High mortality, violence and crime in alcohol dependents: 5 years after seeking treatment in a Brazilian underprivileged suburban community

Revista Brasileira de Psiquiatria

Alta mortalidade, violência e crime em dependentes de álcool: seguimento após 5 anos de tratamento em periferia brasileira

Jairo Valentim Silva; Viviane de Castro; Ronaldo Laranjeira; Neliana Buzi Figlie

Introduction

Alcoholism is a compelling disorder that raises important public health concerns: 3.8% of all global deaths were attributable to alcohol use, of which 6.2% were accounted for the deaths of males and 1.1% for the deaths of females.¹ Alcohol consumption played a role in the majority of deaths involving injuries and illness such as cancer, cardiovascular disorders and liver cirrhosis.¹ Intentional and unintentional injuries accounted for 42% of all deaths involving alcohol consumption.¹ Moreover, a recent study confirmed a positive relation between years of life and mental health disorders. According to Schmidt et al.,² 19% of general mental illness including alcohol dependence, depression and psychosis are correlated to lower quality and years of life.

Print version ISSN 1516-4446

Rev. Bras. Psiquiatr. vol.34 no.2 São Paulo June 2012

http://dx.doi.org/10.1590/S1516-44462012000200004 ORIGINAL ARTICLE

High mortality, violence and crime in alcohol dependents: 5 years after seeking treatment in a Brazilian underprivileged suburban community

Alta mortalidade, violência e crime em dependentes de álcool: seguimento após 5 anos de tratamento em periferia brasileira

Jairo Valentim Silva^I; Viviane de Castro^II; Ronaldo Laranjeira^III; Neliana Buzi Figlie^IV

^IPost-graduate student at Universidade Federal de São Paulo (UNIFESP), Department of Psychiatry, Masters Program, Brazil
^IIAffiliate Research Psychologist of UNIFESP, Department of Psychiatry; Instituto Nacional de Políticas Públicas do Álcool e Outras Drogas (National Institute on Alcohol and Drug Policies – INPAD), Brazil; Assistant Professor of the International Psychology PhD Program, Chicago School of Professional Psychology, USA
^IIIINPAD Director; Full Professor of Psychiatry at Universidade Federal de São Paulo – Department of Psychiatry, Brazil
^IVPost-graduation Tutor Supervisor of UNIFESP, Department of Psychiatry, and INPAD Senior Researcher, Brazil

Corresponding author

---

ABSTRACT

OBJECTIVE: To investigate the results of alcohol-related consequences in an underprivileged area of São Paulo.
METHOD: One hundred and ninety one adult patients who sought alcohol treatment in 2002 were reassessed in 2007 regarding alcohol use and involvement with crime. The interview consisted of demographic questions and questionnaires assessing alcohol dependence and pattern of alcohol use. Risk and protective factors and involvement with crime were further explored.
RESULTS High mortality rate (16.9%, n = 41) was found in this sample and 97.4% were identified as being severe alcohol dependents. The sample consisted of a homogeneous group, average age of 42, 81.9% male, 57.5% black, 52.2% unemployed and 100% of low socioeconomic status. Individuals ageing 35 or younger, not engaged in religious activities and with intense alcohol consumption in the last month had 2.7 times more chance on committing crimes (95% CI = [1.22; 5.93] p = 0.014). Subjects who consumed alcohol in the last month also had a 4.1 greater chance of becoming involved in crime (95% CI = [1.2; 14.24] p = 0.024).
CONCLUSION: Alcohol dependence within an underprivileged community was associated with high rates of crime and mortality. Religious affiliation was negatively associated with delinquent behavior.

Descriptors: alcohol; violence/aggression; religion; outpatient psychiatry; other psychosocial techniques/ treatments.

---

RESUMO

OBJETIVO: Explorar as consequências relacionadas ao uso de álcool na periferia de São Paulo.
MÉTODO: Pacientes que procuraram tratamento para alcoolismo em 2002 foram convidados para reavaliação em 2007 para estudo de seguimento retrospectivo. A entrevista consistiu de questões sociodemográficas e questionários que avaliaram a dependência alcoólica e o padrão do consumo. Pesquisa adicional sobre fatores de risco e de proteção e envolvimento com crime foi contemplada neste estudo.
RESULTADOS: A alta taxa de mortalidade (16,9% n = 41) e a dependência grave de álcool foi confirmada em 97,4% da amostra. O grupo se mostrou homogêneo, média de idade (42 anos), sexo masculino (81,9%), raça negra (57,5%), desempregados (52,2%), proveniente de classe social E e D (100%). Indivíduos com até 35 anos sem afiliação religiosa e tendo consumido muito álcool no ultimo mês apresentaram 2,7 mais chances de cometer crimes (95% IC = [1,22; 5,93] p = 0,014). Indivíduos que consumiram álcool no ultimo mês também tiveram suas chances 4,1 vezes maiores de estarem envolvidos com o crime (95% IC = [1,2; 14,24] p = 0,024).
CONCLUSÃO: O consumo alcoólico na periferia foi associado a altas taxas de criminalidade e mortalidade, e a afiliação religiosa foi associada negativamente com comportamento delinquente.

Descritores: álcool; violência/agressividade; religião; psiquiatria ambulatorial; outras técnicas/outros tratamentos psicossociais.

---

Introduction

Alcoholism is a compelling disorder that raises important public health concerns: 3.8% of all global deaths were attributable to alcohol use, of which 6.2% were accounted for the deaths of males and 1.1% for the deaths of females.¹ Alcohol consumption played a role in the majority of deaths involving injuries and illness such as cancer, cardiovascular disorders and liver cirrhosis.¹ Intentional and unintentional injuries accounted for 42% of all deaths involving alcohol consumption.¹ Moreover, a recent study confirmed a positive relation between years of life and mental health disorders. According to Schmidt et al.,² 19% of general mental illness including alcohol dependence, depression and psychosis are correlated to lower quality and years of life.

Alcohol use is associated with risk behavior which includes: drinking and driving, unprotected sex and multiple substance use. Data indicates that alcohol consumption is detrimental to cognitive behavior and emotional development.³ Social consequences of alcohol abuse impose at least as much of a burden as the alcohol effects on health. Alcohol abuse has a relevant association with violent behaviors.⁴ It contributes to increased crime rate, violence and prevalence of injures related to drinking and driving, thus, impacting society in negative and complex ways.⁴ A 60-year prospective study on alcohol dependents found alcohol use to be a major risk factor for mortality.⁵

In Latin America, alcohol consumption is considered the leading risk factor for mortality, being accountable for 16% of total disease burden.¹ In Brazil, alcohol is one of the most consumed substances. A Brazilian national study reveals that 52% of the population is actively consuming alcohol, 29% drinks 5 or more doses per occasion and 28% binge drink. It is worthwhile to mention the fact that 38% of the population in the category of 5 drinks or more was composed mainly of males. Additionally, 25% reported to have problems related to alcohol consumption, 3% already meet criteria for alcohol abuse and 9% for alcohol dependence.⁶

In unstable economic countries, in particular Brazil, 90% of reported violence cases due to alcohol use involved young adult black males, ranging in age between 15 and 44 years, lacking basic education, and living in underprivileged suburban areas.⁷ Interestingly, general literature confirms that, more than any other demographic group, young adult black males were concerned in alcohol related violence, such as homicides and assaults.⁸ Harmful effects of alcohol misuse severely affect underprivileged communities. In these communities, the wide availability of an unrestricted number of outlets selling liquor significantly increase alcohol consumption and, consequently, violence and crime.⁹

The objective of this retrospective observational study was to investigate alcohol consumption-related consequences within an underprivileged violent suburban area of São Paulo, on an attempt to foster the understanding of risk and protective factors associated with subjects’ involvement with crime.

Methods

Design

This is a retrospective research conducted in 2007 and 2008 with 303 outpatient subjects who sought alcohol dependence treatment at Jardim Ângela’s psychosocial treatment center for alcohol and drug abuse during the year of 2002. Among 303 subjects, 34 did not agree to be part of the study, and 37 were not located. The final sample comprised 232 subjects. All individuals met the criteria for alcohol dependence according to DSM-IV¹⁰ during treatment admission. It is important to mention that patients were evaluated and diagnosed by psychiatrists in charge of the unit. Patients’ medical records such as diagnoses, prognoses and treatment progress were obtained in patients’ charts and made available for the principal investigator of the present study by Jardim Ângela’s treatment center.

Forty one deaths (16.9%) were confirmed by subjects’ families. The prime cause of death was due to liver and pancreas disorders (66% n = 27) followed by accidents and/or violence (e.g. homicide) (34% n = 14). One hundred ninety one participants (63%) responded questionnaires by themselves.

The following are criteria for the inclusion of patients in the current study: (1) fulfill criteria for alcohol dependence according to DSM-IV¹⁰ during treatment admission (patients were assessed and diagnosed by psychiatrists in charge of the unit); (2) to be a patient of Jardim Ângela’s psychosocial treatment center for alcohol dependence during the year of 2002; and (3) to be 18 years old or older. Exclusion criteria comprised the following: (1) diagnosis of dependence on an additional substance other than alcohol; and (2) individuals younger than 18 years old.

Setting

The study was conducted at Jardim Ângela’s specialized treatment center for alcohol and drug abuse. Jardim Ângela is a suburban district located at South São Paulo, Brazil, currently comprising 300,000 inhabitants and encompassing an area of 37.4 Km². It is worth mentioning that besides religious and self-support groups, Jardim Ângela’s population mostly relies on this single public treatment center to assist alcohol and drug dependents. The Jardim Ângela community is a conglomerate of slums located in the southern region of São Paulo. According to Laranjeira et al.⁹ this area comprised the highest alcohol outlet density reported in literature during 2002. In the late 90s, the United Nations (UN) ranked this area as the most violent neighborhood in the world involving 120 homicides per 100 thousand inhabitants.

Measures

The study was conducted during the years of 2007 and 2008. The first contact with patients and/or families was established by telephone calls and/or home visits. After the first contact, a senior psychologist conducted a 60-minute face-to-face interview with subjects at Jardim Angela’s Treatment Center and/or at subjects’ residencies.

The interview was composed of a sociodemographic questionnaire which included a crime assessment, and four other instruments described below:

a. Directory of client outcome measures for addiction treatment programs.¹¹ The instrument was used on a crack/cocaine Brazilian follow-up study in 2006.¹² The current study used the instrument developed for the crack/cocaine follow-up and tailored it to investigate alcohol-related variables such as: age at onset, last intake episode, lifetime use, physical complications and client outcome measures for addiction treatment programs.

b. Alcohol dependence data questionnaire (SADD¹³). In this study a Portuguese version of SAAD was used. Jorge et al.¹⁴ (1985)¹⁴ culturally adapted this instrument for the Brazilian population. The Brazilian version of SADD is a 15-item questionnaire which investigates alcohol dependence, categorizing it in 3 distinct levels: mild, moderate and severe.

c. Pattern of alcohol consumption was documented using the interview schedule developed for the WHO/ISBRA Collaborative Study on State and Trait Markers in Alcoholism.¹⁵ The questions addressed the pattern of alcohol consumption, the quantity and frequency of use in the last 30 days as well as the heaviest alcohol-consumption period of life time.

d. Drinker inventory of consequences (DrInC-2L). Instrument was developed by Project MATCH.¹⁶ It is composed by 50 items which evaluates lifetime problems related to alcohol use. A Portuguese version is available for use in Brazil.¹⁷

Ethical considerations

The current study was approved by the Ethical Institutional Review Board committee 0261\08 of the Universidade Federal de São Paulo. The work done with human subjects reported in this study complies with the guideline policies and principles for experimental procedures of the Helsinki Declaration (2002).¹⁸ Participants were properly advised concerning the study design and signed informed consent forms.

Statistical analysis

Statistical Package for Social Scientists (SPSS) descriptive statistical program for Windows, version 8.0.1 (1997)¹⁹ was used to analyze data. A 5% significance level was adopted for all statistical analysis.

Categorical variables were described as absolute, while relative frequency and numerical variables as mean, minimum and maximum, quartiles and standard deviation. Student t -test for independent samples was used to compare means and Person’s chi-square test was employed to perform comparisons between categorical variables distributions. A predictive investigation using logistic regression model was performed to explore the correlation of sociodemographic and alcohol consumption data on crime and violence. Classification tree statistical analyses were conducted to examine relationships between variables and further analyze the possibility of including variables in the logistic regression model.

Classification tree analysis is a tree-building, non-parametric method based on repeated partitioning of a sample into subgroups. Analyses are performed so that the most significant predictor at each step is used to divide the sample into subgroups. This process continues until no significant statistical differences are acknowledged.²⁰ Exhaustive chi-square automatic interaction detector algorithm was part of the classification tree analysis. Results are presented as classification and/or decisions in order to identify groups expected to experience a given outcome. It is considered a powerful technique that further explores variables relations. Within the context of alcohol-related violence, it may help identify relevant protective factors.

Several studies are conducted to investigate the interrelations between the research variables through Lineal Models as Lineal Regression, Logistics, Analysis of Variances etc. In this study, a classification tree method was adopted to offer objective responses fostering a practical interpretation. It is important to emphasize that these methods demonstrate the relationship between the response variable (categorical or continuous) to other research variables (categorical or numeric explanatory variables). Classification tree methodology present results in different groups of responses related to the variable under investigation.²⁰

Results

The group demonstrated to be homogeneous with similar sociodemographic characteristics. The average age was 42 (SD = 11years). The majority of subjects were males 81.9% (n = 154), black 57.5% (n = 96), unemployed 52.2% (n = 82), with up to 8 years of basic schooling 91,6% (n=175), and the whole sample (N = 191) were of low socioeconomic status, where 79.6% was from class E (n = 152) and 20.4% from class D (n = 39). About 51.8% (n = 99) were married and/or in a committed relationship and 68.6% (n = 129) were involved with religious groups/activities.

Almost the entire sample (97.4%, n = 186) was identified by SADD as being severe alcohol dependent. Subjects who visited the emergency room after treatment discharge accounted for 63.2% (n = 120) of the sample and 31% (n = 54) was re-admitted in inpatient programs. Moreover, 59% (n = 111) was currently attending alcohol dependence treatment, of which 96.4% (n = 107) in outpatient programs (Table 1).

A comparative analysis investigation was conducted considering the sociodemographic variables of subjects who were involved in crime versus subjects who did not commit any sort of transgression through the chi-squared test (χ²). Results show that individuals under the age of 35 (35.4%, n = 17) were more likely to commit crimes and be involved in violence (p = 0.004) when compared to subjects 36 years old or older (15.4% n = 22). On the other hand, individuals engaged in religious activities (84.5% n = 109) reported less involvement in crime (p = 0.017). Worth of mention is the fact that, on average, subjects involved with criminal practices increased alcohol intake at age 24 (SD = 5.2), when compared to those not implicated in crime (p = 0.008).

The initial independent variables involving crime regression entering into the model were: gender, age, marital status, race, SES, employment status, alcohol consumption within the last month as well as 12 months prior, total and sub-scale scores of DrInC-2L scale, treatment, and engagement in religious activities – a total of 20 independent variables. Variables entered a backward regression until reaching significant correlation coefficient (p = 0.05). Hosmer et al.²¹ test for goodness-of-fit was calculated to determine model adequacy. The test indicated the regression model demonstrated a good fit (p = 0.920) (Table 2).

Treatment program subjects entered classification tree analysis. The following independent variables were controlled to conduct analysis on delinquent behavior: gender, age, marital status, race, SES, employment status, alcohol consumption within the last month as well as 12 months prior, total and sub-scale scores of DrInC-2L, treatment, and engagement in religious activities. Figure 1 illustrates the results from classification analysis comprising the classification variable and the cutoff point of each division. Each node encompassed the proportion of participants who were being treated for alcohol and drug-related problems. Root node – or node 0 – was based on whether participants were involved in crime and/or violence after entering alcohol and drug treatment. Thirty nine participants (20.4%) were involved in crime after alcohol and drug treatment. The first node split related to individuals’ religious group participation. Among those who were involved in religious activities, 15.5% (n = 20) were also implicated in crime, and a meaningful 30.6% (n = 19) were not involved in any sort of transgression (p = 0.045). The second node split included participants involved in religious activities and alcohol consumption within the last month. Among those who were participating in religious groups, 84.5% (n = 109) did not use alcohol during the last month (p = 0.023). Final node split explored the association between individuals not involved with religious activities and alcohol consumption within the last month. Among subjects who used alcohol, 44.8% (n = 13) were also involved with crime, versus 18.2% (n = 6) that did not commit any sort of transgression.

Discussion

Current studies pinpoint an elevated mortality rate (16.9%) among alcohol dependents. Alcohol-related crimes, accidents and violence accounted for 32.5% of deaths in this research. Indirect consequences of alcohol dependence have a great impact in society, demanding urgent reform in Brazilian public policies to reduce alcohol-related harm. Preventive interventions focusing on decreasing the number of alcohol outlets, increasing taxation and banishing media advertisement should be considered by governmental public policymakers.^3,22 Greater availability of specialized alcohol and drug treatment centers should also be considered on an attempt to manage alcohol-related hazards in Brazil.

Alcohol-related crimes have a great impact in several countries including developed nations. In 2001, the United States estimated that 40,933 deaths by injury were associated with excessive use of alcohol.²³ Injured patients in emergency rooms presenting blood alcohol concentrations (BAC) of 0.08 g/dL had three times more chances to present aggressive behaviors and experience violent injuries. Previous studies found that the severity of the crime was directly proportionate to the likelihood of alcohol being involved.^24,25 A significant relationship is observed between alcohol use and violent offences.

This study identifies several relevant findings concerning the effects imposed by alcohol consumption in an underprivileged community of a developing country. The correlation of alcohol consumption and crime was an expected outcome underscored in this investigation.

Worthy of mention is the fact that the current study was conducted at a single treatment center for alcohol and drug problems located in Jardim Ângela, São Paulo. It is anticipated that high rates of alcohol-related crime would be predictable in a deprived and violent neighborhood such as Jardim Ângela.

Regarding sociodemographic profile, the sample was similar: young black males coming from low socioeconomical status. It is important to stress that subjects’ alcohol intake was intensified during their yearly 20s. Regression models corroborates with early results presented in this study that individuals aged 35 or younger have a 2.7 times greater chance of committing crimes and getting involved with violence.

Studies on young African American populations reported alcohol use as a main contributor of three leading causes of death in United States: homicide, unintentional injuries (which include car crashes) and suicide.²⁶ Racial identity was considered a strong predictor of substance seeking behavior among African Americans, which explains the behavior being present in 20% of this population.²⁷ According to Mäkelä,²⁸ the poorest people in society are often the most affected by alcohol-related deaths. Moreover, alcohol abuse among blacks was strongly associated with poverty, violence and mortality.⁷

Severe alcohol dependence is associated with a range of health, social and legal problems.²⁹ Alcohol dependence in its most severe form leads to social disintegration. This in turn affects work and relationships. In this investigation, almost the entire sample (97.4%) was identified as severe alcohol dependent and about 52.2% was unemployed.

Religious group adherence appears to be an important alcohol abuse protector factor, thus preventing delinquent behavior. Classification tree analyses demonstrated that non-religious individuals were more likely to be involved in criminal activities. It seems that crime involvement is inversely related to religious commitment. An epidemiological study with a qualitative focus concluded that religious engagement is significantly associated with a decline in alcohol consumption and increased successful recovery rates.³⁰

According to Sullivan³¹ people who report no religious preference have higher levels of mortality. The mortality rates can be associated with religious affiliation and socioeconomic status (particularly with years of education and household income), involvement with religious services, or the practice of unhealthy behaviors (e.g. cigarette smoking and alcohol consumption).³¹ Several studies suggest a positive relation between religious affiliation and increased levels of life satisfaction, hope, optimism, marriage stability, and decreased rates of anxiety, depression and substance abuse.³² Individuals involved in religious activities tend to avoid use or abuse of alcohol, tobacco, drugs, and are generally not implicated in crime or delinquent behavior.³³

In the current study, regression models confirmed that individuals not engaged in religious activities and with intensive alcohol consumption within the last 30 days had a 2.7 times greater chance to be implicated in felonies. Subjects also had a 4.1 times greater chance of being involved in crime than those who did not consume alcohol within the last month.

Participation in religious activities may enhance self-control, build coping strategies and promote self-discipline. Religious communities could be considered an important source of social control which improves individuals’ health bonds with society and discourages transgressive behavior. A Brazilian National survey confirmed that religious affiliation did not have a direct relation with black racial identity or the fact of coming from low socioeconomic strata.³⁴ Rather, the involvement in religious practice was pursued in its majority by females and older people. Results from this current study supports data presented by a Brazilian national survey, since young black males that came from low socioeconomic strata and were not affiliated to religious practices had greater chances of getting involved with crime. Furthermore, a high mortality rate was also observed in this particular population.

Strengths and limitations

An important strength point of the study concerns the fact that this is a 5-year retrospective observational study conducted in an underprivileged violent area based on the follow-up of individuals who sought treatment for alcohol and drug problems. The findings of this study should be read with caution, due to the fact that the data collected in the study comprised a convenient sample assessed in one alcohol and drug treatment center alone. Additional limitations include the fact that subjects’ diagnoses were obtained from hard copies of patients’ medical records and charts did not include the history of mental health comorbidities. Interviews also relied on families’ reports to confirm subjects’ death. Research outcomes should not be generalized to the general population.

Acknowledgments

This work was funded by Ford International Foundation and Instituto Carlos Chagas. The authors would like to thank Jardim Ângela’s Psychosocial Treatment Center for Alcohol and Drug Abuse staff, specially Camila Garcia Grandi and Thais dos Reis Vilela, all of whom extensively contributed for the success of this study.

References

1. World Health Organization. Global status report on alcohol and health. Geneva: World Health Organization, Department of Mental Health and Substance Abuse 2011.         [ Links ]

2. Schmidt MI, Duncan BB, Silva GA, Menezes AM, Monteiro CA, Barreto SM, Chor D, Menezes PR. Chronic non-communicable diseases in Brazil: burden and current challenges. The Lancet, 2011: 6736(11)60135-9 [doi:10.1016/S0140]         [ Links ].

3. Chaloupka FJ, Grossman M, Saffer H. The effects of price on alcohol consumption and alcohol-related problems. Alcohol Res Health. 2002;26(1):22-34.         [ Links ]

4. Miller TR, Levy DT, Cohen MA, Cox KLC. Costs of alcohol and drug-involved crime. Prev Science. 2006;7:333-42.         [ Links ]

5. Vaillant GE. The Natural History of Alcoholism Revisited. Cambridge MA: Harvard University Press 1995.         [ Links ]

6. Laranjeira R, Pinsky I, Zaleski M, Caetano R. I levantamento nacional sobre os padrões de consumo de álcool na população brasileira. Secretaria Nacional Antidrogas. Revisão técnica científica: Paulina do Carmo Arruda Vieira Duarte. Secretaria Nacional Antidrogas, Brasília, 2007. http://www.uniad.org.br/docs/ILevAlcool.PDF (acessado em 25/Abril/2011).         [ Links ]

7. Mari JJ, Mello MF, Figueira I. The impact of urban violence on mental health. Rev Bras Psiquiatr. 2008;30:183-4.         [ Links ]

8. Collins, J.J., Messerschmidt, P.M. Epidemiology of alcohol-related violence. Alcohol Health Res World. 1993;17:93-100.         [ Links ]

9. Laranjeira R, Hinkly D. Evaluation of alcohol outlet density and its relation with violence. Rev Bras Psiquiatr. 2002;36:455-61.         [ Links ]

10. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (4^th edition). Washington DC: Author 1994.         [ Links ]

11. Addiction Research Foundation. Directory of client outcome measures for addition treatment programs. Toronto: Addiction Research Foundation 1994.         [ Links ]

12. Dias AC, Ribeiro MR, Dunn J, Sesso R, Laranjeira R. Follow up study of crack cocaine users: situation of the patients after 2, 5, and 12 years. Subst Abus. 2008;29:71-9.         [ Links ]

13. Raistrick D, Dunbar G, Davidson R. Development of a questionnaire to measure alcohol dependence. Br J Addict. 1983;78:89-95.         [ Links ]

14. Jorge MR, Masur J. The use of the short-form alcohol dependence data questionnaire (SADD) in Brazilian alcoholic patients. Br J Addict. 1985;80:301-5.         [ Links ]

15. Tabakoff B, Dongier M. The WHO/ISBRA study on state and trait markers in alcoholism: progress report. Alcohol Clin Exp Res. 1996;20:243-7.         [ Links ]

16. Miller WR, Tonigan JS, Longabaugh R. The Drinker Inventory of Consequences (DrInC): an instrument for assessing adverse consequences of alcohol abuse. Test manual NIAAA Project MATCH. Monograph Series volume 4 NIH PUB. No. 95-3911. Washington: Government Printing Office 1995.         [ Links ]

17. Figlie NB, Dunn J, Gomes LCS, Turisco J, Payá R, Laranjeira R. Motivation to change drinking behavior among outpatients with and without gastric disease. Sao Paulo Med J. 2005;123:223-8.         [ Links ]

18. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. J Postgrad Med. 2002;48:206-8.         [ Links ]

19. SPSS Inc. SPSS Statistical Package for Social Sciences. Version 8.0.1 for Windows. Chicago (IL): SPSS Inc. 1997.         [ Links ]

20. Hair JF, Anderson RE, Tatham RL, Black WC. Multivariate Data Analysis. New Jersey: Prentice Hall 1998.         [ Links ]

21. Hosmer DW, Lemeshow S. Applied Logistic Regression, New York: Wiley 2000.         [ Links ]

22. Anderson P, Chisholm D, Fuhr DC. Effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet. 2009;27:2234-46.         [ Links ]

23. Centers for Disease Control (CDC) and Prevention, National Center for Chronic Disease prevention and Health Promotion. Alcohol attributable deaths and years of potential life lost-United States 2001. MMWR Morb Mortal Wkly Rep. 2004;57:866-70.         [ Links ]

24. Roizen J. Epidemiological issues in alcohol-related violence. Recent Dev Alcohol. 1997;13:7-40.         [ Links ]

25. Greenfeld LA. Alcohol and crime: an analysis of national data on the prevalence of alcohol involvement in crime (Bureau of Justice Statistics: Report Prepared for the Assistant Attorney General’s National Symposium on Alcohol and Crime – 1998). Retrieved Aug. 6, 2009 from www.ojp.usdoj.gov/bjs/pub/pdf/ac.pdf        [ Links ]

26. National Center for Injury Prevention and Control, “10 Leading Causes of Death, United States: 2002, Black, Both Sexes”. In WISQARS Leading Causes of Death Reports, 1999-2002.         [ Links ] National Research Council and Institute of Medicine, Reducing Underage Drinking: A Collective Responsibility. Washington, DC: National Academies Press 2004, pp. 60-61.         [ Links ]

27. Thomas D, Townsend T, Belgrave F. The influence of cultural and racial identification on the psychosocial adjustment of innercity African American children in school. Am J Community Psychol. 2003;32:217-28.         [ Links ]

28. Mäkelä P. Alcohol-related mortality as a function of socio-economic status. Addiction. 2009;94:867-86.         [ Links ]

29. Rehm J, Gmel G, Sempos CT, Trevisan M. Alcohol-related morbidity and mortality. Alcohol Res Health. 2003a;27:39-51.         [ Links ]

30. Sanchez ZVDM, Nappo SA. Religious intervention and recovery from drug addiction. Rev Bras Psiquiatr. 2008;42:265-72.         [ Links ]

31. Sullivan AR. Mortality differentials and religion in the U.S.: Religious Affiliation and Attendance. J Sci Study Relig. 2010;49(4):740-53.         [ Links ]

32. Koenig HG. Religion and medicine IV: religion, physical health, and clinical implications. Int J Psychiatry Med. 2001;31(3):321-36.         [ Links ]

33. Koenig HG, McCullough M, Larson D. Handbook of religion and health. Oxford University, 2001.         [ Links ]

34. Moreira-Almeida A, Pinsky I, Zaleski M, Laranjeira R. Religiuos envolvement and sociodemografic factors: a Brazilian national survey. Rev Psiq Clin. 2010;37(1):22-5.         [ Links ]

Corresponding author:
Neliana Buzi Figlie
R. Leandro Dupret, 204, conj. 92
São Paulo, SP, Brazil; CEP: 04025-011
Phone/Fax: (+55 11) 3807-3327, (+55 11) 2864-8781
E-mail: neliana_figlie@uol.com.br

Received on March 3, 2011; accepted on July 25, 2011