Substance Abuse / Illegal Drug Abuse Problems

Paper presented at 37th International Conference on Reduction of Drug Related Harm,

Introduction

Harm to individuals and to society resulting from the use of drugs can take many forms. Drug overdoses, accidents resulting from intoxication, and violence associated with involvement in the drug trade are clear examples of ways that drug use causes harm to individuals (SAMHSA 1994a, SAMHSA 1994b, NHTSA 1992, Maguire et al., 1993). Society overall is harmed by reduced productivity of students and employees impaired by drugs and by the substantial drug-related costs involved with the criminal justice and health care systems (Rice, 1990).

A number of descriptive studies have demonstrated that people who use drugs are more likely to have mental disorders, physical health problems, and family problems (NIDA 1991, Schober and Schade, 1991). However, it is not always clear from these studies that the drug use actually caused or was even strongly associated with the problem.

In this paper, the relationships between drug use and various measurements of harm are shown. For this analysis, harm is defined as harm to the individual using the drugs (in terms of adverse psychosocial and health consequences resulting from drug use), reduced productivity (as a result of missed or skipped days from work), and potential harm to society (as measured by driving under the influence of alcohol or drugs and having had an automobile accident. The analysis is based on data from the National Household Survey on Drug Abuse (NHSDA). This paper is focused primarily on the frequency of use of drugs and its relationship with the reported problems, including problems that respondents report as being caused by their drug use.

Methods

The National Household Survey on Drug Abuse (NHSDA), conducted by the Substance Abuse and Mental Health Services Administration (or SAMHSA), provides estimates of the prevalence, consequences, and patterns of drug use and abuse in the United States. It is the primary source of statistical information on the use of illegal drugs by the United States population age 12 and older. Conducted periodically by the Federal Government since 1971, the survey collects data by administering questionnaires to a representative sample of persons living in the Nation (SAMHSA, 1993).

The respondent universe includes residents of noninstitutional group quarters such as shelters, rooming houses, dormitories and residents of civilian housing on military bases. Persons excluded from the universe include the homeless not found in shelters, residents of institutional quarters, such as jails and hospitals, and active military personnel. The survey employs a multistage area probability sample design that includes oversampling of young people, African-Americans, and Hispanics. Sample households are screened in a 5-minute interview to roster all household members aged 12 and older. Two, one, or no household members are then selected tobe interviewed using a random sampling procedure.

The household interview takes about an hour to complete, and includes a combination of interviewer-administered and self-administered questions. With this procedure, the answers to sensitive questions (such as those on illicit drug use) are recorded on separate answer sheets by the respondent and are not seen or reviewed by the interviewer. After the answer sheets are completed, they are placed by the respondent in an envelope, which is sealed and mailed with no name or address information included.

A concern of NHSDA data users is that the data are based on self-reports of drug use, and their value depends on respondents' truthfulness and memory. Although many studies have generally established the validity of self-report data and the NHSDA procedures were designed to encourage honesty and recall, some underreporting may have taken place (Harrell et al., 1986). The methodology used in the NHSDA has been shown to produce more valid results than other self-report methods such as interviews by telephone (Turner et al., 1992). However, comparisons of NHSDA data with data from surveys conducted in classrooms suggest that underreporting of drug use by youths in their homes may be substantial (Gfroerer, 1993).

For this study, data from the 1991 and 1992 NHSDA data sets combined were used. NHSDA analytic weights were divided by 2 to produce average annual yearly estimates for the combined 1991-1992 data set. Questionnaires and data collection and estimation methodologies were essentially the same in those two years. The 1991 NHSDA included a sample of 32,594 respondents. Of selected households, 97 percent completed the screening form, while 84 percent of sampled persons completed the interview. The 1992 screening and interview response rates were 95 percent and 83 percent, respectively, for a sample size of 28,832 respondents. Thus, the combined 1991-92 data file included 61,426 respondents. Of these, the analysis focused primarily on the 7,541 respondents who reported using marijuana and the 2,340 respondents who reported using cocaine in the past 12 months. This represents an estimated annual average of 15 million past year marijuana users and 4 million past year cocaine users.

For several measures of harm, the bivariate relationship between levels of marijuana and cocaine use and the prevalence of problems was studied. The first set of harm measures is a series of social, psychological and health problems that were reported by respondents as being caused by their use of drugs. Other harm measures were health status indicators, problems at work, and driving behaviors. After examining each of these bivariate relationships, logistic regression models were constructed which tested if the relationship between the harm measure and the drug use measure was significant after controlling for possible confounding factors. For most regression modes, the primary independent variable of interest was frequency of drug use during the past year. This variable was scaled so that the maximum value was 1, representing daily use of the drug, and the minimum value was 0, representing no use of the drug. For models of days missed or skipped from work during the past month, the primary independent variable of interest was the dichotomy of use in the past month (1=use, 0=no use).

Possible confounding factors that were included as independent variables in the regression models were age, race, sex, education, and marital status. Additional controls for past year cigarette smoking status and number of years smoked daily were included in models for healthstatus and days missed or skipped from work. In logistic regression models for driving under the influence of alcohol and drugs and for having had an automobile accident, the number of days drunk in the past year was included.

Test statistics were adjusted to account for the complex sample design of the NHSDA through the use of the SUDAAN software. The resulting adjusted odds ratios for drug use estimate the effects associated with the use of marijuana and cocaine after controlling for the other factors in the model.

Characteristics of Past Year Marijuana and Cocaine Users

Respondents were asked how often in the past 12 months they used marijuana and cocaine (Table 1). About half of the marijuana users (48%) and 70 percent of cocaine users used the drug every other month or so or less often. 19 percent of marijuana users were daily or almost daily users, as were 6 percent of cocaine users. Table 2 shows the demographic characteristics of marijuana and cocaine users by frequency of use.

Self-Reported Problems Associated with Drug Use

The survey included a set of questions concerning social, psychological, and medical problems that could be related to the use of drugs. Respondents were asked, "As a result of drug use at any time in your life, did you in the past 12 months [for example] become depressed or lose interest in things..." If they answered "yes" to any problem, they were subsequently asked the name of the drug(s) that caused the problem. This analysis focuses on problems that respondents said were caused by the use of marijuana or cocaine. Of past year marijuana users, fifteen percent said they had at least one of the problems shown as a result of their marijuana use; eight percent reported two or more problems. Of those who used cocaine in the past year, nineteen percent reported they had at least one of the problems as a result of their cocaine use; fourteen percent reported two or more problems.

The rate of having one or more problem was highly correlated to frequency of use. One-third of daily/almost daily marijuana users said they experienced one or more of the 11 problems, and the rate of having problems declined as frequency of use declined. Two-thirds of daily/almost daily cocaine users said they experienced one or more of the 11 problems due to their use of cocaine and, as with marijuana, the rate of having problems generally declined as the frequency of use declined.

This relationship was examined in more detail by looking at rates for each drug-related problem by past year frequency of use (Table 3). In general, as the use of marijuana or cocaine became less frequent, the rate of reporting problems fell.

The daily/almost daily users of marijuana were more likely than infrequent marijuana users to say they found it difficult to think clearly, they became depressed or lost interest in things and they got less work done at school or on the job. Daily/almost daily marijuana users were also more likely to say they felt suspicious and mistrustful of people.

For all 11 of the problems analyzed, daily/almost daily cocaine users were more likely than infrequent users to report the problem. About half of the daily/almost daily cocaine users said that they found it harder to handle their problems and difficult to think clearly and felt suspicious and mistrustful of people due to their cocaine use.

The relationships between frequency of use and problems were found to be significant in logistic regression models that controlled for possible confounding factors (Table 4). For problems due to marijuana use, odds ratios for frequency of marijuana use ranged from 2.4 to 9.8. Odds ratios indicated that the likelihood of reporting a problem was 6.3 to 39.3 times as great for the most frequent users of cocaine than it was for the least frequent users.

Health Status

NHSDA respondents reported whether their overall health during the past year was excellent, very good, good, fair, or poor. Cigarette smokers and nonsmokers were analyzed separately for persons aged 18 to 49 (Table 5). About two-thirds of marijuana and cocaine users were cigarette smokers. Marijuana users who did not smoke cigarettes were about equally likely as non-users to describe their health as poorer, regardless of frequency of use. Among cigarette smokers, daily/almost daily marijuana users and cocaine users were more likely to describe their health as poorer compared with those who used marijuana every other month or less often.

Among cocaine users, the non-smoking infrequent cocaine users were the least likely to report poorer health. Logistic regression models showed that among cigarette smokers, frequency of use was a significant predictor of health status and indicated that more frequent marijuana users and cocaine users were more likely to report their health as poorer than less frequent users (Table 6). Among cigarette non-smokers, there was no clear relationship between self-perception of health and frequency of marijuana and cocaine use.

Days Lost From Work

A measure of harm that can be related to drug use is days lost from work. Because work loss days are reported for the past month in the NHSDA, past month marijuana and cocaine use (any use vs. nonuse) were used as drug use measures for this analysis which was restricted to 18-49 year olds who were employed full-time.

Among cigarette smokers and nonsmokers, marijuana users were more likely to have missed one or more days from work due to illness or injury and to have skipped 1 or more days from work just because they did not want to be there than non-users of marijuana. Results were the same for comparisons of cocaine users and non-users (Table 7).

The logistic regression model showed that the relationships between past month cocaine use or marijuana use and having missed days of work due to illness or injury were significant only for cigarette smokers. Similarly, among cigarette smokers, there were significant relationships between past month cocaine use or marijuana use and having skipped one or more days of work. Among cigarette non-smokers, there were also significant relationships between past month cocaine use or marijuana use and having skipped 1 or more days of work.

Driving Behaviors

Another measure of harm to oneself and possible harm to others is driving under the influence of alcohol or other drugs. NHSDA respondents were asked if they drove under the influence of alcohol or other drugs at least once in the past year. Therefore, for respondents who said they drove under the influence, it is known from alcohol and drug questions which substances they used in the past year. However, if more than one substance was used, it is not known specifically which substance the respondent used prior to driving under the influence.

This analysis was restricted to marijuana or cocaine users between the ages of 16 and 49 who reported driving a motor vehicle in the past year. Forty-four percent of marijuana users and 58 percent of cocaine users said they drove under the influence of alcohol or other drugs in the past year compared with only 9 percent of persons who did not use marijuana and 13 percent of those who did not use cocaine (Table 8).

Logistic regression models showed positive relationships between frequency of past year cocaine use or marijuana use and driving under the influence of alcohol or other drugs. These models included a control for the number of times a person was drunk in the past year.

Respondents were asked about their involvement in a motor vehicle accident in the past year. The responses for persons between the ages of 16 and 49 were analyzed comparing those who used marijuana or cocaine with nonusers (Table 9). Among 16-20 year olds, those who used marijuana or cocaine were more likely to have had a motor vehicle accident than non-users.

In the regression model, frequency of marijuana or cocaine use was not significantly associated with having had a motor vehicle accident. To control for the effect of heavy alcohol use, a variable showing the number of times a persons was drunk in the past year was included in the model.

Conclusions

This analysis shows that levels of marijuana and cocaine use are strongly associated with rates of several measures of harm. In this analysis, harm was defined as physical, mental, or social problems resulting from drug use, and harm to society in terms of reduced productivity and automobile accidents. For some social, psychological and physical problem measures, respondents directly reported that their drug use was the cause of these problems. Furthermore, rates of these reported problems increase as the frequencies of marijuana and cocaine use increase. All of these relationships were significant in regression models that controlled for confounding factors.

For other measures of harm such as self-perceived health status, days missed from work, and driving under the influence of alcohol or other drugs, some significant relationships were found between these measures of harm and drug use, even though respondents did not directly report that the problems were caused by their drug use.

For health status and absences from work, it cannot be concluded that drug use was a cause of the harm. However, the importance of frequency of use in the regression models does suggest that drug use may be directly linked to health status and absences from work, particularly among cigarette smokers.

The implication of these results is that a reduction in the frequency of drug use could result in a reduction in many of the problems cited. Another implication of this analysis is that smoking cigarettes and frequent use of marijuana or cocaine would seem to exacerbate ill affects on health.

References

Gfroerer, J. (1993). An Overview of the National Household Survey on Drug Abuse and Related Methodological Research. Proceedings of the Survey Research Section of the American Statistical Association, Joint Statistical Meetings, Boston, Massachusetts, August, 1992. American Statistical Association, 1993.

Harrell, A.V., Kapsak, K.A., Cisin, I.H., and Wirtz, P.W. (1986). The Validity of Self-Reported Drug Use Data: The Accuracy of Responses on Confidential Self-Administered Answer Sheets. Prepared for the National Institute on Drug Abuse, Contract Number 271-85-8305.

Maguire, K., Pastore, A.L., and Flanagan, T.J., eds. (1993) Sourcebook of Criminal Justice Statistics 1992, U.S. Department of Justice, Bureau of Justice Statistics, Washington, DC: USGPO.

National Highway Traffic Safety Administration (1992). The Incidence and Role of Drugs in Fatally Injured Drivers. Final Report September 1988 - October 1992. DOT HS 808 065.

National Institute on Drug Abuse (1991). Third Triennial Report to Congress on Drug Abuse and Drug Abuse Research (1991), DHHS Pub. No. (ADM) 91-1704. Washington, DC: Supt. of Docs., U.S. Govt. Print. Off.

Rice, D.P., Kelman S., Miller L.S., Dunmeyer, L., (1990) The Economic costs of alcohol and drug abuse and mental illness: 1985. Rockville, MD: Alcohol, Drug Abuse and Mental Health Administration.

Schober, S., and Schade, C., eds. (1991) The Epidemiology of Cocaine Use and Abuse, Research Monograph 110, 1991. DHHS Pub. No. (ADM) 91-1787. Washington, DC: Supt. of Docs., U.S. Govt. Print. Off.

Substance Abuse and Mental Health Services Administration (1993). National Household Survey on Drug Abuse: Main Findings 1991. DHHS Pub. No. (SMA) 93 1980, Washington, DC: Supt. of Docs., U.S. Govt. Print. Off.

Substance Abuse and Mental Health Services Administration (1994a). Advance Report 6. Preliminary Estimates from the Drug Abuse Warning Network. Office of Applied Studies, March 1994.

Substance Abuse and Mental Health Services Administration (1994b). Annual Emergency Room Data 1992. DHHS Pub. No. (SMA) 94-2080, Washington, DC: Supt. of Docs., U.S. Govt. Print. Off.

Turner, C.F., Lessler, J.T., and Gfroerer, J.C. (1992). Survey Measurement of Drug Use: Methodological Studies. National Institute on Drug Abuse. DHHS Pub. No. (ADM) 92-1929.

Table 1:Frequency of Use Among Past Year Marijuana and Cocaine Users, 1991-92

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse.

Table 2:Demographic Characteristics of Past Year Marijuana and Cocaine Users by Frequency of Use, 1991-1992

Table 2 (continued):

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

Table 3:Percentage of Past Year Marijuana and Cocaine Users Reporting Problems in the Past Year Caused by Their Marijuana or Cocaine Use, by Frequency of Use: 1991-1992

Table 3 (continued):

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

Table 4:Adjusted Odds Ratios and 95% Confidence Intervals for frequency of Marijuana and Cocaine Use as a Predictor of Reported Problems, 1991-1992

Note:The Adjusted Odds Ratios were computed using logistic regression models that controlled for the effects of age, race, sex, education, and marital status. The Odds Ratio approximates how much more likely it is someone who used marijuana or cocaine daily to have reported problems than it is for someone who did not use these drugs at all.

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

Table 5:Percentage Reporting Health in Past Year as Below "Very Good" Among Past Year Marijuana and Cocaine Users and Non-users Age 18-49 by Cigarette Smoking Status and Frequency of Use of Marijuana and Cocaine, 1991-1992

Table 6:Adjusted Odds Ratios and 95% Confidence Intervals for the Frequency of Cocaine or Marijuana Use as a Predictor of Reporting Health Status as Below "Very Good" Among Past Year Users of Marijuana and Cocaine Age 18-49 by Cigarette Smoking Status, 1991-1992

Adjusted Odds Ratio For Frequency of Drug Use

Frequency of Use

Cigarette Smokers

Cigarette Non-Smokers

Odds Ratio

95% C.I.

Odds Ratio

95% C.I.

Marijuana Use

Cocaine Use

1.5

2.5

(1.0 - 2.2)

(1.1 - 6.1)

1.0

4.5

(0.5 - 1.8)

(0.5 - 37.2)

Note:The adjusted Odds Ratios were computed separately for cigarette smokers and non-smokers using logistic regression models that included frequency of drug use, age, race, sex, education, marital status, and number of years smoked daily. The Odds Ratio approximates how much more likely it is for someone who uses marijuana or cocaine daily to have reported their health as below "very good" than it is for someone who does not use these drugs at all.

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

Table 7:Percentage of Full-Time Employed Persons Age 18-49 Who Reported Missing or Skipping at Least One Day From Work in Past Month by Past Month Cigarette Smoking Status and Past Month Use of Marijuana and Cocaine, 1991-1992

Table 7 (continued):

Adjusted Odds Ratios and 95% Confidence Intervals for Past Month Marijuana and Cocaine Use as a Predictor of Skipping or Missing Work in Past Month, 1991-1992

Adjusted Odds Ratios for Past Month Use

Cigarette Smoker

Cigarette Non-Smoker

Odds Ratio

95% C.I.

Odds Ratio

95% C.I.

Missed 1 or more days from work due to injury/illness

Marijuana Use

Cocaine Use

Skipped 1 or more days from work

Marijuana Use

Cocaine Use

1.6

1.8

2.0

2.8

(1.2 - 2.1)

(1.1 - 2.9)

(1.5 - 2.7)

(1.8 - 4.4)

1.3

1.4

2.7

2.4

(0.9 - 1.0)

(0.7 - 2.7)

(1.7 - 4.4)

(1.2 - 4.8)

Note:The Adjusted Odds Ratios were computed separately for cigarette smokers and non-smokers using logistic regression models that controlled for the effects of age, race, sex, education, marital status, and number of years smoked daily. The Adjusted Odds Ratio approximates how much more likely it is for past month marijuana or cocaine users to have reported they missed or skipped at least one day from work than it is for someone who did not use these drugs at all.

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

Table 8:Percentage of Persons age 16 to 49 who Reported Driving Under the Influence of Alcohol or Drugs in Past Year by Past Year Marijuana and Cocaine Use, 1991-1992

Percentage of Persons Age 18 to 49 Who Reported Having Had a Motor Vehicle Accident in Past Year by Past Year Marijuana and Cocaine Use, 1991-1992

Past Year Drug Use

Marijuana

Cocaine

Used

Did Not Use

Used

Did Not Use

Age

16-20

21-34

35-49

Total

26.1%

14.1

10.1

15.7

17.6%

11.2

8.2

10.4

32.9%

15.5

6.9

15.7

19.0%

11.5

8.3

10.9

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

Table 9:Adjusted Odds Ratios and 95% Confidence Intervals for Frequency of Cocaine or Marijuana Use as a Predictor of Reporting Having Driven Under the Influence of Alcohol or Other Drugs or Having Had a Motor Vehicle Accident in the Past Year among Past Year Users of Marijuana and Cocaine Age 16-49, 1991-1992

Note:The Odds Ratios were computed using logistic regression models that included past year drug use, age, race, sex, education, marital status, and number of times drunk in the past year. The Odds Ratio approximates how much more likely it is for someone who used cocaine or marijuana daily to have reported they drove under the influence of alcohol or other drugs or had a motor vehicle accident than it was for someone who did not use these drugs at all.

Source:Office of Applied Studies, SAMHSA, National Household Survey on Drug Abuse

SAMHSA, an agency in the Department of Health and Human Services, is the Federal Government's lead agency for improving the quality and availability of substance abuse prevention, addiction treatment, and mental health services in the United States.

* Adobe™ PDF and MS Office™ formatted files require software viewer programs to properly read them. Click here to download these FREE programs now

Part III

Health and Other Problems Associated with Drug Use

5. Self-Reported Problems Associated with Drug Use

Introduction

Methods

Characteristics of Past Year Marijuana and Cocaine Users

Self-Reported Problems Associated with Drug Use

Health Status

Days Lost From Work

Driving Behaviors

Conclusions

References

Table 1:Frequency of Use Among Past Year Marijuana and Cocaine Users, 1991-92

Table 2:Demographic Characteristics of Past Year Marijuana and Cocaine Users by Frequency of Use, 1991-1992

Table 2 (continued):

Table 3:Percentage of Past Year Marijuana and Cocaine Users Reporting Problems in the Past Year Caused by Their Marijuana or Cocaine Use, by Frequency of Use: 1991-1992

Table 3 (continued):

Table 4:Adjusted Odds Ratios and 95% Confidence Intervals for frequency of Marijuana and Cocaine Use as a Predictor of Reported Problems, 1991-1992

Table 5:Percentage Reporting Health in Past Year as Below "Very Good" Among Past Year Marijuana and Cocaine Users and Non-users Age 18-49 by Cigarette Smoking Status and Frequency of Use of Marijuana and Cocaine, 1991-1992

Table 7:Percentage of Full-Time Employed Persons Age 18-49 Who Reported Missing or Skipping at Least One Day From Work in Past Month by Past Month Cigarette Smoking Status and Past Month Use of Marijuana and Cocaine, 1991-1992

Table 7 (continued):

Table 8:Percentage of Persons age 16 to 49 who Reported Driving Under the Influence of Alcohol or Drugs in Past Year by Past Year Marijuana and Cocaine Use, 1991-1992