Introduction

The aim of this study is to make a simple descrip tion, based on a 310 necropsy studies taking into account homicides that occurred in the Department of Atlántico and Magdalena, Colombia which is a country that has been affected by the situation created by an internal war, as well as, by murders committed by hired killers. This article presents a summary with descriptive values that will help to get a better view of how these murders occur in these regions.

In Colombia, murders are classified as: simple, aggravated, felony, for mercy, induction or assisted sui cide, manslaughter, abusive, artificial insemination or fertilized ovulum transferred without consent. In this case, within all of these categories, the hired murderer commits an aggravated homicide because the homicide committed "by price, profitable promise or due to abject or futile fact" (Act 40 of 1993, 1993) is found aggravated under these circumstances.

Different scientists have stated the existence of di verse factors that may affect the incidence of this crime in a given situation. Among these factors, there are cul ture (^{Ousey & Lee, 2010}), social and cultural structure (^{DeFronzo et al., 2007}) resource deprivation (^{Pridemore & Trent, 2010}; ^{McCall & Nieuwbeerta}, 2007), the num ber of inhabitants (^{McCall, Land & Parker, 2010}), the type of zone (Weisheit & Wells, 2005), the level of eco nomic inequality (^{Chamlin & Cochran, 2006}), divorce rates (^{Beaulieu & Messner, 2010}), among others.

Homicides in the world

The WHO (2002), in 2002 estimated that there had been about 520 000 deaths caused by murders, equivalent to a rate of 8.8 homicides per 100 000 people in the world. Homicide is the second leading cause of violent deaths with a total share of 31.3 % after suicide with 49.1 %. It is important to emphasize that low-in come countries are the ones with more deaths caused by violence with a percentage of 91.1 %. From the total number of killings by homicide, 77 % were men, hav ing a rate three times bigger than women. Also, men between 15 and 29 years old had the highest homicide rates (19.4 per 100 000 population) followed by men aged between 30 and 44 (18.7 per 100 000 inhabitants).

Moreover, the highest rates of youth homicide occur in Africa and Latin America, and also, the lowest in West ern Europe, parts of Asia and the Pacific.

Criminal behavior in Colombia

According to ^{Espino-Duque (2010}), in Colom bia in 2009, a total of 454 030 criminal behaviors were recorded. Comparing these figures with those of the previous year, where 460 135 behaviors were registered, there was a downward variation by 1 %. Also, accord ing to this author, there was a decrease of criminal be havior in 8 of the 18 Penal Code titles compared to the last two years; these decreases were "against life and personal integrity" by 17 %; "against persons and ob jects protected under the international humanitarian law" by 9 %; "against economic assets" by 3 %; "against copyright" by 4 %, "against the economic and social or der" by 2 % "against natural resources and the environ ment" by 5 %, "against effective and righteous adminis tration of justice" and 13 % "against the existence and security of the State" going from two murders in 2008 to one in 2009. Meanwhile, the other titles of the Penal Code show increase in various types of crimes.

It is important to highlight that only crimes "against life and personal integrity" had a 22.27 % of the total counting 101 135 cases in 2009. Within this title, the criminal conducts that occurred most often were common personal injury, personal injury inflicted during a trafficaccident and homicide, totaling 97 099 cases, accumulating a percentage of 96 % in the total number of crimes that threaten "against life and per sonal integrity."

On the other hand, we find relevant to add that this sort of criminal behaviors represent life meaning for these people, keeping in mind that we need to address the subjectivity of the killers.

Homicide in Colombia

According to the records of murder in Colombia, for the last 12 years there has been a decline in the num ber of this crime. In 1997 there were 24 306 cases of murder, the highest rate of homicides occurred in 2002 with a total of 27 829 (^{Acero Alvarez, 2010}).

According to official data of violence in Colom bia, in 2010 there were a total of 17 459 homicides, a rate of 38.36 homicides per every 100 000 habitants. The age group with the highest rate of homicide victims is between 25 and 29 years old, with a rate of 90.64 per 100 000 populations. Meanwhile, homicide victims were male in a total of 16 015 cases. Another important fact is the marital status of the victims, in which there was a high prevalence of homicides in the single population, representing 50.16 % (^{Acero Alvarez, 2010}).

In Colombia, the most common method used by criminals is the firearm constituting 77.6 % of the to tal number of homicides registered in 2010, being this method the one used in 13 549 opportunities. It follows the stabbing of 2 444 cases and a 14 %. In 2010, in the departments of Atlántico and Magdalena there were a total of 958 homicides.

In this study, a descriptive analysis of the reports in the Atlantic and Magdalena departments located in the Colombian Caribbean region is presented.

Method

Data was obtained from historical records of the Institute of Legal Medicine and Forensic Sciences. The sample (N = 310) is comprised of homicide victims in 2007, in the Atlantic and Magdalena Departments. The sample was selected according to two criteria: 1) the murder was committed with a firearm, and 2) the mur der was committed by hired killers. The information is taken from the Statistical Information System and Op erational Delinquency (SIEDCO) of the National Police of Colombia, from where information from each victim was obtained and from their corresponding forensic au topsy reports from the National Institute of Legal Medi cine.

The data were processed in SPSS (version 18) and analyzed according to epidemiological factors such as sex, age, marital status, weekday death scene, number of input impacts firearm, among others. At the same time, turn crosses were made for research relevant variables.

Results

From the total number of victims killed by hired murders with a firearm, it was found that male participa tion is the most significant, representing 93.5 % (290) of the total, while only 5.2 % (16) corresponds to the female gender.

The sample selected for the study presented as a result the age range of the homicide victims at the time of their death. It is noted that the 38.7 % of the victims were between 21 and 30 years of age while 31 % were within the range between 31-40 years old, followed by the victims ranging between 41 and 50 years old repre senting 18.5 % of the victims. Moreover, the ages with lower incidence were those younger than 10 years old, people between 51 and 60 years old and the victims be tween 11 and 20 years old.

The most frequent marital status of the victims was cohabiting with a percentage of 32.9 %. The second most frequent was "single" with a percentage of 27.4 %. Only 16.8 % of the victims were married. On the other hand, the marital status data was not obtained for 20.6 % per centage of the victims.

The day of the week with the highest frequency of deaths was Monday with a percentage of 19.2 %. The day with the second highest incidence of death was Fri day with18.4 %. The days with less frequent deaths were Tuesday (11 %) and Sunday (11 %).

The most frequent site of the murder was "in pub lic" with a percentage of 33.2 %. Following, "Close / Facing a business establishment" with 8.1 %. The places with the lowest frequencies were "in a restaurant" with a 1.3 % and "on a farm / Vereda" also with 1.3 %.

The scene where a higher frequency of victims oc curred was Santa Marta with a percentage of 51.6 %. Barranquilla followed with 31 % of victims.

Of the two departments studied, it was found that Magdalena had a higher incidence rate of homicide victims with a percentage of 54.2 %, equivalent to a total of 168 victims. Meanwhile, the Atlantic showed a percentage of 44.2 % with a number of 137 victims.

The most frequent number of input impacts was "2" with 21.3 %, followed by "1" with a percentage of 18.4% and "3" with 17.3 %. Within the less frequent number of impacts are "12" and "14" both with a per centage of 0.3 %.

Most gunshot fire did not leave the body of the vic tim as 90 % showed "0" output impacts. The second most frequently impacts on output was "1" with 3.9 %,corresponding to 12 victims.

The highest frequency of impacts received accord ing to regions of the body was the head with a total of 469 impacts, equivalent to 46 %. The body received 341 impacts equivalent to 34 % and the extremities received a total of 204 impacts representing 20 % of the total. Specifying by area of all the head impacts, the areas that had the highest input frequency impacts were the left occipital area with a percentage of 17.5 %, followed by the right occipital area with 16.6 %. The most frequent body areas were the suprascapular impact left with a per centage of 8.8 % equivalent to 27 impacts and neck on the right side (infrascapular) also with 8.1 % with a total of 25 impacts. Meanwhile, the area of the limbs with higher input frequency of impacts was the left anterior arm don a percentage of 5.5 % equivalents to a total of 17 input impacts.

According to the data obtained, in both the At lantic and the Magdalena departments, the "public area" is the most common place of homicide. On the other hand, in the Magdalena a trend was observer, different from the Atlantic, where killings were carried out "in a hospital", of which Magdalena had 16 and the Atlantic did not have any. Another major difference is in the ten dency to commit murder by hired killers "near their resi dence" since in the Atlantic 8 of these were committed, while in Magdalena 0 homicides occurred in places with this feature. Also in the Magdalena many more murders were committed by hired killers "outside city limits" than in the Atlantic, with a difference of 15-1.

The right occipital area had the highest input fre quency of impacts on Thursday. While the left occipital showed their greatest number of input impacts on Mon days, Thursdays and Saturdays.

most frequent was "single" with a percentage of 27.4 %. Only 16.8 % of the victims were married. On the other hand, the marital status data was not obtained for 20.6 % percentage of the victims.

The body areas with the most input impacts were the left suprascapular area with a total of 8 hits and also the right scapular area with a total of 8 hits. On the left suprascapular area, the most often impacts were received on Wednesday and Friday with 2 hits each. Meanwhile, the left scapular area received the highest number of im pacts on Monday with a total of 3.

The most frequent marital status of the victims was cohabiting with a percentage of 32.9 %. The second

Discussion

An autopsy is a fundamental instrument in ho micide cases with an unknown perpetrator, because it can provide information about the possible psychologi cal print related to the motivation and intentionality of the murderer, from the type of injury, the number of projectiles impacted and the area of the bodyinjured, providing important clues to better understand this type of homicide.

From a medical approach, ^{Hamilton and col leagues (2008}), (^{Padubidri, Menezes, Pant, & Shetty, 2013}), ^{Lucena, et al., (2009}); ^{Hebert, Maleki, Vasovic, Arnold, Steinberg & Prystowsky (2014}); ^{Park, Huh, Piao, Kim, & Hwang, (2013}); ^{Colville-Ebeling, Freeman, Banner, & Lynnerup, (2014}); ^{Sinard (2013}) and ^{Tette, Yawson, & Tettey (2014}); ^{Elder, (2007}); ^{Yayci, Pakis, Karapirli, Celik, Uysal, & Polat, (2011}); ^{Subedi, Yadav, Jha, Paudel, & Regmi, (2013}) used autopsies to verify diagnosis of dementia with Lewy bodies, Parkinson's and Alzheimer's disease, aneurysms and several other diseases or to address educational practices in order to improve autopsy adult and paediatrical education.

^{Newton, Coffin, Clark, & Lowichik, (2004}) and ^{Hull, Nazarian, Wheeler, Black-Schaffer, & Mark, (2007}), ^{Wong, Chan, Beh, Yau, Yip, & Hawton, (2010}); ^{Flach and colleagues, (2014}); ^{Mohammed & Kharo- shah, (2014}); ^{Inokuchi, and colleagues, (2014}); ^{Yeow, Mahmud, & Raj, (2014}); ^{Matsumoto, Sengoku, Saito, Kakuta, Murayama, & Imafuku, (2014}); ^{Ifteni, Correl, Burtea, Kane, & Manu, (2014}); ^{Le Blanc-Louvry, Th- ureau, Lagroy de Croutte, Ledoux, Dacher, & Proust, (2014}) and ^{Kodaka, and colleagues, (2014}) note that au topsies still remain important to medicine to strength en the pathologist-internist collaboration and to make clearer diagnosis or unexpected pathologic findings. In exactly this same direction, ^{Burton & Underwood, (2007}) state that autopsy has been often underused in modern clinical practice but it is an important proce dure with potential to advance medical knowledge and part of its relevance comes from its epidemiological, educational, forensic and clinical value.

There has been other topics related to autopsies like the knowledge, attitudes, perceptions and the way relatives react to that forensic procedure, like those made by ^{Ogata, Nishi, & Maeda, (2009}) and ^{Oluwasola, Fawole, Otegbayo, Ogun, Adebamowo, & Bamigboye, (2009}) that assesed this type of constructs finding that it is difficult to obtain consent from relatives of deceased patients in a Nigerian Tertuary Institution and that only 38 % of relatives had satisfactory knowledge about the procedure.

From a psychological perspective, there are psy chological autopsies that help to understand the possible reasons behind a suicide or suspicious deaths, like the studies conducted by ^{van Spijker, Graafsma; Dullaart & Kerkhof (2009)}; ^{Wong, Chan, Beh, Yau, Yip, & Haw ton, (2010}) and ^{Yucel Beyaztas, Butun, Ozer, & Celik, (2013}), ^{Kizza, Hjelmeland, Kinyanda, & Knizek, (2012}), ^{Paraschakis, Michopoulos, Douzenis, Christodoulou, Koutsaftis, & Lykouras, (2012}), ^{Ahmed, de Jager, Haigh, & Garrard, (2013}) and ^{Henry & Greenfield, (2009}), that analyze personality factors involved on the deaths and suicides retrospectivily in order to make predictions around their deaths based on their personality profile.

This article is only intended to provide data to the scientific investigation of the killings, so its interest is purely academic, it doesn't attempt to specifically inves tigate in itself any information relating to any particular criminal.

In the present study we found a high incidence of male homicide victims in killings (94.2 %). This, shows a concordance with global figures of homicides where men comprise the vast majority (77 %). However, it is clear that global figures do not specify the modalities of homicides, so its incidence in males, despite having a majority in both studies, in the present study significant ly exceeds global figures. Given this lack of information provided by global data shows, there is a need to investi gate the variable "homicide killings" to be more specific as to the methods of murder. Other studies also show a high incidence of homicide victims in males (^{Abrams, Leon, Tardiff, Marzuk & Sutherland, 2007}).

The area of the body with the most number of impacts was the head, so one can deduce a special prefer ence for injuring the victim specifically in the Occipital area where you will find 1 or 2 shells.

Another important result found is the high in cidence of homicide victims that is in the age range between 21 and 30 years old (38.7 %) and in the age range of 31 and 40 years old (31.3 %). Other studies have found similar rates in these ranges. Amongstthese studies is the one carried out by ^{Goren, Subasi, Tirasci & Kemaloglu (2003}), where the majority of homicide victims belonged to the age group between 16-25 years of age (38.7 %) as well as the study of ^{Kumar, Li, Zanial, Lee & Salleh (2005}) with 63 % of victims between 20-39 years old. Considering the WHO data, the results found in this study have a strong correlation with the global in cidence where the two ranges with the highest incidence of homicide victims are, first, between 15 and 29 years old and in Second, between 30 and 44 years old of age.

In Colombia, social exclusion is a phenomenon that affects poor people in a deep manner. It limits op portunities for young people and it is harder for them to get educated and find jobs. If they do actually get a job, salaries in Colombia are very low and employees with low salaries cannot even pay for the basic goods they need, so many of these youths find illegal gangs as a way out and this makes it more plausible for them to kill or get killed.

In terms of the most common marital status of the homicide victims, a result was found that may be highly relevant. The lowest incidence of homicide vic tims of killings is occupied by those whose marital sta tus was married. This result is of great importance since there are lines of research regarding homicide that have linked it with divorce rates (^{Beaulieu & Messner, 2010}), so it would be possible to further clarify and add to the literature that has found a relationship between these variables, murder and marital status.

At last, we propose that it is necessary to create educational programs that involve families so that there can be a change in the way families conceive violence, thus educate them to adopt non-violent behaviors that could promote, explicitly, non-violent practices as a healthy alternative, understanding that through experi ence people can learn to understand the culture of peace (^{Polo, Pineda y Romero, 2008}).