Article Text

Download PDFPDF

Recent trends in television tip over-related injuries among children aged 0–9 years
  1. K J Murray1,
  2. R Griffin1,2,
  3. L W Rue III1,
  4. G McGwin, Jr1,2
  1. 1
    Center for Injury Sciences at UAB and Section of Trauma, Burns, and Surgical Critical Care, Department of Surgery, School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA
  2. 2
    Department of Epidemiology, School of Public Health, University of Alabama at Birmingham, Birmingham, Alabama, USA
  1. Mr R Griffin, Center for Injury Sciences, University of Alabama at Birmingham, 120 Kracke Building, 1922 7th Avenue South, Birmingham, AL 35294, USA; russellg{at}


Objective: To describe recent trends in television tip over-related injuries among children aged 0–9 years, and to compare injury rates with sales of newer digital televisions.

Methods: Digital television sales data were obtained from marketing data provided by the Television Bureau of Advertising. Data regarding television tip over-related injuries among children aged 0–9 years were obtained from the 1998–2007 National Electronic Injury Surveillance System. A Wald χ2 test, estimated from logistic analysis, was used to determine whether the distribution of injury types differed by age group. Pearson’s correlation was used to estimate the association between digital television sales and television tip over-related injuries.

Results: An estimated 42 122 (95% CI 35 199 to 49 122) injuries from television tip-overs were treated in US emergency departments from 1998 to 2007. The injury rate was highest for children aged 1–4 years (18.6/100 000). A majority of injuries (63.9%) involved the head and neck for children under 1 year of age, while a higher proportion of injuries among children aged 1–4 involved the hip and lower extremity (42.9% and 31.0%, respectively), and shoulder and upper extremity (16.8%) for children aged 5–9. A strong, positive correlation was observed between television sales and annual injury rates (r = 0.89, p<0.001).

Conclusion: Estimates of injury rates were similar to previously reported estimates, particularly for the increased proportion of head and neck injuries among very young children. While digital television sales were strongly correlated with increased injury rates, the lack of information regarding the type of television involved prevents inference regarding causation.

Statistics from

Furniture tip overs are responsible for 22 deaths and approximately 3000 injuries each year in the United States.1 Chief among these incidents are television tip overs, which occur most frequently among those between 0 and 5 years of age, and tend to result in head, extremity and thoracic injuries in all age groups.24 The majority of parents whose children experience a television tip over-related injury are unaware that such events pose a significant risk for their children; televisions are often placed in less stable places, thereby increasing the risk for injury.5

Since 2001, the number of television sets in US homes has increased by over 50 million sets—an increasing proportion of which are digital televisions—while the proportion of households with at least one television remained at 98.2%, indicating that many households are getting their second or third television.67 Despite the recent increase in digital television household presence, the dominant television technology remains the cathode-ray tube (CRT), which comprises the heaviest part of this type of television. This results in the centre of gravity being located in the front of the television, and as the screen sizes increase, the televisions become disproportionately front end heavy.8 As a result, even a small amount of force applied to the television or the furniture supporting it can induce a tip over.

With the advent of flat-panel televisions (FPTs) (such as liquid crystal display (LCD) and plasma), these televisions are lighter and thinner, and are therefore thought to significantly reduce the risk of tip overs.2 The slim design and lighter weight of FPTs, however, may allow children to better grasp and move the television, making it possible for them to tip the television over. Thus, whether the increase in FPTs increases or decreases tip over-related injury rates remains unclear.

To date there have been no studies that have examined the impact of the newer television technologies (and the resulting decrease in television set depth) on the risk of television tip over-related injuries. A previous report analysed rates of paediatric television tip over injuries in the United States, but used data from 2001 when the newer technologies were still prohibitively expensive.9 This study noted both an increase in paediatric television-related injuries and a modest increase in analogue colour television sales over the same period. A 2006 study reported that the most frequent injuries occurred with 20–30 inch televisions, which is an increase from a 2001 study’s finding of 19 inches as the most common screen size.45 This increase in screen size over time likely reflects the trend in the purchase of larger televisions. With the drastic price drops in television technologies in recent years, ownership of larger FPTs has increased.6 This study will examine the impact of the increased prevalence of these flat-panel televisions on the incidence of paediatric television-tip over injury since 1998, when FPTs were first introduced to the public.


Data sources

Data for this study were obtained from the 1998–2007 National Electronic Injury Surveillance System (NEISS). Details regarding NEISS methodology have been described elsewhere.1011 Briefly, NEISS is a probability sample of hospitals with emergency departments, and provides national estimates of consumer product-related injuries by collecting information for patients with a consumer product-related injury who seek medical attention at one of the sampled hospitals. All injuries involving televisions (code 0572) among children aged 0–9 years were selected and reviewed to ensure that the injury could be attributed to a television tip over.

