A modified two-process Knox test for investigating the relationship between law enforcement opioid seizures and overdoses

Mohler, George; Mishra, Swapnil; Ray, Bradley; Magee, Lauren A.; Huynh, Philip; M, Canada; O’Donnell, Daniel; Flaxman, Seth

doi:10.1098/rspa.2021.0195

Cited by 10 publications

(8 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the Knox test, we use parameters

δ = 250

m and

τ = 21

days for the Knox statistic

κ false(τ, δ false)

(similar to the values used in Mohler et al, (2021)). In Figure 9, we see similar agreement between the

L

functions of random permutations, SOP permutations and the

L

function of the data.…”

Section: Methodsmentioning

confidence: 99%

“…Data consist of latitude, longitude, date and time of the incident and come from Indianapolis, Indiana, from 1 July 2018 to 31 December 2018. A space-time positive association between law enforcement drug seizures and overdose was observed in Mohler et al (2021). One hypothesis is that when law enforcement officers make an arrest for drug dealing and seize drugs, users may need to seek out alternative sources and may be at higher risk of overdose in the near future.…”

Section: Association Between Drug Seizures and Overdose In Indianapolismentioning

confidence: 99%

“…For example, in Klauber (Klauber, 1971), a two‐sample randomization test was introduced to test for interactions between underground nuclear tests and subsequent earthquakes. Similar methods have been applied to test for interactions between firearm arrests and shootings (Wyant et al, 2012), homicide and other violent or property crimes (Flaxman et al, 2013) and law enforcement drug seizures and overdose (Mohler et al, 2021). In a more general multivariate fashion, the interest is in detecting and modelling cross‐interactions among any two marginal processes, as shown in Jalilian et al (2015) and Waagepetersen et al (2016).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Second‐order preserving point process permutations

Mohler

Mateu

2023

Stat

Self Cite

View full text Add to dashboard Cite

While random permutations of point processes are useful for generating counterfactuals in bivariate interaction tests, such permutations require that the underlying intensity be separable. In many real‐world datasets where clustering or inhibition is present, such an assumption does not hold. Here, we introduce a simple combinatorial optimization algorithm that generates second‐order preserving (SOP) point process permutations, for example, permutations of the times of events such that the function of the permuted process matches the function of the data. We apply the algorithm to synthetic data generated by a self‐exciting Hawkes process and a self‐avoiding point process, along with data from Los Angeles on earthquakes and arsons and data from Indianapolis on law enforcement drug seizures and overdoses. In all cases, we are able to generate a diverse sample of permuted point processes where the distribution of the functions closely matches that of the data. We then show how SOP point process permutations can be used in two applications: (1) bivariate Knox tests and (2) data augmentation to improve deep learning‐based space‐time forecasts.

show abstract

“…For the Knox test, we use parameters

δ = 250

m and

τ = 21

days for the Knox statistic

κ false(τ, δ false)

(similar to the values used in Mohler et al, (2021)). In Figure 9, we see similar agreement between the

L

functions of random permutations, SOP permutations and the

L

function of the data.…”

Section: Methodsmentioning

confidence: 99%

Section: Association Between Drug Seizures and Overdose In Indianapolismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Second‐order preserving point process permutations

Mohler

Mateu

2023

Stat

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hence, the model allows capturing how the intensity of the process elevates locally (in space and time) after the occurrence of each crime event. In addition, the use of such models has allowed the development of new versions of the Knox test that take into account the self‐exciting nature of crime data (Mohler et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

A mechanistic bivariate point process model for crime pattern analysis

Briz‐Redón

Mateu

2023

Stat

View full text Add to dashboard Cite

The statistical analysis of crime data has gained attention in the last decade. In particular, the availability of spatio‐temporal crime data at the event level allows us to model the incidence of crime with high precision. Point process models are the natural tool to study crime patterns. As it is well‐known that crime events often spread as a contagion process, mechanistic self‐exciting models are usually considered in this context. In this paper, we propose a mechanistic bivariate spatio‐temporal model for the first‐order intensity function of the point processes associated with the intensity of two crime types. Specifically, the model includes separate estimates of the overall temporal and spatial intensities of crime and a spatio‐temporal interaction term for each of the crime types under analysis. Regarding the spatio‐temporal term, we model how the occurrence of previous crime events (from any of the two types) influences the intensity of each type of crime under study. We consider a dataset of crime events recorded in Valencia (Spain) during the year 2017 and focus on two crime types for the analysis: property crime and robbery. The results show that there is an association between the recent occurrence of either property crimes or robberies and the intensity of both crime types. Several spatio‐temporal monitoring tools are described and discussed as well.

show abstract

“…Overdose events geographically cluster and likewise (Marshall et al 2020 ; Mohler et al 2021 ), firearm injury events cluster in (Branas et al 2017 ; Braga et al 2009 ) relatively small areas of cities (i.e., hot spots) (Magee 2020 ; Koper et al 2015 ). Thus, research on firearm injury, including homicide, has consistently focused on neighborhood conditions which include economic disadvantage (including deprivation and a lack of resources), residential instability, racial segregation, and the prevalence of firearm ownership, aggregated to neighborhood levels (i.e., census tract, county level), as causal mechanisms (Hepburn and Hemenway 2004 ; Sampson et al 1997 ).…”

Section: Introductionmentioning

confidence: 99%

Dual public health crises: the overlap of drug overdose and firearm injury in Indianapolis, Indiana, 2018–2020

et al. 2022

View full text Add to dashboard Cite

Background Drug overdose and firearm injury are two of the United States (US) most unrelenting public health crises, both of which have been compounded by the COVID-19 pandemic. Programs and policies typically focus on each epidemic, alone, which may produce less efficient interventions if overlap does exist. The objective is to examine whether drug overdose correlates with and is associated with firearm injury at the census tract level while controlling for neighborhood characteristics. Methods An ecological study of census tracts in Indianapolis, Indiana from 2018 to 2020. Population rates per 100,000 and census tracts with the highest overlap of overdose and firearm injury were identified based on spatial clusters. Bivariate association between census tract characteristic and drug overdose and firearm violence rate within spatial clusters. Zero-inflated negative binominal regression was used to estimate if the drug overdose activity is associated with higher future firearm injury. Results In high overdose—high firearm injury census tracts, rates of firearm injury and drug overdose are two times higher compared to city wide rates. Indicators of structural disadvantage and structural racism are higher in high overdose—high firearm injury census tracts compared to city-wide averages. Drug overdoses are associated with higher rates of firearm injury in the following year (IRR: 1.004, 95% CI 1.001, 1.007, p < 0.05), adjusting for census tract characteristics and spatial dependence. Conclusions Drug overdose and firearm injury co-spatially concentrate within census tracts. Moreover, drug overdoses are associated with future firearm injury. Interventions to reduce firearm injuries and drug overdoses should be a co-response in high drug overdose—high firearm injury communities.

show abstract

A modified two-process Knox test for investigating the relationship between law enforcement opioid seizures and overdoses

Cited by 10 publications

References 56 publications

Second‐order preserving point process permutations

Second‐order preserving point process permutations

A mechanistic bivariate point process model for crime pattern analysis

Dual public health crises: the overlap of drug overdose and firearm injury in Indianapolis, Indiana, 2018–2020

Contact Info

Product

Resources

About