Discussion
As a novel contribution to the literature, we examined parent and child DRD scores across 6 months to characterize their stability and we explored the associations between parent and child DRD levels and child HbA1c levels over time. Regarding our first objective, we found stable or increasing DRD levels for 69% of parents and 80% of children. For our second objective, in a regression model including parent and child DRD scores at baseline, controlling for child HbA1c level at baseline, and predicting child HbA1c at 6 months, we found that only child HbA1c and child DRD scores at baseline were significant predictors.
While we believe that we are the first to examine DRD levels in parents and school-age children over time, cross-sectional studies have previously identified DRD as a potential risk factor for elevated glycemic levels in families of school-age children with type 1 diabetes.6 8 20 Although some degree of transient DRD is expected in both parents and children, parents of school-aged children may be at risk of experiencing chronic DRD because of the significant role they play in helping to manage their child’s type 1 diabetes.19 Likewise, we hypothesize that school-aged children may be vulnerable to experiencing chronic DRD because many of them are beginning to assume more responsibility for their own diabetes care.1 3 Within the broader literature, studies have explored the rates of chronic DRD in adults and adolescents living with type 1 diabetes. Specifically, in adults with type 1 diabetes, researchers have found rates of stable or increasing DRD among 51% of participants,21 while another study found significant chronic DRD over 16 months in approximately 40% of adolescents with type 1 diabetes.22 Our results add to this literature and appear to reinforce the concern that DRD is chronic and may even increase over time in some persons living with type 1 diabetes as well as among some persons who care for them. Because we identify relatively high rates of chronic or increasing DRD among school-age children with type 1 diabetes and their parents, our results also underscore the need for additional and longer prospective studies of DRD in these populations. This may allow us to obtain a better sense of the burden that families carry related to chronic DRD by identifying patterns of DRD over time and to identify targets for age-appropriate interventions to help school-age children to manage DRD.
Specific to our second objective, the results of our regression models provide more information about how parent and child DRD levels may relate to children’s glycemic levels over time. As noted previously, our model found that only child HbA1c and DRD scores at baseline were significant predictors of child HbA1c at 6 months. Together, these two variables explained 75% of the variance in child HbA1c at 6 months. Not surprisingly, children’s HbA1c at baseline was most strongly associated with their HbA1c at 6 months. As this was not an intervention study, and because of cumulative effects of blood glucose over time, we would expect a strong association between children’s baseline and 6-month HbA1c levels. Moreover, a prospective study among similarly aged children with type 1 diabetes previously found relatively stable HbA1c levels in 64% of children across 30 months.23 Nevertheless, we think it is remarkable that even controlling for the association with baseline HbA1c, child DRD at baseline remained a significant predictor of their HbA1c at 6 months, which confirms that child DRD is indeed a risk factor for later glycemic elevations.6 20 Interpreting the unstandardized beta weight for children’s DRD scores suggests that for every 10-unit decrease, we would predict an approximately 0.13% decrease in child HbA1c level at 6 months. This result further reinforces the possibility that treating child DRD could also help them to achieve a lower HbA1c, which would have long-term health benefits.
It was surprising that parents’ DRD scores at baseline did not remain a significant predictor of child HbA1c at 6 months in our full model, as a previous longitudinal study from the Netherlands found an association between parent DRD and child HbA1c over time.24 There are also data from a previous cross-sectional study identifying parents’ DRD as a stronger predictor of children’s HbA1c than children’s insulin pump use and type 1 diabetes duration.8 One explanation for our finding may be that there exist other variables (eg, treatment engagement, resilience, self-efficacy, coping, or shared responsibility for type 1 diabetes management) that potentially mediate the association between parents’ DRD at baseline and children’s HbA1c at 6 months.25 26 That is, perhaps some parents use their resilience and/or coping skills to manage any feelings of DRD so that they do not interfere with daily type 1 diabetes treatment engagement.27 If so, it might be helpful to deploy strategies targeting these variables as well as parent DRD in a parent-focused intervention to improve child HbA1c. In the future, it is possible that a larger and longer study, including more variables theoretically related to children’s HbA1c, might help us to better understand the associations between parent and child DRD and children’s HbA1c over time.
With respect to clinical implications, our results underscore current clinical guidelines to routinely screen for DRD in families of children with type 1 diabetes beginning at 7–8 years old.1–3 Routine screening of DRD in parents and school-age children can provide the opportunity for earlier detection and possibly treatment of DRD, which may help families achieve better quality of life. Moreover, it is possible our results provide some guidance as to how frequently clinics should screen families for DRD. Specifically, because we found 55% of parents and 60% of children had stable DRD across 6 months, it is possible clinics may be able to reduce burden by screening most families for DRD no more than twice a year, though we acknowledge that families with chronically high levels of DRD may still benefit from more frequent monitoring. There are validated measures of DRD available for school-aged children,6 adolescents,28 29 and parents.6 30 These measures tend to be short and, therefore, could be practical for use in a busy clinic. Within the literature, there is emerging evidence supporting the efficacy of cognitive–behavioral therapy strategies and positive psychology to reduce DRD in parents,31 32 adolescents,33 and families of school-aged children.34 It is possible that with additional research and refinement, we can disseminate these treatments more broadly, thereby giving clinics and families access to treatments to help them reduce their feelings of DRD and achieve lower child HbA1c.
Our study has some limitations. We only recruited families who were English speaking, which may reduce the generalizability of our results. We recruited a sample that was majority White, preventing us from examining whether a family’s race or ethnicity could also be an important covariate when characterizing parent and child DRD levels and associating their DRD to children’s HbA1c levels over time.20 35 Though children’s mean HbA1c levels at baseline and 6 months were lower than US registry data,36 nearly three-quarters of children had a HbA1c level >7.0%.1 Thus, we do not believe our sample was overly enriched for children in tighter glycemic control. Nonetheless, in the future, it might be helpful to purposefully recruit more families of children with higher HbA1c levels to more fully examine the associations between parent and child DRD and children’s HbA1c. We acknowledge that the parents in our sample were mostly mothers, and that future research is needed to describe DRD levels in fathers and other caregivers and to associate these with children’s HbA1c. We acknowledge that we did not collect data on the type of insulin pump children used. Therefore, in a future study, it may be helpful to retest our model and include type of insulin pump as a covariate. We acknowledge that we did not collect children’s personal continuous glucose monitor data at each of the study visits, which might provide a more nuanced assessment of how parent and child DRD relate to children’s glycemia. In a future investigation, it will be important to retest our model with continuous glucose monitor metrics versus child HbA1c as the dependent variable. Because our study recruited only families of children between the ages of 8–12 years old, we recognize that our findings may not generalize to families of older or younger children. Finally, we acknowledge that this study employed an observational cohort design and that, while we did not attempt to intervene on parent and child DRD from baseline to 6 months, families were able to seek treatment for DRD on their own, which could impact our results. Of note, at the time we conducted the study, routine DRD screening was not a part of standard of care for the participating clinics. Thus, our IRB determined our study design and methods involved “no more than minimal risk” and identified no ethical concerns in observing families’ DRD levels over time. Some of the strengths of our study included its large sample size of parents and children, our use of validated measures of DRD and children’s HbA1c, our use of a single laboratory for analyzing children’s HbA1c, and our prospective design.
To conclude, our study suggests that school-age children living with type 1 diabetes and their parents may experience chronic levels of DRD. Our results are consistent with previous research conducted with adolescents22 and adults21 living with type 1 diabetes. Moreover, because our results identified child DRD at baseline as a significant predictor of child HbA1c at 6 months despite the variance explained by children’s baseline HbA1c, they contribute to the growing body of evidence suggesting that treating DRD may be one pathway to lowering child HbA1c. However, we assert a need for future research to examine these associations over a longer period and for studies that include other variables theoretically related to child HbA1c (eg, treatment engagement, treatment responsibility) to better understand how parent and child DRD relates to child HbA1c.25 26 DRD is treatable, and there are evidence-based treatments available for parents and children.31–34 Therefore, it is possible that through a greater understanding of how parent and child DRD relate to children’s HbA1c over time, we can teach parents and children strategies to reduce their feelings of DRD and achieve and maintain lower HbA1c levels.
Leave a Reply