Abstract Introduction To mitigate the harms of arrest and incarceration on health and racial equity, jurisdictions are increasingly enacting reforms to decriminalize drug possession via prosecutorial discretion (de facto). Impacts on health outcomes rely on whether this policy can reduce exposure to the carceral system among people who use drugs (PWUD); however, data evaluating effects on arrest are lacking. This study explores possible impacts of Baltimore City's enactment of de facto decriminalization on arrests by race. Methods Police and court records were used to explore possible impacts of Baltimore City's de facto decriminalization on street arrests and (“processed”) arrests advancing through the courts among PWUD. Interrupted time series models were used to compare pre-policy (January 2018–March 2020) to post-policy trends (April 2020–December 2021) in arrests for possession of drugs/paraphernalia and estimate racial disparities in street arrests (Black vs other races). Analyses were performed February-May, 2022. Results The policy was associated with a significant and immediate decline in street and processed arrests for possession which was not seen for other crime categories. While declines were concentrated in the Black community, disparities in arresting persisted after the policy. Conclusions De facto decriminalization may be a promising strategy to reduce exposure to the carceral system, an established risk factor for overdose and other drug-related sequalae and driver of racial disparities in the US. Further research is needed to elucidate drivers of persisting racial disparities and disentangle policy effects from pandemic-related closures.
Introduction The overdose crisis continues to worsen in the United States (US), with fatalities increasing most steeply among racial minorities.1 The predominant policy approach to substance use has been criminalization of drug possession although empirical data demonstrate this to be associated with worse health outcomes among people who use drugs (PWUD). Criminalization increases likelihood of high-risk drug use behaviors such as injection, deters help-seeking during overdose emergencies and erects obstacles to treatment access and continuity. Access to treatment for substance use disorder is limited in jails and prisons 2 and risk of fatal overdose is significantly elevated post-detention.3 Drug criminalization has also been a key driver of racial disparities. President Nixon's launch of the ‘War in Drugs’ in 1971 and its expansion by subsequent administrations triggered unprecedented investment in enforcement of drug interdiction and militarization of police departments nationwide. The number of people incarcerated for drug offenses increased by over 1200% from 1980-2008.4,5 Racial and ethnic minorities and especially Black individuals have been disproportionately targeted by drug law enforcement, despite no evidence of elevated drug use among them. Consequently, health and social ramifications of carceral involvement are concentrated in these communities. These deleterious consequences are theorized to operate through numerous pathways.6 Police encounters are associated with physiological and psychological distress, including cardiovascular disease, anxiety and depression;7 some result in police brutality which incurs additional trauma and can in rare circumstances be lethal.8 Indirect exposure via residence in overpoliced neighborhoods and media exposure to police brutality is also associated with physical and mental health.9 Arrests advancing through the carceral system resulting in jail, prison, probation or parole incur additional consequences including, but not limited to, interrupted access to healthcare, employment and housing opportunities, disruption of families and social networks, exposure to violence and increased risk of substance use initiation and related outcomes.10,11 Discriminatory drug enforcement results in increased risk of both police encounters and advancement through the system among Black individuals, substantially amplifying these sequelae in their communities.12 Amid escalating overdoses and racial inequities, advocates have called for decriminalization13-15 of the possession of all illicit drugs, including those implicated in the overdose crisis (e.g. opioids, stimulants). Decriminalization eliminates criminal penalties for drug possession, replacing them with either civil penalties, health referrals, or no action at all.16,17 This is distinct from legalization, which legally sanctions the substance, addressing issues of supply. Drug decriminalization can be achieved legislatively [de jure], which occurred for the first time in the US in Oregon in 2020. Alternatively, de facto decriminalization may be enacted through blanket non-prosecution of drug possession by chief prosecutors (e.g. State or District Attorneys). Police discretion in applying these reforms, which may be influenced by factors including policy knowledge, buy-in and competing priorities or incentives, is a known implementation lever which can influence drug policy effectiveness18; relative to legislation, de facto decriminalization may rely more heavily on cooperation of law enforcement officials who are not legally bound to adhere to prosecutorial directives. While a movement to leverage prosecutorial discretion is growing (known as ‘progressive prosecution’)19, there is limited empirical evidence to support claims of its critics and proponents alike. To understand the long-term implications of de facto reforms for health and racial equity, it is necessary to evaluate whether it effectively reduces arrests, an indicator of police adherence and key step on the presumed causal pathway linking decriminalization to health outcomes. The current paper addresses this gap by examining the impact of a blanket non-prosecution policy for all drug and paraphernalia possession on arrests in Baltimore City, Maryland, a jurisdiction historically impacted by racial disparities in substance use and mass incarceration. Fatal overdoses increased by 262% between 2015-2020, with provisional data signaling further increases since the COVID-19 pandemic. In Maryland, fatalities decreased among white residents by 11% but increased by 35% for Black residents between 2017-2019.20 Black residents account for 29% of the state population but 71% of those incarcerated.21,22 In 2015, tensions between police and the community erupted in Baltimore after the death of a Black man named Freddie Gray in police custody. The State's Attorney's Office (SAO) indicted six officers on charges ranging from false imprisonment to 2nd degree depraved heart murder. Though all were acquitted, relations between the Baltimore Police Department (BPD) and the SAO were strained. In 2016, the US Department of Justice issued a report detailing BPD's engagement in systematic civil rights violations targeting Black communities, including unconstitutional searches, arrests, and excessive force.23 The City of Baltimore and the BPD entered into a federal court-enforceable consent decree signaling commitment to reforming egregious policing practices.24 However, subsequent years were marked by further police misconduct, including planting, stealing, and distributing narcotics and firearms,25 and ongoing tension between police, the SAO, and the community. In March 2020, amid concerns about COVID transmission in detention centers, the SAO announced a non-prosecution policy for several low-level non-violent offenses, including illicit drug possession, possession of drug paraphernalia, prostitution, loitering/trespassing, minor traffic offenses, open container, rogue and vagabond, and public urination/defecation. The decision was endorsed by BPD leadership as a response to the pandemic,26 and formalized by the SAO one year later. While cannabis possession of up to 10g for any purpose was legislatively decriminalized in 2014, this policy represented a significant shift in its de facto extension to all illicit drugs. This study aimed to measure the impacts of implementing de facto decriminalization of all drugs via prosecutorial discretion for reducing carceral involvement among people who use drugs in Baltimore, Maryland. The hypothesis was that if the policy directive was adhered to by police, its implementation would be associated with reductions in drug possession arrests in the post-policy, relative to the pre-policy period, and that these reductions would be concentrated in the Black community. The broader goal of this study was to generate evidence informing drug and criminal legal policy in communities experiencing the intersecting crises of overdose and racialized policing in the US.
Methods Study Sample Individuals arrested in Baltimore City, Maryland between January 1, 2018 and December 31, 2021, and whose arrests appeared in Baltimore Police Department or Maryland Courts’ Judicial Information System, were included in the analysis. Arrest trends were compared before and after policy implementation to examine whether policy implementation in March 2020 was associated with reduced arrests overall and by race. Our primary outcome of interest was number of monthly arrests for drug or paraphernalia possession, excluding cannabis (‘possession arrests’). To account for broader changes to street activity and policing due to pandemic-related restrictions, arrests not covered by the policy change were examined (‘control’ crimes: robbery, assault, and guns/weapons charges). Measures Two distinct datasets were utilized to measure arrests trends. The first was publicly available arrest data (individual, de-identified record of arrest, date, time, location, charge description, and age, race and sex of arrestee) reported by the BPD through the Open Baltimore data initiative were utilized.27 These reflect all arrests made by police, irrespective of outcome, and are henceforth referred to as ‘street arrests.’ The second source of arrest data (incident-level records containing arrest date, charge description, age, race, sex and unique de-identified numeric code of arrestee) was accessed by the SAO through the Maryland Courts’ Judicial Information System (JIS) and acquired by the study team through a data sharing agreement. JIS data reflects arrests whose processing at central booking resulted in the advancement of the case, resulting in any combination of later diversion/dismissal, prosecution, conviction, and/or sentencing, and are referred to as ‘processed arrests.’ These data were considered markers of fundamentally different exposures; where street arrests are a marker of police encounters, processed arrests denote additional carceral exposure. All records from 1/1/2018 to 12/31/2021 were categorized by charge description. Drug arrests were coded as ‘possession of controlled dangerous substances’ or ‘possession of a controlled substance – marijuana.’ The latter category was excluded because cannabis was already decriminalized and qualitatively differs from other narcotics with respect to substance use disorder, overdose risk and treatment indications. Statistical analysis Interrupted time-series (ITS) modeling28,29 was conducted to evaluate changes to monthly street and processed arrests before and after the policy change (the “interruption”). ITS uses pre-interruption data to project trends in the post-interruption period, providing a counterfactual against which observed post-interruption data are compared. ITS models employed ordinary least squares estimation30,31 to produce linear trends in the pre-policy period (i.e., January 2018 to March 2020), the post-policy period (i.e., April 2020 to December 2021), and estimate the level of immediate change (i.e., change in monthly arrests between March – April 2020). A smoothing technique was used averaging data from one month before and after each timepoint to account for short-term fluctuations. Analyses were stratified by race (Black vs White/Hispanic/Asian/Other individuals). ITS was used to estimate pre- and post-interruption trends for each racial stratum and test for significant differences in their slopes. Due high missingness for race (59%) in the JIS data, analyses on racial disparities were not conducted on processed arrests. Race was reported for 100% of street arrests. The predicted number of street arrests averted was calculated for each group by subtracting observed from predicted street drug arrests post-interruption. To understand whether police continued to arrest PWUD using other charges (i.e., evidence of ‘replacement’ arrests), sensitivity analyses for i) all arrests; ii) arrests for possession with intent to distribute were conducted. Sensitivity analyses were only run using street arrests to examine police behaviors. All models were run in Stata/SE v.15.1. P-values below 0.05 were considered statistically significant. This study was approved by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board.
Results Table 1 Total arrests examined before and after de facto decriminalization in Baltimore, Maryland Reason for arrest Street arrests (Baltimore Police Data) Processed arrests (Maryland Court Data) Pre-policy period
(Jan 2018-Mar 2020) Post-policy period
(Apr 2020-Dec 2021) Pre-policy period
(Jan 2018-Mar 2020) Post-policy period
(Apr 2020-Dec 2021) Possession arrests Illicit drugs 6,570 820 2,959 131 Paraphernalia 13 0 59 0 ‘Control’ crimes Robbery 673 433 753 243 Assault 1,364 1,324 670 481 Weapons 2,469 2,247 220 41 Figure 1 Associations between de facto decriminalization policy and (A) street arrests and (B) processed arrests in Baltimore, Maryland Table 2 Associations between de facto decriminalization policy and arrests by category Baltimore, Maryland Reason for arrest Pre-policy
(January 2018 – March 2020) Post-policy
(April 2020 – December 2021) Slope (95% CI) p Slope (95% CI) p Drug/paraphernalia possession Street arrests -7.70 (-9.78, -5.62) <0.001 -0.90 (-1.54, -0.25) 0.01 Processed arrests 5.45 (2.81, 8.09) <0.001 -0.76 (-1.21, -0.30) 0.002 Robbery Street arrests -0.23 (-0.73, 0.28) 0.37 -0.11 (-0.56, 0.34) 0.61 Processed arrests -0.28 (-0.61, 0.05) 0.09 0.14 (-0.32, 0.59) 0.55 Assault Street arrests 0.20 (-0.24, 0.64) 0.37 1.24 (0.52, 1.96) 0.001 Processed arrests -0.43 (-0.60, -0.25) <0.001 -0.18 (-0.81, 0.46) 0.57 Guns/weapons Street arrests -1.01 (-1.74, -0.29) 0.01 -0.98 (-2.30, 0.34) 0.14 Processed arrests -0.40 (-0.59, -0.22) <0.001 0.01 (-0.14, 0.16) 0.86 Table 2 shows the slope (change in) arrests during the pre-policy period and the post-policy period. Boldface indicates statistical significance (P-values <0.05), denoting a positive increasing or decreasing trend in monthly arrests during the respective time period. A total of 66,964 street arrest records between January 1, 2018 to December 31, 2021 were extracted from the BPD's Open Data resource. The number of arrest records obtained is shown in Table 1 . Prior to policy implementation there were an average of 244 (s.d.=72.5) possession arrests per month ( Fig. 1 ), which were significantly declining by 7.7 arrests per month (p<0.001). The policy was associated with an immediate significant reduction in street drug arrests (p<0.001) between March and April 2020. After policy implementation, there were an average of 39 (s.d.=12.5) street arrests reported per month, which continued to decrease significantly decreasing by nearly 1 per month (p=0.01). Figure 1 depicts arrest trends; Table 2 shows model estimates before and after policy implementation. In contrast to street arrests, processed arrests averaged 111 (s.d.=63.7) per month in the pre-policy which were significantly increasing by 5.5 arrests per month (p<0.001). The policy was associated with an immediate significant reduction in processed possession arrests (p<0.001) between March and April 2020. After policy implementation there were an average of 6 (s.d.=6.2) possession arrests processed per month, continuing to decline by an average of 0.8 per month p=0.002). Differences in arrests for control crimes before and after the policy change are reported in Table 2 . The policy was not associated with decreased street arrests for robberies, first-degree assault, or weapons arrests between March and April 2020. After policy implementation, the average number of monthly street arrests remained stable for robberies and guns or other weapons and showed an increasing trend (p=0.01) for assaults. Similarly, policy implementation was not associated with a significant change in processed arrests for first-degree assault or weapons arrests between March and April 2020, and these remained stable during the post-policy period. There was a significant reduction in processed robbery arrests between March and April 2020, after which they remained stable. Figure 2 Racial differences in street possession arrests before and after prosecutorial policy change, comparing Black individuals to all other races in Baltimore, Maryland Street possession arrests are shown by race in Figure 2 . Before policy implementation, there were approximately five times more street possession arrests among Black individuals (N=5,512; 204 per month) than those of all other races (N=1,071; 40 per month). Arrests were significantly decreasing by 5.6/month among Black individuals (p<0.001) and 1.1/month among individuals of other races (p=0.001). The policy interruption was significantly associated with immediate declines in street possession arrests in both groups (p<0.001). After policy implementation, these continued to significantly decrease in both groups (Black: -0.74, p=0.02; other: -0.16, p=0.04). However, average monthly arrests were 12 times higher for Black individuals (36/month) than other races (3/month). Observed arrests (n=820) were subtracted from expected arrests (n=1,282) in the post-policy period, resulting in an estimated 462 possession arrests averted after policy implementation (a reduction of 36%). Of these, 362 arrests were estimated to be in the Black community, relative to 103 averted arrests among other individuals. To test whether reductions in possession arrests were offset by arrests for other charges, sensitivity analyses were run using street arrest data. If drug possession arrests were being replaced by other arrests, overall arrests should remain constant despite decreases in possession arrests. However, immediate, significant declines in overall arrests (p<0.001) were observed. No significant increases to arrests for drug distribution or other low-level misdemeanors were observed.
Discussion After decades of investment in punitive interdiction, jurisdictions across the US are considering decriminalization drug possession to usher substance use into the purview of public health. De facto decriminalization can be adopted expeditiously, without the encumbrance of the legislative process; however, evaluation of its impacts on carceral involvement is lacking. This study reports significant reductions in both police- and court-reported arrests for drug and paraphernalia possession after a de facto policy decriminalizing low-level drug offenses through a blanket non-prosecution order in Baltimore. Reductions continued over time and were concentrated in the Black community, an important finding given racial inequities in policing, incarceration, and overdose. Evidence on alternatives to drug criminalization in the US is largely limited to single substances (e.g., cannabis, buprenorphine)32,33. This study presents data from the Baltimore City's blanket non-prosecution policy for possession of all drugs and paraphernalia, which is to the authors’ knowledge among the first examples of comprehensive de facto decriminalization in the US. Findings indicate that, even without accompanying legislation and amid nuanced relationships between prosecutors and police, this directive was associated with substantial reductions in arrests at the street level and in such cases advancing through the court system. Demonstrating impacts on arrests is the first necessary step to evaluating downstream effects of decriminalization policies on health outcomes and racial equity. The same trends were not observed for other crime categories, suggesting that findings were not solely attributable to COVID-19-related changes to policing and prosecution. Nonetheless, the pandemic may have impacted offending and enforcement differentially by crime category, impeding causal inference of the policy's impact on arrests. No evidence of systematic criminalization of PWUD using different charge categories (i.e., ‘replacement arrests’) was detected, perhaps in part due to extension of the directive to other low-level crimes associated with poverty (e.g., loitering). Nationally, overall arrests, arrests for drug sales and overall prison admissions have decreased significantly over the past decade, but drug possession arrests have remained stable and exceed 1 million annually.34 Taken together with data from Baltimore and elsewhere35,36 refuting concerns that non-prosecution of low-level misdemeanors increases crime, findings indicate that de facto decriminalization of drug possession may be useful for reducing possession arrests expeditiously without compromising public safety. Two datasets were used to examine distinct domains of carceral exposure. Before policy implementation, street-level possession arrests were already significantly decreasing while those in the court system were significantly increasing. There are several plausible explanations. First, efforts to improve precision of police arrests pursuant to the consent decree may have resulted in a reduction of spurious street arrests and increased likelihood that those presented to central booking could be processed further. Second, where police data list the top reason for arrest, court data may reflect additional reasons for arrest for consideration at the charging stage. Third, criminal legal data quality is an ongoing challenge and lags in reporting, reporting bias, and random or systematic measurement error may be implicated here. Notably, after policy implementation, both datasets exhibited an immediate significant decline which continued to reduce thereafter, an important finding for characterizing the post-pandemic risk environment for PWUD in Baltimore where policing has previously been demonstrated to drive overdose risk37-39. The use of distinct indicators and data sources reported by different stakeholders served to triangulate and increase confidence in results. Black Americans are overrepresented in the carceral system and are experiencing steeper overdose increases than their white counterparts. Black individuals accounted for 82% of the observed decline and 78% of estimated averted street arrests following the policy change. Though declines were concentrated among them, Black residents had a smaller relative decline (86%) compared with others (95%), and the ratio of street arrests in Black to non-Black individuals increased from 5:1 to 12:1. This observation is consistent with research on other reforms intended to mitigate Drug War policies, including sentencing reform40,41, overdose Good Samaritan Laws42,43 and cannabis decriminalization44. Together these findings suggests that policy change may be necessary but insufficient for equity, particularly when implementation is reliant on police discretion.18,44 Reliance on police discretion may be especially pronounced under a de facto model, which does not inherently change the laws of policing on the books. Comparing these data to evaluations of de jure decriminalization (Oregon), and examining policy implementation and dissemination by police leadership, and knowledge and buy-in among rank-and-file officers, are important avenues for future study. Research additionally highlights neighborhood-level factors like racial composition, economic disadvantage, and gentrification in determining post-reform racial disparities45-48; further work is needed to elucidate the role of these factors here. Limitations This study has several limitations. Missingness in court data precluded examination of policy impacts by race among processed arrests. Police reported street arrest data may be subject to reporting bias and had quality issues: arrests records classified as ‘unknown’ were increasingly prevalent throughout the study period and may not have occurred at random thereby posing a threat to validity. Though possession arrests excluded those cannabis, coding schemes precluded further examination of arrest by drug type. Overlap of policy implementation with COVID-19 is a limitation for causal inference. Extension of analysis over time and replication in other settings with different implementation dates is necessary for understanding relative impacts of pandemic versus policy impacts on drug-related arrests.
Conclusions The public health and racial equity implications of continued investment in Drug War-era policies are dire. Progressive prosecution is increasingly being deployed to mitigate low-level drug arrests, but given recent backlash and political mobilization against this strategy49, evidence of its impacts is urgently needed. This work is an important and timely contribution to the nascent body of literature evaluating de facto drug decriminalization in the US; findings indicate it is a promising first step to reduce carceral exposure, which has been demonstrated to exacerbate overdose risks and health and racial inequities in society. However, widening racial disparities in the post-policy highlight the need for additional examination of police discretion in applying prosecutorial directives. Further research is also needed to assess arrest reductions translate to better health and social outcomes, which also rely on mobilization of other resources (e.g. harm reduction, substance use treatment, housing).
References 1 Overdose death rates increased significantly for Black, American Indian/Alaska Native people in 2020 [press release]. Centers for Disease Control and Prevention2022.
2 Opioid Use Disorder Treatment in Jails and Prisons. The Pew Charitable Trusts;2020.
3 Binswanger IA, Stern MF, Deyo RA, et al. Release from prison-a high risk of death for former inmates. N Engl J Med. 2007;356(2):157-165. DOI: 10.1056/NEJMsa064115
4 Correctional Populations in the United States. Office of Justice Programs, U.S. Department of Justice 1995.
5 Heather C. West WJS, Sarah J. Greenman. Prisoners in 2009. In: Bureau of Justice Statistics; 2011.
6 Simckes M, Willits D, McFarland M, McFarland C, Rowhani-Rahbar A, Hajat A. The adverse effects of policing on population health: A conceptual model. Soc Sci Med. 2021;281:114103. DOI: 10.1016/j.socscimed.2021.114103
7 Alang S, McAlpine D, McClain M. Police Encounters as Stressors: Associations with Depression and Anxiety across Race. Socius. 2021;7:2378023121998128. DOI:10.1177/2378023121998128
8 Miller TR, Lawrence BA, Carlson NN, et al. Perils of police action: a cautionary tale from US data sets. Inj Prev. 2017;23(1):27-32. DOI: 10.1136/injuryprev-2016-042023
9 Bor J, Venkataramani AS, Williams DR, Tsai AC. Police killings and their spillover effects on the mental health of black Americans: a population-based, quasi-experimental study. The Lancet. 2018;392(10144):302-310. DOI: 10.1016/S0140-6736(18)31130-9
10 How Mass Incarceration Makes Us All Sick. Health Affairs Blog 2021; 10.1377/hblog20210526.678786. DOI:10.1377/forefront.20210526.678786
11 Brinkley-Rubinstein L. Incarceration as a catalyst for worsening health. Health & Justice. 2013;1(1):3. DOI: 10.1186/2194-7899-1-3
12 Hinton E, Henderson L, Reed C. An unjust burden: The disparate treatment of Black Americans in the criminal justice system. Vera Institute of Justice. 2018:1-20.
13 Csete J, Kamarulzaman A, Kazatchkine M, et al. Public health and international drug policy. Lancet. 2016;387(10026):1427-1480. DOI: 10.1016/S0140-6736(16)00619-X
14 Virani HN, Haines-Saah RJ. Drug decriminalization: A matter of justice and equity, not just health. Am J Prev Med. 2020;58(1):161-164. DOI: 10.1016/j.amepre.2019.08.012
15 Earp BD, Lewis J, Hart CL. Racial Justice Requires Ending the War on Drugs. Am J Bioeth. 2021;21(4):4-19. DOI: 10.1080/15265161.2020.1861364
16 Stevens A, Hughes CE, Hulme S, Cassidy R. Depenalization, diversion and decriminalization: A realist review and programme theory of alternatives to criminalization for simple drug possession. Eur J Crim. 2022;19(1):29-54. DOI: 10.1177/1477370819887514
17 Greer A, Bonn M, Shane C, Stevens A, Tousenard N, Ritter A. The details of decriminalization: Designing a non-criminal response to the possession of drugs for personal use. Intl J of Drug Policy 2022;102:103605. DOI: 10.1016/j.drugpo.2022.103605
18 del Pozo B, Sightes E, Goulka J, et al. Police discretion in encounters with people who use drugs: operationalizing the theory of planned behavior. Harm Reduct J 2021;18(1):132. DOI: 10.1186/s12954-021-00583-4
19 Davis AJ. Reimagining prosecution: A growing progressive movement. UCLA Criminal Justice Law Review. 2019;3(1).
20 Substance-Related Intoxication Fatality Demographics. In: Maryland Opioid Operational Command Center; 2020.
21 Nellis A. The color of justice: Racial and ethnic disparity in state prisons. 2016.
22 New data reveals where people in Maryland prisons come from. 2022; https://www.prisonpolicy.org/blog/2022/06/27/mdorigin/
23 Justice Department Announces Findings of Investigation into Baltimore Police Department [press release]. Civil Rights Division, Department of Justice2016.
24 City of Baltimore Consent Decree. 2017; https://consentdecree.baltimorecity.gov/
25 Michael R. Bromwich JMW, Rachel B. Peck, Katherine M. Dubyak, William G. Fletcher, James M. Purce, Troy D. Shephard. Anatomy of the Gun Trace Task Force Scandal: Its Origins, Causes, and Consequences. 2022.
26 Anderson J. Baltimore Police officers to use discretion for low-level arrests to limit coronavirus exposure. 2020. https://www.baltimoresun.com/coronavirus/bs-md-police-coronavirus-policy-20200319-7iz2zuwmnbdz5pnrkbiq2r4dre-story.html
27 Baltimore Police Department Open Data: Arrests. In: Baltimore Police Department BP, ed2022. Available at: https://www.baltimorepolice.org/crime-stats/open-data . Access date: January 15, 2022
28 Hartmann DP, Gottman JM, Jones RR, Gardner W, Kazdin AE, Vaught RS. Interrupted time-series analysis and its application to behavioral data. J Appl Behav Anal. 1980;13(4):543-559. DOI: 10.1901/jaba.1980.13-543
29 Beard E, Marsden J, Brown J, et al. Understanding and using time series analyses in addiction research. Addiction. 2019;114(10):1866-1884. DOI: 10.1111/add.14643
30 Box GE, Jenkins GM, Reinsel GC, Ljung GM. Time series analysis: forecasting and control. John Wiley & Sons; 2015.
31 Velicer WF, Harrop J. The reliability and accuracy of time series model identification. Evaluation Review. 1983;7(4):551-560.
32 Pozo Bd, Krasner LS, George SF. Decriminalization of Diverted Buprenorphine in Burlington, Vermont and Philadelphia: An Intervention to Reduce Opioid Overdose Deaths. J Law Med Ethics. 2020;48(2):373-375. DOI: 10.1177/1073110520935353
33 Grucza RA, Vuolo M, Krauss MJ, et al. Cannabis decriminalization: A study of recent policy change in five U.S. states. Intl J of Drug Policy. 2018;59:67-75. DOI: 10.1016/j.drugpo.2018.06.016
34 Jake Horowitz JW. Drug Arrests Stayed High Even as Imprisonment Fell From 2009 to 2019. PEW Charitable Trust 2022.
35 Sherman SG, Tomko C, Rouhani S. De Facto Decriminalization for Drug Possession and Sex Work in Baltimore, Maryland. Am J of Prev Med. 2023; 64(4):567-568 DOI: 10.1016/j.amepre.2022.10.021
36 Agan AY DJ, Harvey A. Misemeanor Prosecution. National Bureau of Economic Research. 2022 (March 2021).
37 Latimore AD, Bergstein RS. "Caught with a body" yet protected by law? Calling 911 for opioid overdose in the context of the Good Samaritan Law. Int J Drug Policy. 2017;50:82-89. DOI: 10.1016/j.drugpo.2017.09.010
38 Beletsky L, Cochrane J, Sawyer AL, et al. Police Encounters Among Needle Exchange Clients in Baltimore: Drug Law Enforcement as a Structural Determinant of Health. Am J Public Health. 2015;105(9):1872-1879. DOI: 10.2105/AJPH.2015.302681
39 Flath N, Tobin K, King K, Lee A, Latkin C. Enduring Consequences From the War on Drugs: How Policing Practices Impact HIV Risk Among People Who Inject Drugs in Baltimore City. Subst Use Misuse. 2017;52(8):1003-1010. DOI: 10.1080/10826084.2016.1268630
40 Mooney AC, Giannella E, Glymour MM, et al. Racial/Ethnic Disparities in Arrests for Drug Possession After California Proposition 47, 2011–2016. Am J Public Health. 2018;108(8):987-993. DOI: 10.2105/AJPH.2018.304445
41 Lofstrom M, Martin B, Raphael S. Effect of sentencing reform on racial and ethnic disparities in involvement with the criminal justice system: The case of California's proposition 47. Crimino Public Policy. 2020;19(4):1165-1207. DOI: 10.1111/1745-9133.12527
42 Pamplin JR RS, Davis CS, King C, Townsend TS. Persistent criminalization and structural racism inc US drug policy: The case of overdose Good Samaritan laws. Am J Public Health. 2023;113(S1):S43-S48. DOI: 10.2105/AJPH.2022.307037
43 Rouhani S, Schneider KE, Rao A, et al. Perceived vulnerability to overdose-related arrests among people who use drugs in Maryland. Int J Drug Policy. 2021;98:103426. DOI: 10.1016/j.drugpo.2021.103426
44 Tran NK, Goldstein ND, Purtle J, et al. The heterogeneous effect of marijuana decriminalization policy on arrest rates in Philadelphia, Pennsylvania, 2009-2018. Drug Alcohol Depend 2020;212:108058. DOI: 10.1016/j.drugpo.2021.103426
45 Chamberlain AW, Boggess LN, Walker J. Drunk, Drugged and Disorderly: Examining the Non-Linear Effects of Neighborhood Racial Composition on Race-Specific Arrests. Justice Quarterly. 2022;39(6):1263-1286. DOI:10.1080/07418825.2021.2012234
46 Donnelly EA WJ, Anderson TL, O'Connell D. Revisiting Neighborhood Context and Racial Disparities in Drug Arrests Under the Opioid Epidemic. Race Justice. 2019;12(2). DOI: 10.1177/2153368719877222
47 Ramirez MM. City as borderland: Gentrification and the policing of Black and Latinx geographies in Oakland. Environ Plan D: Society and Space. 2019;38(1). DOI: 10.1177/0263775819843924
48 Collins CR, Stuart F, Janulis P. Policing gentrification or policing displacement? Testing the relationship between order maintenance policing and neighbourhood change in Los Angeles. Urban Stud. 2021;59(2):414-433. DOI: 10.1177/0042098021993354
49 Tensley EMB. What critics of progressive prosecution get wrong about crime spikes and the reform movement. 2022. https://www.cnn.com/2022/11/06/us/progressive-prosecution-midterms-reaj
CRediT authorship contribution statement Saba Rouhani: Conceptualization, Methodology, Formal analysis, Writing – original draft, Writing – review & editing. Catherine Tomko: Methodology, Validation, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Bradley E Silberzahn: Methodology, Validation, Data curation, Writing – review & editing. Noelle P Weicker: . Susan G Sherman: Conceptualization, Writing – review & editing, Supervision, Funding acquisition.
Acknowledgements SR received a pilot grant from Johns Hopkins University to conduct the study. CT is supported by the National Institute on Drug Abuse (T32DA007292). SGS is funded by an endowment from the Bloomberg American Health Initiative. Funders were not involved in study design, data management, analysis, and interpretation, or manuscript preparation and review. No financial disclosures were reported by the authors of this paper. SGS serves as an expert witness in opioid litigation. No other authors have conflicts or financial disclosures.