Weather and homicide in Hong Kong

Ambient temperature and intentional homicide: A multi-city case-crossover study in the US

BACKGROUND

There has been an increasing interest in the association between ambient temperature and violence and crime, in the context of global warming. We aimed to evaluate the association between daily ambient temperature and intentional homicide-a proxy for overall inter-personal violence.

METHODS

We collected daily weather and crime data from 9 large US cities (Chicago, Detroit, Fort Worth, Kansas City, Los Angeles, Louisville, New York, Tucson and Virginia Beach) from 2007 to 2017. A time-stratified case-crossover design was used. The associations were quantified by conditional logistic regression with distributed lag models, adjusting for relative humidity, precipitation and effects of public holidays. City-specific odds ratios (OR) were used to calculate the attributable fractions in each city.

RESULTS

Based on 19,523 intentional homicide cases, we found a linear temperature-homicide association. Every 5 °C increase in daily mean temperature was associated with a 9.5% [95% confidence interval (CI): 4.3-15.0%] and 8.8% (95% CI: 1.5-16.6%) increase in intentional homicide over lag 0-7 days in Chicago and New York, respectively. The association was not statistically significant in the other seven cities and seemed to be stronger for cases that happened during the hot season, at night (18:00-06:00) and on the street. During the study period, 8.7% (95%CI: 4.3-12.7%) and 7.1% (95% CI: 1.4-12.0%) intentional homicide cases could be attributed to temperatures above city-specific median temperatures, corresponding to 488 and 316 excess cases in Chicago and New York, respectively.

CONCLUSIONS

Our study suggests that the interpersonal violence might increase with temperature in some US cities. We also provide some insights into the mechanisms and targeted prevention strategies for heat-related violence.

Predictive Extrinsic Factors in Multiple Victim Shootings

Although researchers have found support for a relationship between temperature and violence and evidence of temporal patterns in violent crime, research on homicide shows less consistent results and no research on mass murder has been conducted. We address this by examining predictive factors in multi-victim shootings (those with four or more victims, including injured), a more general crime category than mass murder, but one with likely similar predictive factors. We used data from the Gun Violence Archive to understand the relationship between multi-victim shootings and temperature as well as other extrinsic factors. To avoid the confound between season and temperature, we employed temperature anomaly (the difference between actual and expected temperature) as a predictor of daily shooting rate. Using a generalized linear model for the daily count of multi-victim shootings in the U.S., we found that these events are significantly more frequent on weekends, some major holidays, hotter seasons, and when the temperature is higher than usual. Like other crimes, rates of multi-victim shooting vary systematically.

Prediction of crime occurrence from multi-modal data using deep learning

In recent years, various studies have been conducted on the prediction of crime occurrences. This predictive capability is intended to assist in crime prevention by facilitating effective implementation of police patrols. Previous studies have used data from multiple domains such as demographics, economics, and education. Their prediction models treat data from different domains equally. These methods have problems in crime occurrence prediction, such as difficulty in discovering highly nonlinear relationships, redundancies, and dependencies between multiple datasets. In order to enhance crime prediction models, we consider environmental context information, such as broken windows theory and crime prevention through environmental design. In this paper, we propose a feature-level data fusion method with environmental context based on a deep neural network (DNN). Our dataset consists of data collected from various online databases of crime statistics, demographic and meteorological data, and images in Chicago, Illinois. Prior to generating training data, we select crime-related data by conducting statistical analyses. Finally, we train our DNN, which consists of the following four kinds of layers: spatial, temporal, environmental context, and joint feature representation layers. Coupled with crucial data extracted from various domains, our fusion DNN is a product of an efficient decision-making process that statistically analyzes data redundancy. Experimental performance results show that our DNN model is more accurate in predicting crime occurrence than other prediction models.

