1 Introduction

While it is widely understood that the consumption of illegal substances imposes harms on the user as well as society at
large, considerable debate exists regarding the nature of those harms, the actual burden they place on individuals and society,
and the extent to which their existence and magnitude justify government action or are influenced by government policies.
Some of the debate stems from philosophical differences with respect to the importance of personal liberties and the proper
domain of government. However, a good portion of the debate also comes from observations made across various countries,
where it is clear that fundamentally different approaches have been taken towards managing drug users and the harms they
impose on others. Given these “natural experiments,” there is a desire by scientists as well as policy makers to evaluate and
compare the relative burden of illicit drug use across nations adopting different strategies in hopes that such a comparison
would generate useful insights regarding successful approaches for balancing the harms from use and the harms from society’s
response to use. A careful cost-effectiveness analysis using information across countries would be the best way to obtain
these insights. One of the first steps in conducting such an analysis involves the consistent assessment of the burden of drug
use across countries.
Considerable work has been done in some developed countries to quantify the social cost of drug abuse (Rehm et al., 2007;
Collins & Lapsley, 2002, 2008; Godfrey et al., 2002; ONDCP, 2002). Although these studies all employ the same end metric
for valuing harm (currency), they use a variety of approaches that make it impossible to directly compare results. Political
and social environments influence not only the types of harms considered in these calculations but also the relationship
between drug use and the harm (e.g. harm reduction strategies influencing the relationship between injection drug use and
the spread of HIV/AIDS). Methodological differences in the measurement of harms, the inclusion of intangible costs, and the
time horizon in which harms are evaluated leads to further inconsistencies. In light of these differences, it becomes difficult
to learn much from drawing comparisons across these studies even though they ultimately measure the problem using the
same final outcome metric.
This report attempts to consider the feasibility of constructing a new estimate of the cost of drug abuse by focusing on a small
subset of harm indicators that one might reasonably expect to be systematically and consistently measured across countries.
It lays out the steps that were taken in an attempt to construct such an estimate for a select number of developed countries.
While the report provides the components to estimate these costs for selected countries, the validity of the estimates for
cross-country comparisons is highly questionable and strongly discouraged. It became readily apparent while constructing
these estimates that most of the data indicators were not truly consistent, particularly across countries. A given variable, even
when defined in a consistent manner, is not measured the same way across countries. The pitfalls and assumptions necessary
to construct a comparable estimate across countries are quite significant and described in detail throughout this report. We
conclude that it is not possible at this time to develop a meaningful comparative estimate of the cost of drug use across
countries or to aggregate these costs to the regional or global level.
The rest of this report is organized as follows. Chapter 2 compares the published national studies on the cost of drug abuse
and highlights key differences in definitions, and measurement even when common methodological approaches are adopted.
It demonstrates why simply aggregating existing studies to generate a global burden of the drug problem is problematic.
Chapter 3 presents the conceptual framework for a simplified approach for considering the economic burden of drug abuse
consistently across countries, identifying key cost components that should be obtainable in a consistent fashion across many
countries. While national estimates of the cost of drug use generally include additional indicators of the cost of drug abuse,
these indicators are unlikely to be collected systematically for all countries (as in the case of social welfare costs). Further, it is
even less likely that consistent measures of unit cost estimates are available (as in the case of the value of lost time at work
due to drug-related absenteeism). Therefore, the conceptual framework presented is necessarily less comprehensive than
existing national estimates. Nonetheless, it captures many of the key cost drivers demonstrated in those national studies.
In chapter 4, issues discovered in trying to implement even this simplified conceptual framework are discussed that led to a
further narrowing of the number of countries and costs actually considered. In chapter 5, issues related to obtaining consistent
estimates of the unit cost of the harms for just a relatively small number of developed countries are discussed. Finally, in
chapter 6, after examining inconsistencies in the measurement of indicators and costs even for a subset of countries with
relatively good data, it is determined that it is not yet possible to construct a comparable estimate of the cost of drug use in a
manner that would enable cross-country comparisons in a systematic and scientifically consistent way. Such an effort requires
the coordinated effort of countries interested and willing to engage in such an exercise, such as that being undertaken by the

EMCDDA for the European Community, so that indicators and cost information is obtained in a fashion that would enable
cross-country comparisons. The work of harmonizing indicators is a very difficult process, however, and one that takes time.
More time is needed to expand the harmonization across more countries, including the U.S., Australia and Canada, before
any serious attempt can be made.