Trauma is a significant healthcare burden. Mortality risk increases with increasing injury severity. An Injury Severity Score of 16 (ISS > 15) or more is considered life threatening, a severe injury.1 Age and gender are two well-recognised risk factors for traumatic injury and death. Some ethnic groups have higher injury and mortality rates.2-5Ethnicity has been seen as a risk factor in several population studies. Mortality rates are higher in some ethnic groups as a reflection of interpersonal violence. In Los Angeles County the homicide mortality rates for African Americans, (40.4/100,000), was higher than the total population mortality rate (30.9/100,000).2The mechanism of injury can be similar to that of the reference population with some causes having higher relative risk. In the Calgary Health Region, Aboriginal Canadians had higher severe injury rates than the reference population (relative risk 3.7). Motor vehicle crashes, assault and suicide had highest relative risks.7Native Americans and Australian Aboriginals had differing injury mortality profiles when compared by Stevenson et al.8 In different populations the mechanism of injury for different ethnic groups varies. Specific interventions, for specific ethnicity and mechanism, can be effective.9Auckland severe injury and mortality rates, for the year 2004, are 14.4/100,000 and 33.6/100,000.10 As both the reference and study populations had ethnicity as part of the data set an ad hoc analysis was performed.Methods The boundaries of the Auckland, Waitemata and Counties Manukau District Health Boards were used to define the study population. This population is served by four major metropolitan hospitals: North Shore Hospital, Starship Children's Hospital, Auckland City Hospital and Middlemore Hospital. Statistics New Zealand population projections (age, gender and ethnicity) for the resident population, 2004, were utilised to identify the baseline population characteristics.11 Trauma registries at the four major hospitals were utilised to identify ISS >15 patients and inpatient deaths from injury. Pre-hospital traumatic mortality was identified from coroner's autopsy records (New Zealand law requires coronial autopsy for all pre-hospital traumatic deaths). Self-harm from physical cause was included. No individual had more than one severe injury in the time frame. A residential address was available for all individuals. Prioritised ethnicity was used for both the trauma and resident populations. Age, gender, ethnicity, injury severity score, mortality and mechanism were identified. Injury mechanism group was prioritised: road traffic crash (including pedestrian), falls, hanging, assault, burns other. The inclusion criteria were: Auckland resident, death from injury, life threatening injury (ISS >15), injury occurring during the 2004 calendar year and treatment at one of the four major hospitals. Exclusion criteria were: injury occurring outside of geographic catchment and the exclusion criteria of the hospital trauma registries (examples of registry exclusions include: poisoning, drowning, and overdose as mechanisms, and fractured neck of femur or chronic subdural as injuries).12 The Auckland Ethics committee approved the study. Results In Auckland 2004, Māori comprised 11% of the population, had 27% of the injuries and 23% of the injury mortality. Pacific comprised 14% of the population, had 15% of the injuries and 15% of the injury mortality. With univariate analysis, Māori showed the highest risk of both severe injury and mortality. In multivariate analysis these risks increased after controlling for age and gender, compared to the non-Māori, non-Pacific group, Māori had a relative risk of 2.38 (95%CI 1.88-3.02) of severe injury and 2.80 (95%CI 1.96-3.99) for mortality. A similar effect was seen in respect of the Pacific group, however the relative risks were not as high, 1.49 (95% CI 1.11-1.95) for severe injury and 1.59 (95% CI 1.05-2.41) (Table 1). Table 1. Injury and mortality rates and relative risk by ethnicity Variables Rate (/100,000) Univariate risk RR (95% CI) Multivariable risk RR (95% CI) Severe injury Māori Pacific Other 61.4 38.6 28.5 