Child Maltreatment and the Adolescent Patient With Severe Obesity
Child Maltreatment and the Adolescent Patient With Severe Obesity
Twenty-nine percent of females and 12% of males with severe obesity undergoing WLS or nonsurgical treatment self-reported a history of CM. Emotional abuse was the most prevalent, although approximately 1 in 10 females reported a history of moderate/severe sexual or physical abuse. These rates are similar to self-reported rates of CM in adolescent community samples in the United States (Finkelhor, Turner, Shattuck, & Hamby, 2013), as well as in Canada when using the CTQ (Crooks, Scott, Ellis, & Wolfe, 2011), suggesting no greater CM history risk than would be expected in the broader adolescent population. Sample size and low prevalence of CM in males precluded examination of CM correlates and severity. However, females with a CM history appear to be a significant minority who carry greater psychosocial burden into a clinical weight management setting.
Females with a history of CM had significantly greater internalizing (e.g., depressive and anxiety) and externalizing symptomatology (e.g., oppositional or defiant behaviors), as well as a greater likelihood of current use of psychiatric medications compared with those without a CM history. However, although neither group had mean levels of internalizing or externalizing symptoms (YSR, CBCL) in a clinical range, both groups reported psychiatric medication use rates (maltreated 28%; nonmaltreated 20%) that were higher than national base rates (7%) (Olfson, He, & Merikangas, 2013). Females with a history of CM were also more likely to be current smokers and report alcohol use, and were characterized by lower self-esteem and weight-related quality of life, as well as greater family dysfunction. Finally, a greater number of maltreatment types (CM-Load) was associated with a range of psychosocial impairments, with unique associations when considering severity within each CM domain.
These data have clear clinical implications, highlighting modifiable intervention and prevention targets (e.g., psychopathology, alcohol use, risky sexual behaviors, low self-esteem) for an at-risk adolescent subgroup to improve patient outcomes. While CM is known to increase obesity risk, no empirical base indicates that a CM history serves as a barrier to an adolescent successfully engaging in supervised weight loss treatment. The adult obesity treatment literature is equivocal regarding this issue (Steinig, Wagner, Shang, Dolemeyer, & Kersting, 2012). More likely, CM may play an indirect role in WLS outcomes given its associated psychosocial sequelae. For example, depression has been shown to mediate the links between CM and later obesity (Danese & Tan, 2014). Severe obesity, depression (Zeller et al., 2004) as well as family dysfunction (Williams et al., 2010) have each been shown to increase the likelihood of pediatric lifestyle modification dropout. For adults, greater psychiatric impairment has been linked to program dropout before WLS (Merrell, Ashton, Windover, & Heinberg, 2012), as well as poorer weight loss outcomes in surgical and nonsurgical interventions (Legenbauer, 2009; Wood & Ogden, 2012).
Providers in surgical and nonsurgical pediatric weight management programs are already primed to assess and monitor the clinical needs of patients with severe obesity, and thus uniquely positioned to assess a patient's maltreatment history and risk. Providers may find the American Academy of Pediatrics trauma guide a helpful resource (see www.aap.org/traumaguide). At a minimum, providers should be knowledgeable of adjunctive referral resources to assist adolescents and their families presenting with dysfunction and/or distress. Evidence-based treatments have also emerged to promote resilience in youth who have experienced CM (i.e., trauma-focused cognitive behavior therapy; Mannarino, Cohen, Deblinger, Runyon, & Steer, 2012). Pediatric psychologists can play a crucial role in facilitating appropriate referrals to adjunctive care.
The present study also included some unanticipated findings. Specifically, CM rates were consistently lower in adolescents undergoing WLS compared with those in lifestyle modification. These cohort differences may be a downstream effect of a complex and often lengthy process to achieve WLS candidacy. This includes decision-makers on multiple levels (i.e., referring physician, adolescent, family, clinical team, insurance provider; Inge et al., 2014). In addition, the adolescent must maintain a stable psychiatric status (i.e., symptoms well-managed by collaborating providers; Austin, Smith, & Ward, 2013). Thus, it is conceivable that maltreated youth presenting with poorly managed psychopathology and family dysfunction may be less likely to progress to achieving WLS candidacy. While beyond the scope of this study, this is an important empirical question to be addressed by future studies examining access to care and preoperative program attrition.
Alternately, it remains possible that WLS adolescents may have minimized their CM history. A post hoc examination of the CTQ Minimization/Denial scale, an indicator of possible underreporting of CM (score range 0–3; Bernstein et al., 2003), suggested that a significantly greater number of WLS participants (26%) than comparisons (13%; p = .038) had high minimization scores (i.e., score of 2 or 3). While all participants had been approved for surgery before study recruitment, adolescents may have minimized CM history severity to avoid mandated follow-up by research staff at such a critical time (i.e., within 30 days of WLS). Mandated reporting for minors may also be an important consideration when comparing the present adolescent findings to the higher CM rates reported in the adult WLS literature, as adult retrospective reporting bears few consequences (Grilo et al., 2005; Wildes et al., 2008). Moreover, adolescents in the present study remain in the age window to still experience CM, and thus, the present data may underestimate CM prevalence in adolescent WLS patients.
