4
Epidemiology

Although originally concerned solely with communicable diseases, epidemiology has broadened its scope with time to encompass the study of the incidence, prevalence, causes, and consequences of a range of health problems and health behaviors (Rogers, 1965; NIDA, 1994a).1 The application of epidemiology to the study of drug use and abuse is relatively recent. During the outbreak of heroin abuse in the late 1960s, the term ''epidemic" began to be used (Kozel and Adams, 1986). After the National Institute on Drug Abuse (NIDA) was established in 1974, epidemiology became one of its earliest priorities, and one of the first publications in NIDA's Research Monograph Series was Epidemiology of Drug Abuse: Current Issues (NIDA, 1976). Epidemiological research continues to be one of the most active research programs at NIDA.

Traditionally, two classic triads of concepts serve as an organizational framework for epidemiology: the "epidemiologic triangle" of agent, host, and environment for analytic epidemiology and the specification of the rates of disease by person, place, and time for descriptive epidemiology (Lillienfeld and Stolley, 1994). In the application of the epidemiologic triangle, one would consider the drug as the agent of exposure, the host as the individual taking the drug, and the environment as the setting of

1  

Incidence is the number of new cases of a condition (such as heroin use) in a defined population within a specified period of time. Prevalence is the number of instances of a given condition (such as heroin use) in a defined population at a specified time (Last, 1983).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research 4 Epidemiology Although originally concerned solely with communicable diseases, epidemiology has broadened its scope with time to encompass the study of the incidence, prevalence, causes, and consequences of a range of health problems and health behaviors (Rogers, 1965; NIDA, 1994a).1 The application of epidemiology to the study of drug use and abuse is relatively recent. During the outbreak of heroin abuse in the late 1960s, the term ''epidemic" began to be used (Kozel and Adams, 1986). After the National Institute on Drug Abuse (NIDA) was established in 1974, epidemiology became one of its earliest priorities, and one of the first publications in NIDA's Research Monograph Series was Epidemiology of Drug Abuse: Current Issues (NIDA, 1976). Epidemiological research continues to be one of the most active research programs at NIDA. Traditionally, two classic triads of concepts serve as an organizational framework for epidemiology: the "epidemiologic triangle" of agent, host, and environment for analytic epidemiology and the specification of the rates of disease by person, place, and time for descriptive epidemiology (Lillienfeld and Stolley, 1994). In the application of the epidemiologic triangle, one would consider the drug as the agent of exposure, the host as the individual taking the drug, and the environment as the setting of 1   Incidence is the number of new cases of a condition (such as heroin use) in a defined population within a specified period of time. Prevalence is the number of instances of a given condition (such as heroin use) in a defined population at a specified time (Last, 1983).

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research exposure. Early research applied a narrow traditional epidemiological framework to the study of heroin use (de Alarcon, 1969; Hughes and Crawford, 1972); however, recent studies employ a broader concept of epidemiology in which both descriptive and analytic epidemiological studies are used to address the problems of drug use and abuse in society. Epidemiological research provides information essential for defining the scope of the problem by identifying populations at risk. Epidemiological data on trends in illicit drug use and abuse over time help to measure the effectiveness of the national drug control program. Epidemiological research provides insights into the etiology of drug initiation and use (Chapter 5). Additionally, epidemiological research provides information on the nature and extent of the multiple consequences of drug abuse (Chapter 7). Data on drug availability and demographics allow prevention and treatment programs to target the needs of those populations identified as at risk for increased alcohol and illicit drug use. This chapter describes the variety of data systems currently in place that address different aspects of the drug use problem in the United States and discusses accomplishments and future directions in epidemiological research. DEFINITIONS One way in which the epidemiology of drug abuse differs from more traditional epidemiological studies of infectious diseases is that drug abuse is not universally accepted as a medical condition. As indicated in Chapter 1, there are differences of opinion about applying the medical model to drug abuse. Researchers and clinicians commonly distinguish three levels of drug behavior: use, abuse, and dependence (see Chapter 1 and Appendix C). The stages of abuse and dependence are most clearly amenable to a diagnostic perspective, whereas the use stage is more readily characterized by its frequency, quantity, and duration. Therefore, use is generally the stage most easily and accurately measured outside clinical practice as described by epidemiological research. A complication in research, however, is that different drugs have different patterns of use, and the transition from use to abuse to dependence may be very different for different drugs (e.g., heroin as compared to marijuana). ACCOMPLISHMENTS Clearly, one of the major accomplishments in epidemiology has been to establish a variety of data systems that measure different aspects of drug use and abuse. Two major data systems provide broad-based statis-

