Rates and external causes of blunt head trauma in Ontario:

Analysis and review of Ontario Trauma Registry datasets

Vol. 25 No. 1, 2004

William Pickett, Kelly Simpson and Robert J Brison

Abstract

Contemporary studies of blunt head trauma and its determinants are important for prevention. It is also important to understand the strengths and limitations of the common sources of data used for the ongoing study of these injuries. Using the Ontario Trauma Registry, we described frequent patterns of blunt head trauma and identified priorities for prevention and research. A review of methodological issues that arose during the analysis of these trauma registry data is also provided. Blunt head trauma cases were identified within two data sets of the Ontario Trauma Registry. The Minimal Data Set is population- based and contains acute care injury hospitalizations, and the Comprehensive Data Set contains "major injuries" treated at a lead trauma hospital. Injury control priorities varied by age group, sex and data set and these are profiled in the manuscript. The results indicate the importance of examining multiple sources of surveillance data in establishing injury control priorities. The methodological review demonstrated the need to critically examine the completeness and accuracy of trauma registry data in arriving at decisions about priorities.

Key words: blunt head trauma; head injury; injury surveillance; neurotrauma prevention

Introduction

Contemporary studies of blunt head trauma and its determinants are important to the prevention of these injuries and their clinical management. These forms of neurotrauma contribute to high levels of morbidity, long-term disability, mortality, and associated economic burdens.1-3 Injuries are generally preventable, non-random events,4 therefore prevention efforts aimed at reducing the magnitude and minimizing the consequences of these head injuries are important. Formal quantification of the magnitude of the problem and assessment of injury patterns are necessary steps in the development of prevention efforts, and basic epidemiological analyses are helpful in this regard.

Reported incidence rates for head injury range from 17 to 444 per 100,000 population annually.5-16 Definitions and terminology used in the study of these injuries vary, which contributes to the disparity in results and makes it difficult to compare study findings.17,18 Published case series use definitions of head trauma that range from mild to severe forms.19 Potential sources of cases vary from records of emergency department and outpatient visits to those that describe hospitalizations and deaths.18 Common patterns of injury experienced vary with the severity of the cases under study. For example, the proportion of head injuries attributable to motor vehicle crashes increases with severity of injury18,20-23 whereas falls result in severe head trauma less often but are a recurrent cause of minor head injury.3,18,22

There are no published studies describing contemporary rates and patterns of head injury for a large Canadian population. Existing studies are limited to non-peer review reports,15,16 smaller scale studies,7,24,25 or have restricted their focus to specific age groups21,26-29 or sports related causes of injury.30 Existence of a provincial trauma registry in Ontario provided a practical opportunity to describe the occurrence of blunt head trauma for a large Canadian population. This study fills a void in the existing neurotrauma literature by comparing the descriptive epidemiological results of two data sets and also by providing a means for comparison with the injury experiences within other jurisdictions.

In this analysis, we examined two data sets associated with the Ontario Trauma Registry. The Minimal Data Set contains records for all acute care hospitalizations in the province of Ontario and the Comprehensive Data Set contains records for "major injuries" treated at any Ontario lead trauma hospital. Comparison of records from these data sets provides an opportunity to investigate whether different injury prevention and research priorities emerge from these two "windows" of surveillance. This in turn may have implications for policy and associated priorities for prevention. (Note: deaths occurring outside of hospital caused by head injury and outpatient visits for treatment of head injury were not considered in this analysis. These injuries may have patterns of occurrence that are distinct from the hospitalized injuries examined here, and offer different opportunities for prevention).

Our specific objectives were to: 1) calculate rates and describe contemporary patterns of blunt head trauma for the province of Ontario; and, 2) compare priorities for focused prevention and research initiatives derived from the Minimal and Comprehensive Data Sets. We also took this opportunity to discuss methodological issues that arose during the use and application of data from this registry. These issues are relevant to researchers using trauma registry data and can be used to form the basis of future studies investigating data quality.

Methods

Data sources

The Ontario Trauma Registry is a provincial initiative funded by the government of Ontario and managed by the Canadian Institute for Health Information.31 The goal of the Ontario Trauma Registry is to "facilitate the reduction of injuries by clearly identifying, describing, and quantifying the nature and scope of injuries in the province of Ontario."31

Inclusion and exclusion criteria for the Ontario Trauma Registry are fully documented in technical reports from the Canadian Institute for Health Information.31,32 In short, injuries contained in the Minimal and Comprehensive Data Sets include those that resulted from the transfer of energy. All cases are coded according to International Classification of Diseases 9th Revision (ICD-9) external cause of injury codes (E codes) and nature of illness codes (N codes).33

The Minimal Data Set contains records for all acute care injury hospitalizations in Ontario. The Comprehensive Data Set contains records for "major injuries", defined as an Injury Severity Score (ISS) greater than 12,34 treated at a lead trauma hospital in Ontario. Patients included are those who are admitted as inpatients, treated in the emergency department, or who die in the emergency department after treatment of a major injury is initiated in a lead trauma hospital. Hospitals included in the Comprehensive Data Set are situated in major urban Ontario centres as follows: Hamilton, Kingston, London, Ottawa, Sudbury, Thunder Bay, Toronto, and Windsor.

