www.elsevier.com/locate/jpedsurg

Journal of Pediatric Surgery (2012) 47, 991–995

Television-related injuries in children—the BritishColumbia experienceJessica Mills a, Jeremy Grushka b, Sonia Butterworth a,⁎

aDepartment of Surgery, BC Children's Hospital and the University of British Columbia, Vancouver, Canada V6H 3V4bDepartment of Surgery, McGill University, Montreal, Canada H3G 1AH

Received 20 January 2012; accepted 26 January 2012

C

0d

Key words:Pediatric;Trauma;Television;Unintentional injury;Ethnicity

AbstractPurpose: In Canada, mortality from falling televisions (TVs) is the 15th leading cause of childhooddeath owing to injury. Frequency, characteristics, and outcomes of TV childhood injuries wereexamined to determine any at risk populations.Methods: All TV-related traumas at a tertiary children's hospital from 1997 to 2011 were identifiedusing the Canadian Hospitals Injury Reporting and Prevention Program database and the hospital'strauma database. Charts of admitted patients were reviewed.Results: Analysis of 179 injuries (10-24 per year) revealed a high frequency of injury in the home and apreponderance of head and neck injuries. Toddlers were the most commonly injured age group. Elevenadmitted patients were identified; 6 were admitted to intensive care unit with significant head injuries, 2of whom required surgery. More than half of admitted patients were First Nations or recent immigrants.The length of stay for a ward vs intensive care unit admission was 1.3 days (range, b1-2 days) comparedwith 7.6 days (range, b1-20 days), respectively. One child had residual deficits requiring rehabilitation,but there were no mortalities.Conclusion: Injury severity appeared higher in patients from First Nations and recent immigrantfamilies. Television injury would likely have been prevented by a securing device or support.© 2012 Elsevier Inc. All rights reserved.

In 2009, American households had more television (TV) In September of 2006, the US Consumer Products Safety

sets than people, with an average of 2.93 TVs to 2.54 people[1]. Although the number of sets per household had droppedto 2.5 in 2010, this still represents 116 million householdswith at least 1 TV and 31% of these with 4 or more sets [2].The Canadian situation is similar, with 95% of all Canadianhouseholds in 2007 having at least 1 TV and 60% having 2 ormore [3].

⁎ Corresponding author. Division of Pediatric Surgery, Vancouver, BC,anada V6H 3V4. Tel.: +1 604 875 2667; fax: +1 604 875 2721.E-mail address: sbutterworth@cw.bc.ca (S. Butterworth).

022-3468/$ – see front matter © 2012 Elsevier Inc. All rights reserved.oi:10.1016/j.jpedsurg.2012.01.062

Commission [4] reported a startling increase in the number ofpediatric deaths related to TV trauma and warned of thedangers of large TV sets resting on improper supports suchas dressers or inadequate TV stands. Since that time, studiesin several countries have confirmed the concerning findingthat TV-related injuries are a significant source of pediatrictrauma and that the rate of these injuries is increasing [5-13].Literature on the Canadian experience with these types ofinjuries is lacking. Our aim was to delineate the provincialtrauma burden owing to TV-related injury and to determine ifany high-risk groups could be identified that might be targetsfor prevention education.

Table 1 Population demographics for nonadmitted vsadmitted TV trauma patients

Variable Nonadmitted Admitted

No. of patients 168 11SexMale 112 (67%) 6 (55%)Female 55 (33%) 5 (45%)Age (mo)Mean 42.2 24.7Median 30.5 19Minimum/maximum 8/194 9/76Age groupsb12 mo 13 (8%) 2 (18%)12-23 mo 47 (28%) 5 (46%)2-4 y 75 (45%) 3 (27%)5-9 y 24 (14%) 1 (9%)10-14 y 7 (4%) 015-19 y 2 (1%) 0

992 J. Mills et al.

1. Methods

A database review of all TV-related traumas seen at thesingle provincial tertiary children's hospital from January1997 to December 2007 was carried out using data abstractedfrom the Canadian Hospitals Injury Reporting and Preven-tion Program (CHIRPP) database. The CHIRPP is anationwide database collecting data on injuries treated at10 pediatric and 4 general hospitals in Canada. It obtainsinformation regarding the circumstances surrounding theinjury as well as information regarding the nature andtreatment of the injury. Data fields include patient de-mographics, environment of injury, and type, mechanism,and severity of injury as well as patient disposition. Thesefields were used to characterize the incidence, environment,and characteristics of pediatric TV trauma in the province.

