Traumatic and other non-natural childhood deaths in Manitoba, Canada: a retrospective autopsy analysis (1989-2010).
Herath, Jayantha C. ; Kalikias, Saman ; Phillips, Susan M. 等
Manner of death is categorized as natural, accident, suicide,
homicide and undetermined. Death may be due to disease (i.e., natural)
or external factors. Statistics Canada reported that during the period
2004-08, the leading causes of death in children were accidental
injuries (~30%), malignant neoplasms (~16%), congenital anomalies (~8%),
suicide (~4%), heart disease (~4%) and assault (~3%). (1) Injuries are
the leading cause of childhood death in developed nations, accounting
for almost 40% of deaths between 1 and 14 years. (2) In Canada in 2004,
there were >13,000 deaths and 211,000 hospitalizations due to injury
with a total cost of CAD $19.8 billion; children aged 15-19 years
accounted for a substantial proportion of these. (3,4) In Manitoba,
1994-97 administrative data indicate that injuries accounted for >50%
of all deaths at 1-9 years of age and 70% of deaths at 10-19 years of
age. (5)
Most mortality statistics are derived from administrative
databases, which are subject to coding errors. Retrospective autopsy
review potentially offers more detailed and accurate information.
Statistics Canada data indicate that the percentage of deaths subjected
to autopsy in 1991 was 16.9% and had declined to a national average of
5.7% in 2007. In Manitoba, the autopsy rate was considerably higher, at
10.5%, although comparisons for the pediatric population are not
available. (6) Manitoba has highly centralized health care. The death
investigation system is coordinated by the Office of the Chief Medical
Examiner (OCME). In Manitoba, all child deaths (nonnatural and natural
deaths of individuals <19 years of age) are reportable to the OCME.
Manitoba has Canada's oldest multidisciplinary child death review
committee, formed in 1992 through the provincial Advisory Committee on
Child Abuse.
Because of the centralized death investigation system, it is
possible to analyze all childhood deaths retrospectively. During the
study period, almost all autopsies were performed in two
university-affiliated hospitals in Winnipeg and a small proportion in a
third hospital in Brandon. More than 95% of the pediatric autopsies were
done at a single centre in Winnipeg. Most were performed by pathologists
with broad forensic experience, often in consultation with a
neuropathologist. The goals of analyzing all non-natural childhood
deaths in Manitoba for the 22-year period are to document temporal
trends that might be influenced by changes in society and to highlight
preventable causes of death.
METHODS
This is a retrospective study of deaths among children aged 0-18
years inclusive residing in the Province of Manitoba from 1989 to 2010.
The study was approved by the Research Ethics Board of the University of
Manitoba (protocol H2009:338). Databases used included the pediatric
autopsy records and files of the OCME. Most autopsy records were located
at the Health Sciences Centre, where >95% of child medicolegal
autopsies have been conducted.
The OCME records were used to enumerate the manner of childhood
death, which is a legal designation of natural, accident, suicide,
homicide and undetermined. The cause of death is a biological
consideration that comprises a multitude of diseases as well as external
causes (including accidental trauma, poisoning, intentional self-harm
and assaults). Information included year of death, age at time of
injury/death, sex, manner of death (homicide, suicide, accidental,
undetermined), cause of death, body regions affected by trauma (head,
neck, chest, abdomen, pelvis, upper limbs, lower limbs) and survival
time between the event and death. The age grouping was 0-1 years, 2-4
years, 5-9 years, 10-14 years and 15- [less than or equal to] 18 years
(rounded to the nearest integer). Causes were categorized as trauma due
to motor vehicle/traffic/pedestrian accidents, blunt or sharp force
assault or falling; and poisoning, choking, hanging, drowning,
environmental exposure, burns or smoke inhalation sustained in (mainly
house) fires, animal attack, gunshot and electrocution. Deaths due to
natural causes or the complications of medical management were excluded.
The location of the event that led to death (not the location of death)
was divided into four geographic regions: 1) Winnipeg Metropolitan
Region, 2) smaller urban centres (>5,000 population) in southern and
central Manitoba (Brandon, Steinbach, portage la prairie, Winkler,
Selkirk, Dauphin, Morden), 3) rural south and central Manitoba and 4)
northern Manitoba corresponding to health authority regions (Norman,
Burntwood, Churchill) located north of the major lakes. During the study
period, the population of Manitoba gradually increased from 1.104
million to 1.236 million, and the percentage of children 0-19 years
gradually declined from 29.4% to 26.0%. Using population data from 1995
and 2010, the regional population and the percentage of children within
each region were determined. There was no significant change in the
regional distributions between the two census periods. In 2010,
approximately 55% of the population were located within the Winnipeg
Metropolitan Region, 29% in rural regions and the remainder in scattered
small cities and towns (>5,000 population).
Data acquisition was done by manual review of autopsy reports.
Mortality, population and demographic data were extracted from the
Canadian Socioeconomic Information Management (CANSIM) database
available online from Statistics Canada. Data from the autopsy reports,
OCME reports and CANSIM were entered into a statistical database (JMP
10.0.1; SAS Institute Inc., Cary, NC) for comparisons. Note that CANSIM
data for children are reported in 5-year aggregates to the end of the
19th year. OCME data are summarized in decades (to the end of the 19th
year), as well as in the aggregate period to the end of the 17th year;
the latter are used unless otherwise stated. Data are reported as mean
[+ or -] standard error of mean.
RESULTS
Demographic factors
During the 22-year period analyzed, 990 autopsies were performed
for non-natural childhood deaths ([less than or equal to] 18 years of
age). This represents approximately 22% of the total childhood deaths.
During the entire period, there were 234 autopsies in the 0-1 year
group, 204 in the 2-4 years group, 127 in the 5-9 years group, 164 in
the 10-14 years group and 261 in the 15-18 years group. There were 581
males and 409 females; over the years, the annual proportion of males
ranged from 42% to 69%.
Analysis by geographic region showed no obvious temporal changes.
During the 22-year study period, 13.4 [+ or -] 1.0 non-natural deaths
subjected to autopsy per year (30% of total) were from the Winnipeg
region (51% of children in Manitoba), 4.0 [+ or -] 0.5 deaths per year
(9% of total) were in smaller urban centres (14% of children in
Manitoba), 14.2 [+ or -] 1.3 deaths per year (32% of total) were in
south and central rural regions (26% of children in Manitoba), and 13.9
[+ or -] 0.8 deaths per year (29% of total) were in northern rural
regions (9% of children in Manitoba).
Manner and causes of childhood death
Within the reported administrative datasets, approximately 16% of
the total childhood deaths were accident, 6% were suicide, and 3% were
homicide. The overall distribution for manner of death in the autopsies
(excluding natural causes) was 50.9% accident, 20.3% undetermined, 17.1%
suicide and 11.7% homicide. Accurate proportions in relation to all
childhood deaths could not be determined because not all natural deaths
are reported to the OCME (and a minority undergo autopsy), and the
administrative databases are aggregated by decade rather than a limit of
[less than or equal to] 18 years. The accidental death rate increased
gradually, peaking in 1999, and has generally declined since then. The
suicidal death rate was quite variable but has generally increased,
peaking in 2005 and showing no general evidence of decline. The homicide
rate was generally low with no obvious pattern (Table 1). The manner of
most deaths in the 15-18 year age group was accident (especially injury)
or suicide.
Road traffic incidents represented the major cause of all
accidental deaths (47.8%). The majority of those who died in traffic
incidents were vehicle passengers (n=94) and pedestrians (n=76), and the
minority were vehicle drivers (n=42) and bicycle riders (n=32) (Table
2). There were no specific temporal trends. A mix of head, torso and
limb injuries occurred with no single region of the body accounting for
the majority. Other causes of accidental deaths included drowning
(n=106), burn/smoke inhalation (n=66), poisoning (n=22), fall from a
height (n=11), choking on food or foreign object (n=9), gunshot (n=7)
and animal attack (n=1).
Suicide was apparently rare in the initial years of this survey
(Table l). Suicides, especially in the 15-18 years group, then began to
increase steadily, peaking at 20 in 2005 (mean 17.1 [+ or -] 2.2% of
non-natural deaths per year). Almost all were due to hanging (n=174)
(Table 2), with rare self-inflicted gunshot injuries (n=7) and
poisoning/drug overdose (n=4). Prior to 2006 males accounted for the
majority of suicides, but by 2006 there was approximate gender parity.
The frequency of homicides was highly variable (mean 11.7 [+ or -]
1.3% of non-natural deaths per year, range 1-18) (Table l).
The largest single category of homicide was assault on infants and
children <3 years old (n= 52, 0-6 per year) (Table 2). In most of
these, the cause of death had been designated as "shaken baby
syndrome" or "non-accidental head injury" in early years
and "head injury" or "blunt force injury" in more
recent years. The shift over time reflected the need to use terminology
that was objective and not prejudicial. Excluding this category,
homicidal deaths were categorized according to type of force used,
including blunt force trauma (n=2l; e.g., beating), sharp force (n=18;
e.g., stabbing), mixed sharp and blunt trauma (n=8) and gunshot wound
(n=16).
The majority (75.8%) of deaths classified as undetermined were
infant (0-l years) sleep-associated deaths at home. In the early years
of this study, many had been designated as sudden infant death syndrome
(SIDS). The OCME database showed the frequency of SIDS cases to be 12-34
per year (mean 22.5) from 1980 to 1990, 11-15 per year (mean 12.4) from
1991 to 1997 and 1-8 per year (mean 4.3) from 1998 to 2010 (Table 2).
Note, however, that the frequency of postneonatal death had declined
less slowly and seems to have stabilized at ~30 deaths per year.
Increasingly utilized are the terms "sudden unexplained infant
death" or "undetermined cause of death", often with a
comment about unsafe sleeping circumstances (e.g., face down position in
bed, sleeping in bed with parents, parental alcohol ingestion). In rare
cases of infant deaths (<5), the cause was designated as asphyxiation
(e.g., by smothering) and the manner as accidental.
Survival period following fatal event
Most deaths occurred relatively soon after the event. Death was
most often pronounced at the scene of the incident (847 of 990) or
within hours thereafter (63 of 990), usually in a hospital emergency
department following transport and attempted resuscitation. A small
number of deaths occurred 1 to 7 days (n=57), 7 to 30 days (n=14) or
longer (n=9) after the event. Deaths that occurred very long periods
after a childhood traumatic event may not have been captured in this
analysis if they were classified as natural (e.g., pneumonia after
survival for years in a vegetative state following head injury) or
because the individuals had become adults. Anecdotal information
suggests that 1-2 autopsies per year fell into this category; more might
have been missed because autopsies were not done.
DISCUSSION
We sought to analyze non-natural childhood deaths in Manitoba for
the period of 1989 to 2010 using autopsy reports and the OCME database
to document temporal trends and to highlight potentially preventable
causes of death. The major causes of non-natural death among Manitoba
children were accidental road traffic incidents, suicide by hanging,
accidental drowning, accidents related to house fires, homicides through
child abuse and sleep environment-related deaths in infants. The
proportion of rural inhabitants in Manitoba remained fairly stable at
~30% from 1989 to 2010. As has been previously documented using
administrative data from 19842007, (5,7) non-natural childhood deaths
occurred with disproportionate frequency in rural settings. This might
reflect the socio-economic features of rural Manitoba. (8,9) It is known
that impoverished rural communities, many of which are populated by
First Nations, suffer from a higher frequency of drowning, (10) suicide,
(11-13) infant mortality (14) and severe trauma. (15-17)
Injury (including road traffic injuries, falls and assault) is the
leading cause of deaths and hospitalizations among Canadian adolescents.
(18,19) Overall, the rate of accidental child death in Manitoba peaked
in 1999, and we observed an encouraging trend with declining traffic
fatalities. However, 2010 was marred by a substantial increase,
suggesting that education about seatbelt use (many passenger deaths were
caused by ejection from the vehicle) and driving while intoxicated
remains only partially effective. The other major contributors to
accidental deaths were fire-related burns or smoke inhalation and
drowning. Both occurred mainly in rural communities, where at-risk
housing is common and there are many unmonitored bodies of water. (20)
Prevention remains the mainstay of reducing death and disability from
accidental injuries. (21)
Across all ages, including adults, the suicide rate in Manitoba was
similar to that in Canada overall for the years 2000-2008 (12.1/100,000
population). (22) Across Canada during the same period, there was an
overall decline in childhood and adolescent suicide, mainly due to the
decreasing frequency of firearm use by males. Hanging
("suffocation") is now the leading means of suicide among
Canadian adolescents, accounting for ~70%. (23) The observed increase in
the number of Manitoba childhood suicides by hanging during the late
1990s, with no decline, is disturbing. The fact that alternative methods
of suicide (e.g., gunshot and poisoning) did not increase suggests that
factors other than adolescent depression are involved. In part this can
be attributed to the relative ease with which an intoxicated child can
obtain a piece of rope, shoelace or belt. One must also consider the
role that mimicry plays in the social clustering of adolescent suicide.
(24-27) Designation of suicide as the manner of death may be incomplete,
because intent cannot always be inferred from the circumstance of death.
For example, some drug-related deaths were recorded as accidental or
undetermined. The factors that contribute to car crashes are never fully
known. With respect to hanging, there was evidence for intoxication in
many cases. The very rare cases of death associated with apparent
autoerotic asphyxiation were categorized as accidental.
In contrast to suicide, across all ages the mortality rate for
homicide in Manitoba was more than two times that in Canada overall (3.9
vs. 1.6/100,000 population/year for 2000-2008). (22) Child abuse-related
homicides in Manitoba were unfortunately relatively common (1.7
cases/million/year) in comparison to those reported in a prior detailed
study from Ontario (0.3 cases/million/year). (28) Violence against
children and youth, as reported to the police, is highest in
Saskatchewan and Manitoba. (29) The demographic factors discussed above
may be contributory.
A large proportion of deaths classified as undetermined were
sleep-related infant deaths. Judging by the gradual decline in
postneonatal deaths overall (from ~40 annually in the 1990s to ~30
annually in the 2000s), part of the decreased incidence in SIDS might be
real, the result of public education concerning a safe infant sleeping
environment. (30) Most of the apparent decrease in SIDS is an artifact
of reclassification that resulted from more careful consideration of the
circumstances of death. (31) The OCME stopped classifying infant death
cases as SIDS in 2011, now officially using the term "sudden
unexplained infant death". However, "no anatomical cause of
death" was the most common cause in the infant deaths of
undetermined manner.
Unlike administrative datasets, this autopsy database report lacks
the direct links to socio-economic status, racial and ethnic factors,
and prior health service utilization. However, it offers the view of a
longer time span, which makes temporal trends more obvious. It also
provides more detail about injury types and survival. The data presented
here mirror data held in administrative databases, such as the Canadian
Vital Statistics system at Statistics Canada and that of the Manitoba
Centre for Health Policy. (8,32,33) These data supplement such
administrative databases. For example, the Canadian Hospitals Injury
Reporting and Prevention Program is known to underreport fatal injuries.
(34) Administrative databases may be limited by inaccurate coding and
may not be able to determine specific causes of death. (35) Linkage
between the medical examiner's autopsy records and administrative
data are likely to yield a more complete view of the circumstances of
death for population research. (36) Statistics Canada recently initiated
a Canadian Coroner and Medical Examiner Database to incorporate detailed
data from reported deaths, including autopsy information. However, data
have been reported only for 2006-2008, and information from Manitoba has
not yet been incorporated. (37)
Because this was a 22-year data analysis, we were forced to use
multiple resources to gather information. For most of the medical
datasets, the definition of the end of childhood is the 18th birthday,
but the population datasets were clustered by decade. This can lead to a
slight discrepancy in the proportions calculated. Had we included
autopsies on individuals in their 18th and 19th years, the proportion of
deaths related to traffic accidents and violence would likely have
increased. Regardless of these uncertainties, the temporal features are
not altered.
Non-natural deaths remain a major problem among Manitoba children,
particularly in rural areas. In the ideal world, accidental deaths (road
traffic, drowning, fire-related) should be preventable by education and
reasonable safety measures. Reduced numbers of infant deaths through
improved education about safe sleeping conditions should also be
feasible. Because the majority of deaths occur at the scene or within
the first few hours, improved hospital care would not be expected to
play a major role in reducing non-natural childhood deaths, although we
cannot comment on the outcomes of severe non-fatal insults. Given that
these non-natural deaths are more likely in rural and impoverished
communities, improved societal support might be expected to achieve
positive results; however, the contributing socio-economic factors are
difficult to modify. Linkage between autopsy and administrative
databases should amplify the value of both and, it is hoped, would offer
a tool for social change.
Received: July 18, 2013 Accepted: January 26, 2014
Acknowledgements: We thank the Office of the Chief Medical Examiner
of Manitoba for the provision of data. This work was made possible by a
grant from Diagnostic Services Manitoba, which was used to create a
searchable autopsy database. Dr. Del Bigio holds the Canada Research
Chair in Developmental Neuropathology.
REFERENCES_
(1.) Statistics Canada. CANSIM database Table 102-0561 - Leading
causes of death, total population, by age group and sex, Canada, annual.
2009. Available at: http://www5.statcan.gc.ca/cansim/a05?lang=eng&id=1020561 (Accessed January 20, 2013).
(2.) UNICEF. A League Table of Child Deaths by Injury in Rich
Nations, Innocenti Report Card No. 2. Florence, Italy: UNICEF Innocenti
Research Centre, February 2001.
(3.) SMARTRISK. The Economic Burden of Injury in Canada. Toronto,
ON: SMARTRISK, 2009.
(4.) Chen Y, Mo F, Yi QL, Jiang Y, Mao Y. Unintentional injury
mortality and external causes in Canada from 2001 to 2007. Chronic Dis
Inj Can 2013; 33(2):95-102.
(5.) Martens PJ, Derksen S. A matter of life and death for
Manitoba's children: An overview of birth rates and mortality
rates. Can J Public Health 2002; 93(Suppl 2):S2l-26.
(6.) Statistics Canada. CANSIM database Table 102-0510 - Deaths
subject to autopsy. 2009. Available at:
http://www5.statcan.gc.ca/cansim/ a05?lang=eng&id=1020561 (Accessed
November 3, 2012).
(7.) Martens P, Brownell M, Au W, MacWilliam L, Prior H, Schultz J,
et al. Health Inequities in Manitoba: Is the Socioeconomic Gap Widening
or Narrowing Over Time? Winnipeg, MB: Manitoba Centre for Health Policy,
2010.
(8.) Brownell M, De Coster C, Penfold R, Derksen S, Au W, Schultz
J, et al. Manitoba Child Health Atlas Update. Winnipeg: Manitoba Centre
for Health Policy, 2008.
(9.) Chen W-H. Comparing low income of Canada's regions: A
stochastic dominance approach. Table 5, Headcount low-income rankings
for selected low-income lines, 2000. Statistics Canada. 2008. Available
at: http://www.statcan.gc.ca/pub/75f0002m/75f0002m2008006-eng.pdf.
(10.) Bristow KM, Carson JB, Warda L, Wartman R. Childhood drowning
in Manitoba: A 10-year review of provincial Paediatric Death Review
Committee data. Paediatr Child Health 2002; 7(9):637-41.
(11.) MacNeil MS. An epidemiologic study of aboriginal adolescent
risk in Canada: The meaning of suicide. J Child AdolescPsychiatr Nurs
2008; 2l(l):3-l2.
(12.) Malchy B, Enns MW, Young TK, Cox BJ. Suicide among
Manitoba's aboriginal people, 1988 to 1994. CMAJ 1997;
l56(8):1133-38.
(13.) Katz LY, Au W, Singal D, Brownell M, Roos N, Martens PJ, et
al. Suicide and suicide attempts in children and adolescents in the
child welfare system. CMAJ 20ll; 183(l7):1977-8l.
(14.) Luo ZC, Wilkins R, Heaman M, Martens P, Smylie J, Hart L, et
al. Birth outcomes and infant mortality by the degree of rural isolation
among first nations and non-first nations in Manitoba, Canada. J Rural
Health 2010; 26(2):175-81.
(15.) Karmali S, Laupland K, Harrop AR, Findlay C, Kirkpatrick AW,
Winston B, et al. Epidemiology of severe trauma among status Aboriginal
Canadians: A population-based study. CMAJ2005; 172(8):1007-11.
(16.) Brownell M, Friesen D, Mayer T. Childhood injury rates in
Manitoba: Socioeconomic influences. Can J Public Health 2002; 93(Suppl
2):S50-56.
(17.) Canadian Institute for Health Information. Injury
Hospitalizations and SocioEconomic Status. Ottawa, ON, 2010.
(18.) Pan SY, Desmeules M, Morrison H, Semenciw R, Ugnat AM,
Thompson W, et al. Adolescent injury deaths and hospitalization in
Canada: Magnitude and temporal trends (1979-2003). J Adolesc Health
2007; 41(1):84-92.
(19.) Condello AS, Hancock BJ, Hoppensack M, Tenenbein M, Stewart
TC, Kirwin D, et al. Pediatric trauma registries: The foundation of
quality care. JPediatr Surg 200l; 36(5):685-89.
(20.) Warda L, Tenenbein M, Moffatt ME. House fire injury
prevention update. Part I. A review of risk factors for fatal and
non-fatal house fire injury. Inj Prev 1999; 5(2):145-50.
(21.) SMARTRISK. The Economic Burden of Unintentional Injury in
Manitoba. Toronto: SMARTRISK, 2003.
(22.) Statistics Canada. CANSIM database Table 102-0552 - Deaths
and mortality rate, by selected grouped causes and sex, Canada,
provinces and territories, annual. 2009. Available at:
http://www5.statcan.gc.ca/cansim/ a05?lang=eng&id=102056l (Accessed
November 3, 2012).
(23.) Skinner R, McFaull S. Suicide among children and adolescents
in Canada: Trends and sex differences, 1980-2008. CMAJ 2012;
184(9):1029-34.
(24.) Austin AE, van den Heuvel C, Byard RW. Cluster hanging
suicides in the young in South Australia. J Forensic Sci 20ll;
56(6):1528-30.
(25.) Bernburg JG, Thorlindsson T, Sigfusdottir ID. The spreading
of suicidal behavior: The contextual effect of community household
poverty on adolescent suicidal behavior and the mediating role of
suicide suggestion. Soc Sci Med 2009; 68(2):380-89.
(26.) Wissow LS, Walkup J, Barlow A, Reid R, Kane S. Cluster and
regional influences on suicide in a Southwestern American Indian tribe.
Soc Sci Med 200l; 53(9):1115-24.
(27.) Swanson SA, Colman I. Association between exposure to suicide
and suicidality outcomes in youth. CMAJ 2013; l85(l0):870-77.
(28.) Pollanen MS, Smith CR, Chiasson DA, Cairns JT, Young J. Fatal
child abusemaltreatment syndrome. A retrospective study in Ontario,
Canada, 1990-1995. Forensic Sci Int 2002; 126(2):l0l-4.
(29.) Ogrodnik L. Child and Youth Victims of Police-reported
Violent Crime, 2008. Ottawa, ON: Statistics Canada, 2010. Catalogue no.
85F0033M, no. 23.
(30.) Moon RY. SIDS and other sleep-related infant deaths:
Expansion of recommendations for a safe infant sleeping environment.
Pediatrics 20ll; 128(5):l030-39.
(31.) Byard RW. Changing infant death rates: Diagnostic shift,
success story, or both? Forensic Science, Medicine, and Pathology 20l3;
9(l):l-2.
(32.) Pan SY, Ugnat AM, Semenciw R, Desmeules M, Mao Y, Macleod M.
Trends in childhood injury mortality in Canada, 1979-2002. Inj Prev
2006; 12(3):155-60.
(33.) Brownell M, Mayer T, Martens PJ, Kozyrskyj A, Fergusson P,
Bodnarchuk J, et al. Using a population-based health information system
to study child health. Can J Public Health 2002; 93(Suppl 2):S9-S14.
(34.) Public Health Agency of Canada. Child and Youth Injury in
Review, 2009 Edition - Spotlight on Consumer Product Safety. Ottawa:
PHAC, 2009.
(35.) Christianson-Wood J, Murray JL. Child Death Reviews and Child
Mortality Data Collection in Canada. Ottawa: PHAC, 1999.
(36.) Campbell LA, Jackson L, Bassett R, Bowes MJ, Donahue M,
Cartwright J, et al. Can we use medical examiners' records for
suicide surveillance and prevention research in Nova Scotia? Chronic Dis
Inj Can 20ll; 31(4):165-71.
(37.) Anonymous. Canadian Coroner and Medical Examiner Database:
Annual Report 2006 to 2008. Ottawa: Statistics Canada.
Jayantha C. Herath, MD, FRCPC, [1] Saman Kalikias, MD, [2] Susan M.
Phillips, MD, FRCPC, [3] Marc R. Del Bigio, MD, PhD, FRCPC [3]
Author Affiliations
[1.] Department of Pathology, University of Manitoba, and
Diagnostic Services Manitoba, Winnipeg, MB (at time of study; currently
with Ontario Forensic Pathology Service and University of Toronto,
Toronto, ON)
[2.] Faculty of Medicine (Postgraduate Medical Education),
University of Manitoba, Winnipeg, MB
[3.] Department of Pathology, University of Manitoba, and
Diagnostic Services Manitoba, Winnipeg, MB
Correspondence: Marc R. Del Bigio, Department of Pathology,
University of Manitoba, 401 Brodie Centre, 727 McDermot Ave., Winnipeg,
MB R3E 3P5, Tel: 204-789-3378, E-mail: marc.delbigio@med.umanitoba.ca
Conflict of Interest: None to declare.
Table 1. Childhood deaths in Manitoba--Manner and mortality rate *
Year Total Total Total Accident
postneonatal deaths childhood (0-18 y)
deaths 0-19 y mortality
1 to 12 ([dagger]) rate/100,000
months population
([dagger]) ([double dagger])
1989 nr nr -- 19
1990 nr nr -- 29
1991 45 261 23.5 28
1992 43 220 19.8 29
1993 38 234 20.9 26
1994 43 245 21.8 41
1995 42 235 20.8 37
1996 37 216 19.0 38
1997 39 246 21.7 47
1998 30 210 18.5 47
1999 36 236 20.7 55
2000 38 210 18.3 39
2001 35 206 17.9 38
2002 35 200 17.3 35
2003 30 213 18.3 30
2004 29 200 17.0 38
2005 27 212 18.0 28
2006 25 192 16.2 32
2007 36 221 18.5 28
2008 34 200 16.6 25
2009 26 217 17.8 22
2010 27 202 16.3 28
Year Child Suicide Child Homicide
accident (0-18 y) suicide (0-18 y)
mortality mortality
rate/100,000 rate/100,000
population population
1989 1.7 0 0 1
1990 2.6 0 0 3
1991 2.5 13 1.2 4
1992 2.6 3 0.3 5
1993 2.3 13 1.2 3
1994 3.7 14 1.2 2
1995 3.3 12 1.1 4
1996 3.4 18 1.6 9
1997 4.1 14 1.2 2
1998 4.1 7 0.6 6
1999 4.8 14 1.2 16
2000 3.4 17 1.5 18
2001 3.3 13 1.1 4
2002 3 14 1.2 7
2003 2.6 12 1.0 8
2004 3.2 18 1.5 5
2005 2.4 25 2.1 11
2006 2.7 14 1.2 4
2007 2.3 13 1.1 11
2008 2.1 12 1.0 7
2009 1.8 21 1.7 11
2010 2.3 16 1.3 4
Year Child Undetermined Complete
homicide manner * autopsies
mortality ([section]) on
rate/100,000 children
population ([parallel])
1989 0.1 -- 20
1990 0.3 -- 32
1991 0.4 -- 36
1992 0.4 -- 36
1993 0.3 -- 32
1994 0.2 -- 47
1995 0.4 -- 47
1996 0.8 24 49
1997 0.2 30 51
1998 0.5 18 49
1999 1.4 1 51
2000 1.6 8 52
2001 0.3 17 55
2002 0.6 11 52
2003 0.7 15 53
2004 0.4 12 56
2005 0.9 9 61
2006 0.3 15 49
2007 0.9 16 43
2008 0.6 9 32
2009 0.9 11 45
2010 0.3 16 47
* Manner of death includes accident, suicide, homicide and
undetermined, as recorded by the Office of the Chief Medical
Examiner of Manitoba for individuals <18 years of age.
([dagger]) Total deaths represent the sum of all manners, including
natural causes, as reported by Statistics Canada (CANSIM database)
for each year (nr = not reported). Postneonatal deaths do not
include stillbirths or deaths prior to 1 month of age (which
includes mainly premature births and obstetric complications, i.e.,
natural). CANSIM data are aggregated by decade, therefore "0-19 y"
is <20 y.
([double dagger]) Childhood mortality rates are calculated against
the annual total Manitoba population as reported by Statistics
Canada.
([section]) Recording of undetermined manner of death by the OCME was
not summarized consistently prior to 1995.
([parallel]) Autopsies performed on individuals <18 years of age at
the Health Sciences Centre (HSC), Winnipeg, excluding natural
causes. Prior to 1994, some infant and child Autopsies performed on
individuals <18 years of age at the Health Sciences Centre (HSC),
Winnipeg, excluding natural causes. Prior to 1994, some infant and
child autopsies, including sudden infant death syndrome (i.e.,
predominantly natural and undetermined manner), were performed at
other sites. After 1994, the pediatric autopsies were concentrated
at the HSC site.
Table 2. Childhood deaths in Manitoba--Selected major causes
documented at autopsy*
Year Vehicle Pedestrian- Bicycle Drowning
passenger vehicle (accident) (accident)
or driver (accident)
(accident)
1989 1 1 1 3
1990 4 6 2 3
1991 1 8 1 7
1992 3 3 2 4
1993 4 2 1 6
1994 8 3 2 5
1995 9 3 2 11
1996 9 4 4 5
1997 7 3 1 7
1998 13 5 0 9
1999 11 6 2 2
2000 10 1 0 8
2001 8 4 1 2
2002 9 2 2 2
2003 4 6 0 3
2004 4 4 1 8
2005 7 3 3 6
2006 6 2 3 1
2007 4 5 1 1
2008 0 1 1 5
2009 3 1 0 5
2010 11 3 2 3
Year Burns Hanging Child SIDS
or smoke (suicide) abuse (undeter-
inhalation (homicide) mined) ([double
(accident) ([dagger]) dagger])
1989 6 0 1 21
1990 3 0 2 24
1991 0 1 1 12
1992 1 2 2 14
1993 0 2 1 11
1994 13 3 1 12
1995 0 6 3 15
1996 4 5 5 11
1997 3 7 1 12
1998 5 4 4 8
1999 7 11 2 7
2000 4 15 6 5
2001 5 11 0 6
2002 2 12 4 3
2003 0 9 4 2
2004 2 11 3 1
2005 0 20 3 2
2006 0 12 2 5
2007 0 7 3 4
2008 5 7 1 2
2009 2 17 1 2
2010 4 12 2 4
* Autopsies performed on individuals <18 years of age at the Health
Sciences Centre (HSC), Winnipeg. The manner of death is shown in
parentheses. ([dagger]) All are <3 years of age, the majority are
blunt trauma to the head and/or blunt trauma to the abdomen.
([double dagger]) Sudden infant death syndrome reported by the Office
of the Chief Medical Examiner of Manitoba. Not all autopsies were
performed at the HSC. See Discussion for explanation of apparent
decline. Use of the term SIDS was discontinued in 2011 and replaced
by sudden unexplained infant deaths with risk factors specified
(e.g., unsafe sleeping arrangements).
