO, Kosterman R, McCarty CA, Hill KG, Hawkins JD. Can patterns of alcohol use disorder in young adulthood help explain gender differences in depression? Comprehensive Psychiatry. 2012;53(8):1071–1077. doi: 10.1016/j.comppsych.2012.03.012. doi:10.1016/j.comppsych.2012.03.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. McCutcheon AL. Basic concepts and procedures in single-and multiple-group latent class analysis. In: Hagenaars JA, McCutcheon AL, editors. Applied latent class analysis. Cambridge University Press; New York, NY: 2002. pp. 56–88. [Google Scholar]
  37. Mennen FE, Kim K, Sang J, Trickett PK. Child neglect: Definition and identification of youth’s experiences in official reports of maltreatment. Child Abuse & Neglect. 2010;34(9):647–658. doi: 10.1016/j.chiabu.2010.02.007. doi: http://dx.doi.org/10.1016/j.chiabu.2010.02.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Merritt DH. Child abuse potential: Correlates with child maltreatment rates and structural measures of neighborhoods. Children and Youth Services Review. 2009;31(8):927–934. doi:10.1016/j.childyouth.2009.04.009. [Google Scholar]
  39. Merritt DH, Klein S. Do early care and education services improve language development for maltreated children? Evidence from a national child welfare sample. Child Abuse & Neglect. 2015;39:185–196. doi: 10.1016/j.chiabu.2014.10.011. doi:10.1016/j.chiabu.2014.10.011. [DOI] [PubMed] [Google Scholar]
  40. Mullings JL, Hartley DJ, Marquart JW. Exploring the relationship between alcohol use, child maltreatment, and treatment needs among female prisoners. Substance Use & Misuse. 2004;39(2):277–305. doi: 10.1081/ja-120028491. doi:10.1081/JA-120028491. [DOI] [PubMed] [Google Scholar]
  41. NIAAA . Women and alcohol. National Institute on Alcohol Abuse and Alcoholism; Bethesda, MD: 2013. Retrieved March 12, 2015, from http://pubs.niaaa.nih.gov/publications/womensfact/womensfact.htm. [Google Scholar]
  42. NIAAA . Drinking levels defined. National Institute on Alcohol Abuse and Alcoholism; Bethesda, MD: n.d. Retrieved March 12, 2015, from http://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/moderate-binge-drinking. [Google Scholar]
  43. Nylund K, Bellmore A, Nishina A, Graham S. Subtypes, severity, and structural stability of peer victimization: What does latent class analysis say? Child Development. 2007;78(6):1706–1722. doi: 10.1111/j.1467-8624.2007.01097.x. doi:10.1111/j.1467-8624.2007.01097.x. [DOI] [PubMed] [Google Scholar]
  44. Patock-Peckham JA, Morgan-Lopez AA. College drinking behaviors: Meditational links between parenting styles, parental bonds, depression, and alcohol problems. Psychology of Addictive Behaviors. 2007;21(3):297. doi: 10.1037/0893-164X.21.3.297. [DOI] [PubMed] [Google Scholar]
  45. SAMHSA Results from the 2012 National Survey on Drug Use and Health: Summary of national findings. 2013 Retrieved from http://www.samhsa.gov/data/NSDUH/2k10NSDUH/2k10Results.pdf.
  46. SAMHSA . Results from the 2013 National Survey on Drug Use and Health summary of national findings NSDUH Series H-38A. Substance Abuse and Mental Health Services Administration; Rockville, MD: 2014. [Google Scholar]
  47. Schwartz SJ, Forthun LF, Ravert RD, Zamboanga BL, Umaña-Taylor AJ, Filton BJ, Hudson M. Identity consolidation and health risk behaviors in college students. American Journal of Health Behavior. 2010;34(2):214–224. [PMC free article] [PubMed] [Google Scholar]
  48. Shin SH, Edwards EM, Heeren T. Child abuse and neglect: Relations to adolescent binge drinking in the National Longitudinal Study of Adolescent Health (AddHealth) Study. Addictive Behaviors. 2009;34(3):277–280. doi: 10.1016/j.addbeh.2008.10.023. doi:10.1016/j.addbeh.2008.10.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Snyder SM, Gwaltney AY, Landeck EA. What bonds influence of patterns of substance use among child-welfare involved youth? Journal of Psychoactive Drugs. 2015;47:308–316. doi: 10.1080/02791072.2015.1075091. doi:10.1080/02791072.2015.1075091. [DOI] [PubMed] [Google Scholar]
  50. Snyder SM, Merritt DH. The influence of supervisory neglect on subtypes of emerging adult substance use after controlling for familial factors, relationship status, and individual traits. Substance Abuse, (just-accepted) 2015:00–00. doi: 10.1080/08897077.2014.997911. doi:10.1080/08897077.2014.997911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Snyder SM, Monroe CR. Do child physical abuse and adolescent peer relationships influence typologies of illegal and substance-use behaviors during emerging adulthood? Journal of the Society for Social Work and Research. 2013;4(3):214–243. doi:10.5243/jsswr.2013.15. [Google Scholar]
  52. Snyder SM, Rubenstein CR. Do incest, depression, parental drinking, serious romantic relationships, and living with parents influence patterns of substance use during emerging adulthood? Journal of Psychoactive Drugs. 2014;46(3):188–197. doi: 10.1080/02791072.2014.914610. doi:10.1080/02791072.2014.91461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Snyder SM, Smith RE. Do physical abuse, depression, and parental substance use influence patterns of substance use among child welfare involved youth? Substance Use & Misuse. 2014;50(2):226–235. doi: 10.3109/10826084.2014.966845. doi:10.3109/10826084.2014.966845. [DOI] [PubMed] [Google Scholar]
  54. Snyder SM, Smith RE. Do youth with substantiated child maltreatment investigations have distinct patterns of problem behaviors? Children and Youth Services Review. in press. [Google Scholar]
  55. Weitzman ER. Poor mental health, depression, and associations with alcohol consumption, harm, and abuse in a national sample of young adults in college. The Journal of Nervous and Mental Disease. 2004;192(4):269–277. doi: 10.1097/01.nmd.0000120885.17362.94. [DOI] [PubMed] [Google Scholar]
  56. White HR, Jackson KM. Social and psychological influences on emerging adult drinking behavior. Alcohol Research & Health; Alcohol Research & Health. 2004;28(4):182–190. [Google Scholar]
  57. White HR, McMorris BJ, Catalano RF, Fleming CB, Haggerty KP, Abbott RD. Increases in alcohol and marijuana use during the transition out of high school into emerging adulthood: The effects of leaving home, going to college, and high school protective factors. Journal of Studies on Alcohol. 2006;67(6):810–822. doi: 10.15288/jsa.2006.67.810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Widom CS, Hiller-Sturmhofel S. Alcohol abuse as a risk factor for and consequence of child abuse. Alcohol Research and Health. 2001;25(1):52–57. [PMC free article] [PubMed] [Google Scholar]
  59. Widom CS, Ireland TO, Glynn PJ. Alcohol abuse in abused and neglected children followed-up: Are they at increased risk? Journal of Studies on Alcohol. 1995;56(2):207–217. doi: 10.15288/jsa.1995.56.207. [DOI] [PubMed] [Google Scholar]
  60. Widom CS, White HR, Czaja SJ, Marmorstein NR. Long-term effects of child abuse and neglect on alcohol use and excessive drinking in middle adulthood. Journal of Studies on Alcohol and Drugs. 2007;68(3):317–326. doi: 10.15288/jsad.2007.68.317. doi:10.15288/jsad.2007.68.317. [DOI] [PubMed] [Google Scholar]
  61. Williams JH, Van Dorn RA, Ayers CD, Bright CL, Abbott RD, Hawkins JD. Understanding race and gender differences in delinquent acts and alcohol and marijuana use: A developmental analysis of initiation. Social Work Research. 2007;31(2):71–81. [Google Scholar]
  62. Wilsnack RW, Wilsnack SC. Gender and alcohol: consumption and consequences. In: Boyle P, Boffetta P, Lowenfels AB, Burns H, Brawley O, Zatonski W, Rehm J, editors. Alcohol: Science, policy and public health. Oxford University Press; Oxford, United Kingdom: 2013. pp. 153–160. [Google Scholar]
  63. Wilson HW, Widom CS. Predictors of drug-use patterns in maltreated children and matched controls followed up into middle adulthood. Journal of Studies on Alcohol and Drugs. 2010;71(6):801. doi: 10.15288/jsad.2010.71.801. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES