Definition of Cost Categories

The following descriptions of cost factors are taken verbatim from "The Economic Costs of Drug Abuse in the United States 1992-1998," Office of National Drug Control Policy, September 2001

Costs Attributed to Health Care
  • Community-Based Treatment

    Community-based specialty treatment includes all specialty drug abuse treatment that is not funded through a federal agency.

  • Federally-Provided Treatment

    Federal specialty treatment costs consist of costs for specialty treatment funded through a federal agency. Specifically, cost estimates for the Department of Defense, Bureau of Indian Affairs, Bureau of Prisons, and Veterans Affairs were obtained from ONDCP National Drug Control Strategy: Budget Summary.

  • Federal Prevention

    The actual level of federal spending on prevention is published annually by ONDCP. This estimate includes funding for prevention of alcohol abuse, however, as well as drug abuse. The spending estimate was apportioned between alcohol and drug abuse based on data from analyses of the primary reason for treatment among clients in the Uniform Facility Data Set (UFDS). Analysis of these data indicated that 48 percent of current clients were treated for alcohol abuse only, 28 percent for drug abuse only, and 24 percent for both alcohol and drugs. Splitting the comorbid group in half, the cost of alcohol and drug abuse treatment was apportioned 60-percent to alcohol and 40-percent drugs (Harwood et al., 1998).

  • State and Local Prevention

    State and local government spending for drug abuse prevention can be estimated based on data included in an annual report called State Resources and Services Related to Alcohol and Other Drug Problems published by the National Association of State Alcohol and Drug Abuse Directors (NASADAD). Ideally, an estimate of spending on prevention that is limited to drug abuse and that includes only spending funded by state and local government would be used. The prevention spending estimates listed in the NASADAD report however include funding for prevention of alcohol abuse and include federal funding and funding from other nongovernmental sources. To apportion the spending between alcohol and drug abuse, the same ratios that were applied for federal spending on prevention were used. To apportion the prevention spending between state and local government spending and spending from the federal government and other sources, estimates of the proportion of total spending included in the NASADAD (i.e., spending for treatment, prevention, and all other activities) for alcohol and drug abuse that is from state and local government funds were used. This proportion ranges from 41.1 to 44.9 percent depending on the year.

  • Training

    The cost of training includes initial and continuing education for specialists in drug and alcohol treatment as well as for other health professionals, law enforcement officials, criminal justice professionals, and clergy. No published data specific to these costs were available. Therefore, this estimate was updated based on two components. The first component measures real change based on the change in the U.S. population age 18 years of age or older. The second component measures the price change based on the change in the Consumer Price Index.

  • Prevention Research

    Research spending can be decomposed into prevention and treatment related research. ONDCP publishes annual estimates for these two types of research costs in the National Drug Control Strategy: Budget Summary. Actual spending estimates are available for these two components through 1999. For 2000, the budget request is used as the estimate of spending for these components.

  • Treatment Research

    Research spending can be decomposed into prevention and treatment related research. ONDCP publishes annual estimates for these two types of research costs in the National Drug Control Strategy: Budget Summary. Actual spending estimates are available for these two components through 1999. For 2000, the budget request is used as the estimate of spending for these components.

  • Insurance Administration

    The Health Care Financing Administration (HCFA) estimates the percentage of health care spending that is for administration as part of the national health accounts. These percentages were adopted for 1992 through 1998 to estimate the administrative costs related to specialty drug treatment. These percentages are applied to the estimates of the cost of specialty treatment in each respective year to determine the cost of administration related to the specialty treatment.

  • Hospital and Ambulatory Care

    There were no published data available which could be used to re-estimate the hospital and ambulatory care costs related to the medical consequences of drug abuse through simple tabulations. Therefore, the baseline 1992 estimate of these costs is updated based on two factors. The real change is measured based on changes in the number of adults reporting more than 100 days of using marijuana or cocaine during their lifetime as displayed in Table IV-6. The derivation of the estimates presented in Table IV-6 is based on data from the NHSDA as presented in Appendix Table B-1. This number fluctuated between 1995 and 1998. Since there is no consistent historical trend in this measure, it was assumed that it would be constant from 1998 through 2000. While trends in drug abuse related illness are expected to parallel trends in the number of heavy drug users, the measure of heavy drug users from the NHSDA is limited, because of sampling error and because it is self-reported.

  • Drug Exposed Infants

    Ideally, the real change in this estimate would be measured based on the change in the number of drug-exposed infants born each year, but this information was not available. As an alternative an estimate of the number of pregnant women reporting current illicit drug use was considered. The NHSDA includes pregnant respondents who are asked about current drug use. The sample of pregnant women in the NHSDA is small however. To calculate statistically valid estimates, two years worth of data on pregnant women are combined. The estimates of the percent of pregnant women who use illicit drugs are small. They vary from 2.3 to 3.2 percent depending on the years and the estimates fluctuate from year to year. Because the sample size is small and the estimates fluctuate from year to year, data from the NHSDA was not used to update the cost of drugexposed infants.

  • Tuberculosis

    The real change in health care costs attributable to tuberculosis (TB) was measured as the change in the number of cases of TB that were attributable to injecting or non-injecting drug use according to the Center for Disease Control (CDC). The number of TB cases was available from the CDC for each year between 1992 and 1999. However, the percentage of cases attributable to injecting or non-injecting drug use was only available for 1996 through 1999. Between 1992 and 1996, the percentage of cases attributable to injecting or non-injecting drug use was assumed to be fixed at the 1996 level. Also, since 2000 values were not yet available from the CDC, the change in the number of cases observed between 1999 and 2000 was assumed to be the same as that observed between 1998 and 1999.

  • HIV/AIDS

    The most recent tabulation of the cost of caring for individuals with HIV/AIDS is Hellinger and Fleishman (2000). These authors estimate that the cost of treating all people with HIV disease in 1996 was between $6.7 and $7.8 billion. This range was calculated through two different approaches (i.e., payer-based and provider-based). The estimates calculated under each approach were compared. The mid-point between these two estimates, $7.25 billion, is used as the estimate for total medical spending on HIV/AIDS in 1996.

    Not all cases of HIV/AIDS are attributable to drug abuse. The portion of HIV/AIDS spending attributable to drug abuse was estimated based on data from the CDC. The CDC monitors the exposure category of each reported case. In 1996, 37 percent of individuals living with AIDS had an exposure category related to intravenous drug use (IVDU). Thus, 37 percent of the 7.25 billion or $2.68 billion are assumed to be related to drug abuse.

  • Hepatitis B and C

    To update the estimate for drug abuse related hepatitis spending, a real and a price change factor were calculated. The real change was based on the change in the number of hepatitis cases attributed to IVDU between 1992 and 1998. Table IV-11 displays the total number of hepatitis B and C cases in each year between 1992 and 1998. This table also displays the percentage of cases attributable to drug use for 1993 through 1995. The percentage of cases attributable to drug abuse is not available for 1992 nor for 1996 through 1998. The percentage of cases attributable to IVDU in 1992 was assumed to be the same as in 1993. Similarly, the percentage of cases attributable to IVDU in 1996 through 1998 was assumed to be the same as in 1995. Between 1995 and 1998 the number of hepatitis cases associated with IVDU declined at a five percent annual rate. However, there was considerable fluctuation in the rate of growth across these years. Since there was no consistent trend in the historical data, the number of cases was assumed to remain at the 1998 level in 1999 and 2000.

  • Violent Crime

    The attribution factors are based directly on estimates that about 30 percent of prison inmates serving time for robbery, burglary, and larceny (income-generating, or acquisitive crimes) and two to five percent of violent offenders reported that they committed the crime for which they were incarcerated in order to get drugs (U.S. Department of Justice,1994a). The 15.8 percent homicide factor is based on FBI analyses. Their study of crime circumstances and offender relationships found 15.8 percent of homicides either listed drugs as a factor, or were "gang" related. This study is found in U.S. Department of Justice (1994b). Crime in the United States, 1993. Washington DC, Federal Bureau of Investigation. These attribution factors might be termed "conservative", given that somewhat higher proportions of inmates reported use of drugs at the time they committed their crime, and more than twice as many arrestees tested positive for drug use based on a study found in U.S. Department of Justice (1994a). Fact Sheet: Drug-Related Crime. Office of Justice Programs, Bureau of Justice Statistics, Publication NCJ-149286. However, the issues of attribution and causality are quite complex with criminal behavior, particularly given that many offenders also drink alcohol at the time of offenses.

  • Health Insurance Administration

    Similar to the calculation of health administration costs related to specialty care, health administration costs related to the medical consequences of drug abuse were calculated as a percentage of the total medical service costs related to medical consequences of drug abuse.

Costs Attributed to Lost Productivity

  • Premature Death

    The same list of diagnoses and attribution factors that was used by Harwood et al. (1998) to calculate the baseline 1992 estimate was used.10 The initial list of diagnoses and attribution factors was obtained from the National Institute on Drug Abuse which developed the list for the Drug Abuse Warning System (Gottshalk et al. 1977 and 1979). This list includes diagnoses representing abuse of and dependence on psychoactive drugs as well as accidental and intentional (i.e., suicide) poisoning by a broad range of drugs and medications-psychoactive and otherwise. The Harwood et al (1998) study then added TB, hepatitis B and C and HIV/AIDS to the list of diagnoses attributable to drug abuse.

    The number of deaths for each age/sex category was multiplied by the estimated value of lifetime earnings. The lifetime earnings table was obtained in personal communication with Dorothy Rice (1997). The formula for these calculations appears in Rice et al. (1990). The estimates for the expected value of lifetime earnings for 1992 are trended to future years based on the Bureau of Labor Statistics (BLS) hourly compensation index (HCI).

  • Drug Abuse Related Illness

    Two factors were used to update this estimate. The real change in the estimated cost of lost productivity related to drug abuse related illness is measured as the change in the number of persons reporting more than one hundred days of marijuana or cocaine use in their lifetime in the NHSDA as displayed in Table IV-6. The price change in the estimate is measured based on the BLS HCI as displayed in Table IV-16.

    The real change in lost productivity due to drug related illness is measured based on the number of individuals reporting 100 or more days of marijuana or cocaine use in their lifetime. Since this is a self-reported measure included in the NHSDA, it may be subject to changing views on drug use. Also, changes in the number of individuals using marijuana and cocaine for more than 100 days may not be closely related to drug abuse related illness. [Note: Emphasis added by truth:the Anti-drugwar]

  • Institutionalization/Hospitalization

    The estimates of lost productivity due to institutionalization/hospitalization were updated based on two factors. The real change in these costs was measured based on the number of clients using inpatient hospital or residential treatment services. The count of clients using these two types of services in 1992 through 1998 is displayed in Table IV-17.

    There was not a consistent trend in this measure between 1992 and 1998. Therefore, the projections for 1999 and 2000 assume that the number of clients will remain constant at the 1998 level.

    The price change was measured based on the BLS HCI as discussed in Section IV.B.1 and displayed in Table IV-16.

  • Productivity Loss of Victims of Crime

    This estimate was updated based on two factors. First, the real change in this component is measured based on the estimated change in the number of hours lost by crime victims due to drug abuse. The number of hours lost by crime victims due to drug abuse in each year between 1992 and 1998 was calculated based on three components:

     The number of victimizations by type of crime (e.g., burglary, rape);
     The average number of days of productivity lost by crime type; and
     The percentage of crimes of each type that are attributable to drug abuse.

    The estimated number of days lost by crime type is not updated annually. The most recent published estimates were from 1992 and were applied to estimate the number of days lost to crime for each year between 1992 and 1998. The value of each day lost was estimated to be $133 in 1992. To estimate the value of hours lost in 1993 through 2000, this value was updated via the BLS's HCI as presented in Table IV-16.

  • Incarceration

    Similar to the update factors derived for the lost productivity of crime victims the productivity lost due to incarceration was updated based on two factors. The number of individuals under incarceration for drug abuse related crime in each year between 1992 and 1998 was calculated based on three components:

     The number of individuals under incarceration on June 30th of the year;
     The distribution of individuals under incarceration by primary offense; and
     The percentage of crimes of each type that are attributable to drug abuse.

    The price change in the cost of lost productivity due to drug abuse related crime was measured via the BLS's HCI as presented in Table IV-16.

  • Crime Careers

    The estimate of lost productivity due to drug related crime careers was updated based on two factors. The real change was measured based on the change in the number of chronic hardcore users of drugs. The number of chronic hardcore users of drugs was estimated annually in the NHSDA between 1992 and 1996. The estimates derived from this survey are displayed in Table IV-20.

    The number of hardcore drug users for 1999 and 2000 are projected. The number of hardcore drug users was fairly level between 1992 and 1995 and then there was a sharp 14 percent increase between 1995 and 1996. The projections assume that the number of hardcore users remained constant between 1998 and 2000. The price change in the cost of lost productivity due to drug abuse related crime was measured via the BLS's HCI as presented in Table IV-16.

Costs Attributed to "Other" Costs
  • Police Protection

    Total police protection and legal adjudication costs are published periodically by the Bureau of Justice Statistics. There is a substantial lag in the publication of these estimates. The most recent estimates available were for 1996. The estimates for 1992 through 1996 are listed in Appendix Table B-7. For 1997 through 2000, the costs were projected to increase at a six-percent annual rate. This rate of increase is consistent with the magnitude of the increases in police protection and legal adjudication costs that were observed between 1992 and 1996. Police protection and legal adjudication costs were attributed to drug abuse based on the percentage of arrests attributable to drug abuse.

  • Legal Adjudication

    Because of the lag in publication of the police protection and legal adjudication costs, the 1992 estimate published in Harwood et al. (1998) was derived by projecting forward the 1990 estimates for police protection and legal adjudication costs. Because the observed value for 1992 is now available, the 1992 estimate was updated as well as deriving the estimates for subsequent years. The estimates produced by Harwood et al (1998) were $4,644 and $1,210 million for police protection and legal adjudication, respectively. The estimates produced in this study based on the observed cost estimates for 1992 are substantially higher, $5,348 and $2,716, because the actual costs for police protection and legal adjudication were substantially higher than the projected costs.

  • State and Fedeal Corrections

    Corrections costs for local jails were attributed to drug abuse based on the percentage of arrests that are associated with drug abuse. Corrections costs for state and federal prisons were attributed to drug abuse based on the percentage of persons under incarceration in state and federal prisons that were attributable to drug abuse. Appendix Table B-11 displays these percentages for 1992 through 1999. The number of people incarcerated is expected to increase between 1999 and 2000 at the same rate of increase that was observed between 1998 and 1999. The percentage of persons under incarceration whose primary offense is drug related was assumed to remain constant between 1999 and 2000.

  • Local Corrections

    Corrections costs for local jails were attributed to drug abuse based on the percentage of arrests that are associated with drug abuse. Corrections costs for state and federal prisons were attributed to drug abuse based on the percentage of persons under incarceration in state and federal prisons that were attributable to drug abuse. Appendix Table B-11 displays these percentages for 1992 through 1999. The number of people incarcerated is expected to increase between 1999 and 2000 at the same rate of increase that was observed between 1998 and 1999. The percentage of persons under incarceration whose primary offense is drug related was assumed to remain constant between 1999 and 2000.

  • Supply Reduction

    ONDCP reports annually on federal funds spent to reduce the supply of drugs. The detailed components of this funding for 1992 through 2000 are presented in Appendix Table B-12. In Table IV-24, the overall total spending for each year between 1992 and 2000 are presented. The estimates for 1992 through 1999 indicate actual spending. Since actual spending for 2000 is not available, the 2000 costs indicate the budget request.

  • Private Legal Defense

    The cost of private legal defense attributable to drug abuse was calculated based on three components:

     total annual revenue for legal services as reported by the Bureau of the Census;
     the percentage of lawyers practicing criminal law; and
     the percentage of arrests attributed to drug abuse.

  • Property Damage

    Property lost due to crime was estimated based on the following three components:

     Number of victimizations by offense;
     Percentage of victimizations for each offense attributed to drug abuse; and
     Estimated mean property loss per crime by offense.

  • Social Welfare

    This category was divided into two separate components for updating Supplemental Security Income (SSI) payments and all other social welfare payments. Beginning in 1997 drug abuse related disorders no longer constituted an acceptable basis for SSI eligibility, and this value was set to zero. Other changes in social welfare becoming effective in 1997 have significantly reduced payments to beneficiaries. This trend has been estimated/represented by the change in Food Stamp benefits as presented in Table IV-27. While the Social Security Administration (SSA) compiles national (federal plus state) estimates of social welfare outlays (and particularly public aid), publication of these data lags about 4 to 5 years -- 1994 data were the most recent available for this study.

truth: the Anti-drugwar Anti-drugwar Arsenal Accept the $1 Challenge