A retrospective chart review of all TV trauma admissionsto the hospital between January 2007 and April 2011 was alsocarried out. Admissions from January 1997 to December2007 were identified using the CHIRPP database, whereasadmissions from January 2008 onward were identified usingthe hospital's own trauma database. Specific details abstract-ed included patient demographics; environment of injury;type, mechanism, and severity of injury; days admitted tointensive care unit (ICU); days of mechanical ventilation;operative interventions required; length of stay; patientdisposition; and outcome. Descriptive statistics were used.

Both the database review and retrospective chart reviewwere sanctioned by the institutional review board.

Table 2 Summary of injury environment for nonadmitted vsadmitted TV trauma patients

Variable Nonadmitted Admitted

Mechanism of injuryPatient collided with TV/stand 118/168 a

(70%)1/11a

(9%)TV/stand fell on patient 50 (30%) 10 (91%)Injury environmentChild's home (house, apartment,cottage)

138/162a

(85%)10/11a

(91%)Other home 15 (9%) 1 (9%)Educational (school, preschool, 2 (1%) 0

2. Results

A total of 179 children sustained TV-related trauma that,at minimum, required emergency department evaluationduring the study period. The mean annual incidenceincreased from 14.3 injuries per year from 1997 to 2002 to17.8 injuries per year from 2003 to 2007 (Fig. 1). The medianage of injured children was 30 months (8-194 months), with38% (n = 67) younger than 2 years and 81% (n = 148)younger than 5 years (Table 1). More than half (51%) ofinjuries occurred in toddlers (12-35 months of age inclusive),

0

5

10

15

20

25

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Hospital Admissions Emergency Visits

Fig. 1 Incidence of TV trauma emergency visits and hospitaladmissions from 1997 to 2007. This figure does not include hospitaladmissions from 2008 to 2011.

and 64% of admissions were in this age group. Injuries weremore common on weekdays (66%), and half of the injuriesoccurred in the early evening between 4:00 PM and 8:00 PM.Most injuries occurred in the home environment, mostcommonly in the living room, with a very small minority ineducational or recreational spaces (Table 2).

The most common injury mechanism was that of a childcolliding with the TV or TV stand. This resulted from severaldifferent activities including tripping while walking orrunning, falling from another piece of furniture or off anotherperson, or being pushed while playing, roughhousing, or

daycare)Recreational (community/artscenter, shopping mall)

6 (4%) 0

Hotel 1 (1%) 0Injury roomBedroom 26/106a

(24%)1/7a

(14%)Classroom 1 (1%) 0Kitchen 3 (3%) 0Living room 74 (70%) 6 (86%)Basement 1 (1%) 0Other indoors 1 (1%) 0

a Total number of records available for analysis.

Table 4 Ethnicity comparison between study population andcatchment area population

Ethnicity Nonadmitted(n = 61 known)

Admitted(n = 8 known)

GVRD(2006)

English/NorthAmerican

42 (69%) 2 (25%) 1.4 million

European 3 (5%) 0 890 000Semetic 1 (2%) 0 21 500South Asian 12 (20%) 0 210 000Southeast/EastAsian

6 (10%) 1 (12.5%) 590 000

First Nations ? 3 (37.5%) 43 000African ? 1 (12.5%) N/AOceanian ? 1 (12.5%) N/A

N/A indicates not available; GVRD, Greater Vancouver Regional District(main catchment area of population).

993TV-related injuries in children

fighting. The other major mechanism was a TV, with orwithout its stand, falling or being dropped on a child. Themost common body area injured was the head and neckregions (78%), with extremity trauma a distant second (20%)and almost no torso trauma reported (2%) (Table 3).

Eleven children were admitted to hospital for injuriessustained in TV-related trauma including 5 admitted to theward and 6 admitted to the ICU. Mean ICU stay was 3.6 days(0.5-10 days), and mean hospital length of stay was 4.7 days(0.5-20 days). Total mean length of hospital stay was 1.3days (range b1-2 days) and 7.6 days (range b1-20 days) forpatients admitted to the ward and intensive care unitrespectively. Four of the ICU patients required ventilationfor an average of 3.4 days (0.5-10 days). Ward admissionswere divided between extremity fractures, all of whomrequired operative repair, and closed head injuries, none ofwhom required operative intervention. All ICU admissionswere related to significant head injuries, 2 of which requiredoperative treatment. Skull fractures and intracranial injurieswere present in 45% and 36% of all admissions, respectively,and in 67% and 50% of ICU admissions, respectively.Admitted children were younger, with more than half agedyounger than 2 years, and were more likely to be injured by afalling TV rather than a collision. Cathode ray TVs wereinvolved in all admitted cases where the type of TV wasrecorded; however, this information was only available in 5cases. Head and neck traumas were, again, the most commonbody region involved, but admitted children were far morelikely to have skull fractures and/or intracranial injuriescompared with nonadmitted patients.

The database field for ethnicity was poorly recorded withonly 67 entries (Table 4). There were no entries for First

Table 3 Summary of TV injury characteristics in nonadmittedvs admitted patients

Variable Nonadmitted Admitted

Body region injuredHead and neck 128/163 a (79%) 8/11 a (73%)Upper extremity 10 (6%) 3 (27%)Torso 3 (2%) 0Lower extremity 22 (13%) 0Injury classification b

Soft tissue c 145 4Fracture 5 9 (3 extremity,

6 skull)Eye injury 1 1Dental injury 5 0Minor head injury d 27 3Intracranial injury 0 4No injury found 3 0

a Total number of records available for analysis (accounting formissing fields).

b Each patient had up to 3 injury codes to account for combinationinjuries, for example, facial laceration (soft tissue) with zygomaticfracture (fracture) and subdural hematoma (intracranial injury).

c Superficial wound, openwound, crush injury, and soft tissue injury.d Minor head injury and concussion.

Nations patients, although from our admitted patient data, itis clear that this ethnic group did make up part of our studypopulation. Ethnicity was available for 8 admitted patientsand was notable for a relative overrepresentation of FirstNations patients and recent immigrant patients (1 African, 1Fijian, and 1 East/Southeast Asian).

There were no deaths related to TV trauma in our series.Full functional recovery was achieved or expected in 10 of11 admitted patients. One three-and-a-half–year-old boywith an unwitnessed fall of a TV onto his head had extensiveskull and facial fractures, pneumocephalus, and multipleepidural hematomas. He was left with ataxia and balanceproblems as well as a sixth cranial nerve defect, a left afferentpapillary defect, and a presumed cerebrospinal fluid (CSF)leak from his right ear. His family refused transfer to arehabilitation facility and was therefore discharged to anoutpatient rehabilitation program.

3. Discussion

In North America, as in most of the developing world,unintentional injury is the leading cause of death in childrenolder than 1 year [14,15]. Although the most commonunintentional injuries include motor vehicle collisions, drown-ing, burns, and suffocation [15], the danger of furniture tip-over is being recognized [4,16]. The danger of TV tip-over wasfirst brought to light in 1998 [17], but it was not until 2006 thatthe US Consumer Product Safety Commission released awarning about the danger of large and heavy furniture, such asTV sets, tipping over, and causing serious pediatric injury anddeath [4]. In Canada, TV tip-over has been recognized as oneof the top 15 causes of fatal pediatric injury [18], whereas in theUnited States, the US Consumer Product Safety Commissionranked the TV as 1 of the top 5 hazards in the home and rankedthird in cause of morbidity and mortality [19].

Over the years, the data regarding the international burdenof morbidity and mortality caused by TV tip-over have been

994 J. Mills et al.

accumulating [5-13]. Several series have shown an increas-ing incidence of TV tip-over injuries [5,10,11], although itremains unclear whether this is related to the increasedprevalence of household TVs, increased numbers of TVs perhousehold, less supervision of toddlers, less use of propersecuring devices/safety stands, or the transition to flat-screenTVs. Our study echoes the increasing incidence of TV-related injury.

It is accepted that cathode ray tube models are inherentlyunstable because they are both heavy and possess a forward-placed center of mass. The flat-screens TVs, which havereplaced the older models, are generally more stable;however, their mass is centered on the screen. The largerthe screen, therefore, the more forward the center of gravityis displaced and the more unstable the TV becomes. Typicalflat screens of 42-in diameter can weigh from 25 to 45 kg,posing significant risk to young children if not properlysecured. It might be expected that with the ability to wallmount flat-screen TVs, these sets would be more commonlysecured and, therefore, safer to use. As with the older cathoderay sets, however, data regarding the rate of use of securingdevices and the actual security provided by wall mountdevices are lacking. With the explosion of technology andconsumers purchasing newer models, there is an expectationthat cathode ray models will become obsolete not only inelectronic store selves but also in the home. This expectationmay not materialize, however, because many may keep theolder models in other rooms in the home. Once a new flatscreen is purchased, older models may be placed innoncentral rooms with less supervision or be improperlysupported. Data regarding the types of TVs in the home arelacking. From the available data for admitted patients in thisstudy, it is clear that cathode ray tube models continue tocause severe injury.