2.14 (1.70-2.70) 1.35 (1.04-1.77) 1.00 2.38 (1.88-3.02) 1.49 (1.13-1.95) 1.00 Mortality Māori Pacific Other 28.4 16.4 11.9 2.37 (1.68-3.35) 1.37 (0.91-2.07) 1.00 2.80 (1.96-3.99) 1.59 (1.05-2.41) 1.00 CI=confidence interval. When relative risks for ethnicity by age group were analysed, the relative risks varied notably by age group. In the older age groups the numbers of Pacific and Māori were small and statistical analysis was unable to be carried out. For Māori there were statistically significantly increased risks associated with severe injury for all age groups below 45 and mortality for the 15-29 and 30-44 age groups. The point estimates of the other groups were increased however did not reach statistical significance. For severe injury an increased risk was seen in the 15-29 year age group, those for the 0-14 and 30-44 year groups were increased but did not quite reach significance at the 5% level. The only age group showing an increased risk of mortality amongst the Pacific group was those 15-29 years of age (Table 2). Table 2. Relative risk, injury and mortality for Māori and Pacific in 15-year age groups Age group Māori injury Pacific injury Māori mortality Pacific mortality 0-14 2.66 (1.89-3.74) 1.48 (0.98-2.23) 2.58 (0.69-9.62) 1.23 (0.24-6.35) 15-29 2.87 (1.46-5.63) 2.57 (1.29-5.09) 3.03 (1.88-4.91) 1.86 (1.07-3.25) 30-44 2.83 (1.71-4.69) 1.73 (0.96-3.11) 3.84 (1.78-8.25) 1.40 (0.48-4.12) 45-59 1.71 (0.85-3.45) 0.82 (0.33-2.05) 2.00 (0.69-5.81) 1.30 (0.39-4.35) 60-74 Numbers too small Numbers too small 75+ Numbers too small Numbers too small The risk of mortality was increased in both Māori and Pacific Island males compared to those of other ethnicities, whilst the risk of injury was only significantly increased amongst Māori male. For females, Māori were at increased risk of both mortality and injury, whilst there was no significant difference in risk compared to those of non-Māori/non-Pacific ethnicity for Pacific women, in fact the point estimated for mortality was in a protective direction (Table 3). Table 3. Relative risk, injury and mortality for Māori and Pacific by gender Gender Māori injury Pacific injury Māori mortality Pacific mortality Male 1.97 (1.37-2.83) 1.36 (0.91-2.03) 2.18 (1.43-3.32) 1.76 (1.14-2.73) Female 2.03 (1.09-3.78) 1.29 (0.63-2.66) 2.90 (1.58-5.35) 0.36 (0.09-1.49) Mortality rates are highest for Māori male, and lowest for Pacific female (Figure 1). Severe injury rates are highest for 15-29 year old Māori male (Figure 2). The lowest rates are seen in elderly Māori and Pacific. The denominator in these groups is small. The Pacific female graph (Figure 3) is different from all other groups, the peak injury incidence is in childhood, when it is higher than pacific male for the same age range. Figure 1. Mortality rates for gender and ethnicity Figure 2. Male injury rates, age groups and ethnicity Figure 3. Female injury rates: age groups and ethnicity The most common cause for injury, accounting for 50% of the injuries across all ethnic groups, was road traffic crashes including pedestrian injuries (Table 4). This was followed by falls (18%), hanging (15%) and assault (11%). There was, however, a significant difference of injury between ethnicities (p.=0.0013). Hanging (25%) and assault (18%) were the second and third most common causes for Māori; falls (23%) and assault (14%) were second and third for Pacific. (The non-Māori, non-Pacific group followed the overall pattern.) Table 4. Number (percent) of injury type by ethnicity Injury type Māori Pacific Other Road traffic crashes Falls Hanging Assault Burns Other 43 (45.3) 6 (6.3) 24 (25.3) 17 (17.9) 2 (2.1) 3 (17.7) 29 (43.9) 15 (22.7) 8 (12.1) 9 (13.6) 3. (4.6) 2 (3.0) 150 (52.3) 62 (21.6) 35 (12.2) 22 (7.7) 6 (2.1) 12 (4.2) Discussion This paper shows statistically significant differences for Māori and Pacific injury and mortality rates from the remaining population. Although the mechanisms of injury are the same, Māori and Pacific have relatively more assaults and hangings. Pacific women have lower injury rates than all other ethnic and gender groups. The basis for this is not known and is beyond the scope of this paper. Using