Notably, our post hoc sensitivity analyses of race/ethnicity only utilized the years 1989-2013, indicating that the discrepancies between our findings and Cil’s probably are not due to the different time frames. Strengths and Limitations This is the first study to examine all policies related to alcohol use in pregnancy simultaneously across all 50 states using a time frame long enough to capture the period before any laws were enacted . Furthermore, for most of the time frame the data include the entire population of singleton births born in the United States and for the years 1972-1984 include a 50% sample, which makes questions regarding inference and generalizability essentially irrelevant. Another major advantage of these data over, for example, survey data regarding alcohol use during pregnancy, is that biases due to self-report are not present here. Finally, our results were robust across various model specifications, further strengthening our conclusions. The main limitation of this study is that Vital Statistics birth certificate data are not collected for research purposes; therefore, we cannot adjust for maternal-level alcohol or tobacco use. Although maternal alcohol and tobacco use have been recorded on birth certificates since 1989, these data have been shown to be invalid We adjusted for state-level alcohol and tobacco consumption instead. Another limitation is that race has been measured inconsistently on birth certificate data over time. Only in 1989 did states begin to document ethnicity as well as race, although this was phased in over the 1990s. Our primary analyses did not account for ethnicity, e.g. White Hispanic and White Non-Hispanic women are in a single group. Such an approach is reasonable because birth outcomes are similar between White nonHispanic and Hispanic births,vegetables vertical farming both of which differ from Black birth outcomes.

Measurement of key outcome variables – particularly gestational age – changed over time as well. We applied approaches developed later to correct for implausible gestational age values to earlier years of Vital Statistics to improve consistency. Also, for these analyses, we focused specifically on policies targeting alcohol use during pregnancy.Future research is needed to explore whether the findings generalize to policies targeting drug use during pregnancy. Framing drug policy in language of supply- vs. demand-side programs reflects the increasing diffusion of economic thinking from the business place to other domains of American life. The idea is that some interventions involve supply while others involve demand , and that there is a drug control budget pie that can be sliced along these lines. But there are some drawbacks to this framing. As Murphy has documented, the notion that we can simply shift monies from one portion of a federal drug budget to another is naïve; there is no single allocating authority, and the “budget” is a mythical post-hoc construction assembled from a variety of conflicting sources and entities. And supply and demand factors are clearly interdependent and endogenous. The alternative idea of a “public health” framing of drug policy is refreshing, but in practice it tends to devolve to the “demand reduction” frame. Instead, we will try to keep the focus on strategies, rather than tactics; goals rather than programs. Our framework for doing so is sketched here and is developed in greater detail elsewhere Our perspective will not appeal to everyone. In particular, our framework is irrelevant for people who hold that certain moral beliefs trump any consideration of consequences.

There are two such deontological positions. One is the libertarian belief that ingesting psychoactive substances is our birthright. At the other extreme is legal moralism – the belief that drug intoxication is intrinsically immoral. Based on an extensive analysis of drug policy rhetoric , we conclude that few people are strict libertarians or pure legal moralists with respect to drugs. Most people who argue that either drug use or drug prohibition is immoral usually cite empirical arguments in support of their positions. At this point, we bid pure libertarians and legal moralists adieu. For the consequentialists, we suggest three broad goals: Prevalence reduction , quantity reduction , and micro harm reduction . Practices and concepts most readily identified with prevalence reduction include abstinence, prevention, deterrence, and incapacitation. Practices and concepts most readily identified with harm reduction include safe-use and safe-sex educational materials, needle exchanges, and the free distribution of condoms to students . Traditional discussions of prevention, treatment, deterrence, and incapacitation focus almost exclusively on the first category, with the implicit assumption that the best way to eliminate harm is to eliminate prevalence — turning users into non-users. This is logically correct, but not very realistic. Prevalence reduction may be employed in the hope of reducing drug-related harms, but because it directly targets use, any influence on harm is indirect. Harm reduction directly targets harms; any influence on use is indirect. From an analytic standpoint, all three strategies contribute to a broader goal, macro harm reduction . For tangible harms, Macro Harm = Micro Harm x Prevalence x Quantity, summed across types of harm .

