Physical fighting is a manifestation of interpersonal violence and has become a concerning public health issue worldwide among young people. (1,2) In studies from the United States, nearly one third of adolescents reported participating in at least one physical fight in the previous year. (3-5)

Physical fighting puts adolescents at increased risk for injury. (6-8) Injury is one of the most common reasons for death, hospitalization and disability in young people. (9-11) Fighting is the fifth most common reason for injury in Canadian youth, and the third most common reason for injuries requiring significant medical treatment or overnight hospital admission. Only vehicle- and bicycle-related injuries are more common among young people for medically treated injuries. (9) Due to the potential for harm, it is important for researchers to understand who is at risk for fighting and fighting-related injury, and what reasons or mechanisms underpin these outcomes.

There is limited epidemiological information in a Canadian context that is recent and nationally representative. In Canada there is also minimal information on demographic and socio-economic variation of physical fight involvement and fighting-related injuries. Few studies consider the contextual factors of fighting and fighting-related injuries, such as where the event occurred and who was involved. There is also little information regarding trends that indicate whether physical fights and fighting-related injuries have changed over time. International studies report that physical fighting has decreased over time, although it is difficult to generalize these findings to Canada. (2)

The objectives of this study were to: 1) describe physical fighting and fighting-related injury among Canadian adolescents by sex, grade and subjective social status; and 2) investigate potential trends of physical fighting and fighting-related injuries between 1993-2010 in the overall population and within subgroups.

METHODS

Data sources and sample

This study used Canadian data from the Health Behaviour in School-aged Children (HBSC) study. (6) HBSC is a nationally representative study that was developed in collaboration with the World Health Organization for the purpose of understanding health determinants and behaviours in young people. Canada has collected data every four years since 1998 through questionnaires administered to 11-15 year old students. Six HBSC cycles have been undertaken in Canada during the period 1990-2010. Cycles 2-6 (1993-2010) were used since data for the outcomes of interest were not available for the first cycle. Table 1 includes information about the response rates and sampling strategy for each survey cycle.

The study included a total sample of 61,465 students in grades 6-10. Analysis was undertaken independently for each cycle due to differences in sampling procedures and slight variability in wording and coded responses for each survey item.

Study variables

Physical fighting data were available for cycles 4-6. Participants were asked how many times they were in a physical fight in the previous 12 months. Students who completed this question were categorized as follows: no physical fights, 1 time, 2-3 times, and 4 or more times. A dichotomous variable ("No physical fights" and "One or more physical fights") was created for cross-tabulations.

Fighting-related injury was the second main outcome. Data were available in cycles 2-6. Participants were asked whether they had been injured in the previous 12 months, and the cause of their most serious injury. Responses that said "Yes" to being injured with "Fighting" chosen as the cause were coded as a fighting-related injury.

Data for sex, grade and subjective social status were available in all datasets and used as descriptor subgroups. Participants were asked early in the questionnaire to indicate their sex and grade (response groups: grade <6, grade 7-8 and grade >9). Grade categories correspond roughly with students in primary, middle and secondary schools. Subjective social status was measured by the question "How well off do you think your family is?" with five possible responses, which were then grouped into three categories: high social status ("Very well off", "Well off"), average ("Average"), and low social status ("Not well off", "Not at all well off").

Scenario variables

Variables describing the context of the injury and physical fight encounter included: who the fight was with; whether the injury caused missed school or activities; the type of injury; whether medical treatment was required; the location where the injury occurred; whether the injury happened during an activity or club; the season during which the injury happened; and where treatment was received.

Time trends of physical fighting and fighting-related injury were assessed in terms of differences in overall prevalence and prevalence within subgroups for each one-year time point.

Statistical analysis and survey weights

SAS 9.3 (SAS Inc., Cary, NC) was used for all analyses. Each HBSC cycle had various sampling procedures. In accordance with international protocols, cycles 2-5 were designed to be self-weighted and considered characteristics of the Canadian population for the sampling process. (12,13) Cycle 6 includes survey weights by province and territory within grade groups. Over-represented provinces and territories were given weights of <1, and under-represented groups were given weights of >1. Survey weights ranged from 0.017 to 3.655. (6,11) Prevalence estimates were derived from specific time cycles, and cross-tabulations were performed to obtain frequencies, percentages and prevalence ratios with 95% confidence intervals). The Cochrane-Armitage test for trend was used to establish the significance of increasing or decreasing prevalence trends over time. Ethical approval for this analysis was obtained from the Queen's University Health Sciences Research Ethics Board (File # 6011541).

RESULTS

Distribution of the participants by demographic characteristics and outcomes can be found in Table 2.

Prevalence of physical fighting

The prevalence of physical fighting one or more times ranged from 35.1%-41.2%, and peaked at 41.2% in 2005/2006 (Figure 1). The prevalence from 2001-2010 displays a significant change over time despite the prevalence decrease in 2009/2010 ([p.sub.trend] = 0.015). A significant difference in prevalence from 2001-2010 was also observed in specific groups, including females ([p.sub.trend] < 0.001), grade 7-8 students ([p.sub.trend] = 0.008) and high subjective social status individuals ([p.sub.trend] = 0.002). We do recognize however that further time points are necessary to determine the extent and direction of this trend with certainty.