A portion of the records included in the Minimal Data Set are contained in the Comprehensive Data Set, but the latter contains detail about the external causes and circumstances of injury events beyond that which is available in the Minimal Data Set. The two data sets are not mutually exclusive although each contains potentially different injury patterns that are helpful for prevention. Individual identifiers that would allow one to link the two datasets for research purposes are not made available by the Registry to external researchers such as ourselves.

Case selection

The following ICD-9 diagnostic codes are consistent with blunt head trauma, and records for cases were abstracted from the Minimal and Comprehensive Data Sets if at least one these codes were present in any diagnostic field (up to 16 and 27 diagnostic fields were available for review in the Minimal and Comprehensive Data Sets, respectively): 1) N800 (fracture of the vault of the skull); 2) N801 (fracture of the base of the skull); 3) N803 (other and unqualified skull fractures); 4) N804 (multiple fractures involving skull of face with other bones); 5) N850 (concussion); 6) N851 (cerebral laceration and contusion); 6) N852 (subarachnoid, subdural and extradural hemorrhage, following injury); 7) N853 (other and unspecified intracranial hemorrhage, following injury); 8) N854 (intracranial injury of other and unspecified nature). Data abstracted from the Minimal Data Set were for the fiscal years of 1994/95 through 1998/99, while data from the Comprehensive Data Set were for 1994/95 through 1997/98.

Statistical analysis

Annual age-standardized hospitalization rates of blunt head trauma were calculated for cases from the Minimal Data Set (MDS) of the Ontario Trauma Registry. Age-specific rates (five-year age groups) were first calculated by sex using cumulative counts of injuries over five years from the MDS in the numerator (1994/95 through 1998/99 estimates), and Ontario denominator data for the same time period (1994-1998 population estimates) from the Canada Census of Population.35 The demographic structure of the 1991 general population of Canada35 was used in the calculation of age standardized rates for the five year period of study. Confidence intervals surrounding these rates were calculated according to procedures outlined by Breslow and Day.36 Mean annual age-specific rates were calculated by sex and by ten-year age group.

Patterns of head injury in the Minimal and Comprehensive Data Sets were described via frequencies and cross-tabulations that examined external causes (primary E Code only), age group (<20, 20-59, 60+), sex, and most responsible diagnosis (Minimal Data Set only). The ages were broadly classified into three age groups because similar injury patterns were observed within these categories. Using the Minimal Data Set, specific rates were calculated for the external causes by age group and sex, then by region of Ontario (Southwest, Central South, Central West, Toronto, East and North). It was not possible to calculate rates for the Comprehensive Data Set as it is not a population-based data source (only data from the lead trauma hospitals are included). All analyses were conducted in SPSS (v.11.0, Chicago, IL).

Results

For the years under study, approximately 12% of patients in the Minimal Data Set and 70% of patients in the Comprehensive Data Set sustained at least one head injury. Annual age-standardized hospitalization rates for head injury declined over time, from 85.3 per 100,000 in 1994/95 (95% CI: 83.6-87.0) to 62.7 per 100,000 in 1998/99 (95% CI: 61.2-64.1; Figure 1). Males experienced higher injury rates than females within each age group (Figure 2). Rates of head trauma were highest for both sexes among the elderly (70+ age group) although a modest peak was observed among males aged 10-19 years.

FIGURE 1
Age standardized rates of head injury hospitalization in Ontario, 1994/95 through 1998/99 (from the Minimal Data Set of the Ontario Trauma Registry)

FIGURE 2
Age-specific rates of head injury hospitalization in Ontario, overall 1994/95 through 1998/1999 (from the Minimal Data Set of the Ontario Trauma Registry)

Leading external causes of injury differed between the Minimal and Comprehensive Data Sets (Table 2). In the Minimal Data Set, unintentional falls were the leading external cause of blunt head trauma (19,423/40,392; 48.1%) followed by transport incidents (14,249/40,392; 35.3%) and unintentionally being struck by or against an object or person (2,721/40,392; 6.7%; Table 1). In the Comprehensive Data Set, transport incidents were the primary external cause of injury (4,938/8,512; 58.0%) followed by unintentional falls (2,413/8,512; 28.3%) and injury purposely inflicted by another person (497/8,512; 5.8%).