Although injuries occur in all age groups, the burden ofdisease is carried by the toddlers who are at higher risk forboth general [5,6,9] and severe TV trauma. It is thought thattoddlers are the highest risk age group because they oftenspend most of the time at home, are mobile, and areinquisitive but have poor coordination and balance coupledwith poor judgment. Our study echoed earlier publications,with most injuries in our series (52%) occurring in childrenbetween the ages of 12 months and 3 years. This age groupmade up an even higher proportion of admitted trauma (64%).

Head injury has been identified as the most common TV-related injury [5,6,8-11]. It has also been identified as themain cause of significant morbidity and the only reportedcause of mortality related to TV trauma [5,6,11]. Thereported rates of head and neck injuries range from 54% to83% [6,8-11], and studies looking specifically at TV-relatedhead injury have shown a skull fracture rate of 73% to 89%and an intracranial injury rate of 29% to 56% [7,12,13]. Ouroverall head and neck injuries' rate of 78% is comparablewith other series. Similarly, our admitted head injury patients(n = 8) had a skull fracture rate of 63% and an intracranialinjury rate of 50%, which fit with prior data.

The home has previously been identified as the mainlocation for TV trauma in previous series [5,9]. Our study wasconsistent with these findings, with more than 85% of injuriesoccurring in the child's own home and 95% of injuriesoccurring in a home environment. In 2006, Ota et al [8]canvassed the parents/caregivers of children presenting withTV-related injuries and looked at the parental awareness of thedanger of the TV in the home. Eighty-five percent of parentsreported no awareness of the potential danger of a falling TV.This parallels data showing that parents of toddlers identifyless than half of the hazards present in the home environmentand that they see their child as less vulnerable to in-homehazards than toddlers in general [20]. Certainly, despitewidespread media coverage with several consumer protectionwarnings andmainstreammagazine coverage, the incidence ofthese injuries is, if anything, increasing. It is clear thatprevention education must be targeted at caregivers specifi-cally focusing on the home TV environment.

Although the data regarding ethnicity were not wellrecorded for the database as a whole, it is certainlyconcerning that more than one third of the admittedpatients were of First Nations ancestry and another thirdhad a primary caregiver who was a new immigrant toCanada with poor English language skills. These 2populations are dramatically overrepresented in our sampleof admitted patients compared with the demographics ofthe catchment area [21]. Canadian First Nations havepreviously been identified as a high-risk group for pediatricinjury [22], but being the child of an immigrant family haspreviously shown a protective effect against unintentionalchildhood injury [23]. Our study does not include anysocioeconomic data that permit conclusions regardingwhich factors might contribute to make these groups athigher risk for severe TV trauma. It does, however,suggest that these 2 populations would be good targets forrisk reduction or prevention education.

Our study was limited in its retrospective nature and theresulting absence of relevant data. In particular, the type of TVinvolved and the use of a securing device were not available.Missing records were also a problem, especially when lookingat ethnicity, particularly in First Nations patients. Given thatthis ethnic group has already been identified as high risk forpediatric injury, it is significant that First Nations ethnicity wasnot recorded in the CHIRPP database. Finally, our serieswould be more complete if we had database records to matchour chart review up to 2011. This would have allowed us to putthe recent TV injury admissions into the broader context ofoverall TV trauma admissions.

Our series of TV-related injuries is the first to report thatTV trauma is increasing in Canada. The threat posed tochildren is especially pronounced in toddlers who almostexclusively have both long-term morbidity and mortalitybecause of TV tip-over. Although inexpensive TV securingdevices are easily obtainable, we have no data as to thefrequency of their use. It is unclear whether the problem isdriven by a lack of parental knowledge regarding either the

995TV-related injuries in children

risk posed to children by large TVs or the availability of riskreduction measures or by an inaccurate parental perception ofrisk, but all 3 mechanisms are likely involved. Directedprevention campaigns that educate parents of toddlers mayavert TV-related injuries. In addition, the industry needs to beurged to ensure that proper securing devices are providedwith all new TV sets, with emphasis placed on ensuring allhome TVs are properly secured. In our region, both FirstNations families and recent immigrant families are atincreased risk and may have the most to gain from targeted-education strategies. The alarming increases in injury causedby TVs will only be stemmed by targeted, concerted,sustained education efforts for both parents and the industry.

References

[1] Television Audience 2009. In: The Nielsen Company 2010. http://blog.nielsen.com/nielsenwire/wp-content/uploads/2010/04/TVA_2009-for-Wire.pdf. Last accessed June 15, 2011.

[2] Nielsen's State of the Media 2010 - U.S. Audiences and devices. TheNielsen Company. http://blog.nielsen.com/nielsenwire/wp-content/uploads/2011/01/nielsen-media-fact-sheet-jan-11.pdf. Last accessedJune 15, 2011.

[3] Office of Energy Efficiency. 2007 Survey of Household Energy Use(SHEU) summary report. Natural Resources Canada 2007. http://oee.nrcan-rncan.gc.ca/publications/statistics/sheu-summary07/index.cfm.Last accessed June 16, 2011.

[4] CPSC Warns about TV, large furniture tip-over dangers. U.S.Consumer Product Safety Commission September 12, 2006: Release#06-254.

[5] DiScala C, Marthel M, Sege R. Outcomes from television sets topplingonto toddlers. Arch Pediatr Adolesc Med 2001;155:145-8.

[6] Gokhan S, Kose O, Ozhasenekler A, et al. Mortality and morbidity inchildren caused by falling televisions: a retrospective analysis of 71cases. Int J Emerg Med 2010;3:305-8.

[7] Jea A, Ragheb J, Morrison G. Television tipovers as a significantsource of pediatric head injury. Pediatr Neurosurg 2003;38:191-4.

[8] Ota FS, Maxson RT, Okada PJ. Childhood injuries caused by fallingtelevisions. Acad Emerg Med 2006;13:700-3.

[9] Rutkoski JD, Sippey M, Gaines BA. Traumatic television tip-overs inthe pediatric patient population. J Surg Res 2011;166:199-204.

[10] Scheidler MG, Shultz BL, Schall L, et al. Falling televisions: thehidden danger for children. J Pediatr Surg 2002;37:572-5.

[11] Sikron F, Glasser S, Peleg K. Children injured following TV tipoversin Israel, 1997-2003. Child Care Health Dev 2007;33:45-51.

[12] Suresh N, Harini G, Radhika R, et al. Head injuries in childrenresulting from the fall of television. Indian J Pediatr 2010;77:459-60.

[13] Yahya RR, Dirks P, Humphreys R, et al. Children and televisiontipovers: a significant and preventable cause of long-term neurologicaldeficits. J Neurosurg 2005;103:219-22.

[14] Child and Youth injury Review 2009 Edition. Public Health Agency ofCanada 2009. http://www.phac-aspc.gc.ca/publicat/cyi-bej/2009/pdf/injrep-rapbles2009_eng.pdf. Last accessed June 16, 2011.

[15] WISQARS Leading Causes of Death Reports 1999-2007. CDC 2010.http://www.cdc.gov/injury/wisqars. Last accessed June 16, 2011.

[16] Gottesman BL, McKenzie LB, Connner KA, et al. Injuries fromfurniture tip-overs among children and adolescents in the UnitedStates, 1990-2007. Clin Pediatr 2009;48:851-8.

[17] Bernard PA, Johnston C, Curtis SE, et al. Toppled television sets causesignificant pediatric morbidity and mortality. Pediatrics 1998;102:E32.

[18] Pereira J. 15 accidents that can kill kids. Parents Canada; 2007.August.

[19] Top Five Hidden Home Hazards. US Consumer Product SafetyCommission August 1, 2007:Release #07-256.

[20] Gaines J, Schwebel DC. Recognition of home injury risks by noviceparents of toddlers. Accid Anal Prev 2009;41:1070-4.

[21] Statistics Canada. Population by ethnic origin in Metro Vancouver,2006 Census. http://www.metrovancouver.org/about/publications/Publications/ethnic2006.pdf. Last accessed June 15, 2011.

[22] Gilbride SJ, Wild C, Wilson DR, et al. Socio-economic status andtypes of childhood injury in Alberta: a population based study. BMCPediatr 2006;6:30.

[23] Schwebel DC, Brezausek CM. Language acculturation and pediatricinjury risk. J Immigr Minor Health 2009;11:168-73.