International Journal of Mental Health & PsychiatryISSN: 2471-4372

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Research Article, Int J Ment Health Psychiatry Vol: 3 Issue: 3

Testosterone and Homicide (An African Perspective)

Azza Kamal Alnourani1, Abdelgadir Hussein Osman2* and Abdelgani Alsheikh2

1Tbuuk Medical school, University of Tabuuk, Khartoum, Sudan, Africa

2FRCPsych, associate professor, department of psychiatry, faculty of medicine, Alribat University

*Corresponding Author : Dr. Abdelgadir Hussein Osman, MBBS, MRCP1
FRCPsych, DPM Medical Ed, CCST UK consultant Neuropsychiatrist, Sudan, Africa
Tel: +447393334607, 00249912312117
E-mail: [email protected]

Received: May 18, 2017 Accepted: June 01, 2017 Published: June 06, 2017

Citation: Osman AH, Alnourani AK, Alsheikh A (2017) Testosterone and Homicide (An African Perspective). Int J Ment Health Psychiatry 3:3. doi: 10.4172/2471-4372.1000144


The enigma of the relationship between testosterone and violence remained of blurred boundaries, yet a challenging conundrum for forensic psychiatry and psychology. This article presents the outcome of a case controlled study comparing testosterone blood level from 50 inmates convicted for murder with 50 cross matched control. A positive association was found between inmates convicted for murder and high testosterone blood level. Moreover, the relationship is even stronger between high testosterone blood level and individuals with recurrent aggressive intensions and who experienced recurrent criminal thoughts that are completely opposite with individuals who are possessed with peaceful attitudes, with P value of 0.002, 0.0001 and 0.000 consecutively.

Keywords: Testosterone; Homicide; Murder; Aggressive intensions; Criminal thoughts


Earlier studies have found higher testosterone levels among persons who had committed violent crimes than among those who had committed nonviolent crimes, mostly from Western literature. However only few publications assert this relationship across different cultures [1-3]. This study attempted to shed light on the relationship between violence and testosterone blood level, in Sudan, as an example of an African country.

Basic physiology of Testosterone

The testosterone release control begins in the brain at the site of the hypothalamus, which releases gonadotropin releasing hormone, which travels directly to the anterior pituitary gland where it stimulates luteinizing hormone (LH) release into the blood stream. Luteinizing hormone then stimulates production and release of testosterone from the testes. Testosterone level in the blood can be increased by a number of factors including emotion, energy balance, and stress. There is overwhelming evidence that most of the effects of testosterone in mediating aggression occur after aromatization phase of the synthesis of testosterone. It is believed that actions of testosterone on the brain under increased status threat, enhanced testosterone excretion, up regulating vasopressin gene expression in the amygdala, induces hyper-coupling of the amygdala and the brainstem [4,5]. Testosterone excretion in the blood in stepwise manner and under rising status threat, facilitates processing shift from the orbito-frontal cortex (OFC) towards the brainstem. This processing shift can ultimately lead to social aggression and get compared to a brain-processing shift observed in the case of fear of a proximate threat [6-8].

Testosterone release has surging phase at late adolescence and early adulthood, with a pulsatile pattern of release during the day and night with a peak release early in the morning. There is remarkable drop after middle age [7,8].

Determinations are recommended to overcome the influences caused by variations in transport of proteins on the total testosterone concentration in animals and humans. Testosterone level tends to increase among the winners of the competition while it drops among the losers. In addition, testosterone can increase in response to sexual stimuli [8,9].

It is widely believed that testosterone not withstanding its physiological role during sexual function in men incriminates in influencing sex violence in some ways which is not clearly understood. Saliva testosterone concentrations of incarcerated rapists, child molesters and non-incarcerated volunteers was compared using radioimmunoassay techniques [9,10]. Testosterone levels were higher in rapists than in control subjects and child molesters.

In addition to the formerly widely used antiandrogens medroxyprogesterone acetate (MPA) and cyproterone acetate, most recently LHRH agonists have been prescribed, especially for treatment of paedophilic patients. A systematic review on the use of LHRH agonists to treat paraphilic patients was reported [11-13]. Testosterone circulates in the blood which gets bound to three proteins: sex hormone binding globulin (60-80%), albumin, and cortisol binding globulin [14]. Only about 1-2% of the total circulating testosterone remains unbound.

One important aspect of the social landscape that affects testosterone production is stress. Research indicated that an elevation in cortisol (a stress hormone) lowers testosterone, and decrease in cortisol level enhances testosterone level.

Regarding testosterone and competition men appear to adjust their pre-competition rise in testosterone to the perceived strength of the opponent and the importance of the event. The greater the challenge, the greater the hormone level gets elevated [15-17].

Studies of male rowers suggested that elevation in testosterone during the event is associated with superior performance, as indicated by the amount of time it takes to row 2000 meters [18-20].

Post-event research indicates that winners have higher testosterone than losers. This effect has been replicated several times, not only in physically vigorous competitions but also, in chess competitions as well. The extent of hormonal rise post match seems to depend on the importance of the win and the participants’ evaluations of their own performances. When male competitors believe that they have won by luck rather than personal effort, there is little elevation in testosterone if any post-win rise which was proved by a study conducted on 2008 investigated voters’ testosterone responses to the outcome of the United States Presidential election. 183 participants provided multiple saliva samples before and after the winner was announced on election night. The results showed that male Barack Obama voters (winners) had stable post-outcome testosterone levels, whereas testosterone levels dropped in male John McCain and Robert Barr voters (losers). There were no significant effects in female voters [21-23].

Objective of the Study

Objective of the study is to examine the association between testosterone levels and severe violence leading to murder among Sudanese population. Moreover, to correlate testosterone hormone level with, aggressive impulses, criminal thoughts and peaceful attitude.


This is a cross-sectional descriptive study, conducted on inmates of a high security prison, who were convicted for murder during the study period. Wadmadani High security prison is the second biggest prison in Sudan, situated in Wadmadani city, the second largest city in the country after the capital (Khartoum). Wadmadani city is by the Blue Nile River, in the heart of Sudan.

A sample of 50 inmates were studied, who, were randomly chosen of 153 inmates convicted for murder at the time of the study. Selected randomly from a list of prisoner’s names; using random cluster sampling (every third name was chosen). And 50 crossed matched control randomly selected from a group of men referred for Khartoum Blood Bank for routine investigation prior to Blood donation. For selection of control we used, random cluster sampling (at day 0 the first donor was selected then every third donor, the same thing was done at day 1 and day 2 at the morning hours). The two groups were named Group A and Group B respectively. Only, consented, inmates and control candidates were included in the study.

A blood sample was taken, via vein-section of each candidate in the study, between 10.00 and 11.00 am, then was analyzed in Central Khartoum Laboratory for total and free plasma testosterone level and results were later calculated on computed across identified risk and demographic factors. Each candidate completed socio-demographic questionnaire, and Likert-type scale giving three choices for propensity for aggressive intentions, recurrent criminal thoughts and peaceful attitudes. Each character has three choices describing frequencies of the occurrence of character or thoughts, as very frequent, sometimes and rarely ever. The study was conducted in five months period from May to September 2012.

Inclusion criteria

The study included male prisoners in Wadmadani prison, aged (25-45) years, serving sentences for murder, during the study period from May to September, 2012, who consented to participate in the study.

Exclusion criteria

Prisoners who refused to participate in the study, prisoners older than 45 years of age, prisoners with physical or psychiatric illness, and prisoners with history of alcohol and drug abuse.

Instruments and study tools

A Socio-demographic questionnaire for all participants interviewed inside the prison by the first author and trained assistants psychologists for completing the questionnaire that contained sociodemographic data, and relevant questions assessing intentions of aggression, criminal thoughts and peaceful attitude.

Blood sample of 3 mls, was collected from all subjects in group A and likewise from the control group, by an expert laboratory technician. Each labeled sample was kept in blank container, centrifuged and plasma serum was separated, frozen in small lab unit inside the prison for group A which was then brought to a laboratory for processing by auto analyzer machine to measure testosterone level. The same technique was applied for group B.

Statistical analysis

Data was analyzed by an expert statistician using computer software program, the statistical package for social sciences (SPSS). Spearman’s correlation, student t-test and Chi-square test were computed to compare groups and test of significance of results and differences. Results were expressed in Tables 1 and 2.

Variables Group A Group B P Values Coefficient Pearson's R
  Number % Number %
Age group   .000






25 – 34 50 100% 35 70%
35 – 44 0 0 15 30%
45 ≥ 0 0 0 0%
Total 50 100% 50 100%
Marital Status  
Single 18 36% 15 30%
Married 31 62% 17 34%
Divorced 01 02% 18 36%
Total 50 100% 50 100%
Education Level  
Illiterate 20 40% 08 16%
Primary 30 60% 26 52%
Secondary/Uni. 00 00% 16 32%
Total 50 100% 50 100%
Family Income  
Good 3 6% 2 4%
Average 21 42% 33 66%
Low 26 52% 15 30%
Total 50 100% 50 100%

Table 1: Shows Socio-demographic correlates of the study group A and the Control group B.

Variables Group A Group B P Values Coefficient Pearson's R
TH Blood level TH ≥ 800 ng TH ≤ 800 ng TH ≥ 800 ng TH ≤ 800 ng
Number % Number % Number % Number %
45 90 5 10 35 70 15 30 .000 .000c
Aggressive Intensions No.*2 % No.*2 % No.*2 % No.*2 % .001 0.00c
Very Often 26 52 4 8 5 10 0 0
Sometimes 14 28 1 2 20 40 10 20
None 5 10 0 0 5 10 10 20
Total 45 90 5 10 30 60 20 40
TH and experiencing Criminal thoughts Number % Number % Number % Number % .001 0.00c
Yes Recurrent 25 50 2 4 5 10 5 10
Only sometimes 19 38 3 6 10 20 10 20
No rarely ever 1 2 0 0 15 30 5 10
Total 45 90 5 10 30 60 20 40
Peaceful attitudes Number % Number % Number % Number % .001 .713c
Yes 5 10 1 2 25 50 17 34
No 32 64 3 6 2 4 1 2
Sometimes 8 16 1 2 3 6 2 4
Total 45 90 5 10 30 60 20 40

Table 2: Testosterone Hormone (TH) Blood Level Versus Aggressive & Peaceful attitude study: Group A Study group B and Control Group.

Ethical Considerations

The study was approved by Sudanese Medical Specialization Board. Written permission was obtained from administration of Wadmadani prison and Khartoum blood bank. Written consents were obtained from all participants. Written information about the study was given beforehand. It was explained to all participants that, participation was optional with guaranteed confidentiality.


The total number of candidates participated in the study was, (100) males. 50 inmates convicted for murder and 50 controls. Participants ages ranged between (25-45) years as shown in Table 1. Most of inmates were illiterate 20 (40%), or ended their education at a primary level 30 (60%). On the other hand Most of control, (Group B), received primary education or above up to university level 42 (84%), (P-value for difference 0.000), as shown in Table 1. The later commensurate with National Statistics for the country. More inmates came from socially deprived background 26(52%), with an income below National average ($1800-2000 per annum), while only 15 (30%) of the control sample were from low family income. Majority of our sample of inmate were married 31 (62%), compared to control group as only 17 (34%) (p-value is 0.000) as shown in Table 1.

Regarding testosterone blood level (T.H), 45(90%) of inmates convicted for murder in this sample showed level equal or over 800 nanogram, while only 35(70%) of the control group did so (p-value is0.000), as shown in Table 2.

Importantly, however 30 (60%) of inmates reported to have very often experiencing aggressive thoughts and intensions and 27 (54%) recurrently experienced criminal thoughts in the last two weeks prior to the study. On the other hand only 6 (12%) of the control group did so for aggressive intensions and 10(20%) experienced peaceful attitude 5(10%), when 25 (50%) of the control group described themselves as peaceful, as shown in Table 2.


In this research several findings deserves special note: We only screened candidates between 25 and 45 years of age, to avoid the high surge periods of testosterone in the late adolescence and early adulthood, and periods of physiological reduction of body release of the hormone. Despite the range of age groups of the inmates in the high security prison with murder conviction, skewed towards the younger category of mid-twenties to early thirties 25-35, 100% of the inmates and only 70% in the control group.

Married, individuals constituted 31(36%) of inmates convicted for murder in this sample, compared 17 (34%), of the control group which is more representative to National characteristics of this group characteristic in the country and statistics from similar studies in the West [24,25]. This discrepancy may be explained by the nature of the crime and its circumstance, as, one third of culprits committed murder following tribal and family feud in Sudan. On the other hand studies from Western European culture, showed murder crimes often related to drug and for financial gains [22,23].

In line with many studies from Western background, most of our inmate group received little education, 40% were illiterate, and the other 60% did not exceed primary level of education. On the other hand 32% of the control group studied up to University level. There was no statistical difference between the two groups in relation to family income value 0.56. This is in contravention to studies from Western culture, reason of which was hypothesized above.

To test the hypothesis that men with high testosterone have a tendency towards aggressive impulses, separate computation was carried out on men with high testosterone from both group A and B, versus low testosterone. This revealed that men with high testosterone level from both groups have more tendency to report aggressive impulses than those with low testosterone (p-value, 0.001).

Prior research asserted that, younger males tend to commit more crimes, especially more violent crime, and show far show higher levels of recidivism than older males. Testosterone is also higher in younger males. This fact was evident in the outcome of the findings of this work, as, 45 (90%), inmate showed high testosterone level, compared to 35 (70%) of the control group with p value of 0.01 for the difference. Our findings showed mean blood testosterone concentrations to be higher among inmates convicted of violent crimes than among those without convictions, despite the fact that testosterone level dropped right after incarceration as proved by study done by Thompson, Dabbs and Frady and, the majority of our prisoners who had spent more than 2years in person, 90 % of them showed high testosterone level [23,25,26].

The current evidence suggests that, relationship between testosterone and aggressive behaviour is quite complex and nonlinear. It is clear that an enormous amount of work still remains to completely understand the relationship between testosterone and human behaviour. Traditional research questions have focused on the role of testosterone in male behaviours, but we are becoming increasingly aware that this is not a simple linear relationship; as the social context and individual behaviours might also play a significant role in determining hormone levels [26-28].

Important finding in this study is the high association between (P=0.01), individuals that reported high frequency of aggressive intentions and high testosterone level. This positive relationship was detected even with the control group. Likewise, both inmates and control group showed their character as peaceful and with less frequent aggressive intentions showed tendency towards low testosterone blood level. Our present finding showed that testosterone levels were relatively high in both, the violent and in the nonviolent men which implies, that a high level of testosterone is associated with aggressive intentions and character that entertain the aggressive thoughts or possibly its environmental difference to our sample. However, the former, elucidate personality traits that is relevant to high testosterone excretion in the blood.


We wish to express several cautions about this work on testosterone and violence. The first is that our testosterone scores were based upon one serum sample from each subject. Testosterone has a pulsatile and cyclical quality. It varies randomly from moment to moment and day to day and it varies in a number of ultradian, circadian, and infradian cycles.

Research is yet to be conducted to establish the relative magnitudes of intra- and inter subject variation, as, little is known of, whether it is better to think in terms of mean scores or cyclical quality.

The second caution Little is known of the complexity of casual relationships between testosterone and violent behavior; as violent behavior may lead to higher testosterone; or the other way round or even more vicious circular interaction [27-29].

Our third caution is the concern about the fact that these data were collected in a place where violence is the norm. Under more benign conditions, testosterone might enter into quite different relationships, which could reflect a limitation on this study as the two respondent groups were subjected to different levels of stress [30,31].

Research into human forms of behavior such as aggression is complicated by the complex manner in which humans tend to behave, so much, aggressive encounters are influenced by learning, i.e. the outcome of the first encounter influenced subsequent fight. Therefore, induction of testosterone secretion by external stimuli of aggression should be seen to be dependent on additional factors, such as culture and education. Social background can influence these reactions, as shown in students from the Northern and Southern parts of the USA, the latter described themselves as having grown up in a `culture of honor’ in which insults diminish a man’s reputation. When the volunteers were subjected to verbal insult, the Southerners reacted with significantly more aggressive and dominant behavior, which was correlated with their rising serum testosterone [24,32,33].


This study confirms the strong positive correlation between serum testosterone levels and criminal behavior. It also correlates positively with violent drives and aggressive intentions. Individuals that identified themselves as peaceful tend to have low testosterone level.


Track Your Manuscript