Injuries among adolescents in Greenland: behavioural and socio-economic correlates among a nationally representative sample

Introduction

Injuries among adolescents has emerged as a global problem of public health importance. It is estimated by the World Health Organisation (WHO), that injuries account for almost one million deaths annually among young people under the age of 18 years worldwide, making it one of the most urgent adolescent health concerns (Djerboua, Chen & Davison, 2016). From the age of one year until adulthood, injuries account for a major portion of adolescent deaths. Even though ways to decrease the likelihood and severity of injuries are known, prevention and awareness still trail behind the current state of research (Peden et al., 2009). Considering that effective measures to prevent injuries do exist, it must be a focal point of injury research to determine what puts adolescents at risk for and protects them against injuries.

Injuries can be divided into two major areas: those which are violence-related or intentional injuries and unintentional injuries, so called ‘accidents’. As both the World Report on Violence against Children and the World Report on Child Injury Prevention state, prevention of violence and injury respectively is the key for improving adolescents’ well-being (Peden et al., 2009; Pinheiro & Children UNS-GsSoVA, 2006).

The most common causes of unintentional injuries among adolescents are road traffic injuries, drowning, and other causes (including categories such as smothering, asphyxiation, bites, hypo/hyperthermia and natural disasters). Risk factors these causes share are poverty, lack of supervision or care and often but not always male gender (Peden et al., 2009). In high-income countries road traffic and drowning related injuries account for more than two thirds of all unintentional injury deaths among people aged less than 20 years. To compare, in low- and middle-income countries other causes (including categories such as smothering, asphyxiation, bites, hypo/hyperthermia and natural disasters) and road traffic injuries account for more than 60% of all deaths in under 20 year olds (Harvey et al., 2009).

Interpersonal violence and injuries can lead to harm that threatens the physical and psychological integrity in ways that range from harmless to possibly life-threatening. The devastating effect of adolescent violence was described by David-Ferdon & Simon (2014), when they showed that more U.S. youth die from homicide each year than from cancer, heart disease, birth defects, flu and pneumonia, respiratory diseases, stroke, and diabetes combined. International research has shown that social marginalisation and adverse familiar circumstances (e.g. familiar abuse and neglect) as well as drug abuse and economic inequality can be considered as risk factors for being involved in violence (David-Ferdon & Simon, 2014; Kobusingye et al., 2010). Protective factors include a safe and care-giving familiar environment, strong social bonds outside of the family, an understanding of responsible drug use, and high socio-economic status (SES) (Bushman et al., 2016).

Despite the importance of research on injury and violence among Greenlandic adolescents, research on it remains sparse. The last English-language paper addressing the topic of adolescent aggression, which is one of many risk factors for adolescent injuries (Laflamme et al., 2002), was written by Schnohr & Niclasen (2006). The authors reported an increase in the prevalence of bullying in Greenlandic schools. In comparison to 35 other countries (located in Europe and North America) Greenland ranked 7th in terms of bullying others several times during the month preceding survey. Greenlandic schoolchildren showed above-average percentages for bullying others as well (Schnohr & Niclasen, 2006). Those numbers stress the need of further research to be conducted on the topic of injuries among Greenlandic adolescents and their socio-economic and behavioural correlates. The aim of this study was to examine the socio-economic and behavioural correlates for injuries among a nationally representative sample of school-attending adolescents in Greenland.

Materials and Methods

Results

Within the recall period (1 year), 30% of sampled adolescents reported having been injured, 24% of whom reported 1–2 injuries and 6% reported three or more injuries. Of the 1–2 Injuries group 57% were male and of the 3+ Injuries group 62% were male. The overall mean age of our sample was 13.4 (SD = 1.7). Those reporting alcohol misuse constituted 11.2%, 36% reported smoking tobacco and 12% reported being involved in a physical fight within the last 12 months. Being a victim of bullying was affirmed by 37.5% while 42.7% reported having bullied others. Of the participating adolescents 78.7% lived together with their mother. A total of 60.6% of the participating adolescents confirmed living together with their father. Living together with both parents was affirmed by 52.5%. A skilled maternal job was confirmed by 47.1%, while 41.3 % classified their father’s job as skilled. When asked to assess their own family’s wealth 73.5% assessed it as average, 7.3% as not well-off, and 19.2% as well-off. Practising VPA on 4 or more days a week was affirmed by 49.7%, while 50.8% practised VPA outside of school 4 days or more a week.

Table 1 depicts the distribution of the independent variables within the three categories of the dependent variable. Significance was found in the psychological and physical aggression category (Bullying victimisation, Bullying perpetrator, Physical fighting) and in social categories such as living together with their mother. Furthermore, being male and practising VPA outside of school settings yielded higher percentages of injury occurrence. Of the aforementioned results only living together with the mother had a protective effect against injury occurrence. All other factors were found to have increased the occurrence of self-reported injuries. No significant differences were found with respect to age, substance use, parental jobs or self-assessed wealth. Further factors that did not prove to be significant were living with fathers, VPA and talking to friends of the same and opposite gender.

After adjusting for all covariates, the first regression model delivered the following results (Table 2). Those who were injured once or twice within the one year preceding the survey were found to been bullied more often (OR = 1.58; CI [1.25–1.99]). Involvement in physical fighting increased the risk for injuries significantly (OR = 1.58; CI [1.14–2.20]). Males were overrepresented among those who reported being injured once or twice (OR = 1.54; CI [1.23–1.93]).

The significant associations between getting injured three times or more were being male (OR = 1.62; CI [1.05–2.51]), being bullied (OR = 2.29; CI [1.48–3.54]) and being involved in a physical fight (OR = 3.42; CI [2.09–5.59]).

The second regression model was adjusted only for age and gender. The results are summarised below in Table 3. Those who reported being in a fight (OR = 1.82; CI [1.33–2.47]), being bullied (OR = 1.81; CI [1.47–2.24]) and having bullied others (OR = 1.53; CI [1.25–1.89]) were more likely to have been injured once or twice. Males were also at a higher risk to be injured once or twice (OR = 1.70; CI [1.39–2.09]).

Significant associations were found between being injured 3 times or more and being male (OR = 2.13; CI [1.44–3.14]), physical fighting (OR = 4.47; CI [2.86–7.01]), being bullied (OR = 2.43; CI [1.65–3.57]) and bullying others (OR = 1.67; CI [1.13–2.45]). Living without a mother was found to be associated with the experience of three or more injuries as well (OR = 1.63; CI [1.05–2.52]).

Discussion

This study results suggest that the incidence in Greenland for having experienced a medically attended injury at least once within a one-year period of recall was 24.2%, whereas the average cross-national incidence for injuries among school-attending adolescents from all countries that participated in the HBSC survey ranks higher with 34.2%. The Greenlandic incidence for injuries among school-attending adolescents from 9 to 19 years for being injured three times or more was 5.6%. That is also lower than the average cross-national incidence of 9.4% (Currie et al., 2008). Most of the participating countries (full list of the participating countries: http://www.hbsc.org/) had a similar or higher GDP and HDI than Greenland at the time the survey was conducted. Greenlandic students also reported lower rates of witnessing other people being injured or killed in comparison to Iceland, the Faeroe Islands or Denmark (Karsberg, Lasgaard & Elklit, 2012). Transportation in many areas of Greenland during the summer can only be achieved by air or sea, whilst in the winter travelling on sea ice becomes a valuable traffic component as well (Grydehøj, 2014). The lack of large roads and car traffic compared to other highly developed countries may have implications for the occurrence of road traffic related injuries in Greenland. Since those make up around 57% of all unintentional childhood injury deaths, it could be a possible explanation for the observed lower injury incidence rate in Greenland (Peden et al., 2009).

Injuries have long been known to be more prevalent amongst males, who accounted for about 68% of all injury-related deaths in 2010 (Lozano et al., 2012). It was shown that male adolescents accounted for 57% (1–2 Injuries) and 62% (3+ Injuries) of all injuries in the respective groups.

We did not find evidence for injuries to be associated with alcohol misuse or smoking. Historically, Greenlandic adolescents have been more likely to engage in smoking and binge drinking behaviour (Niclasen & Bjerregaard, 2007). Even though the impact of inebriation on injury severity remains unclear, research has found that alcohol consumption and smoking have been linked to the suffering of injuries in adolescents (Cherpitel et al., 2015; Knapik & Bedno, 2018; Peden et al., 2009; Valdez et al., 2016). The mean age of our study’s participants was 13.4 years (SD = 1.74). It was pointed out that alcohol-attributable deaths among adolescents (0–15 years of age) only account for 0.1 percent of all adolescent deaths worldwide (Rehm & Shield, 2013). Skala & Walter (2013) also considered the age of 15 to be a threshold after which a rise in repetitive excessive alcohol consumption can be observed. Further, the HBSC international report from 2005/06 points out significantly higher rates of drunkenness among 15 year olds than among 11 or 13 year olds (Currie et al., 2008). This potentially illustrates that the present study sample may have been too young to be exposed to excessive alcohol consumption behaviour, hence significant associations could not be found. Lastly, it remains difficult to assess alcohol consumption through a questionnaire as the imprecise recall of alcohol consumption may drive an association to the null (Wilson et al., 2012). The prevalence of smoking in the present sample was found to be higher than in other high-income countries. Almost 26% of all participants reported smoking daily and 36% report smoking more or less regularly. In comparison, daily smoking prevalence for US American high school students ranks between 9% and 22% (Siqueira et al., 2000). The association between smoking and injury has been subset into four categories of reasoning: direct toxicity, distractibility, associated medical conditions, and confounding factors (Sacks & Nelson, 1994). Since the theories of direct toxicity and distractibility are either still questionable or only applicable to drivers, they do not aid to explain a possible relationship between injuries and smoking within our study’s cohort. Medical conditions associated with the toxic effects of smoking such as cardiovascular diseases or cancer, usually do not appear in teenagers but rather during adulthood (Burns, 2003). The most frequent injuries resulting from smoking are burns, but burns only make up about 3.3% of all unintentional childhood injury deaths in high-income countries (Peden et al., 2009; Sacks & Nelson, 1994). The missing applicability of the study conducted by Sacks and Nelson as well as the rather small share of burns in injury deaths could explain why the present study is silent on an association between smoking and injuries.

Consistent with existing research, bullying victimisation as well as bullying others was associated with increased injury occurrence. Being involved in bullying as a victim or perpetrator goes along with an increased risk of physical fighting (Rudatsikira, Muula & Siziya, 2008) which can lead to injuries requiring medical attention. Physical fighting was, as mentioned, a significant risk factor for suffering from injury in the present study.

The finding that living without mother increases risk for injury is supported by evidence which shows that an increase of adults living within a household goes along with a decrease in the risk of injury (Haynes, Reading & Gale, 2003). Parents also play a deciding role in teaching their children about injury prevention, thus the reduction of parental contact may put adolescents at risk for injuries (Wong & Breslin, 2017). Moreover, a study examining parenting differences between mothers and fathers came to the conclusion that especially uninvolved maternal parenting can lead to delinquency, depression, and decreased school commitment (Simons & Conger, 2007). These findings may aid to demonstrate the protective effect involved parenting can have on the risk of adolescent injury occurrence. Our results only partially support these claims since we could show that living without a mother was associated with an increase in injury occurrence whereas living without a father showed no association with injury occurrence.

This study provides valuable insight into a largely neglected area of global health research. To our knowledge this is the only study to have examined the epidemiology of injuries among a school-based sample in Greenland. Despite that, the current study also had the following limitations, since, besides giving valuable insight, it also remains confined and silent on several topics.

First, the study does only include students that go to school and it does not include adolescents that do not attend school at all. Research shows that those who are not in school may be at even greater risk for injuries, yet assessing this would go beyond the scope of this study (Dingli et al., 2012).

Secondly, some questions such as drug use, bullying, or physical fighting might have altered our findings in in the form recall bias, as the participants had to recall information within a rather large time frame which might have reduced their recall ability and accuracy (Celedonia et al., 2013; Wilson et al., 2012). Also, all the questions were self-reported, which may have been subject to social desirability and non-response bias (Acquah, Wilson & Doku, 2014).

Lastly, the cross-sectional nature of the dataset prevents us from drawing causal inference from our findings. The results must be interpreted within their own time context. Longitudinal data is necessary to establish the aforementioned causal pathways and to take into account the nature of change over time. Still our results have value not only for the historical aggregation of data, but they also play a role in understanding how young populations, and their behaviours, might change over time.

The body of research on injuries in circumpolar regions, especially Greenland, is rather limited at present. Therefore, this study also stresses the need for further examination of this topic.

Conclusions

The results of this study suggest that interpersonal relationships and conflicts are heavily linked with injury occurrence among school-attending adolescents in Greenland. Future programmes should consider the risk behaviours that were found to have a negative association with the adolescents’ well-being such as bullying and physical fighting. Moreover, further research would benefit from exploring the kinds of injuries Greenlandic adolescents suffer to identify factors that might modify the associations pointed out in this study.

Supplemental Information