Reprinted from Diabetes Care


An Office-Based Intervention to Maintain Parent-Adolescent Teamwork in Diabetes Management: Impact on Parent Involvement, Family Conflict, and Subsequent Glycemic Control

Barbara J. Anderson, PHD, Julienne Brackett, BA, Joyce Ho, BA, Lori M.B. Laffel, MD, MPH, Joslin Diabetes Center, Children's Hospital, and Harvard Medical School, Boston, Massachusetts.

[Diabetes Care 22(5):713-721, 1999. 1999 American Diabetes Association, Inc.]


Objective: To design and evaluate an office-based intervention aimed at maintaining parent-adolescent teamwork in diabetes management tasks without increasing diabetes-related family conflict.

Research Design and Methods: There were 85 patients (aged 10-15 years, mean 12.6 years) with type 1 diabetes (mean duration 5.5 years; mean HbA
1c 8.5%) who were randomly assigned to one of three study groups:teamwork, attention control, and standard care:and followed for 24 months. At each visit, parent involvement in insulin administration and blood glucose monitoring was assessed. The teamwork and attention control interventions were integrated into routine ambulatory visits over the first 12 months (four medical visits). Measures of diabetes-related family conflict were collected at baseline and after 12 months. All patients were followed for an additional 12 months with respect to glycemic control.
Results: In the teamwork group, there was no major deterioration (0%) in parent involvement in insulin administration, in contrast to 16% major deterioration in the combined comparison (attention control and standard care) group (
P< 0.03). Similarly, no teamwork families showed major deterioration in parent involvement with blood glucose monitoring versus 11% in the comparison group (P< 0.07). On both the Diabetes Family Conflict Scale and the Diabetes Family Behavior Checklist, teamwork families reported significantly less conflict at 12 months. An analysis of HbA
1c over the 12- to 24-month follow-up period indicated that more adolescents in the teamwork group (68%) than in the comparison group (47%) improved their HbA1c (P< 0.07).

Conclusions: The data demonstrate that parent involvement in diabetes management tasks can be strengthened through a low-intensity intervention integrated into routine follow-up diabetes care. Moreover, despite increased engagement between teen and parent centered around diabetes tasks, the teamwork families showed decreased diabetes-related family conflict. Within the context of a broader cultural recognition of the protective function of parent involvement in the lives of adolescents, the findings of this study reinforce the potential value of a parent-adolescent partnership in managing chronic disease.


There is a consensus among empirical studies of individuals with type 1 diabetes that adolescents as a group display the worst glycemic control when compared with other age-groups.[1-3] Moreover, the pubertal years during adolescence are consistently identified as a period of deteriorating blood glucose control[2] and heightened family conflict over diabetes management.[4,5]

Clearly, fundamental conflicts exist between the tasks of managing diabetes and the normative developmental tasks facing the young adolescent -- striving to be comfortable with a rapidly maturing body, defining an identity, establishing a new role in the family, and gaining positive acceptance by peers.[6,7] Cross-sectional and prospective studies reveal that both diabetes-specific family conflict and general family conflict are associated with lower adherence rates in adolescents.[8-12] Moreover, these struggles occur within the broader context of increased expectations by parents and health care providers for the young adolescent to assume more independence in self-care responsibilities.[13,14] In the post-Diabetes Control and Complications Trial era, parents, as well as health care providers, have increased expectations for good glycemic control in adolescents with diabetes.[15]

In recent family studies, it has been documented that there is an erosion of parental involvement and support for diabetes management tasks over the early adolescent years.[13,16] Furthermore, there is consistent agreement among empirical studies that children and adolescents who assume early responsibility for their diabetes management are less adherent, have more mistakes in their self-care, and are in poorer glycemic control than those whose parents remain involved.[13,16-18]

A consensus appears to be developing among investigators concerned with general adolescent development that separation from parents during early adolescence increases their vulnerability to negative peer influence[19,20], while engagement with parents enhances ego development and individuation in young adolescents and has positive health, academic, and social-emotional outcomes.[21,22] Current research supports the corollary that early distancing from parents puts the young adolescent at risk for health-compromising behaviors.[22]

Parental involvement, however, can also lead to conflict and stress in the parent-adolescent relationship. It is clear that supportive behaviors must be individualized depending on the adolescent's developmental level, temperament, and the circumstances of each family.[23] Parent involvement can, at times, undermine healthy adolescent self-care behavior, as in the behavioral interaction cycle of "miscarried helping"[24] that occurs when the offered help shames, blames, or humiliates the patient. In the context of a chronic illness like diabetes, "miscarried helping" can escalate parent-adolescent conflict, which undermines adolescent adherence and positive medical outcomes.[5,25]

Two priorities stand out from the recent research literature on adolescents with diabetes and their families: sustaining parent involvement and minimizing parent-adolescent conflict are both important for positive health outcomes over the early adolescent years. In addition, there is a pressing need for family-focused interventions, which are more feasible than group interventions, and which avoid the prohibitive cost of individual counseling.[26] These interventions must focus on family behaviors as well as on glycemic control.[27]

The purpose of this study was to design and evaluate an office-based intervention aimed at maintaining parent-adolescent teamwork in diabetes management tasks without increasing diabetes-related conflict between parent and teen. Priority was given to the development of a family-focused and low-cost intervention that could be integrated into the regular follow-up appointments of youth with diabetes. In this era of health care reform and cost containment, our focus was also to design an intervention that would provide recommendations for health care providers that realistically could be translated into office-based care on a wide-scale basis.

Three research questions guided this study. First, will a brief low-intensity intervention be able to maintain parent involvement in the tasks of diabetes management? Second, if parent involvement is maintained over the 12-month study period, can this intervention also prevent escalation of diabetes-related family conflict? Third, what is the impact of this family intervention on subsequent glycemic control?

Research Design and Methods


Study participants were pre- and young adolescents with type 1 diabetes, aged 10-15 years, and their parents, who were followed at the Pediatric Unit of the Joslin Diabetes Center. Patient records were reviewed for the following eligibility criteria: duration of type 1 diabetes >1 year, reasonable glycemic control (HbA1c from 6.6 to 10.4% [reference range 4.0-6.0%]), no documented serious medical or psychiatric condition in the patient or their parents (as defined by a medical diagnosis recorded in the patient's chart by the patient's physician), residence in New England or New York, at least one outpatient medical visit in the previous year, and ability to come to Joslin for medical visits 3-4 times over the next calendar year.

Letters of introduction were mailed to each eligible family and followed up by telephone contact. Of the 140 eligible families, 89 (or 64% of eligible families) agreed to participate. Most families who refused participation were not able to make the time commitment for 3-4 outpatient visits over the 1-year study period. There were no significant differences between study families and those who declined study participation with respect to attained age, disease duration, frequency of injections per day, or glycemic control measured as HbA1c. This study was approved by Joslin's Committee on Human Subjects, and written informed consent was obtained from all families before entry.


Subjects were randomly assigned to one of three study groups (teamwork intervention, attention control, or standard care), stratified according to age and sex to insure equal representation of younger (10-12 years) and older (13-15 years) male and female patients in each group. Two families (one in standard care and one in attention control) were lost to follow-up because of changes in residence and health insurance coverage after the first study visit and were excluded. Two additional families from the teamwork intervention were excluded because of the identification of serious maternal psychiatric conditions (exclusion criteria). Thus, the final sample included 85 families: 28 in the teamwork group, 30 in attention control, and 27 in standard care.


During the 12-month study period, all subjects in each of the three study groups had four routine ambulatory appointments for their diabetes care at 3- to 4-month intervals from members of the pediatric diabetes team. Families randomly assigned to the teamwork intervention and attention control conditions also met individually with a research assistant for 20- to 30-min intervention sessions, immediately before or after the routine medical appointment during the 12-month study period, for a total of four sessions. All patients and families in the teamwork and attention control groups received all four sessions. (The topics for each intervention module are shown in Table 1.) The authors created written teaching modules that were administered at each session. Research assistants implemented these modules with a scripted protocol to insure consistency of intervention delivery. The research assistants received diabetes training by observing a multidisciplinary pediatric team for 2 months before study onset. Families in all three groups were followed up for an additional 12-month period, with ascertainment of glycemic control at each visit during the entire 24-month period.

For all patients, adherence assessments and clinical data were gathered by the health care team at each visit. In addition, at each visit, a research assistant conducted a brief joint patient and parent interview to update demographic information and to assess the division of responsibility for diabetes management tasks in the family over the preceding month. Self-reported questionnaire data were gathered from all families at the beginning and end of the 12-month study period.

