Abstract
Objectives. We described nonfatal injuries and calculated injury rates among middle-school and high-school students in Guangxi, China.
Methods. Students were selected using multistage randomizing techniques, and nonfatal injuries in 1840 students from February 2002 to January 2003 were monitored with standardized injury forms. Risk factors for injury were identified in multivariate analyses.
Results. The annual overall injury rate was 32.3 per 100 students. Boys had a significantly higher injury rate than the girls (34.8 vs 30.3 per 100 students), and a significantly higher proportion of injuries in boys was caused by other students (28.1% vs 19.4%). A higher proportion of injuries in girls (40.7%) occurred at home. For both boys and girls, sports were the most common activities associated with injury. Injuries from falls were the leading cause of injury, and extremities were most frequently injured. Gender, age, ethnicity, and family income levels were identified as significant risk factors for injury in multivariate analyses.
Conclusions. Nonfatal injuries in middle-school and high-school students should be recognized as a significant public health concern in China.
Injuries are the leading cause of death and disability for children around the world.1–4 In addition to lost lives and permanent disabilities, nonfatal injuries in students are associated with serious social and economic consequences, including health care costs, psychological distress in families, missed school days, and parents’ missed work days.5–8
Previous studies have estimated that annual injury rates in middle-school and high-school students range from 5 to 50 injuries per 100 students.3,9–11 One study showed that school-aged athletes had a higher injury rate than nonathletes,12 and another study showed substantial differences in injury rates of adolescents by gender and school year.13 However, the majority of school-related injuries occurred in classrooms or on school playgrounds.3,14,15
Results from several studies among middle-school and high-school students indicate that injury type, cause of injury, injured body part, and location of injury vary by student demographics and family socioeconomic status.16–19 An international comparison study of medically attended, nonfatal injuries in adolescents in 24 developed countries across Europe and North America showed a wide range of injury patterns.11 Although epidemiological characteristics of nonfatal injuries in middle-school and high-school students have been well documented in the United States and other developed countries, little has been done in China. We used a large, randomly selected sample of middle-school and high-school students in Baise City, Guangxi Zhuang Autonomous Region of China from February 2002 to January 2003 to conduct a prospective study with the specific aim of investigating nonfatal unintentional injuries. We present results from descriptive epidemiological analyses of nonfatal injuries in this student population with a focus on describing injury characteristics and identifying risk factors for injury.
This prospective cohort study was designed to monitor nonfatal injuries in middle-school and high-school students in Baise City, Guangxi Zhuang Autonomous Region of China. The schools were selected using multistage randomizing techniques on the basis of the geographical location and the total number of classes of the schools in three districts of Baise City. Once schools were selected, invitations were sent to school principals and nurses via the Baise City Education Bureau. Each school principal reviewed the study design and signed a cooperative agreement approving participation in the study. Consent to participate in the study was obtained from parents of the selected students, and assent was obtained from the students. The students were allowed to withdraw from the study at any time.
Students aged 11 to 18 years were randomly selected in middle and high schools in Baise City, Guangxi Zhuang Autonomous Region of China. Overall, 1855 students were randomly selected from 36 classes in 9 middle and high schools.
A reportable injury in this study was defined as any injury meeting at least one of the following criteria: (1) an injury for which the student received medical treatment at the school nurse’s office, (2) an injury for which the student received emergency medical care from a doctor at a hospital or a private medical office, (3) an injury for which the student received first aid from his/her parent and missed a half day or more of school or regular activities, or (4) an injury that was not treated but caused the student to miss a half day or more of school or regular activities.
The injury definition criteria, taught during training for the study, were used by school nurses to confirm all injuries. The focus of this study was unintentional injuries; therefore, medically diagnosed child abuse, suicide, and homicide cases were excluded.
