Extent, patterns and demographic correlates for physical fighting among school-attending adolescents in Namibia: examination of the 2013 Global School-based Health Survey

Introduction

Physical fighting is consistently associated with serious injury and among adolescents is particularly problematic as it is strongly associated with both bullying (perpetration and victimization), family violence and weakened academic achievement (Pickett et al., 2002; Pickett et al., 2005). Additionally, individuals who experience physical fighting in adolescence often experience substance use, poor relationships with peers and decreased overall life satisfaction, that extend well into adult life (Krug et al., 2002; Rudatsikira, Muula & Siziya, 2008a; Patton et al., 2009; Sousa et al., 2010).

Several risk factors for physical fighting have been identified in adolescents, such as male sex, poor mental health, drug use, bullying and younger age (Alikasifoglu et al., 2004; Rudatsikira, Muula & Siziya, 2008a; Pickett et al., 2012). However, these studies mainly take place in high income countries and quality information regarding the prevalence and general risk factors for physical fighting is lacking in low-income and sub-Saharan African countries. Detailed contextual information on physical fighting risk factors is essential in order to better target adolescents in sub-Saharan African countries who may at particular risk of participating in physical fighting.

In 2004, 51% of adolescents in Namibia reported having been in at least one physical fight that year (Rudatsikira et al., 2007b). We aim to provide an updated overview of the prevalence and risk factors for physical fighting among adolescents in Namibia using the Namibia Global School-based Health Survey (GSHS) conducted in 2013. The current study builds upon a previously published study by Rudatsikira et al., where we aim to examine a more contemporary cohort of Namibian adolescents who exhibit physical fighting behavior and include additional risk factors not previously included in the Rudatsikira et al. (2007b) study.

Methods

Results

4,531 students in grades 7 to 12 aged 13 to 17 (52.7% female) completed the survey questionnaire. Twenty-one students were excluded with missing ‘physical fights’ information, resulting in a final sample size of 4,510 adolescents. For comparison purposes we measured 1 or more and 2 more physical fights in the past year, but use 2 or more physical fights as the main outcome throughout the study. Within the recall period, 32.5% (unweighted count: 1,493) of participants reported being involved in 1 or more physical fights and 16.9% (unweighted count: 785) of participants reported being involved in two or more physical fights in the past year, most of whom were male (59.7%). Parents or guardians understanding of respondent’s problems and worries was reported as either never, rarely, or sometimes by 59.6%. Having made a plan about how one would attempt suicide was reported by 25.5%. Regarding use of addictive substances and having sex; 21.0% reported early sexual debut, defined as having had sex before turning 15 years old. Alcohol use, marijuana use and smoking cigarette for one or more days, during the past 30 days was reported by 33.6%, 5.3%, and 9.6% respectively.

Table 2 provides the weighted distribution of selected factors according to physical fighting category. In the bivariate analyses, eleven out of the seventeen variables were statistically significantly associated with participants who had been involved in two or more physical fights. The age and sex adjusted analyses (Table 3) for all the variables found to be statistically significant in bivariate analysis, revealed statistically significant associations for all selected variables.

Table 4 provides results of adjusted analysis for all covariates in the model. Compared to those who did not report being involved in physical fighting, those who had been involved in physical fights were males (OR 2.06; CI [1.54–2.76]), more likely to have made suicide plan (OR 2.10; CI [1.66–2.65]), were truant (OR 1.60; CI [1.08–2.38]), food deprived (OR 1.90; CI [1.41–2.57]), having been bullied (OR 2.35; CI [1.79–3.09]), had early sexual debut (OR 1.43; CI [1.20–1.71]), and were cigarette smokers (OR 1.99; CI [1.19–3.32]). While age bestowed protection against involvement in physical fighting (OR 0.89; CI [0.84–0.94]). Anxiety, alcohol use, and marijuana users were not found to be statistically significantly associated with physical fighting. The goodness-of-fit test revealed that this was a good multivariable logistic model for physical fight in Namibian students (F: 0.93, p-value: 0.5264). With the exception of having been truant and smoked cigarettes, all other covariates that were statistically significant at p-value of less than 0.05.

Discussion

This study described the prevalence and factors associated with physical fighting among school-attending adolescents residing in Namibia. We found the prevalence of participating in at least two physical fights among school attending adolescents in Namibia was almost 17%. Factors associated with an increased odds of physical fighting included having a suicide plan, anxiety, truancy, food deprivation, being bullied, early sexual debut and cigarette smoking. Increased age was associated with a decreased odds of physical fighting.

Namibia was found to have a relatively low prevalence of physical fighting, especially compared to other sub-Saharan African countries, and even slightly lower when compared to some high-income countries (Pickett et al., 2012). For example, a similar survey in Canada found that 19.3% of adolescents participated in two or more fights (Djerboua, Chen & Davison, 2016). Interestingly, relative to other sub-Saharan African countries, rates of physical fighting appear to be particularly low in Namibia (Rudatsikira et al., 2007b; Acquah et al., 2014a). For reference, in Ghana, approximately 32% of adolescents participated in two or more physical fights in the year prior to the survey (Acquah et al., 2014a, Acquah et al., 2014b). A similar study in Namibia found that 50% of adolescents had been in one physical fight in the last 12 months from the time of the survey, compared to the present study which found only 32.5% (Rudatsikira et al., 2007b). This may reflect an improvement in physical fighting between 2007 and 2013, or differences in respondent characteristics between the years.

Cultural or educational differences in Namibia may result in lower rates of physical fighting when compared to other sub-Saharan African countries. Although the human development index in Namibia is low, and poverty as well as income disparity have been shown to be associated with increased physical fighting, there may be other positive influences that discourage physical fighting among adolescents in the country. For instance, school environment has been shown in association with physical fighting among school-aged children (Larsen, 2003; Limbos & Casteel, 2008). Namibia has free education and spent 9.2% of GDP on education in 2014, perhaps indicating a positive school environment that may mitigate physical fighting in the country (Ministry of Arts, Culture and Education, 2017).

We found several factors associated with physical fighting. Physical fighting is predominantly present in young adolescent males; Sex and age have both been established as common risk factors for physical fighting (Pickett et al., 2005; Pickett et al., 2012). In line with previous literature in low-income countries, we identified the following factors in association with increased physical fighting: having a suicide plan, anxiety, truancy, food deprivation, being bullied, early sexual debut and cigarette smoking (Rudatsikira, Muula & Siziya, 2008b; Rudatsikira et al., 2008; Celedonia et al., 2013). Previous studies have demonstrated clustering of risky behaviors, such as truancy, cigarette smoking and alcohol abuse, which may explain the association of physical fighting with suicide planning and truancy (Petridou et al., 1997; Pickett et al., 2002; Acquah et al., 2014a; Yang et al., 2017). Being bullied has been previously shown in association with violence and physical fighting, indicating that those individuals who are victimized are more likely to perpetuate that violence elsewhere (Rudatsikira et al., 2007a; Pickett et al., 2012). This may also be associated with factors of low income, such as food deprivation. Lower socioeconomic status and income disparities have been found in association with physical fighting in other studies (Pickett et al., 2012).

Interestingly, none of the factors we hypothesized to have a positive influence on adolescents, such as supportive parents, positive peers, physical activity and close friends, were significantly associated with physical fighting on descriptive analyses, although having supportive parents was close (p-value = 0.06). In this study, loneliness was associated with decreased odds of physical fighting. One hypothesis may be that adolescents with a smaller social circle are less likely to engage in delinquent behavior such as physical fighting (Haynie, 2002).

Although we found relatively low rates of physical fighting in Namibia, this number is still high at almost 17% of adolescents participating in at least two fights and represents an area for intervention. By identifying factors associated with physical fighting policy makers are able to better target populations who may be at particular risk of participating in this risky behavior, such as younger adolescent males and those who may be dealing with other social issues, such as anxiety, suicidal thoughts and bullying. School programming has shown success in reducing physical fighting among school-aged youth and those programs that specifically target high risk youth are particularly successful (Wilson, Lipsey & Derzon, 2003; Espelage et al., 2013).

This study has several limitations. First, the cross-sectional nature of the survey means that we cannot make any reference to causation, only association. However, this still allows us to identify potential groups at higher risk for physical fighting. Second, the survey does not capture students who were not present on the day of the survey or those students who do not attend school for other reasons. These students may be part of vulnerable populations, such as low-income groups, and may be more likely to express risky behavior, thus potentially underestimating the risk of physical fighting. Third, the survey is self-report which, although anonymous, may result in social desirability response bias, which would bias our results towards the null. Fourth, we were limited to the variables collected by the survey and therefore may be missing important protective factors that may be associated with physical fighting, such as the potential protective effects of family life (e.g., participating in family meals, important discussions with parents and friends) or participating in community activities (Fontaine et al., 2016; Ttofi et al., 2016; Vassallo, Edwards & Forrest, 2016). Finally, we used a robust definition of two or more physical fights in the last year to capture more repetitive behavior, however, this definition makes comparisons to other classifications more difficult. We believe that the associations between risk factors and physical fighting would not change significantly between these definitions, except in strength of association. Despite these limitations, this study provides information on physical fighting in a large updated and representative sample of all school-attending adolescents aged 11-16 in Namibia.

Conclusion

This study provides an updated description of physical fighting in Namibia as well as factors that can be used to identify populations for intervention programs. Although physical fighting was found to be relatively low in Namibia compared to other sub-Saharan countries, it is still a modifiable risky behavior that can be altered in this young population. We recommend public health and school-based programming that simultaneously targets risk behaviors and conflict resolution to reduce rates of physical fighting. Future iterations of the survey would benefit from the inclusion of more family and peer-related factors which may help to identify additional correlates that may mitigate aggressive behaviors.