JPP Student Journal Club Commentary: The Value of Assessing Patterns of Psychosocial Risk and Resilience in Glycemic Control Trajectory Models

Maintaining optimal glycemic control is particularly challenging for adolescents and young adults with type 1 diabetes (T1D), as up to 80% of them fail to meet recommended hemoglobin A1c (HbA1c) values (Wood et al., 2013). The developmental and social changes of adolescence and the transition of T1D management responsibilities from caregiver to youth highlight the need for a comprehensive understanding of variables that affect glycemic control. However, most previous studies focus on cross-sectional data. Helgeson and colleagues (Helgeson et al., 2017) describe an underused approach to studying long-term health-care outcomes in adolescents with T1D. They followed 132 youth with T1D for 11 years, beginning in early adolescence. At the start of the study, youth completed measures of parent relationship quality; friend support and conflict; unmitigated communion; symptoms of depression, anxiety, and anger; and self-care behaviors. Parent responses on a broadband measure of youth functioning yielded internalizing and externalizing composite scores.

HbA1c values collected throughout the study were used to identify five trajectories: stable on target, stable above target, volatile late peak, stable high, and inverted U. Social status, household structure, and insulin pump usage distinguished the stable on target group from others; members of the stable on target group were more likely to come from higher social status, two-parent families, and to use an insulin pump than members of more “problematic” groups. After controlling for social status and household structure, individuals in the stable on target group reported less friend conflict, less psychological distress, fewer internalizing problems, fewer externalizing problems, more unmitigated communion, and poorer self-care than those in other groups, particularly the inverted U group.

This study represents an important area of research that holds promise for improving our approach to youth T1D management. Nevertheless, three issues warrant consideration in future research. First, while the trajectory-based methodology clarifies how diabetes-related outcomes change over time, aspects of the study design may limit the applicability and generalizability of the results. Second, the incorporation of a risk and resistance framework warrants equal emphasis on risk factors and resistance factors. Finally, despite individual differences in unmitigated communion, it is currently an understudied trait and its clinical applicability is unknown.

The trajectory-based methodology is an advantageous approach for examining T1D glycemic control during a critical developmental stage. Previous research examining glycemic trajectories in adolescents have found both consistencies and variability in resulting patterns (Hilliard, Wu, Rausch, Dolan, & Hood, 2013; Luyckx & Seiffge-Krenke, 2009; Schwandt et al., 2017). While some youth maintain stable glycemic control over adolescence and young adulthood (either poorly or well controlled), other subgroups have deteriorating HbA1c values that eventually peak well above the recommended range. Across studies, relatively few participants were in the “problematic” glycemic trajectories. In the present study, the volatile late peak group included only 7.5% of participants, and the inverted U group included just 9.8%. While this is encouraging from a clinical standpoint, it is difficult to draw definitive conclusions about such small samples. Replication of trajectory analyses would confirm whether patterns generalize across populations. Additionally, this study showed significant variability in the number of HbA1c values for each participant; some had as few as four measurements over the 11 years. Although statistical analyses can account for this variability, it is important to understand the reasons for the range in number of HbA1c values. A greater number of HbA1c values was related to higher social status and use of an insulin pump. It is important to consider potential reasons for missing values including fewer medical visits or receiving care at another facility, especially for youth transitioning from pediatric to adult care. Notably, psychosocial and self-care variables were assessed only at the beginning of the 11-year study. It would be increasingly valuable to assess these variables repeatedly throughout a study. Further investigation would reveal how psychosocial and self-care variables change over time and the relationship between glycemic control and these variables throughout adolescence and young adulthood.

Wallander, Varni, Babani, Banis, and Wilcox’s (1989) risk and resistance framework was proposed to identify variables related to glycemic control patterns. Although friend support and lower levels of psychological distress may be conceptualized as resistance factors, a complex combination of individual, family, and social/contextual processes potentially impact diabetes competence and health outcomes (Hilliard, Harris, & Weissberg-Benchell, 2012). A more thorough examination of modifiable protective factors could guide interventions. Identifying and promoting diabetes-specific protective processes are especially necessary (Hilliard et al., 2012). How do risk and protective processes change during adolescence and young adulthood? Do the relationships between risk and protective factors change over time for the different trajectory groups? Do the same factors differentiate the groups 11 years later? With limited research in this area, there is a need to further examine both unique and interrelated factors that promote optimal glycemic control.

A personality trait, unmitigated communion, was explored as a possible contributing factor to poor glycemic control. Unmitigated communion differentiated the stable on target group from the inverted U group, which maintained the highest levels of HbA1c through adolescence. It is unclear how behaviors associated with unmitigated communion are distinct from adherence behaviors assessed with widely used, well-validated measures. While unmitigated communion describes a potentially relevant trait, it is unknown how providers might respond to an adolescent with high levels of unmitigated communion. Is it valid to assess this construct in adolescents when personality characteristics are still developing? Personality characteristics might play an important role, but whether unmitigated control accounts for the most variability in adolescent self-care behaviors is unclear. Further investigation should examine if unmitigated communion is a unique contributor to the trajectories of adolescent glycemic control.

This study furthers our understanding of glycemic control through adolescence and young adulthood and the demographic and psychosocial variables that may play a role in these changes over time. Refined assessment of psychosocial characteristics is imperative. Future research will improve the ability to identify and intervene with youth at risk for developing “problematic” trajectories and to support youth with stable in-range trajectories.

Conflicts of interest: None declared.

References