Factors influencing temporal patterns in crime in a large American city: A predictive analytics perspective

Does darkness increase the risk of certain types of crime? A registered report protocol

Evidence about the relationship between lighting and crime is mixed. Although a review of evidence found that improved road / street lighting was associated with reductions in crime, these reductions occurred in daylight as well as after dark, suggesting any effect was not due only to changes in visual conditions. One limitation of previous studies is that crime data are reported in aggregate and thus previous analyses were required to make simplifications concerning types of crimes or locations. We will overcome that by working with a UK police force to access records of individual crimes. We will use these data to determine whether the risk of crime at a specific time of day is greater after dark than during daylight. If no difference is found, this would suggest improvements to visual conditions after dark through lighting would have no effect. If however the risk of crime occurring after dark was greater than during daylight, quantifying this effect would provide a measure to assess the potential effectiveness of lighting in reducing crime risk after dark. We will use a case and control approach to analyse ten years of crime data. We will compare counts of crimes in 'case' hours, that are in daylight and darkness at different times of the year, and 'control' hours, that are in daylight throughout the year. From these counts we will calculate odds ratios as a measure of the effect of darkness on risk of crime, using these to answer three questions: 1) Is the risk of overall crime occurring greater after dark than during daylight? 2) Does the risk of crime occurring after dark vary depending on the category of crime? 3) Does the risk of crime occurring after dark vary depending on the geographical area?

Climate attribution of interpersonal violence: International evidence

Anthropogenic climate change is increasingly threatening interpersonal violence, yet global evidence for related impacts and potential transmission mechanisms remains limited. We examine whether and how climate change, particularly climate extremes, affects interpersonal violence. Using the panel data of 140 countries and regions from 2000 to 2019, we find that hot and wet extremes precipitated increase in homicide rates globally. Economic level, inequality, and resources scarcity were important intermediaries through which climate extremes affected homicide, while the direct effects still dominated the total effects. We then reveal the heterogeneous effects of climate extremes, further suggesting that poor countries and regions with relatively small contributions to climate change were particularly sensitive to climate extremes. These findings elucidate a strong climate-violence link, helping explain implications of facilitating violence prevention and mitigating climate change.

Investigating the impact of the COVID-19 pandemic on crime incidents number in different cities

The COVID-19 pandemic is strongly affecting many aspects of human life and society around the world. To investigate whether this pandemic also influences crime, the differences in crime incidents numbers before and during the pandemic in four large cities (namely Washington DC, Chicago, New York City and Los Angeles) are investigated. Moreover, the Granger causal relationships between crime incident numbers and new cases of COVID-19 are also examined. Based on that, new cases of COVID-19 with significant Granger causal correlations are used to improve the crime prediction performance. The results show that crime is generally impacted by the COVID-19 pandemic, but it varies in different cities and with different crime types. Most types of crimes have seen fewer incidents numbers during the pandemic than before. Several Granger causal correlations are found between the COVID-19 cases and crime incidents in these cities. More specifically, crime incidents numbers of theft in Washington DC, Chicago and New York City, fraud in Washington DC and Los Angeles, assault in Chicago and New York City, and robbery in Los Angeles and New York City, are significantly Granger caused by the new case of COVID-19. These results may be partially explained by the Routine Activity theory and Opportunity theory that people may prefer to stay at home to avoid being infected with COVID-19 during the pandemic, giving fewer chances for crimes. In addition, involving new cases of COVID-19 as a variable can slightly improve the performance of crime prediction in terms of some specific types of crime. This study is expected to obtain deeper insights into the relationships between the pandemic and crime in different cities, and to provide new attempts for crime prediction during the pandemic.

Holiday effect on injuries sustained by assault victims seen in US emergency departments

PURPOSE

There has been a limited and inconsistent analysis of assault-related injury patterns associated with holidays. We investigated the temporal variation in assault-related injuries presenting to US emergency departments (ED) around holidays.

METHODS

We examined data from the National Electronic Injury Surveillance System Database between 2005 and 2017 for six categories of assault-related injuries: altercation, sexual assault, robbery, intimate partner violence (IPV), other specified, and unknown. Differences between holiday and non-holiday periods were analyzed for each assault type.

RESULTS

There was a significant difference in overall assault-related injury visits between holiday and non-holiday periods (p < 0.00001). Of over 21 million assault-related injury visits, 14.9% occurred during holiday periods and 85% during non-holiday periods. The difference between the daily number of assault-related ED visits was also significantly higher during the holiday period than baseline non-holiday period (p < 0.00001). Altercations and IPV were significantly higher than baseline for New Year's Eve (highest), St. Patrick's Day, July 4th, and Labor Day. IPV also remained significantly higher than baseline during Thanksgiving and Christmas. Sexual assaults were significantly higher than baseline during the New Year's Eve period but lower than baseline during Christmas and Easter.

CONCLUSIONS

Holidays are associated with increase in several assault-related injuries. The information can help allocate healthcare resources and guide prevention strategies.

Assault related injury visits in US emergency departments: An analysis by weekday, month and weekday-by-month

This study investigated the temporal variation (month and day) in assault-related injuries presenting to the US Emergency Departments (ED). An IRB exempt, retrospective review of prospectively collected data was performed using the National Electronic Injury Surveillance System Database from 2005 to 2017 for six categories of assault-related injuries-altercation, sexual assault, robbery, intimate partner violence (IPV), other specified assaults and unknown assault types. National estimates of injuries and associated variables were obtained using SUDAAN software, followed by cosinor analyses for the variation of month and weekday of injury. Three-dimensional topographic representations for weekday-by-month analyses were also created. Over this 13-year time span, there were more than 21 million injury visits due to assault, accounting for 6.57% of all ED visits. While there was no change in the incidence of total number of assaults over the study period, there was a significant increase in the annual percentage incidence of IPV (1.17%; p = .0094) and robbery (2.56%; p = .0001). Cosinor analyses demonstrated a mid-summer peak for all assault types except for robberies (late summer). All assault types showed a weekend peak (late Saturday or early Sunday). Topographical contours exhibited a peak in July and August on early Sunday for all assaults, however the month varied by the type of assault, with weekend peaks in the spring and winter for IPV. This information can be used in prospective resource planning for management and prevention strategies.

Utilizing Text Mining, Data Linkage and Deep Learning in Police and Health Records to Predict Future Offenses in Family and Domestic Violence

Family and Domestic violence (FDV) is a global problem with significant social, economic, and health consequences for victims including increased health care costs, mental trauma, and social stigmatization. In Australia, the estimated annual cost of FDV is $22 billion, with one woman being murdered by a current or former partner every week. Despite this, tools that can predict future FDV based on the features of the person of interest (POI) and victim are lacking. The New South Wales Police Force attends thousands of FDV events each year and records details as fixed fields (e.g., demographic information for individuals involved in the event) and as text narratives which describe abuse types, victim injuries, threats, including the mental health status for POIs and victims. This information within the narratives is mostly untapped for research and reporting purposes. After applying a text mining methodology to extract information from 492,393 FDV event narratives (abuse types, victim injuries, mental illness mentions), we linked these characteristics with the respective fixed fields and with actual mental health diagnoses obtained from the NSW Ministry of Health for the same cohort to form a comprehensive FDV dataset. These data were input into five deep learning models (MLP, LSTM, Bi-LSTM, Bi-GRU, BERT) to predict three FDV offense types ("hands-on," "hands-off," "Apprehended Domestic Violence Order (ADVO) breach"). The transformer model with BERT embeddings returned the best performance (69.00% accuracy; 66.76% ROC) for "ADVO breach" in a multilabel classification setup while the binary classification setup generated similar results. "Hands-off" offenses proved the hardest offense type to predict (60.72% accuracy; 57.86% ROC using BERT) but showed potential to improve with fine-tuning of binary classification setups. "Hands-on" offenses benefitted least from the contextual information gained through BERT embeddings in which MLP with categorical embeddings outperformed it in three out of four metrics (65.95% accuracy; 78.03% F1-score; 70.00% precision). The encouraging results indicate that future FDV offenses can be predicted using deep learning on a large corpus of police and health data. Incorporating additional data sources will likely increase the performance which can assist those working on FDV and law enforcement to improve outcomes and better manage FDV events.

Human Responses and Adaptation in a Changing Climate: A Framework Integrating Biological, Psychological, and Behavioural Aspects

Climate change is one of the biggest challenges of our times. Its impact on human populations is not yet completely understood. Many studies have focused on single aspects with contradictory observations. However, climate change is a complex phenomenon that cannot be adequately addressed from a single discipline’s perspective. Hence, we propose a comprehensive conceptual framework on the relationships between climate change and human responses. This framework includes biological, psychological, and behavioural aspects and provides a multidisciplinary overview and critical information for focused interventions. The role of tipping points and regime shifts is explored, and a historical perspective is presented to describe the relationship between climate evolution and socio-cultural crisis. Vulnerability, resilience, and adaptation are analysed from an individual and a community point of view. Finally, emergent behaviours and mass effect phenomena are examined that account for mental maladjustment and conflicts.