Strengths of the present study include the multisite and controlled design, standardized data collection, and a comprehensive and age-salient assessment battery. However, this study is not without limitations that can inform future work. Consistent with adult WLS trends (Belle et al., 2013), the Teen-LABS patient population is primarily female and White, which combined with the study's design to recruit a demographically similar comparison cohort, resulted in limited information regarding males and other race/ethnic groups (e.g., Hispanic, non-Hispanic Black, Native American) known be at heightened risk for severe obesity (Kelly et al., 2013), as well as to have experienced CM by the age of 18 years (Wildeman et al., 2014). Further, adolescents who demographically "matched" and participated in the nonoperative comparison group may have been different than those who did not match, or those who initially declined being listed as potential matches on the nonsurgical registry. Finally, these findings may not be representative of adolescents with severe obesity in nonclinical settings.
The presence of CM was based on adolescent self-report on the CTQ-SF, a widely used psychometrically sound screening tool. An alternate approach would be to examine CM substantiated by CPS criterion. This would be challenging in the current multisite context representing patients from eight states, with reporting laws based on varying definitions of abuse and neglect. We did not assess additional CM characteristics that are important to understanding psychosocial correlates, such as developmental timing or frequency of CM experiences (Jackson, Gabrielli, Fleming, Tunno, & Makanui, 2014). In addition, the present study focused exclusively on CM, arguably an exemplar of only one aspect of adverse child experiences (e.g., parental death, divorce, parental psychopathology, family violence, crime, poverty, victimization) that may co-occur and contribute to adolescent health outcomes (McLaughlin et al., 2012). Finally, the present data are cross-sectional and causality cannot be inferred.
Given maltreated youth are not homogeneous in their experiences or their outcomes (Jackson et al., 2014), we cannot assume a simple CM to WLS outcomes link, whether the outcomes be in the domain of physical or psychosocial health. These outcome pathways are likely more complex and best examined through prospective longitudinal studies where one must consider CM playing a mediating or moderating role. What role CM plays in adolescent WLS outcomes is unknown and part of important ongoing ancillary work currently being executed in collaboration with the Teen-LABS consortium. In addition, the present study shows that there is concern for maltreated adolescents with severe obesity who are not undergoing WLS, who are an understudied subpopulation, and for whom there is a bleak forecast of health and well-being in adulthood. The present authors are in no way advocating that youth with a history of CM be excluded from weight management care, nor should CM be perceived as a barrier to WLS candidacy. Rather, as for any pediatric patient in any clinical setting, CM history should be indicator for trauma-informed care.
Discussion
Twenty-nine percent of females and 12% of males with severe obesity undergoing WLS or nonsurgical treatment self-reported a history of CM. Emotional abuse was the most prevalent, although approximately 1 in 10 females reported a history of moderate/severe sexual or physical abuse. These rates are similar to self-reported rates of CM in adolescent community samples in the United States (Finkelhor, Turner, Shattuck, & Hamby, 2013), as well as in Canada when using the CTQ (Crooks, Scott, Ellis, & Wolfe, 2011), suggesting no greater CM history risk than would be expected in the broader adolescent population. Sample size and low prevalence of CM in males precluded examination of CM correlates and severity. However, females with a CM history appear to be a significant minority who carry greater psychosocial burden into a clinical weight management setting.
Females with a history of CM had significantly greater internalizing (e.g., depressive and anxiety) and externalizing symptomatology (e.g., oppositional or defiant behaviors), as well as a greater likelihood of current use of psychiatric medications compared with those without a CM history. However, although neither group had mean levels of internalizing or externalizing symptoms (YSR, CBCL) in a clinical range, both groups reported psychiatric medication use rates (maltreated 28%; nonmaltreated 20%) that were higher than national base rates (7%) (Olfson, He, & Merikangas, 2013). Females with a history of CM were also more likely to be current smokers and report alcohol use, and were characterized by lower self-esteem and weight-related quality of life, as well as greater family dysfunction. Finally, a greater number of maltreatment types (CM-Load) was associated with a range of psychosocial impairments, with unique associations when considering severity within each CM domain.
These data have clear clinical implications, highlighting modifiable intervention and prevention targets (e.g., psychopathology, alcohol use, risky sexual behaviors, low self-esteem) for an at-risk adolescent subgroup to improve patient outcomes. While CM is known to increase obesity risk, no empirical base indicates that a CM history serves as a barrier to an adolescent successfully engaging in supervised weight loss treatment. The adult obesity treatment literature is equivocal regarding this issue (Steinig, Wagner, Shang, Dolemeyer, & Kersting, 2012). More likely, CM may play an indirect role in WLS outcomes given its associated psychosocial sequelae. For example, depression has been shown to mediate the links between CM and later obesity (Danese & Tan, 2014). Severe obesity, depression (Zeller et al., 2004) as well as family dysfunction (Williams et al., 2010) have each been shown to increase the likelihood of pediatric lifestyle modification dropout. For adults, greater psychiatric impairment has been linked to program dropout before WLS (Merrell, Ashton, Windover, & Heinberg, 2012), as well as poorer weight loss outcomes in surgical and nonsurgical interventions (Legenbauer, 2009; Wood & Ogden, 2012).
Providers in surgical and nonsurgical pediatric weight management programs are already primed to assess and monitor the clinical needs of patients with severe obesity, and thus uniquely positioned to assess a patient's maltreatment history and risk. Providers may find the American Academy of Pediatrics trauma guide a helpful resource (see www.aap.org/traumaguide). At a minimum, providers should be knowledgeable of adjunctive referral resources to assist adolescents and their families presenting with dysfunction and/or distress. Evidence-based treatments have also emerged to promote resilience in youth who have experienced CM (i.e., trauma-focused cognitive behavior therapy; Mannarino, Cohen, Deblinger, Runyon, & Steer, 2012). Pediatric psychologists can play a crucial role in facilitating appropriate referrals to adjunctive care.
The present study also included some unanticipated findings. Specifically, CM rates were consistently lower in adolescents undergoing WLS compared with those in lifestyle modification. These cohort differences may be a downstream effect of a complex and often lengthy process to achieve WLS candidacy. This includes decision-makers on multiple levels (i.e., referring physician, adolescent, family, clinical team, insurance provider; Inge et al., 2014). In addition, the adolescent must maintain a stable psychiatric status (i.e., symptoms well-managed by collaborating providers; Austin, Smith, & Ward, 2013). Thus, it is conceivable that maltreated youth presenting with poorly managed psychopathology and family dysfunction may be less likely to progress to achieving WLS candidacy. While beyond the scope of this study, this is an important empirical question to be addressed by future studies examining access to care and preoperative program attrition.
Alternately, it remains possible that WLS adolescents may have minimized their CM history. A post hoc examination of the CTQ Minimization/Denial scale, an indicator of possible underreporting of CM (score range 0–3; Bernstein et al., 2003), suggested that a significantly greater number of WLS participants (26%) than comparisons (13%; p = .038) had high minimization scores (i.e., score of 2 or 3). While all participants had been approved for surgery before study recruitment, adolescents may have minimized CM history severity to avoid mandated follow-up by research staff at such a critical time (i.e., within 30 days of WLS). Mandated reporting for minors may also be an important consideration when comparing the present adolescent findings to the higher CM rates reported in the adult WLS literature, as adult retrospective reporting bears few consequences (Grilo et al., 2005; Wildes et al., 2008). Moreover, adolescents in the present study remain in the age window to still experience CM, and thus, the present data may underestimate CM prevalence in adolescent WLS patients.
Strengths of the present study include the multisite and controlled design, standardized data collection, and a comprehensive and age-salient assessment battery. However, this study is not without limitations that can inform future work. Consistent with adult WLS trends (Belle et al., 2013), the Teen-LABS patient population is primarily female and White, which combined with the study's design to recruit a demographically similar comparison cohort, resulted in limited information regarding males and other race/ethnic groups (e.g., Hispanic, non-Hispanic Black, Native American) known be at heightened risk for severe obesity (Kelly et al., 2013), as well as to have experienced CM by the age of 18 years (Wildeman et al., 2014). Further, adolescents who demographically "matched" and participated in the nonoperative comparison group may have been different than those who did not match, or those who initially declined being listed as potential matches on the nonsurgical registry. Finally, these findings may not be representative of adolescents with severe obesity in nonclinical settings.
The presence of CM was based on adolescent self-report on the CTQ-SF, a widely used psychometrically sound screening tool. An alternate approach would be to examine CM substantiated by CPS criterion. This would be challenging in the current multisite context representing patients from eight states, with reporting laws based on varying definitions of abuse and neglect. We did not assess additional CM characteristics that are important to understanding psychosocial correlates, such as developmental timing or frequency of CM experiences (Jackson, Gabrielli, Fleming, Tunno, & Makanui, 2014). In addition, the present study focused exclusively on CM, arguably an exemplar of only one aspect of adverse child experiences (e.g., parental death, divorce, parental psychopathology, family violence, crime, poverty, victimization) that may co-occur and contribute to adolescent health outcomes (McLaughlin et al., 2012). Finally, the present data are cross-sectional and causality cannot be inferred.
Conclusions and Future Directions
Given maltreated youth are not homogeneous in their experiences or their outcomes (Jackson et al., 2014), we cannot assume a simple CM to WLS outcomes link, whether the outcomes be in the domain of physical or psychosocial health. These outcome pathways are likely more complex and best examined through prospective longitudinal studies where one must consider CM playing a mediating or moderating role. What role CM plays in adolescent WLS outcomes is unknown and part of important ongoing ancillary work currently being executed in collaboration with the Teen-LABS consortium. In addition, the present study shows that there is concern for maltreated adolescents with severe obesity who are not undergoing WLS, who are an understudied subpopulation, and for whom there is a bleak forecast of health and well-being in adulthood. The present authors are in no way advocating that youth with a history of CM be excluded from weight management care, nor should CM be perceived as a barrier to WLS candidacy. Rather, as for any pediatric patient in any clinical setting, CM history should be indicator for trauma-informed care.
Source...