Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jul 13.
Published in final edited form as: Drug Alcohol Depend. 2014 Jan 3;136:92–99. doi: 10.1016/j.drugalcdep.2013.12.012

DSM-5 latent classes of alcohol users in a population-based sample: Results from the São Paulo Megacity Mental Health Survey, Brazil✩

João Mauricio Castaldelli-Maia a,b,c,*, Camila M Silveira a,b, Erica R Siu a, Yuan-Pang Wang a, Igor A Milhorança a, Clóvis Alexandrino-Silva a, Guilherme Borges d, Maria C Viana e, Arthur G Andrade b,c, Laura H Andrade a,1, Silvia S Martins f,1
PMCID: PMC4943751  NIHMSID: NIHMS692695  PMID: 24440273

Abstract

Background

We aimed to identify different categorical phenotypes based upon the DSM-V criteria of alcohol use disorders (AUD) among alcohol users who had at least one drink per week in the past year (n = 948).

Methods

Data are from the São Paulo Megacity Mental Health Survey collected in 2005–2007, as part of the World Mental Health Survey Initiative. A latent class analysis of the 11 DSM-5-AUD criteria was performed using Mplus, taking into account complex survey design features. Weighted logistic regression models were used to examine demographic correlates of the DSM-5-AUD latent classes.

Results

The best latent-class model was a three-class model. We found a “non-symptomatic class” (69.7%), a “use in larger amounts class” (23.2%), defined by high probability (>70%) of the “use in larger amounts” criterion only, and a “high-moderate symptomatic class” (7.1%), defined by high-moderate probability of all the 11 AUD criteria. Compared to those in the non-symptomatic class, individuals in the “high-moderate symptomatic class” were more likely to have been married, have lower educational attainment and to be unemployed or in non-regular/informal employment. Those on the “use in larger amounts class” were more likely to have been married or never married.

Conclusion

The two symptomatic classes clearly represented the dimensionality of the new proposed AUD criteria, and could be more specifically targeted by different prevention or treatment strategies. DSM-5-AUD has the advantage of shedding light on risky drinkers included in the “use in larger amounts class”, allowing for preventive interventions, which will reach a large number of individuals.

Keywords: Alcohol use disorders (AUD), DSM-5 criteria, Latent class analysis, São Paulo, Brazil

1. Introduction

The recently released Fifth Edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) brings a revised alcohol use disorder (DSM-5 AUD) diagnosis that combines DSM-IV alcohol abuse and dependence criteria, with the following notations: the exclusion of the legal problems abuse criterion (due to its low prevalence among alcohol users; Keyes and Hasin, 2008; Saha et al., 2006; Babor and Caetano, 2008) and the inclusion of the craving criterion from the ICD-10 (International Classification of Diseases – Tenth Edition; Keyes et al., 2011; Casey et al., 2012). As reported by nearly a quarter of all current alcohol users (Cherpitel et al., 2010), this totals 11 criteria (O’Brien, 2010). There is the need to explore the applicability of this new diagnosis in different populations and countries (Mewton et al., 2011; Borges et al., 2010; Shmulewitz et al., 2010), including urban populations in middle-income countries such as Brazil, the largest country in Latin America with approximately 190 million inhabitants (The World Bank, 2013).

This new unidimensional diagnosis for AUD addresses some important issues. Firstly, the idea that the abuse diagnosis would always precede the diagnosis of dependence has been disproven by a number of prospective studies (Grant et al., 2001; Schuckit et al., 2001, 2008). For instance, in a recent paper about transitions across alcohol use stages in the São Paulo Megacity Health Survey (SPMHS), without imposing the DSM-IV hierarchy of abuse and dependence, we identified three different sequences of transitions in AUD: (i) around 15% of the positive cases for both AUD had concurrent onsets of abuse and dependence when the first DSM-IV non-dependent alcohol abuse (NDAA) occurs for the first time during the same year of life as the first DSM-IV alcohol dependence problem; (ii) around 10% had dependence before abuse, when the first dependence problem predates the first NDAA problem and (iii) around 75% had abuse before dependence, when the first NDAA problem predates the first dependence problem (Silveira et al., 2011). Secondly, the two distinct diagnostic categories in DSM-IV leave a number of individuals with severe problems related to alcohol use as “diagnostic orphans” (Hasin and Paykin, 1998; McBride et al., 2009; Caetano et al., 2011; Peer et al., 2013), because they do not meet the minimum of three criteria required to have dependence nor have any criteria of abuse (Degenhardt et al., 2002; Lynskey and Agrawal, 2007; Martin et al., 2008). This is overcome by DSM-5-AUD with the threshold for AUD diagnosis set at two or more criteria.

Notwithstanding recent research (Mewton et al., 2011; Borges et al., 2010; Shmulewitz et al., 2010; Hagman and Cohn, 2011) reinforcing the unidimensional model, there is also strong evidence of the existence of different classes of alcohol dependents (Moss et al., 2007; Delucchi et al., 2004; Hedden et al., 2010) or alcohol users (Ko et al., 2010; McBride et al., 2011). These studies explored the endorsement of DSM-IV alcohol use disorder criteria (Moss et al., 2007; Ko et al., 2010; McBride et al., 2011), longitudinal drinking patterns (Delucchi et al., 2004), and problems associated with different classes of individuals with AUD (Hedden et al., 2010). In addition, substance use disorders are well-known clinically heterogeneous syndromes (Kendler et al., 2013), as described by studies using latent class analysis (LCA), a statistical method that generates categorical phenotypes within a single unidimensional diagnostic construct (Muthén and Asparouhov, 2006). However, none of these studies used the DSM-5 AUD criteria, with the inclusion of craving criterion and exclusion of legal problems criterion, therefore, it is important to further extend this line of research by examining DSM-5 AUD criteria via LCA in other populations of alcohol users.

The current report presents results of this type of analysis using data from the Sao Paulo Megacity Health Survey (SPMHS), the Brazilian counterpart of the World Mental Health Survey Initiative (WMHS). Different from other WMHS countries, the SPMHS data were collected in an ungated approach, i.e., individuals did not need to meet the criteria for alcohol abuse to be asked about dependence symptoms (Degenhardt et al., 2008; Silveira et al., 2011). This provided a unique opportunity of investigating all DSM-5 AUD criteria in this setting. We aimed to explore latent classes among alcohol users who had at least 1 drink per week in the past year in the Brazilian general adult population. As secondary aims, we explored the sociodemographic characteristics associated with identified categorical phenotypes within the new diagnosis for AUD and the distribution of these phenotypes within DSM-5 alcohol use disorders.

