Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Jan 28.
Published in final edited form as: Subst Use Misuse. 2016 Oct 21;52(2):214–222. doi: 10.1080/10826084.2016.1223689

The Spatial-Temporal Pattern of Policing Following a Drug Policy Reform: Triangulating Self-Reported Arrests With Official Crime Statistics

Tommi L Gaines 1, Daniel Werb 1,2, Jaime Arredondo 1, Victor M Alaniz 3, Carlos Vilalta 4, Leo Beletsky 1,5
PMCID: PMC5378066  NIHMSID: NIHMS850392  PMID: 27767367

Abstract

Background

In 2009, Mexico enacted a drug policy reform (Narcomenudeo) designed to divert persons possessing small amounts of illicit drugs to treatment rather than incarceration. To assess reform impact, this study examines the spatial-temporal trends of drug-related policing in Tijuana, Mexico post-enactment.

Method

Location of self-reported arrests (N=1,160) among a prospective, community-recruited cohort of people who inject drugs (PWID) in Tijuana (N=552) was mapped across city neighborhoods. Official police reports detailing drug-related arrests was triangulated with PWID self-reported arrests. Exploratory spatial data analysis examined the distribution of arrests and spatial association between both datasets across three successive years, 2011–2013.

Results

In 2011, over half of PWID reported being detained but not officially charged with a criminal offense; in 2013, 90% of arrests led to criminal charges. Official drug-related arrests increased by 67.8% (p < 0.01) from 2011 to 2013 despite overall arrest rates remaining stable throughout Tijuana. For each successive year, we identified a high degree of spatial association between the location of self-reported and official arrests (p < 0.05).

Conclusion

Two independent data sources suggest that intensity of drug law enforcement had risen in Tijuana despite the promulgation of a public health-oriented drug policy reform. The highest concentrations of arrests were in areas traditionally characterized by higher rates of drug crime. High correlation between self-reported and official arrest data underscores opportunities for future research on the role of policing as a structural determinant of public health.

Keywords: drug policy, police, drug laws, spatial-temporal analysis, Mexico

INTRODUCTION

From 2007–2012, Mexico experienced a surge in drug-related violence, which left over 100,000 individuals dead (Espinal-Enriquez & Larralde, 2015). As part of its response, in November 2009, the Mexican government passed a public health-oriented drug policy that reformed its handling of illicit drug use. Known as the Narcomenudeo reform (i.e., “drug dealing in very small quantities”), the federal policy sought to prioritize a harm reduction approach by decriminalizing the possession of small amounts of illicit drugs for personal consumption and diverting drug-using individuals to treatment instead of incarceration (Consejo Nacional Contras las Addicciones, 2010). Under the reform, police are not to arrest individuals found to be in possession of drugs under a specified threshold. Instead, police are required to report these individuals to health authorities and upon a third apprehension to refer these individuals to drug treatment. Thus, the reform is intended to free up Mexico’s limited law enforcement resources away from minor drug offenders and towards large-scale dealers and traffickers (Moreno, Licea, & Ajenjo, 2010; Werb et al., 2014).

For areas with a legacy of elevated drug-related violence, the drug policy reform provided the promise of a more balanced approach to the policing of drug markets while also improving public safety. These potential benefits held special relevance to locales situated along major drug trafficking routes, such as Tijuana, Mexico. However, this legislation came at a time of major law enforcement deployment in Mexico’s struggle with drug cartels, which precipitated human right violations throughout the country (Meyer, 2014). For example, in Tijuana, people who inject drugs (PWID) are generally affected by arbitrary and abusive policing practices with recent studies questioning whether the policing of low-level drug offenders had changed under Mexico’s drug policy reform (Beletsky et al., 2015; Gaines et al., 2015; Pinedo, Burgos, et al., 2015). More broadly, these studies reflect a considerable global evidence base highlighting pervasive implementation gaps in the enforcement of drug policies and other laws (Burris et al., 2004; Global Commission on Drug Policy, 2015; Kerr, Small, & Wood, 2005; Werb, Rowell, et al., 2011; Strathdee, Beletsky, Kerr, 2015).

Although intensified policing in places where drugs are consumed would appear to increase public safety, research has demonstrated that it inadvertently exacerbates health risk among PWID (Pollini et al., 2008; Werb et al., 2008; Strathdee et al., 2011; Volkmann et al., 2011). However, these and other studies (Cooper, Bossak, Tempalski, Jarlais, & Friedman, 2009; Cooper, Des Jarlais, et al., 2012) have generally relied on either self-report or official crime statistics to understand the role of place, in terms of the physical setting, on the health of PWID. To our knowledge, no studies have combined multiple sources of data that report on drug-related policing activities even though different sources of data have the potential to fill in gaps regarding physical spaces and its association with negative health outcomes for PWID. It is rare and difficult for information related to criminalized activity, such as illicit drug use, to be collected (Peters, Kremling, & Hunt, 2015); and geographic data recording such criminalized activity are seldom triangulated from multiple data sources (Hibdon & Groff, 2014).

Accordingly, the current study has two major goals. First, we examine the spatial-temporal patterns of drug-related arrests across three successive years (2011–2013) in Tijuana, Mexico, as states and municipalities were provided a one-year grace period to implement the reform. Based on local news reports of police raids in places with street-based PWID (Nieves, 2014; Archibold, 2014; Quiones, 2014), we hypothesize that policing practices around low-level drug and nondrug related offenses increased (rather than decreased) during the 3-year period, and that drug-related arrests would be highest in neighborhoods with open-air drug markets. Second, we sought to determine whether official crime statistics, provided through police activity reports, could approximate self-reports of drug users’ encounters with police, given the potential methodological implications of employing official crime statistics as an alternative to survey data collection. We hypothesize a spatial overlap between the two data sources based on prior studies demonstrating greater police presence in places frequented by PWID in other settings (Cooper et al., 2009; Cooper et al., 2012).

METHODS

Units of Analysis

The units of analysis for this study are colonias or administrative neighborhood units (n=724) in Tijuana that vary in size and population density. According to the 2010 Mexican Census, there are 1.5 million people in Tijuana with a median of 5,220 people per square kilometer per colonia (interquartile range: 2,155 – 8,935). Approximately 13% (n=96) of the colonias have a population count less than 10, and these colonias are generally located in rural or highly industrialized areas. Colonias are nested within larger geographic regions called delegaciones; administrative boundaries that divide Tijuana into nine boroughs that also vary in land area (range: 17.8 km2 – 103.7 km2) and population size (83,886 – 445,934). The current study was approved by the Institutional Review Board of the University of California, San Diego and the Ethics Board of the Colegio de la Frontera Norte, Tijuana.

Data Sources

Self-Reported Arrest Data

Beginning in 2011, PWID residing in Tijuana were enrolled into a prospective cohort study (Proyecto El Cuete IV) examining the social epidemiology of HIV and other infectious disease. The methods have been described elsewhere (Robertson et al., 2014) but in brief, individuals were recruited in neighborhoods throughout Tijuana where PWID were known to reside. Eligibility criteria included being 18 years or older; being a Spanish or English speaker; having injected illicit drugs within the past month; having no plans to move from the city in the next 24 months; currently not participating in other intervention studies; and providing informed consent. At baseline and every 6 months thereafter, participants completed an interviewer-administered survey. Those who reported being arrested for any offense were asked to provide the date and geographic location of the respective arrest. Location was confirmed by showing participants Google Street View maps to identify the exact geographical coordinates of the arrest, which were then recorded (Beletsky et al., 2015). In the current study, a total of 552 cohort participants reported 1,160 arrest incidents within Tijuana city limits from January 1, 2011 to December 31, 2013. These self-reported arrests were aggregated to the colonia-level for each respective year and used in subsequent analysis.

Official Crime Statistics

Police activity reports reflecting the number of drug possession stops and arrests that included the possession, sale, and/or distribution of illegal drugs, was obtained from the Tijuana Municipal Police Department, hereafter referred to as official drug-related arrests. The department provided this information in tabular form detailing the monthly number of official drug-related arrests by colonia covering the period from January 1, 2011 to December 31, 2013, as part of a Memorandum of Understanding between the Tijuana Municipal Police Department and the Division of Global Public Health at the University of California San Diego. We converted the textual information identifying colonias with an arrest into a digital geographic representation using digital maps (i.e., shapefiles) provided by the Tijuana Municipal Planning Institute. There were a total of 8,727 official drug-related arrests recorded in the police activity reports from 2011–2013 of which 8,025 were successfully matched to a colonia using the maps provided by the Tijuana Municipal Planning Institute.

Dependent Variable

The dependent variable is the number of self-reported and official drug-related arrests at the colonia-level occurring from 2011–2013 identified from the community-recruited cohort of PWID and the Tijuana Municipal Police Department.

Exploratory Spatial Data Analysis

We applied multiple analytic methods to examine the spatial-temporal patterns of self-report and official drug-related arrests. First, we described the distribution of self-reported arrests among the cohort of PWID in terms of arrest frequency, reason for arrest, and criminal charges resulting from the arrest. Similarly, we generated descriptive statistics on official drug-related arrests by calculating arrest rates at the delegacion-level. We chose not to construct an arrest rate at the colonia-level since arrest rates would have been artificially inflated by including colonias with smaller population sizes (i.e., population < 10).

Second, we applied a GIS-based analysis to identify hotspots of official drug-related arrests (i.e., counts of drug-related arrests per year). The Getis-Ord Gi* statistics measures statistically significant spatial clustering of high values (i.e., hotspots of arrests) or low values (i.e., coldspots of arrests) by returning a z-score for each colonia. Large z-scores (GiZScore > 1.96; p<0.05) identified clusters of hotspots while low z-scores (GIZScore < −1.96; p< 0.05) identified clusters of coldspots. Further, limiting the analysis to colonias identified as hotspots, we examined trends in the monthly counts of official arrests from 2011–2013 using a generalized linear model with a negative binomial distribution and Newey West standard errors to account for serial autocorrelation and heteroscedasticity of the error terms (Newey & West, 1987). The purpose of this analysis was to address our first hypothesis of determining whether the same neighborhoods, particularly those known to have open air drug markets, were consistently identified as hotspots of drug-related arrests during the study period. Regression analysis was carried out in STATA 11 (StataCorp, LP, College Station, TX, USA).

Third, we assessed the spatial-temporal relationship of official drug-related arrests between 2011 and 2013 using the Global Bivariate Moran’s I statistic generated in GeoDa (Anselin, Syabri, & Kho, 2006). This statistic quantifies spatial autocorrelation by determining to what extent the number of official drug-related arrests in one location at one point in time is linearly associated with a weighted average of official drug-related arrests of neighboring locations at a different point in time (i.e., spatial lag of arrests). Significant autocorrelation (p<0.05) indicated spatial clustering of official drug-related arrests over time.

Lastly, we analyzed spatial similarities between self-reported and official arrests by layering the geographical coordinates of PWID self-reported arrest incidents onto the map generated from the hotspot analysis of official drug-related arrests. For each successive year, the Pearson correlation coefficient was calculated to measure the degree of association between the number of self-reported arrests and official drug-related arrests within the same colonia. The Global Bivariate Moran’s I statistic was then calculated to determine whether self-reported arrests and official drug-related arrests spatially clustered such that a high (or low) number of arrests in both data sets were located in and around the same colonias; statistical significance of the Moran’s I statistic was based on a permutation test (p<0.05). The purpose of this analysis was to address our second hypothesis of determining whether police activity reports, could approximate self-reports of drug users’ encounters with police. The spatial analyses were carried out in ArcMap 10.1 and GeoDA.

RESULTS

Patterns of Self-Reported Arrests

Of the 552 PWID arrested from 2011–2013, 38.8% (n=214) were arrested once, 29.7% (n=164) were arrested twice, and 31.5% (n=174) were arrested three to five times during the study period; this resulted in a total of 1,160 arrest incidents. Approximately 14% (n=78) of PWID were arrested and charged with possession of drugs or drug paraphernalia over the 3-year period but none had a drug-related apprehension prior to their initial drug-related arrest (data not shown due to small cell sizes).

The distribution of the 1,160 arrest incidents reported by PWID is shown in Table 1. In 2011, 42% of the self-reported arrests were detainments in which PWID were apprehended by the police, perceivably often lacking any official reason, and eventually released with no subsequent criminal charges. But in 2013, arrest incidents overwhelmingly led to criminal charges with the majority (70.3%) being linked to a minor offense. Approximately 14% (n=159) of all self-reported arrests from 2011–2013 involved drugs or drug paraphernalia possession (Table 1). Half of these arrests (53%, n = 85) led to charges of drug possession, loitering, trespassing, disturbing the peace, or lack of proper identification whereas the remaining incidents were charged with some other criminal offenses (11%, n = 18) or detainment (35%, n = 56).

Table 1.

Distribution of PWID Self-Reported Arrests in Tijuana from 2011–2013 (n=1,160)

Variable Year
2011
n=323
% (n)		 2012
n= 382
% (n)		 2013
n = 455
% (n)
Type of offense
 No Charge 57.7 (186) 42.9 (164) 10.3 (47)
 Minor a 26.6 (86) 31.4 (120) 70.3 (320)
 Drug or drug paraphernalia possession 6.2 (20) 8.4 (32) 6.8 (31)
 Other b 9.6 (31) 17.3 (66) 12.5 (57)
Drugs or Needle possession at time of arrest 16.4 (53) 18.6 (71) 7.7 (35)
Open in a new tab
a

arrests for loitering, trespassing, disturbing the peace, and lack of proper identification

b

arrests for weapon possession, public intoxication, stealing, resisting arrest, violating parole, routine check, being in a conflict zone, and drug selling/trafficking.

Patterns of Official Arrest Rates

Table 2 displays the rate of official drug-related arrests from 2011 to 2013 in Tijuana, Mexico. Overall, rates increased during this period from 162.5 per 100,000 in 2011 to 272.7 per 100,000 in 2013. The highest rate of drug possession arrests were reported in the Centro delegacion (an area with several open air drug markets) with arrests ranging from 532.7 in 2011 to 820 in 2013 per 100,000.

Table 2.

Official drug-related arrest rate per 100,000 population in Tijuana, Mexico from 2011–2013

Delegacion Population Land Area Km2 No. of Colonias 2011 2012 2013
Centenario 140,291 37.9 75 144.7 233.1 171.9
Centro 119,388 24.9 86 532.7 589.7 820.0
Cerro Colorado 112,517 23.5 52 111.1 153.8 200.9
La Mesa 150,300 30.5 156 140.4 233.5 320.7
Mesa de Otay 83,886 17.8 41 426.8 441.1 243.2
Playas de Tijuana 138,837 61.7 95 67.0 167.8 206.0
San Antonio de los Buenos 211,100 62.5 68 123.2 175.3 267.7
Sanchez Taboada 191,740 42.6 73 121.5 207.6 199.8
La Presa 445,934 103.7 78 105.6 172.7 219.9
Overall 1,593,993 405.2 724 162.5 231.87 272.7
Open in a new tab

Hotspots of Official Arrests

At the colonia-level, we observed significant hotspots of official drug-related arrests, most of which were located along the Mexico-U.S. border and primarily in the Centro delegacion, which also had the highest number of self-reported arrests by PWID (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Hotspot analysis on the number of official drug-related arrests by colonia with the point location of PWID self-reported arrest for any offense in Tijuana, Mexico, 2011–2013

The number of hotspots varied over the 3-year period and accounted for 22.6% to 30.9% of all the official drug-related arrests in Tijuana (Table 3). Within hotspots, the number of official drug-related arrests increased annually and this trend was significant over the study period according to the negative binomial regression [βyear(s.e.) = 0.11(0.32); p < 0.01]. In contrast, the number of overall official arrests occurring within hotspots, including violent and property offenses, decreased over the 3-year period from 3,545 in 2011 (or 24.7% of all arrests in Tijuana) to 2,478 (or 16% of all arrest in Tijuana) and this decreasing trend was significant [βyear(s.e.) = 0.18(0.32); p < 0.01].

Table 3.

Number of official arrests and official drug-related arrests from 2011–2013

Characteristics Year
2011 2012 2013
Total number of arrestsa in Tijuana 14,659 20,283 15,459
Total number of drug-related arrests in Tijuana 1,985 2,634 3,406
Number of colonias identified as hotspots of drug-related arrests 21 23 12
Total arrests in hotspots 3,545 3,116 2,478
Total drug-related arrests in hotspots 613 710 769
Open in a new tab
a

Includes arrests rape, murder, assault, robbery, gun possession, burglary, auto theft, domestic violence, human trafficking, and others.

In total, 24 colonias were identified as hotspots of official drug-related arrests with over four-fifths (n=20) of these colonias observed as hotspots across multiple years. Of these 20 colonias, 45% experienced an increase in drug-related arrests, 35% had a decrease, and 20% experienced no change. Notably, the decrease of drug-related arrests was not substantial since arrests within these colonias declined by 1 to 4 arrests (the median decrease was 1). However, the increase of drug-related arrests was quite noticeable, with colonias experiencing as little as 1 additional arrest to as many as 202 additional arrests (the median increase was 7); and according to the Wilcoxon matched-pairs signed rank test, the differences between 2011 and 2013 were significant (p = 0.012). None of the colonias met the criteria of being a significant coldspot (i.e., all colonias had a GiZScore > −1.96).

Spatial Autocorrelation of Official and Self-Reported Arrests

Results from the global bivariate Moran’s I demonstrate a significant and positive spatial autocorrelation (Table 4) indicating that the pattern of official drug-related arrests at the colonia-level are associated across space and over time (i.e., arrests in one colonia are associated with arrests in neighboring colonias over time). Pearson’s correlation coefficients measuring the association of arrest between official and PWID self-reported data were estimated as 0.785, 0.710, and 0.851 for 2011, 2102, and 2013, respectively, suggesting moderate to high levels of correlation between the two data sources. The global bivariate Moran’s I further indicated significant spatial clustering of arrest patterns between self-reported and official arrest data (Table 4); colonias with a high number of self-reported arrests were surrounded by colonias with similarly high number of official drug-related arrests.

Table 4.

Spatial autocorrelation of official drug-related arrests and spatial association between self-reported and official arrests, 2011–2013

Variable Spatial Laga Bivariate Moran’s I p-value
Spatial-Temporal Association
Drug-related arrests in 2011 Drug-related arrest in 2012 0.1809 0.004
Drug-related arrests in 2011 Drug-related arrests in 2013 0.1680 0.008
Drug-related arrests in 2012 Drug-related arrests in 2013 0.1544 0.006
Spatial Dependency
Self-reported arrest in 2011 Drug-related arrest in 2011 0.2076 0.005
Self-reported arrest in 2012 Drug-related arrest in 2012 0.1942 0.006
Self-reported arrest in 2013 Drug-related arrest in 2013 0.1959 0.005
Open in a new tab
a

Spatial lag reflects an average of the number of arrests in neighboring colonias

DISCUSSION

This study found that the policing of drug users and the location of drug-related arrests were consistently concentrated in a small number of colonias in Tijuana and that this spatial pattern remained stable over three successive years. In general, colonias (i.e., neighborhoods) with a high number of self-reported arrests were also characterized by a high number of official drug-related arrests. Further, although police recorded a similar number of arrests for violent and property offenses over the 3-year period, we found that the number of drug-related arrests increased almost 2-fold during this timeframe. In 2011, 13.5% of all arrests in Tijuana were drug-related; in 2013, two years after the drug decriminalization reform was enacted, 22% of all arrests were drug-related.

These findings have two major implications. First, they suggest that despite passage of the Narcomenudeo reform, which prioritizes a public health response to drug-related harms in Mexico, drug-related policing appears to have intensified, particularly in neighborhoods with known street-based drug markets. Although the concentration of drug crimes within a limited number of neighborhoods is a commonly observed phenomenon in Mexico (Espinal-Enriquez & Larralde, 2015; Espinosa & Rubin, 2015; Vilalta, 2009) and elsewhere (Rengert & Lockwood, 2009), the remarkable consistency in the arrest patterns reported by PWID, in both hotspots and non-hotspots of official drug-related arrests, may indicate that police are using discretionary tools to maintain public order. Notably, less than half of the arrests reported by PWID led to criminal charges in 2011. However, this changed markedly in 2013 when 90% of self-reported arrests resulted in a criminal offense that was most often minor (i.e., loitering, trespassing, disturbing the peace, and lacking proper identification). Further, PWID who possessed drugs or drug paraphernalia at the time of their arrest were more likely to be charged with a crime than they were to be released by police, despite the Narcomenudeo reform that focuses on diverting drug-dependent individuals to treatment.

Collectively, these analyses parallel our prior findings that the tenets of the Narcomenudeo reform have not been meaningfully adopted by law enforcement in Tijuana (Beletsky et al., 2015; Werb et al., 2015). These results also highlight the challenge of implementing a drug policy reform where the “law on the books” can drastically differ from policing practices on the street (Burris et al., 2004). It may be that in an attempt to reduce more serious crimes, Tijuana police adopted a zero-tolerance approach by aggressively policing low-level nondrug related offenses in places with high drug activity (e.g., among PWID, using city ordinances to criminalize minor offenses such as loitering and vagrancy). This is concerning since intensified policing in places where drugs are consumed can present additional barriers to accessing medical care and harm reduction services among PWID, as has been observed in Tijuana (Pinedo, Burgos, Ojeda, Fitzgerald, & Ojeda, 2015) and elsewhere (Hammett et al., 2014).

Apprehending drug users has been a policing tactic used in other settings to disrupt illicit drug markets and maintain order and safety (Dixon & Coffin, 1999; Wagner, Simon-Freeman, & Bluthenthal, 2013; Werb et al., 2011). Although this might seem to deter criminal behavior, evidence shows that leaning on stringent law enforcement tactics does not reduce drug-related crime (Braga, Welsh, & Schnell, 2015) and instead may displace drug markets (Kerr et al., 2005). Further, although we found that hotspots of official drug-related arrests remained concentrated in the same colonias from 2011–2013, prior evidence suggests that “police sweeps” in places where PWID reside and use drugs may lead to displacement, thus increasing risk behaviors for blood-borne infections in new settings (Brouwer et al., 2012). Therefore, effective implementation of the Narcomenudeo reform necessitates better police management and training that has shown promise in other drug use settings (Beletsky, Agrawal, et al., 2011).

To that end, our research team is currently implementing a police education program to align policing practices with HIV prevention efforts in Tijuana (Strathdee et al., 2015). Although the initial results are positive (Beletsky et al., 2016), institutional barriers, including an inadequate system that keeps track of prior drug offenses, insufficient expertise to assess drug dependence, and lack of funding for scaling up addiction treatment and policing, must be addressed to ensure that Tijuana’s criminal justice system has the resources to appropriately respond and implement the drug policy reform (Werb et al., 2015).

The second major finding is the potential utility of geospatial data from official crime statistics to identify places where high risk activity may be occurring and where interventions designed to align police practices with public health should be prioritized. Although surveys of illicit drug users provide the most accurate assessment of their encounters with law enforcement, the costs and limitations of surveying such hard-to-reach populations (Rudolph et al., 2011) can be prohibitive in many settings (Spooner & Flaherty, 1993). Official crime statistics, by contrast, are routinely collected in many settings at the population level but likely oversamples highly marginalized subgroups of PWID (i.e., homeless) and under samples others (i.e., drug users that regularly migrate across settings) (Linton, Jennings, Latkin, Gomez, & Mehta, 2014). Nevertheless, our results are promising in that we observed a modest and positive spatial autocorrelation in the number of arrests reported by PWID and those reported by police. In general, the number of self-reported arrests increased in places where the number of official drug-related arrests increased.

Given that official crime statistics are generally accessible to the public through freedom of information initiatives, this potentially represents an opportunity to save resources by foregoing primary research with high risk groups, which can be difficult and resource intensive. Further, as found in this study, different data sources can provide similar information of where the policing of drug-related crime is occurring and as a result can be useful for evaluating drug policies that impact populations of PWID. Consequently, official crime statistics should be considered among those investigating HIV risk associated with drug use and the criminal justice system and particularly among those who are working in settings in which self-reported PWID data are not available.

This study is subject to limitations consistent with the use of self-reported and criminal justice data. First, official police activity reports did not distinguish between arrest for possession versus sale or distribution of drugs and therefore we cannot conclude which, if any, type of drug-related arrest was responsible for the overall increase. Second, approximately 8% (n=702) of all the official drug-related arrests could not be linked to a colonia due to police activity reports providing locations that were not geographically identifiable in the maps obtained from the Tijuana Municipal Planning Institute. Nevertheless, these data did provide insight into which colonias experienced an increase in drug-related policing activities following the drug policy reform, which we also found to be positively correlated with PWID self-reported arrests. Third, while we had access to official police activity reports from the Tijuana police department, we note that other law enforcement agencies (i.e., federal, state, and military police) were also policing drug-related activity in Tijuana during the study period. Our analysis, though, was likely unaffected given that less than 5% (n=47) of the sample of PWID self-reported arrests were carried out by federal or state agencies, suggesting that the majority of street-level policing was being conducted by the Tijuana Municipal Police Department during the 3-year period. Lastly, the spatial patterns of both self-reported and official arrests we observed are limited to Tijuana and cannot be generalized to other parts of Mexico for which the federal drug policy reform applies.

In summary, although this study is more descriptive and exploratory, rather than explanatory, its strength is in providing a spatial-temporal analysis of drug-related arrest patterns that could help to improve the management and accountability of police departments, as well as provide confirmation of the utility of official crime statistics to triangulate self-reported data collected among PWID. This, in turn, holds promise for improving the reliability of future studies examining the legal and policing environment shaping drug-related harms.

Acknowledgments

This research was supported by NIDA grants K01DA034523, R37 DA019829, R01DA028692, DP2 DA040256-01; the Canadian Institutes of Health Research MOP-79297; and the Fogarty International Center of the National Institutes of Health under Award Number D43TW008633. Mr. Alaniz and Dr. Vilalta are participating in this research of their own capacity. The findings and conclusions in this report are those of the authors and do not necessarily reflect the views of the National Institutes of Health, Canadian Institute of Health, Fogarty International, the Tijuana Municipal Police Department or the Centro de Investigacion y Docencia Economica.

Footnotes

Declaration of Interest: The authors report no conflicts of interest. The authors along are responsible for the content and writing of the article.

References

  1. Anselin L, Syabri I, Kho Y. GeoDa: An introduction to spatial data analysis. Geographical Analysis. 2006;38(1):5–22. [Google Scholar]
  2. Archibold RC. As Mexican Border Town Tries to Move On, Some are Stuck in Limbo. [Accessed on July 10, 2015];New York Times Report, November 27, 2014. 2014 from http://www.nytimes.com/2014/11/28/world/americas/a-border-limbo-for-the-displaced-at-home-in-mexico-no-more-.html.
  3. Beletsky L, Agrawal A, Moreau B, Kumar P, Weiss-Laxer N, Heimer R. Police training to align law enforcement and HIV prevention: Preliminary evidence from the field. American Journal of Public Health. 2011;101(11):2012–2015. doi: 10.2105/AJPH.2011.300254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Beletsky L, Arredondo J, Rolón ML, Patiño E, Rocha T, Abramovitz D, … Strathdee SA. Shifting Police Practices to Support HIV Prevention: Initial Findings from an Assessment of a Police Education Program in a Mexican Setting. 21th International AIDS Conference (AIDS 2016).2016. [Google Scholar]
  5. Beletsky L, Vera A, Gaines TL, Arredondo J, Werb D, Bañuelos A, … Strathdee SA. Utilization of Google Earth to Georeference Survey Data among People who Inject Drugs: Strategic Application for HIV Research. 8th International AIDS Society Conference on HIV Pathogenesis, Treatment & Prevention; July 21, 2015; Vancouver, Canada. 2015. [Google Scholar]
  6. Beletsky L, Wagner K, Arredondo J, Palinkas L, Rodríguez CM, Kalic N, … Strathdee SA. Implementing Mexico’s “Narcomenudeo” drug law reform: A mixed-methods assessment of early experiences among people who inject drugs. Journal of Mixed Methods. 2015:1–18. [Google Scholar]
  7. Braga AA, Welsh BC, Schnell C. Can policing disorder reduce crime? A systematic review and meta-analysis. Journal of Research in Crime and Delinquency. 2015;52(4):567–588. [Google Scholar]
  8. Brouwer KC, Rusch ML, Weeks JR, Lozada R, Vera A, Magis-Rodriguez C, Strathdee SA. Spatial epidemiology of HIV among injection drug users in Tijuana, Mexico. Annals of the Association of American Geographers. 2012;102(5):1190–1199. doi: 10.1080/00045608.2012.674896. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Burris H, Blankenship KM, Donoghoe M, Sherman S, Vernick JS, Case P, … Koester S. Addressing the “risk environment” for injection drug users: The mysterious case of the missing cop. The Milbank Quarterly. 2004;82(1):125–156. doi: 10.1111/j.0887-378X.2004.00304.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Consejo Nacional Contras las Addicciones. [Accessed on April 1, 2014];Normatividad y legislacion. 2010 at http://www.conadic.salud.gob.mx/interior/normas.html.
  11. Cooper HLF, Des Jarlais DC, Tempalski B, Bossak BH, Ross Z, Friedman SR. Drug-related arrest rates and spatial access to syringe exchange programs in New York City health districts: Combined effects on the risk of injection-related infections among injectors. Health & Place. 2012;18(2):218–228. doi: 10.1016/j.healthplace.2011.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cooper HLF, Bossak B, Tempalski B, Jarlais DCD, Friedman SR. Geographic approaches to quantifying the risk environment: Drug-related law enforcement and access to syringe exchange programmes. International Journal of Drug Policy. 2009;20(3):217–226. doi: 10.1016/j.drugpo.2008.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dixon D, Coffin P. Zero tolerance policing of illegal drug markets. Drug and Alcohol Review. 1999;18(4):477–486. [Google Scholar]
  14. Espinal-Enriquez J, Larralde H. Analysis of Mexico’s narco-war network (2007–2011) Plos One. 2015;10(5):15. doi: 10.1371/journal.pone.0126503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Espinosa V, Rubin DB. Did the military interventions in the Mexican drug war increase violence? American Statistician. 2015;69(1):17–27. [Google Scholar]
  16. Gaines TL, Beletsky L, Arredondo J, Werb D, Rangel G, Vera A, Brouwer K. Examining the spatial distribution of law enforcement encounters among people who inject drugs after implementation of Mexico’s drug policy reform. Journal of Urban Health-Bulletin of the New York Academy of Medicine. 2015;92(2):338–351. doi: 10.1007/s11524-014-9907-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Global Commission on Drug Policy. [Accessed on March 15, 2016];The negative impace of drug control on public health: The global crisis of avoidable pain. 2015 at http://www.globalcommissionondrugs.org/reports/
  18. Hammett TM, Phan S, Gaggin J, Case P, Zaller N, Lutnick A, … Jarlais DCD. Pharamcies as providers of expanded health services for people who inject drugs: A review of laws, policies, and barriers in six countries. BMC Health Services Research. 2014;14:261. doi: 10.1186/1472-6963-14-261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hibdon J, Groff ER. What you find depends on where you look: Using emergency medical services call data to target illicit drug use hot spots. Journal of Contemporary Criminal Justice. 2014;30(2):169–185. [Google Scholar]
  20. Kerr T, Small W, Wood E. The public health and social impacts of drug market enforcement: A review of the evidence. International Journal of Drug Policy. 2005;16(4):210–220. [Google Scholar]
  21. Linton SL, Jennings JM, Latkin CA, Gomez MB, Mehta SH. Application of space-time scan statistics to describe geographic and temporal clustering of visible drug activity. Journal of Urban Health-Bulletin of the New York Academy of Medicine. 2014;91(5):940–956. doi: 10.1007/s11524-014-9890-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Meyer M. Mexico’s police. Many reforms little progress. [Accessed on August 12, 2014];Washington Office on Latin America; 2014. 2014 at http://www.wola.org/sites/default/files/Mexicos%20Police.pdf.
  23. Moreno JGB, Licea JAI, Ajenjo CR. Tackling HIV and drug addiction in Mexico. Lancet. 2010;376(9740):493–495. doi: 10.1016/S0140-6736(10)60883-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Newey WK, West KD. A simple, positive semidfinite, heteroskedasticity and autocorrelation consistent covariance-matrix. Econometrica. 1987;55(3):703–708. [Google Scholar]
  25. Nieves R. Stuck between two countries. [Accessed on July 10, 2015];CNN Report, February 22, 2014. 2014 from http://www.cnn.com/2014/02/21/us/u-s-mexico-border-purgatory/
  26. Peters RH, Kremling J, Hunt E. Accuracy of self-reported drug use among offenders: Findings from the arrestee drug abuse monitoring-II program. Criminal Justice and Behavior. 2015;42(6):623–643. [Google Scholar]
  27. Pinedo M, Burgos JL, Ojeda AV, Fitzgerald D, Ojeda V. The role of visual markers in police victimization among structurally vulnerable persons in Tijuana, Mexico. International Journal of Drug Policy. 2015;26(5):501–508. doi: 10.1016/j.drugpo.2014.08.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Pinedo M, Burgos JL, Zuniga ML, Perez R, Macera CA, Ojeda VD. Police victimization among people who inject drugs along the U.S.-Mexico border. Journal of Studies on Alcohol and Drugs. 2015;76(5):758–763. doi: 10.15288/jsad.2015.76.758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pollini RA, Brouwer KC, Lozada RM, Ramos R, Cruz MF, Magis-Rodriguez C, … Strathdee SA. Syringe possession arrests are associated with receptive syringe sharing in two Mexico-US border cities. Addiction. 2008;103(1):101–108. doi: 10.1111/j.1360-0443.2007.02051.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Quiones S. Tijuana, Mexicans Deported by U.S. Struggle to Find ‘Home’. [Accessed on July 10, 2015];National Geographic Report, November 21, 2014. 2014 from http://news.nationalgeographic.com/news/special-features/2014/11/141121-tijuana-deportees-immigrants-mexico-border/
  31. Rengert GF, Lockwood B. Geographical units of analysis and the analysis of crime. Putting Crime in Its Place: Units of Analysis in Geographic Criminology. 2009:109–122. [Google Scholar]
  32. Robertson AM, Garfein RS, Wagner KD, Mehta SR, Magis-Rodriguez C, Cuevas-Mota J, … Stahr Evaluating the impact of Mexico’s drug policy reforms on people who inject drugs in Tijuana, BC, Mexico, and San Diego, CA, United States: A binational mixed methods research agenda. Harm Reduction Journal. 2014;11:7. doi: 10.1186/1477-7517-11-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rudolph AE, Crawford ND, Latkin C, Heimer R, Benjamin EO, Jones KC, Fuller CM. Subpopulations of illicit drug users eeached by targeted street outreach and respondent-driven sampling strategies: Implications for research and public health practice. Annals of Epidemiology. 2011;21(4):280–289. doi: 10.1016/j.annepidem.2010.11.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Spooner C, Flaherty B. Comparison of 3 data collection-methodologies for the study of young illicit drug-users. Australian Journal of Public Health. 1993;17(3):195–202. [PubMed] [Google Scholar]
  35. Strathdee SA, Lozada R, Martinez G, Vera A, Rusch M, Nguyen L, … Patterson TL. Social and structural factors associated with HIV infection among female sex workers who inject drugs in the Mexico-US border region. Plos One. 2011;6(4):8. doi: 10.1371/journal.pone.0019048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Strathdee SA, Arredondo J, Rocha T, Abramovitz DA, Rolón ML, Patiño ME, … Beletsky L. Implementation of a police education program to integrate occupational safety and HIV prevention: Protocol for a modified stepped-wedge study design with parallel prospective cohorts to assess behavioral outcomes. BMJ Open. 2015;5(8):e008958. doi: 10.1136/bmjopen-2015-008958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Strathdee SA, Beletsky L, Kerr T. HIV, drugs and the legal environment. International Journal of Drug Policy. 2015;26:S27–S32. doi: 10.1016/j.drugpo.2014.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Vilalta CJ. La geografía local del narcomenudeo: Patrones, procesos y recomendaciones de política urbana. Estudios Demográficos y Urbanos. 2009;24(1):49–77. [Google Scholar]
  39. Volkmann T, Lozada R, Anderson CM, Patterson TL, Vera A, Strathdee SA. Factors associated with drug-related harms related to policing in Tijuana, Mexico. Harm Reduction Journal. 2011;8:7. doi: 10.1186/1477-7517-8-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wagner KD, Simon-Freeman R, Bluthenthal RN. The association between law enforcement encounters and syringe sharing among IDUs on skid row: A mixed methods analysis. AIDS and Behavior. 2013;17(8):2637–2643. doi: 10.1007/s10461-013-0488-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Werb D, Mora MEM, Beletsky L, Rafful C, Mackey T, Arredondo J, Strathdee SA. Mexico’s drug policy reform: Cutting edge success or crisis in the making? International Journal of Drug Policy. 2014;25(5):823–825. doi: 10.1016/j.drugpo.2014.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Werb D, Rowell G, Guyatt G, Kerr T, Montaner J, Wood E. Effect of drug law enforcement on drug market violence: A systematic review. International Journal of Drug Policy. 2011;22(2):87–94. doi: 10.1016/j.drugpo.2011.02.002. [DOI] [PubMed] [Google Scholar]
  43. Werb D, Strathdee SA, Meza E, Range-Gomez MG, Palinkas L, Medina-Mora ME, Beletsky L. Institutional stakeholder perceptions of barriers to addiciton treatment under Mexico’s drug policy reform. Global Public Health. 2015 doi: 10.1080/17441692.2015.1093524. In Press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Werb D, Wood E, Small W, Strathdee S, Li K, Montaner J, Kerr T. Effects of police confiscation of illicit drugs and syringes among injection drug users in Vancouver. International Journal of Drug Policy. 2008;19(4):332–338. doi: 10.1016/j.drugpo.2007.08.004. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES