|Home | About | Journals | Submit | Contact Us | Français|
The current study examined the impact of collaborative and intrusive parenting on depression and glycemic control in children with Type 1 diabetes (T1D). Research examining the association between parenting and child adjustment in this population has been limited by a reliance primarily on questionnaire data and cross-sectional analyses. To address these gaps, the current study used an observational coding system to measure the effects of parenting on child adjustment over a 1-year period.
Youth (10 to 16 years old) with T1D and their mothers (N = 81) were recruited from an outpatient pediatrics diabetes clinic. Mothers' symptoms of anxiety and depression and children's depressive symptoms were assessed by self- reports; parenting behaviors were assessed via video-recorded observations coded using the Iowa Family Interaction Rating Scales; and adolescents' glycosylated hemoglobin (HbA1c) was obtained from medical records.
Bivariate correlations and linear regression analyses revealed that higher levels of observed collaborative parenting were related to significantly lower HbA1c 12 months later, and higher levels of observed overinvolved parenting were related to significantly greater child depressive symptoms 12 months later. Further, age and treatment type moderated the relation between overinvolved parenting and child depressive symptoms.
Collaborative and overinvolved parenting appears important for adolescents in predicting both psychological and health-related outcomes over time. Parenting behaviors may serve as an important target for future interventions to enhance adjustment in these children.
Type 1 diabetes (T1D) is the third most common chronic illness in children and adolescents in the United States, with approximately 18,000 new patients diagnosed each year (Hamman et al., 2014). Effective management of T1D requires adherence to a complex daily treatment regimen involving glucose checks, insulin injections/boluses and careful monitoring of diet and exercise (American Diabetes Association, 2015). Adherence to this intensive regimen keeps blood sugars as close to normal as possible, which is necessary for the prevention of both short- and long-term health complications (e.g., ketoacidosis and cardiovascular disease; Diabetes Control and Complications Trial Research Group, 1994). However, adolescence is associated with deteriorating glycemic control, due to physical and hormonal changes and problems with adherence (Borus & Laffel, 2010; Hood et al., 2014). Additionally, adolescents with T1D are at increased risk for psychological problems, especially depression (Hood et al., 2006). The responsibility of managing diabetes generally falls to the individual and his or her family; consequently, parenting and family functioning has been consistently related to outcomes in youth with T1D (Young, Lord, Patel, Gruhn, & Jaser, 2014).
Although adolescents often take over more diabetes treatment-related responsibilities as they mature, parental involvement remains integral to maintaining good psychological health and metabolic control, even among older adolescents (Helgeson, Reynolds, Siminerio, Escobar, & Becker, 2008). Thus the type, rather than level, of parental involvement may change over time; instead of being directly involved in diabetes care tasks (e.g., changing pump sites), parents may become more indirectly involved as children reach adolescence through increased monitoring (e.g., supervising child changing pump site; Berg et al., 2008; Ellis et al., 2007). This shift to more collaborative involvement enables the child to gain self-efficacy and autonomy while the parent continues to monitor the completion of essential tasks (Ivey, Wright, & Dashiff, 2009). Although this transition begins during adolescence, the actual age depends on the child's maturity and demonstrated responsibility (Wiebe et al., 2014). Further, some parents feel unable to trust their child to manage diabetes properly and therefore may continue to be directly, rather than collaboratively, involved to insure that treatment demands are being met (Hilliard et al., 2011; Ivey et al., 2009). High levels of diabetes-related stress in parents—especially mothers—may also contribute to overinvolved parenting behaviors due to increased parental anxiety and depressive symptoms (Butler et al., 2009; Jaser & Grey, 2010). The current study will focus on these two distinct types of parenting—collaborative parenting and overinvolved parenting—and their relationship to psychological and physical health outcomes in adolescents with T1D.
Empirical studies have established a robust association between intrusive/overinvolved parenting and increased internalizing symptoms in offspring, including psychological distress and feelings of hopelessness and worthlessness (McKee et al., 2007). In the diabetes literature specifically, parent–child relationships characterized by overprotection, strict rules/limits, harshness and low emotional support are associated with poorer adherence, worse glycemic control, and increased depressive symptoms in youth (Cameron et al., 2008; Hood, Butler, Anderson, & Laffel, 2007). Intrusive parenting practices may be especially harmful when directed toward older adolescents by impeding their ability to attain proper self-care behaviors (Duke et al., 2008; Sweenie, Mackey, & Streisand, 2014).
In contrast, collaborative involvement from parents of adolescents with T1D may serve as a protective factor against the negative health and psychological outcomes commonly observed in this population. Unlike overinvolved parenting, collaborative parenting is characterized by open communication, emotional support, and independence encouragement. This type of parenting style enables the child to gain self-efficacy and autonomy by gradually allowing them to be included in the various aspects of diabetes care while the parent continues to monitor the completion of essential tasks (Ivey et al., 2009). Maternal parent–child relationships characterized by collaboration are associated with greater quality of life (Jaser & Grey, 2010), adherence and glycemic control (Greene, Mandleco, Roper, Marshall, & Dyches, 2010; Jaser & Grey, 2010; Shorer et al., 2011).
To date, several studies have investigated the associations between parenting and adolescent adjustment in the Type 1 diabetes population. For example, Butler, Skinner, Gelfand, Berg, and Wiebe (2007) found that adolescents' perceptions of firm maternal psychological control were associated with greater depressed mood among older adolescents, and that adolescents' perceptions of maternal acceptance were associated with lower depressive symptoms and better self-efficacy for diabetes management, particularly for girls. A second study, conducted by Nansel et al. (2009), also established an association between collaborative parenting behaviors with children's adherence and quality of life. Overall, the literature on this topic has been limited by reliance on cross-sectional data, prohibiting strong conclusions regarding the determination of causes or effects (Maxwell & Cole, 2007). Prior work has also consisted mainly of retrospective adolescent or parent self-report of parenting that can be influenced by recall bias, social desirability and disagreement among respondents (De Los Reyes & Kazdin, 2005). Observational studies reduce issues of shared method variance for studies solely relying on questionnaire measures and may provide a more objective measure of parenting. Although some studies on this topic have used observational methods (e.g., Wysocki et al., 2009; Weissberg-Benchell et al., 2009), analysis of this data has been based on broad communication styles rather than specific parenting behaviors (e.g., “family communication”). No studies to date have analyzed the effects of specific, observed parenting behaviors on child adjustment over time. Therefore, the current study focuses on understanding the prospective effects of observed collaborative and overinvolved parenting on adolescent's glycemic control and depressive symptoms in a population of youth with T1D.
Given the links between parental distress and increased negative/intrusive parenting cited in the literature, it is hypothesized that maternal symptoms of depression, anxiety, and diabetes-related stress will be related to increased levels of observed overinvolved parenting and lower levels of collaborative parenting. Second, it is expected that collaborative parenting at baseline will predict lower HbA1c (i.e., better glycemic control) and decreased depressive symptoms 12 months later, while overinvolved parenting will predict higher HbA1c and increased depressive symptoms at 12 months. Additionally, adolescent sex, treatment type (e.g., pump vs. injections), and age will be tested as possible moderators of the relation between parenting and child adjustment, as these factors have been linked to differences in depression and glycemic control in youth (Hood, Rausch, & Dolan, 2011).
Participants included 93 adolescents with T1D and their mothers, recruited from an outpatient pediatrics diabetes clinic of a university medical center. To be eligible for the study, adolescents had a diagnosis of T1D for at least 6 months, no other major health problems, and were able to speak and read English. The age range for adolescents was 10–16 years, chosen to capture the developmental transition when responsibility for diabetes management begins shifting from parent to child (Anderson et al., 2002), and when glycemic control and adherence to the recommended treatment regimen often suffer a decline (Borus & Laffel, 2010). Mothers were eligible to participate if they were currently living with the adolescent and were able to speak and read English. The focus of the current study is on mothers, as they are typically the primary caregivers and tend to report more diabetes-related distress than fathers (Whittemore, Jaser, Chao, Jang, & Grey, 2012). Demographic characteristics are reported in Table 1.
Families were approached during their regular quarterly clinic visits, and interested families scheduled a separate visit to the laboratory. Before participating, parents and adolescents gave written informed consent/assent, following procedures approved by the university's Institutional Review Board. At the baseline assessment, adolescents and their mothers completed questionnaire data and participated in a 15-min videotaped discussion of diabetes stress. The topic for the video interaction was based on parent and child ratings of 10 diabetes-specific stressors (e.g., child dealing with diabetes care, parents bugging child about taking care of his or her self) listed on the Response to Stress Questionnaire (Connor-Smith, Compas, Wadsworth, Thomsen, & Saltzman, 2000). Mothers and adolescents rated the frequency of each stressor (e.g., worrying about “bad” numbers, feeling different from others) on a 4-point Likert scale ranging from 0 (never) to 3 (almost every day). Ratings were summed across mother and adolescent report to determine the highest-rated stressor, and each dyad was given a cue card with questions regarding this stressor (e.g., What happened the last time you [felt different from others], What kind of emotions do you have when you [are feeling different from others]? How can we reduce this stress?) to guide discussion. The most common topics were “dealing with diabetes care” and “parents reminding me to take care of myself.” After the 15-min period, the participants had an opportunity to provide feedback regarding their visit and were compensated for their time. Participants completed all questionnaires again during regularly scheduled clinic visits 12 months after baseline. Of the 93 participants, 12 failed to return for a clinic appointment within the study's time frame; therefore, the final sample for longitudinal analyses included 81 dyads. There were no significant differences between those who completed follow-up information and those who did not related to age, sex, family income, duration of diabetes, therapy type, or glycemic control.
A global coding system, the Iowa Family Interaction Rating Scales (IFIRS; Melby et al., 1998), was used to code observed parenting behaviors via videotaped conversations between the mother–child dyads. The validity of the IFIRS system has been established with correlational and confirmatory factor analyses (Alderfer et al., 2008; Melby & Conger, 2001) and the IFIRS coding system is one of five observational coding systems designed to measure family functioning deemed “well established” for use in pediatric populations (Alderfer et al., 2008).
Two trained bachelor's-level coders viewed each video interaction and scored behaviors on a scale from 1 (absent) to 9 (high level), based on the behavior's frequency, context, affect, intensity, and proportion. The mean of the two coders' scores were used in analyses. Following procedures used previously with the IFIRS codes (e.g., Compas et al., 2010; Lim, Wood, & Miller, 2008), composite codes were created to measure the two constructs of interest: collaborative parenting (α = .83) and overinvolved parenting (α = .71). The collaborative parenting composite included the following codes: (a) Communication, the extent to which the mother uses appropriate explanations and gives appropriate feedback (e.g., “I know that you're busy, but I think you could check your blood sugar more”), intraclass correlation (ICC) = .68; (b) Positive Reinforcement, the extent to which the mother responds to her child's desired behavior with praise or rewards (e.g., “I'm really proud of how you've dealt with all of this”), ICC = .77; and (c) Child Centered, the degree to which the mother displays an awareness of the child's needs, moods, interests, and capabilities, and is “in sync” with the child's interactions, ICC = .35. Scores on the collaborative parenting composite ranged from 9–26, with higher scores indicating higher levels of collaborative parenting. The codes selected to measure overinvolved/intrusive parenting included: (a) Parental Influence, parental attempts to regulate, control or influence the child's behavior (e.g., “You need to check your blood sugar before you drive to the game tonight”), ICC = .45; (b) Intrusiveness, overcontrolling behaviors that are parent-centered and emphasize task completion rather than promoting the child's autonomy (e.g., “I don't care if you don't want to talk about your high blood sugars, we have to”), ICC = .26; and (c) Lecturing/Moralizing, the extent to which the parent tells the child how to think in a way that assumes superior wisdom and provides little opportunity for the child to think independently (e.g., “You should know better than to leave home without your meter”), ICC = .67. Scores on the Overinvolved/Intrusive composite range from 4–23, with higher scores indicating greater observed overinvolved/intrusive parenting.
Adolescents completed the Child Depression Inventory (CDI, Kovacs, 1985) as a measure of current depressive symptoms. The CDI consists of 27 items, with scores ranging from 0 to 54; higher scores indicate greater depressive symptoms. Scores ≥13 suggest clinical levels of depression, and adolescents in the current study who scored at or above this cutoff were evaluated for depression by a licensed clinical psychologist, with referrals made as needed. Internal consistency for the current sample was .90 at Time 1 and .89 at Time 2.
Data was collected on treatment type (pump vs. injection) from adolescents' medical records because use of insulin pumps in adolescents has been associated with improved glycemic control (Sherr & Tamborlane, 2008). In addition, diagnosis date was obtained from adolescents' medical records, as greater duration of diabetes has been associated with poorer glycemic control (Hood et al., 2014).
Mothers completed the Center for Epidemiologic Studies of Depression Scale (CES-D) as a self-report measure of depressive symptoms (Radloff, 1977). The CES-D consists of 20 items, with scores ranging from 0–60; higher scores indicate more depressive symptoms. Internal consistency for the CES-D in the present sample was excellent (α=.90).
Mothers completed the Type I Diabetes version of the Responses to Stress Questionnaire (RSQ) as a measure of stress related to diabetes (Connor-Smith et al., 2000). As noted above, the RSQ includes 10 items on which mothers report how often they experienced a particular diabetes stressor (e.g., dealing with diabetes care, frequently reminding adolescent to take care of him/herself). Items are summed to calculate a Total Stress Score, with scores ranging from 0–30; higher scores indicate greater diabetes-related stress. Internal consistency for the RSQ distress score was α = .63.
Mothers' current levels of anxiety were measured by the State Trait Anxiety Inventory–State Scale (Spielberg, Gorsuch, Lushene, Vagg, & Jacobs, 1983). The STAI consists of 20 items in which respondents rate current anxiety-related feelings on a 4-point Likert scale ranging from not at all to very much so, with higher scores indicating greater symptoms of anxiety. The internal consistency for the state scale in the current sample was excellent (α = .91).
Glycosylated hemoglobin (HbA1c) was used as an objective measure of adolescents' glycemic control over the most recent 8–12 weeks. The Bayer Diagnostics DCA2000® machine was used to conduct analyses of HbA1c; a normal range is considered 4.2–6.3%. A target HbA1c of <7.5% is recommended for children and adolescents (American Diabetes Association, 2015).
Analyses were performed using SPSS (v. 16.0, Chicago, IL) statistical packages. Means, standard deviations, and ranges of scores for mothers' symptoms of anxiety and depression, diabetes-related stress, observed parenting behaviors, adolescents' depressive symptoms and HbA1c were calculated. T tests were conducted to test for differences related to child sex and treatment type on all key variables (i.e., parenting constructs, HbA1c, and CDI). Bivariate Pearson's correlations were calculated to examine associations among observed collaborative and overinvolved/intrusive parenting behaviors, parental stress/distress, and adolescents' depressive symptoms, glycemic control, age, and duration of T1D. Regression models were conducted to examine parenting as a predictor of glycemic control (HbA1c) and depressive symptoms (CDI), adjusting for baseline levels and covariates (sex, treatment type, income, age). Finally, child age, sex, and treatment type (insulin pump vs. injections) were examined as potential moderators of parenting by centering age and parenting variables and creating interaction terms.
Descriptive statistics and bivariate correlations for parenting behaviors and parents' diabetes-related stress, depressive symptoms, and anxiety are presented in Table 2. HbA1c increased significantly, F = 4.97, p = .03 from Time 1 (M = 7.62) to Time 2 (M = 7.84); 52.7% of the adolescents in the sample were in the target range (HbA1c <7.5%) at baseline, and 44.7% of the adolescents were in the target range at 12 months. While relatively well controlled, this is similar to the glycemic control seen in other samples of similar aged youth with T1D (e.g., Luyckx, Seiffge-Krenke, & Hampson, 2010). Mean CDI scores decreased slightly (Time 1: M = 5.11; Time 2: M = 4.31, p = .56), yet 10.9% of the adolescents scored above the clinical cutoff for depressive symptoms (CDI score ≥13) at baseline, and 12.2% scored above the cutoff at 12 months. This is lower than percentages found in past research with similar, but slightly older samples (Lawrence et al., 2006; Hood et al., 2006), but close to the rates found in more recently in Bernstein, Stockwell, Gallagher, Rosenthal, and Soren (2013). Of a possible score of 27, the mean IFIRS score for observed collaborative parenting was 15.95, indicating that collaborative parenting was “somewhat characteristic” (i.e., the behavioral occasionally occurs and is at a low to moderate level of intensity) in this sample. The mean level overinvolved parenting was 11.10, indicating that overinvolved parenting occurred in between “minimally” (the behavior rarely occurs and is of low intensity) and “somewhat characteristic” in this sample (Melby et al., 1998).
There was a trend toward a significant difference in parenting by treatment type, t = 1.78, p = .07, such that higher levels of collaborative parenting were observed in mothers of adolescents using insulin pumps (M = 16.25) compared to mothers of adolescents using injections (M = 14.37). There were no significant differences on key variables related to adolescent sex. Adolescent age was significantly related to children's depressive symptoms at 12 months, r = .28, p < .05, but duration of T1D was unrelated to any of the constructs measured.
All of the mothers in the current sample reported some diabetes-related stress on the RSQ; scores ranged from 5 to 22, with a mean total stress score of 12.4 (±3.4). Approximately 18% of the mothers scored above the clinical cutoff for depression, with a mean score of 10.2 (±8.3) on the CES-D. Approximately 13% of the mothers scored above the clinical cutoff for state anxiety, with a mean score of 32.4 (±9.3). These rates of distress were similar to other studies with parents of adolescents with T1D (Whittemore et al., 2012).
As hypothesized, maternal symptoms of anxiety (r = .24) and diabetes-related stress (r = .21) were significantly related to observed overinvolved parenting (see Table 2). However, maternal depressive symptoms, as measured by the CESD, were not significantly related to overinvolved parenting, and observed collaborative parenting was not significantly related to measures assessing maternal stress and distress (ps > .05).
Bivariate correlations indicated that observed collaborative parenting was significantly related to lower HbA1c at 12 months, r = −.28, p < .05, but was not significantly related to lower depressive symptoms at 12 months. Conversely, overinvolved parent behavior observed at baseline was related to greater depressive symptoms 12 months later, r = .28, p < .05, but was not significantly associated with HbA1c.
Multiple linear regression analyses were conducted to test observed parenting as a predictor of changes in depressive symptoms (CDI) and glycemic control (HbA1c) over time. First, a linear regression analysis was conducted to predict HbA1c at 12 months, adjusting for baseline levels of HbA1c and potential covariates (child sex, treatment type, family income, and child age) in Step 1, observed collaborative parenting in Step 2, and interaction terms in Step 3 (age and parenting variables were centered). The overall model was significant, F = 6.65, p < .001, explaining 39% of the variance in HbA1c (see Table 3). Analyses revealed that higher observed levels of collaborative parenting at baseline significantly predicted lower HbA1c levels in adolescents 12 months later when adjusting for demographic variables (β = −.19, p = .05). No interaction terms were significant. A similar model was analyzed to determine if overinvolved parenting was a significant predictor of HbA1c at 12 months. Although the overall model was significant, F = 5.54, p < .001, explaining 33% of the variance in HbA1c, the only significant predictor was baseline HbA1c (β = .30, p < .001).
Next, a linear regression analysis was conducted to predict adolescents' depressive symptoms over 12 months, adjusting for baseline depressive symptoms and covariates (child sex, treatment type, family income, and child age) in Step 1, observed collaborative parenting (Step 2), and interaction terms (Step 3). The model predicting child depressive symptoms was significant, F = 2.76, p = .009, explaining 18% of the variance in CDI scores at 12 months. However, in the final model, only baseline CDI score (β=.40, p = .001) and child age (β=.26, p = .030) significantly predicted depressive symptoms 12 months later. Collaborative parenting was not a significant predictor of adolescent depressive symptoms, nor was any of the interaction terms.
Finally, a regression model was analyzed to determine if overinvolved parenting was a significant predictor of CDI at 12 months (see Table 4). The overall model was significant, F = 4.43, p < .001, explaining 30% of the variance in CDI. After adjusting for baseline CDI and covariates, overinvolved parenting was significantly related to child depressive symptoms 12 months later (β = .23, p = .033). In addition, the interaction between overinvolved parenting and treatment type was significant (β = .77, p = .034), such that higher levels of overinvolved parenting in adolescents using injections were related to significantly higher depressive symptoms than in adolescents using insulin pumps (see Figure 1). The interaction between overinvolved parenting and child age was also significant (β = −.22, p = .042), such that overinvolved parenting was a stronger predictor of depressive symptoms in younger adolescents than in older adolescents (see Figure 2).
Findings from the current study extend previous research by examining the associations between maternal stress/distress and observed parenting behaviors, as well as the prospective effects of observed collaborative and overinvolved parenting on mental and physical health outcomes in adolescents with T1D over a 1-year period. Mothers in the current sample exhibited clinically significant levels of distress similar to the levels found in other studies that have examined depression and anxiety in mothers of children with T1D (e.g., Eckshtain, Ellis, Kolmodin, & Naar-King, 2010; Jaser, Whittemore, Ambrosino, Lindemann, & Grey, 2009). As hypothesized, maternal symptoms of anxiety and diabetes-related stress were significantly related to overinvolved parenting in this sample, indicating that mothers experiencing significant stress or anxiety may engage in more intrusive parenting behaviors. This finding is in line with prior work suggesting that parental anxiety may prime heightened interference with offspring (Dix & Meunier, 2009), and suggests that interventions designed to reduce the anxiety and stress associated with caring for a child with T1D have the potential to improve parenting behaviors.
Results from the current study also support previous research citing collaborative parenting as a protective factor against deteriorating adherence and HbA1c (e.g., Greene et al., 2010; Shorer et al., 2011). In this sample, longitudinal analyses indicated that observed collaborative parenting at baseline was significantly related to lower HbA1c (i.e., better glycemic control) 12 months later. It is thought that collaborative parenting reinforces proper self-managing behaviors in youth by incentivizing diabetes management tasks through parental communication, praise, and positive feedback (Stoeckel & Duke, 2015). Thus, findings from the current study support that interventions promoting effective communication skills, increasing positive reinforcement of child behaviors, and improving attentive listening skills may offer promise as a way to improve health outcomes in youth.
Conversely, observed overinvolved parenting was related to an increase in adolescents' depressive symptoms over a 1-year period. This finding is in line with the general literature citing an association between overly intrusive parenting behaviors and negative psychological outcomes in youth (e.g., McKee et al., 2007). Notably, past studies of the effects of intrusive parenting have largely focused on outwardly hostile parenting behaviors (e.g., Gershoff, Lansford, Sexton, Davis-Kean, & Sameroff, 2012), while other studies have focused on the negative impact of the lack of parental involvement (e.g., Eckshtain et al., 2010). The current study suggests that intrusive parenting in the form of lecturing and behavioral control may be just as aversive for youth, predicting similar outcomes (i.e., depressive symptoms) over time.
Further, overinvolved parenting was a stronger predictor of depressive symptoms in adolescents using insulin injections (vs. pumps). Given that adolescents may be discouraged from using insulin pumps if providers perceive that they are not exhibiting good diabetes management (Valenzuela, La Greca, Hsin, Taylor, & Delamater, 2011), this finding may indicate that parents were responding to adolescents' poor adherence with more critical, overinvolved parenting. Notably, the relationship between parent and child behavior has been found to be bidirectional (Kiff, Lengua, & Zalewski, 2011); thus, analyzing effects of children's behavior on parenting should be included in future studies.
Lastly, child age was significantly associated with depressive symptoms in youth, and the interaction between overinvolved parenting and age was significant. These findings suggest that, although older age is generally associated with greater depressive symptoms, overinvolved parenting, characterized by nagging, lecturing, and interrogating, may be especially harmful for younger adolescents. Future studies should test age as a moderator between parenting and mental and physical health outcomes. Interestingly, child age was not significantly associated with observed collaborative or overinvolved parenting behaviors in correlation analyses. This could indicate that mothers display similar parenting styles regardless of age. However, it is likely that a larger sample is needed to detect age-related differences in parenting.
Taken together, these findings suggest a possible pattern of specificity, by which collaborative behaviors by parents are linked to physical health outcomes while overinvolved behaviors are linked to psychological outcomes. Thus, to have the greatest broadband impact, interventions should focus not only on decreasing overinvolved parenting behaviors, but additionally promoting certain positive parenting behaviors, specifically those characterized by responsiveness to children.
The present study has several limitations that should be noted. First, the sample is limited by the relatively small percentage of minority families, which did not allow us to examine differences in specific racial or ethnic groups, or to identify parenting differences related to race/ethnicity. In the context of T1D, ethnicity has been found to moderate the association between the parent–child relationship and diabetes management in youth; for example, one study found that Latino adolescents reported lower levels of parental acceptance and higher levels diabetes conflict than Caucasian adolescents, but that conflict was only associated with poorer adherence in Caucasians (Main et al., 2014). While the observational design was a strength of the study, the laboratory setting may result in less representative parenting behaviors than in the home environment and may not capture uninvolved parenting. Additionally, the current sample had a relatively high income and good glycemic control, which may impact findings. For example, the relatively lower rate of depressive symptoms in this sample may be related to adolescents' better overall glycemic control. Thus, results of the current study should be replicated in other samples to determine if findings generalize to lower income adolescents or those with poorly controlled diabetes. Finally, while the longitudinal design of the present study is a strength, it is possible that problems with child behavior related to diabetes management or glycemic control predated the study period.
Overall, the current study extends previous research by examining two distinct types of parenting—collaborative and overinvolved—observationally as predictors of psychological and physical health outcomes in youth over time. Findings suggest that there may be a specific effect by which collaborative parenting is related to improvements in glycemic control while overinvolved parenting is related to increases in depressive symptoms one year. Implications for intervention include: (a) decreasing anxiety and diabetes related stress in mothers of children with T1D, (b) promoting parental communication, listener responsiveness, and communication to facilitate optimal glycemic control in youth, and (c) decreasing intrusive parenting behaviors to decrease depressive symptoms and promote positive psychological health in youth.
This research was supported in part by grants from the National Institute of Diabetes and Digestive and Kidney Diseases (K23 NK088454) and the National Center for Research Resources (UL1 RR024139).