Rebellious Behaviors in Adolescents With Epilepsy

Aimee W. Smith, PHD; Constance Mara, PHD; Shannon Ollier, PSY.D; Angela Combs, BS; Avani C. Modi, PHD


J Pediatr Psychol. 2018;43(1):52-60. 

In This Article

Abstract and Introduction


Objectives The study aims are to (1) examine the prevalence of risk-taking (i.e., behaviors that can be categorized as rebellious or reckless) and (2) determine the influence of risk-taking on adherence, seizures, and health-related quality of life (HRQOL) in adolescents with epilepsy. An exploratory aim was to identify predictors of risk-taking.

Methods Fifty-four adolescents with epilepsy (M = 15.33 ± 1.46 years) and caregivers completed questionnaires on demographics, risk-taking, parent–child relations, adolescent inattention/hyperactivity, and HRQOL at four time points across 1 year. Seizure occurrence and electronically monitored adherence were also collected.

Results Rebellious behaviors were normative and stable over 1 year in adolescents with epilepsy. Higher rebelliousness was related to poorer adolescent-reported memory HRQOL. The only significant positive predictor of rebellious behaviors was adolescent age.

Conclusions Adolescents with epilepsy endorsed normative levels of rebelliousness, which is negatively related to HRQOL. Older adolescents may warrant clinical attention.


Risky behaviors can be categorized as either rebellious or reckless in nature (Gullone, Moore, Moss, & Boyd, 2000). Rebellious behaviors, such as underage drinking, taking drugs, and staying out late, are a means of experimentation with potential legal and social consequences (Gullone, Moore, et al., 2000). In contrast, reckless behaviors, such as drinking and driving, stealing cars, and having unprotected sex, can result in life-threatening consequences or have long-term impacts (e.g., severe injury, imprisonment, unwanted pregnancy; Arnett, 1992). Both rebellious and reckless behaviors peak during adolescence, with 32% of high school students trying smoking, 63% alcohol, 39% marijuana, 2–17% other illegal substances (e.g., heroin, cocaine, ecstasy, illegal prescription drugs), and 8% of drivers engaging in drunk driving. Furthermore, 91% of sexually active high schoolers reported not using appropriate birth control to prevent both pregnancy and sexually transmitted infections (Kann et al., 2016). Over the course of development, adolescents' risk-taking behavior increases (Gardner & Steinberg, 2005).

Findings from developmental neuroscience demonstrate that adolescents are motivated by novel experiences and are predisposed to impulsive and risky behaviors owing to an immature inhibitory control system (Chambers, Taylor, & Potenza, 2003). In addition, neuroimaging studies have revealed two pathways for risk-taking behavior: a social-emotional system (i.e., more automatic and reactive) and a cognitive control system (i.e., thinking through, inhibiting impulses) (Albert, Chein, & Steinberg, 2013). For adolescents, the social-emotional system develops before the cognitive control system, and this imbalance coincides with an increase in salience in rewards and difficulty delaying gratification, culminating in a time of heightened likelihood of risk-taking behaviors for adolescents (Albert et al., 2013; Steinberg, 2004). Changes in these systems are salient for risk-taking, which is defined as behavior "which involves potential negative consequences (loss) but is balanced in some way by perceived positive consequences (gain)" (Gullone, Moore, et al., 2000, p. 347).

In addition to developmental changes, adolescence is often accompanied by decreased parental monitoring (Modi, Marciel, Slater, Drotar, & Quittner, 2008), as well as increased desires for autonomy, feelings of invincibility, and susceptibility to peer influence (Albert et al., 2013). Adolescents are more likely to take risks (e.g., driving, drinking alcohol) in peer groups than when alone (Chassin, Presson, & Sherman, 1989; Gardner & Steinberg, 2005; Simons-Morton, Lerner, & Singer, 2005), especially when compared with adults (Gardner & Steinberg, 2005). In fact, children as young as 6 years are influenced by peer social norms when making decisions involving risk (Morrongiello, McArthur, Kane, & Fleury, 2013). Examination of typical risk-taking behaviors is particularly important for adolescents with chronic illnesses because those who engage in one risky behavior are more likely to engage in other risky behaviors (Dryfoos, 1993), which could lead to poor illness management and adherence. For example, in pediatric diabetes, risk-taking has been conceptualized as a factor within self-management (Wasserman, Anderson, & Schwartz, 2017). That is, nonadherence itself represents a risk-taking behavior in adolescence, as the biological systems related to control and motivation may play a role in whether adolescents choose to adhere to their treatment recommendations. Additionally, authors conceptualize biological rewards (sex, food, social approval, avoidance of pain) as motivations for avoiding or delaying adherence.

Research findings regarding risk-taking in pediatric illnesses is equivocal. A study of pediatric diabetes, as well as a population-based study of adolescents with chronic illnesses found higher or equal engagement in risky behaviors compared with healthy peers (Scaramuzza et al., 2010; Suris, Michaud, Akre, & Sawyer, 2008). In contrast, adolescents with chronic kidney disease and spina bifida demonstrated lower rates of risky behaviors (Dodson, Diener-West, Gerson, Kaskel, & Furth, 2007; Murray et al., 2014). While the impact of risk-taking in adolescents with chronic illness has been understudied, there is some evidence that less risk-taking is associated with better adherence and health-related quality of life (HRQOL). For example, lower risk behavior in adolescents with diabetes predicted better adherence (Hackworth et al., 2013). However, in adolescents with chronic kidney disease, more risk-taking behaviors accounted for higher HRQOL (Dodson et al., 2007).

Adolescents with epilepsy, a common neurological disease affecting 1.4% of youth aged 12–17 years (Russ, Larson, & Halfon, 2012) are particularly vulnerable to engaging in risk-taking behaviors. This is likely due to executive functioning deficits (Hoie et al., 2008; Jackson et al., 2013) and neurobehavioral comorbidities like attention deficit hyperactivity disorder (Davies, Heyman, & Goodman, 2003; Russ et al., 2012) that result in higher impulsivity and poor decision-making, especially in those with ongoing seizures (Wandschneider et al., 2013). Comorbid depression can also make adolescents with epilepsy more vulnerable to risk-taking, such as suicide attempts (Plioplys, 2003). It remains unknown whether adolescents with epilepsy look similar to children with other neurological disorders who demonstrate lower risk-taking (Murray et al., 2014), or are more comparable with children with broader chronic illnesses who engage in higher risk-taking compared with healthy peers (Suris et al., 2008). Only one study examined risky behaviors in adolescents with epilepsy and found that 8% engaged in daily alcohol consumption and 12% tried illegal substances, with boys at higher risk than girls (Alfstad et al., 2011). While this study identified some aspects of risk-taking, it did not examine broader rebellious and reckless behaviors of adolescence nor their longitudinal relationship to health outcomes (e.g., adherence, seizure occurrence, and HRQOL).

Combined, the neurobiological underpinnings of risk and the increased likelihood for executive dysfunction create a "perfect storm" for adolescents with epilepsy to engage in rebellious and reckless behaviors. These risk-taking behaviors can have significant consequences for adolescents with epilepsy, including increased seizures. For example, the efficacy of antiepileptic drugs (AEDs) can be compromised with alcohol or drug exposure in individuals with epilepsy (Aird, 1983; Gordon & Devinsky, 2001; Nakken et al., 2005). Further, staying up late or increased stress can also serve as seizure triggers in epilepsy (Aird, 1983; Brodie & French, 2000; Malow, Fromes, & Aldrich, 1997; Nakken et al., 2005). The link between risk-taking behaviors and AED nonadherence is unknown in adolescents with epilepsy. However, poor adherence is related to worse seizure outcomes (Modi, Wu, Rausch, Peugh, & Glauser, 2014), which drives poorer HRQOL (Ferro, 2014). Therefore, risk-taking behaviors may be a key driver to poor outcomes in adolescents with epilepsy. The neurobehavioral stage of development makes adolescents with epilepsy more likely to engage in behaviors that have immediate rewards, especially in the context of their peers (e.g., skipping taking medication when out with friends), and less likely to be concerned about the negative consequences of their decisions for the future (e.g., will have a seizure) (Ernst, Pine, & Hardin, 2006; Gardner & Steinberg, 2005).

Due to to the significant potential negative impact of risk-taking, it is important to identify those most likely to engage in risk-taking. Previous pediatric literature implicates demographic factors (age, sex, race, family socioeconomic status [SES]), family factors (e.g., parent monitoring), and executive functioning skills (e.g., impulsivity) in risk-taking behaviors. Higher risk-taking is associated with older age (Gardner & Steinberg, 2005; Madsen, Roisman, & Collins, 2002; Scaramuzza et al., 2010), male gender (Alfstad et al., 2011; Gardner & Steinberg, 2005; Van Dale et al., 2014), non-White race (Gardner & Steinberg, 2005), lower self-esteem (Jackman & MacPhee, 2017; Van Dale et al., 2014), and lower family incomes (Alfstad et al., 2011; Van Dale et al., 2014). Family factors, such as higher parental monitoring (i.e., control over child's behavior, adolescent–parent communication, and adolescent disclosure of information to parents), are negatively related to risky behaviors for youth. Furthermore, poor executive functioning is related to higher endorsement of hypothetical risky behaviors, while poor emotion regulation is related to increased participation in risky behaviors for typical adolescents (Magar, Phillips, & Hosie, 2008). Although many isolated factors have been identified to affect risk-taking, demographic, family, and executive functioning have not been studied simultaneously to determine the most salient predictors of rebellious and reckless behaviors in adolescents with epilepsy. Even less is known about the extent to which risky behaviors are maintained over time. While the initiation of risk-taking behaviors are associated with factors previously mentioned (e.g., age, sex, race, SES, family factors, and executive functioning), the maintenance of risky behavior is likely influenced by parental monitoring (Wang et al., 2013).

The current study has three aims. Our first aim is to examine the prevalence of risk-taking in adolescents with epilepsy and changes over 1 year. Because older age and later adolescence are associated with increased risk-taking, it is hypothesized that risk-taking behaviors will increase over the year. The second aim is to examine the impact of risk-taking on health outcomes (adherence, seizure occurrence, and HRQOL). Higher risk-taking behaviors are expected to be associated with poorer adherence, increased occurrence of seizures, and worse HRQOL. An exploratory aim is to identify significant predictors of risk-taking in adolescents with epilepsy. Age, sex, race, SES, adolescent inattention/hyperactivity, parent–adolescent conflict, and parent involvement were hypothesized to predict risk-taking.