The Convention on the Human Rights of the Child stipulates that countries have an obligation to take all necessary steps to protect children from all forms of injury.1 Unintentional injury is the leading cause of death in children (1–14 years) in New Zealand; there are approximately 8.4 unintentional injury-related deaths per 100,000 children (aged 0 to 14) annually.2 Ethnic and gender disparities are evident, with the unintentional injury mortality rate for children 3.5 times higher among Māori than non-Māori, and 1.6 times higher in male children compared to female children.3 A 2010 report by O’Dea and Wren estimated the economic cost of a child fatality at $8.05 million, significantly higher than that for an adult fatality ($5.74 million).4 The most recent United Nations International Children’s Emergency Fund (UNICEF) report on child poverty in rich countries, which includes injury data, indicates New Zealand has the highest rate of childhood fatal injury among 21 OECD countries,5 highlighting the need for renewed efforts to improve child safety to a level that is at least consistent with other high-income countries. Furthermore, a study comparing New Zealand child injury safety with 24 European countries found that New Zealand ranked 25th for Moped/motor scooter safety, 23rd for passenger/driver safety and 19th equal for pedestrian safety.6

A potential effective child injury prevention strategy is the provision of child safety messages via print and online news media. News media, unlike TV mass media advertising, has the advantage of not having to compete with commercial advertising for viewer attention,7 and is widely accessible with the potential to raise public awareness of childhood injuries and how to prevent them. A 2004 systematic review of the effectiveness of mass media campaigns for reducing alcohol-involved crashes found that a well-executed mass media campaign can contribute to a reduction in alcohol-related crashes.8 Similarly, a US study found that news media accounts played a significant role in a campaign to increase helmet usage among school-age children with rates increasing from 4% among children under 15 years to 54% among those ages five to nine years, and 38% among 10–14 year olds over 51 months.9

A study by Kool et al of New Zealand fatal domestic fire-related child injury deaths found that all 14 fatal fire-events recorded by the fire service during a 10-year period were reported in the national newspaper with a high degree of detail and accuracy.10 However, only around one quarter of articles informed readers of specific measures that could prevent such events, suggesting a potential missed opportunity.

The potential of “teachable moments” for advocating for health behaviour change is well documented.11 Health professionals can potentially play an important role in promoting prevention messages to accompany media accounts of public health interest. For instance, the need for greater advocacy partnerships between the media and public health professionals for awareness of skin cancer prevention12 or in changing the way that motor vehicle crashes are framed.13

The aim of this study was to review media accounts of fatal child unintentional injury events reported in leading New Zealand newspapers to explore the content of these reports and their potential for the delivery of evidence-based injury prevention messages. The findings have the potential to inform changes to the way in which print and other media outlets report serious child injury events and to ensure that public health professionals are more proactive in engaging with key media outlets (print, online and television) to promote child safety agendas.

Method

Using a content-analysis approach we examined New Zealand print media accounts of fatal unintentional child (0–14 years) injury events over a five-year period (1 January 2011 to 31 December 2015). New Zealand’s four largest daily newspapers (the New Zealand Herald [readership: 549,000], The Dominion Post [readership: 256,000], The Press [readership: 188,000] and The Otago Daily Times [readership: 93,000])14 were searched for relevant articles. Two news clippings services (Media-Monitors and Isentia) accessed via Safekids Aotearoa were used to locate the relevant articles. Eligible articles included those that were published during the five-year period of interest and reported the death of a child or children (<15 years of age) as a result of unintentional injury. Treatment injuries were excluded. We only analysed accounts that were published in the immediate period following the injury event. In situations where the same event was mentioned by multiple sources the event was only counted once in the analysis.15

Content analysis is the “systematic, objective, quantitative analysis of message characteristics”, and can be used to examine messages in a range of mediums, including: advertising, face-to-face human interactions, blogs, political speeches and news media.16 For the purposes of this research, information extracted from each media account included: the newspaper/s which featured the media account, the date it appeared, prominence of the media account in the newspaper (page number, presence of a photograph) and presence or absence of prevention messages. Information specific to the injury event included: the date and location of the event, mechanism of injury, mention of prevention measures, and the age, gender and ethnicity of victims. The mechanism of injury was classified by members of the research team using the Centre for Disease Control’s International Classification of Diseases (ICD-10) external cause codes (V01-X59).

Prevention messages were classified into two categories for clarity: clear or ambiguous. ‘Clear’ messages were those that directly stated the prevention method, for example “police recommend parents keep children at arm’s length when in and around water”. Prevention messages commonly quoted sources such as the police or SafeKids Aoteoaroa. ‘Ambiguous’ messages were those that loosely referenced prevention or carelessness involved, for example a newspaper account where alcohol involvement in a road traffic crash that killed a child was noted, however there was no direct mention of the known risks associated with driving under the influence of alcohol or the current legal blood or breath alcohol levels. In addition, as a proxy to assess the completeness of case ascertainment by the media accounts, the total number of injury deaths for the two leading causes of injury (transport [V01-V09] and drowning [W65-W74]) recorded in the National Injury Query System (NIQS) were compared with the number of deaths captured by the media accounts. The NIQS (http://psm-dm.otago.ac.nz/niqs/) is maintained by the University of Otago and is a simple online injury query system which allows the user to produce New Zealand non-fatal and fatal injury-related statistics based on data from national data collections maintained by the Ministry of Health (the Mortality Collection, and the National Minimum Dataset [NMDS], which consists of public hospital discharge data). Due to the lengthy coronial process in New Zealand, there is usually a two-year lag period before the public release of the mortality data. Therefore, we compared the media account data for the two most common causes of fatal unintentional child injury with the three most recent years of complete mortality data (2011–2013) contained in the NIQS.

Results

Over the five-year period reviewed, a total of 122 unique unintentional childhood fatal injury events covering 133 fatalities were reported by the four leading newspapers reviewed, an average of 47.6 per year. The four leading newspapers contained a total of 242 articles relating to these events. Of the four major papers, the New Zealand Herald reported the most articles relating to the events of interest (37.2%; n=90) while The Press produced the least (14.5%; n=35).

Aside from a peak in 2012, there appeared to be a downward trend in the number of fatal child injury-related media accounts reported in the four major newspapers in New Zealand for the five-year period reviewed (Figure 1).

Figure 1: Trends in the number of unintentional fatal child injury event deaths covered per year in media accounts by the four major New Zealand newspapers, 2011–2015 inclusive (n=122).

Characteristics of fatal unintentional child injury events reported in the media

The leading mechanisms of injury in the newspaper accounts of unintentional child injury deaths were transport-related events (56.5%; n=69), followed by drowning (23.0%; n=28) (Table 1). Among the transport-related deaths, ‘car occupant’ events were the most common (n=26/69), followed by ‘pedestrian injured in collision with vehicle’ (n=21/69) and ‘all-terrain vehicle occupant’ (n=7/69) events. In all the drowning deaths, the location where the drowning occurred was reported. The most commonly reported location where injury events occurred was in lakes or rivers (n=9) followed by the beach or sea (n=8).

Table 1: Mechanism of injury of unintentional childhood injury fatal events reported by the four major New Zealand newspapers, 2011–2015 inclusive.

‍

* ‘Pedestrian injured in collision with vehicle’ includes: pedestrian non-traffic vs car, bus, truck, ute; and pedestrian traffic vs car, motorbike.
† ‘Pedal cycle’ includes: pedal cycle non-collision, pedal cycle vs car, bus.
†† ‘Other transport’ includes: go-kart; plane crash; non-collision car passenger, van passenger; motorcycle non-traffic; passenger non-traffic.
‡ ‘Other’ includes: cutting/piercing, electrocution, fall, firearm, allergy to insects and arachnids, exposure to animate mechanical forces, forces of nature-excessive heat.

The majority of the fatal child injury events reported in the newspapers were single fatality events (94.2%, n=115). There were seven multiple fatality events: five road traffic events resulting in 11 deaths, the Christchurch earthquake in 2011 accounted for four deaths, and a single poisoning event resulting in three deaths.

Completeness of media accounts

The majority (85.0%; n=113) of child fatalities were identified by name in the media accounts reviewed (Table 2). Suffocation/strangulation events were least likely to report the child’s name (n=1/3). In almost all (97.7%; n=130) of the accounts, the victim’s age was reported, and in 95.5% (n=127) gender was reported. Ethnicity was reported for only six victims (5%) who were involved in four events (two drownings, one poisoning, and one fire/flame-related death).

Table 2: Completeness of reporting of unintentional fatal childhood injury events by injury type in leading New Zealand newspapers, 2011–2015 inclusive.