Exploring the Immediate Effects of COVID-19 Containment Policies on Crime: an Empirical Analysis of the Short-Term Aftermath in Los Angeles

This work investigates whether and how COVID-19 containment policies had an immediate impact on crime trends in Los Angeles. The analysis is conducted using Bayesian structural time-series and focuses on nine crime categories and on the overall crime count, daily monitored from January 1st 2017 to March 28th 2020. We concentrate on two post-intervention time windows-from March 4th to March 16th and from March 4th to March 28th 2020-to dynamically assess the short-term effects of mild and strict policies. In Los Angeles, overall crime has significantly decreased, as well as robbery, shoplifting, theft, and battery. No significant effect has been detected for vehicle theft, burglary, assault with a deadly weapon, intimate partner assault, and homicide. Results suggest that, in the first weeks after the interventions are put in place, social distancing impacts more directly on instrumental and less serious crimes. Policy implications are also discussed.

QSAR without borders

Prediction of chemical bioactivity and physical properties has been one of the most important applications of statistical and more recently, machine learning and artificial intelligence methods in chemical sciences. This field of research, broadly known as quantitative structure-activity relationships (QSAR) modeling, has developed many important algorithms and has found a broad range of applications in physical organic and medicinal chemistry in the past 55+ years. This Perspective summarizes recent technological advances in QSAR modeling but it also highlights the applicability of algorithms, modeling methods, and validation practices developed in QSAR to a wide range of research areas outside of traditional QSAR boundaries including synthesis planning, nanotechnology, materials science, biomaterials, and clinical informatics. As modern research methods generate rapidly increasing amounts of data, the knowledge of robust data-driven modelling methods professed within the QSAR field can become essential for scientists working both within and outside of chemical research. We hope that this contribution highlighting the generalizable components of QSAR modeling will serve to address this challenge.

The Impact of Climate Change on Mental Health: A Systematic Descriptive Review

BACKGROUND

Climate change is one of the great challenges of our time. The consequences of climate change on exposed biological subjects, as well as on vulnerable societies, are a concern for the entire scientific community. Rising temperatures, heat waves, floods, tornadoes, hurricanes, droughts, fires, loss of forest, and glaciers, along with disappearance of rivers and desertification, can directly and indirectly cause human pathologies that are physical and mental. However, there is a clear lack in psychiatric studies on mental disorders linked to climate change.

METHODS

Literature available on PubMed, EMBASE, and Cochrane library until end of June 2019 were reviewed. The total number of articles and association reports was 445. From these, 163 were selected. We looked for the association between classical psychiatric disorders such as anxiety schizophrenia, mood disorder and depression, suicide, aggressive behaviors, despair for the loss of usual landscape, and phenomena related to climate change and extreme weather. Review of literature was then divided into specific areas: the course of change in mental health, temperature, water, air pollution, drought, as well as the exposure of certain groups and critical psychological adaptations.

RESULTS

Climate change has an impact on a large part of the population, in different geographical areas and with different types of threats to public health. However, the delay in studies on climate change and mental health consequences is an important aspect. Lack of literature is perhaps due to the complexity and novelty of this issue. It has been shown that climate change acts on mental health with different timing. The phenomenology of the effects of climate change differs greatly-some mental disorders are common and others more specific in relation to atypical climatic conditions. Moreover, climate change also affects different population groups who are directly exposed and more vulnerable in their geographical conditions, as well as a lack of access to resources, information, and protection. Perhaps it is also worth underlining that in some papers the connection between climatic events and mental disorders was described through the introduction of new terms, coined only recently: ecoanxiety, ecoguilt, ecopsychology, ecological grief, solastalgia, biospheric concern, etc.

CONCLUSIONS

The effects of climate change can be direct or indirect, short-term or long-term. Acute events can act through mechanisms similar to that of traumatic stress, leading to well-understood psychopathological patterns. In addition, the consequences of exposure to extreme or prolonged weather-related events can also be delayed, encompassing disorders such as posttraumatic stress, or even transmitted to later generations.

Homicides during the Barranquilla Carnival, Colombia: A 10 Year Time-Series Analysis

Homicides are currently the third leading cause of death among young adults, and an increase has been reported during holidays. The aim of the present study was to explore whether an association exists between Carnival in Barranquilla, Colombia, and an increase in homicides in the city. We used mortality records to identify the number of daily homicides of men and women throughout the week of Carnival, and we compared those with records from all of standard days between 1 January 2005 and 31 December 2015. Conditional fixed-effects models were used, stratified by time and adjusted by weather variables. The average number of homicides on Carnival days was found to be higher than on a standard day, with an OR of 2.34 (CI 95%: 1.19–4.58) for the occurrence of at least one male homicide per day during Carnival, and 1.22 (CI 95%: 1.22–7.36) for female homicides, adjusted by weather variables. The occurrence of homicides during Carnival was observed and was similar to findings for other holidays. Given that violence is a multifactorial phenomenon, the identification of the factors involved serves as a basis for evaluating whether current strategies have a positive effect on controlling it.