2. Methods

2.1. Ethics committee approval

The procedures for recruitment, obtaining informed consent, and protection of human subjects involved during field procedures of São Paulo Megacity Health Survey were approved by the Research and Ethics Committee of the University of São Paulo Medical School (Project number 792/03).

2.2. Sample

The São Paulo Megacity Mental Health Survey (SPMHS) is a cross-sectional population-based study, designed to evaluate psychiatric morbidity in a representative sample in the general population, aged 18 years or more, living in the São Paulo Metropolitan Area (SPMA) (Andrade et al., 2012; Viana and Andrade, 2012). Respondents were selected through a multistage probabilistic process covering the 39 municipalities of SPMA, without replacement. Respondents were assessed using the World Mental Health Study (World Mental Health Survey) Composite International Diagnostic Interview (WMH-CIDI) of the World Health Organization, which was translated and adapted to Brazilian Portuguese (Viana et al., 2009). Data collection occurred between May 2005 and April 2007, by trained interviewers. The final sample assessed was composed of 5037 individuals, with a response rate of 81.3%. The current analyses were restricted to 948 subjects who endorsed drinking at least 1 drink per week in the past year. This sub-sample was chosen because of the low drinking levels in South America (Rehm et al., 2003) and the usual criteria for selecting a past-year drinking sample was too restrictive for this survey sample.

The WMH-CIDI is composed of clinical and non-clinical sections, arranged in two parts, generating diagnoses according to DSM-IV and ICD-10. All respondents received the assessment modules of mood, anxiety, and impulse control disorders, along with substance abuse and suicidal behavior, considered nuclear disorders, as well as a sociodemographic module. Blind clinical re-appraisal using the Structured Clinical Interview for DSM-IV Axis I disorder (SCID-I) for last 12-month DSM-IV Disorders in a probabilistic subsample of WMH respondents found generally good agreement between WMH-CIDI diagnoses and SCID diagnoses (Haro et al., 2006). Preliminary results of the clinical reappraisal study in the SPMHS with a probability subsample of 775 respondents, not included in the previous validation study, showed good total classification accuracy (range: 76–99%) and an area under the Receiver Operating Characteristics curve around 0.7 for any disorder (data available from the authors).

2.3. Measures

The 948 subjects who endorsed drinking at least one drink per week in the past year were asked a series of questions derived from DSM-IV/ICD-10 alcohol abuse/dependence criteria embedded in the WMH-CIDI (see supplementary materials for question details2). Twelve individuals did not answer these questions and were excluded from the statistical analysis (n = 936). These questions were combined to generate 11 dichotomous variables matching the DSM-5 (Diagnostic and Statistical Manual of Mental Disorders, 2013) criteria (see supplementary materials for the 11 criteria3). The abbreviation of the names of the variables was based on previously published studies (Hasin and Beseler, 2009; Shmulewitz et al., 2010). Positive cases of DSM-5 AUD had to endorse at least 2 of the following 11 criteria at any point in their lifetime: Tolerance, Withdrawal, Larger/Longer, Quit/Control, Time Spent, Activities Given Up, Physical/Psychological, Neglect Roles, Social/Interpersonal, Hazardous Use and Craving. In addition, DSM-5 AUD cases were classified according to three levels of severity based on the number of criteria endorsed: mild (2 or 3 criteria); moderate (4 or 5 criteria); or severe (6 or more criteria).

The following sociodemographic correlates were considered: gender (female and male), age (categorized as 18–34, 35–54 and ≥55); education (0–8, 9–11, ≥12); marital status (married/cohabiting, previously married (separated/divorced/widowed), never married); employment status (working/student, homemaker, retired, unemployed/other: non-regular or informal employment.); and household income, defined in categories (low = up to 3918 US dollars/year, low-average = from 3919 to 7050 US dollars/year, high-average = from 7051 to 14,826 US dollars/year, high = more than 14,826 US dollars/year) based on the respondent’s household income per family member divided by the median income-per-family member in the entire sample.

2.4. Statistical analysis

All analyses were performed with Mplus version 6 (Muthén and Muthén, 1998–2010), using sampling weights and complex survey design measures. Descriptive statistics were used to describe the sample. Specifically, counts and percentages were used to describe categorical variables.

Latent class analysis (LCA) was conducted with Mplus version 6.0, using maximum likelihood ratio estimation. The random option in Mplus was applied to ensure convergence for the most successful LCA models. Specifically, 500 sets of random starting values were used in the initial phase, and 10 optimizations were used in the final stage of convergence. This process ensured that the best log-likelihood (LL) value for each model was replicated several times. Several statistical indices were used to assess model fit, including LL, Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and sample-size-adjusted BIC (SSABIC). A value of LL together with smaller amounts in AIC, BIC and SSABIC may reflect a parsimonious model (Muthén, 2006). However, the BIC value has been shown to be more reliable than other information (Nylund et al., 2007).

Once the number of classes was determined, we used polytomous logistic regression models (e.g., Ko et al., 2010) in Mplus (pseudo-class draw method) to investigate the association between latent classes and demographic characteristics. The Mplus auxiliary option was used to identify the DSM-5 variables (DSM-5 AUD, Mild DSM-5 AUD, Moderate DSM-5 AUD, and Severe DSM-5 AUD) for which the equality of means across latent classes was tested using pseudo-class-based multiple imputations (Asparouhov, 2007).

3. Results

Table 1 presents the prevalence of DSM-5 AUD criteria endorsement and the distribution of sociodemographic variables in the sample. Similar to a previous study investigating lifetime alcohol use (Silveira et al., 2011), there was a remarkably higher prevalence of men among past-year alcohol users compared to the total SPMHS sample. Other significant differences were found in age, household income, employment status and education.

Table 1.

Prevalence of socio-demographic variables of individuals who had at least 1 drink/week in the past-year (n = 936) and from the total representative sample (n = 5037) of the São Paulo metropolitan area, 2005–2007.

