In addition to assessing medical and metabolic outcomes, DCCT investigators measured diabetes-related quality of life (DRQoL) as an index of patients' satisfaction with their diabetes treatment, as well as their worries related to diabetes and its impact on daily life. Among adults, recent studies have shown that better DRQoL is associated with higher levels of social support, self-efficacy, physical activity, education, income, and with the absence of health complications and co-morbid psychiatric disorders.^4-6 Another recent study found that DRQoL and treatment satisfaction improved after intensifying adults' insulin therapy regimens with the use of insulin pens or pumps. Improved DRQoL was likely due to patients' greater perceived flexibility in leisure time activities and diet.⁷

It is also important to consider QoL in children and adolescents with diabetes. Diabetes imposes considerable demands that may interfere with children's ability to negotiate important developmental tasks and achieve good psychosocial adjustment. This is especially the case in the post-DCCT era, when the expectation for many diabetic youths is to initiate intensive insulin regimens in order to attain and maintain optimal levels of glycemic control and reduce their risk of health complications.⁸

DRQoL refers to those aspects of child and adolescent functioning that are directly affected by diabetes and its treatment. This construct may include physical symptoms, functional status, psychological functioning, and social functioning.⁹

The purpose of this article is to review the results of studies that have focused on DRQoL in youths. Because relatively little research has directly considered DRQoL, studies targeting behavioral and psychosocial functioning, which indirectly consider QoL, will also be reviewed. The implications of these findings for current clinical practice and future research will then be discussed.

The demographic and clinical correlates of DRQoL were evaluated in a study of 74 children and adolescents from culturally and socially diverse backgrounds. The mean age and diabetes duration of the study sample was 16.0 and 5.6 years, respectively. Age was not associated with DRQoL, but shorter diabetes duration was associated with greater disease impact, and girls reported more disease worries than boys. None of the DRQoL scales was related to glycemic control as measured by glycohemoglobin. However, self-rated health status was correlated with glycohemoglobin.

Three recent studies have examined DRQoL using the measure developed by Ingersoll and Marrero (1991). Grey and colleagues¹¹ studied the personal and family correlates of DRQoL in 52 adolescents. These youths had a mean age of 16.1 years, mean diabetes duration of 8.2 years, and were primarily a Caucasian sample of middle to upper socioeconomic (SES) status. Youths who reported greater disease impact and less satisfaction felt that diabetes was more upsetting and diabetes management was more difficult. Similarly, those who reported more disease-related worries felt that diabetes was more upsetting and diabetes management was more difficult. Additionally, more disease-related worries were associated with lower levels of self-efficacy, higher levels of depression, and less family support. Multiple regression analyses revealed that DRQoL was best predicted by depression, and to a lesser extent, difficulty coping with diabetes. Glyco-hemoglobin levels were not associated with any aspect of DRQoL.

Guttmann-Bauman, Flaherty, Strugger, and McEvoy¹² examined the relationship between metabolic control and DRQoL in a sample of 69 youths with a mean age of 15 years and mean disease duration of 7 years. One-third of the study sample was composed of minority youth (His-panic and African American), but the SES of the sample was not reported. Metabolic control was assessed both by glycohemoglobin and by the frequency of acute events such as diabetes-related emergency room visits or hospital admissions. Using a total score from the sum of the three DRQoL scales, significant correlations were obtained between DRQoL and glycohemoglobin, as well as between DRQoL and acute events, such that youths who reported better DRQoL had lower levels of glycohemoglobin and fewer acute events.

Analyses of the three DRQoL scales revealed significant associations of glycohemoglobin with satisfaction and disease impact, but not with worries. Acute events were significantly related to impact and worries. Associations with glycohemoglobin were stronger when the mean glycohemoglobin for the year was used in analyses, rather than just a single measure. The analyses controlled for the effects of age, disease duration, gender, and ethnicity, but unfortunately, these effects on DRQoL were not reported.

It is important to focus on minority patients because they are at increased risk for metabolic control problems.^13-15 Such findings suggest the potential for lower DRQoL in poor and minority youth compared with other social groups. My colleagues and I¹⁶ studied DRQoL in a culturally diverse sample of 96 youths comprising the full range of SES. Participants included 23 Caucasian, 29 African-American, and 44 Hispanic youths. Caucasian youths had significantly higher SES than African-American and Hispanic youths. We used the modified DRQoL measure and examined DRQoL in relation to demographic, clinical, and psychosocial factors.

Our findings showed significantly lower DRQoL (i.e., greater impact and worries) in youths from single-parent families and those with lower SES. African-American youths reported more disease worries than did Caucasian youths, but when SES was controlled for, this effect was attenuated. A number of psychosocial factors were significantly related to DRQoL. Specifically, poorer DRQoL was associated with maladaptive coping, increased diabetes-related stress, more behavior problems, and greater levels of nonsupportive family behavior.

Multivariate analyses identified SES and diabetes-related stress to be the best predictors of DRQoL impact and worry scores. Age, diabetes duration, gender, and glycohemoglobin were unrelated to DRQoL. Compared with the sample reported by Ingersoll and Marrero,¹⁰ our participants reported significantly greater DRQoL impact and less satisfaction, but fewer diabetes-related worries.

In summary, the results of these studies provide some initial evidence regarding the demographic, psychosocial, and clinical correlates of DRQoL in youths.

Worse DRQoL has been observed among youths from single-parent families and lower SES groups. In one study, girls reported more disease-related worries than boys. Poorer DRQoL also appears to be associated with youths' perceptions that diabetes is upsetting, difficult to manage, and stressful, and is related to higher levels of self-reported depression. There is one report of greater DRQoL impact among youths with shorter disease duration, and in only one of the four studies was a relationship observed between DRQoL and metabolic control.

Challenges at diagnosis. Several longitudinal studies suggest that the diagnosis of diabetes may pose considerable challenges for children and families.¹⁸ Kovacs and colleagues¹⁹ found that 36% of newly diagnosed children had diagnosable psychiatric disorders soon after diabetes onset, most commonly adjustment disorders. However, the majority of these cases had resolved within the first year after diagnosis. Jacobson and colleagues²⁰ also investigated children with newly diagnosed diabetes. Evaluations conducted within the first 5 months after diagnosis found only one difference between the group of diabetic children and a medical control group on a variety of behavioral and psychological measures: decreased school-related competence among the children with diabetes. Grey and colleagues²¹ found that mild adjustment problems had dissipated by the end of the first year, but some problems reappeared by the end of the second year.

Ongoing challenges of diabetes. Early studies comparing children with established diabetes to other groups on psychological variables generally did not find significant differences,¹⁷ although one report found boys with later onset of diabetes to have more behavioral problems.²² More recent studies do indicate, however, that children with diabetes, like children with various chronic diseases, may be at increased risk for psychological problems.²³ Blanz, Rensch-Riemann, Fritz-Sigmund, and Schmidt²⁴ found that one-third of their sample of 93 diabetic adolescents had psychiatric disorders (mostly internalizing symptoms), compared with 10% of the subjects in a control group. In a 10-year follow-up study of newly diagnosed youth, Kovacs and colleagues²⁵ found that nearly half of the study sample had a psychiatric diagnosis, the most frequent being major depression, conduct disorder, and generalized anxiety disorder. In another 10-year follow-up study of newly-diagnosed children and adolescents, Jacobson and colleagues²⁶ found that as young adults, participants reported lower self-esteem relative to control participants. Poorer adjustment to diabetes during adolescence may persist into early adulthood.²⁷ There is also evidence that diabetic youths, particularly females, are at increased risk for eating disorders, and that eating disorders are associated with poor metabolic control.^28,29

Psychosocial correlates of regimen adherence and glycemic control. Psychological and family factors have been related to regimen adherence and metabolic control within samples of children to diabetes.¹⁷ Results indicate youths in poor metabolic control have increased anxiety and lower self-concept,³⁰ more peer relationship problems and depression,^31,32 and greater levels of stress.^33-35 Studies of attributional and coping styles suggest that youths in poor metabolic control are more likely to use the learned helplessness style³⁶ and engage in avoidance and wishful thinking in response to stress.³⁷ Inappropriate health beliefs³⁸ and lower self-efficacy³⁹ have also been associated with poor metabolic control.

Families of youths in poor metabolic control have more conflict and financial problems and less cohesion and stability than those of youths in good metabolic control.^30,33,40 Improved levels of metabolic control have been related to better family communication and conflict resolution skill,⁴¹ agreement about family responsibilities and appropriate involvement in diabetes management tasks,^42,43 and more structured and controlling family environments.⁴⁴ Clinical studies of adolescents selected for chronically poor metabolic control have revealed significant family dysfunction in the majority of families.^45,46

There is also evidence that family relations are linked to regimen adherence. Adherence levels of youths have been associated with general and regimen-specific family support^47,48 as well as communication skills⁴⁹ and conflict.⁵⁰ While youths receive instrumental support from their families, they receive considerable emotional support from their friends.⁴⁸

Neuropsychological functioning. A number of studies have examined neuropsychological functioning in children with diabetes. Children with early onset of diabetes and a history of hypoglycemic episodes have been shown to have significantly poorer cognitive performance, particularly for visual-spatial functioning.^51,52 Other studies provide additional data suggesting children who develop diabetes early in life may be at increased risk for later neurocognitive deficits, both verbal and visual-spatial.^53,54 Ryan, Longstreet, and Morrow⁵⁵ found that children with diabetes missed twice as much school as their peers who did not have diabetes, and lower reading achievement was associated with more school absences. One study found diabetic youths were more likely to have learning problems, with such problems more frequent among boys than girls.⁵⁶ Rovet and Alvarez⁵⁷ found poorer attentional functioning and lower verbal intelligence in diabetic children and adolescents, which were related to history of hypoglycemic seizures.

In a neuropsychological study of newly diagnosed children, Kovacs, Goldston, and Iyengar⁵⁸ found that verbal intelligence and school grades were average shortly after diagnosis, but declined significantly over time. This decline in verbal skill was predicted in part by memory dysfunc-tion.⁵⁹ In a recent prospective study of newly diagnosed children, mild neuropsychological deficits were observed 2 years after diagnosis, with reduced speed of information processing and decrements in conceptual reasoning and acquisition of new knowledge.⁶⁰

In summary, research findings indicate that:

Diabetic youths are at risk for adjustment problems during the initial period of adaptation after diagnosis, and to the degree that there are such problems, the risks for continued difficulties related to adjustment and self-esteem appear to be increased.

In particular, diabetic youths, especially girls, seem to have a greater incidence of depression and eating disorders.

Psychosocial problems, including anxiety, depression, poor self-esteem, high levels of stress, and maladaptive coping styles, have all been associated with poorer metabolic control.

In addition, family conflict, poor communication, and lack of appropriate parental involvement and support for diabetes care are also linked with poor metabolic control and regimen adherence.

Studies of neuropsychological functioning and school performance indicate that diabetic youths are also at risk for learning problems.

All of these factors are important aspects of quality of life and indicate the challenges that diabetic youths face in simultaneously facing the expected tasks of development and daily management of diabetes.

Given the fact that research has shown DRQoL is significantly associated with stress related to diabetes management and general levels of depression, interventions to reduce such stress and depression through improvement in coping abilities are clearly indicated. Empirical support for such approaches is already available. For example, stress management interventions for adolescents have resulted in decreased disease-related stress and improved social interaction.^61,62 A recent study of coping skills training showed improvements in DRQoL, diabetes self-efficacy, and coping, as well as in glycemic control, in youths 3 months after beginning intensive insulin therapies.⁶³ It is important to note that these stress management and coping skills programs integrate several behavioral intervention strategies, including modeling, role-play, behavioral rehearsal, self-monitoring, relaxation, and cognitive restructuring.

Because research has shown that youths who do not manage diabetes well are more likely to have psychological problems, family conflict, and less than optimal parental involvement and support for diabetes tasks, interventions must also consider these psychosocial factors. Empirical support for these approaches is also available. For example, interventions to maintain parent-adolescent teamwork in diabetes management resulted in less conflict and nonsupportive behaviors.⁶⁴ Family-based behavioral intervention utilizing communication skills, goal-setting with behavioral contracts, and problem-solving strategies resulted in improvements in adherence and parent-teen relationships.⁶⁵ Interventions to improve these kinds of psychosocial factors are also likely to improve DRQoL.

More studies are needed with the modified DRQoL for youth measure in order to generate norms for disease impact, worries, and satisfaction, considering youths' age, gender, family income, and ethnic status. Additional data on the behavioral, psychosocial, and health correlates of DRQoL are also needed. With such data, it will be possible to determine the levels at which DRQoL can be considered impaired. At that point, the measure will have more clinical utility. Further studies of the psychometric properties of this measure are also needed, including factor analyses.

While research with diabetic youths has begun to examine disease-specific QoL, there are few data currently available utilizing the multidimensional approach to DRQoL, including generic measures of health status, functional status, disease state, and psychological and social functioning. Because children's QoL is affected greatly by their peer and family relationships, as well as their achievement in academic, athletic, and leisure activities (hobbies, clubs, etc.), it is important for new studies to measure these aspects of diabetic children's functioning.

It is important to consider DRQoL as an essential outcome measure in clinical trials.⁶⁷ More intervention research is needed that specifically targets improvements in these generic aspects of QoL (e.g., physical activity levels, peer and family support, depression) and their impact on DRQoL. Future research should clarify the relationship between DRQoL and metabolic control. It may be that DRQoL and metabolic control are related, but not in a linear fashion. It is especially important to determine the effects on QoL of more intensive, DCCT-type interventions.

Further research should develop norms using this measure that can ultimately identify levels at which DRQoL can be considered impaired. In addition, studies should consider more generic measures of psychological and social functioning that are related to QoL.

Because DRQoL is an important outcome measure in its own right, its measurement should be routinely included in clinical practice and in clinical research. Future research needs include the development and refinement of interventions to improve QoL in youths with diabetes.