Teamwork intervention condition. The intervention for families in the teamwork condition focused on the importance of parent-teen responsibility sharing for diabetes tasks and ways to avoid conflicts that undermine such teamwork. The modules focused on common conflicts or issues that may interfere with parent-adolescent teamwork around diabetes management (Table 1). The current teamwork intervention was based on written materials that can be transported to any office-based practice and can guide and support recommendations for encouraging parent-adolescent teamwork from diagnosis through various stages of development. These written materials emphasized three key points: 1) the multiple causes of high and low blood glucose levels during early adolescence; 2) realistic expectations for blood glucose levels and for behaviors during early adolescence; and 3) the importance of parents maintaining involvement with insulin injections and blood glucose monitoring without shaming and blaming the young teen.

From a public health perspective, a low-cost intervention integrating these themes into routine diabetes medical care may be appropriate for families for whom group-based interventions or therapy approaches are neither acceptable nor affordable.[28]

At each visit, the research assistant encouraged active family discussion and provided brief written materials designed to reinforce the module topic. At the conclusion of the first session, the young adolescent and the parent negotiated a responsibility-sharing plan. This plan outlined who would be responsible for the different tasks involved in insulin injections, such as deciding the insulin dose, drawing up insulin, and doing the injection, as well as for the tasks of blood glucose monitoring. The plan also indicated whether a parent would supervise the injection and/or know the blood sugar value. The plan emphasized the need for the family to work as a team to manage diabetes, with the parents offering hands-on as well as emotional support to the adolescent. This plan was reviewed, reinforced, and/or renegotiated at each subsequent visit during the 12-month study period.

Attention control condition. The attention control group received time and attention from the research assistant equivalent to that provided to families in the teamwork group. The attention control sessions provided didactic "traditional" diabetes education, with no focus on parental involvement (Table 1). No plan for parent-adolescent teamwork was negotiated.

Standard care condition. Families randomly assigned to the standard care condition received routine clinical care from the diabetes team every 3-4 months over the 12-month study period. Families in this group had no intervention sessions with a research assistant.


Three levels of data were measured in this project: parent involvement in diabetes management, diabetes-related family conflict, and glycemic control.

Measures of parental involvement in diabetes management tasks. To assess in detail the division of responsibility within families during a typical day for two major tasks of diabetes management, insulin injections and blood glucose monitoring, the investigators developed an interview to ascertain the current insulin and blood glucose monitoring routines in the family. The specific coding details for these two measures have been reported previously.[29] Briefly, families were asked who was usually (in the past month) responsible for five components of injecting insulin and for four components of monitoring blood sugars. Because there is no single set of parent behaviors that constitutes an "involved parent," the components of parent involvement in insulin injections or blood sugar monitoring were combined to create the Insulin Routine Score and the Blood Glucose Monitoring Score. Component behaviors for insulin administration and blood glucose monitoring were coded with the creation of two composite scores ranging from 1 to 4, with 1 = no parental involvement (adolescent has total responsibility), 2 = minimal parental involvement, 3 = moderate parental involvement, and 4 = maximum parental involvement (adolescent has no responsibility). Because the coding systems involved combining component behaviors into the higher order combinations of parent involvement, reliability of this coding system was checked independently by two trained research assistants. Comparison of the coding by each research assistant revealed 94% inter-rater reliability. Reliability checks were maintained throughout the study period.

Measures of diabetes-related conflict. At baseline and 12 months, parents completed the Diabetes Family Conflict Scale to assess the degree of family conflict in 17 diabetes management tasks.[30] Recent work by the developers of this measure have revealed excellent internal validity of the scale, with alpha = 0.90 for parental responses (M. Peyrot, personal communication). The level of conflict in the family over diabetes-specific tasks was rated on a three-point scale, with 1 = always hassle and 3 = never hassle. We chose, however, to sum the number of items in which any level of conflict was acknowledged (1 or 2) because the impact of social desirability on parent responses makes it difficult to distinguish a meaningful difference between families reporting 1s and 2s as responses. Therefore, scores could range from 0-17, with a 17 indicating conflict on all items.

At baseline and 12 months, the parents also completed the Diabetes Family Behavior Checklist.[31] This 16-item checklist assessed the parent's perceptions of the frequency of both supportive and unsupportive parent responses with respect to their child's diabetes management behavior. The Unsupportive Behavior Subscale for the Diabetes Family Behavior Checklist was selected to provide a second measure of negative family interaction around diabetes, thereby providing cross-validation for the Diabetes Family Conflict Scale. Two items from the Unsupportive Behavior Subscale were omitted because they had become outdated (following a diabetic "diet," etc.). Behaviors are rated on a five-point scale ranging from "at least once a day" to "never." Both Supportive and Unsupportive Subscale scores were generated, but this report focuses solely on the Unsupportive Behavior Subscale.

Measure of glycemic control. Blood sampling to assess glycemic control occurred at each visit. Initially, total glycosylated hemoglobin (HbA1) was measured by electrophoresis (reference range 5.4-7.4%; Corning Medical and Scientific, Corning, NY), but during the study, lab methodology changed to a method measuring HbA1c (reference range 4.0-6.0%) using high-performance liquid chromatography (Bio-Rad Variant, Hercules, CA). To allow for comparison between HbA1 and HbA1c values, a conversion formula derived from a regression analysis of 700 samples analyzed by both methods was used (HbA1c = 0.77 X HbA1+ 0.44). All glycemic control data are reported as HbA1c values.

Statistical analysis. Statistical analysis of the data was performed using SAS for Windows (Release Version 6.12). Means SD are presented unless otherwise indicated. The analyses included unpaired t tests, analysis of variance (ANOVA), Pearson bivariate correlations, X2, and multivariate analysis. In general, univariate analyses were examined first. The complex relationships among the developmental, behavioral, and biological variables called for multivariate analyses to control for potentially confounding covariates. P values <0.05 were considered significant.



Baseline values were examined to ensure comparability of study groups at entry. There were no significant differences in demographic or clinical characteristics across the three study groups (Table 2). In addition, the occupational status of parents was similar in the teamwork, attention control, and standard care groups, 3.1, 3.4, and 2.8, respectively, with the mean for all groups indicating skilled workers according to Hollingshead. Patients were ~12.6 years of age, with a diabetes duration of ~5.5 years. More than 50% of patients in each group were receiving three injections per day and were checking blood sugars three or more times per day. It is important to note the striking homogeneity and relative intensity in home-care behaviors among the patients across the three study groups. This intensity is due, in part, to the eligibility criteria established for this longitudinal research sample, and the homogeneity reflects adequate randomization.

Table 3 presents baseline values for measures of parent involvement in insulin injections and blood glucose monitoring for the three study groups. There were no significant differences between the three study groups at baseline with respect to level of parent involvement in insulin or blood glucose monitoring tasks. More than 50% of parents in each of the study groups demonstrated moderate or maximum involvement in insulin administration and >40% of parents in each group demonstrated moderate or maximum involvement in blood glucose monitoring at baseline.

Table 3 also presents baseline values for the three study groups with respect to the two measures of diabetes-related conflict, Rubin's Diabetes Conflict Scale and the Unsupportive Behavior Subscale of the Diabetes Family Behavior Checklist. On Rubin's Diabetes Conflict Scale, parents endorsed ~4 of l7 conflict items. For the Unsupportive Behavior Subscale of the Diabetes Family Behavior Checklist, at baseline, parents endorsed ~12.5 of a possible 25 Unsupportive Behavior Score. For both measures of parent-reported diabetes-related family conflict, there were no statistically significant differences across the three study groups at baseline.

Outcomes of the 12-Month Study Period: Impact on Teamwork and Conflict

Table 3 also displays the 12-month study data for parent involvement in insulin administration and blood glucose monitoring, family conflict, and glycemic control. A series of X2 analyses and repeated measures ANOVAs were carried out examining changes in parent involvement, family conflict, and glycemic control over the 12-month study period. We found no statistically significant differences between families in the attention control and standard care groups after the 12-month study period on key outcome variables, such as parent involvement, family conflict, and glycemic control. For additional longitudinal data analyses and presentation of results, we combined study groups to form a single comparison group with n = 57 to increase the power of our study.[32]

The primary goal of this intervention was to prevent the deterioration in parental involvement with diabetes management tasks that often occurs during early adolescence. With respect to involvement in both insulin administration and blood glucose monitoring, we defined deterioration in parental involvement as decreasing involvement by one or more categories, as defined earlier, in measures of parental involvement. Figure 1 presents the percentage of parents demonstrating increased involvement, no change, and deterioration of parent involvement in insulin administration and blood glucose monitoring from baseline to the end of the 12-month intervention period for the teamwork group contrasted with the comparison group.

There was no major deterioration in parental involvement in insulin administration in any family in the teamwork group (Fig. 1A). Significantly more parents in the comparison group (16%) showed major deterioration in parental involvement in insulin administration (X2 = 4.95, df = 1, P< 0.03).

Figure 1B presents the percentage of parents demonstrating increased involvement, no change, or major deterioration of parent involvement in blood glucose monitoring over the 12-month study period for the two groups. Again, there was no major deterioration in parental involvement in any family in the teamwork group compared with 11% of parents in the comparison group who showed major deterioration in parental involvement in blood glucose monitoring. There was a trend for this change in parental involvement to reach statistical significance (X2 = 3.17, df = 1, P< 0.075).

The second research question concerned the impact of the teamwork intervention on the level of diabetes-related conflict in the family. Because the teamwork parents were more engaged with their adolescents around diabetes tasks, they had more potential for family conflict over diabetes-related tasks than did families in the comparison group. Figure 2 presents the mean level of diabetes-related family conflict from the Diabetes Conflict Scale by group at baseline and at the end of the 12-month study period. ANOVA revealed that teamwork families reported a significantly greater decrease in diabetes-specific conflict at the end of the study period, as shown by a significant group by time interaction (F = 4.97 df = 1, P< 0.02). No significant change in the level of conflict was reported in the comparison group.

Next, we examined changes in the Unsupportive Behavior Subscale of the Diabetes Family Behavior Checklist over the 12-month study period (Fig. 3). ANOVA revealed a significant effect for time (F = 10.35, df = 1, P< 0.002), as well as a significant group X time interaction (F = 5.66, df = 1, P< 0.02). While parents in the study showed a decrease in diabetes-related unsupportive behavior over the study period, parents in the teamwork group reported a significantly greater decrease in their negative behavior than did parents in the comparison group.

Follow-Up Period of 12-24 Months: Impact on Glycemic Control

The final research question concerned the impact of increased family teamwork on subsequent glycemic control. We did not hypothesize changes in HbA1c during the 12-month intervention period, since the primary goal of this intervention was on changing patterns of parent involvement in diabetes management tasks and not directly on factors more likely to impact immediately on glycemia, such as intensity of the diabetes treatment regimen. However, once family behavior patterns concerning teamwork and conflict were changed, we hypothesized an impact on glycemic control over the subsequent 12-month period, from 12 to 24 months. While there was no significant difference in mean HbA1c between groups at 24 months, 8.7 1.2 and 8.8 1.1% for the teamwork and comparison groups, respectively, there was a trend for more patients in the teamwork intervention to improve. Change scores for HbA1c from 12 to 24 months revealed a mean change in the teamwork group of -0.20 1.1 versus 0.11 1.1 in the comparison group. We categorized adolescents in the teamwork and comparison groups on the quality of their HbA1c change as improved (or no change) versus deteriorated (defined as a greater than one percentage point increase), using a standard strategy for classification of individual HbA1c data.[2,33] There was a trend for more adolescents in the teamwork group (68%) to improve, in contrast with the comparison group, where 47% improved their HbA1c in the 12 months following the intervention period (X2 = 3.17, df = 1, P< 0.07). Calculating an odds ratio, we determined that adolescents had 2.4 times the chance of improving their glycemic control if they were in the teamwork group than in the comparison group (P< 0.07).


Over the course of this study, an erosion in parental involvement in diabetes tasks occurred in the families who were not exposed to the intervention focused on fostering parent-adolescent teamwork at their regular diabetes follow-up visits. These results are consistent with other reports that parent involvement in the tasks of diabetes management deteriorates in young adolescents over time with increasing age and disease duration.[13,16,18] Diabetes clinicians are only beginning to recognize, however, the importance of continued parent involvement on glycemic and behavioral outcomes in adolescents.

We designed a relatively low-cost low-intensity intervention that was integrated into the routine diabetes follow-up medical care of young adolescent patients. This teamwork intervention was dramatically effective in preventing the expected erosion in parental involvement in diabetes management. We found that within the teamwork intervention condition, parents and teens were able to sustain shared responsibility for insulin administration significantly more than families in the attention control and standard care conditions. In addition, there was a trend for parents and teens in the teamwork condition, compared with families in the comparison groups, to sustain, and even increase, involvement with blood glucose monitoring tasks. The current study was carried out within the context of current developmental theories that conceptualize the major task of the adolescent period as movement away from dependence on the family, not toward independence, but rather toward interdependence. Interdependence does not require adolescents to distance themselves emotionally from parents, but rather requires a reorganization in which family members renegotiate and redistribute responsibilities and obligations.[34] The large empirical study, the National Longitudinal Study of Adolescent Health, investigated a representative sample of 12,000 adolescents and concluded that parental involvement was the single most important predictor of positive adolescent outcomes, such as school success and avoidance of drug use and teen pregnancy.[22]

Because investigators have recently demonstrated that both parent and child reports of increased family conflict related to diabetes adherence problems and poorer blood glucose control[5], our clinical concern was that the teamwork intervention not lead to increased family conflict. One of the most important findings in this study is that families in the teamwork group with sustained parental involvement in both insulin tasks and blood glucose monitoring did not experience an increase in diabetes-related family conflict. In fact, on two separate measures of diabetes-related family conflict, families in the teamwork condition, who were more engaged around diabetes tasks and thus had more potential for conflict, reported a greater decrease in conflict over time than did families in the comparison group. This is important, given new research from the general adolescent developmental literature that the affective intensity (i.e., the level of hurt and pain) of parent-adolescent conflicts typically increases between the early adolescent period and the mid- to late-adolescent period.[20] This suggests that if pediatric diabetes teams can intervene with anticipatory guidance to encourage more positive parent-child patterns of responsibility sharing during the preadolescent and early adolescent years, when family behaviors are being established, some of the entrenched and intense diabetes conflicts seen frequently between older adolescents and their parents can possibly be prevented. Future research must examine optimal levels of parent involvement at specific ages over the early adolescent period.

It is not family conflict, however, but glycemic control that is receiving the most attention in this new era, with clear demonstration from the Diabetes Control and Complications Trial that improved glycemic control during adolescence has the potential to prevent and/or delay early physiological complications.[35,36] Thus, there is a pressing need for interventions focused on this glycemically at-risk group. It is important to recall that the immediate targets of the teamwork intervention were not glycemic parameters, but rather family behavior patterns. Therefore, it is not surprising that the teamwork intervention did not lead to immediate or dramatic improvements in glycemic control among the participating adolescent patients. However, after 24 months, when patients were examined from the perspective of improving, remaining unchanged, or deteriorating their glycemic control, there was a trend for more patients in the teamwork group to improve. These data suggest that after changes in family interaction patterns have been consolidated, parental involvement in diabetes care and lower levels of family conflict may lead to improved glycemia. Indeed, this may provide the opportunity to intensify management strategies, since the patients can be supported by their parents to improve glycemic control. Similarly, Delamater et al.[37], reporting on results of a pioneering family-based group intervention with adolescents with diabetes, have also suggested that decreasing family conflict may have positive effects on the glycemic and behavioral outcomes of adolescents with type 1 diabetes. One valuable application of this teamwork intervention strategy may be in combining it with more medically based approaches that intensify insulin therapy to improve glycemic control, such as the group-based approach reported by Grey et al.[15], which used coping skills training to improve the efficacy of medical treatment intensification with adolescent patients. For adolescents, who are our most metabolically at-risk group of patients with type 1 diabetes, new pediatric diabetes intervention models may be most effectively built from a synthesis of several current approaches, and target not only insulin regimens, but also family interaction patterns.

Family-based intervention strategies may seem unrealistic, however, given the realities of the tighter allocation of resources within the current health care environment. It is crucial for intervention approaches in chronic disease today to target at-risk groups and to be generalizable across broad groups of patients and health care delivery systems.[28]

However, several cautions must be emphasized in discussing the results of this study. First, these findings are based on a small sample size of 85 families. One consequence of the eligibility criteria set for this longitudinal intervention research was a very homogeneous group of participating families. This intervention needs to be replicated with a larger, more heterogeneous sample of families. In fact, the benefits may be greater if more at-risk patients, adolescents with HbA1c>8.0%, were included in the study. Secondly, the intervention period of this study was relatively brief, and this approach should be replicated over longer intervention and follow-up time periods.

In conclusion, contemporary theories of adolescent development focus on the transformation of the parent-adolescent relationship and the processes that foster continuity of parental influences and minimize the disruption of these influences. The recently published report Great Transitions by the Carnegie Council on Adolescent Development emphasizes re-engaging parents with their young adolescents to promote educational achievement and good health outcomes.[38] Interventions focused on adolescents with diabetes will benefit from this broader cultural awareness that parent involvement protects youth from many high-risk behaviors and negative developmental outcomes. The model of pediatric diabetes care for the new millennium will need to incorporate a public health approach and be built on new developmental constructs of the adolescent-parent relationship, which educate both health care teams and families about the value of a parent-adolescent partnership in managing chronic disease.


This study was supported financially by a grant (DK-46887) from the National Institute of Diabetes, Digestive and Kidney Diseases (to B.A.) and by the Charles H. Hood Foundation.

We wish to acknowledge contributions of the clinical team in the Pediatric Unit of the Joslin Diabetes Center: Dr. Joan Mansfield, Dr. Alyne Ricker, and Dr. Joseph Wolfsdorf, as well as Louise Crescenzi, Paula Michel Fanizzi, Cindy Pasquarello, and Kristen Rice, and the biostatistical consultation of Dr. Dianne Finkelstein, Harvard School of Public Health.


  1. Gowers SG, Jones JC, Kiana S, North CD, Price DA: Family functioning: a correlate of diabetic control? J Child Psychol Psychiatry 36:993-1001, 1995
  2. Daneman D, Wolfson DH, Becker DJ, Drash AL: Factors affecting glycosylated hemoglobin values in children with insulin-dependent diabetes. J Pediatr 99:847-853, 1981
  3. Anderson BJ, Laffel L: Behavioral and family aspects of the treatment of children and adolescents with IDDM. In Ellenberg and Rifkin's Diabetes Mellitus. 5th ed. Porte D, Sherwin R, Rifkin H, Eds. Stamford, CT, Appleton & Lange, 1996, p. 811-825
  4. Anderson BJ: The impact of diabetes on the developmental tasks of childhood and adolescence: a research perspective. In Recent Advances in Diabetes. Nattrass M, Santiago JV, Eds. London, Churchill Livingston, 1984, p. 165-171
  5. Miller-Johnson S, Emery RE, Marvin RS, Clarke W, Lovinger R, Martin M: Parent-child relationships and the management of insulin-dependent diabetes mellitus. J Consult Clin Psychol 62:603-610, 1994
  6. Delamater AM: Compliance interventions for children with diabetes and other chronic diseases. In Developmental Aspects of Health Compliance Behavior.Krasnegor NA, Epstein E, Johnson SB, Yaffe SJ, Eds. Hillsdale, NJ, Lawrence Erlbaum, 1993, p. 336-364
  7. Greydanus DE, Hofmann AD: A perspective on the brittle teenage diabetic. J Fam Pract 9:1007-1012, 1979
  8. Anderson BJ, Miller JP, Auslander WF, Santiago JV: Family characteristics of diabetic adolescents: relationship to glycemic control. Diabetes Care 4:586-594, 1981
  9. Hanson CL, Henggeler SW, Burghen GA: Model of associations between psychosocial variables and health outcome measures of adolescents with IDDM. Diabetes Care 10:752-756, 1987
  10. Jacobson AM, Hauser ST, Wolfsdorf JI, Houlihan J, Milley JE, Herskowitz RD, Wertlieb D, Watt BA: Psychologic predictors of compliance in children with recent onset of diabetes mellitus. J Pediatr 110:805-811, 1987
  11. Hauser ST, Jacobson AM, Lavori P, Wolfsdorf JI, Herskowitz RD, Milley JE, Bliss R, Wertlieb D, Stein J: Adherence among children and adolescents with insulin-dependent diabetes mellitus over a four-year longitudinal follow-up. II. Immediate and long-term linkages with the family milieu. J Pediatr Psychol 15:527-542, 1990
  12. Jacobson AM, Hauser ST, Lavori P, Wolfsdorf JI, Herskowitz RD, Milley JE, Bliss R, Gelfand E, Wertlieb D, Stein J: Adherence among children and adolescents with insulin-dependent diabetes mellitus over a four-year longitudinal follow-up. I. The influence of patient coping and adjustment. J Pediatr Psychol 15:511-526, 1990
  13. Anderson BJ, Auslander WF, Jung KC, Miller JP, Santiago JV: Assessing family sharing of diabetes responsibilities. J Pediatr Psychol 15:477-492, 1990
  14. Follansbee DS: Assuming responsibility for diabetes management: what age? what price? Diabetes Educ 15:347-352, 1989
  15. Grey M, Goland EA, Davidson M, Yu C, Sullivan-Bolyai S, Tamborlane WV: Short-term effects of coping skills training as adjunct to intensive therapy in adolescents. Diabetes Care 21:902-908, 1998
  16. Wysocki T, Taylor A, Hough BS, Linscheid TR, Yeates KO, Naglieri JA: Deviation from developmentally appropriate self-care autonomy. Diabetes Care 19:119-125, 1996
  17. Burns KL, Green P, Chase HP: Psychosocial correlates of glycemic control as a function of age in youth with IDDM. J Adolesc Health 7:311-319, 1986
  18. Weissberg-Benchell J, Glasgow AM, Tynan WD, Wirtz P, Turek J, Ward J: Adolescent diabetes management and mismanagement. Diabetes Care 18:77-82, 1995
  19. Irwin CE, Ed: Adolescent Social Behavior and Health. San Francisco, CA, Jossey-Bass, 1987
  20. Laursen B, Coy KC, Collins WA: Reconsidering changes in parent-child conflict across adolescence: a meta-analysis. Child Dev 69:817-832, 1998
  21. Grolnick WS, Slowiaczek ML: Parents' involvement in children's schooling: a multidimensional conceptualization and motivational model. Child Dev 65:237-252, 1994
  22. Resnick MD, Bearman PS, Blum RW, Bauman KE, Harris KM, Jones J, Tabor J, Beuhring T, Sieving RE, Shew M, Ireland M, Bearinger LH, Udry JR: Protecting adolescents from harm: finding for the National Longitudinal Study on Adolescent Health. JAMA 278:823-832, 1997
  23. Holmbeck GN: A model of family relational transformations during the transition to adolescence: parent-adolescent conflict and adaptation. In Transitions Through Adolescence: Interpersonal Domains and Context. Graber JA, Brooks-Gunn J, Petersen AC, Eds. Mahwah, NJ, Lawrence Erlbaum, 1993, p. 167-199
  24. Coyne JC, Anderson BJ: The "psychosomatic family" reconsidered: diabetes in context. J Marital Fam Ther 14:113-123, 1988
  25. Anderson BJ, Coyne JC: Family context and compliance behavior in chronically ill children. In Developmental Aspects of Health Compliance Behavior.Krasnegor N, Epstein L, Johnson SB, Yaffe SJ, Eds. Hillsdale, NJ, Lawrence Erlbaum, 1993, p. 77-89
  26. Glasgow RE, Osteen V: Evaluating diabetes education: are we measuring the most important outcomes? Diabetes Care 15:423-4232, 1992
  27. Drotar D: Intervention research: pushing back the frontiers of pediatric psychology. J Pediatr Psychol 22:593-606, 1997
  28. Glasgow RE, Kaplan RM, Smith L, Wagner EH, Vinicor F, Norman J. If diabetes is a public health problem, why not treat it as one? a population-based approach to chronic illness. Ann Behav Med. In press
  29. Anderson BJ, Ho J, Brackett J, Finkelstein D, Laffel L: Parental involvement in diabetes management tasks: relationships to blood glucose monitoring adherence and metabolic control in young adolescents with IDDM. J Pediatr 130:257-265, 1997
  30. Rubin R, Young-Hyman D, Peyrot M: Parent-child responsibility and conflict in diabetes care (Abstract). Diabetes 38 (Suppl. 2):28A, 1989
  31. Schafer LC, McCaul KD, Glasgow RE: Supportive and nonsupportive family behaviors: Relationships to adherence and metabolic control in persons with type I diabetes. Diabetes Care 9:179-185, 1986
  32. Rosner B: Fundamentals of Biostatistics. Belmont, CA, Duxbury Press, 1995
  33. Anderson BJ, Wolf FM, Burkhart MT, Cornell RG, Bacon GE: Metabolic effects of a peer group intervention with adolescents with insulin-dependent diabetes mellitus: a randomized controlled study in an outpatient setting. Diabetes Care 12:179-183, 1989
  34. Baumrind D: A developmental perspective on adolescent risk taking in contemporary America. In Adolescent Social Behavior and Health. Irwin CE, Ed. San Francisco, CA, Jossey-Bass, 1987, p. 93-126
  35. Diabetes Control and Complications Trial Research Group: Effect of intensive diabetes treatment on the development and progression of long-term complications in adolescents with insulin-dependent diabetes mellitus: Diabetes Control and Complications Trial. J Pediatr 125:177-188, 1994
  36. Drash AL: The child, the adolescent, and the Diabetes Control and Complications Trial (Editorial). Diabetes Care 16:1515-1516, 1993
  37. Delamater AM, Smith JA, Bubb J, Green-Davis S, Gamble T, White NH, Santiago JV: Family-based behavior therapy for diabetic adolescents. In Advances in Child Health Psychology. Johnson JH, Johnson SB, Eds. Gainesville, University of Florida Press, 1991, p. 293-306
  38. Carnegie Council on Adolescent Development: Great Transitions: Preparing Adolescents for a New Century. New York, Carnegie Corporation of New York, 1995

Home Page
Reload The National Diabetes Center Site