Variables Past-year weekly alcohol users
Total sample
na % SE na % SE λ p
Age (years)
18–34 390 41.69 1.75 2252 44.70 0.73 14.4 <0.01
35–54 414 44.28 2.01 1916 38.03 0.90
≥55 131 14.03 1.54 869 17.26 0.63
Gender
Female 235 25.13 1.55 2669 52.99 0.97 246.0 <0.01
Male 701 74.86 1.55 2368 47.00 0.97
Household incomeb
Low 176 18.78 2.18 1132 22.47 0.63 15.4 <0.01
Low-average 237 25.31 1.21 1388 27.56 0.53
High-average 230 24.57 1.68 1225 24.32 0.75
High 293 31.34 1.91 1292 25.64 0.71
Employment status
Homemaker 44 4.66 0.97 668 13.27 0.44 64.1 <0.01
Working/student 674 71.97 1.90 3274 64.99 0.84
Retired 60 6.37 0.83 412 8.19 0.39
Unemployment/otherc 159 17.00 1.74 682 13.54 0.73
Education (years)
Low (≤4) 181 19.29 1.64 1111 22.05 0.73 9.4 0.02
Low-average (5–8) 250 26.69 1.60 1189 23.61 0.68
High-average (9–11) 324 34.70 2.51 1874 37.20 0.94
High (≥12) 181 19.32 1.95 863 17.13 0.74
Marital status
Married/cohabiting 568 60.72 1.45 2971 58.97 1.00 1.5 0.45
Previously married 147 15.71 1.63 780 15.49 0.57
Never married 221 23.56 1.73 1286 25.53 0.99
Open in a new tab
a

Estimated value of n since after de imputation of weights/clusters/strata it is not possible to calculate the exact value of n.

b

Household income: low = up to 3918 US dollars/year; low-average = from 3919 to 7050 US dollars/year; high-average = from 7051 to 14,826 US dollars/year; high = more than 14,826 US dollars/year.

c

Other = non-regular/informal employment.

Table 2 presents the comparison between fit statistics according to the number of latent classes based upon the DSM-5 AUD criteria among past-year alcohol users who drank alcohol at least once a week. The three-class model had the lowest BIC value. However, the lowest value of SSABIC was found in the five-class model. The AIC and LL values were better as the number of factors increased but the opposite was true for the p-value of Vuong–Mendel–Rubin and degrees of freedom. The higher entropy values were found in the two-class and four-class models, but all the models had an acceptable value (>0.8). Considering the value of BIC as the most reliable measure (over SSABIC), and the opposite trend of the other variables (AIC/LL versus p/df), we selected the three-class model as the most parsimonious one for the DSM-5 AUD. In addition, the three-class model had acceptable values in all other statistical indices.

Table 2.

Comparison between fit statistics according to the number of LCA with the DSM-5 AUD criteria among alcohol users who had at least 1 drink/week in the last year in São Paulo, 2005–2007.

DSM-5 (AUD – alcohol use disorder) criteria
Model LL AIC BIC SSABIC p-Vuong Entropy df
One-class −3942.706 7907.411 7960.669 7925.734 N.A. N.A. 1968
Two-class −2943.857 5933.714 6045.071 5972.025 0.000 0.913 2015
Three-class −2795.417 5660.834 5830.290 5719.133 0.358 0.896 2003
Four-class −2765.747 5625.495 5853.051 5703.783 1.000 0.914 1991
Five-class −2735.091 5588.182 5873.838 5686.458 1.000 0.877 1979
Six-class −2717.583 5577.165 5920.920 5695.430 1.000 0.867 1967
Open in a new tab

LL = log-likelihood; AIC = Akaike Information Criteria; BIC = Bayesian Information Criteria; SSABIC = sample size adjusted BIC; p-Vuong = p value of Vuong–Mendel–Rubin; df = degrees of freedom.

Table 3 presents the weighted probability of DSM-5 AUD criteria per latent class of best model fit for the DSM-5 (three-class model). The three classes were labeled as: “non-symptomatic class” (corresponding to 69.7% of the respondents); “use in larger amounts class” (23.2%); and “high-moderate symptomatic class” (7.1%). Respondents in the “non-symptomatic class” had a very low probability of endorsing all considered criteria. Those in the “use in larger amounts class” had low probability (<30%) of having all DSM-5 AUD criteria, with the exception of a high probability of endorsing the Larger/Longer criterion (85.4%) and a moderate probability of endorsing the Hazardous Use criterion (40.8%). The “high-symptomatic class” included individuals with moderate to high probabilities (60.5–100.0%) of endorsing all DSM-5 AUD criteria. Among all past-year weekly alcohol users, the Larger/Longer criterion had the highest prevalence of endorsement (35.53%), followed by Hazardous Use (16.99%), Neglect Roles (14.10%) and Craving (14.06%) criteria. Fig. 1 displays the probability of endorsement of the eleven DSM-5 AUD criteria per each latent class in the three-class model.

Table 3.

Conditional prevalence DSM-5 AUD criteria endorsement per each latent class, among alcohol users who had at least 1 drinks/week in the past year in São Paulo, 2005–2007.

Criteria “Non-symptomatic class”
“Use in larger amounts class”
“High-moderate symptomatic class”
Total
na % SE na % SE na % SE na %
Craving 5 0.7 0.01 64 29.3 0.05 44 67.1 0.12 121 12.9
Tolerance 6 0.9 0.01 54 25.0 0.03 41 63.3 0.10 110 11.7
Withdrawal 2 0.3 0.01 45 20.7 0.05 42 63.8 0.08 96 10.3
Larger/Longer 94 14.4 0.02 185 85.4 0.03 66 100.0 0.00 358 38.2
Quit/Control 1 0.1 0.01 63 29.1 0.05 51 77.4 0.12 124 13.3
Time Spent 0 0.0 0.00 12 5.4 0.02 53 80.2 0.13 74 7.9
Activities Given Up 1 0.1 0.01 7 3.4 0.02 38 57.3 0.12 52 5.6
Physical/Psychological 0 0.0 0.00 36 16.4 0.05 51 78.0 0.11 96 10.3
Neglect Roles 5 0.8 0.01 73 33.8 0.05 40 60.5 0.09 126 13.5
Social/Interpersonal 9 1.4 0.01 61 28.1 0.06 52 79.2 0.06 131 14.1
Hazardous Use 22 3.4 0.01 89 40.8 0.05 46 69.5 0.08 165 17.7
Total 652 69.7 0.02 217 23.2 0.01 66 7.1 0.01 936 100.0
Open in a new tab
a

Estimated value of n since after de imputation of weights/clusters/strata it is not possible to calculate the exact value of n.

Fig. 1.

Fig. 1

Open in a new tab

DSM-5 alcohol use disorder criteria in a three-class model. Weighted probability of endorsing DSM-5 AUD criteria given latent class among alcohol users who had at least 1 drink/week in the past year in São Paulo, 2005–2007.

Table 4 shows the covariate estimates model with the “non-symptomatic class” as the reference category. Those in the “high-moderate symptomatic class” were more likely to have been previously married (aOR = 2.45, 95%CI = 1.10–5.44, p = 0.027), have lower educational attainment (aOR = 14.35, 95%CI = 2.73–75.41, p = 0.002, aOR = 10.81, 95%CI = 2.01–58.13, p = 0.006, and aOR = 8.51, 95%CI = 2.24–32.34, p = 0.002) and to be unemployed/other (aOR = 8.84, 95%CI = 1.38–56.80, p = 0.021). Those in the “use in larger amounts class” were more likely to have been previously married (aOR = 1.99, 95%CI = 1.17–3.40, p = 0.011) or never married (aOR = 2.26, 95%CI = 1.19–4.30, p = 0.012), and less likely to be in the high-average income category (aOR = 0.39, 95%CI = 0.17–0.87, p = 0.028).

Table 4.

Logisitic regression models results of sociodemographic correlates within latent classes among individuals who had at least 1 drink/week in the past-year in São Paulo, 2005–2007.

Variable “Use in larger amounts class”* “High-moderate symptomatic class”* p
aOR 95%CI p aOR 95%CI
Gender
 Female 1.00 1.00
 Male 1.29 0.69–2.40 0.423 1.81 0.65–5.06 0.255
Household income**
 Low 0.99 0.43–2.29 0.995 1.79 0.56–5.66 0.318
 Low-average 0.92 0.45–1.85 0.776 1.58 0.56–4.41 0.382
 High-average 0.39 0.17–0.87 0.028 0.77 0.20–2.94 0.703
 High 1.00 1.00
Employment status
 Homemaker 1.00 1.00
 Working/student 1.56 0.34–7.13 0.564 2.12 0.34–12.96 0.413
 Retired 0.39 0.04–3.67 0.411 1.28 0.04–41.77 0.886
 Unemployed/other 1.24 0.25–6.15 0.792 8.84 1.38–56.60 0.021
Education (in years)
 Low (0–4) 2.38 0.93–6.08 0.068 14.35 2.73–75.41 0.002
 Low-average (5–8) 2.25 0.97–5.24 0.059 10.81 2.01–58.13 0.006
 High-average (9–11) 1.55 0.77–3.10 0.214 8.51 2.24–32.34 0.002
 High (>11) 1.00 1.00
Marital status
 Married/cohabiting 1.00 1.00
 Previously married 1.99 1.17–3.40 0.011 2.45 1.10–5.44 0.027
 Never married 2.26 1.19–4.30 0.012 0.74 0.20–2.74 0.654
Age (in years)
 18–34 0.93 0.38–2.29 0.881 1.20 0.20–7.02 0.838
 35–54 1.47 0.59–3.69 0.405 0.90 0.17–4.63 0.900
 >54 1.00 1.00
Open in a new tab
*

Non-symptomatic class was the reference category for other latent classes.

**

Household income: low = up to 3918 dollars/year; low-average = from 3919 to 7050 dollars/year; high-average = from 7051 to 14,826 dollars/year; high = more than 14,826 dollars/year. Bold values are significant.

Table 5 presents the proportion of DSM-5 AUD estimated in each of the classes. The majority of the past-year alcohol users who drank at least once a week was within the “non-symptomatic class” and did not meet criteria for DSM-5 AUD. The vast majority of the respondents in the 2 problematic latent classes (92.5% and 100%) met AUD diagnosis. Most of the individuals from the “use in larger amounts class” were diagnosed with Mild (54.8%) or Moderate (28.3%) DSM-5 AUD, while almost all individuals from the “high-moderate symptomatic class” were diagnosed as having Severe DSM-5 AUD (96.7%).

Table 5.

Prevalence DSM-5 diagnosis per four latent classes of alcohol users who had at least 1 drink/week in the past year in São Paulo, 2005–2007.

Latent class Sample %(SE) DSM-5 AUD %(SE) Mild DSM-5 AUD %(SE) Moderate DSM-5 AUD %(SE) Severe DSM-5 AUD %(SE)
“Non-symptomatic class” 69.7 (0.02) 1.1 (0.01) 1.1 (0.01) 0.0 (0.00) 0.0 (0.00)
“Use in larger amounts class” 23.2 (0.01) 92.5 (0.02) 54.8 (0.03) 28.3 (0.03) 9.4 (0.02)
“High-moderate symptomatic class” 7.1 (0.01) 100.0 (0.00) 0.0 (0.01) 3.3 (0.03) 96.7 (0.03)
Open in a new tab

SE = linearized standard error.

4. Discussion

This is the first LCA study in past-year alcohol users who drank at least once a week from a representative sample of the general population using DSM-5 AUD criteria. Our findings indicate that within the new proposed DSM-5 AUD there are three different latent classes of past-year alcohol users who drank at least one a week (the best LCA solution): the vast majority of these alcohol users were within the “non-symptomatic class” (nearly 70%) with less than 1% of respondents in this class receiving a DSM-5 AUD diagnosis. In contrast, the two symptomatic classes well represented the dimensionality of the new proposed AUD criteria. The “use in larger amounts class” reflected the less severe level of this disorder, with more than 90% being captured by DSM-5 AUD (representing Mild and Moderate DSM-5 AUD). At the upper end of the spectrum is the “high-moderate symptomatic class”, representing Severe DSM-5 AUD. The difference between these two classes was regarding the probability (low versus moderate/high, respectively) of endorsement of all criteria other than Larger/Longer (high in both classes).

The similarities among our latent classes obtained with DSM-5 criteria, and the latent classes obtained using DSM-IV criteria from previous studies (Ko et al., 2010; McBride et al., 2011), support the existence of three or four latent classes of alcohol users: a non-symptomatic class, a high-moderate symptomatic class, and an intermediate class or classes with high probability of endorsing Larger/Longer criterion. One recent study that investigated DSM-IV abuse and dependence criteria together in a representative sample of individuals who drank on ≥3 days a week and/or have usually consumed ≥3 drinks on the days they were drinking during lifetime of the population of Australia (McBride et al., 2011), found the model with three phenotypic classes was the most parsimonious: (1) a non-symptomatic class (68.1%), (2) a moderate-high symptomatic class (4.8%) and (3) a class (27.1%) with a high probability of endorsing Larger/Longer criterion, and with a moderate probability of endorsing the Hazardous Use criterion, which resembled a lot this study’s findings.

It is important to discuss some previous problematic issues (Martin et al., 2008) with the criterion Larger/Longer criterion, as this criterion was very important for the definition of latent classes in the present study with DSM-5 criteria, and in the previous LCA studies with DSM-IV criteria (Ko et al., 2010; McBride et al., 2011). Moss et al. (2007) state that Larger/Longer criterion assignments could be false among those who do not report a compulsive pattern of substance use. Future LCA studies that take into account the Moss et al. (2007) suggestions of defining whether using more or longer durations of alcohol due to compulsive use and not merely due to social reasons, could possibly generate different latent classes, which could represent better the different phenotypes of alcohol users. Item response theory (IRT) analysis could also generate interesting findings to further advance our knowledge on the severity and discrimination of Larger/Longer and the other DSM-5 AUD criteria (McBride et al., 2011).

