As the contributions illustrate, there are important future avenues to consider. specifically, the articles in the Special Section provide recommendations for future MOBC research to improve the feasibility of experimental manipulation, collect and analyze data with greater temporal resolution, maximize the use of existing data sets to extract valid and reliable information to inform clinical practice, consider the importance of pretreatment behavior change, and examine multiple levels of analysis. Greater use of experiments, which is advocated by both the MOBC and Science of Behavior Change initiatives, would move the field a step closer to garnering convergent evidence for multilevel mechanisms of drinking behavior change. The use of biofeedback to modify emotion regulation is one such approach. Also, future trials can incorporate data collection methods, such as ecological momentary assessment to assess person-level change in the natural environment, that allow greater attention to be paid to how behaviors vary with time, how behavior outside treatment occurs, and how pretreatment behavior change affects later change processes. Each of these components of change is likely important for a more complete understanding of MOBCs. Moreover, as the world becomes more connected via mobile devices and applications, ecological momentary assessment and other intensive momentary data collection methods are becoming increasingly affordable, sophisticated, and user friendly. Finally, there remains a need to examine existing data from clinical trials that have already been conducted with high methodological rigor, even when no main effects of treatment were observed. Analyses of extant data sets can additionally examine theory-based, cannabis square pot pretreatment moderators of MOBCs or changes in potential mechanisms during treatment. In conclusion, significant strides have been made in understanding how and under what conditions individuals change addictive behavior.
The question is deceptively simple, whereas the multifaceted nuances in behavior change require rigorous and complex design, sophisticated analytic methods, and strong theoretical rationale. This Special Section touches on a number of these important advances that can inform the emerging science of behavior change. Early U.S. studies generally found that decriminalization had no statistically significant impact on use in the United States. These studies focused on the years immediately following the passage of decriminalization statutes in eleven states. The most ambitious analysis found no significant association with use in both cross-sectional and longitudinal comparisons, using micro-level data from the late 1970s Monitoring the Future Survey of High School Seniors . Other state-specific analyses found similar null results . Studies examining the early years of decriminalization in several Australian jurisdictions also failed to find an effect on prevalence in simple cross-sectional and longitudinal comparisons . Drawing on a sparse set of cross-sectional and longitudinal indicators, MacCoun and Reuter argued that Dutch decriminalization in the 1970s had no measurable impact on levels of use over the first decade, though they tentatively attributed a later increase in prevalence to the rapid expansion of the number of commercial retail coffee shop outlets for cannabis. Only one study has suggested an effect of state decriminalization during the 1970s. Model analyzed the effect of marijuana decriminalization on drug mentions in hospital emergency room episodes using data from the 1975-1978 Drug Abuse Warning Network. Her analyses showed that cities in states that had decriminalized marijuana experienced higher marijuana ER mentions and lower other drug mentions than nondecriminalized cities. Model did not estimate a demand function directly, but her results implied that under decriminalization, drug users might have substituted marijuana for hard drugs. More recent statistical analysis have generated mixed findings, with some studies showing no effect while others showed a positive and statistically significant effect. For example, DiNardo and Lemieux found no effects of state marijuana decriminalization using state-level aggregated data from the 1980-1989 Monitoring the Future Survey.
They estimated log-linear and bivariate probit models of the likelihood of using alcohol and marijuana, so unlike previous models, their model considered the possible relationship between alcohol and marijuana use. Thies and Register found no significant impact of decriminalization in their analysis of data on young adults from the 1984 and 1988 National Longitudinal Survey of Youth . They estimated logit and to bit specifications of the demand for marijuana, binge drinking, and cocaine and included cross-price effects in all of the regressions. Finally, Pacula found no significant effect of decriminalization policy in her two-part model specification of the demand for marijuana using data from just the 1984 NLSY. Her model differed from that of Thies and Register in that it included additional proxies for the price of marijuana and other substances. Saffer and Chaloupka also found a significant decriminalization effect in individual level prevalence equations for past year and past month use of marijuana, alcohol, cocaine, and heroin using data from the 1988, 1990, and 1991 National Household Survey on Drug Abuse. Unlike other analyses, Saffer and Chaloupka’s work controlled for various measures of the monetary price of legal and illicit drugs in addition to controlling for whether a state had a formalized decriminalization policy. Additional analyses finding evidence of a statistically positive association in nationally representative samples of youth and young adults in the United States include Williams et al , DeSimone and Farrelly and Pacula et al . Another possibility is that the period in which the policies were evaluated may matter. This inconsistency in years evaluated may be generating differences due to cohort effects or unidentified policy changes that are not captured fully in the analysis. Cohort effects are likely to exist due to the fact that public awareness of specific policies generally declines over time as we move farther away from the period in which the policy was discussed or adopted. There are a number of other unidentified policy changes that could also be occurring during the time period, such as changes in enforcement practices associated with marijuana offences.
