|Abstract / Summary
The thesis is based on five studies conducted in cooperation between the Accident Analysis Group at Odense University Hospital, the Institute of Clinical Research at University of Southern Denmark, the Institute of Forensic Medicine at University of Southern Denmark, the National Centre for Register-based Research at University of Aarhus, and Odense Police. Studies 1-4 Objectives The objectives of the thesis were: 1. To describe the epidemiology of deliberate physical interpersonal violence and repeat violent victimization in the Odense Municipality 1991-2002. 2. To estimate the lifetime risk of violent victimization in the Odense Municipality. 3. To describe the demographic and socioeconomic status of adult victims of violence in the Odense Municipality 1991-2002 compared to population-based controls using a case-control design. Methods and material Included were all victims of violence from the Odense Municipality attending the A&E department at Odense University Hospital and/or subjected to medicolegal autopsy at the Institute of Forensic Medicine at University of Southern Denmark 1991-2002. Based on mid-year population counts age and gender specific annual incidence rates were estimated and Poisson regression was used to analyse changes in the study period. Time trends in lesion types, injured body regions, weapon use, and mortality rates were used as indicators of changes in severity of violence in the study period. Using the Danish Civil Person Registration Number and including a five year run-in period all victims with more than one contact to the A&E department and/or the Institute of Forensic Medicine were identified as repeat victims and recurrent proportions were estimated as the proportion of individuals with more than one contact due to violence in the study period. Kaplan-Meier survival analyses were made with time of observation from first to second contact due to violence. A repetition percentage was estimated as the proportion with repeated characteristics of the incident leading to the second contact compared to the incident leading to the first contact. Based on age and gender specific annual incidence rates of first time contact with the A&E department and/or IFM and the survival rates of the background population the lifetime risk of contact with the A&E department and/or the IFM through exposure to violence was estimated. For every adult victim aged 20 years or more five age, gender, and date matched controls were randomly selected from the total population of Odense Municipality. Based on the Danish Civil Person Registration Number demographic and socioeconomic information about cases and controls was extracted from two national longitudinal registers. Different models of demographic and socioeconomic risk factors were constructed and unconditional logistic regression analyses were performed. Results In the study period 14,316 victims attended the A&E department at OUH and/or were subjected to medicolegal autopsy at the IFM. The gender specific annual incidence rate was 9.9 (95% CI: 9.7-10.1) per 1000 population/year for males and 3.4 (95% CI: 3.3-3.5) per 1000 population/year for females. For males the rate decreased significantly and for females the rate was unchanged in the study period (Poisson regression, p=0.001 and p=0.588). For both males and females the percentage of less serious lesions increased in the study period (nonparametric test for trend, p=0.01) whereas the percentage of bone fractures and deep lesions decreased significantly for both males (nonparametric test for trend, p=0.01 and p=0.02) and females (nonparametric test for trend, p=0.02 and p=0.03). Three per cent of the victims were stabbed with a knife and 0.4 % had lesions from firearms. The percentages did not change in the study period (nonparametric test for trend, p=0.91 and p=0.94). Overall 22 % of the individuals were repeat victims who were responsible for 44 % of all contacts in the study period. For individuals with repeat contacts the median number of contacts in the study period was two (range 2-15). The median time span from first to the second contact was 1.75 years (range 0-11.44 years) for males and 1.64 years (range 1.00-9.91) for females. Within four years 79 % of the victims had had the next contact. The median time span from index to the second contact decreased significantly with increasing age. The repetition percentage for weapon use was 17 % for males and 7 % for females. Potentially severe lesions such as bone fractures and internal lesions occurred with a repetition percentage of 28 % and 19 % for males and 23 % and 25 % for females. The estimated lifetime risk was 0.440 (95 % CI: 0.438-0.442) for males and 0.180 (95% CI: 0.178-0.181) for females. In the adolescence (15-24 years) the cumulative risk was 0.192 (95% CI: 0.188.0.195) for males, and 0.048 (95% CI: 0.046-0.051) for females respectively. The risk factors most strongly associated with A&E department and/or IFM contact due to adult violent victimization were: not living with a partner (OR: 2.93; 95% CI: 2.74-3.13), an annual household gross income of less than 10,000 € (OR: 2.38; 95% CI: 2.16-2.63), a high lifetime unemployment rate (OR: 1.92; 95% CI: 1.82-2.02), an annual household gross income between 10,000 € and 19,999 € (OR: 1.87; 95% CI: 1.71-2.04), a household crowding of less than 24 m2 per resident (OR: 1.86; 95% CI: 1.68-2.05), an age difference to partner of five years or more (OR:1.56; 95% CI: 1.43-1.69), and being a pensioner (OR: 1.49; 95% CI: 1.36-1.63). The most important factors negatively associated with A&E department and/or IFM contact due to adult violent victimization were: being a student (OR: 0.23; 95% CI: 0.20-0.26), being self-employed or a manager (OR: 0.32; 95% CI: 0.29-0.35), being a skilled or unskilled worker (OR: 0.45; 95% CI: 0.42-0.48). Foreign citizenship in a country outside Europe was not significantly associated with A&E department and/or IFM contact due to adult violent victimization (OR: 1.08; 95% CI: 0.96-1.22). Study 5 Objective 1. To describe the completeness of A&E department data for the illumination of the frequency of recorded violence in Odense Municipality. Methods and material Included were all victims of violence from the Odense Municipality attending the A&E department at OUH and/or recorded by Odense Police due to violent victimization occurring in the period from 1 March 2003 to 29 February 2004. Data from the A&E department and the police were compared and the completeness of data was evaluated by estimating the proportion of all registered violence (by both the A&E department and the police) that was registered at the A&E department and by the police in the community. Results In the one-year study period 2003-2004 1403 victims of violence from Odense Municipality attended the A&E department at OUH and/or were recorded by the police. A total of 87 % of the male victims and 82 % of the female victims were registered at the A&E department (Fisher’s exact, p=0.015), and 35 % of the male victims and 44 % of the female victims were registered in the police records (Fisher’s exact, p=0.003). The proportion of victims attending the A&E department and also recorded by the police was 26 % for males and 31 % for females, respectively (Fisher exact, p=0.048). Conclusion Based on A&E department and IFM data the current study did not reveal any evidence of an increase in the incidence of deliberate interpersonal violence in the study period 1991-2002. Furthermore, we found no evidence supporting the prevailing view that injuries from deliberate interpersonal violence have become more severe. The overall recurrence proportion was 22 % and repeat victims were responsible for 44 % of all contacts in the twelve year study period. The median time span from the first contact to the second contact with the same A&E department and/or an IFM due to violent victimization was more than 18 months and varied with age group. There were some similarities between recurrent incidents of violence. The lifetime risk of A&E department and/or IFM contact due to violent victimization was 44 % for males and 18 % for females. Merging of data from the health care system with demographic and socioeconomic data revealed that factors such as: not living with a partner, low annual household gross income, high lifetime unemployment rate, high level of household crowding, and a large age difference between victim and partner were strongly associated with adult violent victimization.