Child Death Reviews: a Gold Mine for Injury Prevention and Control

Home , Injury prevention

276 Injury Prevention 1999;5:276–279 Inj Prev: first published as 10.1136/ip.5.4.276 on 1 December 1999. Downloaded from Child death reviews: a gold mine for injury prevention and control

Chukwudi Onwuachi-Saunders, Samuel N Forjuoh, Patricia West, Cimon Brooks

Abstract lence, compared with their counterparts in Objectives—The purpose of this study was other industrialized nations in the world.3 Vio- to demonstrate how child death review lence is the second leading cause of death for teams can be used to prevent future deaths youth aged 15–24 years in the US.2 These grim through retrospective, multiagency case statistics highlight the serious health eVects of analysis and recommendations for edu- injury on children. What is even more startling cational programs and policy change. is that many of these injury deaths are Methods—A listing of all deaths to persons eminently preventable. ages 21 years and younger in Philadelphia The challenge for the injury arena has been that occurred in 1995 was compiled by the to move from analysis to action. The problem is Philadelphia Interdisciplinary Youth Fa- clear, but the solutions are still hazy. Applying tality Review Team (PIYFRT), a multia- the public health response to solving this issue gency, multidisciplinary, community would be logical.4 The steps to this approach based group created in 1993 with the mis- include defining the issue through data collec- sion to prevent future deaths through tion or surveillance; analyzing data to identify review, analysis, and initiation of correc- potential risk factors, enabling factors, and tive actions. Data were collected on demo- barriers; developing interventions based on the graphic variables, as well as the analysis; implementing interventions through circumstantial variables on injuries such community based programs; and using evalua- as weapon type, alcohol and drug use, and tion results to modify and re-evaluate original contact with the criminal justice system, interventions. among others. Each case was reviewed Although child death review teams have used thoroughly to determine whether or not this approach in the US, in part, to identify the death was preventable. Selected injury causes of death, surrounding circumstance and http://injuryprevention.bmj.com/ related death cases were analyzed further the needed policy changes,5–9 and no doubt by demographic and circumstantial vari- other countries may have this process, to our ables. knowledge, the child death review process has Results—In 1995, 607 children ages 21 not been adequately described with regards to years and younger died in Philadelphia policy response. The Philadelphia Interdisci- from natural causes (61.6%), uninten- plinary Youth Fatality Review Team (PIY- tional injuries (16.3%), homicide (18.6%), FRT), created in June of 1993, has many suicide (2.3%), and undetermined causes elements of this approach. It represents a new, (1.2%). More than a third (37.2%) of all multiagency, multidisciplinary, community deaths were considered preventable. Of based process by which data are routinely and the injury deaths (n=224), 95% were systematically collected, and analyzed to high- judged to be preventable. Preventable fire/ light potential risk factors, as well as to Delta Consultancy burn injury deaths (n=29) were associated determine the enabling factors and barriers for on September 25, 2021 by guest. Protected copyright. Group, Washington, with lack of a smoke detector, non- DC, 901 6th Street, youth fatalities. Most importantly, this process SW, #401, Washington, supervision of children, and faulty home fosters the development and implementation of DC 20024, USA appliances. Violent deaths were associated interventions to prevent injury mortality and C Onwuachi-Saunders with substance abuse, gang involvement, morbidity among children in Philadelphia. C Brooks chronic truancy, academic failure, and Philadelphia is the fifth largest city in the US access to weapons. with approximately 1.6 million persons: 54% Department of Family Conclusions—Relevant policies for these Medicine, Texas A&M white, 40% African Americans, and 6% University System preventable or intervenable deaths are Latinos and Asians. Philadelphia, like the rest Health Science Center, discussed such as use of non-battery pow- of the nation, has experienced an increase in Scott and White Clinic ered smoke detectors. the number of youth who have died as a result and Memorial (Injury Prevention 1999;5:276–279) Hospital, Temple, of both unintentional injuries and violence. Texas Keywords: death reviews; child death reviews; violence Subsequently, Operation Peace in Philadelphia S N Forjuoh (OPP), a citywide community based antiviolence/peace promotion initiative was Public Health Injury remains the leading cause of death and developed to help reduce this epidemic. OPP’s Consultant, components include a long term community Philadelphia, disability among children world wide. In the Pennsylvania United States, injuries claim nearly 20 000 based collaboration; an ongoing media aware- P West lives and permanently disable 30 000 children ness campaign; and an innovative information each year.12A 1983 report by the US Centers and data system that includes the PIYFRT. Correspondence and reprint for Disease Control and Prevention highlighted The PIYFRT represents a comprehensive requests to: Dr Onwuachi-Saunders (e-mail: that children in the US suVer disproportion- eVort to enhance research on youth [email protected]) ately from injury deaths, particularly from vio- fatalities.10 11 It is unique because the death Child death reviews 277 Inj Prev: first published as 10.1136/ip.5.4.276 on 1 December 1999. Downloaded from Table 1 Preventability of youth deaths by manner of death, Philadelphia, 1995 one in which, with retrospective analysis, a rea- sonable intervention may have prevented the Preventability death. “Reasonableness” was defined to take Need more into consideration the condition, circum- Manner of death Yes (%) No (%) information (%) Total stances, or resources available. At no point Natural 10 (2.7) 337 (90.8) 24 (6.5) 371 during these deliberations was any determina- Unintentional injury 95 (96.0) 3 (3.0) 1 (1.0) 99 tion of preventability construed to equate pre- Suicide 13 (100.0) — — 13 dictability. If the death was judged to be Homicide 105 (93.8) — 7 (6.3) 112 Undetermined 1 (14.3) — 6 7 preventable, the committee then discussed or Total 224 (37.2) 340 (56.5) 38 (6.3) 602 developed interventions and policies to be implemented at the individual, agency and/or community level. The potential intervention review process is linked to the city’s violence strategies and the aggregated data were re- prevention and peace promotion initiative. In viewed at quarterly policy meetings involving addition, it is the only child death review proc- team members, as well as invited guests such as ess in the nation with a process focusing solely city council members, community activists, on youth homicides and suicides. Its mission is and agency leaders. Selected injury related to prevent future child deaths by review, analy- death cases were analyzed by demographic sis, and corrective actions. The PIYFRT characteristics and circumstantial variables. achieves its mission by collecting comprehensive information about victims, including the circumstances preceding the death in order to Results understand more about prevention. The PIY- In 1995, 607 youth aged 21 years and younger FRT process, modeled after the traditional died in Philadelphia. This number represented child death reviews, shifts the paradigm by an 8% decrease from 1994 (662 deaths). Sixty calling upon broader agency, multidisciplinary per cent were due to natural deaths, 18.6% participation. Two separate teams review resulted from homicides, 16.3% resulted from deaths based on how the child died: homicides unintentional injuries, 2.3% were due to and suicides are reviewed by one team, while suicides, and 1.2% were undetermined. This natural deaths, unintentional injury deaths, distribution was similar to that in 1994. The and undetermined deaths are reviewed by leading causes for unintentional injury deaths another. The members of both teams include were residential fires/burns (29.3%), occu- one Hispanic and one predominantly African pancy of motor vehicles (15.2%), drowning American community based organization and (10.1%), and pedestrian injuries (7.1%).

many other organizations such as the district http://injuryprevention.bmj.com/ attorney’s oYce, ﬁre and police departments, PREVENTABLE DEATHS oYce of the medical examiner, and several Of all deaths, 37% were judged to be prevent- hospitals. able. While the majority of natural deaths were In this paper we demonstrate the use of child judged to be non-preventable, all suicide death reviews to prevent future deaths. deaths and the majority of homicides and unintentional deaths were judged to be preventable. Because of incomplete information Methods on a few of the cases, preventability was not Each month, all team members received a list determined for 7% of the cases (table 1). of all deaths involving youth 21 years and younger from the Philadelphia vital statistics oYce. To assure that all deaths were reviewed HOMICIDES (N=112)

accurately, the lists were compared with other Among the leading causes of death, homicides on September 25, 2021 by guest. Protected copyright. lists obtained from the state vital statistics, ranked first in youth ages 13–21 years. Even medical examiners’ database, and various par- among infants, unintentional injuries and ticipating agencies. Although many non- homicides tied for the fifth position among the Philadelphia County residents died in the city, 10 leading causes of death. The majority of only deaths involving county residents were homicide victims were African American males reviewed. Members were responsible for (81.4%). Most homicides were perpetrated by searching their own agencies for all information firearms, largely automatic handguns (43.4%), on the decedents. Because the information dis- while a few were perpetrated by fists, knives, cussed by the fatality review members was and through hanging or suVocation. Most highly confidential, all members signed a homicide victims had histories of substance confidentiality statement. To ensure further abuse, prior arrests, problems at school such as confidentiality, information was only shared truancy and poor grades, as well as prior verbally. weapon related injuries. At the monthly meetings, agencies presented their information and a staV person entered the FIRE/BURN DEATHS (N=29) relevant information on each of the variables All 29 deaths due to fire/burns were judged to onto specific data forms. There were separate be preventable with more than half involving forms for homicides, suicides, and uninten- situations where the fire was started by a child tional or natural deaths. Once all the available aged 3–6 years. In half of the cases, the source data were collected on each death, the of the fire was matches while the child was committee determined whether the death was playing, and in the remaining cases the source preventable. A preventable death was defined as was a cigarette (n=4) or a lighter (n=3). Other 278 Onwuachi-Saunders, Forjuoh, West,et al Inj Prev: first published as 10.1136/ip.5.4.276 on 1 December 1999. Downloaded from Table 2 Demographic and circumstantial characteristics of fire/burn youth deaths, mothers the potential risk of assuming that Philadelphia, 1995 all intimate partners are appropriate caregivers to their young children; and iden- Age of person (years) Smoke Activity of person starting fire detector starting fire tifying the need for multidisciplinary, multiagency school intervention teams. Source of fire No 3 4 5 6 Yes No Playing Other x Prevention of unintentional injuries through Matches 8 — 2 4 2 3 5 7 1 increasing the enforcement of the child Cigarette 5 4 1 — — 4 1 1 4 safety seat law; debating traYc regulations Lighter 3 2 1 — — 3 — 3 — Spaceheater3 **** —3 — 3 of public transportation buses used as Electrical wire 5 **** 3 2 — 5 school buses during school hours; increas- Other 5 **** 2 3 — 5 ing school based pedestrian safety educa- *Not applicable. tion; recognizing the need for “Fire Starter” programs for children ages 3–7; and recog- sources of the fire included space heaters and nizing the need for non-battery powered faulty electrical wires (table 2). smoke detectors. This process allows for innovative data collection. The diverse data from many agen- Discussion cies allow a broader perspective and multidisci- The concept of using the multidisciplinary plinary input, thus the steps to more complete death review team to analyze child deaths and research. To illustrate this point, a scenario is from the data develop cost eVective interven- the 5 year old who dies after playing with tions is important. Regardless of the age matches. When all fire related deaths are group, injury was found to be a significant aggregated, most reviews by researchers would cause of youth death in Philadelphia. Child identify the lack of a smoke detector as the death review teams have contributed signifi- “potential” risk factor and a working smoke cantly to knowledge about child abuse and detector use campaign as the primary preven- neglect. These traditional child death review tion strategy. There is a need to better processes have often been referred to as a understand the barriers and enabling factors as “witch hunt” by some practitioners outside of they relate to behavior. Through the PIYFRT the child abuse and neglect arena because of process, the “barrier” identified was inad- their focus on identifying the perpetrators. equate supervision by a family member who The PIYFRT process, on the other hand, has left matches available toa5yearold. Both been expanded beyond this perspective and reviews would have come to the same reflects a public health mission to prevent conclusion—that the death was preventable.

youth mortality, but most importantly to pro- But only the PIYFRT process would identify http://injuryprevention.bmj.com/ mote a healthy lifestyle. This process can be the “barrier,” the lack of supervision by the viewed as a “gold mine”. It is a gold mine in as caretaker and the intervention strategy—that much as by examining deaths, a cost eVective is, highlighting the need for programs for alternative injury surveillance methodology young children who start fires. In this era of can be created to better understand youth scarcity, understanding barriers and enabling fatality in any community. In addition, this factors is essential in making decisions about process alleviates or decreases “turfing” and eYcient resource allocations. increases collaboration among agencies, the Many times an agency appears with what stakeholders of children’s health. Perhaps the they believe is the “only story” as opposed to greatest asset is that these data, in conjunction understanding that they have only one piece of with data from other communities (currently the puzzle. The process of sharing confidential child death review teams now exist in almost information frequently involves exposing the every state and many local communities), offer limitations of agency systems and the uncom- on September 25, 2021 by guest. Protected copyright. a wealth of information to examine the many fortable realization that many of those limita- factors that increase the likelihood of injury tions fail to protect children. For team that may lead to death. members to become eVective in this process, a As a result of this process, the following critical component is the willingness to partici- products and policy changes were generated pate. Participation increases as members de- and are being implemented in Philadelphia: velop trust in the process and better under- x Improved coordination among PIYFRT stand that the limitations of failures become members. the basis for system change. x More comprehensive data collection to The PIYFRT review process involving com- reduce the number of missing data; a munity participation with agencies (public and decrease in inadequate death certificates12;a private) allows many of the pieces of the puzzle link between birth and infant death certifi- to be brought together. The comprehensive cates; use of height/weight charts in the data generated serves to complete the puzzle medical examiner’s oYce; and access to the needed by decision makers to formulate 911 history for selected deaths. solutions. The benefits of sharing information x Prevention of interpersonal violence through can be viewed through the development of pre- development of a program for first time vention strategies, interventions, products, juvenile violators of the Uniform Firearm policies, etc. Act; increasing agency communication on The challenge is to expand child death juvenile bench warrants; recognizing the reviews beyond the focus of child abuse and need for school nurses’ training on domestic neglect to one of public health so as to identify violence; recognizing the need to highlight to preventable child deaths and achieve eVective Child death reviews 279 Inj Prev: first published as 10.1136/ip.5.4.276 on 1 December 1999. Downloaded from prevention. Presently, this process may serve as 3 Centers for Disease Control. Homicide surveillance, 1970– 1978. Atlanta, GA: Centers for Disease Control and the one source of comprehensive data on the Prevention, 1983. many factors contributing to youth deaths. It 4 Moore M, Prothrow-Stith D, Guyer B, et al. Violence and intentional injuries: criminal justice and public health per- also oVers a menu of what can be done, as well spectives on an urgent national problem. In: Reiss AJ, Roth as providing the necessary information for JA, eds. Understanding and preventing violence. Vol 4. Consequences and control. Washington, DC: National Acad- decision makers competing for scarce public emy Press, 1997: 167–216. health resources. The PIYFRT serves as a 5 Colorado Child Fatality Review Committee. Annual report and conference proceedings. Denver: Colorado Child Fatality model on the local level of a public health/ Review Committee, 1991. community based approach which generates 6 Committee on Child Abuse and Neglect and Committee on science based information and converts it into Community Health Services. Investigation and review of unexpected infant and child deaths. Pediatrics 1993;92: concrete action, a gold mine for injury preven- 734–5. tion and control. 7 Durfee M, Gellert GA, Tilton-Durfee D. Origins and clini- cal relevance of child death review teams. JAMA 1991;267: 3172–5. We wish to acknowledge the members of both PIYFRT review 8 Ewigman B, Kiviahan C, Land G. The Missouri Child teams and the members of the task forces who are implement- Fatality Study: under reporting of maltreatment fatalities ing the gold. We would also like to acknowledge the support of among children younger than five years of age, 1983 Robert K Ross, MD who had the foresight to declare violence through 1986. Pediatrics 1993;91:330–7. among youth a public health emergency in Philadelphia and 9 Granik LA, Durfee M, Wells SJ. Child death review teams: a Rueben Warren, DDS, MPH, DrPH of the Centers for Disease manual for design and implementation. Chicago: American Control and Prevention who supported that eVort. We wish to Bar Association, 1991. thank Dawn Berney of the Philadelphia Health Management 10 Onwuachi-Saunders C, West P, Berney D. Youth as victims Corporation who provided technical support for the data of homicides: prevention of child homicide with an collection process. interdisciplinary youth fatality review team. Data needs in an era of health reform. Proceedings of the 15th Public Health Conference on Records and Statistics. Washington, DC: US 1 Center for Disease Control and Prevention. Childhood inju- Department of Health and Human Services, 1995: 25–7. ries in the United States. Atlanta, GA: Centers for Disease 11 Philadelphia Department of Public Health, Operation Peace Control and Prevention, 1990. in Philadelphia. Annual report. Philadelphia, PA: Philadel- 2 Children’s Safety Network. A data book of child and phia Interdisciplinary Youth Fatality Review Team, 1997. adolescent injury. Washington, DC: National Center for 12 Hanzlick R. Improving accuracy of death certificates. JAMA Education in Maternal and Child Health, 1991. 1995;274:537–8.

Faculty position in injury prevention Department of Occupational and Environmental Health, University of Iowa College of Public Health The Department of Occupational and Environmental Health at the University of Iowa Col- http://injuryprevention.bmj.com/ lege of Public Health invites applications for a tenure track faculty position in injury prevention at the level of assistant or associate professor. Applications may also be considered at the level of faculty associate. The successful applicant will join the faculty of the Department of Occupational and Environmental Health and of the CDC-funded Injury Prevention Research Center. A secondary appointment in the Department of Epidemiology, or in the Division of Emergency Medicine is possible. Applicants should have an MD, PhD, or other advanced professional degree with research experience related to the epidemiology of injury prevention. Strong epidemiological skills are required. The candidate should have publications in the peer reviewed literature and demon- strated teaching ability. Applicants should demonstrate outstanding research productivity or promise. Interest and experience in preventing injuries in a rural context is desirable. Dem-

onstrated ability to work eVectively with interdisciplinary research teams is desirable. on September 25, 2021 by guest. Protected copyright. The University of Iowa oVers a rich environment for collaborative research including a federally funded center in agricultural health and safety, a world class driving simulator, and a level 1 trauma center. Please send a curriculum vitae, a statement of research interest, and names of three refer- ences to Dr Craig Zwerling, Chair, Search Committee, (#F1850) Attn: Wendy Jackson, Department of Occupational and Environmental Health, College of Public Health, Univer- sity of Iowa, 100 Oakdale Campus, 124 IREH, Iowa City, IA 52242-5000, USA. Our department is committed to increasing faculty diversity. The University of Iowa is an equal opportunity and aYrmative action employer. We strongly encourage women and minorities to apply.