Custodial deaths seen in the Office of the Chief Medical Examiner,Lagos,Nigeria: An 11-year autopsy study

Abstract

Introduction

Deaths in custody are a matter of global concern. However, such information is often missing in developing countries. This study aimed to examine retrospectively the profile, cause and manner of deaths amongst cases of custodial deaths in Lagos State, Nigeria.

Method

An 11-year study (June 2008–June 2019) was done of all autopsy cases of custodial deaths in the Office of the Chief Medical Examiner. Variables including age, sex, offence, place of death, duration in custody prior to death and cause and manner of death were extracted from the records. Results were analysed using frequencies and percentages.

Results

Out of 9894 autopsies over the study period, 45 custodial deaths were identified. Males and females constituted 84.4% and 15.6%, respectively (M:F = 5.4:1). Ages ranged from 20 to 64 years, with a mean age of 37 ± 11.0 years. These deaths were most common in the third decade. Armed robbery and financial crime were the two leading reasons for arrest, while most deaths occurred within 24 hours of arrival in custody. The two leading causes of death were acute cardiac failure from hypertensive heart disease and cranio-cerebral injury from blunt-force trauma.

Conclusion

Deaths in custody need to be properly investigated and particular attention needs to be paid to unlawful deaths if and when they arise.

Keywords

Deaths incarceration post-mortem examination medical examiner

Introduction

Although deaths in custody occur in all societies, they are given greater attention in well-developed democracies. While the number of such deaths is small compared to the overall national mortality figures, they often attract greater attention within the society, and this attention can be a good index of the quality of democracy and the degree of respect for the sanctity of life.¹

The definition of deaths in custody is not universal, however. Rather, it varies, depending on the legal system of each country.^2,3 Such deaths can be defined as situations in which the circumstances place the deceased in either direct or indirect contact with law enforcement agent at the time of their death, including incarceration, arrest or pursuit. They are often referred to as in-custody deaths, legal intervention deaths or apprehension deaths.^4,5

Recent interest in this topic by the media and the public at large has led to an attempt to establish the roles that law enforcement agents play in these deaths, as well as attempting to understand how they have been managed by medical examiners or forensic pathologists. For these reasons, the National Association of Medical Examiner in the USA proposed a definition for deaths in custody.⁶ It identifies three broad categories: pre-custody, in-custody and judicial executions. Pre-custody deaths include those occurring prior to physical restraints associated with arrest, while in-custody deaths include those occurring in police custody and correctional centres, including pretrial and sentenced custody.⁶ Although judicial execution does occur in Nigeria, the investigation of these deaths by the medical examiner is not required by law. For this reason, these deaths were excluded from this study.

Custodial deaths, whether pre-custodial or in-custody, are a complex issue. They may be followed by media attention and legal action from the family of the deceased or the involvement of human right organisations. Depending on the circumstances, this can generate considerable public scrutiny and controversy.

For these reasons, the authors decided to look at custodial deaths in Lagos State, Nigeria. Although similar reports exist within the Western world and Asia, the only report from Nigeria was a one-year review (report of two cases) done in Calabar, Southern Nigeria.⁷ Further search of the literature revealed few cases from South Africa⁸ and only one from Mauritius.⁹ It is therefore apparent that this information is largely missing from the literature from developing societies. Given this situation, the authors aimed to examine retrospectively the epidemiological profile and cause and manner of custodial deaths in Lagos, Nigeria. It is noteworthy that within Nigeria, Lagos State is the only jurisdiction that has an updated Coroner’s Law that established the Office of the Chief Medical Examiner and specifically mentions custodial deaths as a reportable death.¹⁰

Methods

We undertook an 11-year retrospective study of all autopsy cases of custodial deaths that were conducted by the Office of the Chief Medical Examiner based within the Department of Pathology and Forensic Medicine, of the Lagos State University Teaching Hospital, Ikeja, Lagos, Nigeria, between June 2008 and June 2019. The department has an experienced and British-trained forensic pathologist who conducts both hospital and coroner’s autopsies for the state as well as for the teaching hospital. By law, all reportable deaths within the state are referred to the Office of the Chief Medical Examiner for post-mortem examination.

A standard neck dissection is part of the protocol for all custodial deaths investigated in the department. This is preceded by radiological examination and general external examination for bruises and other marks of violence or any evidence of medical intervention. These cases also undergo soft-tissue examination for deep bruises. Toxicological samples were retained in few cases because of the absence of the required equipment within the Office of the Chief Medical Examiner. These tests are done in exceptional situations and at great cost to the family of the deceased.

For the purpose of clarity, the authors adopted the National Association of Medical Examiners’ (NAME’s)⁶ definition in identifying the custodial deaths. Consequently, the variables extracted from the autopsy reports included age, sex, offence necessitating the arrest, place of death, duration of incarceration prior to heath, cause of death and the manner of death. This information was recorded in a predesigned form and analysed with IBM SPSS Statistics for Windows v20.0 (IBM Corp., Armonk, NY).

Research materials

The Chief Medical Examiner approved the use of archival materials in the office. These included post-mortem reports, police extracts and clinical notes where applicable.

Study population

Lagos State has a population of approximately 20 million people of diverse background. There are three correction facilities for males and one for females, with the combined number of inmates totalling 9033, although these centres were originally built to accommodate 4087 inmates. Therefore, they are considerably overcrowded. In addition, there is a juvenile correction centre for both males and females.¹¹

Inclusion criteria

All in-custody deaths⁶ in police formations, corrections homes, National Drug Law Enforcement Agency (NDLEA) and National Agency for the Prohibition of Trafficking in Person (NAPTIP) were included. Victims dying in hospital following transfer from the custodial sites were also included.

Exclusion criteria

Exclusion criteria included all deaths from orphanages, mental health institutions and homes accommodating disabled youths.

Results

Out of 9894 autopsies performed during the study period, 45 deaths were identified, representing 0.45% of all investigated cases. Of these, 84.4% were males, while 15.6% were females, giving a M:F ratio of 5.4:1. The ages ranged from 20 to 64 years, with a mean age of 37 ± 11.0 years. Deaths were most common in the third decade, that is, the 21- to 30-year age group, which accounted for 35.6% of cases. This was closely followed by the fourth decade, that is, the 31- to 40-year age group, which accounted for 28.9% of cases. Only two deaths occurred in people >60 years of age.

Figure 1 shows the yearly distribution of cases. Five cases were seen in the second half of 2008, while six cases were seen in 2010, and six were reported during the first half of 2019. Therefore, assuming that there was no under-reporting, 2008 and 2019 recorded more cases than the other years. Only two cases were recorded per year in 2014, 2017 and 2018.

Figure 1.

Number of custodial deaths per year.

Table 2 shows the range of offences warranting arrest. A significant number of the victims were alleged to have committed armed robbery (28.9%), while the same proportion had unspecified offences (28.9%). Financial crime ranked third, representing 15.5%. An unusual scenario was the death of the wife of an alleged thief. The police had arrested her in place of her husband who had committed an offence but could not be located. She was 48 years old and hypertensive, and she died of intracerebral haemorrhage with accompanying intraventricular haemorrhage. There was also the case of a 52-year-old woman who was diabetic and hypertensive and who suffered from deep vein thrombosis (DVT). She had yet to meet her bail conditions while in police custody when she embolised fragments of a deep vein thrombus and succumbed to a massive pulmonary embolism.

Table 1.

Distribution of sex and age.

Parameter	n	%
Sex
Male	38	84.4
Female	7	15.6
Age group (year)
21–30	16	35.6
31–40	13	28.9
41–50	10	22.2
51–60	4	8.9
61–70	2	4.4

Table 2.

Distribution of the alleged offence.

Alleged offence	Frequency	%
Armed robbery	13	28.9
Arson	1	2.2
Child trafficking	1	2.2
Cultism	1	2.2
Drug-related offence	1	2.2
Impersonation	1	2.2
Kidnapping	2	4.4
Police raid	2	4.4
Financial crime	7	15.5
Spouse of alleged thief	1	2.2
Forced eviction of tenant	1	2.2
Pipeline vandalisation	1	2.2
Not provided	13	28.9

Toxicological tests were done in only one of the asphyxial deaths, but the tests revealed what was concluded to be endogenous alcohol.

Table 3 shows that deaths occurring within 24 hours of arrest accounted for 33.4% of cases, while the interval was unspecified in 44.4% of the cases. It is noteworthy that despite the statutory provisions and several complaints, the custodial authorities rarely invite the pathologists to the death scene before transferring the bodies to the morgue.

Table 3.

Time spent in incarceration prior to death.

Duration in the custody	Frequency	%
Unspecified	20	44.4
<24 hours	15	33.3
1–7 days	4	8.9
1–4 weeks	2	4.4
≥1 month	4	8.9
Total	45	100.0

The two leading causes of death were acute cardiac failure from hypertensive heart disease (n = 17) and cranio-cerebral injury from blunt-force trauma (n = 7). In seven cases, the cause of death could not be determined because these bodies had decomposed and had no skeletal injuries (Figure 2).

Figure 2.

Distribution of the various causes of death.

Most deaths occurred in police custody (n = 36) representing 80% of cases. Other places such as hospital and the correctional centre each represented 6.7%. The most common manner of death was natural (57.8%). This was followed by homicide, accounting for 20%, while the manner of death was undetermined in 22.2% of the victims (Figure 3).

Figure 3.

Distribution of the manner of deaths.

Discussion

When people are deprived of their freedom, the responsibility for their well-being rests solely on the custodian, who must guarantee their safety. If death occurs, an independent investigation will not only determine the cause and circumstances leading to the death but will also provide succour and relief to family members of the deceased as to what might have transpired while the deceased was in confinement.¹²

This is the first study that examines the epidemiological features, duration in custody, cause and manner of custodial deaths in Lagos, Nigeria. A six-year study of custodial deaths in Karachi, Pakistan, revealed that they were predominantly male, constituting 95.1%, while females accounted for 4.9%.¹³ The present study also reveals a male preponderance (M:F = 5.4:1), which is similarly reflected worldwide.^2,8,12–20 This is obviously due to the internationally recognised tendency for men to be imprisoned for criminal acts compared to females. It is therefore expected that they will be over-represented in custodial environments and consequently among cases of custodial deaths.

In this study, the mean age was 37 ± 11.0 years, with age ranging from 20 to 64 years. These findings are also consistent with some previous studies.^12,13 Custodial deaths usually occur in males in the third and fourth decades of life. This has also been documented in previous reports.^12,14,21,22 Similarly, it was noted in the present study that custodial death predominated in the 21- to 30-year age group (35.6%); this was closely followed by the 31- to 40-year age group, representing 28.9% (Table 1). These are normally the most productive age groups in the average national population. Spousal abuse and aggression in families, warranting arrest, could result from alcohol abuse. This assertion has been corroborated by a previous study in South Africa, showing a strong correlation between alcohol abuse and violence.²³

An average of 4.4 cases of custodial death are seen each year in Lagos, as demonstrated by this study. The years 2008 and 2019 had the highest incidence, with five and six cases, respectively, over just a six-month period. However, the authors think that there is some degree of under-estimation, considering the strong possibility of under-reporting by the police and other custodial authorities in our jurisdiction. The Office of the Chief Medical Examiner can only report cases presented to it by the custodial authorities. Bodies are often received from the police authorities with claims that the death resulted in the course of gunshot exchanges between the police and suspected armed robbers. Mirza et al. reported 10.16 cases per year in Karachi.¹² Their figure appears to be higher than ours, but this might be due to under-reporting in our jurisdiction. As stated earlier, there is this general impression that custodial deaths are still under-reported in Nigeria and perhaps worldwide, especially in under-developed so-called democracies.

It is believed that people are rarely arrested or incarcerated without an alleged offence. According to the statistics provided by the Lagos State Criminal Information System in March 2019,¹¹ stealing, sexual-related offences and armed robbery are the leading offences why people are incarcerated or arrested in Lagos State, Nigeria. It is therefore not surprising that armed robbery was the most common alleged offence in our study (Table 2). This was followed by financial crime, while approximately one third of cases had no offence stated by the custodial authority. The authors wonder whether it was a deliberate attempt by the custodial authority not to have stated the alleged offence. However, in our experience, it is not uncommon for the custodial authority to attempt to conceal the alleged offence of a deceased, especially in instances where death occurred following arrest for a minor offence. Our study is consistent with an Australian report on deaths in custody spanning a period of 21 years. In that study, ‘violent offences’, which include homicide, assault and robbery, were the most common offences.²⁴

The guidelines for the investigation of deaths in custody prepared by the NAME⁶ require that the time of death following an incarceration be ascertained. In our series, most deaths occurred within 24 hours (33.3%), while fewer deaths occurred within one week (8.9%). Interestingly, the time interval was not specified in 44.4% of cases (Table 3).

An important part of an independent investigation of custodial deaths is the post-mortem examination. NAME has recommended that this should be performed by, or be directly supervised by, an experienced forensic pathologist, and strictly following its forensic autopsy performance standard.^6,25 Acute cardiac failure from hypertensive heart disease was the most common cause of death (37.7%). Cranio-cerebral injuries from blunt-force trauma accounted for 15.5%. A particular case had intracerebral haemorrhage with ventricular extension secondary to hypertensive heart disease. What preceded this catastrophic event was not made clear to the Office of the Chief Medical Examiner. Comprehensive toxicological examination, including testing for common illicit and therapeutic drugs, is essential in the investigation of these deaths, as recommended by NAME.⁶ We conducted toxicological examination in one of the cases of asphyxial deaths. The test identified only post-mortem endogenous ethanol production and subsequently could not be concluded. In most cases, family members of the deceased were expected to pay for the toxicology analysis, and because it is still expensive, these suggestions were usually declined. The inability to perform this test forms part of the limitations of this study.

Overall, natural deaths accounted for 57.8% in our work. Studies in Karachi¹² and India^{12,13,15,26–28} corroborated this finding. Poor or lack of accessible health care, overcrowding and issues with hygiene could be attributable to this finding. In this regard, tuberculosis readily comes to mind, as it was seen in this study and in many other studies. It is worthy of note that the study done by Okoye et al.¹⁶ also observed natural causes as the most common cause of death. Acute cardiac failure was the dominant picture among natural cause of death in this study. This is in contrast to what is seen in Western nations, where ischaemic heart disease was notably observed.²¹

It is apparent that there are variations in the manner of deaths across different regions. In the USA, Okoye et al.¹⁶ observed natural causes as the most common cause of death, whereas Wobeser et al.²¹ in Ontario, Canada, observed that violent deaths accounted for 59% (comprise of suicides, poisoning and homicides). These were the predominant manner of death in the male Caucasian population.

In this study, homicides accounted for 20% of all deaths. These deaths largely resulted from cranio-cerebral injuries due to blunt-force trauma to the skull. Furthermore, a large proportion of cases were ruled as undetermined (22.2%). These included the asphyxial deaths in which toxicological tests could not be done. Fewer deaths were observed with respect to other custodial authorities when compared to those that occurred within the police stations.

A limitation of this study is clearly apparent from the inability to conduct toxicological studies on all of these cases. This could have assisted in identifying the causes of the asphyxial deaths, since these deaths appear to have accounted for a large proportion of cases of undetermined deaths. Future studies should attempt to perform toxicological investigations on samples retained at autopsy. Second, documentation from the custodial authority was inadequate. This was notably evidenced in this study by the large percentage of cases in which the period of incarceration prior to death was not provided by the custodial authority (Table 3). However, this study remains relevant in that it has brought to the fore the magnitude of custodial deaths occurring, particularly within the police formations in Nigeria.

Conclusion

The preponderance of males in custodial deaths as observed in this study is most likely due to the sex over-representation in custodial care. The preponderance of cases in the third and fourth decades could be attributable to indulgence in alcohol in that age group. Natural death was the most common manner of death, and this can be attributed to the poor facilities and medical care available to inmates. Acute cardiac failure from hypertensive heart disease contributed largely to the natural causes. Deaths occurred more in police formations and was most common within 24 hours, representing the critical early period of interrogation and possible attempts at forced confession. In-custody deaths remain a major problem in Nigeria, and the judicial system must rise to the challenge.

Footnotes

Declaration of conflicting interests

The authors declare that they have no conflicting interests.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Sunday S Soyemi

References

Hannan

Hearnden

Grace

, et al. Deaths in or following custody: an examination of cases. IPCC Research Series Paper: 17. London: Independent Police Complaints Commission, 2010.

Petschel

Gall

JA.

A profile of deaths in custody in Victoria, 1991–96.

J Clin Forensic Med 2000; 7: 82–87.

Shepherd

RT.

Deaths in custody. In: Stark

(ed.) Clinical forensic medicine: a physician’s guide. Totowa: Humana Press, 2005, pp.401–423.

Graham

MA.

Investigation of deaths temporally associated with law enforcement apprehension.

Acad Forensic Pathol 2014; 4: 366–389.

Gill

Girela-López

Manner of death for in-custody fatalities.

Acad Forensic Pathol 2015; 5: 402–413.

Mitchell

Diaz

Goldfogel

, et al. National Association of Medical Examiners position paper: recommendations for definition, investigation, postmortem examination, and reporting of deaths in custody. Acad Forensic Pathol 2017; 7: 604–618.

Nnoli

Ugbem

A year review of suspected deaths in custody/detention in Calabar, Nigeria. Med Res Chron 2016; 3: 162–166.

Bhana

BD.

Custody related deaths in Durban, South Africa, 1998–2000. Am J Forensic Med Pathol 2003; 24: 202–207.

Agnihotri

Gangadin

SK.

Police custody deaths in Mauritius. Torture 2005; 15: 25–30.

10.

Lagos State House of Assembly. Lagos State Coroner’s System Law, www.lagosstatehouseofassembly.gov.org.ng/download/coroner-system-law-vol 2 (2007, accessed 20 September 2019).

11.

Lagos State Criminal Information System. LCIS Infographics, www.lagosministryofjustice.org/inmates-enrolled-on-lagos-criminal-information-system-surge (2019, accessed March 19, 2021).

12.

Rajesh

Pradeep

Natural deaths in custody: a 10-year mortality study. J Indian Acad Forensic Med 2011; 33: 328–331.

13.

Mirza

Memon

Adil

, et al. Audit of custodial deaths in Karachi – an autopsy based-study. J Pak Med Assoc 2012; 62: 752–755.

14.

Mandar

Kailash

Anand

, et al. Custodial deaths: a ten years experience from Central Maharashtra. J Indian Acad Forensic Med 2016; 38: 299–302.

15.

Bardale

Shrigiriwar

Vyawahare

, et al. Death behind bars: a five-year study of custodial deaths. Med Leg Update 2005; 5: 105–108.

16.

Okoye

Kimmerle

Reinhard

An analysis and report of custodial deaths in Nebraska, USA. J Clin Forensic Med 1999; 6: 77–84.

17.

Dogra

Bhardwaj

Sharma

GAS

, et al. Postmortem examination in cases of custodial death in India. J Indian Med Assoc 2008; 106: 51–53.

18.

Huddleston

Kocoshis

TA.

Death in custody due to a colopericardial fistula. Am J Forensic Med Pathol 1997; 18: 194–198.

19.

Copeland

AR.

Death in custody revisited. Am J Forensic Med Pathol 1984; 5: 121–124.

20.

Southall

Grant

Fowler

, et al. Police custody deaths in Maryland, USA: an examination of 45 cases. J Forensic Leg Med 2008; 15: 227–230.

21.

Wobeser

Datema

Bechard

, et al. Causes of death among people in custody in Ontario, 1990–1999. Can Med Assoc J 2002; 167: 1109–1113.

22.

Singh

Aggarwal

, et al. Profile of disease related deaths in custodial cases: an autopsy based experience. J Indian Acad Forensic Med 2015; 37: 392–395.

23.

Akinmolade

Mokwena

MP.

The problem and awareness of liquor abuse in South Africa. Int J Soc Econ 2010; 37: 54–74.

24.

Collins

Mouzos

Deaths in custody: a gender-specific analysis. Trends and Issues in Crime and Criminal Justice no. 238. Canberra: Australian Institute of Criminology, 2002, pp.1–6.

25.

Peterson

Clark

SC.

Forensic autopsy performance standards. Marceline: National Association of Medical Examiners, 2006, p.27.

26.

Rajesh

Manish

Vyawahare

, et al. Death behind bars: a five-year study of custodial deaths. Med Leg Update 2005; 5: 105–109.

27.

Bansal

Murali

Singh

Custodial deaths – an overview of the prevailing healthcare scenario. J Indian Acad Forensic Med 2010; 32: 315–317.

28.

Vaibhav

A retrospective study of prison deaths in Western Maharashtra (2001–2008). Med Leg Update 2010; 10: 112–114.