The epidemiology of homicide in Italy by season, day of the week and time of day

Circadian and seasonal rhythms have an impact on the risk of violent death in humans, with more evidence available in suicide than in homicide. This study set out to investigate the pattern of deaths by homicide in Italy, taking into account the month, the day of the week and the time of day when the homicidal incident happened. Data based on official statistics on deaths by homicide (n = 3894) from 2003 to 2008 were analysed by spectral analysis and circular statistics techniques based on the Rayleigh test. Homicides showed seasonal distribution with a bimodal pattern, with summer (July and August) and winter (December and January) peaks over the other seasons. Homicides also peaked on Sundays and Mondays and were at their highest at night-time (00:00 to 5:59 hours). These rhythms in homicide victimization could be accounted for by social factors - within the framework of the routine activity theory - by psychological factors, including those leading to domestic violence, and by biological factors, including those impacting on impulse dyscontrol. The identification of risk or protective factors involved in the victim's surviving or succumbing to an assault might be important for the prevention of the worst outcome of injury at large, suicides and accidents included.

Accentuation of suicides but not homicides with rising latitudes of Greenland in the sunny months

BACKGROUND

Seasonal variation in suicides has been shown in many countries. We assessed the seasonality and the variation with latitude in suicides and homicides, and the impact of alcohol on the seasonality in suicides.

METHODS

Official computerized registers on causes of death in all Greenland during 1968-2002 were used. Sales data on beer from one of the major food store chains for July 2005-June 2006 were examined. Seasonal variation was assessed by Rayleigh's test for circular distributions.

RESULTS

There were a total of 1351 suicides and 308 homicides. The suicides rate varied from 4.2/100 000 person-years in 1971 to 128.4/100 000 person-years in 1987. The homicide rate varied from 2.1/100000 person-years in 1969-1970 to 34.8/100 000 person-years in 1988. Out of the 1351 suicides, 80.5% were committed by men and 19.5% by women. Median age was 25 years (n = 1351; Range 11-84 years). Violent methods of suicide were used in 95% of all cases (n = 1286). Out of the 308 homicide victims, 61% were men and 39% were women, and 13% were killed in multiple homicide events.There was a significant seasonal variation with peaks in June and troughs in the winter in all suicide cases (n = 1351, r = 0.07; Z = 7.58, p < 0.001), in violent suicides (n = 1286; r = 0.07; Z = 6.97; p < 0,001), in suicides in men (n = 1087; r = 0.07; Z = 5.39; p < 0.002) , and in women (n = 264; r = 0.10; Z = 2.36; p < 0.05), but not in homicides nor in consumption of beer. There was a bi-phasic seasonal variation in suicide victims where an alcohol-related condition was included in the death certificate. Suicides were more concentrated in the summer months north of the Arctic Circle (n = 577, r = 0.09, Z = 4.45, p < 0.01) than south of it (n = 769, r = 0.07, Z = 3.76, p < 0.002) and most concentrated in North Greenland (n = 33; r = 0.35; Z = 4.11; p < 0.01), where 48% of suicides occurred during the period of constant light. When including astronomical twilight in the constant light period 82% occurred during this time.

CONCLUSION

There was a concentration of suicides but not homicides in the summer months in all Greenland. The concentration was most pronounced at high latitudes.

Suicides in the midnight sun--a study of seasonality in suicides in West Greenland

Greenland is the most extreme of human habitats in regard to annual changes in natural light. From being very low, the rate of suicides in Greenland has increased during the past few decades to be among the highest in the world. Several studies have shown seasonality in suicides with spring or summer peaks, but this has not been previously shown in Greenland. Official data from 1968 to 1995 from West Greenland were pooled. A significant seasonality in the 833 suicides was found using Rayleigh's test, a peak in June and a trough in the winter. Suicides increased during the period, and reached high rates in most age groups, up to 577 per 100,000 person-years in men aged 15-24 years in 1990, and up to 147 per 100,000 person-years in women aged 15-24 years in 1995. Violent methods of suicide were used by 93%. Depression has been reported uncommonly. High alcohol intake and extended periods of light in the summer may contribute to impulsive-aggressive summer suicides.