The strategies are potentially in tension, particularly if efforts to reduce prevalence increase harm , if efforts to reduce quantity discourage abstinence , or if efforts to reduce average harm encourage the prevalence or quantity . Thus, any drug policy intervention should be evaluated with respect to all all three criteria – prevalence reduction, quantity reduction, and harm reduction – because all three contribute to the reduction of total drug harm. Note that our use of “harm reduction” is unusual here, in that we are not referring to specific “harm reduction” programs like needle exchange, but rather to a goal that is served – well or poorly – by any intervention. For that reason, we will discuss harm reduction in the context of traditional interventions like policing, prevention, and treatment. Why is psychoactive drug use a crime? And is there a sensible answer that also explains why tobacco and alcohol are on one side of the legal threshold, while marijuana, cocaine, the opiates, and the psychedelics are on the other? One way of tackling this question is historical, and there are a number of outstanding histories of roles played by race, class, and economic interests in the evolution of drug, tobacco, and alcohol control . Another approach is philosophical. If we were starting a society from scratch, which substances, if any, would we prohibit? The traditional first cut at this question uses John Stuart Mill’s harm principle: “That the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant.” MacCoun, Reuter, and Schelling listed nearly 50 different categories of drug-related harm, falling into three clusters: Health, social and economic functioning, safety and public order, and criminal justice. Many are quantifiable, at least in principle , but others are not . The authors attempted to categorized these harms with respect to two questions: Who is the primary bearer of the harm? And, what is the primary source of the harm? These harms to others meet the Mills criterion, but that hardly nails down the case for prohibition. MacCoun, Reuter, and Schelling argued that for over half of the harm categories, the primary source of the harm was either the illegal status of the drug, or the enforcement of that law, at least under the current prohibition regime. . The notion that prohibition and its enforcement are partially responsible for drug harms is perhaps best illustrated by examining the relationship between an offender’s illicit drug use and his or her involvement in other crimes. A considerable literature on this relationship suggests the following conclusions . Drug use can promote other crimes; criminality can promote drug use; and/or both can be promoted by environmental, situational, dispositional,vertical farming production and/or biological “third variables.” All three pathways have empirical support in at least some settings and populations. But these causal influences are probabilistic, not deterministic. Most drug users are not otherwise involved in serious crime. Finally, the drug-crime link varies across individuals, over time within an individual’s development, across situations, and possibly over time periods . Like many things in life that are bounded at zero, the frequency distribution of drug consumption has a positively skewed log-normal shape . If one plots the proportion of all users as a function of quantity consumed , most users pile up on the low side of the quantity distribution, but the plot will have a long narrow right tail representing a small proportion of user who use very large quantities. As a result, the harmful consequences of substance use are not uniform, but are disproportionately concentrated among the heaviest users.

Everingham and Rydell used these features to explain why cocaine-related harms remained high even as total prevalence was dropping; one sees a similar logic today in methamphetamine statistics. There is a sophisticated treatment of these distributional features and their implications for the targeting of interventions , but far less little discussion in the illicit drug literature . Another distributional consideration is how drug use and drug harms are distributed across geographic, class, and ethnic lines. African Americans use illicit drugs at a rate similar to European Americans, but they bear a disproportionate share of the law enforcement risk and market related violence. This is partly due to the fact that poorer neighborhoods lack the social capital needed to resist open-air drug markets. But it also reflects the deleterious effects of our mandatory minimum sentencing policies, discussed below. Finally, drug problems are distributed over time. Musto argues that drug epidemics are dampened by a generational learning process in which new cohorts observe the harmful results of a drug on older users. Building on this idea, Caulkins and his colleagues have developed sophisticated models of how interventions may provide less or greater leverage at different points in a drug epidemic. They argue, for example, that supply reduction measures will be more effective in the early stages of an epidemic but relatively ineffective in a large, mature, established market. Conversely, prevention and treatment may have limited effectiveness early in an epidemic – prevention because its effects are so lagged, and treatment because it interferes with generational learning about drug harms. This work is necessarily fairly speculative at present; we lack enough long term time-series data to permit serious testing of such hypotheses. But their analyses are valuable in encouraging another dimension of more strategic thinking. Scholars rely heavily on counts of arrest rates and victimization reports to track trends in most categories of illicit behavior. In contrast, the literature on illicit drug use relies much more heavily on surveys of self-reported drug use, and to a lesser extent drug related medical events. This probably reflects the sheer prevalence of drug use in the population , as well as the more diffuse linkage between the criminal act and any harms to innocent victims. The 2007 Monitoring the Future annual school survey has the longest running consistently measured time series for substance use in the US. Figure 1 shows trends in past-month prevalence for selected substances for 12th graders . Several patterns are apparent. First, alcohol remains the most common psychoactive substance among high school seniors. Second, in the most recent year , monthly alcohol and cigarette use each reached their lowest recorded levels. Third, past-month marijuana use reached its peak around 1979, hit a low in 1992, and has stabilized near 20 percent for the past decade. Finally, recent use of cocaine, MDMA, or methamphetamine is fairly rare among high school seniors. MDMA use seems to have peaked at 3.6 percent in the year 2000, cocaine use has remained fairly stable at around 2 percent, and methamphetamine has dropped from 1.9 percent in 2000 to 0.9 percent in 2006. Table 1 shows past-month prevalence of various substances by age category, from the household-based 2006 National Survey on Drug Use and Health . For each substance, young adults were the most frequent recent users, with one exception – heroin. New heroin initiation is rare; the US heroin problem mostly involves an aging cohort of addicts who initiated use in their youth. The overall methamphetamine rates obscure the fact that prevalence remains considerably higher in the Western US than in the Midwest and South and the Northeast , and is higher among whites and Latinos than among African Americans .