Demographic information about the students (e.g., age and gender) and their family (e.g., number of children in the family, family income, marital status, and parents’ education) was collected from the parents at the beginning of the new semester in February 2002 with a survey questionnaire. Students were followed through January 31, 2003. For injuries that occurred at schools, school nurses provided medical treatment and collected detailed information about the injury. Information regarding injuries that occurred outside of the schools was collected during monthly parent–teacher conferences. During these meetings, the teacher asked the parent about any injury that the student experienced in the past month outside of school. If an injury had occurred, the parent reported the injury to a school nurse, who then collected detailed information about the injury from an interview with the parent after the meeting.
Injury data were collected and recorded with Injury Surveillance Survey Forms. These forms recorded the following information for each injury: student’s name and gender, date and time of the injury, cause of the injury, activity when injured, location of the injury, person(s) who caused the injury, body part injured, school days or activities missed because of the injury, and medical advice or treatment received after the injury.
A pilot study was conducted with 45 students at one school in December 2001. Feedback from the pilot study was used to finalize the questionnaire and the Injury Surveillance Survey Form. Before the study, all school nurses and teachers of the selected classes received formal training, which was provided by researchers from the School of Public Health, Wuhan University, Wuhan, China.
The SPSS statistics package (version 11.5)20 was used for all data analyses. Annual injury incidence rates per 100 students were calculated to describe injury incidence by student gender, age group, and ethnicity. Injury incidence rates per 100 students were also calculated by family characteristics, including total number of children in the family, marital status, parents’ education, and monthly family income. According to 2002 local governmental statistics,21 the average monthly household income in Baise City was 2000 Yuan (approximately US$245); therefore, we classified family monthly income into the following three levels: <2000 Yuan, 2000 to 3999 Yuan, and ≥ 4000 Yuan.
We compared injuries between male and female students. Specific injury characteristics compared included cause of the injury, location of the injury, activity when injured, body part injured, person(s) who caused the injury, medical care received after the injury, and school days or activities missed because of the injury.
We used the χ2 statistic to test differences in injury risk among students with different demographics and family characteristics and also the differences in injury characteristics between male and female students. We selected P ≤.05 as the significance level. Multivariate logistic regression models were used to control for potential confounding effects of the selected variables.22,23 Odds ratios and associated 95% confidence intervals were calculated.
Overall, a total of 1840 of the selected 1855 students completed the study, yielding a 99.2% completion rate. The distribution of student demographics and family characteristics was very similar between the selected sample and the total population of students. A total of 595 students reported nonfatal injuries during the 1-year period, generating an overall annual injury incidence rate of 32.3 per 100 students.
Student characteristics and injury incidence rates are summarized in Table 1. Male students had significantly higher injury rates than female students (34.8 vs 30.3 per 100 students; P =.04). Students from nonminority ethnicities (i.e., Han and Zhuang) had almost identical injury rates (31.5 and 31.7 per 100 students). However, the injury incidence rate (45.6 per 100 students) for minority students, including Miao, Bai, Tong, and Yao ethnicities, was significantly higher than that for nonminority students (P =.02). The injury incidence rate decreased with increasing age (i.e., from 41.4 per 100 students aged 11 years to 20.2 per 100 students aged 17–18 years).
Students who were the only child in a family had a significantly higher injury rate than those who had families with more than one child (36.5 vs 27.4 per 100 students; P <.01). There was no significant difference in the injury incidence rate between students whose parents were divorced and those whose parents were married. However, for students with divorced parents, living arrangements had a significant impact on the injury incidence rate. Compared with students who lived with mothers or fathers, those who lived with their grandparents had a significantly lower injury rate (15.0 per 100 students living with grandparents vs 30.8 per 100 students living with mothers and 45.3 per 100 students living with fathers; P =.03). Injury risk was also associated with parents’ education and household income at a significant level (P <.01).
Overall, approximately one third of students (31.4%) had more than one injury during the study period (Table 2). However, compared with female students, a significantly higher proportion of male students had two or more injuries (P =.03).
The characteristics of injury by gender are summarized in Table 3. For both male and female students, falls were the leading cause of injury. Compared with injuries in male students, a significantly higher proportion of injuries in female students occurred at home (40.7% vs 24.3%) or while working in the house (19.4% vs 3.7%). However, a significantly higher proportion of injuries in male students (36.1% vs 23.1%) occurred during sports activities. Related to this, a significantly higher proportion of injuries in male students was also caused by schoolmates (28.1% vs 19.9%).
In both male and female students, more than 67% of injuries were to the lower and upper extremities. During the study period, 23.3% of injuries in female students and 30.1% of injuries in male students required outpatient medical treatment, and a small percentage (3.2% of male injuries and 4.5% of female injuries) resulted in hospitalizations. However, approximately 10% of injuries in both male and female students did not receive medical treatment. Approximately half of the injuries (51.2%) caused students to miss 1 to 3 days of school or activities.
Five logistic regression models were fitted to identify risk factors for injury with control for potential confounding effects of selected variables (Table 4). Of all of the significant variables identified in the univariable analyses in Table 1, only gender, age, and ethnicity were retained in the model, using the method of stepwise autoselection (model 1). After control for ethnicity and age group, male students were significantly more likely than female students to sustain injuries (odds ratio [OR] = 1.25; 95% confidence interval [CI] = 1.02, 1.53). Compared with students aged 17 to 18 years, younger students were significantly more likely to sustain injuries (aged 11–13 years, OR = 1.51; 95% CI = 1.00, 2.26; aged 14–16 years, OR = 2.94; 95% CI = 1.96, 4.42). Consistent with findings in univariable analyses, minority students had approximately a 67% higher risk of injury than Han students (OR = 1.67; 95% CI = 1.05, 2.66).
Although both the total number of children in the family and the educational level of the mother were statistically significant risk factors for injury in univariable analyses, these became nonsignificant variables after control for the potential confounding effects of gender, age, and ethnicity (models 2 and 4). However, household monthly income was still significantly associated with injury risk in students in logistic regression analyses. With control for the potential confounding effects of gender, age, ethnicity, and the mother’s educational level, the odds of injury among students whose families had a median monthly income of 2000 to 4000 Yuan was 1.42 (95% CI=1.11, 1.81) compared with students whose families had a monthly income of less than 2000 Yuan. The interaction effects among selected variables were evaluated; however, none were significant at P ≤.05, and thus the results are not reported here.
We used a prospective cohort study design to collect information on nonfatal injuries from a large random sample of middle-school and high-school students in China. We found that students who were male, younger, and from an ethnic minority/relatively wealthy family had a higher risk of injury. For both male and female students, sports were the most common injury-related activity, falls were the leading cause of injuries, and extremities were the most frequently injured. Compared with injuries in female students, a significantly higher proportion of injuries in male students was caused by classmates. However, females were more likely than males to be injured at home.
We found an overall annual injury rate of 32.3 injuries per 100 students, which is consistent with injury rates of 5 to 50 injuries per 100 students in previous studies.2,3,9,10,12,24–27 The lower injury rates found in some of these studies are related to study design and injury definition. For example, two previous studies10,26 used data collected by the US National Health Interview Survey but only estimated injury rates for a 3-month recall period. The study conducted by Radelet et al. showed only sports-related injuries.12 Our study methodology was most comparable with the methodology used in previous studies in Shantou City in the Guandong province9 and in the United States.6 During a 1-year follow-up of 2559 students aged 7 to 16 years, the study by Li et al. showed an injury rate of 37.9 injuries per 100 students.9 Using a similar study design, Lenaway et al. monitored nonfatal injuries among 5518 elementary-school to high-school students and found an injury rate of 9.2 per 100 students.6 An international comparison study of medically attended nonfatal injuries among adolescents across 24 developed countries in Europe and North America showed that the annual injury rate among students aged 11 to 15 years varied from 10 to 47 injuries per 100 students in different countries.11 Comparison of injury rates between different injury studies is challenging for many reasons, including different study periods,10,12 different injury definitions,12 and different data collection systems.1,2 According to King et al., it may be more useful to look at the types of injuries by age group and by gender and the circumstances in which injuries occur.11
Our results regarding the leading cause of injury, activity when injured, location of injury, and body part injured among this school population were strikingly consistent with findings from other studies conducted around the world.2,3,10 Our findings and those of previous studies highlight the importance of prevention of sports-related injuries in injury control and prevention programs that focus on students. Around the world, health professionals and public health agencies advocate physical activity to combat the rising epidemic of childhood obesity.28 However, numerous studies have documented the substantial economic and social impact of injuries related to sports, recreation, and exercise.29,30 Parents and students need to recognize that injuries are not an inevitable part of participation in sports but are predictable and preventable. Middle-school and high-school students can harvest the tremendous benefits of physical activity when they play safely. Physical education teachers and coaches in schools need to be aware of the high risk of injury associated with sports activities and should design age-appropriate activities for students to reduce the risk of injury.
The inverse relationship of injury rate and age in our study was different from that found in a previous study.9 Injury rate in the study by Li et al. increased from 18.6 injuries per 100 students for those aged 7 to 9 years to 55.4 injuries per 100 students for those aged 13 to 16 years.9 However, our analysis showed that the injury rate decreased from 41.4 per 100 students aged 11 years to 20.2 per 100 students aged 17 to 18 years. Our results were more consistent with findings from a study conducted in Taiwan3 and from studies conducted among American students.6,7 In Taiwan, students in the ninth grade had the lowest injury rate when compared with students in the seventh and eighth grades. Lenaway et al. also found that American high-school students had a lower injury rate than elementary-school students and middle-school/junior high-school students.6 Differences in injury rate by age are more likely because of changes in student developmental stages and changes in the types of activities in which students participate.31 Analysis of injury rates6 revealed that high-school students were injured more frequently in the gym, middle-school/junior high-school students were injured more frequently on the athletic field, and elementary-school students were injured more frequently on the playground. Our results from multivariate regression analysis also showed that when the confounding effect from other variables was controlled, students aged 14 to 16 years had the highest risk of injury. This might indicate that, in our study, students aged 14 to 16 years participated in different types of activities or had more exposure time to high-risk environments than the students in the other two age groups. Future studies comparing sports/ leisure activities and exposure times to these activities between middle-school and high-school students in China may provide additional useful information.
Differences in injury risk and injury patterns in male and female students of different ethnicity in our study might reflect differences in socioeconomic status, social activities, or urban versus rural residence of the students in China. Our findings indicate that minority students had a higher risk of injury than majority students (higher injury rate and OR) and that socioeconomic background has a major impact on injury risk. With regard to gender, more male students than female students in China participate in aggressive sports activities such as football and basketball, whereas female students are more likely to participate in less aggressive activities such as dancing or performing household chores.32 This difference in social activities may explain the significantly higher proportion of sports-related and leisure activity-related injuries among male students in our study. This is also likely to be the reason why a significantly higher proportion of household injuries occurred among female students. Nevertheless, approximately one third of the injuries among both male and female students occurred in schools, and 28.1% of injuries among male students were caused by other children in schools. These findings have implications for injury-prevention initiatives in schools. As suggested by other researchers,3,7,33 schools are social institutions that have a pivotal role in influencing student behaviors as well as academic and social development. Therefore, school administrators are responsible for monitoring the school environment and providing supervision during class breaks and activities after school.
We used a prospective study design to monitor nonfatal unintentional injuries among middle-school and high-school students in China. This study design allowed us to calculate injury rates and compare relative injury risk in students. We also controlled for potential confounding effects of selected variables by conducting logistic regression analyses, which were seldom used in other studies.3,6,7,9,33 Finally, compared with other studies, our participant completion rate was very good. Nevertheless, our findings are subject to at least two limitations. First, underreporting by parents or students may have occurred, particularly when they were asked to report injuries that occurred during breaks outside of the school. In China, students usually have a 45-day break in July and August. Some parents/students may not remember an injury or the details of an injury that occurred during the school break. Another limitation is that our study was conducted in one city in China; therefore, findings in our study population may not reflect the injury risk and patterns of injury among the entire middle-school and high-school student population of China.
In summary, the results from this large prospective cohort study present an epidemiological overview of characteristics and risk factors for nonfatal unintentional injuries experienced by middle-school and high-school students in Gangxi, China. Our data indicate that the demographics and family characteristics of the students had a significant impact on injury risk and injury patterns among school students. Our findings underscore the importance of injury prevention in schools as well as outside of schools. Nonfatal injuries among middle-school and high-school students should be recognized as a significant public health concern in China.
aStatistics of χ2 test. bIncludes 34 students with Miao ethnicity, 20 students with Bai ethnicity, and 12 students with Tong ethnicity. cAll except above categories. aχ2 test between male and female students with two or more injuries. aχ2 test comparing male and female students. bIncludes drowning, firecracker-related injury, electrical injury. cAll places except home, school, and on the way from/to school. dIncludes household chores such as pouring boiled water and cooking. eAll except self and schoolmates. Note. OR = odds ratio; CI = confidence interval.
No. Students No. Injuried (%) χ2 a P Gender Male 836 291 (34.8) 4.28 .04 Female 1004 304 (30.3) Age, y 11 29 12 (41.4) 67.32 < .01 12 337 137 (40.7) 13 375 160 (42.7) 14 201 71 (35.3) 15 294 82 (27.9) 16 436 95 (21.8) 17–18 168 34 (20.2) Ethnicity Han 531 167 (31.5) 7.57 .02 Zhuang 1219 387 (31.7) Minority b 90 41 (45.6) Number of children in the family 1 998 364 (36.5) 17.05 < .01 2 842 231 (27.4) Parents’ marital status Divorced 130 42 (32.3) 0.00 .99 Married 1710 553 (32.3) If parents are divorced, person the child lives with Father 39 12 (30.8) 8.88 .03 Mother 53 24 (45.3) Grandparent 20 3 (15.0) Other c 18 3 (16.7) Father’s education University 620 239 (38.5) 18.60 < .01 Less than university 1220 356 (29.9) Mother’s education University 383 159 (41.5) 16.50 < .01 Less than university 1457 436 (29.9) Family monthly income, Yuan < 2000 984 280 (28.5) 19.60 < .01 2000–3999 442 178 (40.3) ≥ 4000 414 137 (33.1) Injury Episodes per Year Male No. (%) Female No. (%) Total No. (%) P a 1 189 (64.9) 221 (72.7) 410 (68.9) 2 71 (24.4) 67 (22.0) 138 (23.2) .03 ≥ 3 31 (10.7) 16 (5.3) 47 (7.9) Male (n = 426) No. (%) Female (n = 407) No. (%) Total (n = 833) No. (%) P a Cause of injury < .01 Transportation 25 (5.9) 17 (4.1) 42 (5.0) Fire/burning/scald 18 (4.3) 15 (3.6) 33 (4.0) Falls 151 (35.5) 123 (30.3) 274 (33.0) Struck by object or person 108 (25.3) 57 (14.0) 165 (19.8) Animal/insect bite 14 (3.2) 15 (3.6) 29 (3.4) Cut/pierce 46 (10.8) 72 (17.6) 118 (14.1) Choking 9 (2.2) 9 (2.3) 18 (2.2) Poisoning 2 (.5) 4 (.9) 6 (.7) Otherb 53 (12.4) 96 (23.5) 149 (17.8) Location of injury < .01 Home 104 (24.3) 166 (40.7) 270 (32.3) School 164 (38.4) 131 (32.1) 295 (35.3) From/to school 44 (10.3) 37 (9.0) 81 (9.7) Otherc 115 (27.0) 74 (18.1) 189 (22.7) Activity Riding bicycle 78 (18.2) 66 (16.2) 144 (17.3) < .01 Working around the house 16 (3.7) 79 (19.4) 95 (11.4) Sports 154 (36.1) 94 (23.1) 248 (29.8) Leisure activity 62 (14.5) 29 (7.1) 91 (10.9) Walking 23 (5.4) 37 (9.1) 60 (7.2) Climbing/descending stairs 26 (6.1) 25 (6.1) 51 (6.1) Eating/drinking 13 (3.0) 19 (4.7) 32 (3.8) Bathing 4 (1.0) 6 (1.5) 10 (1.2) Otherd 50 (11.8) 52 (12.8) 102 (12.2) Person(s) who caused injury .01 Self 236 (55.4) 234 (57.5) 470 (56.4) Schoolmate(s) 120 (28.1) 81 (19.9) 201 (24.1) Othere 70 (16.5) 92 (22.6) 162 (19.5) Body part injured < .01 Head/face 46 (10.7) 69 (17.0) 115 (13.8) Eyes 14 (3.2) 15 (3.7) 29 (3.4) Teeth 12 (2.7) 4 (0.9) 15 (1.8) Upper limbs/hands 153 (35.8) 147 (36.2) 300 (36.0) Lower limbs/feet 180 (42.2) 127 (31.2) 307 (36.8) Back 5 (1.1) 9 (2.3) 14 (1.7) Chest 12 (2.7) 17 (4.1) 29 (3.4) Multiple body parts 7 (1.6) 19 (4.6) 26 (3.1) Care received after injury .10 Provided by teacher/parent 188 (44.1) 212 (52.0) 400 (48.0) Provided by school medical staff 55 (12.9) 44 (10.8) 99 (11.9) Outpatient visit 128 (30.1) 95 (23.3) 223 (26.8) Hospitalization 14 (3.2) 18 (4.5) 32 (3.8) No care received 41 (9.7) 38 (9.4) 79 (9.6) School days missed after injury .07 < 1 157 (36.8) 132 (32.4) 289 (34.7) 1–3 203 (47.6) 224 (55.0) 427 (51.2) 4–6 25 (5.9) 33 (8.1) 58 (7.0) ≥ 7 41 (9.7) 19 (4.6) 60 (7.2) Model 1 OR (95% CI) Model 2 OR (95% CI) Model 3 OR (95% CI) Model 4 OR (95% CI) Model 5 OR (95% CI) Gender Male 1.25 (1.02, 1.53) 1.23 (1.01, 1.50) 1.24 (1.01, 1.51) 1.24 (1.01, 1.52) 1.22 (1.00, 1.50) Female 1.00 1.00 1.00 1.00 Age group, y 11–13 1.51 (1.00, 2.26) 1.46 (.97, 2.20) 1.45 (.97, 2.18) 1.49 (1.00, 2.24) 1.41 (.94, 2.13) 14–16 2.94 (1.96, 4.42) 2.71 (1.78, 4.13) 2.80 (1.85, 4.23) 2.76 (1.82, 4.18) 2.51 (1.63, 3.85) 17–18 1.00 1.00 1.00 1.00 1.00 Ethnicity group Zhuang 1.11 (.89, 1.39) 1.12 (.89, 1.40) 1.12 (.89, 1.40) 1.09 (.87, 1.37) 1.12 (.89, 1.40) Other 1.67 (1.05, 2.66) 1.71 (1.07, 2.72) 1.65 (1.04, 2.63) 1.62 (1.02, 2.58) 1.64 (1.03, 2.62) Han 1.00 1.00 1.00 1.00 1.00 Children in the family 1 1.18 (.94, 1.46) 1.14 (.91, 1.42) 2 1.00 1.00 Family monthly income, Yuan < 2000 1.00 1.00 2000–3999 1.02 (.79, 1.33) 1.42 (1.11, 1.81) ≥ 4000 1.49 (1.12, 1.98) 0.93 (.71, 1.21) Mother’s education level University 1.21 (.94, 1.56) 1.20 (.92, 1.56) Below university 1.00 1.00
We thank Drs Yingjia Ma and Qijun Wu at the Department of Preventive Medicine, Youjiang Medical College for Nationalities, for assistance with study design and data collection. We also thank teachers at selected middle and high schools for their contributions to the data collection.
Human Participant Protection This research study was reviewed and approved by the Wuhan University School of Public Health. Principals of selected schools signed written consent forms. Parental consent and student assent were obtained before the study.