Statistics for digital and non-digital colour television (DTV) sales were obtained from the Trends in Television documents published by the Television Bureau of Advertising.7 Sales of non-DTVs were defined as the number of sets sold from the factory, and sales of DTVs were defined as the number of sets sold to dealers. Since data regarding FPT sales were not available, sales of DTVs, which encompass both CRTs and FPTs, was used as a proxy for FPTs.


The rate of television tip over-related injury (per 100 000 persons) was calculated as the number of injuries divided by the population estimate derived from annual US census estimates (from 1998 to 2007) provided by the United States Census Bureau.12 Injury rates with associated 95% CIs, calculated from standard error estimates accounting for sampling variation, were estimated by age, gender and race. The anatomic location of injury was categorised into four regions: head and neck, trunk, shoulder/upper extremity, and hip/lower extremity. The distribution of injury types (eg, fracture) was compared among regions overall and by age group. A Wald χ2 test, estimated from a logistic regression model, was used to determine whether the distribution of injury types differed by age group. Poisson regression was used to assess whether an annual trend of tip over-related injury rate was present. Pearson’s correlation was used to estimate the association between DTV sales and television tip over-related injuries. Analyses were weighted to account for the probability sampling of the NEISS and were conducted using SAS V.9.2.13


There were an estimated 42 122 (95% CI 35 199 to 49 122) total cases of television-related injuries among children aged 0–9 years in the United States between 1998 and 2007, resulting in an estimated injury rate of 10.58 per 100 000 persons (95% CI 8.85 to 12.35) (table 1). The rate of injury was higher for boys (12.50 per 100 000, 95% CI 10.55 to 14.45) than for girls (8.58 per 100 000, 95% CI 7.02 to 10.14). The injury rate was highest among blacks (13.22 per 100 000, 95% CI 8.22 to 18.22), followed by whites (8.00 per 100 000, 95% CI 6.10 to 9.90), and was lowest among other races (4.62 per 100 000, 95% CI 2.67 to 6.56). Rates of injury were highest in children aged 1–4 years (18.59 per 100 000, 95% CI 15.76 to 21.42) and lowest in children aged <1 year (1.55 per 100 000, 95% CI 0.79 to 2.31).

Table 1 Estimated number of television-related injuries, and rates (per 100 000 persons) and 95% CI by selected demographic characteristics, 1998–2007

Children aged 1–4 years were the most likely to have a television tip over-related injury (69.5%) compared to those aged 5–9 years (29.0%) and <1 year (1.5%) (table 2). Those aged 1–4 years were more likely to have an injury to the hip and lower extremity (42.9%) compared to those aged <1 (22.7%) and 5–9 years (31.0%) (p = 0.003). Children aged 5–9 years were more likely have the shoulder and upper extremity region injured (16.8%) compared to children aged 1–4 (8.9%) and <1 (1.0%) (p<0.001). Among shoulder and upper extremity injuries, children <1 year were more likely to be diagnosed with a contusion (100%), children 4–9 years were more likely to suffer a fracture (40.5%), and children aged 5–9 years were more likely to be diagnosed with a contusion/abrasion (54.5%). All of these differences were significant (p<0.001); however, no statistical difference existed between shoulder and upper extremity injuries between children aged 1–4 and 5–9 years.

Table 2 Characterisation of television-related injuries by age, 1998–2007

The rate of television tip over-related injuries increased from 8.88 per 100 000 persons in 1998 to 13.15 per 100 000 persons in 2007. While the confidence limits for the rates were similar across years, there was a statistically significant increasing annual trend in the injury rate (p<0.001) (table 3). The increase in digital television sales from 1998 to 2007 correlates very strongly with the increase in television-related injuries in the same period (r = 0.89, p = 0.001).

Table 3 Television tip over-related injury rates by year among children aged <1 to 9 years


The results of the current study indicate an increase in the rate of television tip over-related injuries from 1998 to 2007. These injuries primarily occur in children aged 1–4 years, and are coincident with the increase in DTV sales over the same period. The results also suggest that the distribution of body part injured differs by age, with children <1 year more likely to suffer a head and neck or trunk injury.

There are several possible explanations for the increase in television tip over-related injuries in this time period. First, it is possible that the increase in sales of FPTs is at least partly responsible for the increase in television tip over-related injuries. This assertion, however, is diluted by the fact that CRTs still dominate the market, and only recently have FPT sales started to gain ground relative to CRT sales.14 Therefore, the more plausible explanation is that the number of multi-television households in increasing. It has been estimated that the number of televisions per household has increased from 2.4 in 1998 to 2.7 in 2006, while the percentage of households that own at least one television has remained stable over the same period.6 A survey of households with at least one television reported that 37% of households had a television in a family/recreation room and 20% had a set in a child’s bedroom.15 Thus, the more televisions are in a household, the more a child’s exposure to risk of injury due to television tip over, particularly if the television is in a room where the child may spend a majority of time.

The results of this study are similar to previous reports of television-related injuries being highest among children aged 1–4 years,16 and that a higher proportion of head and neck injuries are observed among younger children.4922 It is unlikely that children under 1 year of age can pull the television onto themselves; it is more likely that the child bumps into the television stand, causing the television to fall onto the child. The first body part hit by the falling television will probably be the head or neck due to the child’s height in relation to the height of the stand. Older children are more likely to be able to pull the television onto themselves or be standing when hitting the stand and causing the television to fall. This could explain the higher proportions of shoulder, upper extremity, hip and lower extremity injuries observed among children aged 1–4 and 5–9 years.

The higher injury rates observed in boys is not surprising, as previous research has suggested that boys are at higher risk of injury than girls.1719 This is particularly true for residential injuries, in which boys have an injury rate 1.4 times that of girls.16 While the same study16 reported that white children and adolescents have a similar rate of residential injuries compared to black children, the current study reports that blacks have the highest rate of television tip over-related injury. This may, however, be due to the fact that the current study included only children 9 years of age or younger, whereas the previous study included children and adolescents aged <1 to 19 years. Research suggests that non-white children aged 9 years or younger have higher all-cause injury rates compared to white children.2021 It has been suggested that such racial differences may not be real but rather reflect confounding due to socioeconomic status.23

Narratives that accompany each NEISS entry frequently mentioned the tip over being caused by a child climbing on the furniture housing the television, and several (108/1991) entries mentioned that the television was sitting on top of a dresser. Since dressers are not designed to support a television, they may be less stable than traditional television stands, and thus may increase the risk of a television tip over-related injury. Television stands, on the other hand, are required to meet specifications for stability and risk of television tip over, and if they fail to do so, may be recalled. Since 1998, three companies have recalled their television stands because of an unacceptable risk of television tip over.2426

The results of this study should be interpreted in light of several limitations. As with any data based on a probability sample, there are concerns about statistical reliability of weighted estimates based on small sample sizes. With NEISS, weighted estimates of <1200 are considered statistically unreliable, and one should be cautious in making inferences for such estimates. The data regarding number of televisions were derived from data collected from factories and dealers. As a result, it is possible that sales may not be a true reflection of televisions in households as dealers and factories could sell the televisions to non-residential customers such as businesses, schools or hotels. Also, there are no data on the Injury Severity Score or Glasgow Coma Scale score (for head injuries); as a result, there is no way to determine whether the severity of injuries has changed. While the injury rate has been increasing, since FPTs are lighter than CRTs, the severity of the injuries may be decreasing if the increase is related to FPTs. Also absent from many of the narratives was information regarding the specifics of the event, such as the size of the television involved, the furniture that was used to support the television, or the height of the furniture supporting the television. Additionally, there was no indication of whether the television was an FPT or CRT; while digital television sales increased during the study period, the effect of the increased presence of these televisions in the home is unclear. That is, it is possible that the increasing annual rate may still be driven by the presence of CRTs in homes, and the true association between FPTs and paediatric injury may not be realised until later years when FPTs are the predominant television type in households.


While the number of sales of DTVs was strongly correlated with an increased television tip over-related injury rate, without the ability to determine what type of television (ie, CRT of FPT) was involved in the injury, we cannot definitively say at this point that the increase is due to DTVs, more specifically FPTs. By addressing this limitation, future research can help to elucidate whether the increasing presence of FPTs in households alters tip over-related injury rates.

In spite of the current study’s limitations, the results suggest that television tip over-related injury rates have been increasing from 1998 to 2007. Because of this, we suggest that the public be educated of the dangers associated with placing televisions in areas that can be reached by children, particularly those aged 1–4 years, who can walk but may lack the refined motor skills to avoid bumping into the television or stand. One of the benefits of FPTs is that their flat shape allows them to be placed on walls, out of the reach of children and on a stable surface that decreases the chance of tip over. Since this may not be a viable choice for all families, manufacturers should not only make buyers aware of the ability to use safety devices to prevent television tip over, but should also incorporate more safety features to ensure that tip over-related injuries do not occur.

What is already known on this topic

  • Television tip over-related injuries are a major cause of head and neck injuries among children.

  • The rate of injuries differs by age.

What this study adds

  • The rate of television tip over-related injuries has risen over the years concomitantly with a rise in sales of flat-panel televisions, although this may be due to an increasing presence of tube televisions in rooms in which children spend more time.

  • The body region injured by the toppled television differs by age, with an increased likelihood of hip and lower extremity injuries more likely among children aged 1–4 years and of shoulder and upper extremity injuries among children aged 5–9 years.



  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.