severe injury, ISS >15, utilises an accepted mechanism for grading injury, and defines the injuries as clearly life threatening.1 Prioritised ethnicity was used in each group. In a similar Canadian Study, being of aboriginal Indian descent was associated with increased risk of injury.7 The higher injury rates noted in Canada may be, in part, due to a lower injury severity score (greater than 12) as a definition of major injury, and partly due to databases characteristics. In Los Angeles county the variation in mortality rates between ethnicities was related almost entirely to interpersonal violence.2 In our study, motor vehicle-related injury was the most common cause of injury and death in all ethnic groups. Bias may be introduced into this study in many ways. This is an ad hoc analysis. Prioritised ethnicity, although consistent between the data groups, may be a methodological issue.14Finally, the data used for the population are projections. A single year of data allows for limited interpretation. In many groups, especially elderly Māori and Pacific, the sub-populations are small. A single injury in the 75+ Māori male group would give an incidence of more than 280/100,000. However, the data collated in this study is the first time that such information has been drawn together in New Zealand, and is the first time that paediatric data is included in such a study. Further research in this area might include more years of data, and be the objective of the research, rather than an ad hoc analysis. In summary, this paper demonstrates the characteristics for severe injury and mortality in the Auckland region for Māori and Pacific people for the 2004 calendar year. Pacific female had the lowest injury rate, although this only reaches significance when children are excluded. Further study is needed to verify or refute this. Māori and Pacific generally have higher rates of major injury (including injury causing death) than the rest of the population. For Māori and Pacific men, ethnicity appears to be a significant risk factor for severe injury and mortality.

Trauma is a significant healthcare burden. Mortality risk increases with increasing injury severity. An Injury Severity Score of 16 (ISS > 15) or more is considered life threatening, a severe injury.1 Age and gender are two well-recognised risk factors for traumatic injury and death. Some ethnic groups have higher injury and mortality rates.2-5Ethnicity has been seen as a risk factor in several population studies. Mortality rates are higher in some ethnic groups as a reflection of interpersonal violence. In Los Angeles County the homicide mortality rates for African Americans, (40.4/100,000), was higher than the total population mortality rate (30.9/100,000).2The mechanism of injury can be similar to that of the reference population with some causes having higher relative risk. In the Calgary Health Region, Aboriginal Canadians had higher severe injury rates than the reference population (relative risk 3.7). Motor vehicle crashes, assault and suicide had highest relative risks.7Native Americans and Australian Aboriginals had differing injury mortality profiles when compared by Stevenson et al.8 In different populations the mechanism of injury for different ethnic groups varies. Specific interventions, for specific ethnicity and mechanism, can be effective.9Auckland severe injury and mortality rates, for the year 2004, are 14.4/100,000 and 33.6/100,000.10 As both the reference and study populations had ethnicity as part of the data set an ad hoc analysis was performed.Methods The boundaries of the Auckland, Waitemata and Counties Manukau District Health Boards were used to define the study population. This population is served by four major metropolitan hospitals: North Shore Hospital, Starship Children's Hospital, Auckland City Hospital and Middlemore Hospital. Statistics New Zealand population projections (age, gender and ethnicity) for the resident population, 2004, were utilised to identify the baseline population characteristics.11 Trauma registries at the four major hospitals were utilised to identify ISS >15 patients and inpatient deaths from injury. Pre-hospital traumatic mortality was identified from coroner's autopsy records (New Zealand law requires coronial autopsy for all pre-hospital traumatic deaths). Self-harm from physical cause was included. No individual had more than one severe injury in the time frame. A residential address was available for all individuals. Prioritised ethnicity was used for both the trauma and resident populations. Age, gender, ethnicity, injury severity score, mortality and mechanism were identified. Injury mechanism group was prioritised: road traffic crash (including pedestrian), falls, hanging, assault, burns other. The inclusion criteria were: Auckland resident, death from injury, life threatening injury (ISS >15), injury occurring during the 2004 calendar year and treatment at one of the four major hospitals. Exclusion criteria were: injury occurring outside of geographic catchment and the exclusion criteria of the hospital trauma registries (examples of registry exclusions include: poisoning, drowning, and overdose as mechanisms, and fractured neck of femur or chronic subdural as injuries).12 The Auckland Ethics committee approved the study. Results In Auckland 2004, Māori comprised 11% of the population, had 27% of the injuries and 23% of the injury mortality. Pacific comprised 14% of the population, had 15% of the injuries and 15% of the injury mortality. With univariate analysis, Māori showed the highest risk of both severe injury and mortality. In multivariate analysis these risks increased after controlling for age and gender, compared to the non-Māori, non-Pacific group, Māori had a relative risk of 2.38 (95%CI 1.88-3.02) of severe injury and 2.80 (95%CI 1.96-3.99) for mortality. A similar effect was seen in respect of the Pacific group, however the relative risks were not as high, 1.49 (95% CI 1.11-1.95) for severe injury and 1.59 (95% CI 1.05-2.41) (Table 1). Table 1. Injury and mortality rates and relative risk by ethnicity Variables Rate (/100,000) Univariate risk RR (95% CI) Multivariable risk RR (95% CI) Severe injury Māori Pacific Other 61.4 38.6 28.5 2.14 (1.70-2.70) 1.35 (1.04-1.77) 1.00 2.38 (1.88-3.02) 1.49 (1.13-1.95) 1.00 Mortality Māori Pacific Other 28.4 16.4 11.9 2.37 (1.68-3.35) 1.37 (0.91-2.07) 1.00 2.80 (1.96-3.99) 1.59 (1.05-2.41) 1.00 CI=confidence interval. When relative risks for ethnicity by age group were analysed, the relative risks varied notably by age group. In the older age groups the numbers of Pacific and Māori were small and statistical analysis was unable to be carried out. For Māori there were statistically significantly increased risks associated with severe injury for all age groups below 45 and mortality for the 15-29 and 30-44 age groups. The point estimates of the other groups were increased however did not reach statistical significance. For severe injury an increased risk was seen in the 15-29 year age group, those for the 0-14 and 30-44 year groups were increased but did not quite reach significance at the 5% level. The only age group showing an increased risk of mortality amongst the Pacific group was those 15-29 years of age (Table 2). Table 2. Relative risk, injury and mortality for Māori and Pacific in 15-year age groups Age group Māori injury Pacific injury Māori mortality Pacific mortality 0-14 2.66 (1.89-3.74) 1.48 (0.98-2.23) 2.58 (0.69-9.62) 1.23 (0.24-6.35) 15-29 2.87 (1.46-5.63) 2.57 (1.29-5.09) 3.03 (1.88-4.91) 1.86 (1.07-3.25) 30-44 2.83 (1.71-4.69) 1.73 (0.96-3.11) 3.84 (1.78-8.25) 1.40 (0.48-4.12) 45-59 1.71 (0.85-3.45) 0.82 (0.33-2.05) 2.00 (0.69-5.81) 1.30 (0.39-4.35) 60-74 Numbers too small Numbers too small 75+ Numbers too small Numbers too small The risk of mortality was increased in both Māori and Pacific Island males compared to those of other ethnicities, whilst the risk of injury was only significantly increased amongst Māori male. For females, Māori were at increased risk of both mortality and injury, whilst there was no significant difference in risk compared to those of non-Māori/non-Pacific ethnicity for Pacific women, in fact the point estimated for mortality was in a protective direction (Table 3). Table 3. Relative risk, injury and mortality for Māori and Pacific by gender Gender Māori injury Pacific injury Māori mortality Pacific mortality Male 1.97 (1.37-2.83) 1.36 (0.91-2.03) 2.18 (1.43-3.32) 1.76 (1.14-2.73) Female 2.03 (1.09-3.78) 1.29 (0.63-2.66) 2.90 (1.58-5.35) 0.36 (0.09-1.49) Mortality rates are highest for Māori male, and lowest for Pacific female (Figure 1). Severe injury rates are highest for 15-29 year old Māori male (Figure 2). The lowest rates are seen in elderly Māori and Pacific. The denominator in these groups is small. The Pacific female graph (Figure 3) is different from all other groups, the peak injury incidence is in childhood, when it is higher than pacific male for the same age range. Figure 1. Mortality rates for gender and ethnicity Figure 2. Male injury rates, age groups and ethnicity Figure 3. Female injury rates: age groups and ethnicity The most common cause for injury, accounting for 50% of the injuries across all ethnic groups, was road traffic crashes including pedestrian injuries (Table 4). This was followed by falls (18%), hanging (15%) and assault (11%). There was, however, a significant difference of injury between ethnicities (p.=0.0013). Hanging (25%) and assault (18%) were the second and third most common causes for Māori; falls (23%) and assault (14%) were second and third for Pacific. (The non-Māori, non-Pacific group followed the overall pattern.) Table 4. Number (percent) of injury type by ethnicity Injury type Māori Pacific Other Road traffic crashes Falls Hanging Assault Burns Other 43 (45.3) 6 (6.3) 24 (25.3) 17 (17.9) 2 (2.1) 3 (17.7) 29 (43.9) 15 (22.7) 8 (12.1) 9 (13.6) 3. (4.6) 2 (3.0) 150 (52.3) 62 (21.6) 35 (12.2) 22 (7.7) 6 (2.1) 12 (4.2) Discussion This paper shows statistically significant differences for Māori and Pacific injury and mortality rates from the remaining population. Although the mechanisms of injury are the same, Māori and Pacific have relatively more assaults and hangings. Pacific women have lower injury rates than all other ethnic and gender groups. The basis for this is not known and is beyond the scope of this paper. Using severe injury, ISS >15, utilises an accepted mechanism for grading injury, and defines the injuries as clearly life threatening.1 Prioritised ethnicity was used in each group. In a similar Canadian Study, being of aboriginal Indian descent was associated with increased risk of injury.7 The higher injury rates noted in Canada may be, in part, due to a lower injury severity score (greater than 12) as a definition of major injury, and partly due to databases characteristics. In Los Angeles county the variation in mortality rates between ethnicities was related almost entirely to interpersonal violence.2 In our study, motor vehicle-related injury was the most common cause of injury and death in all ethnic groups. Bias may be introduced into this study in many ways. This is an ad hoc analysis. Prioritised ethnicity, although consistent between the data groups, may be a methodological issue.14Finally, the data used for the population are projections. A single year of data allows for limited interpretation. In many groups, especially elderly Māori and Pacific, the sub-populations are small. A single injury in the 75+ Māori male group would give an incidence of more than 280/100,000. However, the data collated in this study is the first time that such information has been drawn together in New Zealand, and is the first time that paediatric data is included in such a study. Further research in this area might include more years of data, and be the objective of the research, rather than an ad hoc analysis. In summary, this paper demonstrates the characteristics for severe injury and mortality in the Auckland region for Māori and Pacific people for the 2004 calendar year. Pacific female had the lowest injury rate, although this only reaches significance when children are excluded. Further study is needed to verify or refute this. Māori and Pacific generally have higher rates of major injury (including injury causing death) than the rest of the population. For Māori and Pacific men, ethnicity appears to be a significant risk factor for severe injury and mortality.