TABLE 1
External causes of head injury from the Minimal Data Set (MDS) and
Comprehensive Data Set (CDS) of the Ontario Trauma Registry

External cause of injury MDS
1994/95 through 1998/99
CDS
1994/95 through 1997/98

No.
% of total % of subtotal
No.
% of total % of subtotal
Fall (unintentional) - subtotal 19,423 48.1   2,413 28.3  
  Fall on same level from slip, trip, or stumble 4,458   23.0 418   17.3
Fall on or from stairs or steps 3,385   17.4 604   25.0
Fall on or from ladders or scaffolding 809   4.2 217   9.0
Fall from or out of building or other structure 553   2.8 179   7.4
Fall on same level from collision, push, or shove 430   2.2 21   0.9
Fall into hole or other opening in surface 86   0.4 15   0.6
Other fall from one level to another 3,934   20.3 288   11.9
Other & unspecified fall 5,768   29.7 671   27.8
Transport incident (unintentional) - subtotal 14,249 35.3   4,938 58.0  
  Motor vehicle incident 12,196   85.6 4,679   94.8
  Occupant 8,666     3,269    
Pedestrian 1,919     912    
Motorcyclist 604     246    
Pedal cyclist 556     190    
Other & unspecified 451     62    
Other road vehicle 1,715   12.0 164   3.3
  Pedal cyclist 1,355     132    
Rider of animal 227     25    
Pedestrian 78     5    
Other & unspecified 55     2    
Vehicle incidents not elsewhere classifiable 164   1.2 27   0.5
Water transport incidents 113   0.8 30   0.6
Railway 35   0.2 28   0.6
Air & space transport incidents 26   0.2 10   0.2
Struck by, against (unintentional) - subtotal 2,721 6.7   213 2.5  
  Against or by objects/persons 2,311   84.9 134   62.9
Falling object 410   15.1 79   37.1
Injury purposely inflicted by another person - subtotal 2,520 6.2   497 5.8  
  Unarmed fight or brawl 1,447   57.4 195   39.2
Child battering, other maltreatment 159   6.3 68   13.7
Other & unspecified 914   36.3 234   47.1
Self-inflicted injury - subtotal 230 0.6   193 2.3  
  Jump from high place or before a moving object 100   43.5 95   49.2
Firearms or explosives 71   30.9 59   30.6
Other & unspecified 59   25.7 39   20.2
Other & unspecified 1,249 3.1   258 3.0  
Total 40,392 100.0 100.0 8,512 100.0 100.0

Frequencies of blunt head trauma also differed by age group (Table 2). In the Minimal Data Set, there was a larger proportion of head injuries among those less than 20 years of age (14,024/40,392; 34.7%) when compared with the Comprehensive Data Set (1,822/8,512; 21.4%). Similar proportions (approximately 26%) of head injuries were seen among the elderly (60+ age group) in both data sources. Unintentional falls were most common among the elderly and transport incidents were most common among the 20-59 age group (in both data sets).

TABLE 2
External causes of head injury by age group from the Minimal Data Set (MDS) and
Comprehensive Data Set (CDS) from the Ontario Trauma Registry

External cause of injury MDS
1994/95 through 1998/99
Number of injuries
Age group
CDS
1994/95 through 1997/98
Number of injuries
Age group *
<20 20-59 60+ <20 20-59 60+
Fall (unintentional) 6,341 4,567 8,515 242 796 1,374
  Fall on same level from slip, trip, or stumble 1,072 986 2,400 21 94 303
Fall on or from stairs or steps 963 921 1,501 39 208 357
Fall on or from ladders or scaffolding 50 476 283 3 130 84
Fall from or out of building or other structure 227 269 57 43 110 25
Fall on same level from collision, push, or shove 325 88 17 9 8 4
Fall into hole or other opening in surface 37 41 8 3 9 3
Other fall from one level to another 2,709 501 724 107 82 99
Other & unspecified fall 958 1,285 3,525 17 155 499
Transport incident (unintentional) 5,071 7,409 1,769 1,282 2,943 711
  Motor vehicle incident 3,779 6,752 1,665 1,179 2,810 688
  Occupant 2,282 5,262 1,122 748 2,102 419
Pedestrian 831 669 419 278 393 240
Motorcyclist 166 411 27 38 205 3
Pedal cyclist 344 173 39 98 76 15
Other & unspecified 156 237 58 17 34 11
Other road vehicle 1,097 532 86 67 78 19
  Pedal cyclist 921 373 61 55 62 15
Rider of animal 109 112 6 10 13 2
Pedestrian 42 27 9 1 2 2
Other & unspecified 25 20 10 1 1 0
Vehicle incidents not elsewhere classifiable 144 19 1 22 5 0
Water transport incidents 37 70 6 8 22 0
Railway 7 24 4 6 20 2
Air & space transport incidents 7 12 7 0 8 2
Struck by, against (unintentional) 1,644 882 195 85 102 26
  Against or by objects/persons 1,554 606 151 70 49 15
Falling object 90 276 44 15 53 11
Injury purposely inflicted by another person 627 1,770 123 124 345 26
  Unarmed fight or brawl 322 1,068 57 24 161 9
Child battering, other maltreatment 151 6 2 68 0 0
Other & unspecified 154 696 64 32 184 17
Self-inflicted injury 28 170 32 19 149 25
  Jump from high place or before a moving object 8 88 4 5 83 7
Firearms or explosives 10 42 19 5 40 14
Other & unspecified 10 40 9 9 26 4
Other & unspecified 313 639 297 70 136 52
Total 14,024 15,437 10,931 1,822 4,471 2,214

* There were 5 cases with missing ages.

Specific annual rates of blunt head trauma by external cause varied between age/sex groups (Table 3). With some exceptions, unintentional fall injury rates were highest in the oldest age groups in both sexes. Rates of transport injury were highest among young males, and there was a male predominance in most categories of transport injuries, struck by or against objects, and intentional forms of injury, irrespective of age. There were also striking variations in regional rates of blunt head trauma in the Minimal Data Set (Table 4). Overall rates of injury were highest in Northern Ontario and lowest in Toronto. This pattern held true for all external causes of injury.

TABLE 3
Age-specific rates of head injury by sex and external causes, from
the Minimal Data Set (MDS) of the Ontario Trauma Registry

External cause of injury MDS
1994/95 through 1998/99
Annual rate of injury (per 100,000 population)
Males
Age group
Females
Age group
  <20 20-59 60+ <20 20-59 60+
Fall (unintentional) 52.2 20.2 105.6 32.7 8.8 84.3
  Fall on same level from slip, trip, or stumble 9.0 3.6 27.7 5.3 2.6 25.4
Fall on or from stairs or steps 7.3 3.9 19.2 5.6 2.0 14.4
Fall on or from ladders or scaffolding 0.5 2.8 6.4 0.2 0.2 0.5
Fall from or out of building or other structure 2.2 1.5 1.3 0.9 0.2 *
Fall on same level from collision, push, or shove 3.3 0.4 0.1 1.1 0.1 0.2
Fall into hole or other opening in surface 0.3 0.2 * 0.2 * *
Other fall from one level to another 21.9 2.4 9.1 14.4 0.8 7.1
Other & unspecified fall 7.8 5.3 41.7 5.1 2.8 36.5
Transport incident (unintentional) 43.3 31.6 26.0 24.5 15.4 14.3
  Motor vehicle incident 32.0 28.9 24.0 18.5 13.8 13.9
  Occupant 18.5 21.9 16.5 12.1 11.4 9.0
Pedestrian 6.6 2.7 5.1 4.5 1.5 4.2
Motorcyclist 2.0 2.3 0.6 0.2 0.3 *
Pedal cyclist 3.5 0.9 0.9 1.1 0.2 *
Other & unspecified 1.4 1.1 0.8 0.7 0.4 0.5
Other road vehicle 9.6 2.1 1.7 5.0 1.3 0.4
  Pedal cyclist 8.8 1.7 1.3 3.4 0.6 0.2
Rider of animal 0.3 0.2 0.1 1.2 0.5 *
Pedestrian 0.3 0.1 * 0.3 0.1 0.1
Other & unspecified 0.2 0.1 0.2 0.1 * *
Vehicle incidents not elsewhere classifiable 1.3 0.1 * 0.7 * -
Water transport incidents 0.3 0.3 0.1 0.2 0.2 *
Railway 0.1 0.1 * * * *
Air & space transport incidents 0.1 0.1 0.1 * * *
Struck by, against (unintentional) 15.6 4.4 3.3 6.3 1.2 1.3
  Against or by objects/persons 14.8 2.9 2.3 5.9 0.9 1.2
Falling object 0.7 1.5 1.0 0.5 0.3 0.1
Injury purposely inflicted by another person 6.8 9.5 2.3 1.5 1.7 0.6
  Unarmed fight or brawl 3.8 5.8 1.2 0.4 1.0 0.2
Child battering, other maltreatment 1.3 * * 0.7 * *
Other & unspecified 1.7 3.8 1.0 0.3 0.7 0.4
Self-inflicted injury 0.3 0.7 0.7 0.1 0.3 *
  Jump from high place or before a moving object 0.1 0.3 * * 0.2 *
Firearms or explosives 0.1 0.2 0.5 - * -
Other & unspecified 0.1 0.2 0.2 0.1 0.1 *
Other & unspecified 2.5 3.2 5.0 1.7 0.9 1.9
Total 120.6 69.6 142.9 66.8 28.3 102.5

* Suppressed due to small numbers.

TABLE 4
Regional rates of head injury by external causes, from the Minimal
Data Set (MDS) of the Ontario Trauma Registry

External cause of injury MDS
1994/95 through 1998/99
Annual rate of injury (per 100,000 population)
Region of Ontario
South West Central South Central West Central East Toronto East North
Fall (unintentional) 42.2 41.2 34.1 32.5 26.9 28.8 48.1
  Fall on same level from slip, trip, or stumble 8.5 7.2 8.1 9.2 7.7 5.9 9.4
Fall on or from stairs or steps 6.7 8.1 5.7 5.6 5.0 4.8 7.9
Fall on or from ladders or scaffolding 2.2 1.8 1.3 1.4 1.1 1.0 2.0
Fall from or out of building or other structure 1.4 1.0 0.8 0.7 0.9 0.7 1.4
Fall on same level from collision, push, or shove 1.1 0.7 0.9 1.0 0.5 0.4 1.1
Fall into hole or other opening in surface 0.3 0.2 0.1 0.2 0.1 0.1 0.2
Other fall from one level to another 9.0 8.2 7.7 6.9 4.6 5.6 9.7
Other & unspecified fall 13.1 14.1 9.7 7.6 7.1 10.2 16.4
Transport incident (unintentional) 34.9 29.1 22.6 25.6 16.6 20.0 35.4
  Motor vehicle incident 30.4 24.7 18.8 21.9 14.4 17.3 29.6
  Occupant 23.2 17.0 13.2 16.6 8.0 12.2 22.7
Pedestrian 3.2 3.7 2.9 2.6 4.8 2.5 3.4
Motorcyclist 1.6 1.7 0.9 1.0 0.5 1.0 1.4
Pedal cyclist 1.3 1.6 0.9 0.9 0.7 0.8 1.0
Other & unspecified 1.1 0.8 1.0 0.8 0.4 0.7 1.1
Other road vehicle 3.8 4.1 3.3 2.9 1.9 2.1 4.9
  Pedal cyclist 2.9 3.4 2.7 2.1 1.5 1.6 3.9
Rider of animal 0.6 0.5 0.4 0.6 0.1 0.3 0.6
Pedestrian 0.2 * 0.1 0.2 0.2 0.1 0.2
Other & unspecified 0.1 0.1 0.1 0.1 0.0 0.1 0.2
Vehicle incidents not elsewhere classifiable 0.3 0.2 0.3 0.4 0.1 0.4 0.4
Water transport incidents 0.2 0.1 0.1 0.2 0.1 0.2 0.4
Railway 0.1 * 0.1 * 0.1 * 0.1
Air & space transport incidents 0.1 - * 0.1 0.1 * *
Struck by, against (unintentional) 7.1 5.3 4.4 5.0 2.1 3.8 9.6
  Against or by objects/persons 6.1 4.8 3.9 4.3 1.9 3.0 7.5
Falling object 1.0 0.5 0.5 0.7 0.3 0.8 2.1
Injury purposely inflicted by another person 5.2 5.7 3.4 3.5 3.4 3.0 10.5
  Unarmed fight or brawl 3.4 3.6 2.0 2.3 1.2 1.4 7.0
Child battering, other maltreatment 0.3 0.4 0.2 0.2 0.2 0.3 0.4
Other & unspecified 1.5 1.7 1.1 1.0 2.0 1.3 3.1
Self-inflicted injury 0.5 0.4 0.4 0.3 0.5 0.3 0.3
  Jump from high place or before a moving object 0.1 0.1 0.1 0.1 0.4 0.1 *
Firearms or explosives 0.3 0.2 0.1 0.1 0.1 0.1 0.2
Other & unspecified 0.1 0.1 0.1 0.2 0.1 0.1 0.1
Other & unspecified 2.3 2.1 3.6 1.7 1.6 1.8 2.7
Total 92.3 83.8 68.5 68.7 51.2 57.8 106.7

* Suppressed due to small numbers.

Note: Records are coding according to place of residence. There were 787 records with a residence code outside of Ontario, 114 transients, and 34 with unspecified residence.

Finally, in the Minimal Data Set, 40.3% (11,920/29,570) of the records had a most responsible diagnosis of N854 (intracranial injury of other and unspecified nature) and 19.2% (5,670/29,570) had a diagnosis of N850 (concussion; Table 5).

TABLE 5
Most responsible diagnosis for head injury cases from the
Minimal Data Set (MDS) of the Ontario Trauma Registry

  MDS
1994/95 through
1998/99
Total number of head injury cases 40,392
Total number of cases with a most responsible diagnosis of head injury 29,570
Most responsible head injury diagnosis No. %
N800 (fracture of the vault of the skull) 1,803 6.1
N801 (fracture of the base of the skull) 3,090 10.4
N803 (other and unqualified skull fractures) 891 3.0
N804 (multiple fractures involving skull of face with other bones) 137 0.5
N850 (concussion) 5,670 19.2
N851 (cerebral laceration and contusion) 1,433 4.8
N852 (subarachnoid, subdural and extradural hemorrhage, following injury) 3,603 12.2
N853 (other and unspecified intracranial hemorrhage, following injury) 1,023 3.5
N854 (intracranial injury of other and unspecified nature) 11,920 40.3

Discussion

Epidemiological patterns and trends

This study represents one of the first large-scale epidemiological analyses of contemporary neurotrauma data in Canada. One of the two data sets maintained by the Ontario Trauma Registry (Minimal Data Set) is population-based, while the second (Comprehensive Data Set) can be used to identify frequent patterns of severe forms of head injury treated in a lead trauma hospital. When considered together, results from these analyses are helpful in identifying priorities for focused prevention and research efforts. The results also provide a basis for comparison with other populations.

The magnitude of the head injury problem observed in Ontario, although substantial, fell within the range of rates published elsewhere.5-16 The age- and sex-specific injury trauma rates observed were also consistent with trends observed elsewhere, in that rates among males exceeded those among females in every age group.6,10,11,18,20 The excess rates of injury observed in Northern Ontario suggest that head trauma is an especially important problem in rural and remote parts of the province.

The annual rate of hospitalized injuries due to head trauma declined over the study period. While it is tempting to attribute this temporal decline to existing prevention efforts, the decline may also relate to contemporary medical practice, for example changes in access to diagnostic modalities (e.g., computerized tomography imaging) or admission practices with hospital restructuring and rationalization.3 Upon further investigation the number of "major injuries" in the Comprehensive Data remained stable whereas the number of injury hospitalizations in the Minimal Data Set declined (data not shown). This provides evidence that major head injuries are not declining.

The results indicate the value of examining multiple sources of surveillance data in order to identify leading priorities for prevention. When all hospitalizations for Ontario were examined via the Minimal Data Set, unintentional falls were the leading external cause, representing 48.1% of the reported injuries. This was followed by transport injuries (35.3%), being struck by or against objects (6.7%), and injuries purposely inflicted by another person (6.2%). When the more serious injuries captured via the Comprehensive Data Set were examined, the priorities changed with transport injuries accounting for 58.0% of the injuries observed. Other causes of injury remained important but their relative frequency of occurrence generally decreased. This reinforces the importance of considering the source of injury data when establishing priorities for intervention. By extrapolation, different priorities are likely to emerge if data are obtained from primarily outpatient (e.g., emergency department) sources, versus hospital inpatient or fatality-based records.

Based on the injury frequencies and rates presented here, common external causes that are obvious priorities for focused etiologic and preventive work include: 1) transport injuries (all ages); 2) falls in the elderly (60+ years); 3) unintentionally being struck by an object or person (among those less than 60 years of age); 4) injuries purposely inflicted by another person (e.g., assaults); 5) all external causes of injury in northern and remote areas of Ontario. Caution must be exercised in viewing these as priorities, as others might emerge if different criteria (e.g., evidence surrounding the ability to intervene) and other types of surveillance data (e.g., mortality) are applied to their development.

Strengths, limitations, and methodological issues

Strengths and weaknesses of this epidemiological analysis warrant recognition. Obvious strengths include the large number of cases available for analysis, the population-based features of the Ontario Trauma Registry, and the importance of the topic. Limitations include the use of data collected for administrative purposes as a basis for epidemiological analyses. Several methodological issues require consideration.

First, while there is a centralized agency responsible for record keeping (the Canadian Institute for Health Information or CIHI), and CIHI has training and quality control mechanisms in place for the coding of medical records, this process involves hundreds of hospital medical records departments and potential coders. There is clearly room for error here, and the extent of misclassification of these records should be better understood if the patterns of injury are to be interpreted correctly.

Second, some cases included in the Minimal Data Set are also contained in the Comprehensive Data Set. These data sets and the patterns derived from them cannot be considered mutually exclusive. In this analysis it was not possible to link the data sets. The creation of two mutually exclusive data sets could lead to more refined epidemiological analyses. Despite this limitation there were differences observed between the data sets and these differences would only be enhanced should the two data sets be refined.

Third, blunt head trauma commonly occurs in conjunction with other injuries7,11 and on some occasions the latter may influence the likelihood of hospitalization. Patients included in the Minimal and Comprehensive Data Sets may not be included solely due the effects of their head trauma. This may lead to less focused epidemiological descriptions of injury.

