|
 |
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2007 American Public Health Association
DOI: 10.2105/AJPH.2005.077172
RESEARCH AND PRACTICE |
J. R. Wilkins III is with the Division of Epidemiology, School of Public Health, Ohio State University, Columbus. J. Mac Crawford is with the Division of Environmental Health Sciences, School of Public Health, Ohio State University, Columbus. Lorann Stallones is with the Department of Psychology, Colorado State University, Fort Collins. At the time of the study, Kathleen M. Koechlin was with the Division of Epidemiology, School of Public Health, Ohio State University, Columbus. Lei Shen is with the Division of Biostatistics, School of Public Health, Ohio State University, Columbus. John Hayes is with the Department of Pediatrics, Columbus Children’s Hospital, Columbus. Thomas L. Bean is with the Department of Food, Agricultural, and Biological Engineering, Ohio State University, Columbus.
Correspondence: Requests for reprints should be sent to J. R. Wilkins III, DrPH, Division of Epidemiology, School of Public Health, The Ohio State University, 320 W Tenth Ave, Columbus, OH 43210 (e-mail: wilkins.2{at}osu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Objectives. We conducted a 3-year cohort study of 407 youths aged 9 to 18 years to develop multivariable risk prediction models of agriculture-related injuries.
Methods. Data were obtained via participant event monitoring, with youths self-reporting injuries and exposures in daily diaries over a 13-week period. We evaluated data quality by comparing injury self-reports with other injury data.
Results. Semilogarithmic plots of rates of all unintentional injuries combined (US data from 2000) as well as of agriculture-related injuries (US and Canadian data from 19 previous studies) graphed as a function of injury severity exhibited linearity, as did plots based on the present results. Severity-specific unintentional injury rates were 1.4- to 4.3-times higher than national rates, suggesting that our methodology can significantly reduce injury underreporting. In addition, at each severity level, estimated agriculture-related injury rates were 5.8- to 9.3-times higher than rates from previous national, regional, and state-based studies.
Conclusions. Our approach to participant event monitoring can be implemented with youths aged 9 to 18 years and will yield reliable daily data on unintentional injuries.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Unintentional injuries are a significant childhood health problem in the United States1,2; for example, in 2001 such injuries accounted for almost 45% of all deaths in the population aged 1 to 19 years.3,4 Unintentional injury is the leading cause of death in the 1- to 34-year age group4 and has been for some time. Of the 30 million nonfatal injuries treated in US hospital emergency departments in 2001, 35% involved individuals aged younger than 20 years.5
A significant methodological problem in injury epidemiology is collecting accurate data on events causing tissue damage as well as the tissue damage itself.6,7 Other factors that complicate this process include the varieties of injury-producing events (e.g., slips, trips, falls), the multidimensionality of injury severity, and the question of what constitutes a reportable event. In studies of unintentional injuries, the reporting threshold has commonly been defined in terms of the nature of medical attention or the degree of interference with normal activities. Combined with the relative rarity of high-severity injuries, the ascertainment problem hampers accurate estimation of injury rates and identification of important risk indicators.
It has been suggested that the reporting threshold be lowered, when appropriate, to include minor injuries, which would increase the number of injury events available for study, and that such data be collected on a weekly or even daily basis, which would shorten the recall period and thereby reduce underreporting and misclassification.8–11 Some have argued that minor injuries may serve as proxies for more severe injuries.7
We conducted a study to develop multivariable risk prediction models of work-related injuries among young people aged 9 to 18 years who were exposed to agricultural hazards.12 We report on the quality of data obtained from youths who completed daily diaries during a 13-week reporting period.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Identifying and Recruiting Youths
We targeted young people aged 9 to 18 years residing in the 20-county central Ohio area during 1999 through 2001 who lived or worked on farms or performed agriculture-related chores as part of 4-H (the youth development program of the US Department of Agriculture’s Cooperative Extension System13). Youths and their "parent partners" (their primary caregivers who usually made the agriculture-related chore assignments) were recruited via letters and follow-up telephone calls. This effort proceeded serially, 1 county at a time, and relied on each county’s 4-H infrastructure to identify eligible youths.
Data Collection Procedures
We obtained data from 3 primary sources: (1) prebaseline self-administered questionnaires completed by young people and their parent partners; (2) baseline measurements of known and suspected injury risk factors, including neurobehavioral and anthropometric characteristics of the participating youths (not discussed further but available from the authors); and (3) semistructured diaries in which youths reported "accidents" and injury hazard exposures on a daily basis over a 13-week period. Self-administered questionnaires were designed to obtain data on family demographic and farm or household characteristics, risk attitudes and behaviors, and parenting behaviors in addition to several other youth-related factors (e.g., injury and agricultural work histories, athletic ability, and visual acuity). Most items on the questionnaires were derived from previously used and validated instruments.14,15
Self-administered questionnaires were mailed to participating households before testing to obtain baseline measurements, with the expressed expectation that the questionnaires be completed by the day of testing. Study staff reviewed the returned questionnaires during the testing session and re-examined them later to detect problems not discerned at the testing site, i.e. item nonresponse or illogical or nonsensical responses. Appropriate corrections were made, or, if necessary, a telephone call was made to the child or caregiver to resolve problems. Self-administered questionnaires were coded for data entry at the testing site; this coding was also rechecked.
Daily record books (DRBs) were designed to gather injury and injury hazard exposure data from participating youths on a daily basis. Each DRB contained 7 diaries, 1 for each day of the week. Youths were instructed to complete the daily diary section of the DRB each evening before going to bed and then at the end of each week give each completed DRB to their parent partner to review for accuracy and completeness before mailing it to the project office. Most questions were closed-ended, the exceptions being exposure-related items requiring duration reporting and injury-related items pertaining to bystanders and activity and location at the time of injury.
We considered traditional reporting thresholds too restrictive given the project’s objectives, so the operational definition of a reportable event we used was a modification of the Peterson et al.11 definition of a minor injury: any event with a specific time of onset that leaves a mark for at least 1 hour or that results in pain for at least 15 minutes, namely, any "accident" that caused pain, bleeding, skin redness, or bruising. Because participants were young, we used the more familiar terms "accident" for the event and "accidental" for intent.
Injury events reported in the DRBs were coded according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM),16 and the Murphy et al. Farm and Agricultural Injury Code (FAIC) system.17 As a surrogate measure of severity, all injury episodes were assigned 1 of the following treatment dispositions: no treatment; minor treatment or first aid; treatment at a hospital, clinic, or doctor’s office followed by release; or treatment at a hospital, clinic, or doctor’s office followed by hospitalization.
In instances in which medical attention was sought for an injury, we requested access to the youth’s medical records to validate the youth’s description of the injury. We compared injury descriptions obtained from medical personnel with those provided by youths to determine levels of agreement.
Quality Control Procedures
Given the data collection modality, data quality was a major concern. Consequently, we implemented a multifaceted system of quality control procedures. The quality control efforts described subsequently focused on youths’ interactions with the microcomputer system used called Dr. GOOP (goal and object-oriented programming; designed by J. Hayes); injury reporting; and DRB follow-up.
Dr. GOOP is an interactive microcomputer system that allows automated telephone data collection. Data are collected using custom software with Intel voice boards (Intel, Santa Clara, Calif) having an analog telephone interface. Youths were asked to make a 1-minute toll-free telephone call to this "talking computer" on the final day of each project week. If a call was not made on schedule, Dr. GOOP called the household the following day. Youths were asked whether or not they were filling out their DRBs on a daily basis, whether or not their parent partner reviewed their DRBs for "accuracy and completeness," and whether or not they had forgotten to report any "accidents." The primary functions of Dr. GOOP were to keep participants engaged in the project and to contribute to quality control efforts.
As mentioned in the previous section, we assessed injury reporting concordance by obtaining medical records for injuries reported to have required medical attention. Using the medical record as the gold standard, we measured concordance with respect to type and anatomical site of injury, laterality, and day and date of injury.
We evaluated each DRB for accuracy and completeness within 10 days of receipt. Discernible errors that could not be resolved were usually rectified through a follow-up telephone call to the participant. When such calls were made, youths were again reminded about the correct way to complete DRBs.
Statistical Analysis
Descriptive statistics (frequency counts, percentages, means, and standard deviations) were calculated for injury events. We evaluated the comparability of the injury data we obtained with injury data collected via other methods by statistically and graphically contrasting the severity-specific injury rates estimable from our data with corresponding age-, race-, and severity-specific rates of all unintentional injuries available through the Web-Based Injury Statistics Query and Reporting System (WISQARS)18 and with corresponding age-, race-, and severity-specific rates of agriculture-related injuries among youths reported from relevant US and Canadian studies conducted over the past 20 years.19–37 For these previously reported agriculture-related rates, we fit a linear mixed-effects model with random effects to account for variation among studies.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Recruitment and Retention
During calendar years 1999 through 2001, 3152 households were contacted (55% of the age-eligible residents were girls). Among the 1926 eligible youth–parent dyads identified, 471 consented to participate, yielding an overall response rate of 24.5%. Recruitment effort outcomes (e.g., agreement or refusal to participate, ineligibility) did not vary according to (youth’s) age or gender.
Of the 471 initial responders, 64 contributed no time at risk of injury, because they dropped out early, failed to provide written consent, or failed to complete or return the required prebaseline questionnaires. Comparisons of these 64 youths not providing usable data with the 407 youths who provided such data showed that they were more likely to be boys and that they were slightly older. However, there were no differences between the 2 groups with respect to lifetime history of medically attended injuries or farm residence at the time of baseline testing.
The 169 boys (41.5% of the total) and 238 girls (58.5%) taking part were similar in their age distributions (boys, mean age = 13.1 years, SD= 2.3; girls, mean age = 12.8 years, SD= 2.4). Overall, participating youths completed and returned 4098 DRBs (i.e., 28686 days of injury and activity data). Approximately 56% of boys and girls completed and returned all 13 DRBs. Although the mean number of contributed "youth-weeks" was approximately equal among boys (10.2) and girls (10.0), some variation was observed according to age. The youngest boys were the most responsive, completing and returning 11.0 DRBs on average. The quality of both injury and activity data tended to improve over the 13-week follow-up period, with girls and older youths in general making the fewest errors in their DRBs.12,38
Frequency, Severity, and Nature of Injuries
The 407 participants providing usable data reported 2788 (unintentional) injury-producing events (Table 1
). Approximately two thirds of injuries required no treatment (n = 1888; 67.7%), whereas 27.9% (n = 778) required "minor" treatment (i.e., "first-aid" administered by the youth or a family member at home or by a nurse or teacher at school). Of the 31 events for which medical attention was sought (approximately 1% of the overall number of events), only 1 resulted in hospitalization. No fatal outcomes were reported.
|
Coding according to the FAIC system17 showed that approximately 50% of the injury events were agriculture related. Nonagricultural injury events accounted for about 40% of self-reported events, with approximately 10% considered unclassifiable. Most agriculture-related events occurred while youths were involved in farm production activities (n = 775; 27.8% of all injuries, 55.4% of all agriculture-related injuries). Approximately 13% of all injuries occurred in or around the "farm home," which, it should be emphasized, is a place of residence as well as a part of an agricultural setting where work occurs.
Treatment received varied little according to FAIC category. Girls, as mentioned, made up 58.5% of the sample, but they reported 70.9% of all injuries; they were less likely than boys to report agriculture-related injuries. In general, injury rates increased with age; agriculture-related injury rates were highest among 15- to 18-year-old youths (by a factor of 2 for boys and 1.6 for girls relative to the other age groups).
ICD-9-CM diagnosis codes were assigned to the 3352 injuries produced by the 2788 injury events (2396 events caused 1 injury, and 392 events caused 2 or more). In more than 80% of cases (Table 2), tissue damage was described as superficial (37.6%), as a contusion with intact skin surface (25.4%), or as an open wound (19.0%). In general, youths’ age at the time of the event varied little according to ICD-9-CM code or gender, although boys reporting superficial injuries (ICD codes 910–919), fractures (codes 800–829), and unspecified injuries (codes 958 and 959) were significantly older on average (P < .05) than were girls reporting the same injuries.
|
Log-Transformed Incidence Rates
In Figure 1a
, data from WISQARS (rates of all unintentional injuries) were plotted as small open circles (at injury severity levels 2–4); the small open circle plotted at severity level 1 was the minor injury rate reported by Peterson et al. (for 61 second-grade children).11 Severity-specific rates estimated from our PEM-derived data were also plotted in Figure 1a (dashed linear trend line). Not only are the 2 trend lines approximately parallel, but our PEM-derived rates are 1.4- to 4.3-times higher than national rates (see nonoverlapping confidence intervals for severity levels 1 and 2 in Table 3).
|
|
The small numbers of reported injuries at severity levels 3 and 4 can be interpreted by assuming that the Poisson distribution applies. One severity level 3 injury was observed, the likelihood of which would be only 15% if the true rate were the same as that from WISQARS. No fatal injuries were observed, which is not surprising given the low WISQARS-based rate of 0.02 per 100. Therefore, the present data are compatible with the conclusion that our PEM-derived rates were significantly higher than the WISQARS-based rates.
In Figure 1b
, the same parallelism is demonstrated for agriculture-related rates. As can be seen in Table 3, the agriculture-related rates estimated from our PEM methodology were 5.8- to 9.3-times higher than the rates reported in the relevant US and Canadian studies.19–37 Because these studies typically provided severity-specific data, we fit a linear mixed-effects model to obtain better rate estimates and used random effects to account for variation among studies.
In our analysis, we log-transformed our ratio estimates to stabilize variances, and the resulting estimated agriculture-related injury rates for the studies assessed were 1.84, 0.14, and 0.006 per 100 for severity levels 2, 3, and 4, respectively. Our estimate for severity 2 injuries was higher than all of those observed in the other studies and was 7.8-times higher than the combined estimate from those studies, a statistically significant difference. One severity level 3 injury was observed in this study, an event with only an 11% probability if the true injury rate were the same as the combined estimate from the other studies (0.140 per 100), suggesting an elevated rate in the present study.
Youths’ Interactions With Dr. GOOP
Overall, the results of youths’ interactions with Dr. GOOP suggested that, as intended, this system was successful in keeping participants engaged in the data collection process. In more than 73% of cases, youths initiated contact with Dr. GOOP. Otherwise, Dr. GOOP called and the youth answered or Dr. GOOP called, someone in the household answered the phone, and the youth called Dr. GOOP shortly thereafter.
An increasing trend was seen from year 1 to year 3 in the percentages of youths reporting that they were completing their DRBs each day (83.4%, 89.0%, and 90.6%, respectively). Furthermore, a decreasing trend was seen among youths reporting they had forgotten to record any "accidents" in a given week. In year 1, 12.1% of youths who answered the question about whether or not they had forgotten any accidents reported that they had done so. In year 2 this percentage dropped to 6.5%, and in year 3 only 4.9% of youths reported that they had forgotten to report any accidents. More than 80% of the time (over all 3 years), youths reported that their parent partners had checked their DRBs for accuracy and completeness.
Injury Reporting Concordance
The overall concordance between youths’ self-reports and medical records was 67.2% (43 of 64 cases). The highest concordance rate was that for the anatomical site of the injury (88.2%; 15 of 17 cases); the lowest was that for laterality (52.9%; 9 of 17 cases). The poor laterality concordance may have been a result of the anatomical diagram in the DRB (the diagram’s left was probably interpreted as right, and vice versa). Concordance estimates for injury type and date were 66.7% (10 of 15 cases) and 60.0% (9 of 15 cases), respectively.
Follow-Up
Approximately 45% of the 407 participants required 1 or more DRB follow-up telephone call (n = 181). A total of 547 calls were made (54.5% concerned injuries, and 45.5% concerned activities). Young people who did and did not require follow-up calls were similar with respect to age and gender. Fewer than 10 follow-up calls for self-administered questionnaires were necessary.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
The results shown in Figure 1
and the outcomes of our quality control subanalyses indicate that our approach to PEM can be successfully implemented to collect reliable daily data on unintentional injuries from 9- to 18-year-old youths over a 13-week reporting period. Furthermore, our PEM methodology appeared to significantly reduce injury under-reporting. As can be seen in Table 3, rates estimated from our data were 1.4- to 4.3-times higher than were national rates for all unintentional injuries combined and 5.8- to 9.3-times higher than were aggregated rates for agriculture-related injuries found in previous studies. Although not all injury events were reported by all participants, we agree with Peterson et al. that PEM-derived data "seem to provide a much better estimate of injury frequency than could be obtained in any other fashion."11(p129) However, the observed rate differences might be explained by factors other than underreporting in the comparison populations. For example, although our 4-H youth participants reported exposures similar to those of other youths living or working on farms,39 the higher rates exhibited by these young people might reflect real elevated risks of both unintentional and agriculture-related injuries. It is unlikely that the differences are explained by overreporting of injury events among our participants.
Characteristics of childhood injuries such as multiple, intermittent, and heterogeneous exposures make PEM an attractive data collection methodology. In addition to a significant increase in the number of injury events available for analysis, the time elapsed between injury event and injury event reporting is minimized. Because accuracy of recall of past experiences decreases over time, self-reporting of injury-producing events and injuries themselves has been viewed as problematic.40–43 Nevertheless, accurate self-reporting methods for dietary intake have been developed because obtaining such data through direct observations would be prohibitively expensive.44
Many experts believe, as an example, that valid data on young people’s behaviors can be obtained from parental observations without the need for trained external observers.45 Consequently, our approach to designing daily data collection forms was guided by the methods advocated by Peterson et al.11 and incorporated a pair of approaches shown to have good reliability and validity in nutritional studies: the checklist approach and the diary approach.46,47 As noted earlier, data on interactions with Dr. GOOP indicated that most youths self-reported on most days, suggesting a relatively small adverse effect of recall bias in comparison with other studies.
Although there are contrary views,48 the potential utility of data on minor injuries should not be underestimated. Factors explaining the occurrence of minor injuries are to some extent correlated with the factors that explain the occurrence of serious injuries (i.e., minor injuries may serve as reasonable proxies for serious injuries). Morrongiello et al.49 reported a significant correlation between minor and serious injuries, although in their study, mothers reported on the injury experiences of their 2- and 3-year-old children. Furthermore, minor injuries merit study because of the postevent stress children and adolescents may experience and because such injuries have a significant public health impact in terms of quality-adjusted life-years.9–11,50
There are other advantages of our PEM methodology. Because youths reported their injury and exposure experiences on a daily basis, we were able to obtain relatively more accurate injury and exposure time data than would be available with other methods, allowing for better estimates of injury risks. Also, the diary method is less intrusive than telephone-based techniques.8 Telephone interviews require participants to be available at the time of the call or to return a call, which can be disruptive to a household’s routine and labor intensive, with repeated contact attempts necessary. With PEM, youths were allowed some flexibility in terms of when they could fill out their DRBs.
Given the nature of our data collection methodology, it must be acknowledged that there were several potential sources of error. In the DRB alone, youths were expected to respond to approximately 60 activity items each day, along with 12 items for each injury event. Although every precaution was taken to standardize baseline testing through intensive training, we recognize that there may have been variability between examiners in protocol implementation given that multiple staff members were responsible for testing.
Other disadvantages include the large inputs of time and labor required to check the DRBs for accuracy and completeness. Because approximately 29000 days of activity data were collected, staff reviewed, recorded errors on, coded, and entered data from approximately 29000 single-page forms. Approximately 3000 injuries (with 12 response items per injury) were coded and entered into a database, and telephone calls were made to youths whose errors could not be unambiguously corrected. We recognize that our PEM approach, as implemented here, may not be widely feasible because of its relatively high costs.
Finally, the potential lack of generalizability of our approach is an issue given the relatively low response rate and the nonrandom nature of our sample. We did find that responders and nonresponders were virtually identical with respect to age and gender. Furthermore, the decision to recruit 4-H youths was based on certain characteristics of the 4-H infrastructure.
First, a large number of Ohio children and adolescents with potential exposures to agricultural hazards participate in the state’s 4-H Youth Development program. More than 200 000 young people ranging in age from 5 to 19 years are involved in different activities sponsored by 4-H. Of these youths, approximately 33 000 live on farms. Second, 4-H participants meet regularly in small, organized groups, permitting regular contact with investigators.
Third, each 4-H club has at least one adult volunteer who is committed to youth development and strategically positioned to facilitate recruitment of young people and their primary caregivers. Fourth, a traditional component of youth participation in 4-H is investment of time and effort in "projects" (e.g., raising an animal to show at a county fair) that often require record keeping over a 3- to 12-month period. Finally, as reflected in the comments of the 4-H advisors who participated in a planning focus group, there is a strong expectation that all projects be completed.
The American Academy of Pediatrics recently published recommendations for the prevention of agricultural injuries among children and adolescents, citing the following sobering statistics: each year in the United States there are 104 deaths, 22 000 emergency department visits, and 78000 injuries not treated in emergency departments in this population.51 These troubling facts underscore the need for better injury and exposure data collection methodologies. Although we focused on the problem of childhood agriculture-related injuries in our study, we believe that the lessons learned can be used to guide future research efforts designed to prevent all types of unintentional childhood injuries.
|
Acknowledgments |
|---|
This research was supported by the National Institute for Occupational Safety and Health (grant R01 CCR515580).
We thank Barbara Morrongiello, David Schwebel, and Huiyun Xiang for their thoughtful comments on earlier versions of this article. We also acknowledge the invaluable assistance of the late Lizette Peterson-Homer in the early conceptualization of the study. This article is dedicated to her memory.
Human Participant Protection
All protocols pertaining to human participants, including the informed consent process, were approved by the institutional review board of Ohio State University.
|
Footnotes |
|---|
Peer Reviewed
Contributors
J. R. Wilkins III originated the study, conducted the data analysis, wrote the article, and supervised all aspects of the study’s implementation. J. M. Crawford contributed to conceptual development and oversaw data collection and quality control efforts. L. Stallones contributed to conceptual development and helped write the article. K. M. Koechlin participated in data collection and analysis. L. Shen contributed to the data analysis. J. Hayes developed the telephone-based data collection methodology. T.L. Bean contributed to supervision of field personnel. All of the authors reviewed and edited drafts of the article and helped interpret the findings.
Accepted for publication February 3, 2006.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
1. Miller TR, Romano EO, Spicer RS. The cost of childhood unintentional injuries and the value of prevention. Future Child.2000;10:137–163.[CrossRef][Web of Science][Medline]
2. Crawley-Coha T. Childhood injury: a status report. J Pediatr Nurs.2001;16:371–374.[CrossRef][Medline]
3. Anderson RN, Miniño AM, Fingerhut LA, Warner M, Heinen MA. Deaths: injuries, 2001. Natl Vital Stat Rep. June 2003;52(21).
4. WISQARS Injury Mortality Report. Atlanta, Ga: National Center for Injury Prevention and Control; 2004.
5. WISQARS National Estimates of Nonfatal Injuries Treated in U.S. Hospital Emergency Departments. Atlanta, Ga: National Center for Injury Prevention and Control; 2004.
6. MacKenzie EJ. Epidemiology of injuries: current trends and future challenges. Epidemiol Rev.2000;22: 112–119.
7. Cummings P, Koepsell TD, Mueller BA. Methodological challenges in injury epidemiology and injury prevention research. Annu Rev Public Health.1995;16: 381–400.[CrossRef][Web of Science][Medline]
8. Schwebel DC, Binder SC, Plumert JM. Using an injury diary to describe the ecology of children’s daily injuries. J Safety Res.2002;33:301–319.[CrossRef][Web of Science][Medline]
9. Peterson L, Saldana L, Heiblum N. Quantifying tissue damage from childhood injury: the Minor Injury Severity Scale. J Pediatr Psychol.1996;21:251–267.
10. Peterson L, Heiblum N, Saldana L. Validation of the Minor Injury Severity Scale: expert and novice quantification of minor injury. Behav Ther.1996;27: 515–530.[CrossRef][Web of Science]
11. Peterson L, Brown D, Bartelstone J, Kern T. Methodological considerations in participant event monitoring of low-base-rate events in health psychology: children’s injuries as a model. Health Psychol.1996;15: 124–130.[CrossRef][Web of Science][Medline]
12. Wilkins JR III. Farm Youth Can Be Reliable Reporters of Their Daily Injury Experiences. Saskatoon, Saskatchewan, Canada: Future of Rural Peoples; 2003.
13. US Department of Agriculture, Cooperative State Research, Education, and Extension Service. National 4-H Headquarters [Web site]. Available at: http://www.national4-hheadquarters.gov. Accessed October 22, 2005.
14. National Center for Health Statistics. 1998 National Health Interview Survey (NHIS): questionnaires, datasets, and related documentation, 1997–2005. Available at: http://www.cdc.gov/nchs/about/major/nhis/quest_data_related_doc.htm. Accessed October 22, 2005.
15. National Center for Health Statistics. National Health and Nutrition Examination Survey (NHANES III), Series 11. Available at: http://www.cdc.gov/nchs/about/major/nhanes/nh3data.htm. Accessed October 25, 2005.
16. International Classification of Diseases, Ninth Revision, Clinical Modification. Hyattsville, Md: National Center for Health Statistics; 1980.
17. Murphy DJ, Purschwitz M, Mahoney BS, Hoskin AF. A proposed classification code for farm and agricultural injuries. Am J Public Health.1993;83:736–738.
18. National Center for Injury Prevention and Control. Web-Based Injury Statistics Query and Reporting System (WISQARS). Available at: http://www.cdc.gov/ncipc/wisqars. Accessed November 7, 2006.
19. Hoskin AF, Miller TA. Farm accident surveys: a 21-state summary with emphasis on animal-related injuries. J Safety Res.1979;11:2–13.[Medline]
20. Brison RJ, Pickett CWL. Nonfatal farm injuries in eastern Ontario: a retrospective survey. Accid Anal Prev.1991;23:585–594.[CrossRef][Web of Science][Medline]
21. Rivara FP. Fatal and nonfatal farm injuries to children and adolescents in the United States. Pediatrics.1985;76:567–573.
22. Gerberich SG, Gibson RW, French LR, et al. Injuries among children and youth in farm households: Regional Rural Injury Study I. Inj Prev.2001;7: 117–122.
23. Aherin RA, Todd CM. Accident risk taking behavior and injury experience of farm youth. 1989. Paper presented at: Annual Meeting of the American Society of Agricultural Engineers; December 1989; New Orleans, LA.
24. Hopkins RS. Farm equipment injuries in a rural county, 1980 through 1985: the emergency department as a source of data for prevention. Ann Emerg Med.1989;18:758–762.[CrossRef][Web of Science][Medline]
25. Salmi LR, Weiss HB, Peterson PL, Spengler RF, Sattin RW, Anderson HA. Fatal farm injuries among young children. Pediatrics.1989;83:267–271.
26. Layne LA, Castillo DN, Stout N, Cutlip P. Adolescent occupational injuries requiring hospital emergency department treatment: a nationally representative sample. Am J Public Health.1994;84:657–660.
27. Nordstrom DL, Layde PM, Olson KA, Stueland D, Brand L, Follen MA. Incidence of farm-work-related acute injury in a defined population. Am J Ind Med.1995;28:551–564.[Web of Science][Medline]
28. Schenker MB, Lopez R, Wintemute G. Farm-related fatalities among children in California, 1980 to 1989. Am J Public Health.1995;85:89–92.
29. Centers for Disease Control and Prevention. Youth agricultural work-related injuries treated in emergency departments—United States, October 1995–September 1997. MMWR Morb Mortal Wkly Rep.1998;47: 733–737.[Medline]
30. Rivara FP. Fatal and non-fatal farm injuries to children and adolescents in the United States, 1990–3. Inj Prev.1997;3:190–194.
31. Stueland DT, Lee BC, Nordstrom DL, Layde PM, Wittman LM. A population based case-control study of agricultural injuries in children. Inj Prev.1996;2: 192–196.
32. Myers JR, Hendricks KJ. Injuries Among Youth on Farms in the United States, 1998. Cincinnati, Ohio: National Institute for Occupational Safety and Health; 2001. NIOSH publication 2001-154.
33. Pickett W, Brison RJ, Niezgoda H, Chipman ML. Nonfatal farm injuries in Ontario: a population-based survey. Accid Anal Prev.1995;27:425–433.[CrossRef][Web of Science][Medline]
34. Pickett W, Hartling L, Brison RJ, Guernsey JR. Fatal work-related farm injuries in Canada, 1991–1995. Can Med Assoc J.1999;160:1843–1848.[Abstract]
35. Strickland MJ, Crawford JM, Shen L, Wilkins JR III. Time-dependent recordkeeping fatigue among youths completing health diaries of unintentional injuries. J Safety Res.2006;37:487–492.[CrossRef][Medline]
36. Hartling L, Pickett W, Brison RJ. Non-tractor, agricultural machinery injuries in Ontario. Can J Public Health.1997;88:32–35.[Web of Science][Medline]
37. Pickett W, Schmid H, Boyce WF, et al. Multiple risk behavior and injury: an international analysis of young people. Arch Pediatr Adolesc Med.2002;156: 786–793.
38. Strickland MJ, Crawford JM, Shen L, Wilkins JR III. Time-dependent recordkeeping fatigue among youth completing health diaries of unintentional injuries. J Safety Res.2006;37:487–492.[CrossRef][Medline]
39. Marlenga B, Pickett W, Berg RL. Agricultural work activities reported for children and youth on 498 North American farms. J Agric Safety Health.2001;7: 241–252.[Medline]
40. Braun BL, Gerberich SG, Sidney S. Injury events: utility of self-report in retrospective identification in the USA. J Epidemiol Community Health.1994;48: 604–605.
41. Harel Y, Overpeck MD, Jones DH, et al. The effects of recall on estimating annual nonfatal injury rates for children and adolescents. Am J Public Health.1994;84:599–605.
42. Landen DD, Hendricks S. Effect of recall on reporting of at-work injuries. Public Health Rep.1995; 110:350–354.
43. Jenkins P, Earle-Richardson G, Slingerland DT, May J. Time dependent memory decay. Am J Ind Med.2002;41:98–101.[CrossRef][Web of Science][Medline]
44. Barrett-Conner E. Nutrition epidemiology: how do we know what they ate? Am J Clin Nutr.1991;54: 182S–187S.[Medline]
45. Peterson L, Schick B. Empirically derived injury prevention rules. J Appl Behav Anal.1993;26: 451–460.[CrossRef][Web of Science][Medline]
46. Callmer E, Riboli E, Saracci R, Akesson B. Dietary assessment methods evaluated in the Malmo food study. J Intern Med.1993;233:53–57.[Web of Science][Medline]
47. De Castro JM. Methodology, correlational analysis, and interpretation of diet diary records of the food frequency and fluid intake of free-living humans. Appetite.1994;23:179–192.[CrossRef][Web of Science][Medline]
48. Marsh P, Kendrick D. Near miss and minor injury information—can it be used to plan and evaluate injury prevention programmes? Accid Anal Prev.2000;32: 345–354.[CrossRef][Web of Science][Medline]
49. Morrongiello BA, Ondejko L, Littlejohn A. Understanding toddlers’ in-home injuries: I. Context, correlates, and determinants. J Pediatr Psychol.2004;29: 415–431.
50. McClure RJ, Douglas RM. The public health impact of minor injury. Accid Anal Prev.1996;28: 443–451.[CrossRef][Web of Science][Medline]
51. Prevention of agricultural injuries among children and adolescents Pediatrics. 2001;108:1016–1019.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |