Determination of national pediatric injury prevention priorities using the injury prevention priority score

Determination of National Pediatric Injury Prevention Priorities Using the Injury Prevention Priority Score By Adil H. Haider, Donald Risucci, Saad Omer, Tom Sullivan, Stephen DiRusso, Michel Slim, and Charles Paidas Valhalla, New York and Baltimore, Maryland

Purpose: Previous studies have found that the Injury Prevention Priority Score (IPPS) provides a reliable and valid method to gauge the relative importance of different injury causal mechanisms at individual trauma centers. This study examines its applicability to prioritizing injury mechanisms on a national level and within defined pediatric age groups. Methods: A total of 47,158 patients (age ⬍17) in the National Pediatric Trauma Registry were grouped into common injury mechanisms based on ICD-9 E-Codes. Patients also were stratified by age group. IPPS was calculated for each mechanism and within each age group. Results: Falls of all types account for the greatest number of injuries (n ⫽ 15,042; 32%), whereas child abuse results in the most severe injuries (mean Injury Severity Score, 13.3) How-

T

HE INJURY PREVENTION Priority Score (IPPS) has been introduced recently as a simple and valid method by which trauma centers can identify the relative importance of different injury mechanisms affecting their service population.1,2 The score is based on local injury severity and frequency data present in the center’s trauma registry. This study applies the IPPS to a national registry of pediatric trauma patients in an effort to help delineate national pediatric injury prevention objectives and identify age-specific differences in childhood injury prevention needs. MATERIALS AND METHODS Data from all patients less than 17 years of age, recorded in the National Pediatric Trauma Registry (NPTR)3 between 1995 and 2001,

From the Department of Surgery, New York Medical College/ Westchester Medical Center, Valhalla, NY; Department of International Health, Johns Hopkins Bloomberg School of Public Health; and Division of Pediatric Surgery, Johns Hopkins School of Medicine, Baltimore, MD. Presented at the 55th Annual Meeting of the Section on Surgery of the American Academy of Pediatrics, New Orleans, Louisiana, October 31-November 2, 2003. Address reprint requests to Adil H. Haider, MD, MPH, Senior Resident, Department of Surgery, Westchester Medical Center, 95 Grasslands Rd, Box 489, Valhalla, NY 10595; e-mail: ahaider@ jhsph.edu. © 2004 Elsevier Inc. All rights reserved. 0022-3468/04/3906-0042$30.00/0 doi:10.1016/j.jpedsurg.2004.02.028 976

ever, the most significant mechanisms of injury, according to IPPS, were motor vehicle crashes followed by pedestrian struck by motor vehicles. Certain age groups had specific injury problems including child abuse in infants and assault and gun injuries in adolescents. Conclusions: IPPS provides an objective, quantitative method for determining injury prevention priorities based on both frequency and severity at the national level. It also is sensitive to age-related changes in different mechanisms of injury. J Pediatr Surg 39:976-978. © 2004 Elsevier Inc. All rights reserved. INDEX WORDS: Pediatric injuries, prevention, trauma registry, prioritization.

were included in the analysis. The NPTR was chosen because it is the largest available data repository of injured children requiring hospitalization in the United States. During the years studied, more than 80 large trauma centers from across the country contributed data. To study the impact of different injury mechanisms, patients were grouped according to the Centers for Disease Control and Prevention’s (CDC) recommended framework for classifying different injury mechanisms. This framework is based on International Classification of Diseases version nine (ICD-9), E-codes,4 which reflect the manner and intent of injury. Twenty-two injury mechanisms resulted, eg, motor vehicle crashes (MVC), pedestrians struck by motor vehicles (PEDS STRUCK by MV), falls from a height (Fall-diff), falls on the same level (Fall-same). Only the top 10 injury mechanisms were included in data analysis (Table 1) because the remaining mechanisms contributed only a very small percentage of cases. IPPS was calculated for each injury mechanism based on the sum of standardized scores representing the number of children injured by the particular mechanism (frequency) and their mean Injury Severity Score (ISS).5,6 The computation requires that the frequency and mean ISS for each mechanism is transformed into a Z-score to equalize the scale of both variables before summing them together. The sum then is transformed into a T score (mean, 50; SD, 10), which is the resultant IPPS for that mechanism. The above-described transformations equalize the influence of frequency and severity awarding each mechanism a score representative of both variables. After calculating the IPPS for each mechanism in the NPTR, mechanisms were sorted according to their IPPS in descending order. This created a rank order list of most to least important injury problems. During the initial validation of the IPPS, the chosen measure of ISS was replaced with several alternate measures of injury (eg, revised trauma score7 and Glasgow Coma Score8). The IPPS was found to be robust, (r ⫽ 0.82 to 0.92) supporting the above-described statistical approach. The score was also shown to have substantial cross validation with results in the direction of previously published literature. For Journal of Pediatric Surgery, Vol 39, No 6 (June), 2004: pp 976-978

NATIONAL INJURY PREVENTION PRIORITIES

977

Table 1. National Injury Prevention Priorities as Determined by IPPS (Based on NPTR Data 1995-2001) Rank

Mechanism

No. Injured

Mean ISS

IPPS*

1 2 3 4 5 6 7 8 9 10

MVC PEDS struck Child abuse Falls (diff) Bicycle Falls (same) Gun injuries Assault Sports injuries Playground

10,523 5,752 1,761 8,185 4,240 4,539 309 1,262 1,027 2,498

10.01 9.98 13.3 7.12 8.5 6.01 9.5 7.2 7.09 5.04

69.48 60.08 59.74 58.31 53.43 48.12 47.66 44.20 43.40 36.26

Abbreviations: Bicycle, bicycle injuries; Falls (diff), falls from a height; Falls (same), falls at the same level; MVC, motor vehicle crashes; PEDS struck, pedestrians struck by motor vehicles; Playground, Playground injuries. *Because the mean and standard deviation of the IPPS are known to be 50 and 10, respectively, the relative importance of different injury mechanisms can be easily judged.

example, pedestrians struck by motor vehicles were, on average, more severely injured than victims of falls on the same level. These findings suggest that the IPPS accurately compares different mechanisms of injury on a quantitative basis.1 Patients were also stratified into 5 different age groups (⬍1, 1 to 4, 5 to 9, 10 to 14, 15 to 16 years). IPPS then was computed for each injury mechanism within each age group. Subgroups with 10 or fewer subjects were excluded to prevent spurious results secondary to sampling error.

RESULTS

There were 47,195 pediatric trauma victims included in the NPTR during the 6-year study period. Sixty-five percent of these patients were boys, and the mean age was 9.1 years (SD ⫾ 2.1 years). Table 1 shows the IPPS based ranking of different injury mechanisms for the entire study cohort. Table 2 depicts how IPPS-based rank lists differ for specific age groups. DISCUSSION

The National Center for Injury Prevention and Control (NCIPC) released national injury prevention objectives in 2002.9 This report relied on “Extensive input from its academic research centers, national nonprofit organizations, and other federal agencies with a stake in injury prevention.” When these indirect methods were used,

there was no accepted quantitative method to determine the significance of different injury problems. The IPPS provides such a quantitative measure. Its application to the NPTR may aid in the development of evidence-based national injury prevention priorities for children because it is based on a national registry of pediatric trauma victims admitted to multiple trauma centers from across the country. Basing priorities on data from these institutions is relevant, because they treat the most severely injured children as well as account for a significant portion of the country’s injury prevention activities. IPPS-based ranking of different injury mechanisms also offers the advantage of accounting for the altering level of severity different injury mechanisms typically cause. The importance of this is highlighted by the IPPS based ranking of child abuse. Even though falls from a height were 5 times more common, child abuse ranked higher because these children suffer more severe injuries. In fact, child abuse is the most severe type of injury faced by children today (mean ISS, 13.3), and the IPPS ranks it third overall. If simple frequency tabulations were followed, this mechanism would not even be considered among the top 5 injury problems. As expected, motor vehicle crashes are the number 1 ranked injury problem in the nation followed by pedestrians struck by motor vehicles. This reinforces the need for continuing work in motor vehicle safety and also points out the critical necessity to devise strategies to prevent children from being struck by motor vehicles in motion. Falls from a height and falls at the same level ranked fourth and sixth, respectively. This finding supports dividing “falls” into 2 separate categories, because the higher energy transfer associated with falling from a higher level is reflected in it having a greater mean ISS. Also, prevention strategies for these 2 types of falls are different. Gun injuries ranked much higher than simple frequency tables would suggest because of the intensity of this type of injury. This finding underlines the relative importance of mechanisms that cause extremely severe injuries, high relative morbidity, and high resource utilization. Identifying such mechanisms becomes even more important as preventing them would be more cost

Table 2. Top 5 National Injury Prevention Priorities for Specific Age Groups Ranked According to IPPS Rank

⬍1 yr (n ⫽ 2,205)

1 to 4 yr (n ⫽ 10,042)

4 to 9 yr (n ⫽ 4,560)

9 to 14 yr (n ⫽ 9,647)

15 to 17 yr (n ⫽ 3,076)

All Ages (n ⫽ 34,350)

1 2 3 4 5

Abuse Fall (diff) MVC PEDS Str Bicycle

Fall (diff) Abuse MVC PEDS Str Gun

MVC PEDS Str Fall (diff) Bicycle Gun

MVC PEDS Str Bicycle Fall (same) Gun

MVC PEDS Str Bicycle Assault Fall (same)

MVC PEDS Str Abuse Fall (diff) Bicycle

Abbreviations: Abuse, Child abuse; Bicycle, bicycle injuries; Fall (diff), falls from a height; Fall (same), falls at the same level; Gun, gun injuries; MVC, motor vehicle crashes; PEDS Str, pedestrians struck by motor vehicles. n ⫽ Number of children injured in that specific age group owing to the top 5 causes.

978

HAIDER ET AL

effective. Guns and assault ranked higher than playground and sports injuries. This may be because of the fact that playground injuries have a relatively low ISS, possibly reflecting the effect of intensive injury prevention work in this arena. At the national level, different age groups have specific injury problems. The youngest children suffer most from child abuse and falls. This intensity diminishes with advancing age. As the children grow older, walk, and start riding in motor vehicles, MVC and PEDS STRUCK by MV become more important. Similarly, sports and bicycle injuries become more important causes among school children and young adolescents. For the oldest children, assault and gun injuries become more significant. These findings substantiate similar regional level findings supporting age-specific injury prevention programs.1 Application of the IPPS technique to data from multiple pediatric trauma registries has other potential uses; it may allow individual trauma centers to compare

themselves with national statistics and could also be used to make state versus federal comparisons for resource allocation purposes. Additionally, by applying it serially over different time periods, the effectiveness of different national programs to reduce moderate to severe injuries in specific areas could be assessed. Relative importance of injury mechanisms in pediatric trauma varies considerably across age groups, and highly severe mechanisms such as child abuse remain significant even though they are less common. IPPS provides an objective, quantitative method for determining injury prevention priorities based on both frequency and severity at the national level. It also is sensitive to age-related changes in different mechanisms of injury. ACKNOWLEDGMENT The authors thank the National Pediatric Trauma Registry and its investigators including Carla Di Scala, PhD, Barbara Barlow, MD, Robert Sege, MD, PhD, Arthur Cooper, MD, MS, and Joseph Tepas, MD.

REFERENCES 1. Haider A, Risucci D, Paidas C, et al: Injury prevention priority score: A new method for trauma centers to prioritize injury prevention initiatives. J Am Coll Surgeons 198:2004 (in press) 2. Risucci D, Haider A, Sullivan T, et al: Validation of an Injury Prevention Priority Score. Presented at the sixty second annual meeting of the American Association for the Surgery of Trauma. Orlando, Florida. September 12-14, 2002 3. Tepas J, Ramenofsky M, Di Scala D, et al: National Pediatric Trauma Registry. J Pediatr Surg 24:156-158, 1989 4. CDC Recommended framework for presenting injury mortality data. MMWR Recomm Rep 29; 46(RR-14):1-30, 1997 5. Baker SP, O’Neil B, Haddon W, et al: The injury severity score:

A method for describing patients with multiple injuries and evaluating emergency care. J Trauma 14:187-196, 1974 6. Baker S, O’Neil B: The injury severity score: An update. J Trauma 16:882-885, 1976 7. Champion HR, Sacco WJ, Copes WS: A revision of the Trauma Score. J Trauma 29:624, 1997 8. Healey C, Osler TM, Rogers FB, et al: Improving the Glasgow Coma Scale score: Motor score alone is a better predictor. J Trauma 54:671-678; discussion 678-80, 2003 9. National Center for Injury Prevention and Control, C.D.C. Research Agenda, Atlanta (GA), Centers for Disease Control and Prevention, 2002