Work-related injury among South Texas middle school students: prevalence and patterns.

Background: Work experiences among early adolescents are largely undocumented. Our purpose was to document the prevalence of work and work-related injury among lower-income Hispanic South Texas middle school students.

Methods: Anonymous surveys were conducted in classrooms of sixth- through eighth-grade students, and 3,008 students reported current or recent employment.

Results: The prevalence of work was 56%; mean weekly work hours were 7.7. Increasing weekly work hours were significantly related to work injury (11-20 hours, odds ratio [OR], 1.5; 95% confidence interval [CI], 1.1-1.9; 21+ hours, OR, 2.4; 95% CI, 1.8-3.2, compared with 1-10 hours). The odds of injury were highest for agriculture (OR, 4.4; 95% CI, 3.3-6.0), followed by restaurant (OR, 3.8; 95% CI, 2.7-5.4), construction (OR, 3.6; 95% CI, 2.4-5.2), and yard work (OR, 1.7; 95% CI, 1.4-2.2).

Conclusion: Working more than 20 hours weekly increased the likelihood of injury among middle school students. Parents and professionals should monitor weekly school-year work hours.

Key Words: early adolescence, middle school, occupational injury

**********

With few exceptions, state, national, and international sanctions prohibit the employment of preadolescents and early adolescents and set legal standards for wages, hours, allowable jobs, and working conditions for minor children 14 years of age and older in nonagricultural work. (1-3) Despite contraindications to the employment of young children, emerging evidence suggests that some unknown proportion of young people younger than 14 years of age work during their middle school years. (4-6) However, there is a paucity of information regarding the extent and nature of the work circumstances of these employed preadolescents and early adolescents who, like their older adolescent counterparts, may be assumed to be at risk--possibly greater risk due to the younger teen's lack of job experience and cognitive, psychosocial, and developmental immaturity--for various occupational hazards. (5-14)

Studies of employment among high school students suggest that school-year employment and increased weekly work intensity may adversely influence sleep patterns, health risk behaviors, social life, physical and mental health, adolescent development, school performance/engagement, and the parent-child relationship. (15-20) Occupational injury among high school students is already documented as a significant public health problem, as studies reveal that injuries are very common, sometimes serious, and possibly positively related to weekly work intensity. (5-6), (8-14) However, the results of studies focusing primarily on older adolescent workers cannot be directly generalized to younger workers, because substantial differences may exist between the two populations in job tasks, frequency and intensity of work, developmental abilities, and work-related experience. (5), (7), (9)

Isolated reports of injuries or fatalities among very young workers are occasionally found in the popular or scientific literature. For example, recent worker's compensation data from two states indicate that 1 to 2% of workers 11 to 13 years of age and 7 to 10% of 14- to 15-year-olds file claims. (13), (14) Of the 125 work-related fatalities among Texas children from 1980 to 1990, nearly one-quarter occurred in children 15 years of age or younger. (13), (14) We were unable to locate any published reports from population-based data on the work experiences of young workers 14 years of age or younger. Among older teens, data are also scarce on the work circumstances of young workers of lower socioeconomic status and race or ethnicity other than white. (4), (5), (21), (22) A few studies of older youth have reported that lower income and minority children are less likely to be employed, but when they are employed, work more hours but fewer weeks than higher income children (A Ahituv, M Tienda, VJ Hotz, unpublished data). (5), (23) These studies also indicate that lower income youth are also more likely to work in more hazardous jobs than their more affluent counterparts. (5), (21), (22)

This descriptive article reports on a cross-sectional study of occupational injury in middle school students, a large proportion of whom are economically disadvantaged, bilingual students living in rural South Texas. Our overall goal is to examine the relationship between the prevalence of injury and work intensity level and to identify particularly high-risk subgroups of this population, so that school-based or other appropriate health/safety interventions can be targeted toward preventing injuries in these most vulnerable workers.

Materials and Methods

An important aim of this study was to explore the relationship between weekly work hours and occupational injury using specific thresholds for weekly work hours (ie, 1-10, 11-20, [greater than or equal to]21 h/wk). Another goal was to identify factors associated with work injury among middle school youth. A third aim was to document the prevalence of work injury in a sample with heavy representation by economically disadvantaged, Hispanic students who might be expected to work longer hours, possibly sustaining more injuries, than other students.

Sample

Data in this cross-sectional study were collected as part of the Safe and Drug Free Schools (SDFS) program's regular assessment of the prevalence of substance use among Texas Education Agency Region Two students representing 27 middle schools in 11 contiguous South Texas counties. This data collection provided an opportunity to examine issues of employment in relation to health and other behaviors.

The SDFS program coordinator recruited all 42 school districts in the 11 counties. Fifteen districts declined participation due to time constraints. Participating and nonparticipating districts were located in small towns and rural areas. The distribution of the educational status of parents and student ethnic characteristics was quite similar among participating and nonparticipating districts. Approximately 60% of students in participating districts were eligible to receive free or reduced cost lunches; the median percentage of Hispanic students in participating districts was 72% (24)

The total middle school population in the participating schools was 11,523. At the larger schools, classes were randomly selected by grade from a master list of second-period classes using a random number table. In schools with fewer than 200 students, all students were surveyed. The number of students eligible to complete the survey was 8,757. A total of 7,420 sixth-through eighth-grade students, representing 85% of those eligible, were surveyed. The number of students who responded was 7,302 (2,365 sixth graders, 2,487 seventh graders, and 2,450 eighth graders). Excluded from the analysis for work-related injury questions were nonworkers (n = 2,359) and cases with missing data on one or more variables (n = 1,935), leaving a sample size of 3,008 students reporting working for pay at a usual job(s). (Note: missing data could be explained by time constraints in survey administration; an analysis of excluded cases resulted in findings similar to those of included cases).

The University of Texas Health Science Center at Houston Institutional Review Board approved the study protocol and survey instrument (HSC-SPH-95-018). During the week before survey administration, parents received a passive informed consent form explaining the study and providing an opportunity to decline participation. Following the Centers for Disease Control and Prevention's Youth Risk Behavior Study protocol, teachers received instruction packets and administered the survey during students' regular classes. (25) The survey was anonymous; students recorded their responses directly on a computer-scannable answer sheet.

Analysis and Measures

Survey questions were selected from instruments previously used in other major studies of adolescent development: the North Carolina Teens at Work Questionnaire, the Massachusetts Teens at Work Questionnaire, the Minnesota Department of Health Work Injury Study Questionnaire, and the National Institute for Occupational Safety and Health's Occupational Injuries to Young Workers Questionnaire (unpublished communications with survey authors).

Independent Variables. Current/recent (within the past 6 months) weekly work hours at a paying job was taken from the Temple University Psychology Department's School-year Work Questionnaire (unpublished communication). This variable was expressed as a three-level variable with the following categories: employed 1 to 10 hours weekly (low intensity), 11 to 20 hours (moderate intensity), and 21 or more hours (high intensity). Type of job when the most serious work-related injury occurred and type of injury were taken from the Temple University Psychology Department's Questionnaire and the Minnesota Department of Health Work Injury Study Questionnaire, respectively.

Demographic variables included students' self-reports of gender, grade, race/ethnicity, and maternal/paternal educational status, a measure of socioeconomic level. Race/ethnicity was collapsed into three categories: non-Hispanic white, Hispanic, and all other ethnic groups (black, Asian/Pacific Islander, American Indian/Alaskan Native, and mixed race). Maternal/paternal educational status was indexed into three groups: one/both did not graduate high school, both graduated high school, one/both graduated college.

Because the employment patterns of young workers vary considerably with frequent movement in and out of the labor force, current/recent weekly work hours usually spent at a paying job(s) was considered an indicator of weekly hours usually worked--a typical school-year work pattern. (5) For some analyses, this item was crossed with "ever injured while working," creating an obvious temporality issue, because the injury referred to in "ever injured while working" could have occurred at the current weekly job or when a student was working at a recent job with a different work pattern.

Outcome Measure. "Ever injured while working" was treated as the dependent variable for most analyses. This item was taken from the North Carolina Teens at Work Questionnaire.

Methods of Analysis. The relationships between the outcome measures and independent variables were analyzed using SPSS 8.0 for Windows 95 statistical software package (SPSS, Inc., Chicago, IL). Univariate logistic regression was used to examine the relationship between work-related injury and 1) current/recent weekly work hours, 2) type of job when the injury occurred, and 3) demographic factors. Multivariate logistic regression provided adjusted odds ratios and 95% confidence intervals and was used to identify risk factors for on-the-job injury and the need for medical care after an injury. For the multivariate analysis, variables were selected using the backward elimination method. At each step, variables were evaluated for removal from the model using the conditional likelihood-ratio test.

Results

Characteristics of the Overall Sample

The overall sample consisted of almost equal proportions of boys and girls. Sixth graders made up 29% of the sample; seventh and eighth graders constituted 35% and 36%, respectively. Sixty-three percent of the sample reported Hispanic ethnicity; 37% reported their race as white. Sixty-three percent had parents whose educational backgrounds did not extend past high school.

Characteristics of Working Students

Table 1 presents the prevalence of work and work-related injury among workers and the mean hours worked weekly, all by the demographic variables. Fifty-six percent of the overall sample reported current/recent employment within the past 6 months; the mean hours worked weekly was 7.7 (standard deviation, 7.5). Among workers, approximately 80% of middle school students were employed for 1 to 10 hours weekly; 12% reported working 11 or more hours weekly. A greater proportion of workers and injured workers were boys, and a greater proportion of whites (66%) reported current/recent employment than Hispanics (50%).

Most employed students reported working in informal settings: 36% worked in childcare and almost one-third worked in yards (32%). Three-quarters of employed girls worked as baby-sitters. Boys most often worked in yards (50%), followed by agriculture (16%), construction and restaurant work (10%), and retail stores (9%). Whites most frequently worked in baby-sitting and yard work (40% and 32%, respectively), as did Hispanics (36% and 29%, respectively); however, Hispanics worked in restaurants (10%) more often than whites (5%).

Prevalence of Injury for the Workers

A total of 773 (25%) employed students reported an occupational injury while working for pay. Forty-one percent of injured workers were employed for 21 or more hours weekly, 30% for 11 to 20 hours weekly, and 23% for 1 to 10 hours weekly.

Association of Work Injury and Work Intensity Level, Type of Job, and Demographics 

Table 2 presents the univariate and adjusted odds ratios (ORs) for work injury by type of job, work intensity level, and demographic characteristics. After adjusting for type of job and the demographic variables, students working more than 20 hours per week were almost twice as likely to be injured as those working less than 10 hours per week (OR, 1.8; 95% confidence interval [CI], 1.3-2.6), whereas those working 11 to 20 hours per week did not have an increased odds of injury. Boys were 50% more likely to be injured at work than girls (OR, 1.5; 95% CI, 1.1-2.0). Hispanics were slightly less likely to be injured than whites (OR, 0.8; 95% CI, 0.6-0.9). Student workers involved in farming and ranching (OR, 3.3), restaurant work (OR, 3.2), construction (OR, 2.4), and retail work (OR, 1.7) had greater odds of an injury than baby-sitters.

Association of Severity of Injury with Work Intensity, Demographics, and Type of Job

Approximately one-third of injured workers required medical help (n = 232). Of these, 76% were boys and 63% were Hispanic. Table 3 presents the univariate and adjusted ORs for the association between severity of injury and type of job, work intensity level, and sociodemographic factors. After adjusting for type of job, weekly work intensity level, and sociodemographic factors, restaurant workers (OR, 3.6; 95% CI, 1.7-7.8) and Hispanics (OR, 1.6; 95% CI, 1.1-2.4) had significantly increased odds of a medically attended injury.

Association of Type of Job and Type of Injury

Table 4 presents the type of injury by employment category when the injury occurred. Common work injuries were punctures/jabs/cuts (22%), such as nail or needle sticks; broken bones, breaks, or fractures (18%); sprains, strains, and torn ligaments or muscles (ie, pulled muscles, sprained ankles, or strained backs) (17%); shocks from electricity or burns from grease, fire, or chemicals (15%); bruises or crushes in which the skin was not broken (13%); dislocations of joints, such as a jammed thumb (9%); and amputations (eg, loss of body parts, such as fingers, toes, arms, or legs) (6%).

Thirty-two percent of all injuries reported by middle school students occurred during yard work and 20% were sustained during childcare. Approximately 16% occurred during farm and ranch work; another 13% occurred in restaurants.

Restaurant injuries most frequently involved broken bones (3.8%), followed by shocks/burns (3.2%). Injuries occurring during yard work often included punctures (8.8%), sprains (6.5%), and broken bones (5%). Construction work most often produced broken bones (1.8%) and punctures and sprains (1.7%). Farm/ranch work often included sprains (3.6%), punctures (3.4%), and broken bones (2.8%). Childcare injuries frequently included bruises (5%), punctures (5%), and sprains (4.3%).

Lost Days from School

Slightly more then half of injured workers lost no time from school. However, 41% lost 1 day to 1 week, and 7% lost 8 days or more.

Discussion

To the authors' knowledge, this is the first published study to document the work experiences of U.S. youth attending middle schools. We found that 56% of South Texas middle school students reported working for pay. Occupational injury in middle schoolers is associated with high-intensity weekly work and is more likely to occur in particular jobs, such as agriculture, restaurant work, and construction work.

Prevalence

Our results revealed that one-fourth of South Texas middle school workers were injured on the job, and that almost one-third of the injured were hurt seriously enough to require medical attention. Forty-one percent of the injured lost 1 day to 1 week from school and 7% lost 8 days or more. Unfortunately, very little information about the work experiences of pre-high school youth is available in the scientific literature, making comparisons in the relevant age range unfeasible. Several studies reporting injuries among high school students, however, have been published; these studies reveal findings similar to those of our middle school sample. (5), (6), (8-14)

Perhaps most comparable methodologically to our middle school injury estimates, with the exception of the age differential, are the results of several recent surveys of high school students. These reports indicate that 17% to slightly more than 50% of high school students have been injured on the job. (5), (6), (26), (27) Seven to 16% of high school students were injured seriously enough to require medical attention. For example, at a large urban Massachusetts high school, 35% of 450 students reported having ever been injured at work; of the injured, 10% sought medical help. (27) Of 562 working teens contacted through a statewide North Carolina telephone survey in 1995, 54% reported being injured at least once while working for pay, 16% required medical help for their injuries, and 12% missed at least 1 day of work or school. (6) In our South Texas high school sample, 20% of workers had ever sustained an occupational injury; of the injured, 27% were hurt badly enough to need medical attention (NF Weller, SH Kelder, SP Cooper, et al, unpublished data). The data for the latter study were collected contemporaneously with our middle school data. Noteworthy are the prevalences of injury and serious injury in the high school sample, estimates that are quite similar to those of our middle school sample.

Other sources of injury data also indicate that many teens are injured while working. (5) For instance, extrapolating from 742 teen work injuries reported to the Department of Labor and Industry over a 12-month period (1990-91), officials estimated that 2% of all working students in Minnesota had severe injuries resulting in normal activity limitations, more than 3 days of missed work, or permanent impairment. (26), (28) These and other findings clearly show that preteen and teen work injuries are common and sometimes serious, raising concerns about the health and safety of employed youth.

Weekly Work Intensity and Injury

Many published studies among older adolescents have shown that extensive employment during the school year has untoward consequences on the physical, mental, social, and behavioral health of teens (9), (15-20); therefore, we were interested in investigating the relationship between intense weekly work and occupational injury among our middle school students. Our findings indicate that middle school youth employed more than 20 hours weekly (high-intensity workers) were approximately 1.8 times more likely to have ever been injured at work as students working 1 to 10 hours weekly (P < 0.05); for those employed 11 to 20 hours (intermediate risk), no significant relationship was found (OR, 1.2; 95% CI, 0.87-1.6). Unfortunately, no other studies of preteen employment have examined the relation between weekly work intensity and work injury, making comparisons to age-appropriate populations impossible. However, our South Texas high school study found high-intensity weekly work to be associated with occupational injury, and in a small sample of employed New York State adolescents, an average of 19 hours of weekly work was significantly correlated with work injury (NF Weller, SH Kelder, SP Cooper, et al, unpublished data). (9) These findings prompt concerns about the health and safety of younger youth employed long hours weekly during the academic year.

A possible explanation for the greater risk of work injury in high-intensity middle school workers includes greater exposure time of those working longer hours weekly. Heavy work schedules of 21 or more hours of weekly work during the school year, in addition to a full academic load and a typical preteen social schedule, may also result in sleep deprivation, fatigue and, consequently, a greater risk of injury. The dual effect of moderate- to high-intensity weekly work and certain characteristics of middle school youth, such as their lack of work experience, their premature cognitive and developmental systems, their undeveloped judgment, their more immature psychosocial and emotional development, and the pressure of balancing the demands of school and work, could play a role in the incidence of work-related injury in youngsters employed longer hours weekly. (5), (7), (9) Research pertaining to the association of youth development and work-related injury is, unfortunately, unavailable; however, adult studies find that inexperience on the job contributes to occupational injuries. (29) It would not be surprising, then, if the inexperience of preadolescents and early adolescents at work turns out to be an important factor contributing to work-related injury among middle schoolers. These characteristics of preteens, not coincidentally, constitute the rationale behind federal and state labor prohibitions against formal employment in preadolescents and early adolescents until these youth have attained a more advanced age and, presumably, an enhanced maturational level. (5)

Gender and Work-related Injury

Our middle school results confirmed the findings of many earlier studies among older male adolescents: male adolescents are much more likely to be injured at work than female adolescents. Middle school boys were 50% more likely to have ever had an occupational injury than middle school girls. Numerous other investigations of youth employment patterns also find an injury rate for boys that is approximately double that of girls. (5), (8), (10), (12), (13), (30-33) South Texas high school male workers were also more likely to be injured at work than our high school female workers by a ratio of 1.5:1 (NF Weller, SH Kelder, SP Cooper, unpublished data).

These findings could partly be explained by the disparity in weekly work hours between the genders. Although the proportion of middle school boys and girls working 1 to 10 hours weekly was similar, the proportion of boys working 21 or more hours weekly was twice that of middle school girls. Thus, the increased exposure time of male high-intensity workers could account for a portion of the difference identified between male and female work-injury rates.

Different types of jobs worked by middle school boys and girls might also explain some of the gender difference in the incidence of injury in middle school youth. Middle school boys were more likely to work in occupations that are known to be hazardous for workers of any age, such as agriculture and construction, (5), (10), (23), (33-35) whereas middle school girls were more likely to work in childcare, the type of work least likely to result in injury in our middle school sample. Our middle school boys were also more frequently employed in yard work, restaurants, and retail stores, other types of jobs that are known to produce occupational injuries, especially in young employees. Numerous other studies of youth employment also find associations between gender, type of work, and work-related injury. (5), (10), (23), (33-35) These investigations note that adolescent boys frequently work in riskier work situations with greater exposure to work-related hazards than adolescent girls.

Male gender was also related to severity of injury in our middle school sample (univariate); boys were 70% more likely to have a serious occupational injury than girls. This finding supports the results of several other studies of older adolescent boys in which boys were approximately twice as likely to need medical attention for their work-related injuries as girls. (10), (13)

Race or Ethnicity and Severity of Injury

Noteworthy among our findings are racial and ethnic subgroup differences in the severity of work-related injury needing intervention by health care providers. We found that nonwhite Hispanics were at slightly lower risk of work-related injury than whites, but at higher risk of serious injury requiring medical attention than the white referent group. The types of jobs secured by Hispanic and white youth may partially explain these differences. One published study found that a greater percentage of minority than white youngsters are employed in retail work, including restaurants and grocery stores, industries in which older teenage workers are frequently hurt and sometimes killed (A Ahituv, M Tienda, VJ Hotz, unpublished data). (5), (23) Middle school Hispanics in our study were twice as likely as whites to work in restaurants, and restaurant work was the only type of employment predicting the need for medical attention for a work-related injury (multivariate regression). Although these explanations may partially account for observed differences, it is unlikely that differences in types of jobs and job assignments could fully account for the disparity observed in severity of injury among white and Hispanic middle school workers. However, data to more fully explain these findings are unavailable.

Common Types of Injuries

Common injuries among middle school workers include punctures/cuts/jabs and broken bones, followed by sprains/strains and shocks/burns. Punctures/cuts/jabs most commonly occurred during yard work and childcare, whereas broken bones were more likely to occur during yard and restaurant work. These types of injuries have been documented in multiple studies as common work injuries in older teens. (5)

Study Limitations

Several cautions should be discussed in interpreting study results: 1) data represent public school students only (alternative, private school students, and dropouts are not included); 2) injury experience was self-reported, possibly influenced by recall factors, and not possible to confirm due to the study's anonymous nature; 3) temporal issues in questionnaire items may have caused inconsistent responses (hours of work may have been reported for the current job, whereas injury may have occurred during a former job); 4) estimates of injury frequency may be biased due to inadequate data on the nature, amount, and seasonal variability of work; 5) thresholds for weekly work hours are limited to the categories used in these analyses; 6) a substantial number of cases with missing data were excluded from the analyses; and 7) generalizability of the study is limited to similar populations of middle school students, a portion of whom are low-income Hispanics living in rural geographic areas.

Conclusion

A major contribution of this study is its description of the prevalence of injury in a previously undescribed population of middle school youth, especially our finding that increased weekly work hours are linearly associated with injury. An important addition to the literature is our finding that Hispanic middle schoolers are slightly more protected for work-related injury compared with whites, but are at greater risk of more serious injury than whites. We substantiated the findings of other work injury studies among older youth, including the South Texas high school study, regarding the prevalence of youth injury and several demographic subgroup differences within that prevalence. However, we found no evidence that low-income, Hispanic youth were at greater injury risk than middle schoolers with better financial standing.

The paucity of other reports in the pediatric literature indicates that researchers and public officials have largely neglected employment among very young students. Parents and professionals should closely monitor the work circumstances of middle school youth, and health/safety interventions in schools and other venues should be targeted toward preventing injuries in these most vulnerable workers. (36)

Table 1. Demographic characteristics of South Texas middle school
students by work status

                                          Percentage who  Percentage who
                          Percentage who   work 1-10 h     work 11-20 h
                           do not work        weekly          weekly
                           (n = 2,359)     (n = 2,382)      (n = 365)

Total                          44.0            44.4            6.8

Gender
  Male                         38.0            46.9            8.5
  Female                       49.7            41.9            5.2

Grade
  Sixth                        45.1            45.2            6.1
  Seventh                      43.5            45.0            6.9
  Eighth                       43.5            43.1            7.3

Race
  White                        34.2            53.0            7.4
  Hispanic                     49.8            39.6            6.5

Parents' education level
  College graduate             41.6            46.5            6.3
  High school graduate         45.2            43.9            7.1
  Less than high school        45.7            42.4            7.1

                          Percentage who                 Percentage
                            work 21+ h      Mean (SD)    injured at
                              weekly         weekly         work
                            (n = 261)      work hours    (n = 773)

Total                          4.9          7.7 (7.5)         25.1

Gender
  Male                         6.6          8.3 (8.1) (a)     31.3
  Female                       3.2          7.0 (6.7)         18.1

Grade
  Sixth                        3.6          6.9 (7.0) (a)     26.1
  Seventh                      4.6          7.6 (7.5)         25.4
  Eighth                       6.1          8.4 (8.1)         24.0

Race
  White                        5.4          7.3 (7.4)         25.6
  Hispanic                     4.1          7.7 (7.3)         23.2

Parents' education level
  College graduate             5.5          7.6 (7.7)         25.8
  High school graduate         3.8          7.4 (7.0)         24.5
  Less than high school        4.8          8.0 (7.6)         24.6

(a) P < 0.001.

Table 2. Association between type of job, work intensity,
sociodemographic factors, and work-related injury, South Texas middle
school workers (a)

                                  Univariate  (95% confidence
                                  odds ratio    interval)

Kind of job
  Baby-sitting (referent)            1.0
  Restaurant                         3.8         (2.7-5.4)
  Retail                             2.3         (1.6-3.3)
  Yard work                          1.7         (1.4-2.2)
  Construction                       3.6         (2.4-5.2)
  Farm/ranch                         4.4         (3.3-6.0)

Weekly hours worked
  1-10 (referent)                    1.0
  11-20                              1.5         (1.1-1.9)
  21+                                2.4         (1.8-3.2)

Gender
  Female (referent)                  1.0
  Male                               2.2         (1.8-2.6)

Grade
  Sixth (referent)                   1.0
  Seventh                            1.0        (0.82-1.3)
  Eighth                             0.93       (0.75-1.2)

Race/ethnicity
  White (referent)                   1.0
  Hispanic                           0.9        (0.71-1.0)

Parent education level
  College graduate (referent)        1.0
  High school graduate               0.9        (0.73-1.2)
  Less than high school graduate     0.97       (0.80-1.2)

                                   Adjusted   (95% confidence
                                  odds ratio     interval)

Kind of job
  Baby-sitting (referent)            1.0
  Restaurant                         3.2         (2.1-4.9)
  Retail                             1.7         (1.1-2.6)
  Yard work                          1.3        (0.93-1.9)
  Construction                       2.4         (1.5-3.8)
  Farm/ranch                         3.3         (2.3-4.9)

Weekly hours worked
  1-10 (referent)                    1.0
  11-20                              1.2        (0.87-1.6)
  21+                                1.8         (1.3-2.6)

Gender
  Female (referent)                  1.0
  Male                               1.5         (1.1-2.0)

Grade
  Sixth (referent)                   1.0
  Seventh                            0.96       (0.74-1.2)
  Eighth                             0.90       (0.61-1.2)

Race/ethnicity
  White (referent)                   1.0
  Hispanic                           0.8        (0.61-0.93)

Parent education level
  College graduate (referent)
  High school graduate
  Less than high school graduate

(a) Covariate adjustments were made for gender, grade, and
race/ethnicity. Parent education level was removed from the model.

Table 3. The association between type of job, work intensity level, and
the sociodemographic factors for work-related injuries requiring medical
attention, South Texas middle school workers (a), (b)

                         Univariate   (95% CI)     Adjusted    (95% CI)
                         odds ratio               odds ratio

Kind of job

  Baby-sitting              1.0                      1.0
  (referent)

  Restaurant                5.5       (2.9-10.6)     3.6       (1.7-7.8)

  Retail                    2.8       (1.4-5.7)      1.8      (0.77-4.2)

  Yard work                 1.5      (0.88-2.7)      1.2      (0.58-2.4)

  Construction              2.3       (1.2-4.5)      1.4      (0.59-3.3)

  Farm/ranch                2.4       (1.3-4.4)      2.0      (0.90-4.1)

Weekly hours worked

  1-10 (referent)           1.0                      1.0

  11-20                     1.7       (1.1-2.6)      1.3      (0.76-2.3)

  21+                       1.8       (1.1-2.7)      1.5      (0.83-2.6)

Gender

  Female (referent)         1.0

  Male                      1.7       (1.2-2.4)      1.5      (0.87-2.4)

Grade

  Sixth (referent)          1.0

  Seventh                   1.1      (0.74-1.6)

  Eighth                    1.0      (0.68-1.5)

Race/ethnicity

  White (referent)          1.0                      1.0

  Hispanic                  1.8       (1.2-2.2)      1.6       (1.1-2.4)

Parent education level

  College graduate          1.0
    (referent)

  High school graduate      0.85     (0.55-1.3)

  Less than high school     1.2      (0.85-1.7)
    graduate

(a) CI, confidence interval.
(b) Covariate adjustments were made for gender and race/ethnicity. Grade
and parent education level were removed from the model.

Table 4. Type of employment by type of injury sustained while working,
South Texas middle school workers

                   Type of injury, No. (%)
Kind of job     Bruise    Broken bone     Burn     Dislocation

Childcare      39 (5.0)     16 (2.1)    17 (2.2)       8 (1.0)
Restaurant      8 (1.0)     29 (3.8)    25 (3.2)      12 (1.6)
Retail          5 (0.6)     16 (2.1)    24 (3.1)       8 (1.0)
Yard work      31 (4.0)     39 (5.0)    28 (3.6)      20 (2.6)
Construction    9 (1.2)     14 (1.8)     7 (0.9)      10 (1.3)
Farm/ranch     11 (1.4)     22 (2.8)    14 (1.8)      13 (1.7)
Total         103 (13.3)   136 (17.6)  115 (14.9)     71 (9.2)

                   Type of injury, No. (%)
Kind of job   Amputation   Puncture     Sprain      Total

Childcare        0 (0.0)   38 (4.9)    33 (4.3)   151 (19.5)
Restaurant       7 (0.9)   15 (1.9)     3 (0.4)    99 (12.8)
Retail           8 (1.0)   10 (1.3)     3 (0.4)    74 (9.6)
Yard work       13 (1.7)   68 (8.8)    50 (6.5)   249 (32.2)
Construction     9 (1.2)   13 (1.7)    13 (1.7)    75 (9.7)
Farm/ranch      11 (1.4)   26 (3.4)    28 (3.6)   125 (16.2)
Total           48 (6.2)  170 (22.0)  130 (16.8)  773 (100.0)

Accepted April 23, 2003.

References

(1.) Texas Workforce Commission. Employment of children. 51 Tex. Labor Code subtitle B, subchapter B, [section] 51.011 (Acts 1993, 73rd Leg, ch 269, [section] 1, effective September 1, 1993).

(2.) U.S. Department of Labor, Employment Standards Administration, Wage and Hour Division. Child Labor Requirements in Nonagricultural Occupations Under the Fair Labor Standards Act. Washington, DC, U.S. Department of Labor, 1990.

(3.) Bellamy C (ed): United Nations Children's Fund (UNICEF). The State of the World's Children 1997. New York, Oxford University Press, 1997.

(4.) Steinberg L, Cauffman E. The impact of employment on adolescent development, in Vasta R (ed): Annals of Child Development. London, Jessica Kingsley Publishers, 1995, vol 11, pp 131-166.

(5.) Committee on the Health and Safety Implications of Child Labor, Board on Children, Youth, and Families, Commission on Behavioral and Social Sciences and Education, National Research Council, Institute of Medicine. Protecting Youth at Work: Health, Safety, and Development of Working Children and Adolescents in the United States. Washington, DC, National Academy Press, 1998.

(6.) Dunn KA, Runyan CW, Cohen LR, et al. Teens at work: A statewide study of jobs, hazards, and injuries. J Adolesc Health 1998;22:19-25.

(7.) Evensen CT, Schulman MD, Runyan CW, et al. The downside of adolescent employment: Hazards and injuries among working teens in North Carolina. J Adolesc 2000;23:545-560.

(8.) Miller M. Occupational Injuries among Adolescents in Washington State, 1988-1991: A Review of Workers' Compensation Data (Tech. Rep. 35-1-1995). Olympia, WA, Safety and Health Assessment and Research for Prevention, Washington State Department of Labor and Industries, 1995.

(9.) Frone MR. Predictors of work injuries among employed adolescents. J Appl Psychol 1998;83:565-576.

(10.) Runyan CW, Zakocs RC. Epidemiology and prevention of injuries among adolescent workers in the United States. Annu Rev Public Health 2000;21:247-269.

(11.) Castillo DN, Landen DD, Layne LA. Occupational injury deaths of 16-and 17-year-olds in the United States. Am J Public Health 1994;84:646-649.

(12.) Coleman PJ, Sanderson LM. Surveillance of occupational injuries treated in hospital emergency rooms: United States, 1982, Morb Mortal Wkly Rep Surveill Summ 1983;32(2):31SS-37SS.

(13.) Cooper SP, Burau KD, Robison TB, et al. Adolescent occupational injuries: Texas, 1990-1996. Am J Ind Med 1999;35:43-50.

(14.) Cooper SP, Rothstein MA. Health hazards among working children in Texas. South Med J 1995;88:550-554.

(15.) Steinberg L, Dornbusch SM. Negative correlates of part-time employment during adolescence: Replication and elaboration. Dev Psychol 1991;27:304-313.

(16.) Steinberg L, Fegley S, Dornbusch SM. Negative impact of part-time work on adolescent adjustment: Evidence from a longitudinal study. Dev Psychol 1993;29:171-180.

(17.) Greenberger E, Steinberg L. When Teenagers Work: The Psychological and Social Costs of Adolescent Employment. New York, Basic Books, 1986.

(18.) Rowlett JD, Fairchild VR, Rowlett JD. Adolescents and the workplace. Adolesc Med 2001;12:539-550.

(19.) Bachman JG, Schulenberg J. How part-time work intensity relates to drug use, problem behavior, time use, and satisfaction among high school seniors: Are these consequences or merely correlates? Dev Psychol 1993;29:220-235.

(20.) Bachman J, Schulenberg J. Part-time Work by High School Seniors: Sorting Out Correlates and Possible Consequences (Monitoring the Future Occasional Paper 32). Ann Arbor, MI, Institute for Social Research, 1991.

(21.) U.S. General Accounting Office. Child labor: Characteristics of Working Children (Publication GAO/HRD-91-83BR). Washington, DC, U.S. General Accounting Office, June 1991.

(22.) National Safe Workplace Institute. Sacrificing America's Youth: The Problem of Child Labor and the Response of Government. Chicago, National Safe Workplace Institute, 1992.

(23.) Children's Safety Network National Injury and Violence Prevention Resource Center. Protecting Working Teens: A Public Health Resource Guide. Newton, MA, Children's Safety Network, Education Development Center, Inc., 1995.

(24.) Texas Education Agency. Snapshot '93: 1992-1993 School District Profiles (Doc. GE417001). Austin, TX, Texas Education Agency, 1993.

(25.) Brener ND, Collins JL, Kann L, et al. Reliability of the Youth Risk Behavior Survey Questionnaire. Am J Epidemiol 1995;141:575-580.

(26.) Parker DL, Carl WR, French LR, et al. Nature and incidence of self-reported adolescent work injury in Minnesota. Am J Ind Med 1994;26:529-541.

(27.) Massachusetts Department of Public Health. Newsletter from the Teens at Work: Injury Surveillance and Prevention Project. Boston, Occupational Health Surveillance Program, Massachusetts Department of Public Health, 1998, pp 1-8.

(28.) Parker DL, Clay RL, Mandel JH, et al. Adolescent occupational injuries in Minnesota: A descriptive study. Minn Med 1991;74:25-28.

(29.) Mueller BA, Mohr DL, Rice JC, et al. Factors affecting individual injury experience among petroleum drilling workers. J Occup Med 1987;29:126-131.

(30.) Brooks DR, Davis LK, Gallagher SS. Work-related injuries among Massachusetts children: A study based on emergency department data. Am J Ind Med 1993;24:313-324.

(31.) Schober SE, Handke JL, Halperin WE, et al. Work-related injuries in minors. Am J Ind Med 1988;14:585-595.

(32.) Layne LA, Castillo DN, Stout N, et al. Adolescent occupational injuries requiring hospital emergency department treatment: A nationally representative sample. Am J Public Health 1994;84:657-660.

(33.) Parker DL, Merchant D, Munshi K. Adolescent work patterns and work-related injury incidence in rural Minnesota. Am J Ind Med 2002;42:134-141.

(34.) Reskin B. Sex segregation in the workplace. Annu Rev Sociol 1993;19:241-270.

(35.) American Academy of Pediatrics, Committee on Injury and Poison Prevention and Committee on Community Health Services. Prevention of agricultural injuries among children and adolescents. Pediatrics 2001;108:1016-1019.

(36.) Bowling JM. Training to reduce hazards in youth work: Results from the Brockton survey of teenage occupational safety. Presented at the National Occupational Injury Research Symposium, National Institute for Occupational Health and Safety, October 16, 1997, Morgantown, WV.

RELATED ARTICLE: Key Points

* Fifty-six percent of South Texas middle school students reported working for pay.

* South Texas middle school students reported working an average of 8 hours weekly.

* One-fourth of South Texas middle school students reported a work-related injury.

* Working more than 20 hours weekly increased the likelihood of injury.

Nancy F. Weller, BSN, DRPH, Sharon P. Cooper, PHD, Susan R. Tortolero, PHD, Steven H. Kelder, PHD, MPH, and Sohela Hassan, PHD

From the Center for Health Promotion and Prevention Research, School of Public Health, and the Mental Sciences Institute, University of Texas Health Science Center at Houston, Houston, TX, and the School of Rural Public Health, Texas A&M University, Bryan, TX.

Reprint requests to Nancy F. Weller, DrPH, Department of Family and Community Medicine, Baylor College of Medicine, 3701 Kirby, Suite 600, Hoston, TX 77098. Email: [e-mail]

Copyright [c] 2003 by The Southern Medical Association 0038-4348/03/9612-1213