‍

* ‘Other’ includes: cutting/piercing, electrocution, fall, firearm, allergy to insects and arachnids, exposure to animate mechanical forces, forces of nature-excessive heat.

For the three-year period reviewed where corresponding national mortality data was available (2011–2013), there was complete case ascertainment for the transport-related deaths (n=58) reported in the media, and all but one of the drowning deaths (n=27).

Prevention messages

Just under one half (46.3%; n=112) of the media accounts located included some form of prevention message, in 53.7% of accounts there were no prevention messages (n=130) (Table 3). Clear prevention messages were included in 20.3% of accounts (n=49) such as recommending parents stay within arm’s reach of accompany young children when playing in the water, checking the driveway before moving a vehicle or keeping cots away from curtain cords.

Table 3: Presence or absence of prevention messages in media accounts of fatal child injury events in New Zealand’s four major newspapers, 2011–2015 inclusive (n=242).

‍

The New Zealand Herald had the highest proportion of media accounts with clear prevention messages included (26%; n=23), while The Otago Daily Times (ODT) had the least (14%; n=9). For the five leading injury mechanisms reported (transport, drowning, caught/crushed or jammed, fire/flame, poisoning), media accounts relating to poisonings were most likely to have clear prevention messages included (n=3/7), followed by drownings (n=14/53). Transport-related fatalities were the most commonly reported events, however, only 15.6% (n=21/134) of these accounts contained clear prevention messages. Of interest, there were no prevention messages included in any of the media accounts of electrocution deaths (n=4).

The majority of media accounts were considered to be in prominent locations within the newspaper, with 66% (n=158) appearing in the first three pages. Page three was the most common page for media accounts about unintentional fatal childhood injury events to appear. The mean page number where fatal injury media accounts appeared ranged from page 3.1 (the Press) to page 6.9 (ODT) (Table 4). A total of 36 accounts (14.3%) of fatal child injury events appeared on the first page of the newspapers reviewed. When looking only at accounts which included prevention messages, the most common page number where the account appeared was page three, while the mean page number ranged from 3.2 (Dominion Post) to 9.4 (the ODT). Thirty-six articles (15%) were on the front page, with 12 of them including clear prevention messages

Table 4: Prominence of media accounts of fatal child injury events in New Zealand’s four major newspapers, 2011–2015 inclusive (n=242).

‍

Just over one third (36%; n=86) of media accounts included at least one picture (Table 4). The newspaper most likely to include a photo was the New Zealand Herald (49%; n=44), and the least likely was The Press (29%; n=10). Just over half of the accounts with prevention messages also included pictures with this most commonly occurring in the Dominion Post (n=9/11).

Discussion

This analysis of media accounts of unintentional fatal child injury events in New Zealand reported in the four major newspapers during a five-year period found that while the cause of injury and demographic details of the victims were reported in most articles, only 20% included clear prevention messages, highlighting a missed opportunity for the dissemination of injury prevention messages. Among the most common causes of injury reported in the media accounts, fatal events as a result of poisoning were most likely to include prevention messages, followed by drowning and then transport-related events. Case ascertainment was high for the two most common causes of child fatal unintentional injuries reported, with all of the transport related deaths, and all but one of the drowning deaths captured by the NIQS (collates national mortality statistics) reported in the newspapers reviewed.

The strengths of this study include employing methods adapted from similar international studies.7,10,15,18 However, the findings need to be considered in light of some limitations. We only reviewed the four major daily papers; it is possible that newspapers with a more local distribution or those published in the weekend might report these events in a different way. There is the potential for misclassification bias in our coding of the mechanism of injury from the media accounts as we were limited to the information contained in those accounts. We acknowledge that some deaths classified as injury deaths may have in fact been intentional deaths (eg, assault, filicide or filicide-suicide). We were only able to investigate the completeness of case ascertainment in the media accounts for the first three years of the period being reviewed due to the two-year delay in mortality statistics being publicly available in New Zealand. Analysis of page numbers of articles can be affected by daily variations in thickness of the paper, and affected over time by redesigns in sectioning of the papers. We did not analyse the word count of each newspaper account, which may be of relevance to the prominence of the article. Despite these limitations, this study provides a contemporary snapshot of the state of media reporting of unintentional fatal child injury events in New Zealand.

The low frequency of prevention messages included in media accounts of fatal injury events found in this study is consistent with other published research.10,16,18 A US study of the presentation of injury deaths in the press found that clear prevention messages were reported in only 8% of articles.18 Another US study by Smith et al of newspaper coverage of residential fires found that 36% of deaths (all ages) were reported with accompanying prevention messages.19 A New Zealand study conducted in 2003, investigated the accuracy and public health relevance of the reporting of unintentional fire-related childhood deaths and found injury prevention messages were reported in 29% of news articles.10 This is higher than the 13% found in the present study for fire/flame related deaths.

Transport-related events were the leading mechanism of fatal child injuries reported over the five-year period reviewed in this study. Disappointingly, only 30% of these media accounts included clear prevention strategies. A US study investigating media accounts of fatal motor vehicle crashes, found only 20% of accounts mentioned the use or non-use of seatbelts.13 Drowning was the second most common cause of reported fatal child injury reported in the present study, however, only just over a quarter of these accounts included prevention messages.

Previous research has confirmed that the inclusion of photographs in media accounts draws attention to the text and encourages more extensive reading of the article.17 In the present study, media accounts with prevention messages were more likely to include photographs than those that did not (51% cf. 36%), which may have increased the likelihood of them being read and therefore provided a vehicle for prevention message dissemination.

A New Zealand study by Shepherd et al identified a number of evidence-based injury prevention policy and legislative actions that if implemented in New Zealand could result in a significant reduction in child injury mortality rates with the potential to result in an estimated 81 fewer child injury deaths every year.6 The authors highlight the prioritisation of vehicle passenger safety, pedestrian safety and water safety as “Do Now” policy recommendations. However, despite this, in the present study the inclusion of prevention messages in accounts of transport-related (16%) and drowning (26%) child fatalities was low.

The case ascertainment for the two most common causes of fatal child injury reported in the newspaper accounts reviewed (transport-related and drowning) against national mortality data (NIQS) was encouragingly high. For drowning events, there was almost complete case ascertainment (n=27/28), higher than the 57% found for child (0 to 14 years) drowning deaths reported in a Finnish study that compared the completeness of drowning reporting in newspapers with Statistics Finland data,19 and the 78% reported for drownings (<19 years of age) in a US study examining the accuracy of newspaper accounts with medical examiner reports.20 The high case ascertainment in the reporting of these injury types found in the present study may in part be due to the perceived newsworthiness of these events. This highlights the potential for the inclusion of evidence-based drowning prevention messages in media accounts.

Media accounts of fatal child injury events could be as an effective means of disseminating prevention messages in New Zealand through the development of a set of media guidelines, which include appropriate rules for inclusion of appropriate evidence-based injury prevention messages. An international review of the guidelines for the reporting of violence against women found that most recommend the inclusion of local statistics to highlight the size of the problem.21

The association between unintentional child injury and socio-economic status is well established, with a range of risk factors identified, including: income, family structure, maternal education, accommodation-related factors.22 Injury risk is affected by socio-economic status in a range of ways such as in poor households parents may not be able to afford safety equipment (eg, smoke alarms, child car restraints, etc.) or they may be exposed to more harzardous environments (eg, fast moving traffic, lack of space for safe play, etc.). How the media presents child injury has an impact on “public attitudes”, with many arguing that journalists play a significant role in constructing what society considers acceptable or unacceptable.23,24 The UNICEF report on children’s rights and the media, recommends that all media accounts of children should portray children as subjects rather than objects, and avoid stereotypes or misconceptions in the reporting of the ethnicity of children from lower socio-economic communities.25

Gibson and Zillman in their study of perceptions of visual information in news reports in the US, found that photographic exemplification in media accounts of diseases of any given ethnic group increases the perceived risk to that group while exemplifying two groups equally produced similar estimates of risk for the groups.17 This underlines the importance of the careful choice of photographs to accompany media accounts to avoid the stigmatisation through the appropriate visual representation of ethnic groups.

This study only reviewed New Zealand’s four major daily newspapers; future research could look at the reporting of fatal child injury events in community-specific papers to gain a more comprehensive view of media reporting of these events and to see what type of events are picked up nationally. In addition, an analysis of the headlines of media accounts of these events would provide an additional perspective on the discourse around events of this nature.

The findings of this study highlight the need for changes to the way in which print media outlets report serious child injury events. It emphasises the need for greater advocacy relationships between the media and public health professionals to promote appropriate prevention strategies, and the development of media guidelines for the responsible reporting of unintentional fatal child injury events.

‍