Only one LCA study conducted in U.S. (Ko et al., 2010) investigated sociodemographic aspects of different latent classes in a representative sample of past-year alcohol users. Despite similarities regarding the latent classes found in Ko et al.’ study and this study, the sociodemographic correlates were quite different. The only significant sociodemographic differences in Ko et al.’s study (2010) were: moderate symptomatic and high symptomatic classes tended to be comprised of younger individuals and those with lower income as compared to the other two classes. Adults in Brazil and United States have enormous differences regarding levels of educational attainment (Maia and Sakamoto, 2012), which possibly explains the differences between findings regarding educational attainment in our study and in Ko et al.’s (2010) study. Only 19.1% of adult population in Brazil aged 16 and older have entered college compared to 62.0% in the U.S. (Maia and Sakamoto, 2012).

Lower educational attainment and unemployment status were associated with the “high-moderate symptomatic class”, which seems to be in line with recent investigations in Brazilian samples (Silveira et al., 2011; Castro et al., 2012). A previous report from our research group using the same sample (Silveira et al., 2011), shows that the three lowest levels of education (up to a maximum of 11 years of formal education) predicted the transition from regular alcohol use to abuse. Another study in Brazil (Castro et al., 2012) found similar results: individuals with lower education levels were twice more likely to engage in binge drinking than those at the higher level of education. These associations suggest that in our country there is an association between lower educational attainment with problematic alcohol use (i.e., alcohol abuse, binge drinking and Severe DSM-5 AUD). The association between unemployment and problematic drinking behaviors has been described in different cultures (Caetano et al., 2012; Backhans et al., 2012; Popovici and French, 2013) Despite the fact that there are no specific studies with Brazilian samples, this association seems to be especially important for Latin American male immigrants living in U.S. (Caetano et al., 2012). Our finding that individuals with high-average household income are less likely to be in the “use in larger amounts class” as compared to individuals with higher household income is in line with previous studies in Brazil showing that higher income has a protective effect for heavy drinking, and alcohol abuse or dependence (Moreira et al., 1996; de Lima et al., 2003; Costa et al., 2004; Silveira et al., 2007).

Previously married respondents were more likely to be in both problematic classes (“use in larger amounts” and “high-moderate symptomatic”) than those currently married. This is not surprising since prior studies from different countries (Leonard and Eiden, 2007; Silveira et al., 2007; Dawson et al., 2008; Scott et al., 2009; Boden et al., 2013), a well established two-way relationship has been described, in which drinking in larger amounts affects marital quality and stability, as well as marriage disruption increases alcohol consumption, heavy drinking and related problems.

Remarkably, we did not find any association between gender and the two problematic alcohol classes. Gender effects were expected since male gender has been extensively associated with all patterns of alcohol misuse in Brazil (Silveira et al., 2011, 2012; Castro et al., 2012). Perhaps these differences could be explained by the fact that none of these previous studies investigated gender differences in DSM-5 AUD criteria. In the specific case of gender, the absence of differences may be due to a recent trend of increased alcohol-related problems in women worldwide and in Brazil (Bloomfield et al., 2006; White et al., 2011; Silveira et al., 2012). According to the GENACIS (Gender, Alcohol and Culture: An International Study) project, as women’s position in society has moved toward higher gender equality, differences in consequences of alcohol use between women and men have decreased (Bloomfield et al., 2006).

The identified latent classes and their sociodemographic correlates could guide public policy and clarify the patterns of alcohol use in our country. Considering that the “use in larger amounts” class is the largest problematic class in terms of number of individuals, and that recent Brazilian studies have shown that among alcohol drinkers, a large proportion drink in a heavy pattern (Lima et al., 2007; Guimarães et al., 2010; Laranjeira et al., 2010), the use of alcohol in larger amounts calls for special attention in our country. Prevention strategies for these individuals with low distress related to alcohol use must be the most effective approach to modify this drinking culture in Brazil. In a recent review, Strang et al. (2012) suggest brief interventions to selected individuals at higher risk for drug problems, which seems to fit well for this latent class. On the other hand, individuals in the “high-moderate symptomatic class” deserve specialized treatment in addiction care units with a multidisciplinary team, in view of their high clinically significant impairment caused by alcohol use, with several behavioral and physical symptoms. In these individuals, major areas of life functioning are likely to be impaired (work or school, interpersonal relationships, communication, and health).

This study is not without limitations. The sample is restricted to residents of a large urban area, which precludes the generalization of our findings to the general population who live in rural settings. In addition, lay interviewers may not have the ability to make a refined assessment of alcohol related problems, although they received a five-day standardized training. Furthermore, possible recall bias may have occurred. Another limitation refers to the number (eleven) of disaggregated individual criterion items (which could have occurred in different time-points in a respondents lifetime), which were considered in the latent class model (as analyses with numerous variables increase random error and provide less parsimonious and unreliable latent variables). Finally, the cross-sectional design cannot establish a cause-effect relationship for the observed data.

In summary, this study demonstrates that within the new diagnosis of AUD proposed for DSM-5, LCA distinguishes three distinct categorical phenotypes of Brazilian past-year weekly alcohol users. This new criteria has the advantage of shedding light on risky drinkers included in the “use in larger amounts class,” allowing for preventive and brief interventions, which will reach a large number of individuals. In the light of some problematic issues with Larger/Longer, future LCA studies with alcohol users may distinguish compulsive patterns of this criterion. IRT studies should also further investigate the performance of this and the other ten criteria.

Supplementary Material

appendix1
appendix2

Acknowledgments

The São Paulo Megacity Mental Health Survey is carried out in conjunction with the World Health Organization World Mental Health (WMH) Survey Initiative. The authors thank the WMH staff for assistance with instrumentation, fieldwork, and data analysis. The main coordination center activities, at Harvard University, were supported by the United States National Institutes of Mental Health (R01MH070884), the John D. and Catherine T. MacArthur Foundation, the Pfizer Foundation, the US Public Health Service (R13-MH066849, R01-MH069864, and R01 DA016558), the Fogarty International Center (FIRCA R03-TW006481), the Pan American Health Organization, the Eli Lilly and Company Foundation, Ortho-McNeil Pharmaceutical, Inc., GlaxoSmithKline, Bristol-Myers Squibb, and Shire. The authors declare that the funders of the SPMHS had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors also declare that the commercial funders of the Harvard coordination center had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Role of funding source

The São Paulo Megacity Mental Health Survey was funded by the State of São Paulo Research Foundation, Brazil (FAPESP Grant 03/00204-3, http://www.fapesp.br/materia/176/projeto-tematico/projeto-tematico.htm). Instrument development was supported by the Foundation for Science and Technology of Vitoria, Espírito Santo, Brazil (Fundo de Apoio à Ciência e Tecnologia do Município de Vitória – FACITEC 002/2003). Data analysis was supported by FAPESP Grant 2011/50517-4. Dr. Martins is currently supported by NIH-NICHD 1R01HD060072.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.drugalcdep.2013.12.012.

Footnotes

✩

Supplementary material for this article can be found by accessing the online version of this paper. Please see Appendix A for more information.

2

Supplementary material can be found by accessing the online version of this paper at http://dx.doi.org/10.1016/j.drugalcdep.2013.12.012 and by entering.

3

Supplementary material can be found by accessing the online version of this paper at http://dx.doi.org/10.1016/j.drugalcdep.2013.12.012 and by entering.

Contributors

Authors CMS, ERS, Y-PW, AGA, SSM and LHA designed the study. MCV and LHA wrote protocol, and created the databank. Authors JMC-M, CMS, GB, LHA, and SSM managed the literature searches and summaries of previous related work. JMC-M, IAM, CA-S, and SSM undertook the statistical analysis. Author JMC-M, CMS, LHA and SSM wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.

Conflict of interest

None.

References

  1. Andrade LH, Wang YP, Andreoni S, Silveira CM, Alexandrino-Silva C, Siu ER, Nishimura R, Anthony JC, Gattaz WF, Kessler RC, Viana MC. Mental disorders in megacities: findings from the São Paulo Megacity Mental Health Survey, Brazil. Plos One. 2012;7:e31879. doi: 10.1371/journal.pone.0031879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Asparouhov T. Wald test of mean equality for potential latent class predictors in mixture modeling. 2007 http://www.statmodel.com/download/MeanTest1.pdf (accessed 13.12.13)
  3. Babor TF, Caetano R. The trouble with alcohol abuse: what are we trying to measure, diagnose, count and prevent? Addiction. 2008;103:1057–1059. doi: 10.1111/j.1360-0443.2008.02263.x. [DOI] [PubMed] [Google Scholar]
  4. Backhans MC, Lundin A, Hemmingsson T. Binge drinking – a predictor for or a consequence of unemployment? Alcohol Clin Exp Res. 2012;36:1983–1990. doi: 10.1111/j.1530-0277.2012.01822.x. [DOI] [PubMed] [Google Scholar]
  5. Bloomfield K, Gmel G, Wilsnack S. Introduction to special issue ‘gender, culture and alcohol problems: a multi-national study’. Alcohol Alcohol. 2006;41S:i3–i7. doi: 10.1093/alcalc/agl070. [DOI] [PubMed] [Google Scholar]
  6. Boden JM, Fergusson DM, Horwood LJ. Alcohol misuse and relationship breakdown: findings from a longitudinal birth cohort. Drug Alcohol Depend. 2013;133:115–120. doi: 10.1016/j.drugalcdep.2013.05.023. [DOI] [PubMed] [Google Scholar]
  7. Borges G, Ye Y, Bond J, Cherpitel CJ, Cremonte M, Moskalewicz J, Swiatkiewicz G, Rubio-Stipec M. The dimensionality of alcohol use disorders and alcohol consumption in a cross national perspective. Addiction. 2010;105:240–254. doi: 10.1111/j.1360-0443.2009.02778.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Caetano R, Vaeth PA, Rodriguez LA. The Hispanic Americans Baseline Alcohol Survey (HABLAS): acculturation, birthplace and alcohol-related social problems across Hispanic national groups. Hisp J Behav Sci. 2012;31:95–117. doi: 10.1177/0739986311424040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Caetano R, Baruah J, Chartier KG. Ten-year trends (1992 to 2002) in sociodemographic predictors and indicators of alcohol abuse and dependence among whites, blacks, and Hispanics in the United States. Alcohol Clin Exp Res. 2011;35:1458–1466. doi: 10.1111/j.1530-0277.2011.01482.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Casey M, Adamson G, Shevlin M, McKinney A. The role of craving in AUDs: dimensionality and Differential Functioning in the DSM-5. Drug Alcohol Depend. 2012;125:75–80. doi: 10.1016/j.drugalcdep.2012.03.019. [DOI] [PubMed] [Google Scholar]
  11. Castro DS, Sanchez ZM, Zaleski M, Alves HN, Pinsky I, Caetano R, Laranjeira RR. Sociodemographic characteristics associated with binge drinking among Brazilians. Drug Alcohol Depend. 2012;126:272–276. doi: 10.1016/j.drugalcdep.2012.05.017. [DOI] [PubMed] [Google Scholar]
  12. Cherpitel CJ, Borges G, Ye Y, Bond J, Cremonte M, Moskalewicz J, Swiatkiewicz G. Performance of a craving criterion in DSM alcohol use disorders. J Stud Alcohol Drugs. 2010;71:674–684. doi: 10.15288/jsad.2010.71.674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Costa JS, Silveira MF, Gazalle FK, Oliveira SS, Hallal PC, Menezes AB, Gigante DP, Olinto MT, Macedo S. Heavy alcohol consumption and associated factors: a population-based study. Rev Saude Publica. 2004;38:284–291. doi: 10.1590/s0034-89102004000200019. [DOI] [PubMed] [Google Scholar]
  14. Dawson DA, Goldstein RB, Chou SP, Ruan WJ, Grant BF. Age at first drink and the first incidence of adult-onset DSM-IV alcohol use disorders. Alcohol Clin Exp Res. 2008;32:2149–2160. doi: 10.1111/j.1530-0277.2008.00806.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Degenhardt L, Bohnert KM, Anthony JC. Assessment of cocaine and other drug dependence in the general population: “gated” versus “ungated” approaches. Drug Alcohol Depend. 2008;93:227–232. doi: 10.1016/j.drugalcdep.2007.09.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Degenhardt L, Lynskey M, Coffey C, Patton G. Diagnostic orphans among young adult cannabis users: persons who report dependence symptoms but do not meet diagnostic criteria. Drug Alcohol Depend. 2002;67:205–212. doi: 10.1016/s0376-8716(02)00064-9. [DOI] [PubMed] [Google Scholar]
  17. de Lima MS, Dunn J, Novo IP, Tomasi E, Reisser AA. Gender differences in the use of alcohol and psychotropics in a Brazilian population. Subst Use Misuse. 2003;38:51–65. doi: 10.1081/ja-120016565. [DOI] [PubMed] [Google Scholar]
  18. Delucchi KL, Matzger H, Weisner C. Dependent and problem drinking over 5 years: a latent class growth analysis. Drug Alcohol Depend. 2004;74:235–244. doi: 10.1016/j.drugalcdep.2003.12.014. [DOI] [PubMed] [Google Scholar]
  19. Diagnostic and Statistical Manual of Mental Disorders. 5th. American Psychiatric Association; Arlington, VA: 2013. [Google Scholar]
  20. Grant BF, Stinson FS, Harford TC. Age at onset of alcohol use and DSM-IV alcohol abuse and dependence: a 12-year follow-up. J Subst Abuse. 2001;13:493–504. doi: 10.1016/s0899-3289(01)00096-7. [DOI] [PubMed] [Google Scholar]
  21. Guimarães VV, Florindo AA, Stopa AR, César CLG, Barros MBA, Carandina L, Goldbaum M. Consumo abusivo e dependência de álcool em população adulta no estado de São Paulo, Brasil. Rev Bras Epidemiol. 2010;13:314–325. [Google Scholar]
  22. Hagman BT, Cohn AM. Toward DSM-5: mapping the alcohol use disorder continuum in college students. Drug Alcohol Depend. 2011;118:202–208. doi: 10.1016/j.drugalcdep.2011.03.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Haro JM, Arbabzadeh-Bouchez S, Brugha TS, de Girolamo G, Guyer ME, Jin R, Lepine JP, Mazzi F, Reneses B, Vilagut G, Sampson NA, Kessler RC. Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health surveys. Int J Methods Psychiatr Res. 2006;15:167–180. doi: 10.1002/mpr.196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hasin DS, Beseler CL. Dimensionality of lifetime alcohol abuse, dependence and binge drinking. Drug Alcohol Depend. 2009;101:53–61. doi: 10.1016/j.drugalcdep.2008.10.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hasin D, Paykin A. Dependence symptoms but no diagnosis: diagnostic ‘orphans’ in a community sample. Drug Alcohol Depend. 1998;50:19–26. doi: 10.1016/s0376-8716(98)00007-6. [DOI] [PubMed] [Google Scholar]
  26. Hedden SL, Martins SS, Malcolm RJ, Floyd L, Cavanaugh CE, Latimer WW. Patterns of illegal drug use among an adult alcohol dependent population: results from the National Survey on Drug Use and Health. Drug Alcohol Depend. 2010;106:119–125. doi: 10.1016/j.drugalcdep.2009.08.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kendler KS, Ohlsson H, Sundquiat K, Sundquiast J. A latent class analysis of drug abuse in a national Swedish sample. Psychol Med. 2013;43:2169–2178. doi: 10.1017/S0033291713000081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Keyes KM, Hasin DS. Socio-economic status and problem alcohol use: the positive relationship between income and the DSM-IV alcohol abuse diagnosis. Addiction. 2008;103:1120–1130. doi: 10.1111/j.1360-0443.2008.02218.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Keyes KM, Krueger RF, Grant BF, Hasin DS. Alcohol craving and the dimensionality of alcohol disorders. Psychol Med. 2011;41:629–640. doi: 10.1017/S003329171000053X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ko JY, Martins SS, Kuramoto SJ, Chilcoat HD. Patterns of alcohol-dependence symptoms using a latent empirical approach: associations with treatment usage and other correlates. J Stud Alcohol Drugs. 2010;71:870–878. doi: 10.15288/jsad.2010.71.870. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Laranjeira R, Pinsky I, Sanches M, Zaleski M, Caetano R. Alcohol use patterns among Brazilian adults. Rev Bras Psiquiatr. 2010;32:231–241. doi: 10.1590/s1516-44462009005000012. [DOI] [PubMed] [Google Scholar]
  32. Leonard KE, Eiden RD. Marital and family processes in the context of alcohol use and alcohol disorders. Annu Rev Clin Psychol. 2007;3:285–310. doi: 10.1146/annurev.clinpsy.3.022806.091424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lima MCP, Kerr-Corrêa F, Tucci AM, Simão MO, Oliveira JB, Cavariani MB. Gender difference in heavy alcohol use: a general population survey (The Genacis Project) of São Paulo City, Brazil. Contemp Drug Probl. 2007;34:427–444. [Google Scholar]
  34. Lynskey MT, Agrawal A. Psychometric properties of DSM assessment of illicit drug abuse and dependence: results from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) Psychol Med. 2007;37:1345–1355. doi: 10.1017/S0033291707000396. [DOI] [PubMed] [Google Scholar]
  35. Maia AG, Sakamoto A. Occupational structure and socioeconomic inequality: a comparative study between Brazil and the United States. 16th World Congress of the International Labor and Employment Relations Association (ILERA); Philadelphia, USA. 2012. [Google Scholar]
  36. Martin CS, Chung T, Langenbucher JW. How should we revise diagnostic criteria for substance use disorders in DSM-5? J Abnorm Psychol. 2008;117:561–575. doi: 10.1037/0021-843X.117.3.561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. McBride O, Adamson G, Bunting BP, McCann S. Characteristics of DSM-IV alcohol diagnostic orphans: drinking patterns, physical illness, and negative life events. Drug Alcohol Depend. 2009;99:272–279. doi: 10.1016/j.drugalcdep.2008.08.008. [DOI] [PubMed] [Google Scholar]
  38. McBride O, Teesson M, Baillie A, Slade T. Assessing the dimensionality of lifetime DSM-IV alcohol use disorders and a quantity-frequency alcohol use criterion in the Australian population: a factor mixture modelling approach. Alcohol Alcohol. 2011;46:333–341. doi: 10.1093/alcalc/agr008. [DOI] [PubMed] [Google Scholar]
  39. Mewton L, Slade T, McBride O, Grove R, Teesson M. An evaluation of the proposed DSM-5 alcohol use disorder criteria using Australian national data. Addiction. 2011;106:941–950. doi: 10.1111/j.1360-0443.2010.03340.x. [DOI] [PubMed] [Google Scholar]
  40. Moreira LB, Fuchs FD, Moraes RS, Bredemeier M, Cardozo S, Fuchs SC, Victora CG. Alcoholic beverage consumption and associated factors in Porto Alegre, a southern Brazilian city: a population-based survey. J Stud Alcohol. 1996;57:253–259. doi: 10.15288/jsa.1996.57.253. [DOI] [PubMed] [Google Scholar]
  41. Moss HB, Chen CM, Yi HY. Subtypes of alcohol dependence in a nationally representative sample. Drug Alcohol Depend. 2007;91:149–158. doi: 10.1016/j.drugalcdep.2007.05.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Muthén BO. Should substance use disorders be considered as categorical or dimensional? Addiction. 2006;101(Suppl 1):S6–S16. doi: 10.1111/j.1360-0443.2006.01583.x. [DOI] [PubMed] [Google Scholar]
  43. Muthén BO, Asparouhov T. Item response mixture modeling: application to tobacco dependence criteria. Addict Behav. 2006;31:1050–1066. doi: 10.1016/j.addbeh.2006.03.026. [DOI] [PubMed] [Google Scholar]
  44. Muthén LK, Muthén BO. Mplus User’s Guide 6. Muthén & Muthén; Los Angeles, CA: 2010. [Google Scholar]
  45. Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct Equ Model. 2007;14:535–569. [Google Scholar]
  46. O’Brien C. Addiction and dependence in DSM-5. Addiction. 2010;106:866–867. doi: 10.1111/j.1360-0443.2010.03144.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Peer K, Rennert L, Lynch KG, Farrer L, Gelernter J, Kranzler HR. Prevalence of DSM-IV and DSM-5 alcohol, cocaine, opioid, and cannabis use disorders in a largely substance dependent sample. Drug Alcohol Depend. 2013;127:215–219. doi: 10.1016/j.drugalcdep.2012.07.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Popovici I, French MT. Does unemployment lead to greater alcohol consumption? Ind Relat (Berkeley) 2013;52:444–466. doi: 10.1111/irel.12019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Rehm J, Rehn N, Room R, Monteiro M, Gmel G, Jernigan D, Frick U. The global distribution of average volume of alcohol consumption and patterns of drinking. Eur Addict Res. 2003;9:147–156. doi: 10.1159/000072221. [DOI] [PubMed] [Google Scholar]
  50. Saha TD, Chou SP, Grant BF. Toward an alcohol use disorder continuum using item response theory: results from the National Epidemiologic Survey on alcohol and related conditions. Psychol Med. 2006;36:931–941. doi: 10.1017/S003329170600746X. [DOI] [PubMed] [Google Scholar]
  51. Schuckit M, Smith TL, Danko GP, Bucholz KK, Reich T, Bierut L. Five-year clinical course associated with DSM-IV alcohol abuse or dependence in a large group of men and women. Am J Psychiatry. 2001;158:1084–1090. doi: 10.1176/appi.ajp.158.7.1084. [DOI] [PubMed] [Google Scholar]
  52. Schuckit MA, Danko GP, Smith TL, Bierut LJ, Bucholz KK, Edenberg HJ, Hesselbrock V, Kramer J, Nurnberger JI, Jr, Trim R, Allen R, Kreikebaum S, Hinga B. The prognostic implications of DSM-IV abuse criteria in drinking adolescents. Drug Alcohol Depend. 2008;97:94–104. doi: 10.1016/j.drugalcdep.2008.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Scott KM, Wells JE, Angermeyer M, Brugha TS, Bromet E, Demyttenaere K, de Girolamo G, Gureje O, Haro JM, Jin R, Karam AN, Kovess V, Lara C, Levinson D, Ormel J, Posada-Villa J, Sampson N, Takeshima T, Zhang M, Kessler RC. Gender and the relationship between marital status and first onset of mood, anxiety and substance use disorders. Psychol Med. 2009;26:1–11. doi: 10.1017/S0033291709991942. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Shmulewitz D, Keyes K, Beseler C, Aharonovich E, Aivadyan C, Spivak B, Hasin D. The dimensionality of alcohol use disorders: results from Israel. Drug Alcohol Depend. 2010;111:146–154. doi: 10.1016/j.drugalcdep.2010.04.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Silveira CM, Siu ER, Wang YP, Viana MC, Andrade AG, Andrade LH. Gender differences in drinking patterns and alcohol-related problems in a community sample in São Paulo, Brazil. Clinics. 2012;67:205–212. doi: 10.6061/clinics/2012(03)01. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Silveira CM, Viana MC, Siu ER, de Andrade AG, Anthony JC, Andrade LH. Sociodemographic correlates of transitions from alcohol use to disorders and remission in the São Paulo Megacity Mental Health Survey, Brazil. Alcohol Alcohol. 2011;46:324–332. doi: 10.1093/alcalc/agr007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Silveira CM, Wang YP, Andrade AG, Andrade LH. Heavy episodic drinking in the São Paulo epidemiologic catchment area study in Brazil: gender and sociodemographic correlates. J Stud Alcohol Drugs. 2007;68:18–27. doi: 10.15288/jsad.2007.68.18. [DOI] [PubMed] [Google Scholar]
  58. Strang J, Babor T, Caulkins J, Fischer B, Foxcroft D, Humphreys K. Drug policy and the public good: evidence for effective interventions. Lancet. 2012;379:71–83. doi: 10.1016/S0140-6736(11)61674-7. [DOI] [PubMed] [Google Scholar]
  59. The World Bank. The World Bank working for a world free of poverty website. 2013 http://data.worldbank.org/country/brazil (accessed 11.03.13)
  60. Viana MC, Andrade LH. Lifetime prevalence, age and gender distribution and age-of-onset of psychiatric disorders in the São Paulo Metropolitan Area, Brazil: results from the São Paulo Megacity Mental Health Survey. Rev Bras Psiquiatr. 2012;34:14–25. doi: 10.1016/j.rbp.2012.03.001. [DOI] [PubMed] [Google Scholar]
  61. Viana MC, Teixeira MG, Beraldi F, Bassani IS, Andrade LH. Sao Paulo Megacity Mental Health Survey – a population-based epidemiological study of psychiatric morbidity in the Sao Paulo metropolitan area: aims, design and field implementation. Rev Bras Psiquiatr. 2009;31:375–386. doi: 10.1590/s1516-44462009000400016. [DOI] [PubMed] [Google Scholar]
  62. White AM, Hingson RW, Pan IJ, Yi HY. Hospitalizations for alcohol and drug overdoses in young adults ages 18–24 in the United States, 1999–2008: results from the Nationwide Inpatient Sample. J Stud Alcohol Drugs. 2011;72:774–786. doi: 10.15288/jsad.2011.72.774. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

appendix1
NIHMS692695-supplement-appendix1.xlsx (33.7KB, xlsx)
appendix2
NIHMS692695-supplement-appendix2.xlsx (9.9KB, xlsx)

RESOURCES