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research tics on trends in drug use in the general population:2 the National Household Survey on Drug Abuse (NHSDA) and the Monitoring the Future study (MTF). Additionally, a number of surveys and other data collection efforts describe use and abuse in specific populations (Table 4.1). Other major accomplishments of epidemiological research include the development of valid measures and survey methodologies and the collection and analysis of data on co-occurring psychiatric disorders, natural history, and etiology of drug abuse. General Population Surveys National Household Survey on Drug Abuse NHSDA has been conducted periodically since 1971 and is currently an annual survey.3 It provides national-level estimates of the prevalence of use of illicit drugs, alcohol, and tobacco among members of the household population of the United States (surveys before 1991 excluded Alaska and Hawaii). The survey is estimated to represent 98 percent of the total population age 12 and over; completion rates of the survey in recent years vary between 74 and 84 percent. The subpopulations excluded are homeless persons; persons living in correctional facilities, nursing homes, and treatment centers; and active military personnel (SAMHSA, 1995c). The NHSDA series was designed to measure the prevalence and correlates of alcohol, tobacco, and illicit drug use in the United States; sufficient continuity has been evident in the core questions of NHSDA to be able to chart trends in drug use since 1972. In each survey, similar questions have been asked about the use of illicit drugs, alcohol, and tobacco in the respondent's lifetime, in the past year, and in the past month. Illicit drug use is defined as use of illegal drugs and nonmedical use of prescription-type psychotherapeutic drugs. An important aspect of the epidemiology of drug use is the variation among persons (based on gender, age, and other demographic factors), place (region, population density), and time. Table 4.2 shows the prevalence of illicit drug use in the past month among the household popula- 2   It is important to note that the total number of users results from the rates of use in different age groups in the population and from the demographic structure of the population. The actual number of users may increase while the average or overall rate of use are declining. 3   The survey has been conducted annually since 1990. The National Commission on Marihuana and Drug Abuse sponsored the 1971 and 1972 surveys, NIDA sponsored the NHSDA from 1974 to 1990, and the Substance Abuse and Mental Health Services Administration has been the responsible agency since 1992.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research TABLE 4.1 Major Epidemiologic Data Systems Data System Target Population Current Sponsor Status Estimates of Use in Various Populations     National Household survey on Drug Abuse General population, 12 years old and older SAMHSA Began in 1971; conducted annually Monitoring the future Study Secondary school students, college students, young adults NIDA Began in 1975; conducted annually DC* MADS Population of District of Columbia Metropolitan Statistical Area, 12 years old and older NIDA Conducted National Pregnancy and Health Survey Pregnant women NIDA Conducted in 1992-1993 Survey of Health Related Behaviors Among Military Personnel Active-duty military personnel DOD Conducted in 1991 Drug Use Forecasting Program Arrestees in 23 metropolitan areas NIJ Began in 1987; conducted quarterly Community Epidemiology Work Group Metropolitan areas NIDA Began in 1976; meets semi-annually Estimates of Diagnoses of Abuse and Dependence     National Comorbidity Survey General population, 15-54 years old NIMH, NIDA, W.T. Grant Foundation Conducted in 1990-1992 Estimates of Consequences of Use     Drug Abuse Warning Network Hospital emergency room visits; death SAMHSA Began in 1973; data reported semiannually NOTE: DC* MADS = Washington, D.C., Metropolitan Area Study; DOD = Department of Defense; NIDA = National Institute on Drug Abuse; NIJ = National Institute of Justice; NIMH = National Institute of Mental Health; SAMHSA = Substance Abuse and Mental Health Services Administration.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research TABLE 4.2 Percentages Reporting Past Month Use of Any Illicit Drug by Age Group and Demographic Characteristics, 1994-Ba   Age Group (years) Demographic Characteristic 12-17 18-25 26-34 =35 Total Total 8.2 13.3 8.5 3.2 6.0 Race/Ethnicity White 8.5 14.6 9.0 3.1 6.0 Black 8.3 12.8 10.2 4.4 7.3 Hispanic 8.1 9.1 5.7 2.9 5.4 Other 2.7 b 3.2 b 3.1 Sex Male 8.5 17.1 11.6 4.3 7.9 Female 7.8 9.6 5.6 2.2 4.3 Population densityc 7.6 14.0 9.1 3.0 6.1 Large metro 10.4 13.5 8.3 3.9 6.6 Small metro Nonmetro 5.7 11.8 7.5 2.4 4.8 Region Northeast 6.8 10.8 7.8 2.7 5.1 North Central 9.2 13.8 8.7 2.8 5.8 South 7.7 14.2 8.5 3.5 6.3 West 8.8 13.6 9.0 3.5 6.6 Adult educationd <High school N/A 15.8 13.3 2.2 5.8 High school graduate N/A 12.8 9.3 3.2 5.9 Some college N/A 14.0 7.7 3.5 6.4 College graduate N/A 9.3 5.8 5.9 4.7 Current employmentd Full-time N/A 12.4 8.6 4.6 6.7 Part-time N/A 12.8 7.5 3.8 6.7 Unemployed N/A 19.9 16.4 8.9 13.9 Othere N/A 13.3 5.8 0.8 2.7 NOTES: "Any illicit drug" is defined as nonmedical use of marijuana or hashish, cocaine (including crack), inhalants, hallucinogens (including LSD [lysergic acid diethlyanide] and PCP [phencyclidine]), heroin or psychotherapeutic at least once. The majority of illicit drug use is marijuana use. N/A = not applicable. a Estimates for 1994-B are derived from the NHSDA new-version questionnaire, b Low precision; no estimate reported. c Population density is based on 1990 metropolitan statistical areas (MSA) classifications and their 1990 census of population counts. d Data on adult education and current employment are not shown for persons aged 12-17. Estimates for both adult education and current employment are for persons aged =18. e Retired, disabled, homemaker, student, or "other." SOURCE: SAMHSA (1995c).

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research tion in 1994 by age and other demographic characteristics (SAMHSA, 1995c).4 There are relatively large variations in the prevalence of illicit drug use by age group and by gender. The 18-25 age group is distinctly higher than other age groups, with the group 35 and older being lowest. Males are almost twice as likely as females to have used an illicit drug in the past month. Among other characteristics there are relatively small variations, with the exception that unemployed persons tend to have higher rates of drug use. There is relatively little variation in use rates by population density, indicating that illicit drug use has permeated the society and is not confined to certain areas. One of the more interesting findings from this study and other population-based epidemiological studies is that African Americans under 25 tend to report rates of drug use that are similar to or lower than those of other racial or ethnic groups. This finding is not consistent with the impressions that many Americans have about rates of drug use, although a number of investigations have supported these findings (Bachman et al., 1991; Anthony et al., 1994; Wallace and Bachman, 1994). However, this may not be generally true for African Americans over 25, who tend to have higher rates of some illicit drug use than other racial or ethnic groups (Table 4.2). Trends in the percentage of positive responses provide an indication of changes in the prevalence of use. As shown in Figure 4.1 for four age groups, the percentage of those who used any illicit drug in the previous month declined steadily from 1979 to 1991 in all age groups except the over-35 group, which consistently reported very low rates of illicit drug use in the past 30 days. In 1979, approximately 14 percent of the total household population reported having used one or more illicit drugs in the past month, compared with 6 percent in 1994 (SAMHSA, 1995c). The decreases were particularly dramatic among those aged 18 to 25, the age group that has historically had the highest rates of use. The decreases were more rapid in the earlier part of this period and have since leveled off, particularly in the two younger age groups. The most recent survey, conducted in 1994, however, shows a significant increase among the youngest respondents (ages 12-17) (SAMHSA, 1995c). Monitoring the Future Study The MTF is a series of surveys funded by NIDA that examines the use of alcohol, tobacco, and illicit drugs among young people from grade 8 through age 32. The core of the study, which began in 1975, consists of 4   The majority of illicit drug use is marijuana use.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research FIGURE 4.1 Trends in Monthly Prevalence of Use of Any Illicit Drug by Age. SOURCE: SAMHSA (1995c). annual surveys of nationally representative samples of eighth-, tenth-, and twelfth-grade students;5 in-school questionnaires are administered by professional interviewers to more than 45,000 students in approximately 420 public and private schools each year (Johnston et al., 1995). Since the MTF study targets students in grades 8, 10, and 12, those who have dropped out of school are not eligible. It is estimated that dropouts represent 15-20 percent of the twelfth-grade cohort; dropout rates are much lower for eighth and tenth grades (Johnston et al., 1995). In addition, absentee students are not included; absentee rates average 10 percent for eighth graders, 13 percent for tenth graders, and 16 percent for twelfth graders. School participation rates have ranged from 58 percent to 80 percent, varying by year and grade level. This series documented a dramatic decrease in marijuana use throughout the decade of the 1980s and a general decline in the use of 5   Beginning in 1991, students in grades 8 and 10 were added to the survey.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research FIGURE 4.2 Trends in annual prevalence of marijuana use (1975-1995) by grade level (Monitoring the Future study, 1995) SOURCE: Johnston et al. (1995). other illicit drugs. More recent surveys have produced evidence of an important reversal in this trend in the 1990s, particularly among the younger students, for use of several illicit drugs, with marijuana being most notable (Johnston et al., 1995). Figure 4.2 shows the trend lines for annual use of marijuana in the three student samples; all three grades show recent increases in marijuana use (Johnston et al., 1995). MTF also provides a national sample of college students and a national sample of young adult high school graduates. Those groups also showed dramatic decreases in marijuana use throughout the 1980s, although their rates have remained just about level or increased slightly in the early 1990s. National Comorbidity Survey NHSDA and MTF provide reasonably accurate epidemiological data on the use of alcohol, tobacco, and illicit drugs among the general population and the trends in those measures. However, they do not include measures of the numbers in the general population whose drug use has

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research progressed to a psychiatric diagnosis of abuse or dependence. The National Comorbidity Survey (NCS) used a psychiatric diagnostic perspective to obtain population-based estimated rates of diagnoses of abuse and dependence.6 NCS was sponsored primarily by the National Institute of Mental Health (NIMH), with supplemental support from NIDA and the W.T. Grant Foundation. NCS, conducted from 1990 to 1992, was a collaborative epidemiological investigation of the prevalence, causes, and consequences of psychiatric morbidity and comorbidity in the United States. Results from this survey of more than 8,000 Americans age 15-54 show that a significant 7.5 percent had developed dependence on illicit drugs or inhalants; 4.2 percent were dependent on cannabis and 2.7 percent were dependent on cocaine (Anthony et al., 1994). One in four (24.1 percent) members of the study population had been or were dependent on tobacco; one in seven (14.1 percent) had a diagnosis of dependence on alcohol. Men were more likely than women to be diagnosed as dependent on illicit drugs. One in eleven men (9.2 percent) reached a (lifetime7) diagnosis of dependence on illicit drugs, compared with 5.9 percent of women. Surveys of Specific Populations Washington, D.C., Metropolitan Area Drug Study The 1991 Washington, D.C., Metropolitan Area Drug Study (DC*MADS), funded by NIDA, examined the nature and extent of drug use among all types of persons residing in a single metropolitan area, with a special focus on groups that are underrepresented or unrepresented in NHSDA (NIDA, 1994b). These special samples included homeless people, transients, and institutionalized individuals. The objectives of DC*MADS were to estimate the prevalence, correlates, and consequences of drug use among the diverse populations residing in the metropolitan area and to develop a research model for similar data collection in other major metropolitan areas (NIDA, 1994b). Additionally, DC*MADS data on homeless and institutionalized populations were analyzed to determine the impact of those results on estimates of the number of injection drug users derived from studies such 6   There have been other important and significant efforts to determine the epidemiology of drug abuse and dependence; the most notable is the Epidemiology Catchment Area studies (Eaton and Kessler, 1985), but these were not nationally representative. 7   The term "lifetime" is used to indicate a disorder occurring at any time during the life span to date and does not indicate a chronic, lifelong condition.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research as NHSDA. DC*MADS data on homeless and institutionalized could be added to NHSDA estimates to determine the degree to which the household survey missed injection drug users. The conclusion was that incorporating DC*MADS data would have increased the NHSDA estimated prevalence of past year needle use from 0.2 to 0.3 percent, a small absolute difference (NIDA, 1994b). The number of needle users8 among the household population was estimated at 5,987; after including homeless, transient, and institutionalized populations, the number was 8,740. The difference is too small to change prevalence estimates noticeably but is a significant difference (an increase of approximately 3,000 needle users) in the population estimates often used by providers for estimating the number of people in need of treatment. As with any survey of drug use, there will be some unknown portion of needle users who either deny use or refuse to participate, resulting in some degree of error when estimating injection drug use. Drug Use Forecasting Program The Drug Use Forecasting Program (DUF), launched in 1987 by the National Institute of Justice (NIJ), collects self-report and urinalysis testing information from selected samples of arrestees brought to booking facilities in 23 cities around the country (NIJ, 1994). DUF data identify the illicit drugs that are being used and how usage changes over time among arrestees. The target audiences for these data are state and local policymakers, court administrators, law enforcement officials, and drug treatment program staff. DUF data are collected for two weeks each quarter by trained individuals who interview booked arrestees and obtain voluntary, anonymous urine specimens (NIJ, 1994). Approximately 225 males are sampled quarterly in booking facilities. At some sites, female arrestees, juvenile arrestees, and detainees are also sampled. Typically, more than 90 percent of the arrestees approached agree to be interviewed, and approximately 80 percent of these provide urine specimens (NIJ, 1994). Selection of male adult offenders ensures a wide distribution across charges, with an emphasis on felony charges; female adults and all juvenile arrestees are included, regardless of charge. The DUF program's major strength is its use of urinalysis to validate the self-reports of recent drug use. This is particularly important given the program's population of arrestees who often underreport their recent drug use. Because this is not a random sample and procedures may vary 8   Needle use is defined as injection of cocaine, hallucinogens, heroin, or psychotherapeutics for nonmedical reasons at least once in the previous 12 months.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research between sites, DUF data are not very useful for tracking prevalence or trends, but they are extremely useful regarding the use of drugs by arrestees. In 1993, the percentage of male booked arrestees testing positive for at least one drug ranged from 54 to 81 percent; there was a similar range for female booked arrestees, from 42 to 83 percent (NIJ, 1994).9 In 21 of the 23 data collection sites, more than half of male and female booked adult arrestees tested positive for a drug at the time of arrest. Department of Defense Survey of Health Related Behaviors Among Military Personnel The Survey of Health Related Behaviors Among Military Personnel was conducted in 1980, 1982, 1985, 1988, 1992, and 1995. The eligible population for the survey consists of all active-duty military personnel except recruits, service academy students, persons absent without official leave (AWOL), and those personnel who had a permanent change of station at the time of the survey (Bray et al., 1995). In 1995 the sample consisted of 16,193 personnel selected to represent military personnel in all pay grades of the active-duty U.S. military throughout the world. Military personnel complete self-administered questionnaires, which include questions on the frequency of use of alcohol, tobacco, or illicit drugs in the past 30 days and within the past 12 months. There are also questions on the negative consequences of drug use and questions on a number of other health behaviors. Trends from this survey show significant reductions in alcohol, tobacco, and illicit drug use since 1980. Use of any illicit drugs declined from 27.6 percent in 1980 to 3.0 percent in 1995. Cigarette smoking in the past 30 days prior to the survey declined from 51.0 percent in 1980 to 31.9 percent in 1995, and heavy drinking declined from 20.8 percent in 1980 to 17.1 percent in 1995 (Bray et al., 1995). Marijuana remained the most commonly used illicit drug: 1.7 percent of military personnel reported marijuana use in the past month and 4.6 percent in the past year. A comparison of results of this survey with the general civilian population NHSDA shows consistently lower rates of illicit drug use among military personnel when demographic variations10 are taken into account (Bray et al., 1995). Reasons for these differences include the military's 9   Urine specimens are sent to a central laboratory and analyzed for ten drugs: cocaine, opiates, marijuana, PCP (phencyclidine), methadone, benzodiazepines, methaqualone, propoxyphene, barbiturates, and amphetamines. 10   The military population is predominantly young males, a population with higher rates of illicit drug use.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research Drug Abuse Warning Network From a public health standpoint, it is particularly pertinent to know the mortality and morbidity statistics associated with drug abuse (Crowley, 1988). One mechanism for assembling and analyzing this data is the Drug Abuse Warning Network (DAWN), conducted annually by the Substance Abuse and Mental Health Services Administration (SAMHSA) to collect information on drug-related emergency department visits and medical examiner reports (SAMHSA, 1995a,b)11 The data collected provide information on the estimated number of episodes in which a visit to a hospital emergency room12 was related directly to the use of an illegal drug or the nonmedical use of a legal drug, or in which a death was drug-related or a drug was mentioned in connection with death. Difficulties in data collection, including changes in sample composition, nonresponse from data collectors, changes in data collectors, and coding errors, place limitations on interpreting DAWN data. Other Epidemiological Data Systems A number of other data systems are relevant to the epidemiology of alcohol, tobacco, and illicit drug use and abuse and provide a variety of additional perspectives. The National Longitudinal Survey of Youth, sponsored by the U.S. Department of Labor, annually collects occupational information and asks about the relationship of alcohol use to occupational activities. The 1991 National Health Interview Survey of Drug and Alcohol Use and the Third National Health and Nutrition Examination Survey (NHANES III), sponsored by the National Center for Health Statistics, provide information on the correlation of alcohol, cigarette, and illicit drug use and abuse with other health conditions. The Youth Risk Behavior Surveillance System was developed by the Centers for Disease Control and Prevention. It includes national school-based surveys of high school students and has been conducted biennially since 1991. The National AIDS Demonstration Research Project and the National Cooperative Agreement for AIDS Community Based Outreach Intervention 11   Although DAWN has existed since the early 1970s, prior to 1988 the sample was not representative of eligible hospitals (i.e., nonfederal, short-stay general hospitals that have a 24-hour emergency department). Statistical adjustments have been made to allow for some comparisons to be made from 1978 through 1987. 12   It is important to note that emergency room data are collected on the number of visits to the emergency room but not collected on the extent to which the visits represent repeated contacts with the same patients. Additionally, each drug abuse episode or visit may have multiple drug mentions.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research Research Program (sponsored by NIDA) use a targeted sample approach to study injection drug users who are not in treatment. There are also a variety of attempts to assess the supply of illicit drugs and the success of efforts to reduce the supply. The data of greatest interest from a public health perspective are those that generate informed estimates of the overall amount of illicit drugs consumed and the amount spent by consumers of illicit drugs. Those data collection efforts include the Drug Enforcement Administration (DEA) System to Retrieve Information from Drug Evidence (STRIDE), which compiles data on illicit drugs purchased, seized, or acquired in DEA investigations and includes information on drug purity, street price, and location of confiscation. RESEARCH OPPORTUNITIES Refinement of Data Systems As discussed above, the current epidemiological data collection surveys have made significant contributions to drug abuse research. However, opportunities exist for continued refinement of the data systems to provide more detailed information and analyses of the nature and extent of drug abuse and dependence. This information is critical to determine the extent of need for treatment and to provide information on trends over time in drug abuse and dependence. As noted above, NHSDA and MTF provide reasonably accurate epidemiological data on alcohol, tobacco, and illicit drug use among the general population, but they are limited in assessing the extent of abuse or dependence. Although both surveys ask questions about frequent use and the resulting consequences of use, neither one allows for diagnosis of DSM (Diagnostic and Statistical Manual of Mental Disorders; APA, 1994) disorders; moreover, neither one achieves full coverage of individuals who are drug abusers or drug dependent. Given the variations in drug use patterns (e.g., binges by cocaine users), it is important to collect additional data on the frequency of use within the span of a day or a week. Increasing the focus of data collection on drug use frequency, consequences of drug use, and other items designed to measure DSM criteria for drug abuse and dependence (e.g., withdrawal and relapse) will provide much needed information on the extent of drug abuse and dependence. Co-Occurring Psychiatric Disorders A major accomplishment of epidemiological research has been the collection and analysis of data on the co-occurrence of drug abuse with

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research other illnesses, particularly psychiatric disorders. The Epidemiologic Catchment Area Program sponsored by NIMH documented the high co-occurrence of alcohol and illicit drug use disorders and other psychiatric disorders (Regier et al., 1990). The NCS (see above) provides the first nationally representative estimates on the co-occurrence of drug abuse and psychiatric disorders. The principal finding is that there is a significant overlap in diagnoses. Roughly half of those age 15-54 who had a lifetime addictive disorder (as defined by DSM-III-R) also had a lifetime psychiatric disorder. Similarly, about half of those with a lifetime psychiatric disorder also had a lifetime addictive disorder. The great majority (84 percent) of those with lifetime co-occurrence reported that their first psychiatric disorder occurred prior to their first addictive disorder (Kessler et al., 1996). Further, the NCS found that all the psychiatric disorders (identified) are consistently more strongly associated with dependence as opposed to abuse. Compared to other affective disorders and anxiety disorders, mania is more strongly associated with dependence on alcohol or illicit drugs. Conduct disorders and adult antisocial behavior were more strongly associated with both abuse and dependence than the anxiety disorders or any of the affective disorders other than mania (Kessler et al., 1996). Those extremely high rates of co-occurrence obviously have great implication for treatment, prevention, and understanding the nature of addictive disorders (Chapters 5 and 8). Although the NCS studied comorbidities with psychiatric disorders, addictive disorders also co-occur with other health and physical disorders. For example, tuberculosis, hepatitis, sexually transmitted diseases, and AIDS are all more prevalent among alcoholics or illicit drug abusers than nonabusers. Other conditions such as homelessness and victimization may also be co-occurring, but await further epidemiological study. As with any epidemiological study, the fact of co-occurrence does not indicate a causal connection. There are a variety of possible causal relationships. An important task for future research is to provide analysis of the possible reasons for co-occurring disorders in an effort to improve treatment and prevention strategies. Natural History of Drug Use Epidemiological research also has contributed to knowledge of the natural history, or life course, of drug use and abuse in the general population, including delineation of stages of involvement with different drugs and degrees of involvement with different types of drugs. Such natural histories require longitudinal studies of individuals over time to discern changes with age; however, there are relatively few studies that span a broad age range. Moreover, it is preferable to utilize a cohort sequential

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research design to isolate classes of probable causes and the time period during which they operate in an individual's life. This research methodology allows for differentiation among age-, history-, or cohort-related factors. There has been one recent report from a single cohort delineating the natural history of drug use from adolescence to the midthirties in a general population sample (Chen and Kandel, 1995). That study was based on a cohort of adolescents (born from 1954 to 1956) in grades 10 and 11 in New York State public schools. An important finding was that there was little initiation into illicit drug use after age 29, the age by which most use had ceased. Among heavier users, the proportions of heavy use declined for alcohol and marijuana, but not for cigarettes. Because this is a study of a single cohort (and therefore confounds historical effects with age effects), it will be important to confirm the findings with data from other ongoing and future studies. Currently more is known about the initiation of drug use than about the transition from use to abuse and dependence. Although investigation of the transition between the stages from use to dependence is primarily in the purview of etiological research (see Chapter 5), large-scale epidemiological studies can provide a fertile ground for generating hypotheses, as well as for testing them. The NCS has found differences in the sociodemographic correlates of first use and dependence among users and in the persistence of abuse and dependence (Warner et al., 1995). The cohort effects seen in the NCS emphasize the need to incorporate designs that allow for differentiation of age, cohort, and historical factors. The natural history of drug use, especially focusing on abuse and dependence, is an area for continued research. If the variables associated with the transition from drug use to abuse and dependence can be identified (and are amenable to manipulation), prevention programs and strategies may be developed that target those variables. Causal Relationships and Societal Norms Epidemiological research has provided an important source of information regarding the etiology of drug use and the social attitudes and norms pertaining to use. Data from the MTF study have been used to demonstrate a close link between certain attitudes or beliefs about drugs and use. At the aggregate level there is a very close association between perceived harm and use, and an extensive series of analyses of data at the individual level supported a causal interpretation of the belief-behavior link (Bachman et al., 1988, 1990). There appears to be clear evidence that, as the perceived risk of harm associated with marijuana use increased, the prevalence of marijuana use decreased. It is important to note that the

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research recent upturn in use in 1993 was accompanied, and even preceded, by a downturn in perceived risk. The MTF study also measures perceived availability of alcohol, cigarettes, or illicit drugs, but analyses of that data did not show changes in perceived availability that could account for the changes in use. Without such data and analyses, the declines in marijuana and cocaine use might have been attributed to putative successes in reducing availability. Thus, the epidemiological data provide a means to generate hypotheses and also to refute hypotheses. Epidemiological data can also be used to address more fundamental issues having to do with ''norms" related to illicit drug use.13 Epidemiological research can be very useful in assessing what the norms are for particular drug-related behaviors and how those norms vary by person, place, and time. Most researchers in the field would emphasize the role of social factors, including broad social-cultural norms, in influencing initiation and experimental use of alcohol, tobacco, or illicit drugs, while intraindividual factors (biological and psychological) would be emphasized in influencing the transition to abuse or dependence (Glantz and Pickens, 1992). Epidemiological research in this field continues to go beyond descriptive documentation of occurrences of a condition according to persons, place, and time by contributing causal analyses of the occurrences. Ethnographic Research The patterns, prevalence, and consequences of alcohol, tobacco, and illicit drug use and abuse within youth populations are not fully understood. As mentioned above, young African Americans have significantly lower levels of cigarette and illicit drug use than most of their peers. The 1993 MTF study found that 4.9 percent of African American high school seniors reported daily cigarette use, compared with 22.9 percent of white students (Johnston et al., 1995). The reasons for those differences are not yet understood, and it would be useful to elucidate the reasons for low use among African Americans to determine if that information could be transferred across cultural lines. Further, ethnographic research could provide important data on local-level illicit drug use that could be used to identify potential trends; study the natural history and progression of alcohol, tobacco, and illicit drug use; and provide greater detail than more quantitative surveys. Studying the problem of illicit drug use and abuse in a variety of 13   The term "norm" is used to refer to both the prescriptive sense (i.e., guidelines for behavior considered to be acceptable) and the statistical sense (i.e., behavior that is approved or practiced by a majority).

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research cultures would be useful from a number of perspectives. Comparing prevalences of illicit drug use and abuse across cultures may provide information on the universality or generalizability of risk and protective factors. It may also be easier to study the role of specific factors, such as drug availability, by comparing countries such as Colombia (where cocaine is more readily available) with Asian countries (where heroin is more widely available). Additionally, cultural differences offer the opportunity to learn more about the nature and consequences of specific histories (e.g., by comparing U.S. drug abusers, many of whom are multiple drug abusers, with cultures such as China, where it is easier to find individuals who are heavy users of only one illicit drug). Measurement and Analysis Research Design and Data Analysis Research designs and analytical procedures developed for epidemiological research on drug use have further advanced the field of epidemiology in general. Epidemiological data are often collected under conditions of less than complete control (e.g., missing data, attrition, errors of measurement, including deliberately inaccurate reporting), which contribute to analytic problems. However, various research designs and data analysis methods have been implemented to improve the confidence in epidemiological reporting. For example, a structural equation computer program was developed (Bentler and Wu, 1993) from theoretical statistics, algorithmic experiments, and applied data analysis; multilevel analysis, missing data analysis, and meta-analysis have been developed or improved (Collins et al., 1994). Improved analytical techniques and increasingly sophisticated models will provide more focused and accurate data. Traditional epidemiological survival analysis methods could elucidate aspects of timing of onset of abuse, time to cessation, and time to relapse (Singer and Willett, 1994). Latent growth curve modeling (Duncan and Duncan, 1995), latent class analysis (Uebersax, 1994), and latent transition analysis (Collins et al., 1994) are currently areas of promise for refinement in data analysis. Strategies and software for dealing with missing data are developing at a rapid pace (Graham et al., 1994). A particularly active area is the extension of structural equation methods to accommodate multilevel models; models that do not require traditional assumptions of additivity or linearity; and models that do not require variables that are normally distributed, continuous, and independently measured (e.g., Muthén, 1993). Thus, by using the same level of data collection, it will now be possible to perform increasingly sophisticated analyses that provide additional in-

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research formation regarding alcohol, tobacco, and illicit drug use and abuse problems. Furthermore, it is critical that future research coordinate both quantitative (e.g., probability-based surveys) and qualitative (trained-observer or ethnographic) approaches in an effort to maximize both their unique contributions and their synergistic potential (see section above). Optimizing the use of both approaches will provide a fuller picture of the nature and extent of alcohol, tobacco, and illicit drug use, abuse, and dependence. Self-Reports and Biological Indicators Self-report methods have been shown to be generally reliable and valid when gathered under proper conditions. Such conditions include clear and understandable interview procedures and questionnaires, confidence by the respondent that responses will be kept confidential, and some degree of willingness by the respondent to provide accurate information. However, under some conditions, self-reports are likely to be far less valid. For example, research has shown that higher rates of illicit drug use are found with the use of self-administered questionnaires instead of direct reporting to the interviewer (Turner et al., 1992) and with the use of in-school questionnaires as compared to in-home interviews of adolescents (Rootman and Smart, 1985). Other examples include questioning arrestees or pregnant women about recent drug use. Thus, biological indicators are useful tools to support the validity of self-reports. It would be useful to know the degree to which standard survey techniques underrepresent actual use. Currently, there are a number of efforts to improve the reliability and validity of self-reported data. The use of computers in collecting data is being explored, including the development of computer-assisted self-administered instruments (CASI) that may produce improved reports of alcohol, tobacco, or illicit drug use. Audio-CASI would incorporate an audio component and would allow minimally literate respondents to hear the questions and respond directly to the computer (e.g., verbally, via keypad or touch-screen); this technique would also facilitate the use of multiple languages. Such procedures need to be researched carefully so that there is an assessment of benefits (improved reliability and validity), costs, and impacts on privacy and confidentiality. Another line of selfreport research is the use of principles of cognitive psychology to inform survey methodology (Turner et al., 1992). Those principles include recognition that memory is essentially a reconstructive process, awareness that respondents employ different strategies for reconstructing needed information, and adoption of techniques such as "bounding" to improve

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research memory reconstruction. Substantial changes were made in the 1994 NHSDA as a result of applying those principles. Biological indicators are useful tools to support the validity of self-reports. Urinalysis testing can now detect with a very high degree of accuracy if illicit drugs have been used recently. Institution of random testing of urine samples is likely responsible, in part, for the dramatic decline in use of illicit drugs in the U.S. Armed Forces, indicating that drug testing can be useful both for epidemiological surveillance and for deterrence (Bray et al., 1992). Saliva, sweat patches, and expired breath can also be used to detect a range of licit and illicit drugs. Although biological indicators are generally considered to be more valid than self-reports, they are more complicated and more expensive to implement. Additionally, for drugs that are rapidly metabolized, they may detect use only within a limited time frame. To increase the validity of reports of alcohol, tobacco, and illicit drug use, continued investigation of biological markers will be important. Saliva and hair samples are more easily obtained in field studies than urine or blood samples. Drug testing by hair analysis is a particularly active area of research; NIDA and NIJ have recently published the results of a series of collaborative efforts on hair testing financed under an interagency agreement (Cone et al., 1995). CONCLUSION AND RECOMMENDATION Significant progress has been made in collecting and analyzing data on the extent, incidence, prevalence, and trends of alcohol, tobacco, and illicit drug use. Major data systems are in place that allow incipient changes in use to be detected and provide information for policymakers to prioritize prevention and treatment efforts. However, although use has been studied extensively, it has been more difficult to determine the nature and extent of abuse or dependence. Additional epidemiological research should focus on collecting and analyzing data on the nature and extent of drug abuse and dependence, drug use patterns, co-occurring drug abuse and psychiatric disorders, and refinement of measurement and analytical tools. Continued and refined epidemiological research will provide the data and analyses necessary for the development of treatment and prevention programs and will inform decisions on future allocation of resources to best address the alcohol, tobacco, and illicit drug abuse problem. The committee recommends continued epidemiological research to allow for the assessment of a broader range of issues. Those issues may include the extent of drug abuse and dependence; the nature and extent of drug use and abuse among youth; the nature and

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research extent of co-occurring drug abuse and psychiatric disorders; and improvement in the reliability and validity of the methods for collecting and analyzing the data. REFERENCES Anthony JC, Warner LA, Kessler RC. 1994. Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: Basic findings from the National Comorbidity Study. Experimental and Clinical Psychopharmacology 2:244-268. APA (American Psychiatric Association). 1994. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: APA. Bachman JG, Johnston LD, O'Malley PM, Humphrey RH. 1988. Explaining the recent decline in marijuana use: Differentiating the effects of perceived risks, disapproval, and general lifestyle factors. Journal of Health and Social Behavior 29:92-112. Bachman JG, Johnston LD, O'Malley PM. 1990. Explaining the recent decline in cocaine use among young adults: Further evidence that perceived risks and disapproval lead to reduced drug use. Journal of Health and Social Behavior 31:173-184. Bachman JG, Wallace JM Jr., O'Malley PM, Johnston LD, Kurth CL, Neighbors HW . 1991. Racial/ethnic differences in smoking, drinking, and illicit drug use among American high school seniors, 1976-1989. American Journal of Public Health 81:372-377. Bentler PB, Wu EJC. 1993. EQS/Windows Users Guide. Los Angeles: BMDP Statistical Software. Bray RM, Kroutil LA, Luckey JW, Wheeless SC, lannacchione VG, Anderson DW, Marsden ME, Dunteman GH. 1992. 1992 Worldwide Survey of Substance Abuse and Health Behaviors Among Military Personnel. RTI/5154/06-16FR. Research Triangle Park, NC: Research Triangle Institute. Bray RM, Kroutil LA, Wheeless SC, Marsden ME, Bailey SL, Fairbank JA, Harford TC. 1995. 1995 Department of Defense Survey of Health Related Behaviors Among Military Personnel. RTI/6019/06FR. Research Triangle Park, NC: Research Triangle Institute. Chen K, Kandel DB. 1995. The natural history of drug use from adolescence to the midthirties in a general population sample. American Journal of Public Health 85:41-47. Collins LM, Graham JW, Rousculp SS, Fidler PL, Pan J, Hansen WB. 1994. Latent transition analysis and how it can address prevention research questions. NIDA Research Monograph 142:81-111. Cone EJ, Welch MJ, Babecki MBG, eds. 1995. Hair Testing for Drugs of Abuse: International Research on Standards and Technology. Rockville MD: National Institute on Drug Abuse. NIH Publication No. 95-3727. Crowley TJ. 1988. Learning and unlearning drug abuse in the real world: Clinical treatment and public policy. NIDA Research Monograph 84:100-121. de Alarcon R. 1969. The spread of heroin abuse in a community. Bulletin on Narcotics 21:1722. Duncan TE, Duncan SC. 1995. Modeling the processes of development via latent growth curve methodology. Structural Equation Modeling 2:187-213. Eaton WW, Kessler LG, eds. 1985. Epidemiologic Field Methods in Psychiatry: The NIMH Epidemiologic Catchment Area Program. Orlando FL: Academic Press. Glantz MD, Pickens RW, eds. 1992. Vulnerability to Drug Abuse. Washington, DC: American Psychological Association. Graham JW, Hofer SM, Piccinin AM. 1994. Analysis with missing data in drug prevention research. NIDA Research Monograph 142:13-63.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research Hughes PH, Crawford GA. 1972. A contagious disease model for researching and intervening in heroin epidemics. Archives of General Psychiatry 27:149-155. Johnston LD, O'Malley PM, Bachman JG. 1995. National Survey Results on Drug Use from the Monitoring the Future Study, 1975-1994. Vol. 1, Secondary School Students. Ann Arbor, MI: Institute for Social Research, University of Michigan. Kessler RC, Nelson CB, McGonagle KA, Edlund MJ, Frank RG, Leaf PJ. 1996. The epidemiology of co-occurring addictive and mental disorders in the National Comorbidity Survey: Implications for prevention and service utilization. American Journal of Orthopsychiatry 66(1):17-31. Kozel NJ, Adams EH. 1986. Epidemiology of drug abuse: An overview. Science 234:970-974. Last JM, ed. 1983. A Dictionary of Epidemiology. New York: Oxford University Press. Lillienfeld DE, Stolley PD. 1994. Foundations of Epidemiology. 3rd ed. New York: Oxford University Press. Muthén B. 1993. Latent variable modeling of growth with missing data and multilevel data. In: Cudras CM, Rao CR, eds. Multivariate Analysis: Future Directions. New York: North Holland Publishing. NIDA (National Institute on Drug Abuse). 1976. Epidemiology of Drug Abuse: Current Issues. NIDA Research Monograph 10. Rockville, MD: NIDA. NIDA (National Institute on Drug Abuse). 1994a. 1995 Budget Estimate. Rockville, MD: NIDA. NIDA (National Institute on Drug Abuse). 1994b. Prevalence of Drug Use in the DC Metropolitan Area Household and Nonhousehold Populations: 1991. Technical Report 8. Rockville, MD: NIDA. NIDA (National Institute on Drug Abuse). 1995. Community Epidemiology Work Group, December 1994 Report. Vol. I: Highlights and Executive Summary. Rockville, MD: NIDA. NIDA (National Institute on Drug Abuse). 1996. National Pregnancy and Health Survey: Drug Use Among Women Delivering Livebirths, 1992 . NIH Publication No. 96-3819. Rockville, MD: NIDA. NIJ (National Institute of Justice). 1994. Drug Use Forecasting 1993 Annual Report on Adult Arrestees. NCJ 147411. Washington, DC: NIJ. NRC (National Research Council). 1995. Preventing HIV Transmission: The Role of Sterile Needles and Bleach. Washington, DC: National Academy Press. ONDCP (Office of National Drug Control Policy). 1995. Pulse Check: National Trends in Drug Abuse. Washington, DC: ONDCP. Regier DA, Farmer ME, Rae DS, Locke BZ, Keith SJ, Judd LL, Goodwin FK. 1990. Comorbidity of mental disorders with alcohol and other drug abuse. Results from the Epidemiologic Catchment Area Study. Journal of the American Medical Association, 264(19):2511-2518. Rogers FB, ed. 1965. Studies in Epidemiology. New York: Putnam. Rootman I, Smart RG. 1985. A comparison of alcohol, tobacco, and drug use as determined from household and school surveys. Drug and Alcohol Dependence 16:89-94. SAMHSA (Substance Abuse and Mental Health Services Administration). 1995a. Annual Medical Examiner Data, 1993. Data from the Drug Abuse Warning Network . Statistical Series I, Number 13-B. Rockville, MD: SAMHSA. SAMHSA (Substance Abuse and Mental Health Services Administration). 1995b. Preliminary Estimates from the Drug Abuse Warning Network: 1994 Preliminary Estimates of Drug-Related Emergency Department Episodes. Advance Report Number 11. Rockville, MD: SAMHSA. SAMHSA (Substance Abuse and Mental Health Services Administration). 1995c. National Household Survey on Drug Abuse: Population Estimates 1994. Rockville, MD: SAMHSA.

OCR for page 94
Pathways of Addiction: Opportunities in Drug Abuse Research Singer JD, Willett JB. 1994. Designing and analyzing studies of onset, cessation, and relapse: Using survival analysis in drug abuse prevention research. NIDA Research Monograph 142:196-263. Turner CF, Lessler JT, Gfroerer, JC. 1992. Future directions for research and practice. In: Turner CF, Lessler JT, Gfroerer JC, eds. Survey Measurement of Drug Use: Methodological Studies. Rockville, MD: NIDA. Uebersax JS. 1994. Latent class analysis of substance abuse patterns. NIDA Research Monograph 142:64-80. Wallace JM Jr., Bachman JG. 1994. Validity of self-reports in student-based studies on minority populations: Issues and concerns. NIDA Research Monograph 130:167-200. Warner LA, Kessler RC, Hughes M, Anthony JC, Nelson CB. 1995. Prevalence and correlates of drug use and dependence in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry 52:219-229.