In demographic subgroups, the prevalence of fighting was higher in males than in females (48.3%-53.4% vs. 23.8%-30.3% respectively) (Table 3). Males were also twice as likely to report a physical fight for each time cycle (p < 0.001).

As compared to students in grades [greater than or equal to] 9, students in grades <6 had an approximately 20%-30% higher likelihood of reporting a physical fight in the previous year, while grade 7-8 students had 10%-20% increased risk (Table 4). This decrease of physical fight involvement from lower to higher grades was significant within each cycle ([p.sub.trend] < 0.001).

Compared to students from subjectively high social status families, participants who reported low social status were 20%-30% more likely to report a physical fight in the previous 12 months, while those of self-reported average social status had an approximately 10%-15% higher likelihood. Table 4 shows a significant decline in risk from high to low social status groups for all cycles ([p.sub.trend] < 0.001).

Table 5 reports with whom each individual engaged in a fight. Approximately 43% of those who participated in at least one fight reported fighting with a friend or someone they knew, followed by: brother or sister, total stranger, parent or adult family member, and boyfriend, girlfriend or date.

Prevalence of fighting-related injury

Overall fighting-related injury ranged from 1.1%-2.3% (Figure 2). The trend test indicated a significant increase over time in the overall prevalence of fighting-related injury and within all subgroups ([p.sub.trend] < 0.001).

In Table 4, males had 1.5-2.5 fold increased likelihood of reporting a fighting-related injury compared to females (p < 0.05). Individuals in grades <6 had 30%-60% reduced likelihood of reporting a fighting-related injury compared to those in higher grades (Table 4). This association was only significant for 1993/1994, 2001/2002 and 2009/2010. Grade 7-8 participants for most cycles except 2005/2006 had a 24%-40% lower likelihood of reporting a fighting-related injury compared to participants in grades >9. There is an increasing likelihood for fighting-related injury with higher grades in 1993/1994, 2001/2002, and 2009/2010 ([p.sub.trend] < 0.05).

Table 4 shows that for cycles 4-6, those with low subjective social status were 2-3 times significantly more likely to report a fighting-related injury compared to those with high social status. For all cycles except the last, there was a 13%-60% insignificant higher likelihood of fighting-related injury for individuals of average social status compared to those with higher social status. Lower subjective social status was associated with an increased likelihood of reporting a fighting-related injury ([p.sub.trend] < 0.001).

Context of fighting-related injury

Table 6 shows participants who reported a fighting-related injury in the previous 12 months according to the circumstances of the injury. Some questions were not available for all cycles.

For all available cycles, over 50% of those who reported a fighting-related injury missed at least one day of school or activities due to injury.

The three most common results for fighting-related injury were: broken bones, cut or puncture wounds, and bruises or internal bleeding. Broken bones were most common in 1993/1994, and bruises/internal bleeding in 1997/1998 and 2001/2002.

Information on whether the injury needed medical treatment was available in 3 cycles. For 1993/1994, 57.6% of fighting-related injuries required medical treatment. This decreased to 50.2% in 2005/2006 and 46.8% in 2009/2010.

The earliest three cycles stated that most fighting-related injuries occurred in the home or yard, school, and street or parking lot. There is a shift from 2005/2006 onwards where the most common place an injury occurred was the street followed by the school and home.

Determining whether the injury happened during an organized activity or club was possible for the middle three cycles. For all available cycles, 10.1%-26.8% of fighting-related injuries occurred during an organized activity.

Fighting-related injuries most commonly occurred in autumn and second most commonly in summer for the first two cycles, then in spring for 1993/1994 (16%) and in winter for 1997/1998 (14.9%). 2001/2002 presented a different order, where fights most commonly occurred in spring (43.1%), followed by winter (25.7%), autumn (16.5%) and summer (14.7%).

Data pertaining to where the patient was treated were available for 2001/2002 and 2005/2006. Patients were most commonly treated for their injury in emergency rooms, doctors' offices or health clinics.

DISCUSSION

This study shows that physical fighting in children is a relatively common behaviour within Canada (35%-40%), and that obtaining an injury related to fighting (l%-2%) is more rare. Fighting and fighting-related injury prevalence differed significantly over time overall and within specific subgroups. Males are significantly more likely than females to participate in a physical fight and obtain a fighting-related injury. Students in grades six or below were more likely to participate in fights than their older counterparts, while being in a lower grade was protective against fighting-related injury. Individuals from families with subjective low social status were increasingly more likely to obtain both outcomes compared to participants who reported higher subjective social status.

It is possible that because youth are becoming more exposed to violent media, they are also increasingly desensitized to violence and less inhibited when engaging in aggressive activities. (14) Adolescents who are more prone to violent behaviours may have poorer problem-solving skills and coping strategies for solving confrontations, which can result in physical altercations. (15) This is a plausible explanation for the significantly different prevalence estimates over time. It is difficult to determine whether the prevalence of fighting is increasing or decreasing over time due to the unusually high prevalence in 2005/2006. Therefore, more time points are necessary in future research to confidently examine the time patterns of physical fighting. Further research is also needed to explain the biological and sociological mechanisms for these increases. Fighting prevalence changed significantly over time within specific subgroups (females, middle school students, and higher social status students). Since there were slight variations in sampling strategy and the 2009/2010 cycle used weighted data, the trend test results shall be interpreted with caution. Further research is needed to understand why physical fighting over time has altered in these groups.

While physical fighting has altered significantly over time within females, descriptive analyses showed that males were significantly more likely than females to participate in physical fights and obtain a fighting-related injury for all time points. This gendered difference was also reported in previous international studies. (5,6,16) Scientists have postulated biological reasons for the higher prevalence of violent behaviours in males, including the increases of testosterone associated with puberty and increased strength from muscle mass. (17,18) Individuals who reported their families as of lower social status also had an increased likelihood of obtaining both outcomes. This relationship with physical fighting is consistent with previous studies, although we recognize the various ways to define and measure socio-economic status and affluence. (1) However, for the relationship between subjective social status and fighting-related injuries, further analysis accounting for multiple affluence measurements and covariates needs to be done. (19)

Individuals in grades <6 were more likely to participate in fights than their older counterparts. In contrast, however, being in a lower grade was protective against fighting-related injury when compared to higher school levels. These patterns may be because younger adolescents are still developing mentally and physically, and may use physical confrontations to resolve issues. "Play fighting" or "rough-and-tumble play" is a common behaviour among younger people and these behaviours decrease as students approach adulthood. (20) Concurrently, adolescents also undergo physiological changes such as increased muscle mass and hormonal changes during puberty. The elevated strength in older adolescents may intensify the physical force exerted in fights, which can increase the likelihood of injury. (17,21) The way grade school and high school students interpret situations may vary, where the circumstances and consequences behind physical confrontations may become more severe as adolescents get older; as a result, there may be more intent to harm. (22)

Findings associated with the context of fighting and fighting-related injuries showed that the person participants most frequently fought with was "a friend or someone I know" or a sibling. This is logical since adolescents habitually interact with these people. The next most popular response was "other". There is no information that describes the remaining possible options for this category. It is postulated that the "other" category can represent rivals that the students know sufficiently well to warrant their not being a stranger, but not enough to be "friends". Future surveys should present the option to describe who the "other" person is.

From 1993-2002, the most common places that a fighting-related injury occurred were in a home or yard and at school. Adolescents spend the majority of their time at home and school, which makes the increased likelihood of obtaining a fighting-related injury more plausible. Information was only available for one cycle (1993/1994) with regard to whether the injury occurred during or outside of school hours. This data would be useful for assessing the context of fighting-related injuries, particularly in relation to differentiating physical bullying or domestic violence. From 2005 onwards, the most common response for where the injury took place shifted from the home to the street. It is unknown why there is a shift in location over time.

Earlier cycles (1993-1998) suggest that fighting-related injuries most commonly occur during the fall, followed by summer. This could be because adolescents return to school and are more likely to go outside during these seasons, which allows for more physical interactions. In 2001/2002, fighting-related injuries were most common during the spring, another time when young people engage in more outdoor activities. It is unclear why there is this seasonal shift for this cycle only.

Data suggest that many of the reported injuries are serious enough to require medical attention. The fact that most of these injuries tend to happen outside of activities or clubs suggests that these outcomes may be associated with levels of supervision. A potential solution could be increased student involvement in supervised activities.

Strengths and limitations

Strengths of this study include that this research uses nationally representative data. This supports its generalizability to the Canadian adolescent population. The large sample size in each cycle also provides substantial power to detect significant differences between subgroups. The results from this study can help identify vulnerable groups and contextual factors that may pose risk, and establish priority areas for preventing physical fights and additional injury risk.

This study also contains important limitations. First, the self-reported nature of the data may present potential misclassification for both the exposures and outcomes due to the subjectivity and interpretation of each variable. For example, asking how "well off" somebody is may be a measure of subjective social status, as we have suggested, however it could also be a measure of family income or material wealth. As there is no availability of current validation studies with Canadian students for this question, the exact interpretation is not known. In addition, fighting-related injury can be misclassified as sports-related if an injury occurred because of a fight during a sport or martial arts. This indicates that not all fighting-related injuries are being captured.

Students who were absent on the day of the survey due to injury or suspension for engaging in a fight may not have been selected to participate. Furthermore, the study does not consider adolescents who dropped out of school. These individuals may be fundamentally different in how well off they are and their experiences with violence compared to students still enrolled in school. These can indicate selection bias. (2)

There were also inconsistencies in the collection of data for some variables across HBSC cycles, particularly with respect to the context of the injury, injury severity and care received. This limited our ability to assess some variables for potential confounding or effect modification. Furthermore, the limited number of data points for the trend analysis makes it difficult to assess whether the pattern was linear or exponential. More data points would be informative for trend interpretation.

A last limitation is that the survey only asks participants about the circumstances of their most serious injury. If fighting was not the cause of their most serious injury, data will be truncated. Other less serious fighting injuries may be masked in some circumstances and therefore under-reported in the sample.

CONCLUSION

Physical fighting among young Canadians has become a public health issue of greater concern in the past two decades. Specific subgroups have a higher propensity for physical fight participation and sustaining an injury from these encounters. Understanding the context of these conflicts and injuries can be informative for reducing future occurrences.

REFERENCES

(1.) Pickett W, Molcho M, Elgar FJ, Brooks F, de Looze M, Rathmann K, et al. Trends and socioeconomic correlates of adolescent physical fighting in 30 countries. Pediatrics 2013;131(1):e18-26. PMID: 23209107. doi: 10.1542/peds.2012-1614.

(2.) Krug EG, Dahlberg LL, Mercy JA, Zwi AB, Lozano R. World Report on Violence and Health. Geneva, Switzerland: World Health Organization, 2002.

(3.) Smith-Khuri E, Iachan R, Scheidt PC, Overpeck MD, Gabhainn SN, Pickett W, et al. A cross-national study of violence-related behaviors in adolescents. Arch Pediatr Adolesc Med 2004;158(6):539-44. PMID: 15184216. doi: 10.1001/archpedi.158.6.539.

(4.) Eaton DK, Kann L, Kinchen S, Shanklin S, Flint KH, Hawkins J, et al. Youth risk behavior surveillance - United States, 2011. MMWR Surveill Summ 2012; 61(4):1-162. PMID: 22673000.

(5.) Child Trends Data Bank. Physical Fighting by Youth: Indicators on Children and Youth. Child Trends Data Bank, 2013. Available at: http://www.childtrends.org/?indicators=physical-fighting-by-youth (Accessed March 20, 2014).

(6.) Freeman JG, King M, Pickett W, Craig W, Elgar F, Janssen I, et al. The Health of Canada's Young People: A Mental Health Focus. Ottawa, ON: Public Health Agency of Canada, 2011.

(7.) Borowsky IW, Ireland M. Predictors of future fight-related injury among adolescents. Pediatrics 2004;113(3):530-36. PMID: 14993545. doi: 10.1542/peds.113.3.530.

(8.) Buckley L, Chapman R, Sheehan M. Adolescent involvement in anti-social and delinquent behaviours: Predicting future injury risk. Accid Anal Prev 2012; 48:518-22. PMID: 22664718. doi: 10.1016/j.aap.2012.02.010.

(9.) Davison C, Russell K, Piedt S, Pike I, Pickett W. CIHR Team in Child and Youth Injury Prevention. Injury Among Young Canadians: A National Study of Contextual Determinants. Vancouver, BC: CIHR Team in Child and Youth Injury Prevention, 2013.

(10.) Peden M, Oyegbite K, Ozanne-Smith J, Hyder AA, Branche C, Fazlur Rahman AKM, et al. World Report on Child Injury Prevention. Geneva, Switzerland: WHO, 2008.

(11.) Freeman J, Coe H, King M. Health Behaviour in School-aged Children: Trends Report 1990-2010. Ottawa, ON: PHAC, 2013.

(12.) Roberts C, Freeman J, Samdal O, Schnohr C, Looze M, Nic Gabhainn S, et al. The Health Behaviour in School-aged Children (HBSC) study: Methodological developments and current tensions. Int J Public Health 2009;54 (Suppl 2):140-50. doi: 10.1007/s00038-009-5405-9.

(13.) Boyce W. Young People in Canada: Their Health and Well-being. Ottawa, ON: Health Canada, 2004.

(14.) Browne KD, Hamilton-Giachritsis C. The influence of violent media on children and adolescents: A public-health approach. Lancet 2005; 365(9460):702-10. PMID: 15721477. doi: 10.1016/S0140-6736(05)70938-7.

(15.) Botvin GJ, Griffin KW, Nichols TD. Preventing youth violence and delinquency through a universal school-based prevention approach. Prev Sci 2006;7(4):403-8. PMID: 17136462. doi: 10.1007/s11121-006-0057-y.

(16.) Fein JA, Mollen CJ, Greene MB. The assault-injured youth and the emergency medical system: What can we do? Clin Pediatr Emerg Med 2013;14(1):47-55. doi: 10.1016/j.cpem.2013.01.004.

(17.) Archer J. Sex differences in aggression in real-world settings: A meta-analytic review. Rev Gen Psychol 2004;8(4):291-322. doi: 10.1037/1089-2680.8.4.291.

(18.) Galen BR, Underwood MK. A developmental investigation of social aggression among children. Dev Psychol 1997;33(4):589-600. PMID: 9232374. doi: 10.1037/0012-1649.33.4.589.

(19.) Simpson K, Janssen I, Craig WM, Pickett W. Multilevel analysis of associations between socioeconomic status and injury among Canadian adolescents. J Epidemiol Community Health 2005;59(12):1072-77. PMID: 16286497. doi: 10.1136/jech.2005.036723.

(20.) Smith PK, Hunter T, Carvalho AM, Costabile A. Children's perceptions of playfighting, playchasing and real fighting: A cross-national interview study. Soc Dev 1992;1(3):211-29. doi: 10.1111/j.1467-9507.1992.tb00125.x.

(21.) Munoz-Reyes JA, Gil-Burmann C, Fink B, Turiegano E. Physical strength, fighting ability, and aggressiveness in adolescents. Am J Hum Biol 2012; 24(5):611-17. doi: 10.1002/ajhb.v24.5.

(22.) Cairns RB, Cairns BD, Neckerman HJ, Ferguson LL, Gariepy J-L. Growth and aggression: I. Childhood to early adolescence. Dev Psychol 1989;25(2):320-30. doi: 10.1037/0012-1649.25.2.320.

Received: August 25, 2015

Accepted: March 25, 2016

Maya Djerboua, MSc, [1] Bingshu E. Chen, PhD, [1] Colleen Davison, PhD [1-3]

Author Affiliations

[1.] Department of Public Health Sciences, Queen's University, Kingston, ON

[2.] Clinical Research Centre, Kingston General Hospital, Kingston, ON

[3.] Department of Emergency Medicine, Queen's University, Kingston, ON Correspondence: Colleen Davison, PhD, Department of Public Health Sciences, Queens University, 62 Fifth Field Company Lane, Kingston, ON K7L 3N6, Tel: 613533-6000, ext. 79518, E-mail: davisonc@queensu.ca

Acknowledgements: This study was funded by research grants from the Public Health Agency of Canada (PHAC) (contract 4500307663) and the Canadian Institutes of Health Research Team in Child and Youth Injury Prevention. MD was supported by the Empire Life Child Health Research Fellowship and the Queen's Graduate Award. The Canadian version of the Health Behaviour in School-aged Children (HBSC) study is a part of the international collaborative study developed in collaboration with the World Health Organization. The Canadian HBSC Study was funded by PHAC and Health Canada.

Conflict of Interest: None to declare.

Table 1. Dataset details for the Health Behaviour in
School-aged Children study cycles from 1993-2010

                     Student               School-level
HBSC        Sample   response              response
cycle       size     rate                  rate

1993-1994   7020     Not available         Not available

1997-1998   11,415   Not available         Not available

2001-2002   7235     7235/9780 = 74%       172/231 = 74.5%

2005-2006   9717     9707/13,176 = 73.7%   186/248 = 75%

2009-2010   26,078   26,078/33,868 = 77%   436/765 = 57%

            Province/
HBSC        territory       Description of sampling approach
cycle       response rate

1993-1994   Not available   Systematic cluster sampling, with
                            replacement. All non-private school
                            jurisdictions with students in
                            grades 6, 8 and 10 in a Canadian
                            province or territory were
                            identified. These grades were
                            chosen to correspond to the
                            international study age group
                            targets of 11, 13 and 15 year-olds.
                            A representative sample of schools
                            with students at these levels was
                            generated considering student
                            grade, school enrolment size,
                            community size, location in
                            province or territory, language of
                            instruction and public/Catholic
                            jurisdiction. School administrators
                            of the chosen schools were asked to
                            randomly choose one whole class in
                            each of the targeted grades (if
                            they had them) to participate.
                            Under the international protocol
                            prescribing no weighting of data
                            sets, to be nationally
                            representative, the sample was
                            designed to approximately
                            proportionally represent number of
                            students by grade in each of the
                            provinces and territories.

1997-1998   12/12 = 100%    Same as 1993, however a
                            grade-by-grade sample of grades
                            6-10 was used instead of limiting
                            the dataset to only grades 6, 8 and
                            10.

2001-2002   10/13 = 76.9%   Same as 1997.

2005-2006   12/13 = 92.3%   Same as 1997.

2009-2010   11/13 = 84.6%   Same as 1997, except an attempt at
                            a census for all northern schools
                            was made and an increase in overall
                            sample size was facilitated to
                            allow for reporting at the
                            provincial/territorial level, as
                            well as national reporting. Student
                            data were weighted according to
                            provincial/territorial enrolments
                            to ensure these were nationally
                            representative when national
                            reports were being made.

Table 2. Characteristics of the participants in the
Health Behaviour in School-aged Children study
from 1993-2010

                                    HBSC cycle

                          1993-1994 *          1997-1998 *

Overall                    N = 7020            N = 11,415
                          ([dagger])           ([dagger])
Descriptors
  Age (mean             13.5 [+ or -] 1.7   13.8 [+ or -] 1.6
     [+ or -] SD)
                           n        (%)        n        (%)

  Sex
    Male                 3350     (47.9)     5500     (48.4)
    Female               3644     (52.1)     5870     (51.6)
  Grade
    [less than or        2329     (33.2)     2137     (18.7)
       equal to] 6
    7-8                  2356     (33.5)     4296     (37.7)
    [greater than or     2335     (33.3)     4967     (43.6)
      equal to] 9
  Subjective
      social status
    Low                   923     (13.2)     1562     (13.9)
    Average              2931     (42.0)     4890     (43.5)
    High                 2574     (36.9)     4784     (42.6)
    Unknown               549      (7.9)      --        --
    Total                6977      (100)    11,236     (100)
    Missing               43                  179
Outcomes
  Physical fighting
    None                  --        --        --        --
    1 time
    2-3 times
    4 or more times
  Total
  Missing
  Fighting-related
      injury
    No injury            4371     (63.3)     6962     (63.6)
    Yes                   92       (1.3)      119      (1.1)
    No, injury not       2442     (35.4)     3870     (35.3)
      related to
      fighting
Total                    6905      (100)    10,951     (100)
Missing                   115                 464

                                    HBSC cycle

                          2001-2002           2005-2006

Overall                    N = 7235            N = 9717
                          ([dagger])          ([dagger])
Descriptors
  Age (mean             13.6 [+ or -] 1.5   14.0 [+ or -] 1.5
     [+ or -] SD)
                           n        (%)        n        (%)

  Sex
    Male                 3357     (46.4)     4604     (47.4)
    Female               3878     (53.6)     5111     (52.6)
  Grade
    [less than or        2063     (28.5)     1723     (17.7)
       equal to] 6
    7-8                  2788     (38.5)     3670     (37.8)
    [greater than or     2384     (33.0)     4324     (44.5)
      equal to] 9
  Subjective
      social status
    Low                   597      (8.9)      808      (8.6)
    Average              2315     (34.3)     3003     (32.0)
    High                 3826     (56.8)     5585     (59.4)
    Unknown               --        --        --        --
    Total                6738      (100)     9396      (100)
    Missing               497                 321
Outcomes
  Physical fighting
    None                 4542     (65.0)     5652     (58.8)
    1 time               1192     (17.0)     1705     (17.7)
    2-3 times             787     (11.2)     1372     (14.3)
    4 or more times       472      (6.8)      885      (9.2)
  Total                  6993      (100)     9614      (100)
  Missing                 242                 103
  Fighting-related
      injury
    No injury            3574     (52.0)     5364     (56.0)
    Yes                   116      (1.7)      216      (2.2)
    No, injury not       3184     (46.3)     4001     (41.8)
      related to
      fighting
Total                    6874      (100)     9581      (100)
Missing                   361                 136

                           HBSC cycle

                            2009-2010

Overall                    N = 26,078
                           ([dagger])
Descriptors
  Age (mean             13.8 [+ or -] 1.6
     [+ or -] SD)
                           n        (%)

  Sex
    Male                12,815    (49.2)
    Female              13,254    (50.8)
  Grade
    [less than or        5165     (19.8)
       equal to] 6
    7-8                 10,471    (40.2)
    [greater than or    10,442    (40.0)
      equal to] 9
  Subjective
      social status
    Low                  2339      (9.5)
    Average              8276     (33.6)
    High                13,998    (56.9)
    Unknown               --        --
    Total               24,613     (100)
    Missing              1466
Outcomes
  Physical fighting
    None                16,203    (64.4)
    1 time               4092     (16.3)
    2-3 times            3067     (12.2)
    4 or more times      1786      (7.1)
  Total                 25,148     (100)
  Missing                 930
  Fighting-related
      injury
    No injury           12,959    (51.6)
    Yes                   543      (2.1)
    No, injury not      11,632    (46.3)
      related to
      fighting
Total                   25,134     (100)
Missing                   944

* No physical fighting variables from HBSC
1993-1994 and 1997-1998.

([dagger]) Variable totals may not equal sample
size because of missing data for some variables.

Table 3.
Prevalence of physical fighting (one or more times) and
fighting-related injury for each HBSC cycle (1993-2010)
overall and by demographic characteristic (sex, grade and
subjective social status);trend analysis across HBSC time
cycles also reported

                        Physical fight (1 or      Physical fight (1 or
                           more times) n(%)          more times) n(%)

                                                                [P.sub.
                                                                 trend]
HBSC cycle            2001-2002     2005-2006    2009-2010   ([dagger])

Variable
  Sex
  Male               1552 (48.3)   2421 (53.4)   5944 (48.7)     0.167
  Female             899 (23.8)    1541 (30.3)   2997 (23.2)     0.001
Grade
 [less than or       733 (37.7)    819 (48.1)    1992 (40.9)     0.275
    equal to] 6
 7-8                 984 (36.3)    1612 (44.3)   3652 (36.4)     0.008
 [greater than       734 (31.4)    1531 (35.8)   3300 (32.3)     0.335
    or equal to] 9
Subjective
social status
  Low                239 (40.4)    414 (51.3)    964 (42.3)      0.452
  Average            850 (36.9)    1299 (43.5)   3073 (37.9)     0.367
  High               1272 (33.4)   2132 (38.4)   4504 (32.8)     0.002
  Overall *          2451 (35.1)   3962 (41.2)   8945 (35.6)     0.015

                        Fighting-related        Fighting-related
                          injury n(%)              injury n(%)

HBSC cycle           1993-1994   1997-1998   2001-2002   2005-2006

Variable
  Sex
  Male               57 (1.7)    74 (1.4)    64 (2.0)    149 (3.3)
  Female             34 (0.9)    44 (0.8)    52 (1.4)     67(1.3)
Grade
 [less than or       17 (0.7)    14 (0.7)    24 (1.2)    32 (1.9)
    equal to] 6
 7-8                 28 (1.2)    48 (1.2)    43 (1.6)    106 (2.9)
 [greater than       47 (2.0)    57 (1.2)    49 (2.1)    78 (1.8)
    or equal to] 9
Subjective
social status
  Low                11 (1.2)    20 (1.3)    17 (3.0)    41 (5.1)
  Average            41 (1.4)    55 (1.2)    44 (2.0)    68 (2.3)
  High               32 (1.3)    42 (0.9)    52 (1.4)    102 (1.8)
  Overall *          92 (1.3)    119 (1.1)   116 (1.7)   216 (2.3)

                       Fighting-related
                          injury n(%)

HBSC cycle           2009-2010    [P.sub.
                                   trend]
                                 ([dagger])

Variable
  Sex
  Male               360 (2.9)     <0.001
  Female             183 (1.4)     <0.001
Grade
 [less than or       87 (1.8)      <0.001
    equal to] 6
 7-8                 196 (1.9)     <0.001
 [greater than       260 (2.6)     <0.001
    or equal to] 9
Subjective
social status
  Low                91 (4.0)      <0.001
  Average            210 (2.6)     <0.001
  High               223 (1.6)     0.001
  Overall *          543 (2.2)     <0.001

Note: No physical fighting variables from HBSC 1993-
1994 and 1997-1998.

* Overall n and % calculated from total HBSC population for
that cycle.

([dagger]) Cochrane-Armitage test for trend for physical fighting
and fighting-related injury across HBSC time cycles. 2001-2010
for physical fighting, and 1993-2010 for fighting-related
injuries.

Table 4. Prevalence ratios and 95% confidence intervals for
the outcomes of physical fighting and fighting-related
injury by demographic characteristics across HBSC cycles,
1993-2010

                     Physical fighting (1 or more times)
                     prevalence ratio (95% Cl)

HBSC cycle           2001-2002          2005-2006

Variable
  Sex
    Male             2.03 (1.90-2.17)   1.76 (1.67-1.85)
    Female           Ref.               Ref.
    p-Value          <0.001             <0.001

  Grade
    [greater than    1.20 (1.10-1.30)   1.34 (1.26-1.43)
    or equal to] 6
  7-8                1.16 (1.07-1.25)   1.24 (1.17-1.31)
    [greater than    Ref.               Ref. Ref.
    or equal to] 9
  p-Trend            <0.001             <0.001 <0.001

Subjective
social status
  Low                1.21 (1.09-1.35)   1.34 (1.24-1.44)
  Average            1.10 (1.03-1.18)   1.13 (1.08-1.20)
  High               Ref.               Ref.
  p-Trend            <0.001             <0.001

                     Physical fighting (1 or more times)
                     prevalence ratio (95% Cl)

HBSC cycle           2009-2010          1993-1994

Variable
  Sex
    Male             2.10 (2.03-2.18)   1.84 (1.20-2.80)
    Female           Ref.               Ref.
    p-Value          <0.001             0.004

  Grade
    [greater than    1.27 (1.21-1.32)   0.37 (0.21-0.63)
    or equal to] 6
  7-8                1.13 (1.09-1.17)   0.60 (0.38-0.95)
    [greater than                       Ref.
    or equal to] 9
  p-Trend                               0.001

Subjective
social status
  Low                1.29 (1.22-1.36)   0.96 (0.49-1.90)
  Average            1.16 (1.12-1.20)   1.13 (0.72-1.79)
  High               Ref.               Ref.
  p-Trend            <0.001             0.913

                     Physical fighting  Fighting-related
                     (1 or more times)  injury
                     prevalence         prevalence
                     ratio (95% Cl)     ratio (95% Cl)

HBSC cycle           1997-1998          2001-2002

Variable
  Sex
    Male             1.82 (1.26-2.64)   1.45 (1.01-2.08)
    Female           Ref.               Ref.
    p-Value          0.001              0.044

  Grade
    [greater than    0.58 (0.33-1.04)   0.58 (0.36-0.94)
    or equal to] 6
  7-8                0.98 (0.67-1.44)   0.76 (0.51-1.14)
    [greater than    Ref.               Ref.
    or equal to] 9
  p-Trend            0.116              0.023

Subjective
social status
  Low                1.47 (0.87-2.50)   2.12 (1.23-3.64)
  Average            1.28 (0.86-1.91)   1.41 (0.95-2.10)
  High               Ref.               Ref.
  p-Trend            0.117              0.005

                     Fighting-related injury
                     prevalence ratio (95% Cl)

HBSC cycle           2005-2006          2009-2010

Variable
  Sex
    Male             2.50 (1.88-3.33)   2.08 (1.75-2.49)
    Female           Ref.               Ref.
    p-Value          <0.001             <0.001

  Grade
    [greater than    1.04 (0.69-1.56)   0.68 (0.54-0.87)
    or equal to] 6
  7-8                1.60 (1.20-2.14)   0.76 (0.63-0.91)
    [greater than    Ref.               Ref.
    or equal to] 9
  p-Trend            0.261              <0.001

Subjective
social status
  Low                2.78 (1.95-3.96)   2.43 (1.91-3.09)
  Average            1.24 (0.91-1.68)   1.59 (1.32-1.91)
  High               Ref.               Ref.
  p-Trend            <0.001             <0.001

* No physical fighting variables from HBSC
1993-1994 and 1997-1998.

Table 5. Reports of physical fighting (one or more times) by
with whom they fought from the Canadian HBSC
Survey, 2001-2006

                               Physical fighting (1 or more
                                times) across HBSC cycles

                               2001-2002       2005-2006
                              (N = 7235)       (N = 9717)

                             n      (%)      n      (%)

Overall                     2451   (35.1)   3962   (41.2)
  Who did you fight with?
    Total stranger          222    (9.7)    272    (7.1)
    Parent or adult          30    (1.3)    100    (2.6)
      family member
    Brother or sister       515    (22.4)   1101   (28.6)
    Boyfriend,               47    (2.0)     71    (1.8)
      girlfriend or date
    A friend or             987    (42.9)   1666   (43.3)
      someone I know
    Other                   499    (21.7)   637    (16.6)
Total                       2300            3847
Missing                     151             115

Note: "Who did you fight with" variable is not included
in the 2009-2010 HBSC Survey.

Table 6. Reports of injuries related to fighting by variables
describing the scenario from the Canadian
HBSC Survey, 1993-2010

                            Fighting-related     Fighting-related
                            injuries             injuries
                            across HBSC cycles   across HBSC cycles

                            1993-1994            1997-1998
Variable                        n       (col%)       n       (col%)

Injury caused missed day       51       (55.4)      68       (57.1)
of school or activities
  Results of injury
    Broken bone                33       (36.6)      18       (15.6)
    Sprain/strain               7       (7.8)        7       (6.1)
    Cut or puncture            20       (22.2)      10       (8.7)
      wound
    Concussion or               6       (6.7)        9       (7.8)
      head/neck injury
    Bruises or                 16       (17.8)      44       (38.3)
      internal bleeding
    Burns                       0        (0)         6       (5.2)
    Internal injury            --         --        --         -
      (operation)
    Other                       8       (8.9)       21       (18.3)
  Total                        90                   115
  Missing                      135                  464
Injury needed                  53       (57.6)      --         --
medical treatment
  Place that injury
      occurred
    Home/yard                  28       (31.1)      50       (42.0)
    School                     23       (25.6)      30       (25.2)
    Sports arena/facility       7       (7.8)       10       (8.4)
    Street/parking lot         13       (14.4)      14       (11.8)
    Park                        6       (6.7)       --         --
Commercial/business area        -         --        --         --
  Other                        13       (14.4)      15       (12.6)
  Total                        90                   119
  Missing                      136                  514
Injury happened during          -         --        12       (10.1)
activity, league
or club
  Season that
      injury occurred
    Winter                     11       (13.6)      17       (14.9)
    Spring                     13       (16.0)      16       (14.0)
    Summer                     28       (34.6)      35       (30.7)
    Fall                       29       (35.8)      46       (40.4)
  Total                        81                   114
  Missing                      365                  647
Place treated for
most serious injury
  Doctor's office/             --         --        --         --
    health clinic
  Emergency room
  Hospital admission
    overnight
  School health services
  Other
Total
Missing

                            Fighting-related     Fighting-related
                            injuries             injuries
                            across HBSC cycles   across HBSC cycles

                            2001-2002            2005-2006
Variable                        n       (col%)       n       (col%)

Injury caused missed day       59       (51.8)      116      (57.1)
of school or activities
  Results of injury
    Broken bone                16       (13.8)      --         --
    Sprain/strain              10       (8.6)
    Cut or puncture            14       (12.1)
      wound
    Concussion or               7       (6.0)
      head/neck injury
    Bruises or                 32       (27.6)
      internal bleeding
    Burns                      29       (25.0)
    Internal injury             5       (4.3)
      (operation)
    Other                       3       (2.6)
  Total                        116
  Missing                      692
Injury needed                  --         --        103      (50.2)
medical treatment
  Place that injury
      occurred
    Home/yard                  42       (38.9)      43       (20.5)
    School                     26       (24.0)      50       (23.8)
    Sports arena/facility       7       (6.5)       23       (11.0)
    Street/parking lot         23       (21.3)      62       (29.5)
    Park                       --         --        --         --
Commercial/business area        3       (2.8)       --         --
  Other                         7       (6.5)       32       (15.2)
  Total                        108                  210
  Missing                      533                  148
Injury happened during         13       (11.4)      55       (26.8)
activity, league
or club
  Season that
      injury occurred
    Winter                     28       (25.7)      --         --
    Spring                     47       (43.1)
    Summer                     16       (14.7)
    Fall                       18       (16.5)
  Total                        109
  Missing                      858
Place treated for
most serious injury
  Doctor's office/             30       (30.9)      47       (24.8)
    health clinic
  Emergency room               33       (34.0)      43       (22.8)
  Hospital admission            9       (9.3)       25       (13.2)
    overnight
  School health services        7       (7.2)        6       (3.2)
  Other                        18       (18.6)      68       (36.0)
Total                          97                   189
Missing                        898                  379

                            Fighting-related
                            injuries
                            across HBSC cycles

                            2009-2010
Variable                        n       (col%)

Injury caused missed day       311      (61.5)
of school or activities
  Results of injury
    Broken bone                --         --
    Sprain/strain
    Cut or puncture
      wound
    Concussion or
      head/neck injury
    Bruises or
      internal bleeding
    Burns
    Internal injury
      (operation)
    Other
  Total
  Missing
Injury needed                  245      (46.8)
medical treatment
  Place that injury
      occurred
    Home/yard                  93       (18.5)
    School                     127      (25.3)
    Sports arena/facility      43       (8.6)
    Street/parking lot         142      (28.2)
    Park                       --         --
Commercial/business area       --         --
  Other                        98       (19.4)
  Total                        503
  Missing                     1334
Injury happened during         --         --
activity, league
or club
  Season that
      injury occurred
    Winter                     --         --
    Spring
    Summer
    Fall
  Total
  Missing
Place treated for
most serious injury
  Doctor's office/             --         --
    health clinic
  Emergency room
  Hospital admission
    overnight
  School health services
  Other
Total
Missing

Figure 1. Prevalence of physical fighting (one or more times)
over time (2001-2010). Overall [p.sub.trend] = 0.015

HBSC Cycle (Year)   Prevalence (%)

2001-2002               35.1
2005-2006               41.2
2009-2010               35.6

Note: Table made from line graph.

Figure 2. Prevalence of fighting-related injuries over time
1993-2010). Overall [p.sub.trend] < 0.001

HBSC Cycle (Year)   Prevalence (%)

1993-1994               1.3
1997-1998               1.1
2001-2002               1.7
2005-2006               2.3
2009-2010               2.2

Note: Table made from line graph.