For example, Reuter, Hirsch field and Davies find that one third of those arrested for marijuana possession in three major Maryland counties spend time in jail pre-trial, even though almost none receive a sentence involving incarceration. Thus there may be variations over time in the extent of pre-trial detention that affect perceived penalties even though not targeted at marijuana use. Murphy conducted an analysis of FBI records and showed that 7 out of the 11 states that chose to decriminalize marijuana during the late 1970s ranked in the lowest 21 states in per capita marijuana possession arrests before they enacted their decriminalization law. Two states, Mississippi and North Carolina, were among the top 23 states in per capita arrests before their policy change. Murphy’s analyses of changes in arrest patterns before and after the reform took place suggests that the statutory change had little impact on arrest patterns for any of these states. But survey data from that period, examined below, suggests that youth perceived significantly lower penalties following the legal change, and this shift only occurred in those states changing their laws. We will present evidence that these perceptual differences across states have largely vanished, suggesting either that “decriminalization” and “nondecriminalization” states no longer differ in their actual enforcement patterns, or that citizens no longer perceive the difference – perhaps due to the lower salience of the change over time. Much of the confusion about decriminalization involves terminology. The term “decriminalization” is often seen by the public as a synonym for “legalization.” But this is a mistake; decriminalization refers to penalties for marijuana possession,trim tray and does not imply any change in the legal status of marijuana sales. Also, “decriminalization” literally implies a removal in the criminal status of marijuana possession offences; however, many jurisdictions that are recognized as having decriminalized marijuana in fact merely reduce the penalties associated with possession of specified amounts. In many ways, the term marijuana “depenalization” is a more useful term for describing the diversity in liberalizing policies that have arose across and within countries . Decriminalization, nonetheless, remains the more common term in policy debates. In addition, progress in understanding the effects of marijuana laws has been hindered by an over-reliance on a crude dichotomous “decriminalization” indicator. Recent research demonstrates that this simple dichotomy is quite inadequate for uniquely identifying real differences in the criminal treatment of low-level marijuana offenders in the United States.
Table 1 summarizes statutory penalties in effect as of January 2001 for first time marijuana possession offenders caught in possession of small amounts of marijuana for all fifty states and the District of Columbia . The correspondence between the “decriminalization” label and actual policies is quite variable. Seven states that had actually removed the criminal status of minor possession offences , were not formally recognized as decriminalized states. Five states that are widely recognized as having decriminalization statutes maintain the status of marijuana possession offences as a criminal charge. Some states allow a minor marijuana possession charge to be removed through a formal process called expungement. Many of the states that have expungement provisions are not known as decriminalized states, and only three of the five so-called decriminalized states retaining the criminal status of minor marijuana possession offences allow for the removal of the criminal charge upon completion of mandated punishment. It is also important to note that the decriminalization statutes do not remove criminal penalties for smoking marijuana in public, which has always constituted an important source of possession arrests. In addition to this conceptual confusion, there is empirical uncertainty about the effects of marijuana laws on enforcement patterns. Pacula et al. examined the relationship between state marijuana statutes and actual enforcement during 1991-2000. They report a 264% increase in marijuana possession arrests across all states, mostly occurring between 1991 and 1995. Between 1991 and 2000, there was a dramatic increase in variation across states, with the range increasing from about 30 arrests per 10,000 in 1991 to 110 arrests per 10,000 in 2000. More importantly, by 2000, states that had eliminated the criminal status of possession offences involving amounts of one ounce or less of marijuana did not have systematically lower arrests per capita than those states retaining the criminal status. More than half of the states that do not consider small marijuana possession offences a criminal offence still had per capita arrest rates greater than the national average and they still experienced a significant increase in arrests during the 1992-1995 time period. One interpretation is that these arrests do not reflect simple possession of marijuana but that many are the result of bargaining down from more serious offenses, such as marijuana distribution . But if these arrest rates do correspond to actual legal risks for marijuana possession, then it is puzzling that recent studies find a consistent and statistically significant effect of the simple decriminalization dummy indicator on use even after controlling for enforcement . Because decriminalization is a severity-based intervention, these results may explain those studies failing to detect reliable decriminalization effects. But as we have seen, those studies operationalized decriminalization using an imprecise and somewhat misleading dichotomous indicator. It is also possible that perceptual deterrence studies conducted within a criminalization regime understate the potential effects of decriminalization . The sanction certainty dimension may have important threshold effects.Also, the mere fact that an act is illicit may influence behavior independently of the magnitude of the legal threat. There may be similar discontinuities for the sanction severity dimension. Human judgment is notoriously susceptible to range and anchoring effects . Statutory maxima are an example of the kind of “worst case scenarios” that people tend to weight disproportionately . Whether decriminalization might have a larger than expected effect depends, in part, on whether citizens actually know something about their state’s marijuana laws. Various lines of evidence suggest that citizens may have distorted or biased beliefs about sanctioning threats , but very little work has been done to empirically investigate whether this is true with respect to drug laws. Relatively few studies have ever measured the accuracy of citizens’ beliefs about legal sanctions. Early studies found that the general public tends to exaggerate the risks of arrest and punishment for many crimes . But in accordance with the availability heuristic , personal experiences play an important role in shaping perceived risks.