A: Subcommittee for Review of Epidemiology Portfolio
B: Experts in Epidemiology
C: NIAAA Program Staff
D: NIAAA Staff, Representatives from other NIH Institutes, and Guests
EPIDEMIOLOGY
REPORT OF A SUBCOMMITTEE OF THE NATIONAL ADVISORY COUNCIL
ON ALCOHOL ABUSE AND ALCOHOLISM
EXECUTIVE SUMMARY
The National Institute on Alcohol Abuse and Alcoholism's (NIAAA) Subcommittee for the Review of the Extramural Research Portfolio for Epidemiology met on 5-6 May 1999. The charge to the Subcommittee was to examine the appropriateness of the breadth, coverage, and balance of the epidemiology portfolio, identifying research areas that are well covered and others which are either under-investigated or which otherwise warrant significantly increased attention. The Subcommittee was asked also to provide specific advice and guidance on the scope and direction of the Institute's extramural research activities in the epidemiology area.
The Subcommittee for the Review of the Extramural Research Portfolio for Epidemiology consisted of a chair, NIAAA Advisory Council member, and an advisory group of six individuals. Three of these individuals have demonstrated expertise in alcohol-related areas, and three individuals have demonstrated expertise in non-alcohol-related areas (see Appendix A).
The review process was initiated by having experts (see Appendix B) in epidemiology prepare written assessments of the state of knowledge, gaps in knowledge, and research opportunities. NIAAA program staff (see Appendix C) presented the current extramural portfolio, categorized into the areas of methodology/nosology; moderate alcohol consumption, sexually transmitted diseases, disability, and comorbidity; intentional/ unintentional injuries; and special populations. All information was shared with experts, selected NIAAA staff, and the chair and advisory group before the meeting.
A summary of FY 98 epidemiology awards is detailed below.
|
Epidemiology |
Percentage of Epidemiology to Total |
| |
No. |
Amount
(in thousands) |
No. |
Amount |
|
Research Project Grants1 |
103 |
$30,460 |
18% |
22% |
|
Cooperative Agreements |
0 |
0 |
0% |
0% |
|
Research Centers |
1 |
1,716 |
7% |
8 % |
|
Research Careers |
9 |
705 |
15% |
11% |
|
Research Training |
3 |
210 |
4% |
4% |
|
Total |
116 |
$33,091 |
16% |
19% |
1 includes SBIR awards and reimbursable funds.
On 5-6 May 1999, experts and NIAAA program staff made abbreviated presentations of their material followed by discussion among all of the participants, including representatives from other NIH Institutes and guests (see Appendix D).
PRIORITIES RESULTING FROM REVIEW OF
EPIDEMIOLOGY PORTFOLIO
It is noted that as epidemiology has matured as a discipline, it has evolved from descriptive to hypothesis-driven research. Alcohol-related epidemiological research is encouraged to continue to emphasize an analytic or hypothesis-driven approach.
The following five areas were recommended for major emphasis.
· Identify patterns, risk factors, and protective factors for alcohol use and escalating consumption among 8 - 12 year olds. Secondary analyses of existing data sets as well as new studies are needed.
· Emphasize positive and negative consequences associated with moderate alcohol consumption. Consideration should be given to influence of interacting variables, such as genetic factors, diet, and physical activity. There is minimal need for additional correlative studies of moderate alcohol consumption and coronary heart disease.
· Determine relations among excessive alcohol consumption, hepatitis C, and alcoholic liver disease.
· Conduct longitudinal natural history studies with emphasis on developmental transitions and risk factors that precipitate or protect transitions from moderate to excessive alcohol consumption.
· Determine relations between alcohol consumption and violence, including spousal and child abuse.
Additional recommendations included:
· Identify factors that modify relations between risk factors, alcohol use, and consequences, including multiple levels of context, patterns of consumption, and comorbidity.
· Improve measurement of alcohol consumption and associated outcomes; validity and reliability of measures need to be documented. It would be useful to convene a consensus conference to determine preferred measures of alcohol consumption, especially for longitudinal studies.
· Further refine relations between alcohol consumption and injuries, morbidity, mortality, social roles, social functioning, and disability.
· Conduct a joint workshop between epidemiologists and geneticists to insure that epidemiologists collect data on genetically informative samples.
· Investigators should be encouraged to share epidemiological data with all interested and qualified parties.
Back to Top
OVERVIEW OF EPIDEMIOLOGY RESEARCH AT NIAAA
(Mary Dufour, M.D., M.P.H.)
Epidemiology is the study of a disease, injury, or other health-related condition in human populations and of those factors that increase the likelihood that such a condition will occur. Thus the primary aim of alcohol epidemiology is to identify and explain factors that shape the distribution of alcohol use, abuse and dependence and their consequences in various populations. Such a goal necessarily involves first defining the distribution of alcohol consumption and its related problems in the population. Knowledge of the nature and magnitude of alcohol's impact on the individual and on society is necessary for developing effective prevention, intervention and treatment strategies. It is also critical for informing policy makers and the general public of the need for and importance of alcohol research.
Epidemiology has its roots in the study of infectious diseases. Such research first used the public health model, which involves defining a disease agent, host and environment, to study and control epidemics. During the early and mid-20th century, medical advances and improvements in sanitation led to the mastery of many infectious diseases, such as tuberculosis. Soon thereafter, cancer and heart disease gained recognition as common killers. This resulted in establishing a new discipline--chronic disease epidemiology. In addition, the field of epidemiology's focus shifted to assessing the general health of communities under normal conditions rather than only under specific epidemic circumstances.
Alcohol epidemiology as a unique discipline is a relative newcomer to the field, and it marks a logical progression in the science of epidemiology. The epidemiology of many infectious diseases is relatively straightforward, having a specific, well-defined, easily identifiable etiologic agent (e.g., a bacterium or virus); a clear-cut case identification; a relatively short clinical course; and a quickly effective prevention (vaccination) or treatment (antibiotics). The epidemiology of chronic diseases such as cancer and heart disease is more complex. Alcohol epidemiology, however, is even more intricate and challenging because of the multiplicity of its contributing factors. It encompasses alcohol use, abuse and dependence as well as numerous medical, psychological, social, legal, and economic consequences. The large number and variablilty of definitions in alcohol research adds to the complexity of doing alcohol epidemiology.
Historical Perspective
The Division of Biometry and Epidemiology (DBE) began life as the Laboratory of Population Studies in NIAAA's intramural research program. As the missions and function of the Institute grew and changed, the activities of DBE seemed more compatible with and supportive of the extramural part of NIAAA. Prior to 1986, responsibility for the program portion of all NIAAA extramural grants resided in a single division, the Division of Extramural Research. Following his arrival in October of 1986, Dr. Gordis reorganized the Institute by creating the Division of Clinical and Prevention Research (DCPR) and distributing program responsibility for the Institute's grant portfolio across three Divisions--DCPR, the Division of Basic Research--DBR (formerly the Division of Extramural Research) and DBE. He also commissioned a peer-review of DBE's internal research. The review team of internationally recognized epidemiologists gave the Division's work high marks. They were especially enthusiastic about the data development, collection and surveillance activities and urged that DBE better publicize its endeavors. There are two Branches in DBE--the Biometry Branch and the Epidemiology Branch. Staff in the Biometry Branch primarily do intramural research, while those in the Epidemiology Branch both serve as program staff for epidemiology grants and pursue their own research.
The Alcohol Epidemiologic Data System (AEDS)
Due to the relatively small size of the DBE staff, data collection, analysis and reporting are facilitated by NIAAA's Alcohol Epidemiologic Data System (AEDS)--supported by one of NIAAA's largest and most long-standing contracts. Established in 1977, AEDS monitors trends in alcohol epidemiology through surveillance of apparent per capita alcohol consumption, cirrhosis mortality, alcohol-related morbidity among patients discharged from short-stay community hospitals, and alcohol-related fatal traffic crashes. These trends are reported in annual surveillance reports. AEDS also acquires and analyses major population-based surveys and other data collections that include alcohol as a major or minor component. Some of the surveys are designed specifically to answer alcohol-related questions; other surveys address wider issues but contain alcohol-related data. In order to encourage the fullest possible utilization of these data, every year AEDS publishes a Data Directory-- a listing of these surveys including title, sponsoring agency, contact design, sample characteristics, sample size, alcohol variables, other variables and limitations. AEDS also manages the Quick Facts electronic bulletin board that provides free online access to alcohol-related statistics collected by NIAAA and other researchers.
1988 Extramural Science Advisory Board Review: Incidence and Prevalence
On November 29, 1988, NIAAA's Ad Hoc Extramural Science Advisory Board met to ascertain research needs in the subject area of "Incidence and Prevalence." That meeting, the ninth in a series of external reviews of NIAAA's recent portfolio, marked the completion of the first cycle of review of NIAAA extramural research. Although the primary purpose of the review was to assess the "state of the science" of extramural alcohol epidemiologic research, the recommendations stemming from the meeting were quite global and many have been met through DBE intramural as well as extramural activities, some of which span Divisions. These recommendations are listed below.
General
Launch a major effort to standardize research and diagnostic measures
Increase support for longitudinal and cross-sectional surveys
Facilitate secondary analyses of existing data sets
Encourage interdisciplinary collaboration
Study design and analysis
Encourage the analysis of aggregate data
Support aggregate analyses at the State level
Encourage multi-occasion longitudinal research
Provide training opportunities
Drinking behavior
Support up-to-date reviews on drinking norms and contexts
Support research on long-term changes in drinking norms and contexts
Define drinking problems in terms of norms and contexts
Clarify the roles of formal and informal controls
Investigate international trends
Broaden the scope of drinking surveys
Encourage research on multiple measures of drinking patterns
Investigate the chronicity of heavy drinking
Investigate lagged changes in problem prevalence
Emphasize cultural influences on drinking behavior
Adverse consequences
Continue support for general population and subpopulation surveys
Facilitate the joining of the two traditions of measurement
Study large samples of heavier drinkers
Include alcohol data in other reporting systems
Encourage epidemiological research on cirrhosis
Expand epidemiological research efforts on alcohol and the
immunesystem
Support theoretical studies on vehicular unintentional injuries
Actively encourage research on non-vehicular unintentional injuries
Improve clinical sampling and reporting methods
Investigate the course of alcohol-related disorders
Compile a database on treatment statistics
In the intervening ten years, aspects of most recommendations have been addressed. Under the general recommendations heading, NIAAA has developed the capacity to effectively measure alcohol and other drug use disorders in general population surveys and has fielded it's own first ever national population-based survey, the 1992 National Longitudinal Alcohol Epidemiologic Survey. In May of 1997, DBE hosted an international workshop on consumption measures and models for use in policy development and evaluation that will significantly advance efforts in the alcohol measurement arena. Grantees as well as DBE staff have conducted numerous secondary analyses of existing databases.
Headway has also been made on all of the recommendations regarding study design and analysis. In addition to funding fellowships and individual training grants, DBE has provided long-term funding for two institutional training grants, that of Dr. Nancy Day at the University of Pittsburgh which is currently in its 17th year of funding (T32AA07453 Alcohol Research Training Grant) and that of Dr. Constance Weisner at the University of California, Berkeley which is currently in its 21st year (T32AA07240 Graduate Research Training on Alcohol Problems). Over the years, these two programs have trained at least 50 scientists currently in the alcohol field.
The NIAAA Alcohol Research Center "Epidemiology of Alcohol Problems" also at U.C. Berkeley, among many significant accomplishments, has contributed importantly to the advancement of the study of drinking behavior. Research examining international trends in alcohol use and consequences has increased, as has examination of lagged changes in problem prevalence.
Responding to recommendations in the area of adverse consequences, NIAAA has invested a great deal of effort in inserting quality alcohol measures in many data collection efforts across multiple federal agencies. NIAAA's portfolio on all aspects of alcohol and the immune system has burgeoned. In 1993, NIAAA added a research center more focused on the epidemiology of alcohol-related medical consequences--the "Center for Clinical and Medical Epidemiology" at the Research Institute on Addictions in Buffalo, New York. Extensive additional research has been done on drinking and driving and headway has been made in studying other alcohol-related unintentional injuries. Research on alcohol and intentional injuries including family violence, homicide and suicide is also under way. Clinical sampling and reporting falls under the preview of DCPR while the responsibility for compiling treatment statistics was ceded to SAMHSA when NIAAA, the National Institute on Drug Abuse (NIDA), and the National Institute on Mental Health (NIMH) joined NIH.
The Big Picture
Due to the relatively small size of the DBE grant portfolio and its variability from year to year, and in order to afford the broadest possible overview, the entire portfolio for the ten years since the last review is presented. However, prior to discussing the specifics of DBE's grant portfolio, it is necessary to put this activity into perspective. Although DBE's extramural grant portfolio represents the single largest piece of NIAAA's investment in epidemiologic research, by no stretch of the imagination does it comprise the entire investment.
The ten-year average distribution of the resources committed to the DBE epidemiology program by mechanism is discussed below. Research Project Grants represent 58.5% of DBE's epidemiology investment. Included here are grants having a program class code AE as well as those of other Institutes being co-funded by DBE.
The next largest portion is that of Interagency Agreements with 13.5% of the resources. Items in this category include NIAAA funding provided to the Centers for Disease Control and Prevention's (CDC) National Center for Health Statistics (NCHS) for participation in national surveys such as the 1988 National Health Interview Survey, the National Maternal and Infant Health Survey and the 1993 National Mortality Followback Survey. Funding to the Bureau of the Census to field the 1992 National Longitudinal Alcohol Epidemiologic Survey (NLAES) and the national survey currently under development also fall into this category.
Only slightly smaller, with 13.3% of the resources, is the portion representing NIAAA Research Centers. As mentioned earlier, the two NIAAA Research Centers which focus on epidemiology are "Epidemiology of Alcohol Problems" (P50AA05595) at the University of California, Berkeley and the "Center for Clinical and Medical Epidemiology" (P50AA09802) at the Research Institute on the Addictions in Buffalo, New York.
Research and Development Contracts represent 10.3% and include activities such as the AEDS contract. Careers and Training mechanisms make up the remaining 4.5%. The two previously mentioned long-standing institutional training grants comprise a significant portion of this funding.
Since the answers to epidemiology questions provide the foundation for much of NIAAA-supported research, additional epidemiology research is supported outside DBE. Each year NIAAA grant program staff code each funded application for a variety of variables including keywords, human subjects/special populations, human subjects age and various budget categories such as genetics, epidemiology, treatment of alcoholism, behavior, alcohol and pregnancy, alcohol-related medical disorders, etc. In order to get a sense of the magnitude of this portion of NIAAA epidemiology research, a search was done to find all 1998 projects with > 20% coded to the epidemiology budget category but without AE program class codes. Although the portion of the funding for research project grants (RPG's) shrank a bit relative to the ten-year amounts, it still represents over half (52.4%) of epidemiology funding. Among the RPG's, 14.2% reside outside of DBE with 38% of those having a program class code of AS (Health Services Research) and 49% AP (Prevention). The remaining 3% are AC (treatment) and special collaborative minority programs. Research Center Grants are 25.8%, representing an additional 18.1% in center resources. Resources for training/careers also increase a bit with 21.5% of these funds going for epidemiology training/careers outside of the purview of DBE. The proportions of the resources for DBE research and development contracts and interagency agreements decrease to 5.7% and 9.6%, respectively.
From time to time extramural researchers have voiced concerns that alcohol epidemiology research project grants get shortchanged in favor of intramural epidemiologic research. The above information clearly documents that the bulk of NIAAA epidemiology resources are used to fund RPG's, centers and career/training grants.
Topic Areas
The research supported in DBE falls into a variety of categories. The selection of categories and the placement of grants within categories are to some extent arbitrary. Many other classification schemes are possible. In addition, very few grants belong exclusively in one category: overlap is extensive. For example, relatively few grants are "pure methodology" but grants in every topic area have methodological components. Certain grants included under "Injuries" could just as easily have been placed under
"Family", "Race/Ethnicity" or "Women." Some of the grants under "Women" could be in "Morbidity/Disability/Mortality" or "Elderly." Selected grants in "Race/Ethnicity" could be in "Youth/Family" and vice versa. Following is a table of the eleven broad topic areas selected as well as the number and percentage of grants assigned to each category. Grants from FY 1989 to the present are included.
|
DBE Grants |
Number of DBE grants |
Percent |
|
Alcohol and Pregnancy/FAS |
4 |
3.2 |
|
Genetics |
13 |
10.4 |
|
Methodology/Nosology
(including Dependence) |
15 |
12.0 |
|
Morbidity/Disability/ Mortality |
15 |
12.0 |
|
Injuries/Violence |
11 |
8.8 |
|
Youth/Family |
22 |
17.6 |
|
Elderly |
3 |
2.4 |
|
Women |
9 |
7.2 |
|
Race/Ethnicity |
15 |
12.0 |
|
Workplace |
5 |
4.0 |
|
Other |
13 |
10.4 |
|
DBE Total |
125 |
100 |
FORMAT FOR THE PROGRAM PRESENTATION
For each of the topic areas, a brief introduction is followed by a listing of PAST GRANTS--these are grants for which funding was completed sometime during the 10-year period. Next is a summary of CURRENT GRANTS that includes not only new grants but long-standing continuations that are presently funded. For past and current grants, information provided includes principal investigator, grant number and title and a brief summary. Sections on SELECTED FINDINGS and GAPS complete the narrative. Only certain findings are highlighted to provide a flavor of the nature of the research. Likewise only selected gaps are mentioned. Clearly additional gaps will be identified over the course of the review meeting. The DBE grants in "Alcohol and Pregnancy/FAS" and "Genetics" have already been presented in earlier reviews of these areas. Brief summaries are included here only for the sake of completeness.
FUTURE DIRECTIONS
Substantial progress has been made over the past ten years in alcohol epidemiology. The purpose of this meeting is to assess the "state of the science" of alcohol epidemiology, to identify gaps in DBE's portfolio and suggest future directions for research. In preparing for this review, DBE staff has identified several gaps. Additional gaps will be articulated over the course of the meeting and in the final report of the committee. In addition to identifying gaps in existing areas of research, new areas must be considered as well. Small Area Epidemiology is coming of age.
All areas of alcohol research are rapidly advancing and it is important to utilize the fruits of research in other disciplines to advance alcohol epidemiology. For example, increasing incorporation of the measurement of genetic material and other biomarkers in general population surveys would enhance the value of the data collected.
Back to Top
METHODOLOGY/NOSOLOGY
METHODOLOGY AND ANALYSIS IN ALCOHOL EPIDEMIOLOGY:
RECENT ADVANCES AND OPPORTUNITIES
State of Knowledge (Tom K. Greenfield, Ph.D.)
Epidemiology of alcohol use and alcohol-related consequences descriptively monitors use, abuse, dependence, and acute and chronic problems associated with drinking and investigates mechanisms involved in alcohol consumptive behavior and outcomes, or potential harms. Recently, there have been several methodological developments including (1) improvements in measurement of human alcohol consumption and related problems; (2) use of relevant explanatory constructs such as individual differences in attitudes and expectancies, and features of communities or the contexts where alcohol use occurs; and (3) use of improved methods for analyzing differences between subgroups with varying risk factors on the one hand and changes over time, on the other, or both, using either repeated cross-sectional series or longitudinal panel designs.
Issues concerning the assessment of alcohol consumption (volume, pattern, lifetime estimates) have been reviewed (Rehm, 1998), and research to improve the quality of these data, including cognitive studies of how to ask questions concerning consumption parameters is underway. Measurement of alcohol-related problems is beginning to focus on using a continuum of severity and dissecting relations among various drinking-related problems (Caetano, 1997).
There have been recent conceptual advances in understanding the role of cognitive factors, attitudes, and expectancies in assessing individual differences and risks for developing alcohol-related problems (Goldman et al., in press). However, for a considerable time those concerned with a more complete specification of influences on alcohol consumption and problem rates have recognized that individual-level variables should be supplemented with influences occurring at broader social levels (Johnstone, 1994) based on an ecological understanding of alcohol use (Gruenewald et al., 1993).
Multilevel, contextual, and hierarchical modeling techniques are now being applied to alcohol-related epidemiology and prevention research. A new generation of computer programs (Arbuckle, 1995) allow for analysis of continuous and categorical measures, analysis with multiple groups, and adjustments for non-normal and/or missing data. There has been a convergence of growth curve modeling (study of individual change over time where growth parameters vary) and structural equation modeling (involving relationships between observed and latent variables) within a single analytic framework (Muthen, in press).
There are three specific recommendations: (1) Technology transfer of user friendly, new analytical strategies should be encouraged. (2) Severity of alcohol-related problems needs to be better characterized and measured. (3) It is important to avoid population-specific instrument bias when collecting epidemiological data.
NOSOLOGY
State of Knowledge (John E. Helzer, M.D.)
The major nosological systems for substance use disorders and for mental health in general have been clinically rather than empirically derived. There has been great progress in the last two decades in the development of a much more uniform language for classification in mental health and in the development of instruments to accomplish this systematically. In spite of these advances, there is growing concern about the continuing adequacy of nosological systems. Although the DSM-IV and its predecessors create a nosology that serves the clinical needs of rapid decision-making and communication, it is somewhat problematic for research purposes, which requires a dimensional system that utilizes differential scores on several parameters simultaneously, is predominantly empirical, and can be successfully administered and processed by computer. There is continuing evidence that categorical nosologies are insufficient in capturing important distinctions among alcoholics and they demonstrate poor sensitivity in adolescents (Martin et al., 1995).
In correspondence with the refinement of categorical nosologies, increasingly sophisticated instruments have been developed to gather necessary clinical data. Measurement of current consumption patterns and longitudinal estimates of consumption can be difficult because of retrospective recall. However, results from a variety of new techniques are promising, e.g., computer-assisted collection of sensitive data (Turner et al., 1998); wearable electrochemical devices that monitor alcohol levels over time (Swift, 1993); telephone interactive voice response systems to collect sensitive information on a daily basis (Searles, in press).
Statistical models are being developed that can examine psychiatric signs and symptoms for latent structure (Anthony et al., 1995). These methods can result in models that assume an underlying categorical or a dimensional typology. Another important area for typological improvement is environmental variables. Multiple studies attest to the potential importance of a better understanding and classification of environmental variables in the nosology of alcoholism and in further clarification of the interaction of genetics and environment in the etiology of alcoholism.
Nosological issues are important in cross-national epidemiology where the needs and opportunities are especially great in the field of alcoholism. In the past two decades, the field has moved beyond the problem of making coherent cross-national comparisons based on illness rates defined and assessed in each country independently. There is a growing number of instruments that have been cross-nationally developed and validated including the CIDI, SCAN, AUDADIS-ADR, and AUDIT (Cottler et al., 1997; Allen et al., 1997).
There are two specific recommendations: (1) There is an ongoing need for accurate clinical diagnostic approaches that facilitate rapid decision-making and communication; such classification schemes tend to be categorical. (2) New research diagnostic criteria should be developed empirically and require assessing several parameters simultaneously; criteria tend to be dimensional rather than categorical.
Back to Top
NIAAA PORTFOLIO ON METHODOLOGY/NOSOLOGY
(Mary Dufour, M.D., M.P.H.)
METHODOLOGY
Major foci of the grant program in methodology have been the investigation of factors related to non-response to survey questionnaires concerning alcohol consumption, alcohol use problems and comorbidity; the influence of measurement factors (e.g., order of response options) in self-reports of alcohol use; reliability and validity of commonly used alcohol screening instruments in surveys (i.e., CAGE, Short MAST and AUDIT); and identification of characteristics of abstainers, other than their non-use of alcohol, which may confound associations between drinking and mortality. Another focus has been the use of meta-analysis to predict drinking patterns/problems across time; integrate individual and societal level explanations of drinking patterns/problems; reconcile differences in findings across studies; and synthesize findings on research questions.
Past Grants
Ronald D. Hays (R01AA07852), "Microcomputer Assessment of Alcohol Use" developed an interactive, computerized assessment program for investigating and improving the validity of self-reports of alcohol use. The program was then administered by microcomputer, and measurement factors such as order of items, quantitativeness of responses, and providing feedback were investigated.
Current Grants
Kaye M. Fillmore (R01AA07034), "Collaborative Alcohol Related Longitudinal Project" is continuing the meta-analytic study of an archive of raw data from 40 longitudinal general population data sets and two adoptee studies from 18 countries. The project, now in its 12th year, addresses two substantive domains. The first domain involves social change and gender roles, including examination of cross-study variation in the prediction of gender-specific drinking patterns/problems. The second domain uses meta-analysis to combine research results in longitudinal studies containing information on mortality, morbidity, and other health characteristics, as well as social characteristics. One objective of the research is to identify characteristics of abstainers (e.g., subjective health, social class, social integration, and mental health) that may account for their excess mortality over moderate drinkers. In addition, gender roles are being linked to social structure and to social change, comparing different societies and periods of history. Dr. Fillmore is also in the 6th year of a program of research (K05AA00172), "Situational and Contextual Factors in Drinking Practices" involving synthesis of raw data from multiple, international treatment evaluation studies. The project's aims include the prediction of drinking patterns/problems across time; integrating individual and societal level explanations of drinking patterns/problems; systematically reconciling differences in findings across research studies and methodologies; and synthesizing findings on research questions. Design effect is being used to differentiate the effects of societal change on drinking behavior from confounding factors.
Three new studies are also being supported:
Cheryl S. Alexander (R21AA10939), "Adolescent Drinking-Analytic Models, and Longitudinal Data" is developing and applying marginal, latent variable and transition models to an investigation of alcohol behavior using data from a longitudinal study of rural adolescents. Objectives include investigating whether standard alcohol questions on use, attitudes or associated problems have the same meaning for adolescents at different developmental stages and among different gender or racial/ethnic groups, and whether information about drinking is remembered and retrieved in a similar manner. In addition, the study explores ways these adolescents understand and process questions about their alcohol use and the predictors of use.
Steven S. Henley (R43 AA11607), "Exploiting Hidden Structures in Epidemiological Data" is developing an advanced data recoding algorithm using techniques that combine pattern recognition, stochastic optimization, and genetic algorithms to exploit structural relationships between continuous and categorical predictor variables and categorical outcomes in a Phase I SBIR grant.
Bengt O. Muthen (K02AA00230), "Advanced Analysis of the Development of Alcohol Problems" is developing new statistical methods to answer substantive questions regarding the development and prevention of alcohol problems and collaborating with alcohol researchers on advanced statistical analysis of their longitudinal data sets. Information on the new methods will be disseminated via articles, a book and training sessions.
Selected Findings
Measurement. When drivers court-ordered into alcohol treatment were interviewed by microcomputer about alcohol frequency, analysis revealed that more precise response options and standardized location of items within the questionnaire improved the quality of data. Results of another analysis showed that response quantitativeness, order of presentation of response options, and relative placement of alcohol use items in the questionnaire had minimal influence on the quality of self-report data. When this study was replicated with a sample of university students, results indicated that presentation of high frequency response options prior to lower frequency options increased self-reports of frequency.
Recalled perceptions of intoxication, impairment and reported BAC on the day or night of their arrest were studied in drivers convicted of DUI. Recalled levels of intoxication and impairment corresponded to their reported BAC across drinking pattern categories. Factor analysis identified a factor for "cogent risk-taking" in which people were aware of their level of intoxication and impairment and drove nonetheless.
Non-Response. Demographic characteristics and history of alcohol problems and psychiatric disorders were compared in responders and non-responders in an 11 year follow-up study. Few differences were found. Non-response was greater in men, those with less education and in low users of medical care. Non-response was also greater among those reporting drinking and driving trouble or a history of barbiturate use or dependence. The findings were not changed by refusal conversion, suggesting that efforts to enlist participation of initial refusers might not be needed.
Meta-analyses. Meta-analysis of eight general population surveys of men evaluated all-cause mortality rates by drinking patterns. The most consistent finding was the association of heavy drinking with mortality among youth. Among adults, drinking 43 or more drinks per month and drinking 21 or more times per month were associated with increased mortality risk. No evidence was found to support the hypothesis that abstinence is associated with greater mortality risk than light drinking. There was no association of former drinking with mortality across studies.
Across ten general population studies, adult male former drinkers were consistently more likely to be heavier smokers, depressed, unemployed, lower SES and to have used marijuana than long-term abstainers. Adult female former drinkers were consistently more likely to be heavier smokers, in poorer health, not religious, and unmarried than long-term abstainers. Both long term abstainers and former drinkers tended to be of lower SES than light drinkers and report poorer health. Thus, characteristics of these two groups of abstainers, other than non-use of alcohol, may confound associations found between drinking and mortality.
Meta-analysis of three general population surveys of adult women evaluated all-cause mortality rates by drinking pattern. In models in which age was controlled, odds of death for long-term abstainers and former drinkers was greater than for light drinkers; odds of death for moderate and heavy drinkers was greater than for light drinkers. When other psychosocial attributes were controlled, odds of death for heavy drinkers were greater than those for light drinkers. When interactions of age and the two forms of abstinence were introduced, results were consistent with the hypothesis that characteristics of abstainers other that their non-use of alcohol may account for their higher mortality risk.
Two measures of alcohol consumption were used to predict groups of alcohol problems in 21 general population studies from 11 countries. When only individual-level variables were controlled (age and sex), both quantity and frequency were risk-factors for each drinking problem. However, except in the case of the association of quantity with alcohol treatment, the magnitude of these risks was heterogeneous across studies. When blocking for societal-level traits, each had more relevance than some, but not all of the associations between consumption and problems. These findings suggest that impact of norms and societal stress in groups have different but significant consequences for the relationships between consumption and problems.
Gaps
Research opportunities in methodology abound. The rapid pace of development of computer-based technology makes this a dynamic and exciting area. Better methodologies for measuring alcohol use across the lifespan are needed, as are techniques for calculating alcohol-related risks and benefits in various stages of life. Refinement of methodologies for calculating alcohol-related morbidity and mortality are sorely needed. Many opportunities exist for methodology research to contribute to the advancement of each of the topic areas outlined in this review.
NOSOLOGY
During the past ten years, a major focus of DBE's nosology grant program has been the validity of DSM conceptualizations and measurements of alcohol abuse and dependence. The validity of DSM dependence has generally received support from research evidence, whereas abuse has not. In the past, diagnostic criteria for alcohol use disorders were developed largely by research with adult clinical populations. In order to provide more balance, NIAAA has supported studies of general populations. Other areas of emphasis have been the role of comorbidity (e.g., abuse of other drugs and psychiatric conditions) in the development and course of alcohol use disorders, familial and psychophysiological precursors and other predictors of alcohol problems, predicting drinking patterns/problems across time-periods, and integrating individual and social level explanations of drinking patterns/problems.
Past Grants
Kathleen K. Bucholz (R01AA08752), "New Alcohol Problems and Disorders -- 11 Year ECA Follow-up" followed a group of subjects who were interviewed eleven years earlier as part of the St. Louis ECA study in order to identify new cases of alcohol problems or alcohol use disorders. Specific aims included estimating the 11-year cumulative incidence of alcohol problems and alcohol use disorders; identifying predictors of onset and determining whether predictors would be the same regardless of diagnostic classification system.
Deborah S. Hasin (R01AA08159), "Alcohol Dependence: General Population Validity" examined reliability, validity, and prognostic implications of the DSM-III-R and ICD-10 definitions of alcohol dependence. Dr. Hasin (R01AA08910), "Validating Epidemiologic Measures of Alcohol Dependence" also provided information on the reliability and validity in clinical samples of the Alcohol Use Disorders and Associated Disabilities Interview Schedule (AUDADIS), the survey instrument utilized in NIAAA's large, national, general population survey, the National Longitudinal Alcohol Epidemiology Survey (NLAES), which is administered by lay interviewers.
Vicki Eileen Pollock (R01AA08031), "Alcoholism Risk: Psychobiological Alcohol Effects in Men" assessed the general hypothesis that psychological and physiological characteristics distinguish men at high risk for developing alcoholism from controls while sober, and also after acute alcohol administration. Her related K-award (K02AA00146), "Psychophysiological Factors In Alcoholism Development" tested the validity of predictions drawn from normalization theory which holds that certain individuals are susceptible to developing alcoholism because acute alcohol intake normalizes their psychophysiological functions.
Two additional studies supported were grants by:
Arthur Alterman (R01AA03736), entitled "Risk Factors and Development of Problem Drinking," and Ronald Drabman (R03AA07683) entitled "Taste Preference in High Risk Groups for Alcoholism."
Current Grants
Deborah S. Hasin (K02AA00161), "Epidemiologic Studies of Alcohol Use Disorders" is now in the 6th year of a program of research focused on investigating validity of concepts and definitions of alcohol use disorders; instrument development; measures for the reliable and valid diagnosis of psychiatric disorders in heavy drinkers and drug users; and investigation of the effects of psychiatric comorbidity on the course of alcohol and drug disorders. The current study includes development of reliable and valid measures of HIV sex risk behaviors in alcohol and drug abusing samples and investigation of the effects of acculturation on drinking and the prevalence of alcohol disorders among Russian immigrants to the United States and Israel.
Two new grants are also being supported:
Eric O. Johnson (R03AA11846), "Alcoholism Typology" is refining a typology of alcoholism that has promise for distinguishing subtypes based on relative genetic and environmental influence.
Christopher S. Martin (K02AA00249), "Quantitative Methods in Alcohol Research" is developing expertise in advanced quantitative methods to address theoretical and conceptual issues in ways traditional statistical techniques cannot. In both adolescent and adult data sets, these methods are being used to contrast the validity of existing systems such as DSM-IV and ICD-10 with a new model of the diagnosis and taxonomy of alcohol use disorders in which abuse and dependence diagnoses represent milder and more severe manifestation of the same core addiction constructs.
Selected Findings
Reliability and Validity Studies. A number of findings were based on a sample of community residents screened for heavy drinking in the previous twelve months and diagnosed for alcohol use disorders by different sets of diagnostic criteria (i.e., DSM III, III-R, IV; and ICD-10). Agreement for most comparisons involving diagnoses for current dependence ranged from good to excellent; this finding held across gender, age, and racial subgroups. Evaluation of the association of a set of seven criterion variables external to the alcohol diagnostic criteria and DSM-IV dependence and abuse diagnoses revealed that dependence diagnosis was significantly associated with all criterion variables when compared with no diagnosis, even though cases of dependence were mild. In contrast, the abuse diagnosis did not show a pattern of association with the criterion variables when compared to no diagnosis.
When a modified version of the Structural Clinical Interview for the DSM was used to assess DSM-IV alcohol abuse and dependence symptoms in adolescent drinkers with varying levels of alcohol consumption and alcohol related problems, results generally supported the utility of DSM-IV criteria for alcohol dependence. However, those with DSM-IV alcohol abuse diagnoses displayed quite heterogeneous patterns of symptomatology, suggesting limitations of the criteria for DSM-IV alcohol abuse in adolescents. Although, they are not in the DSM-IV criteria, alcohol-related blackouts, craving and risky sexual behavior were common in adolescents with DSM-IV alcohol dependence and abuse diagnoses and are an important focus for assessment and treatment efforts. Another study involving male and female adolescent drinkers identified three stages of alcohol problems distinguished by time to onset: heavy and heedless drinking with associated social and role obligation problems, psychological dependence, and withdrawal.
Development of Alcohol Abuse and Dependence. An investigation of the influence of gender, comorbidity, drinking history and age on the clinical manifestations of DSM-IV alcohol abuse or dependence based on a sample drawn from the NIMH Epidemiologic Catchment Area Study showed that gender and comorbidity had independent effects on problem drinking. Gender contributed to the age of onset of problem drinking and the rate of its development. Comorbidity, drinking history and age contributed independently to its severity. The effects of these variables in this community sample paralleled those reported in treatment samples. Another study comparing three samples of alcohol-dependent individuals revealed similarities among clinical, family study, and community samples in terms of appearance of milestones of alcohol dependence.
Comorbidity. The study of lifetime prevalence of other psychiatric disorders in three samples (clinical, family study, and community) of alcohol-dependent individuals found that for females ages 45 or younger, lifetime prevalence of major depression was high in both clinical and family study alcoholics compared with community alcoholics; male alcoholics from the community study had excess of drug dependence. Findings suggest that, although alcoholics identified in clinical setting may have more severe alcohol dependence, certain types of psychiatric comorbidity are present to a greater degree in samples of untreated alcoholics.
Analysis of adolescents admitted for inpatient treatment of alcohol abuse showed that 96 percent reported use of drugs other than alcohol during their lives. Findings also indicated that extensive polydrug use characterized the large majority of male and female adolescent alcohol abusers. In another investigation of adolescent drinkers with DSM-IV alcohol dependence, alcohol abuse, and no alcohol diagnosis, findings revealed that the total number of illicit drugs ever used was greater in the alcohol dependence and abuse groups. A consistent pattern of psychoactive substance use was also found in which alcohol was followed by marijuana, which was followed by other drugs. Frequency and extent of polydrug use was associated with being older and having higher levels of behavioral under control and negative emotionality. In national study, a significant association between depression and DSM-III-R alcohol dependence was found.
Gaps
Much research remains to be done in alcohol-related nosology. Development of enhanced measures of alcohol use disorders would be valuable as would better measures of the chronicity of drinking patterns and problems. Additional research on the roles of various comorbid conditions in the etiology and consequences of alcohol problems is also critically needed.
Back to Top
CHRONIC DISEASES, AIDS, DISABILITY, AND COMORBIDITY
THE EPIDEMIOLOGY OF MODERATE ALCOHOL CONSUMPTION
AND RISK OF CHRONIC DISEASE
State of Knowledge (Eric Rimm, Sc.D.)
Twenty-five years ago, Klatsky et al. (1974) observed an inverse association between alcohol and coronary heart disease and later Baboriak et al. (1979) found that moderate alcohol consumption was associated with reduced occlusive disease in patients with coronary angiography. These findings sparked a new era in alcohol research which focused on identifying the health effects (risks and benefits) associated with moderate alcohol consumption.
Today, a large body of evidence from epidemiological studies suggests that alcohol in moderation (2 drinks/day for men and 1 drink/day for women) is associated with a lower risk of coronary heart disease, diabetes, and ischemic stroke. However, there is now convincing evidence that consumption at this level also may lead to modest increases in risk of breast (among women) and colorectal cancer. Because cardiovascular disease is the leading cause of death among men and women, results from large prospective studies of alcohol and all-cause mortality consistently find the lowest risk of death among moderate drinkers.
Further research is needed to help clarify the importance of drinking patterns, dietary interactions, and genetic susceptibility on reported associations between moderate alcohol consumption and risk of chronic disease.
There are two specific recommendations: (1) Develop better measures of alcohol consumption patterns for different populations in ongoing longitudinal studies, with repeated assessment of consumption throughout the lifespan. (2) Determine influence of selected factors (e.g., diet, lifestyle, genetic factors) that may modify relations between alcohol consumption and chronic diseases.
ALCOHOL IN THE EPIDEMIOLOGY OF AIDS AND
OTHER SEXUALLY TRANSMITTED DISEASES
State of Knowledge (Karen F. Trocki, Ph.D.)
A decade and a half ago, when work began on the epidemiology of AIDS and sexually transmitted diseases (STDs), there was a strong focus on "disinhibition" and the possibility that alcohol use in conjunction with sex was a key factor contributing to disease risk because of interference with use of condoms. The finding that alcohol use is not a particularly good predictor of condom use patterns has been fairly consistent (Leigh and Stall, 1993). Less than half of the studies have shown any association whatsoever; even then, the association is often for a few subgroups within the population studied and/or the strength of the association is marginal.
Other aspects of the alcohol/risky sex association have shown strong and consistent patterns. There is a clear relationship between number of sexual partners and use/abuse of alcohol. In virtually every study, there has been a strong association between that aspect of sexual risk and quantity/frequency of drinking (Bailey et al., 1998). Furthermore, heavy alcohol use and/or number of sexual partners is the most important variable marking STD history and predicting future STDs (Ericksen and Trocki, 1994).
Evidence is accumulating that the context of alcohol consumption (e.g., bars, clubs, parties) can be a source of multiple partners, partner change, and other aspects that are known to play an important role in the dynamics of STD transmission (Stall et al., 1990).
There are two specific recommendations: (1) Determine relations among impulsivity, alcohol intoxication, multiple sexual partners, and sexually transmitted diseases (STDs).
(2) Identify environmental risk factors for STDs that interact with alcohol consumption and determine influence of age, gender, and racial/ethnic differences.
EPIDEMIOLOGY OF ALCOHOL-RELATED BEHAVIORS AND DISABILITY
State of Knowledge (Paul M. Roman, Ph.D.)
There are few scientifically sound data on the prevalence of drinking and alcohol problems among those who are disabled. Although most of the available literature has a clinical orientation, there is an apparently large unmet need for dealing with alcohol dependence among the disabled.
Limited survey data suggest there are no differences among disabled and non-disabled in reported lifetime prevalence of alcohol use/abuse (Gilson et al., 1996).
Acquired disability may be more linked to drinking problems than congenital disability, and predictors of post-injury drinking problems may include such variables as sociopathy, thrill-seeking, family background of alcohol problems, and other individual traits that are independent of disabling events (Babor, 1993). Most investigators agree that pre-injury drinking problems are the strongest predictors of post-injury drinking problems, especially in individuals experiencing traumatic brain injury (Kreutzer et al., 1996) and spinal cord injuries (Hubbard et al., 1996).
There are two specific recommendations: (1) Secondary analyses of existing data sets should be conducted in order to gain information on disability, disablement, and alcohol consumption and related problems. (2) A major epidemiological study of alcohol consumption and related problems in the disabled is needed.
COMORBIDITY OF ALCOHOL ABUSE/DEPENDENCE AND OTHER DISORDERS
State of Knowledge (Linda B. Cottler, Ph.D.)
Comorbidity describes the co-occurrence of two or more disorders in a single individual (Feinstein, 1970). Comorbidity is meant to be restricted to disorders, and not symptoms, and each disorder should have the characteristic pattern and etiologic basis typically present when each disorder is found by itself. Illnesses may be described as lifetime comorbid or simultaneously, i.e., currently comorbid. Lifetime comorbidity may describe two or more illnesses that have occurred at some time, but not necessarily at the same time in one's lifetime. Simultaneous or current comorbidity describes two (or more) illnesses in which criteria for each are fully met together at the same time (past year).
Comorbidity of psychiatric disorders and alcohol-use disorders can be explained by (1) psychiatric disorders leading to the use of alcohol and thus alcohol-use disorders, (2) psychiatric disorders and alcohol-use disorders are correlated because both conditions share common etiologic factors, or (3) alcohol use and its consequences can lead to other psychiatric disorders (Reiger et al., 1990). To differentiate among these alternatives, one must understand causal associations.
The Epidemiologic Catchment Area (ECA) study (landmark study of psychiatric disorders among the general population (N=20,000) in the early 1980s) found that, on a lifetime basis, 37% of those who met criteria for alcohol abuse or dependence also met criteria for a mental disorder other than drug abuse or dependence (Reiger et al., 1990), with the most common disorder being any anxiety disorder (19%), followed by antisocial personality disorder (14%) and affective disorder (13%). Conversely, persons with antisocial personality disorder (ASPD) were 21 times as likely to have an alcohol-use disorder than persons without ASPD, with a rate of 74%. Alcohol-use disorders were also common among persons with any affective disorder (22%). The National Comorbidity Survey, conducted in the mid-1990s, found that the most common disorders were alcohol dependence and depression; moreover, the estimates of comorbid disorders were even higher than in the ECA.
Comorbidity of alcohol abuse and dependence with other drug abuse and dependence is 41% in men and 47% in women (Kessler et al., 1997). In addition to the comorbidity described above, the literature also documents comorbidity between alcohol-use disorders and suicide (17%; Berglund, 1984), HIV and other STDs (Shillington et al., 1995), personality disorders (44%; Verheul et al., 1995), pathologic gambling (Crockford and el-Guebaly, 1998), and ADHD (Biederman et al., 1995).
A specific recommendation is that comorbidity needs to be assessed within a longitudinal project involving diverse cultures from early childhood to death.
NIAAA PORTFOLIO ON MORBIDITY/DISABILITY/MORTALITY
(Mary Dufour, M.D., M.P.H.)
This heading subsumes an eclectic collection of grants that fall into the broad categories of Moderate Drinking, Nutrition, HIV/AIDS, and Comorbidity. Comorbidity is a crosscutting theme, aspects of which are interspersed throughout the entire DBE portfolio. The study of the relationship between alcohol use and abuse and Attention Deficit Disorder is included here. Other comorbid conditions are covered elsewhere. DBE has no grants in the area of Disability.
MODERATE DRINKING
Past Grants
In 1990, Douglas Coate's grant (R01AA08366) entitled "Moderate Drinking and Coronary Heart Disease Mortality" was funded to attempt to ascertain whether the statistical association between moderate drinking and coronary heart disease (CHD) mortality could be explained by a correlation with confounding variables such as income, education, life style and diet which underlie the relationship but which had not been controlled for in earlier studies. The investigator also hoped to determine whether the statistical association held up across different socioeconomic, age and sex groups and whether the direction of the causality in the moderate drinking-CHD relationship runs from moderate drinking to improved CHD health or from good health to moderate drinking. This work which utilized two large national data sets, the first National Health and Nutrition Examination Survey (NHANES I) and the NHANES Epidemiologic Follow-up Survey (NHEFS), found evidence for a beneficial effect of moderate drinking in white men with accelerated time-to-failure models showing 3-4% longer life spans for moderate drinkers compared to nondrinkers or light drinkers.
Dorit Carmelli's grant (R01AA08925) entitled "Alcohol Consumption and Mortality in Veteran Twins" was also funded in the early 1990's. Utilizing the National Academy of Sciences-National Research Council's World War II Twin Registry, the work examined alcohol consumption and mortality in 4960 pairs of adult, male, veteran U.S. twins born between 1917 and 1927. Analyzing the first or only deaths in twin pairs discordant for drinking disclosed a relative risk (RR) of 1.93 for death in abstainer twins compared to their light-moderate drinking cotwins. Excess mortality in twin abstainers was also found for deaths from cardiovascular disease (RR=2.0) and other causes of death excluding cancers (RR=3.2). The protective effect of light-to-moderate drinking was not found for twins who were smokers at baseline.
Wendy Carman's small grant (R03AA08630) entitled "Alcohol Use in Tecumseh, Michigan: 1959-1978" which was funded about the same time endeavored to assess the alcohol data in the Tecumseh Study and using database management systems, create alcohol use histories for each individual and family in the study in an attempt to enhance usefulness of the data for the study of the impact of alcohol consumption on diabetes, hypertension, cancer, osteoporosis, osteoarthritis and other aging-related conditions.
Current Grants
In response to a growing body of scientific research and increasing interest in the topic of moderate alcohol consumption, in the summer of 1995, NIAAA issued a Request for Applications (RFA) (AA-95-004) entitled "Moderate Alcohol Consumption: Benefits and Risks". Two grants in the DBE portfolio which were received in response to this RFA are:
Eric Rimm's grant (R01AA11181) entitled "Alcohol Consumption Patterns, Biomarkers and Health" and Arthur Klatsky's grant (R01AA10830) entitled "Alcohol Drinking and Risk of Cerebrovascular Disease." Dr. Rimm is evaluating prospectively the impact of moderate daily alcohol consumption, regular weekly patterns of consumption and usual percent alcohol consumed with meals on the risk of diabetes mellitus, hypertension, cardiovascular disease and total mortality in two large cohorts--the Health Professionals Follow-up Study (51,529 middle-aged men) and the Nurses Health Study II (116,601 women age 25-42 in 1989). The investigators are also planning to examine the impact of moderate alcohol consumption on plasma lipids and thrombotic factors thought to be related to risk for coronary heart disease. Dr. Klatsky and colleagues are prospectively examining the relationship between quantity and type of alcohol consumption and hospitalization and death from cerebrovascular disease in a large (128,934 men and women), multiethnic population in Northern California.
Other current active grants in the DBE portfolio which are examining some aspect of moderate drinking include those of:
Youlian Liao (R01AA11141) entitled "Alcohol--Benefits and Risks--Analytic Studies"; Dr. Ken Mukamal (F32AA05534) entitled "Cost Effectiveness Changes in National Alcohol Use"; and Marcia Russell (R21AA11684) entitled "Guidelines for Moderate Drinking--Influence of Patterns." Dr. Liao is utilizing data from several large national population surveys in order to better define the overall risks and benefits and possible trade-off of moderate drinking for the population as a whole and for specific subgroups. Dr. Mukamal is examining the probable effects of changes in national alcohol consumption through the Coronary Heart Disease Policy Model, a model which uses data from national surveys and published cohort studies to predict coronary heart disease incidence and mortality. In the past, many studies have reported alcohol consumption in terms of average daily amount without assessing volume per occasion. Dr. Russell plans to assess the impact that determining actual volume of alcohol consumption per occasion would have on recommendations for moderate drinking.
NUTRITION
Two of DBE's grants fall under the broad heading of nutrition.
Katherine Flegal's grant (R23AA07018) entitled "Measuring Alcohol as a Risk Factor Using Dietary Data" assessed the reliability and validity of the measures used to assess alcohol consumption in national food consumption surveys. Dr. Flegal found that alcohol is among the best reported of dietary items and that these methods, while not ideal, produce reasonably reliable and valid measures of alcohol consumption. She also recommended revisions of survey items to further improve reporting.
Bryan Johnstone's FIRST award (R29AA08475) entitled "Alcohol and Diet: General Population Relationships" attempted to examine hypothesized relationships between alcohol consumption, intake of alcohol-independent total calories, and principal nutrient and selected food item categories. Also included is body mass index using data from several large national population surveys.
Gaps
To date, few data sources used in moderate drinking research contain adequate measurement of all potential confounds particularly diet and exercise. Broader inclusion and better measurement of potential confounds in a single data source is now needed. In addition, alcohol consumption is often assessed at one point in time and mortality assessed many years hence. More work needs to be done on measurement of alcohol consumption over the lifespan and how this impacts on mortality. Methodological work which permits risk/benefit analysis of moderate alcohol consumption across the lifespan for specific individuals taking gender, various health conditions; etc. into account would be a welcome contribution. Much of epidemiologic research remains to be done on alcohol and diet/nutrition. Improvements in survey instrument design would be of value. Also of interest is the impact of diet composition on alcoholic organ damage. Some animal and human research suggests, for example, that the amount and type of dietary fat intake impacts on the development and severity of alcoholic liver damage. The DBE portfolio currently has no grants in alcohol-related chronic disease epidemiology so this area has great potential for expansion. Refinements of the existing methodologies for deriving alcohol-related mortality and morbidity statistics are especially critical.
HIV/AIDS
Stephen Hulley's grant (R01AA08238) entitled "HIV Risk and Infection Among Alcoholics Seeking Therapy" ascertained the seroprevalence of HIV infection in alcohol dependent individuals seeking treatment in alcohol treatment facilities in the San Francisco Bay area as compared to that of a general population sample drawn from the corresponding area. This work, which was funded in the early 1990's, was among the first to report that the seroprevalence of HIV infection was dramatically higher in the treatment sample than the general population sample. In addition to making a substantive contribution, the investigators also made significant methodological contributions. Since this research began in the relatively early days of AIDS research, the investigators were forced to wrestle with difficult issues of keeping the strictest confidentiality of the patient information and at the same time maintaining the necessary linkages between the patient data and the biological test results. The investigators were extremely creative in devising a coding strategy to meet these disparate needs--so successful that others have since adopted these methods.
Over the past ten years, the primary focus of Karen Trocki's grant (R01AA08564) entitled "Epidemiology of Alcohol Problems--Risk of AIDS" has been to explore how drinking behaviors are associated with sexual risk behaviors. This work has demonstrated that a critical factor in sexually transmitted disease (STD) risk is the number of sexual partners and that the context of drinking strongly influences the availability of certain types of sexual partners and plays a role in determining the nature of the interpersonal relationships. Although still tracking general population trends in disease risk, more detailed work on the role of bars, clubs, taverns, and other locales is currently in progress to elucidate more fully the epidemiology of AIDS and other STDS as related to alcohol consumption. Both qualitative and quantitative methods are being employed to study both heterosexual and homosexual populations in order to gain the broadest overview of specific risk patterns.
Gaps
It is now well established that alcohol consumption contributes to risky behavior, which, in turn, increases risk of exposure to HIV infection. Additional information regarding important individual, situational and contextual factors should be forthcoming from ongoing research mentioned above. Critical questions which remain to be answered include: (1) given exposure to HIV infection, what is the impact of alcohol consumption on the likelihood of acquiring infection and (2) given HIV infection, what is the impact of alcohol consumption on the development and clinical course of AIDS?
COMORBIDITY
Earlier work on alcohol and Attention Deficit Hyperactivity Disorder (ADHD) has suggested that children with ADHD are at increased risk for alcohol abuse relative to children without the disorder; however, small sample sizes and methodological limitations have hampered generalizability.
Brooke Molina's recently awarded grant R0AA11873 "Development of Alcohol Use and Abuse in ADHD Adolescents"aims to address the large knowledge gaps by more adequately describing the risk of ADHD children for alcohol use and abuse and elucidating the extent to which emergence and development of alcohol abuse is associated with other domains of functioning by yearly interviewing 500 children who have been previously diagnosed with ADHD and for whom a large battery of standardized assessments exist. Dr. Molina also has a Scientist Development Award (K21AA00202) "Adolescent Alcohol Abuse and Attention Deficit Disorder"to support a follow-up study comparing adolescents with ADHD with non-ADHD adolescents matched from the probands' schools. ADHD and comorbid psychopathology will be examined as risk factors for adolescent alcohol use, abuse and dependence.
Gaps
Some aspects of comorbidity are addressed in other sections of this review, particularly Nosology and Methodology. Clearly, however, there is ample opportunity for future research. Research on the impact of various comorbid conditions on the clinical course and outcome of alcohol use disorders would be a valuable addition.
DISABILITY
Although DBE has no NIAAA grants in the area of disability, we have played a very active role in a long standing Cooperative Agreement between the World Health Organization (WHO) and NIAAA, the National Institute on Mental Health (NIMH) and the National Institute on Drug Abuse (NIDA): the WHO/NIMH/NIAAA/NIDA Joint Project on Diagnosis and Classification of Disorders (U01MH35883). Among the major accomplishments during the previous 12 years of funding, the Joint Project developed two international diagnostic assessment instruments: the Schedules for Clinical Assessment in Neuropsychiatry (SCAN), designed for use in clinical research, and the Composite International Diagnostic Interview (CIDI), designed for use in epidemiologic and other community based research. The primary purpose of the current five years of funding, began late in 1995, is to develop research instruments that assess "disablement" associated with alcohol, other drug abuse and mental disorders, and to design pilot studies in health services research with these instruments utilizing the existing instrument development methodologies and international research support network created under earlier funding cycles. [Disablement is an umbrella term for "impairments, disabilities, and handicaps" as used in the WHO International Classification of Diseases, Tenth Revision (ICD-10).] Work is also continuing and expanding ongoing in-depth analyses of results from the reliability and validity study of the alcohol and other drug sections of the SCAN and CIDI conducted earlier.
Gaps
Since the above cooperative agreement is our only disability-related research, many research opportunities exist in the area of alcohol-related disability. Extensive methodological work will be a necessary and critical first step, since alcohol-related disability is even more difficult to detect and measure than alcohol-related morbidity.
Back to Top
INTENTIONAL-UNINTENTIONAL INJURIES
EPIDEMIOLOGY OF INTENTIONAL AND UNINTENTIONAL INJURIES
State of Knowledge (Gordon S. Smith, Ch.B., M.P.H.)
Alcohol consumption can impair (1) judgement, (2) coordination, and (3) ability to perceive and respond to hazards, with the risk of injury increasing exponentially with increasing blood alcohol levels. It has also been documented that persons may still be at risk even when blood alcohol levels are zero due to the "hangover" effect. Research examining alcohol habits (i.e., usual drinking quantity and frequency) and chronic abuse is less common than research examining acute involvement in injuries. However, there are data to suggest that chronic alcohol abuse is more likely to result in death from injury than disease (Li et al., 1994).
Alcohol involvement is generally more common in severe injuries than in more minor injuries, e.g., 41% involvement in fatal automobile crashes, 9% in injury crashes, and 5% in property damage crashes (Fell, 1999). Smith et al. (1999) estimated that 21-42% of fatal unintentional injuries involved alcohol, and Hingson and Howland (1993) estimated that 17% of nonfatal unintentional burns and 24% of nonfatal unintentional fall injuries involved alcohol. Alcohol use is also implicated in fatal pedestrian injuries (36%); non-commercial aviation pilot fatalities (10%, Holdener, 1993); bicycle fatalities (25%; Abel et al., 1984); and drowning (21-42%; Hingson and Howland, 1993).
The association of alcohol use and intentional injuries has been known for many years although the relationship is complicated. Alcohol involvement in perpetrators of violence has been estimated to be 28-86% of homicide offenders, 24-37% of assault offenders, 13-60% of sexual offenders, and 6-57% of male domestic offenders (Roizen, 1997). Alcohol involvement in victims of violent events is as prevalent as that for the perpetrator. Alcoholism and alcohol abuse are second to depression and other affective disorders as major risk factors identified for suicide (Blumenthal, 1988).
There are three specific recommendations: (1) Develop better measures of alcohol consumption and alcohol involvement in injuries. (2) Develop more innovative research designs to examine the causal role of alcohol in suicide, homicide, and other violent events. (3) Determine relations among patterns of alcohol consumption, inpulsivity, and injuries.
NIAAA PORTFOLIO ON INJURIES AND VIOLENCE
(Captain Darryl Bertolucci)
Injury and violence research has focused on the role of alcohol in emergency room visits, alcohol and aquatic injury and deaths, costs associated with alcohol-involved injury, and alcohol-related marital violence. The current portfolio expands the focus to bicycle and boating accidents; ethnicity and alcohol-related spousal abuse; alcohol and injury in the U.S. Army; and the relationship between alcohol exposure and injury risk, including risk associated with low levels of consumption.
Past Grants
Cheryl J. Cherpitel (R01AA07102), "Alcohol and Emergency Room Admission in Kaiser Hospitals" conducted a study whose goals were to describe and analyze the role of alcohol in the event which brought the individual to the emergency room and develop a system which would help to predict those who may be at risk for alcohol-related injuries from both emergency room and general population samples.
Jonathan Howland (R01AA08320), "National Survey of Drinking in Aquatic Settings" studied the contribution of alcohol to untoward aquatic events using a national random digit-dial survey of the general population. Data were collected concerning drinking in conjunction with aquatic activities, attitudes towards the propriety of drinking on or near the water, perceptions of drinking as a risk factor for drowning, knowledge of drinking and boating laws, and support for these laws. In addition, the researchers explored whether people are more likely to engage in risky aquatic behaviors when they had been drinking than when they had not.
Glenda K. Kantor (R01AA08269), "Alcohol and Marital Conflict: A Longitudinal Analysis" conducted a longitudinal examination of marital conflict and assaults in alcohol abusing families compared to other families in the general population. Data used included a 1985 nationally representative sample of 6002 American families (National Family Violence Survey), including an oversample of Hispanic and Black families, and panel data collected in 1986 and 1987 on a stratified subsample of 1408 violent and non-violent families.
Kenneth E. Leonard (R01AA07183), "Frequent Heavy Drinking and Marital Violence in Newlyweds" provided epidemiological information concerning the incidence of alcohol-related marital violence and investigated prospectively the impact of frequent heavy drinking, alcohol expectancies, hostility, and marital conflict on marital violence among newly married couples 18-26 years old. Dr. Leonard (R01AA08128), "Experimental Study of Alcohol and Marital Aggression" also examined the impact of alcohol cues and alcohol administration on the marital interactions of physically aggressive married couples, distressed but not aggressive couples, and nondistressed couples.
Ted R. Miller (R01AA09812), "Injuries from Alcohol--Incidence and Cost" developed and applied improved methods for estimating the incidence and costs of injury from alcohol. He also examined various approaches to attribution, and compared cost and incidence estimates for attributable injury to the more commonly used, but methodologically questionable, estimates for alcohol-related injuries.
Current Grants
One current study is a continuation of previous work.
Gordon S. Smith (R01AA07700), "Alcohol-Related Drowning and Serious Boating Injuries" builds on earlier FIRST award (R29AA07700) on alcohol and injuries that emphasized alcohol-related drowning. The research program has successfully developed a comprehensive statewide surveillance system for fatal injuries in collaboration with the Maryland medical examiners' office. Dr. Smith also extended the use of the surveillance system and methodology to investigate occupational injuries. The current research extends the work on drowning to include a case-control study of serious boating injuries (fatalities and hospitalized cases).
In addition, four new studies are being supported.
Nicole C. Bell (R29AA11407), "Alcohol Use and Injury Outcomes in the US Army" is combining data from the U.S. Army Health Risk Appraisal (HRA) data, including self-reports of drinking and other risk taking behaviors, with an existing computerized health-outcome database. HRA data will be linked by social security numbers to hospitalization, death, accident and disability data in order to quantify the relationship between alcohol use and actual injuries and to identify subgroups of men and women particularly at risk because of unhealthy drinking behaviors.
Raul Caetano (R01AA10908), "Ethnicity, Alcohol and Spousal Violence" is using a national survey to examine the association between alcohol and spousal violence among Blacks, Hispanics, and Whites in order to estimate nationwide rates of intimate partner violence with data from both couple members.
Guohua Li (R29AA09963), "The Role of Alcohol in Bicycling Injuries"examines the magnitude and factors related to alcohol involvement in fatal and nonfatal bicycling injuries and tests the hypothesis that alcohol intoxication is associated with significantly increased likelihood of fatality given a serious bicycling injury. Results to date indicate that alcohol plays an important role in fatal and serious bicycling injuries.
Daniel C. Vinson (R01AA11078), "Alcohol and the Risk of Injury--A Case-Crossover Study" is conducting a study to quantify the association between acute alcohol exposure and the risk of injury, including the relative risk at low levels of alcohol consumption. In addition, the research examines the relationship between chronic alcohol involvement and injury risk, with and without simultaneously including acute alcohol exposure in the model, and defines the extent of confounding due to other background variables (such as personality traits) in the association between alcohol and injury. The study utilizes a case-crossover design, in which each injured person is both a case (at one point in time) and a control (at prior times), within a population-based case control study.
Selected Findings Emergency Room (ER) Patients.
Alcohol's role in injury cases seen in the ER differed from its contribution to non-injury ER cases. With the injured more likely than the non-injured to have positive breath-analyzer readings and to report heavy drinking, more frequent drunkenness, prior alcohol-related accidents and prior treatment for an alcohol-related problem, but were no more likely to report harmful consequences of drinking or alcohol dependence experiences during the last year. No differences in drinking patterns, alcohol-related problems or gender or age distribution were found between clinic patients and the non-injured ER patients. Both the noninjured ER patients and the clinic patients were less likely to report heavy and problem drinking, or treatment for an alcohol-related problem during the last year, as compared to those in the general population.
Alcohol and Drowning. In addition to having higher drowning rates than women for most age groups, men had elevated risks for exposure, risk taking, and alcohol use. It was concluded that several factors contribute to their high drowning rates, including a possible interaction between overestimation of abilities and heavy alcohol use.
Violence in the Family. Data analysis, which controlled for a number of confounding risk factors such as low socioeconomic status, found that children who had experienced corporal punishment in adolescence had an increased risk later in life of depressive symptoms, suicidal thoughts, and alcohol abuse. Results of a study on marital violence indicated that problem drinking relates to the quality of the marital relationship even in the early stages of relationship formation, and that it was not the problem drinking, but the levels of anger and femininity associated with the problem drinking that were responsible for less effective relationships. Wives decreased their average alcohol consumption over the first year of marriage; husbands did not decrease their average alcohol consumption; alcohol dependency scores and alcohol problems of both husbands and wives declined; children and pregnancy were related to declining alcohol consumption by wives; the level of wife's intimacy appeared to facilitate alcohol consumption by husbands; the level of husband's intimacy led to lower levels of consumption by wives; and husband's alcohol dependence was correlated to wife's dependence. Results of other work indicated a significant relationship between husband's heavy alcohol use and premarital aggression. In addition, there were significant interactions between husband's heavy alcohol use and marital dissatisfaction and between husband's heavy alcohol use, husband hostility and husband belief in alcohol as an excuse for aggression. Overall, alcohol played a role in marital violence, marital quality, and marital disruptions, but high levels of individual alcohol consumption in a marriage did not uniformly lead to lower marital quality.
Cost of Alcohol-Related Injury. The incidence of alcohol-involved highway crashes (those in which a driver or nonoccupant had been drinking) was estimated from federal databases. The comprehensive cost of alcohol-involved crashes was $148 billion in 1990, including $46 billion in monetary costs and $102 billion in lost quality of life. This represents $1.09 per drink of alcohol consumed. Crashes where blood alcohol concentration (BAC) exceeded .10% accounted for 32% of comprehensive crash costs, and crashes with lower positive BAC accounted for another 8%. Averaged across all drinks, other people collectively pay $0.63 in crash costs every time someone takes a drink. Nonhospitalized medically treated injuries averaged $579 in medical spending per case ($181 per visit, 3.2 visits per injury).
Gaps
Although over the past decade much progress has been made in the study of alcohol- related injuries many gaps remain to be filled.
Better methodologies for ascertainment and surveillance of alcohol-related injuries other than those involving alcohol-related traffic accidents are needed. Better methodologies for determining causal relations between alcohol consumption and various types of injuries are needed. Alcohol and violence is a difficult and sensitive area to study but one in which more work is important. More refined research on the role of childhood physical/sexual abuse in later development of alcohol problems is also critical.
Back to Top
SPECIAL POPULATIONS
EPIDEMIOLOGY AND ETIOLOGY OF ADOLESCENT ALCOHOL INVOLVEMENT IN THE UNITED STATES
State of Knowledge (John E. Donovan, Ph.D.)
Adolescent alcohol involvement includes alcohol use, problem drinking, and alcohol-use disorders.
Information concerning the nationwide prevalence of alcohol use among adolescents comes from two major annual surveys (Monitoring the Future [MTF] and National Household Survey on Drug Abuse [NHSDA]) and a biennial national survey (Youth Risk Behavior Survey [YRBS]). The older the adolescents the larger the percentage that reports experience with alcohol and more likely are reports of having ever been drunk; largest increase is between eighth and tenth grades, where prevalence rates either double or triple. Gender differences emerge in the 10th grade, with males more likely to be engaged in high volume or daily drinking. Acceleration of rates of alcohol use among males but not females occurs from tenth to twelfth grades. Annual prevalence of alcohol use in 12th graders is less among African-American (72%) than among either Caucasian (83%) or Hispanic American (80%). Despite differences in overall level, patterning of use by gender, age, and ethnic/racial background is similar in the two major national surveys.
Beverage alcohol consumed by the greatest number of students (7 - 12th grades) was beer, followed in order by wine coolers, liquor, and wine (OIG, 1991). Average weekly consumption was 3.9 beers for boys, 2.2 beers for girls, 1.4 wine coolers for boys, 1.8 wine coolers for girls, 1.4 glasses of wine for boys, 1.6 glasses for girls, and 1.4 drinks of liquor for boys, 0.9 for girls. Almost two-thirds of underage students who drink reported being able to buy their own alcohol.
There are no current data on the nationwide prevalence of problem drinking among U.S. adolescents. According to MTF survey data from 1994 through 1997, one in three high school seniors, and almost two thirds of those who have consumed alcohol ten or more times in their life, have experienced problems due to their drinking (O'Malley et al., 1998). Three or more problems were experienced by 18% of all seniors and by 34% of those who had consumed alcohol 10 or more times in their life.
There are also no data on the prevalence of alcohol abuse and alcohol dependence, as defined by standardized psychiatric diagnostic criteria, among U.S. adolescents nationwide. In 1996 data from the NHSDA, 12-17 year old adolescents reported one or more problems related to a diagnosis of alcohol dependence (12.7%), two or more problems (12.7%), and three or more problems (4.3%) (SAMSHA, 1998). Kandel et al. (1997) reanalyzed data from the NHSDA to determine proxy measures of alcohol dependence (3 or more problems) and observed an overall prevalence of 2.9%. Although more male than female adolescents had consumed alcohol in the past year, more female than male adolescents reported 3 or more dependence symptoms.
It is useful to distinguish between risk factors for initiation versus escalation of drinking. Risk factors for initiation of drinking include (1) family factors (presence of stepparents, divorced or separated parents, parental drinking, lack of parental support, as well as family density of alcoholism); (2) peer factors (peer involvement in delinquent or drug-using behaviors and perceived peer attitudes toward drug use); (3) personality factors (lower levels of conventionality and higher alcohol expectancies); and (4) behavioral factors (prior involvement in delinquent behavior). Risk factors for the escalation of alcohol use include (1) family factors (greater parental alcohol use, positive parental attitudes toward child drinking, greater sibling alcohol use, and less parental monitoring of child behavior); (2) peer factors (greater susceptibility to peer pressure, association with older peers and working peers, more frequent offers to drink, and exaggerated perceptions of number of friends who drink); (3) personality factors (greater expectations, greater disinhibition, lower law-abidance, less knowledge about alcohol, and lower levels of personal competence and greater social skills); (4) behavior factors (frequent earlier delinquent behavior and lower school grades). Relatively little research has studied risk factors for transitions into problem drinking.
There are three specific recommendations: (1) It is important to collect data on alcohol use, problem drinking, and alcohol-use disorders among adolescents. (2) Determine patterns of alcohol use in preadolescent children. (3) Longitudinal studies of risk factors, beginning in preadolescence, are important to initiate.
NIAAA PORTFOLIO ON YOUTH AND FAMILY
(Vivian Faden, Ph.D.)
YOUTH
Research in this area has focused on how much young people drink, the etiology and development of drinking and problem drinking among youth, vulnerabilities and pre-dispositions among youth in general and among particular groups of young people, and the consequences of drinking by the young. In particular, the relationship between alcohol use and deviant behavior of adolescents has received attention, as has the relationship between adolescent and parental drinking. In addition, DBE is supporting two studies concerning alcohol use in homeless adolescents.
Past Grants
John S. Baer (RO1AA08632), "Prediction of Stability of High Risk Drinking" studied high-risk adolescent drinkers in order to predict chronicity of heavy drinking in this population. As part of this study, subjective intoxication and stress response dampening were evaluated in men and women college seniors during an alcohol challenge study. All were regular drinkers who reported being intoxicated at least twice a month by history.
Marie D. Cornelius (R29AA08284), "Alcohol Use Among Teenagers and Infant Outcome" examined the relationship between alcohol and other drug use by pregnant adolescents (both prior to and during pregnancy) and physical and developmental abnormalities among their offspring. Also explored was the relationship to actual drinking of psychosocial variables, demographics, knowledge and attitudes about drinking during pregnancy.
Frances M. Costa (R01AA10322), "Development of Adolescent Problem Drinking" analyzed data from a four-wave longitudinal study of adolescent health, collected from 1989-1992. The goal of this study was to learn more about the role of psychosocial risk and protective factors in accounting for variation in involvement in and transition to problem drinking among male and female adolescents of different racial/ethnic backgrounds. Risk and protective factors were drawn from the framework of problem behavior theory.
Lisa Crockett (R01AA09678), "Young Adult Consequences of Rural Adolescent Alcohol Use" studied drinking behavior in a sample of rural adolescents as they entered young adulthood. The goals of this three-year study were to identify developmental patterns of alcohol use during this transitional period of life, to examine the continuity/discontinuity in alcohol use from adolescence to adulthood, and to predict young adult alcohol use/abuse and examine the consequences of drinking in this population.
John E. Donovan and Richard Jessor (R01AA08007), "Drink-Driving and Risky Driving in Adolescents and Youth" examined the relationship of risky driving, including drinking-driving and drug-driving, and behavioral and psychosocial conventionality/unconventionality.
Candace M. Fleming (R03AA08211), "Alcohol Abuse and Depression in American Indians" performed secondary data analyses on a longitudinal survey of American Indian boarding school students to establish the incidence and prevalence of symptoms of alcohol and other drug abuse, depression, and anxiety in an American Indian adolescent population.
Joel W. Grube (R01AA08097), "Adolescent Drinking in the United States and Ireland" studied the development and maintenance of drinking, problem drinking and drinking problems among adolescents in the United States and the Republic of Ireland by following 2000 adolescents in each country over a period of three years.
Thomas G. Power (R01AA07740), "Attachment, Autonomy and Patterns of Adolescent Drinking" examined the role of adolescent attachments and autonomy in the development of adolescent alcohol consumption patterns, with a particular focus on the differential prediction of normative and problem drinking.
Valley Rachal (R01AA07766), "Sequencing of Alcohol, Drugs, and Tobacco among Youth" examined patterns of initiation and maintenance of alcohol, other drug, and tobacco use for all youth and subgroups of youth defined by gender, race and ethnicity.
John Schulenberg (R03AA09143), "Stability and Change in Alcohol Use among Youth" analyzed longitudinal data from the national Monitoring the Future project in order to examine adolescent personality and social context predictors of changes in binge drinking during the transition from adolescence to young adulthood.
John Welte (R01AA08157), "Drinking and Delinquency in Young Men" studied drinking, other drug use, and delinquent behavior in a sample of men aged 16-19 to elucidate the role of drinking and other drug use in prolonging and intensifying delinquency in young men.
Current Grants
The Division is currently supporting two new studies of alcohol abuse among homeless youth, aged 13-20.
Ana Marie Cauce's grant (R01AA10253) entitled "Homeless Youth-Alcohol and Psychosocial Risk" is currently in its fifth year. This grant is focusing on the identification and understanding of street youth's family histories, developmental patterns and processes, and socio-environmental conditions that are linked to their victimization on the streets, their involvement in violence against others, and their alcohol use and abuse. Of special interest in this study is the role played by alcohol abuse in placing street youth at risk of becoming victims and/or perpetrators of violence and of becoming homeless adults. This study is being carried out in the Seattle area.
Paul A. Toro's grant (R01AA10597) entitled "Alcohol and Homeless Youth-A Longitudinal Comparison" is a longitudinal comparison of the drinking and other outcomes of homeless adolescents (ages 13-17) and housed adolescents who are matched on gender, age, race, and neighborhood socioeconomic characteristics. This study is being carried out in Detroit.
Selected Findings
Adolescent Problem Drinking. Results of exploratory factor analyses supported the conclusion that adolescent problem drinking is a multidimensional phenomenon. Three dimensions measuring level or frequency of alcohol use, problems related to drinking and symptoms of dependency were identified; these factors were only moderately intercorrelated. Regression models confirmed the uniqueness of the three measures and suggested that the potency of specific risk factors vary for different types of problem drinking.
Among adolescents who were not problem drinkers, higher risk and lower protection accelerated the likelihood of becoming a problem drinker in subsequent years. Protective factors moderated the impact of risk factors in the cross-sectional account of problem drinking involvement, but not in the longitudinal account of the transition into problem drinking. Findings were similar for males and females and among White, Black and Hispanic adolescents. It was concluded that since protective factors play a role, independent of risk factors, in adolescent involvement in and transition to problem drinking, intervention efforts to enhance protection, especially for adolescents exposed to risk, should supplement efforts to reduce risk. In terms of predictors of adolescent drinking, both alcohol behavior and attitudes of parents and peers were significant: Parental attitudes were more important than parental alcohol behavior in predicting adolescent drinking, while peer alcohol behavior was more important than peer attitudes. Overall, peers had more influence on adolescent drinking than parents.
Findings regarding alcohol and violent crime indicated that usual drinking pattern and drinking before offending are related to aggravated assault, and that alcohol may have different roles in explaining different levels of violence. Results also indicated that the age of onset of substance use has significant effects on alcohol use, drug use, association with delinquent peers, and deviant activities. Those who began alcohol and other drug use at an early age are more likely to continue using, to be associated with delinquent peers, and to participate in deviant activities. In addition, early age of delinquency onset was a significant predictor of associating with delinquent peers, alcohol use, and deviant activities. In another study, drinking-driving, drug-driving and risky driving were found to comprise a more general, second order factor of adolescent problem driving behavior. Drinking-driving was found to be related to problem drinking, marijuana use, other illicit drug use and delinquent-type behavior. It was concluded that drinking-driving is part of a more general lifestyle involving behavioral and psychosocial unconventionality.
Among adolescents, drinking decreased during the first and third trimesters of pregnancy while binge drinking rose during the first trimester then fell sharply. Women who were binge drinkers during pregnancy were more likely to be White and heavier users of tobacco, cocaine and marijuana. Rates of first trimester binge drinking were higher among pregnant adolescents than adults. Rates for binge drinking and heavy drinking were highest among White teenagers.
In an international comparison, Irish students began drinking at an older age than American students, but reported higher prevalence rates for lifetime drinking, more frequent drinking and more frequent intoxication. Overall the Irish students were more likely to report alcohol problems. American men and women were more similar in their drinking patterns than Irish men and women.
Transition from Adolescence to Young Adulthood. Different trajectories of frequent binge drinking were observed during the transition to young adulthood: adolescents who were male, had low self-efficacy and drank primarily to get drunk were found to be at greater risk for increased binge drinking over time. In contrast, risk factors such as low conventionality varied according to initial level of binge drinking. Chronic and increasingly frequent binge drinking over time were found to be associated with difficulties in negotiating the transition to young adulthood. Among rural adolescents, a sharp increase was seen in the frequency of getting drunk across the high school years and a more gradual increase after high school for both males and females. Those who were never married as compared to ever-married, and those who were employed as compared to those who were unemployed reported a higher incidence of drunkenness. Antisocial behavior and family problems in rural adolescence were risk factors for problem drinking in adulthood.
Young Adults. During an alcohol challenge study, intersubject variability was greatest early in the study; during ethanol absorption, time to peak breath alcohol concentration (BrAC) varied from 10 to 91 minutes after drinking began and mean BrACs were significantly lower in females than in males. Task persistence after alcohol consumption among young adult offspring of alcoholics (COAs) and young adult offspring of nonalcoholics (nonCOAs) revealed significantly greater acceleration of response latencies after consumption of alcohol among COAs compared with nonCOAs. This study also investigated prevention of alcohol abuse among high-risk colleg