Fourth, it is also possible that patients have experienced multiple head injuries. There is no standard method for presenting the natures of injuries from multiple diagnostic fields. In the analysis of the Minimal Data Set, the most responsible diagnosis was used in order to describe leading natures of head injury. Our rationale for using the most responsible diagnosis is that it represents the diagnosis that was considered, upon discharge, as most responsible for the patient's stay. There is no equivalent "most responsible" diagnosis contained in the Comprehensive Data Set; as such it was not possible to compare diagnoses between the two data sets. This disparity represents an important challenge for comparative research.

Fifth, 40.3% of cases in the Minimal Data Set had a most responsible diagnosis of "intracranial injury of other and unspecified nature". This lack of specificity is a methodological concern because it introduces an element of uncertainty to the injury patterns observed. Diagnoses are coded on the hospital discharge summary by trained medical records personnel. In this study it was not possible to verify the diagnoses as one would need access to each medical record. Analogously, the Comprehensive Data Set contains several variables that could potentially provide a more detailed description of head injury severity. These variables include standard trauma measures such as the Glasgow Coma37 and Outcome38 Scales. Unfortunately, a high proportion of head trauma cases in the Comprehensive Data Set reported inappropriate or missing values for these scales (30% for the Glasgow Coma Scale at the admitting hospital, 37% for the Glasgow Outcome Scale), which obviously limits their utility as descriptors. Possible reasons for inappropriate or missing values for the Glasgow Coma Scale include: 1) the patient being intubated or under the influence of paralytic agents, which makes it impossible to administer the scale; 2) the Glasgow Coma Scale can be difficult to administer under other medical circumstances; and, 3) these scales may have less perceived clinical value for minor head injuries. The extent of missing data is of obvious importance for research, and this should be addressed as the Ontario Trauma Registry is refined and improved.

Conclusion

Blunt head trauma represents an important health issue and its epidemiology warrants further investigation. This study, while a basic form of epidemiological research, provides new data that are valuable for quantifying the magnitude of the problem, outlining leading injury patterns, and identifying specific high-risk groups. Research investigating the entire spectrum of blunt head trauma, from mild forms of injury to fatalities, would be helpful for informing injury prevention and research priorities. Methodological research aimed at enhancing the completeness and accuracy of data available through trauma registries is also warranted.

Acknowledgements

We thank Julian Martalog, Alison Locker and Nicole De Guia at the Canadian Institute for Health Information/Ontario Trauma Registry. This study was financially supported by the Ontario Neurotrauma Foundation through a research award made to Dr. Pickett. Dr. Pickett is a Career Scientist funded by the Ontario Ministry of Health and Long-Term Care.

References

  1. Ivan LP. The impact of head trauma on society. Can J Neurol Sci 1984;11:417-20.
  2. Stening WA. Understanding the epidemic of neurotrauma. Med J Aust 1984;141:5-6.
  3. Thurman DJ, Alverson C, Dunn KA, Guerrero J, Sniezek JE. Traumatic brain injury in the United States: A public health perspective. J Head Trauma Rehabil 1999;14:602-15.
  4. Krug EG, Sharma GK, Lozano R. The global burden of injuries. Am J Public Health 2000;90:523-6.
  5. Kalsbeek WD, McLaurin RL, Harris BS, Miller JD. The National Head and Spinal Cord Injury Survey: major findings. J Neurosurg 1980;53:S19-S31.
  6. Kraus JF, Black MA, Hessol N, et al. The incidence of acute brain injury and serious impairment in a defined population. Am J Epidemiol 1984;119:186-201.
  7. Parkinson D, Stephensen S, Phillips S. Head injuries: a prospective, computerized study. Can J Surg 1985;28:79-83.
  8. Guerrero JL, Thurman DJ, Sniezek JE. Emergency department visits associated with traumatic brain injury: United States, 1995-1996. Brain Inj 2000;14:181-6.
  9. Jager TE, Weiss HB, Coben JH, Pepe PE. Traumatic brain injuries evaluated in U.S. emergency departments, 1992-1994. Acad Emerg Med 2000;7:134-40.
  10. Thurman DJ, Jeppson L, Burnett CL, Beaudoin DE, Rheinberger MM, Sniezek JE. Surveillance of traumatic brain injuries in Utah. West J Med 1996;165:192-6.
  11. Tiret L, Hausherr E, Thicoipe M, et al. The epidemiology of head trauma in Aquitaine (France), 1986: a community-based study of hospital admissions and deaths. Int J Epidemiol 1990;19:133-40.
  12. Jagger J, Levine JI, Jane JA, Rimel RW. Epidemiologic features of head injury in a predominantly rural population. J Trauma 1984;24:40-4.
  13. Masson F, Thicoipe M, Aye P, et al. Epidemiology of severe brain injuries: a prospective population-based study. J Trauma 2001;51:481-9.
  14. Thurman D, Guerrero J. Trends in hospitalization associated with traumatic brain injury. JAMA 1999;282:954-7.
  15. Canadian Institute for Health Information. Head injury admissions in Ontario, 1996/97. Ottawa, ON: Canadian Institute for Health Information, 1999.
  16. Canadian Institute for Health Information. Neurotrauma hospitalizations in Ontario, 1998/99. Ottawa, ON: Canadian Institute for Health Information, 2001.
  17. Kraus JF. Injury to the head and spinal cord. The epidemiological relevance of the medical literature published from 1960 to 1978. J Neurosurg 1980;53:S3-S10.
  18. Annegers JF, Grabow JD, Kurland LT, Laws ERJ. The incidence, causes, and secular trends of head trauma in Olmsted County, Minnesota, 1935-1974. Neurology 1980;30:919.
  19. Kraus JF, McArthur DL. Epidemiologic aspects of brain injury. Neurol Clin 1996;14:435-50.
  20. Brookes M, MacMillan R, Cully S, et al. Head injuries in accident and emergency departments. How different are children from adults? J Epidemiol Community Health 1990;44:147-51.
  21. Klonoff H, Thompson GB. Epidemiology of head injuries in adults: a pilot study. Can Med Assoc J 1969;100:235-41.
  22. Engberg A. Severe traumatic brain injury - epidemiology, external causes, prevention, and rehabilitation of mental and physical sequelae. Acta Neurol Scand Suppl 1995;164:1-151.
  23. Kraus JF, Rock A, Hemyari P. Brain injuries among infants, children, adolescents, and young adults. Am J Dis Child 1990;144:684-91.
  24. Pickett W, Ardern C, Brison RJ. A population-based study of potential brain injuries requiring emergency care. Can Med Assoc J 2001;165:288-92.
  25. Snow WG, Macartney-Filgate MS, Schwartz ML, Klonoff PS, Ridgley BA. Demographic and medical characteristics of adult head injuries in a Canadian setting. Can J Surg 1988;31:191-4.
  26. Wong PP, Dornan J, Schentag CT, Ip R, Keating M. Statistical profile of traumatic brain injury: a Canadian rehabilitation population. Brain Inj 1993;7:283-94.
  27. Hendrick EB, Harwood-Hash DC, Hudson AR. Head injuries in children: a survey of 4465 consecutive cases at the hospital for sick children, Toronto, Canada. Clin Neurosurg 1964;11:46-65.
  28. Ivan LP, Choo SH, Ventureyra EC. Head injuries in childhood: a 2-year survey. Can Med Assoc J 1983;128:281-4.
  29. Klonoff H, Robinson GC. Epidemiology of head injuries in children: a pilot study. Can Med Assoc J 1967;96:1308-11.
  30. Hentschel S, Hader W, Boyd M. Head injuries in skiers and snowboarders in British Columbia. Can J Neurol Sci 2001;28:42-6.
  31. Canadian Institute for Health Information. Ontario Trauma Registry 2001 Report-Hospital Injury Admissions. Ottawa, ON: Canadian Institute for Health Information, 2001.
  32. Canadian Institute for Health Information. Ontario Trauma Registry 2000 Report- Major Injury Ontario (includes 1998/99 data). Ottawa, ON: Canadian Institute for Health Information, 2000.
  33. World Health Organization. International Classification of Diseases, 9th Revision. United States: World Health Organization, 1985.
  34. Baker SP, O'Neill B, Haddon W, Jr., Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14:187-96.
  35. Statistics Canada. Annual Demographic Statistics, 2000. Ottawa, ON: Statistics Canada, 2000. Catalogue No.: 91-213.
  36. Breslow NE, Day NE. Statistical methods in cancer research. Volume II: The design and analysis of cohort studies. Lyon, France: International Agency for Research on Cancer, 1987. IARC Scientific Publications No.: 82, p. 52-61.
  37. Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974;2:81-4.
  38. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1:480-4.

Author References

William Pickett, Kelly Simpson, Department of Emergency Medicine and Department of Community Health and Epidemiology, Queen's University, Kingston, Ontario, Canada

Robert J Brison, Department of Emergency Medicine, Queen's University, Kingston, Ontario, Canada

Correspondence: William Pickett, Kingston General Hospital, Angada 3, 76 Stuart Street, Kingston, Ontario, Canada K7L 2V7; Fax: (613) 548-1381; E-mail: pickettw@post.queensu.ca

Page details

Date modified: