THE RELATIONSHIP BETWEEN FAMI LY FACTORS AND METABOLIC CONTROL IN PEDIATRIC PATIENTS WITH TYPE 2 DIABETES By LAURA MARIE BIMBO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003
ACKNOWLEDGMENTS The preparation and completion of this project would not have been possible without the guidance and invaluable advice from my mentor, Gary Geffken, who above all else, sought to keep my stress level at a minimum. Additionally, I would like to thank Kenny Gelfand for sharing his statistical expertise and Adam Lewin for his moral support. Finally, I would like to acknowledge my family and friends for their never-ending support and encouragement throughout this process. And I would like to especially thank my grandfather, Carl Bimbo, who always expressed his enthusiasm and support for my academic endeavors. ii
TABLE OF CONTENTS Page ACKNOWLEDGEMENTS................................................................................................ii LIST OF TABLES...............................................................................................................v ABSTRACT.......................................................................................................................vi CHAPTER 1 INTRODUCTION...........................................................................................................1 Type 1 Diabetes............................................................................................................1 Type 2 Diabetes............................................................................................................3 Adherence and Metabolic Control................................................................................5 Impact of Family Factors............................................................................................12 Current Study..............................................................................................................17 2 METHOD......................................................................................................................18 Participants.................................................................................................................18 Procedure....................................................................................................................19 Measures.....................................................................................................................20 Demographic Questionnaire................................................................................20 Type 2 Family Factors Questionnaire.................................................................20 Metabolic Control Measure.................................................................................21 3 RESULTS......................................................................................................................22 Demographic Analyses...............................................................................................22 Exploratory Factor Analysis.......................................................................................22 Type 2 Diabetes Family Factors Questionnaire-Parent Form.............................23 Type 2 Diabetes Family Factors Questionnaire-Child Form..............................25 Item Analysis..............................................................................................................26 4 DISCUSSION................................................................................................................27 Interpretation of the Current Study.............................................................................27 iii
Limitations of the Current Study................................................................................30 Future Research Directions.........................................................................................32 APPENDIX A T2FFQ CHILD..............................................................................................................34 B T2FFQ PARENT..........................................................................................................38 REFERENCES..................................................................................................................42 BIOGRAPHICAL SKETCH.............................................................................................46 iv
LIST OF TABLES Table page 1: Demographic Characteristics of Participants................................................................18 2: Means and Standard Deviations for Age and HgbA1c of Participants.........................19 3: Factor Loadings for T2FFQ-Parent Form.....................................................................24 4: Factor Loadings for T2FFQ-Child Form......................................................................25 v
Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Masters of Science THE RELATIONSHIP BETWEEN FAMILY FACTORS AND METABOLIC CONTROL IN PEDIATRIC PATIENTS WITH TYPE 2 DIABETES By Laura Marie Bimbo May 2003 Chair: Gary R.Geffken, Ph.D. Major Department: Clinical and Health Psychology Type 2 diabetes is a growing epidemic in children and adolescents and is attributed to the dramatic increase in pediatric obesity. Recent medical research has emphasized the importance of improved metabolic control in patients with type 2 diabetes in order to prevent or delay long-term diabetes related complications. Studies focusing on pediatric patients with type 1 diabetes have found that family factors, such as warmth and caring, conflict, and parental support related to diabetes care tasks, are strong correlates and predictors of metabolic control. However, this result has yet to be evaluated in the rapidly increasing pediatric population with type 2 diabetes. The purpose of this study was to develop and evaluate a disease-specific questionnaire to assess the role of family factors in diabetes care tasks and their relationship to metabolic control in pediatric patients with type 2 diabetes. Sixty-two adolescent and caregiver pairs from pediatric endocrinology clinics in Gainesville and Jacksonville, Florida, completed the Type 2 Diabetes Family Factors vi
Questionnaire (T2FFQ). This 30-item questionnaire was developed based on 3 measures of family factors designed for use with children and adolescents with type 1 diabetes. Internal consistency was adequate for both the child and parent forms of the T2FFQ. On the child form of the T2FFQ, an exploratory factor analysis revealed 3 factors: Parental Involvement/Emotional Support, Parent Responsibility Taking, and Parental Nagging, which accounted for 32% of the variance. The T2FFQ parent form also revealed 3 factors: Parental Nagging, Parent Responsibility Taking, and Child Responsibility Taking, which accounted for 39% of the variance. Further analyses revealed that both the child reported factor of Parental Nagging and the total T2FFQ child score were significantly correlated with metabolic control. In addition, two items on the T2FFQ child form were significantly correlated with metabolic control. The parent reported factors and the total parent score on the T2FFQ were not significantly related to metabolic control. These results are similar to those previously found in the type 1 diabetes population and suggest the importance of focusing on family variables such as conflict and parental nagging in order to help pediatric patients with type 2 diabetes achieve better metabolic control. Future research should focus on further validation of the T2FFQ using a larger sample with the goal of developing a useful instrument to be used in medical settings. vii
CHAPTER 1 INTRODUCTION Diabetes mellitus is a chronic disease that affects more than 16 million Americans. Approximately 10% of individuals affected by diabetes have a form of the disease known as type 1, or juvenile diabetes, while the other 90% have type 2 diabetes. Type 2 diabetes has been traditionally referred to as adult onset diabetes. However, in recent years, an alarming increase in the number of adolescents diagnosed with type 2 has occurred. This increase has been attributed to the increase in pediatric obesity (Pinhas-Hamiel et al., 1999). Type 1 Diabetes Type 1 diabetes is thought to be caused by an autoimmune process in which the body attacks itâ€™s own beta cells located in the pancreas. These beta cells are responsible for insulin production and once destroyed, an individual becomes insulin deficient. Insulin is an enzyme that allows the bodyâ€™s cells to effectively utilize energy that is obtained from food. Without insulin, the body has no means of converting food into energy and survival is not possible. Therefore, it is necessary for individuals with type 1 diabetes to administer daily doses of insulin, either by injection or through the use of an insulin pump. In addition, the successful management of type 1 diabetes requires a rather complex regimen, incorporating daily blood glucose testing, a dietary plan, and exercise. The treatment goal of individuals with type 1 diabetes is to integrate each aspect of this regimen in order to mimic the typical functioning of the pancreas, which maintains a normal blood glucose level during fluctuations in diet and physical activity. 1
2 Unfortunately, even through the use of a complex diabetes management regimen it is almost impossible to imitate pancreatic functioning and individuals with type 1 diabetes are prone to both hypoglycemia (low blood glucose) and hyperglycemia (high blood glucose). Incidences of hyperglycemia ultimately lead to complications associated with diabetes such as retinopathy, neuropathy, nephropathy, and cardiac disease while hypoglycemia can lead to mental confusion or more severely, to coma, seizures, and death (American Diabetes Association, 2001). Traditionally, type 1 diabetes has been referred to as juvenile diabetes. The National Institutes of Health (1995) estimates that 120,000 individuals under 20 years of age are affected by type 1 diabetes and 13,171 new cases are diagnosed each year. Although puberty is the most common time of diagnosis, the onset of type 1 diabetes can occur in individuals of all ages (Matsushima et al., 1995). Type 1 diabetes is hypothesized to be caused by an interaction between environmental and genetic factors. Genetic factors are not the sole origin of diabetes, given that the concordance rate for monozygotic twins is only 33 %. In fact, 80-85% of individuals with type 1 diabetes have no other family member with the disease. It is hypothesized that environmental factors such as viral exposure, dietary antigens, and toxins also play a role in the onset of type 1 diabetes (Kaufman, 1997). Interestingly, the prevalence of type 1 diabetes is unequal across racial, ethnic, and geographic groups (National Institutes of Heath, 1995). Caucasians are more likely to be affected by this form of diabetes than are African Americans, Asians, or Native Americans. In addition, the geographical distribution of individuals with type 1 diabetes is uneven. For instance, individuals living in Shanghai, China have only a 0.7% chance in 100,000 of developing
3 the disease, while those living in Allegany County, Pennsylvania, develop the disease at the rate of 18.2 cases per 100,000, and individuals from Finland develop type 1 diabetes at a rater of 35.3 out of 100,000 (Kaufman, 1997; Matsushima et al., 1995). Type 2 Diabetes In contrast to type 1, type 2 diabetes is caused by the bodyâ€™s inability to produce enough insulin or the inability to use insulin effectively. Some individuals with type 2 diabetes produce too much insulin but become resistant to the enzyme. For this reason, the management of type 2 diabetes is somewhat different from that of type 1. Type 2 diabetes can often be controlled by diet and exercise alone, although oral diabetes medications are widely prescribed. However, the long-term effects of oral diabetes medications have not yet been evaluated in adolescent populations. Typically, adolescents with type 2 diabetes are first prescribed diabetes medications such as sulfonylureas, metaformin, acarbose, biguanides, and thiazolindinedione, which serve to stimulate the pancreas to produce insulin or help the body use insulin more effectively (Jones, 1998; Rosenbloom et al., 1999). Eventually, however, a number of individuals with type 2 diabetes are required to take insulin injections in order to manage their condition. Long-term complications associated with type 2 diabetes are the same as those seen in individuals with type 1. However, the onset of these complications often comes more quickly in individuals with type 2 diabetes. This may be due to the gradual onset of type 2 diabetes in contrast to the acute nature of type 1 onset. In addition, it is suspected that only half of the cases of type 2 diabetes are diagnosed, leaving millions of individuals without the proper treatment for this serious condition (National Institutes of Health, 1995).
4 In the past, type 2 diabetes was referred to as adult onset diabetes because it was a disease typically diagnosed after age 40. However, in recent years, this has begun to change and the incidence of this disorder among adolescents has increased by 10 times as documented in Cincinnati, Ohio, clinics between 1982 and 1994. In addition, in 1994, adolescents with type 2 diabetes represented almost 40% of the new cases of diabetes among adolescents in that area (Pinhas-Hamiel et al., 1999). Researchers in other countries have also documented the increasing incidence of type 2 diabetes. In Japan, for example, the incidence of type 2 diabetes in elementary school is similar to that of type 1. The incidence of type 2 diabetes in middle school, however, is almost seven times greater than the incidence of type 1 diabetes. These finding suggest that the increasing incidence of type 2 diabetes is especially pervasive among adolescents (Rosenbloom et al., 1999). In contrast to type 1 diabetes, type 2 is more common in minority populations. Information collected from 1990 to 1992 on the National Healthy Interview Survey showed that type 2 diabetes is more common in African Americans. In addition, the Hispanic Health and Nutrition Examination Survey found that incidences of type 2 diabetes were greater in Mexican and Puerto Rican individuals than in Caucasians or African Americans. Native Americans, however, have the highest rates of type 2 diabetes in the United States and throughout the world (National Institutes of Health, 1995). Gender differences in the prevalence of type 2 diabetes have also been documented. Some researchers have found that as many as 80% of youth with type 2 diabetes are female (Dean, 1998). Similar to type 1 diabetes, this form of the disease is most likely caused by both genetic and environmental factors. Among children and adolescents with type 2 diabetes
5 72-85% have multiple family members with the disease (Pinhas-Hamiel et al., 1999). It is unclear whether this high rate of family history is due to genetic or lifestyle influences. In fact, it was found that the family members of adolescents with type 2 diabetes have very similar lifestyle characteristics such as high fat intake, low rates of physical activity, high rates of binge eating, along with a high prevalence of type 2 diabetes in other family members. Other risk factors for type 2 diabetes are: age, obesity, history of gestational diabetes, impaired glucose tolerance, and lack of physical activity (Rosenbloom et al., 1999). Although type1 and type 2 diabetes are different conditions and demand somewhat different treatment regimens, many daily tasks required of individuals with both forms of the disease are the same. For example, individuals with type 1 and type 2 diabetes should perform daily blood glucose testing, adhere to dietary plans, and participate in a daily exercise routine. Adherence and Metabolic Control Due to the complicated treatment regimens required of individuals with type 1 and type 2 diabetes, researchers have sought to elucidate factors that affect the management of this chronic condition. The goal of this research is to understand how individuals with diabetes can improve their quality of life and long-term health status through a better understanding of the variables affecting diabetes care. Within the literature on diabetes, there is a distinction made between adherence to a treatment regimen and metabolic con-trol. Adherence is behavioral measure of diabetes management. According to Haynes (1979) adherence is â€œthe extent to which a personâ€™s behavior (in terms of taking medica-tions, following diets, or executing lifestyle changes) coincides with medical or health adviceâ€ (p. 2-3). In order to measure adherence to diabetes treatment regimens, individuals are asked to report the frequency at which they engage in adherent behaviors
6 such as the self-monitoring of blood glucose levels or exercise as well as the frequency of nonadherent behaviors such as omitting a dose of insulin. In the past, adherence was often referred to as compliance. Although these terms are somewhat interchangeable, adherence is considered to be a more accurate term that reflects a patientâ€™s responsibility for his or her own diabetes care (Johnson, 1994). In contrast to adherence, metabolic control is a medically based assessment of dia-betes care. Today, glycolysated hemoglobin (%HgbA1c) is considered to be the standard method of measuring metabolic control in individuals with diabetes. This laboratory test calculates the average amount of glucose present in the blood during the previous 6 to 12 weeks. Lower HgbA1c values indicate better or â€œtightâ€ metabolic control while higher values indicate poor control. The American Diabetes Association recommends that patients with diabetes strive to obtain an HgbA1c of less than 7%, which is equivalent to an average blood glucose of 150 mg/dl (ADA, 2001). This value is based upon empirical research showing that lower HgbA1c levels are strongly related to delayed and decreased incidence of diabetes related complications (Diabetes Control and Complications Research Trial Group, 1993; Turner, Cull, & Holman, 1996). Intuitively it would seem that adherence behaviors are related to metabolic control in individuals with diabetes. In other words, it is plausible that an individual who closely adheres to the prescribed medical regimen would achieve better metabolic control than an individual with diabetes who is less adherent. Unfortunately, a 1:1 relationship between adherence and control has not been adequately supported empirically (Glasgow et al., 1987; Johnson, 1992, 1993, 1994). In fact, metabolic control is most likely influenced by several factors in addition to adherence such as duration of diabetes, heredity, and
7 hormonal changes (Johnson, 1992). This issue is confounded by the fact that many researchers use adherence and control almost interchangeably despite the fact that these terms refer to different constructs (Brownlee-Duffeck et al., 1987; Johnson 1992). There are several factors that complicate the problem of relating adherence and metabolic control in individuals with diabetes. Adherence has proven to be difficult to conceptualize. The definition of adherence implies there is an ideal standard that individuals with chronic conditions such as diabetes must follow. But researchers define this standard differently. For example, when measuring adherence, behaviors of individuals with diabetes can be compared to recommendations of physicians, more general recommendations of authorities such as the American Diabetes Association, or even to an individualâ€™s personalized standard (Johnson, 1993). Each of these sources may give conflicting advice on the behaviors that are necessary to manage diabetes and therefore complicate the issue of measuring adherence. Physicians, for example, typically record prescriptions for insulin or oral agents in patient's charts but they rarely document advice pertaining to exercise or diet. To further complicate the issue, when giving advice on topics such as diet and exercise physicians often use language that is vague and too general. Due to these problems, researchers may prefer to provide specific recommendations for daily diabetes management. It is also important that the ideal standard used to assess adherence is a valid one. For example, if adherence is measured according to unsound advice from a physician, the relationship between adherence and control is confounded. In other words, adherence to an inadequate medical regimen would not help unravel the association between adherence and metabolic control (Johnson, 1992).
8 In addition, the issue of inadvertent noncompliance further confounds the relationship between adherence and metabolic control in individuals with diabetes. Some individuals with diabetes may report adherence while actually behaving in ways that conflict with their prescribed medical regimen. These acts of nonadherence may not be purposeful but due to misunderstandings between patients and physicians, inaccurate recall of prescriptions, or even skill deficits (Johnson, 1992). For example, a patient may be instructed to inject a specific amount of insulin at a certain time of day. The patient may report adherence to this recommendation but may make inadvertent technical errors and as a result may inject an incorrect dose of insulin. In other words, when measuring adherence, it is important to obtain an accurate report of patient behaviors. When measuring adherence it is important to understand that certain groups of individuals may have more difficulty completing the self-care tasks required for proper diabetes management. For example, in children and adolescents with diabetes, developmental issues may affect diabetes management. Younger children may not possess the cognitive ability needed to master the complicated medial regimen required of individuals with diabetes. Therefore, parents must often assume responsibility for diabetes related tasks. In addition, adolescents are often described as a unique age group in terms of adherence to diabetes regimens. Adherence in adolescents with diabetes is markedly lower than that of younger children (Johnson et al., 1986). This finding in adolescents may be due to an increased need for independence and peer pressure that individuals of this age often experience. However, this assertion has not been empirically supported (Johnson, 1992). It has also been suggested that adolescences with type 2 diabetes may have even greater problems adhering to the medical regimen required to
9 manage diabetes. Jones (1998) attributes the high rate of nonadherence in this population to the fact that adolescents with type 2 diabetes often do not have physical symptoms related to their adherence behaviors. For example, adolescents with type 2 do not feel sick if they skip insulin injections unlike individuals with type 1 diabetes who rapidly experience the unpleasant symptoms of hypergylcemia. In addition, adolescents with type 2 diabetes taking oral hypoglycemic agents generally do not notice any improvements in their health. For these reasons, adolescents with type 2 diabetes may chose to be nonadherent (Jones, 1998). The measures used to assess adherence in diabetic populations are also somewhat problematic. It has been demonstrated that adherence to one aspect of the medical regimen does not predict adherence to other components of the regimen. For example, a patient may adhere to insulin injections but may fail to follow exercise and dietary recommendations. Yet in the literature, global measures of adherence are often used, which fail to take this finding into account (Johnson, 1993). In addition, reports of adherence are often obtained from different sources and through different methods. In the past, researchers have used physician reports of adherence, parental reports, as well as patient self-reports. Various methods have been used to obtain adherence data such as behavioral observations, the counting of permanent products, self-reports, and the 24-hour recall interview (Johnson, 1992). The lack of consistency in the source of adherence data and the method by which this data is obtained makes it difficult to reliably measure the relationship between adherence behaviors and metabolic control. In recent years, the relationship between adherence, control, and complications arising from diabetes has been empirically supported through a landmark ten-year study
10 known as the Diabetes Control and Complications Trial (DCCT) (Diabetes Control and Complications Research Trial Group, 1993). In this longitudinal study of individuals with type 1 diabetes, a control group performed a typical amount of daily management tasks while the intensive therapy group adhered to a stricter, more involved regimen. The intensive therapy group was required to administer three or more injections daily or use an insulin pump, perform at least four blood glucose tests, and adjust their insulin dosage according to their blood glucose levels. In addition, individuals in the intensive therapy group were required to adhere to standards of diet and exercise. Results of this study showed that individuals in the intensive therapy group experienced a delayed onset of diabetes related complications and a slower progression of retinopathy, eye disease often caused by diabetes. In addition, it was found that HgbA1c results were related to the onset of complications. For example, of patients in â€œgoodâ€ control with HgbA1c results < 6.87%, 90% were free of retinopathy. However, among patients with what is considered â€œpoorâ€ metabolic control (HgbA1c > 9.49%), 57% developed retinopathy. The findings of the DCCT, which show a relationship between metabolic control and reduced or delayed diabetes related complications have led physicians to recommend more intensive therapy for individuals with diabetes in order to minimize the risks of complications. In recent years, the relationship between adherence, control, and complications arising from diabetes has been empirically supported through a landmark ten-year study known as the Diabetes Control and Complications Trial (DCCT) (Diabetes Control and Complications Research Trial Group, 1993). In this longitudinal study of individuals with type 1 diabetes, a control group performed a typical amount of daily management
11 tasks while the intensive therapy group adhered to a stricter, more involved regimen. The intensive therapy group was required to administer three or more injections daily or use an insulin pump, perform at least four blood glucose tests, and adjust their insulin dosage according to their blood glucose levels. In addition, individuals in the intensive therapy group were required to adhere to standards of diet and exercise. Results of this study showed that individuals in the intensive therapy group experienced a delayed onset of diabetes related complications and a slower progression of retinopathy, eye disease often caused by diabetes. In addition, it was found that HgbA1c results were related to the onset of complications. For example, of patients in â€œgoodâ€ control with HgbA1c results less than 6.9%, 90% were free of retinopathy. However, among patients with â€œpoorâ€ metabolic control (HgbA1c > 9.5%), 57% developed retinopathy. The findings of the DCCT, which show a relationship between metabolic control and reduced or delayed diabetes related complications have led physicians to recommend more intensive therapy for individuals with diabetes in order to minimize the risks of complications. More recently, the United Kingdom Prospective Diabetes Study (UKPDS) has found similar results in individuals with type 2 Diabetes (Turner et al., 1996). In this study 5,102 individuals with type 2 Diabetes were followed for an average of 10 years. Much like the DCCT findings, the UKPDS showed a link between intensive therapy and lowered HgbA1c levels. However, in individuals with type 2 diabetes the differences in HgbA1c levels between the intensive therapy group and the conventional therapy group were smaller than those found in individuals with type 1 diabetes (average HgbA1c in the intensive control group was 6.7% compared to an average HgbA1c of 7.5% in the conventional therapy group). Despite the somewhat smaller effect of intensive therapy
12 on HgbA1c levels, the UKPDS provides evidence for a link between more intensive adherence behaviors and tighter metabolic control in individuals with type 2 diabetes (Turner et al., 1996). Although results from both the DCCT and the UKPDS suggest that adherence to more intensive treatment regimens can positively affect metabolic control, many other studies suggest that the link between adherence and control is not so simple. In fact, the relationship between these two outcome measures may be affected by other factors such as family functioning. Especially in child and adolescent populations, interactions within family surrounding diabetes related tasks may significantly impact adherence and metabolic control of pediatric patients with diabetes. Impact of Family Factors In addition to research on adherence and outcome, a great deal of research has investigated the relationship between family variables and outcome in children and adolescents with type 1 diabetes. The link between family functioning and disease related functioning in individuals with diabetes is thought to be bi-directional. In other words, the family environment can either facilitate or hinder the management of diabetes and the daily management of diabetes may in turn affect family functioning by for example, causing additional stress and conflict within the family (Susman-Stillman, Hyson, Anderson, Collins, 1997). The current treatment of this relationship will focus on the effect of the family environment on the management of diabetes. While there is a large body of literature dealing with the family environment and disease-related functioning in individuals with diabetes, it is difficult to synthesize the results of specific studies due to numerous methodological variations. For example, in the literature there is a distinction between adherence to the medical regimen required to
13 manage diabetes, a behavioral measure, and metabolic control, which is a biological assessment of outcome. While some researchers opt to investigate both adherence and metabolic control, others use only one of these measures. Previous research has found little relation between adherence and control in individuals with diabetes, making it difficult to compare research utilizing only one of these measures. In addition, the links between adherence, metabolic control, and family functioning have not been fully elucidated. While adherence has been consistently correlated with family functioning, there is some disagreement about the relationship between metabolic control and family factors. It is thought that metabolic control may be indirectly linked with family variables through adherence behaviors (Miller-Johnson et al., 1994). There is also some disagreement within the literature regarding the types of measures that best illustrate the relationship between family variables and adherence and metabolic control in individuals with diabetes. Several researchers have found that general measures of family functioning (e.g. the Family Assessment Measure or the Family Environment Scale) typically do not strongly correlate with adherence or metabolic control (Liss et al., 1998; Hauser et al., 1990). However, Jacobson and colleagues (1994) found a relationship between metabolic control and mother reports of family cohesion, conflict, and expressiveness. Unlike previous research, this study measured family functioning and metabolic control over a four-year period and therefore may better elucidate the relationship between these variables over time. Other measures that are specifically designed to assess functioning in families of individuals with diabetes may be more able to detect the relationships between adherence, control, and family variables. The Diabetes Family Behavior Scale (DFBS)
14 developed by Waller and colleagues (1986) is a disease-specific questionnaire used to assess family functioning. This measure, completed by the child, is designed to evaluate family support using three scales warmth/caring (e.g. â€œMy parent listens to my ideas about taking care of my diabetes), guidance/control (e.g. â€œMy parent gives my insulin shots), and problem solving questions (â€œWhen there is a problem about the diabetes, we call the doctorâ€). Previous research using the DFBS has found a moderate relationship between metabolic control and the warmth/caring and guidance/control subscales. However, problem solving was not correlated with metabolic control (McKelvey et al., 1993). Another disease-specific measure, the Diabetes Family Behavior Checklist (DFBC) has shown relationships between adherence, metabolic control, and family variables (Schafer, McCaul, & Glasgow, 1986). The DFBC assesses supportive parental behaviors (e.g. â€œHow often does he/she praise you for following your dietâ€) and non-supportive parental behaviors (â€œHow often does he/she criticize you for not exercising regularlyâ€) exhibited by the family and the impact of these behaviors on self-care behaviors of the individual with diabetes. More specifically, research with the DFBC has shown that parental report of non-supportive behaviors was associated with poorer adherence while young child report of non-supportive behaviors was somewhat positively correlated with metabolic control (Schafer et al., 1986). Anderson and colleagues (1990) developed the Diabetes Family Responsibility Questionnaire (DFRQ) a child and parent report measure designed to assess which family member takes responsibility for diabetes related tasks. Research using this measure has found that in families in which the child and parent disagree about who takes responsibility for diabetes tasks, children with diabetes have poorer metabolic control.
15 However, the assertion that disease specific measures of family functioning better predict adherence and metabolic control is not universally accepted. Some research has shown that diabetes specific scales do not allow for improved prediction of adherence behaviors or metabolic control when compared with more general measures of family functioning (Miller-Johnson et al., 1994). In addition, it is important to remember that the family functioning measures discussed here are designed for use with patients with type 1 diabetes. There are currently no measures designed to specifically assess diabetes related family functioning in families of patients with type 2 diabetes. An additional problem in this body of literature is the definition of family variables. This definition is not standardized and often varies greatly among different studies. For instance, some researchers choose to focus on the characteristics of families of individuals with diabetes, such as warmth and caring, discipline, control and guidance, problem solving techniques, and conflict (Liss et al., 1998; Miller-Johnson et al., 1994; Waller et al., 1986). In contrast, other research has focused exclusively on the family member who assumes responsibility for diabetes-related tasks (Anderson et al., 1990). The inconsistent definition of family variables leads to further methodological complications and makes it difficult to integrate the results across studies. Despite the lack of consistency in the literature, some generalizations can be made regarding the relationship between family functioning and adherence and metabolic control in individuals with diabetes. Several studies have found that family conflict surrounding diabetes related tasks is correlated with adherence and control in children and adolescents with diabetes (Hauser et al., 1990; Miller-Johnson et al., 1994). In a similar finding, Anderson and colleagues (1990) showed that disagreement between
16 parents and children related to taking responsibility for diabetes related tasks predicted poorer metabolic control. In other words, families in which parents and children disagreed about who was responsible for diabetes-related tasks, such as blood glucose testing, tended to have higher HgbA1c values, indicating poorer metabolic control. Less empirical support has been found for a relationship between other family characteristics such as warmth, discipline, and cohesion and the outcome measure of adherence and metabolic control in the pediatric type 1 diabetes population (Hauser et al., 1990; Miller-Johnson, 1994). Given the inconclusive nature of research on the link between adherence to treatment regimens and metabolic control in pediatric patients with type 1 diabetes, it is clear that other variables, such as family functioning, may contribute to this complicated relationship. Results from some research studies suggest that family functioning is related to both metabolic control and adherence in pediatric diabetes populations. In fact, some researchers have postulated that metabolic control may be indirectly linked with family variables through adherence behaviors (Miller-Johnson, 1994). Although the role that family variables play in diabetes management continues to be an area of research in the type 1 diabetes population, virtually no research has examined this relationship in the drastically increasing population of pediatric patients with type 2 diabetes. Given that treatment regimens for type 1 diabetes and type 2 diabetes involve many similar components, such as diet, exercise, blood glucose testing, and often insulin injections, the role of family variables may be similar to that previously found in the type 1 literature. In addition, due to the recent UKPDS finding that metabolic control is related to delayed and decreased incidence of diabetes-related complications in the type 2 population,
17 research is needed to elucidate the variables that may contribute to metabolic control (Turner et al., 1997). Current Study The current study has two main objectives. First, to develop and assess the psychometric characteristics of a diabetes-specific questionnaire assessing family functioning related to diabetes care tasks in adolescents with type 2 diabetes, the Type 2 Diabetes Family Factors Questionnaire. And second to evaluate the relationship between these family functioning variables and metabolic control in adolescents with type 2 diabetes.
CHAPTER 2 METHOD Participants Participants were recruited from the Pediatric Endocrinology Clinic at Shands Teaching Hospital in Gainesville, Florida, and the Pediatric Endocrinology Clinic at the Nemours Childrenâ€™s Hospital in Jacksonville, Florida. The Pediatric Endocrinology Clinics in Gainesville and Jacksonville are outpatient facilities serving pediatric patients with type 1 diabetes, type 2 diabetes, and other endocrine disorders. Sixty-two adolescent and caregiver pairs participated in the study; 39 recruited from the Gainesville site and 23 from the Jacksonville site. 68 % of participants were female while 32 % were male. In terms of race, 63 % of participants were African American, 31 % were Caucasian, 3 % were Hispanic, and 3 % indicated â€œotherâ€ as their racial category. In addition, 40 % of participants were from single parent families. Table 1 reports the demographic characteristics of the sample. Table 1: Demographic Characteristics of Participants Demographic Variable Category N % Sex Male 20 32.3 Female 42 67.7 Race Caucasian 19 30.6 African American 39 62.9 Hispanic 2 3.2 Other 2 3.2 Family Composition Two-Parent 37 59.7 Single-Parent 25 40.3 18
19 Table 2: Means and Standard Deviations for Age and HgbA1c of Participants M SD Range Age (years) 15.7 2.1 10.4-18.5 HgbA1c (%) 8.0 2.66 5.0-14.0 The mean age of the sample was 15.7 years and the mean HgbA1c value for the patients was 8.0%, considerably higher than the recommended goal of 7.0%. Approximately 70 % of participants reported that their annual gross income was less than $30,000. Mothers represented 74% of the parents participating, followed by fathers, grandparents, and other caregivers. There were no significant differences between participants from the Gainesville and Jacksonville sites based on age, sex, race, or family composition. However, participants recruited in Jacksonville had a significantly higher mean income than those recruited in the Gainesville clinic. This difference is most likely due to the nature of the populations served at the two clinics. Procedure The investigators approached patients and their parents during their routine clinic visit in the endocrinology clinic and requested their participation in the research study. Patients approached met the following inclusion criteria: aged 8-18, a diagnosis of type 2 diabetes for at least six months, living with and accompanied by their primary caregiver, and no evidence of mental retardation. Of families approached, 97.5% agreed to participate in the study and the remaining patient declined because of illness. Parents of the adolescents signed the informed consent, approved by the Institutional Review Board at the University of Florida and the Nemours Childrenâ€™s Hospital, and assent was obtained from patients.
20 Measures Demographic Questionnaire Parents were asked to complete a brief demographic questionnaire. This questionnaire included questions pertaining to the adolescent (age, sex, race), the parent (age and level of education), and the family (single vs. two-parent family and income). Type 2 Family Factors Questionnaire The Type 2 Diabetes Family Factors Questionnaire (T2FFQ) is a questionnaire designed to assess diabetes specific adherence behaviors and family functioning in adolescents with type 2 diabetes. Both the child and the parent form of the T2FFQ contain 15 core items and 20 pilot items (see Appendix A and B). The 15 core items were developed based on similar measures designed to evaluate family functioning related to diabetes care tasks in adolescents with type 1 diabetes, such as blood glucose testing and dietary plans. These 15 questions were adapted from items most related to adherence behaviors and metabolic control in the type 1 population. Pilot T2FFQ items contained additional questions not as highly correlated with outcome in the type 1 population and additional items concerning exercise and â€œdiabetes medications,â€ a term used to describe any medication used to control diabetes including diabetes pills and insulin. Exercise and diabetes medications are two areas of diabetes management considered particularly relevant for pediatric patients with type 2 diabetes. Participants and their parents were both asked to complete the questionnaire. A five-point Likert scale was used and participants were asked to circle the answer â€œthat best describes how these things go on in your family.â€ Possible answers ranged from â€œneverâ€ to â€œat least once a day.â€ Five pilot questions on the T2FFQ were formatted differently and dealt with
21 responsibility taking for diabetes tasks. However, these items were not used in statistical analyses due to lack of variability in responses. Metabolic Control Measure In order to assess the participantsâ€™ level of metabolic control, glycolysated hemoglobin A 1c values were obtained for each patient (%HgbA1c). An HgbA1c test is performed on adolescents during their regular visits to the Pediatric Endocrinology Clinics in Gainesville and Jacksonville. This assay is considered to be the gold standard for the measurement of metabolic control. HgbA1c was measured using a Bayer DCA 2000 with values ranging from 5.0% to 14.0%.
CHAPTER 3 RESULTS Demographic Analyses T-tests were performed on the demographic variables of race, sex, and family composition but no significant differences in HgbA1c values were detected. A univariate analysis of variance also revealed no significant differences in metabolic control based on family income or age of participants. Exploratory Factor Analysis Items with low response variability were first omitted given that these items failed to detect meaningful differences between individual patients. Three items from both the parent and the child questionnaires were removed from further analyses because over 75% of participants responded in the same way. These items were: â€œMy child feels like he/she has no one to talk to about diabetes,â€ â€œI test my childâ€™s blood sugar for him/her,â€ and â€œIâ€™m embarrassed that my child has diabetes.â€ The remaining 25 items on the T2FFQ-Parent form and 26 items on the T2FFQ-Child form were analyzed using an Exploratory Factor Analysis. Despite the relatively low sample size, an exploratory factor analysis was conducted (further attention to this point is included in the discussion section). The unweighted least squares method was used and factor solutions were rotated orthongonally using a Varimax rotation. Using the Kaiser criterion of retaining factors with eigenvalues over 1, a 9-factor solution emerged for both the child and parent form. However, several factors in this solution contained only one item and many factors could not be meaningfully interpreted. 22
23 In addition, many researchers question the applicability of the Kaiser criterion, given that it is highly influenced by the number of items, leads to the retention of too many factors, and does not measure the reliability of the factors (Cliff, 1988; Zwick & Velicer, 1986). Next, the scree plots were examined given that this method of determining the number of factors to retain is frequently used. However, the scree plots were somewhat ambiguous for both the child and parent forms and supported the retention of 2 to 3 factors for both forms of the T2FFQ. Therefore, in order to maximize the variance accounted for and to retain the largest number of meaningful factors, the data was forced into a 3-factor solution. This solution also made sense empirically given that the T2FFQ was constructed using questions from 3 diabetes specific measures designed for children and adolescents with type 1 diabetes. Although a 2-factor solution was considered, Reise, Waller, and Comrey (2000) have postulated that retaining a larger number of meaningful factors is preferred over retaining too few factors. Furthermore, the 3-factor solution provided for a more straightforward interpretation. Items with factor loadings of at least .4 which did not load on either of the other factors within .2 were retained. Type 2 Diabetes Family Factors Questionnaire-Parent Form A 3-factor solution for the T2FFQ-Parent Form accounted for approximately 39% of the variance in responses and included 17 items. Factor loadings for each of the items can be seen in Table 3. The first factor extracted contained 6 items and accounted for approximately 15% of the variance. This factor was called â€œParental Naggingâ€ given that all the items loading on this factor dealt with the parent bugging, reminding, or arguing with the child about their diabetes related tasks. The second factor also contained 6 items and accounted for 13.5% of the variance. This factor was named â€œParental
24 Table 3: Factor loadings for T2FFQ-Parent Form Factors Parental Nagging Parent Responsibility Taking Child Responsibility Taking Item 1 -.128 .252 .570 Item 4 .609 .251 -.087 Item 6 .692 .189 .050 Item 8 .726 .207 -.139 Item 9 .647 .077 .208 Item 12 -.137 .039 .757 Item 14 .170 .220 .405 Item 15 -.018 .659 -.146 Pilot Item 1 .097 .509 .302 Pilot Item 2 .128 .623 .344 Pilot Item 3 .125 .682 .102 Pilot Item 4 .033 .579 -.139 Pilot Item 8 .092 .591 .202 Pilot Item 9 .058 .056 .583 Pilot Item 10 .006 -.120 .430 Pilot Item 13 .703 -.167 -.182 Pilot Item 15 .737 .264 -.138 Responsibility Takingâ€ given that all the items loading on this factor dealt with the parent taking care of the childâ€™s diabetes and close involvement and supervision of diabetes related tasks. Finally, the third extracted factor contained 5 items and accounted for 11% of the variance. This factor, â€œChild Responsibility Taking,â€ contained items concerning the child assuming a more active role in their own diabetes care with support and praise offered by the parent. Reliability analysis of the T2FFQ-Parent form using Cronbachâ€™s alpha revealed adequate reliability ( = .78) (Cronbach, 1951). Further analyses were conducted in order to determine whether either of the 3 identified parent factors or the measure as a whole were correlated with HgbA1c, the outcome measure of metabolic control. These analyses indicated that neither the individual factors scores nor the overall score on the T2FFQ-Parent form were significantly correlated with the outcome measure.
25 Type 2 Diabetes Family Factors Questionnaire-Child Form On the child form, the 3-factor solution accounted for 32% of the variance in responses and included 17 items. See Table 4 for factor loadings of each item. The first factor that emerged included 8 items and accounted for 14.5% of the variance. This factor, â€œParental Involvement/Emotional Support,â€ contained items dealing with the parent providing encouragement and actively participating in the childâ€™s diabetes regimens. The second child factor contained 5 items and accounted for 9.6% of the variance. This factor was termed â€œParental Responsibility Takingâ€ and similar to the parent factor, contained items dealing with the parent taking a more responsible and directive role in the diabetes care tasks. Finally, the third child factor contained 4 items and accounted for 8.2% of the variance. This factor called â€œParental Naggingâ€ contained items that dealt with the parent bugging, arguing, or becoming angry with the child about their diabetes related tasks. Reliability analysis of the T2FFQ-Child form using Cronbachâ€™s alpha revealed adequate reliability ( = .78) (Cronbach, 1951). Table 4: Factor loadings for T2FFQ-Child Form Factors Parental Involvement/ Emotional Support Parental Responsibility Taking Parental Nagging Item 1 .483 -.058 .116 Item 2 .217 .547 .164 Item 3 -.439 -.092 .108 Item 5 .704 .194 -.008 Item 6 .081 .539 .193 Item 8 -.051 .183 .618 Item 9 .145 -.051 .682 Item 11 .420 -.084 .626 Item 12 .157 .496 -.164 Item 14 .581 .229 -.022 Item 15 .136 .571 .012 Pilot Item 3 .426 -.026 .162 Pilot Item 4 .112 .495 .025 Pilot Item 5 .436 .166 .098 Pilot Item 6 .749 .083 .206 Pilot Item 9 .720 .299 .152 Pilot Item 13 -.024 .198 .534
26 Further analyses were also conducted using the T2FFQ-Child form to elucidate the relationship between the factor scores and the total scale score and metabolic control. Although the â€œParental Involvement/Emotional Supportâ€ and â€œParent Responsibility Takingâ€ factors failed to be significantly correlated with HgbA1c results, the â€œParental Naggingâ€ factor was significantly correlated with this outcome measure on the T2FFQ-Child form (r = .277, p < .03). In addition, the total score on the T2FFQ-Child form was significantly correlated with HgbA1c (r = .277, p < .04). Item Analysis An additional analysis was conducted to determine which individual items were most highly correlated with the HgbA1c outcome measure of metabolic control. This analysis revealed that two items on the child form were significantly correlated with HgbA1c values: â€œMy parent reminds me to take my diabetes medicineâ€ (r = .36, p < .004) and â€œMy parent watches while I check my blood sugarâ€ (r = .273, p < .035). In addition, the correlation between HgbA1c and one additional item on the child form and an item on the parent form was approaching significance: â€œMy parent bugs me about exercisingâ€ (r = 320, p < .012) and â€œOur family plans activities to fit in with my childâ€™s diabetes related tasks (like taking pills or checking his/her blood sugar)â€ (r = .328, p < .01).
CHAPTER 4 DISCUSSION Interpretation of the Current Study Type 2 diabetes is a growing epidemic in pediatric populations and recent medical research has shown that improved metabolic control, as measured by lowered HgbA1c values, is strongly predictive of decreased incidence of diabetes related complications such as nephropathy and retinopathy (Turner et al., 1996). Although extensive research has been conducted to elucidate the correlates of metabolic control in the pediatric type 1 diabetes population, research focusing on this issue in the type 2 population has been neglected. Family factors, such as responsibility taking, conflict, and warmth and caring related to diabetes tasks, are related to adherence behaviors and metabolic control in pediatric patients with type 1 diabetes. Given the similarities in the course and treatment regimens of both type 1 and type 2 diabetes, it is plausible that these family factors are also important correlates of adherence and metabolic control in the pediatric type 2 diabetes population. The purpose of the current study was to develop a questionnaire to assess the role of family variables related to diabetes care tasks in a pediatric type 2 diabetes sample. The Type 2 Family Factors Questionnaire was developed using questions adapted from the disease specific type 1 diabetes questionnaires, the DFBC, the DFBS, and the DFRQ in addition to newly developed questions dealing with diet and exercise, two crucial components of the type 2 diabetes treatment regimen. The results of the Exploratory Factor Analysis indicated that family variables that emerge in the pediatric type 2 27
28 diabetes population are very similar to those found in the pediatric type 1 diabetes population. These family variables, parental nagging, parental involvement/emotional support, and responsibility taking have all been previously identified in the type 1 literature and this finding is now also supported in a sample of adolescents with type 2 diabetes. This finding suggests that there are some similarities in family functioning related to disease management for pediatric patients with type 1 and type 2 diabetes. Interestingly, different factors emerged on the parent and child forms of the T2FFQ. Parent reports of family variables included factors related to Parental Nagging, Parental Responsibility Taking, and Child Responsibility Taking while child reports of these variables resulted in a factor solution containing the factors Parental Involvement/Emotional Support, Parental Responsibility Taking, and Parental Nagging. This finding suggests that for parents, a relevant underlying concept is whether the child or the parent takes responsibility for the diabetes related tasks. In addition, parents find the need to nag and remind their child to perform diabetes related tasks a very salient behavior. However, for adolescents with type 2 diabetes, parental involvement and emotional support seems to be a more significant family variable. These results suggest that while parents are most focused on the practical aspects of diabetes care (i.e. making sure diabetes related tasks are performed), adolescents find other family factors important such as their parentsâ€™ perceived investment in their diabetes care and the degree to which their parents are open to discussing diabetes related issues. This finding has practical implications for diabetes care and may help parents and adolescents better understand each otherâ€™s perspective. For instance, a parent who is solely focused on performing
29 diabetes related tasks might not be aware that their child feels that emotional support is a more important part of their relationship related to diabetes care. Another intriguing finding is that one child reported family factor along with the total child score on the T2FFQ was significantly related to metabolic control, as measured by HbA1c. This finding illustrates that the childâ€™s perception of family variables is more related to metabolic control than parentâ€™s perceptions of these family variables related to diabetes care. More specifically, an adolescent with type 2 diabetes who reports that his or her parent nags about their diabetes care, tends to have a more elevated HgbA1c. Given that this finding is only correlational in nature, it is unclear whether having a higher HgbA1c causes parents to nag their children more or whether adolescents with type 2 diabetes who perceive their parents as nagging them subsequently develop poorer metabolic control. Furthermore, it suggests that families in which there is more conflict (e.g. nagging, arguing) surrounding diabetes related tasks, children appear to have poorer metabolic control. In previous literature, Patterson (1974) described a coercive cycle seen in families of aggressive boys in which one family memberâ€™s aggression elicits anotherâ€™s and so on, causing the aggression to continue and escalate. The current finding that there is more conflict in families of patients with poorer metabolic control may also suggest a coercive cycle in these families whereby the parent nags and argues with the child about their diabetes care. In turn, the child may argue with the parent and the cycle continues, with escalating anger and negative interactions subsequently causing a poorer outcome for the child with diabetes, in terms of poorer metabolic control. This interpretation of this finding has practical implications for the medical care provided for pediatric patients with
30 type 2 diabetes. Perhaps clinicians should direct more focus on fostering a positive, supportive relationship between parents and adolescents with type 2 diabetes rather than focusing solely on who performs diabetes related tasks. This finding also implies the need for integrated psychological services within the medical setting for families of adolescents with type 2 diabetes. Physicians, nurses, and other medical professionals have extensive knowledge of the medical components of diabetes management; however, psychologists have an extensive knowledge of family dynamics and can assist families in reducing conflict surrounding diabetes related tasks in order to break the coercive cycle. Given the results of this study, this interdisciplinary approach to diabetes management, with medical and mental health professionals working as a team, may prove more beneficial to adolescents with type 2 diabetes by helping to foster an improved relationship with their parents as well as improved metabolic control. Limitations of the Current Study The major limitation of the current study is the relatively small sample sized used to conduct the exploratory factor analysis. Conventional rules of thumb regarding appropriate sample sizes for a factor analytical study vary widely. In the past, these arbitrary rules were based on N:p ratios (ratio of participants to items) with researcher suggesting ratios ranging from 3:1, 5:1 and up to 10:1 (Cattell, 1978; Gorsuch, 1988). However, more recent research indicates that these arbitrary rules are not useful and sample size should instead be based on certain characteristics of the sample, rather than being equivalent across different types of samples. More specifically, communalities (i.e. the amount of variance in each item accounted for by the factor solution) and overdetermination of factors (i.e. p:r; the ratio of the number of items to the number of factors) have been sited as characteristics that should be considered when evaluating the
31 necessary sample size to conduct a factor analysis (MacCallum, Widaman, Zhang, & Hong, 1999). MacCallum and colleagues (1999) conducted a Monte Carlo study to evaluate the effect of population communalities, factor overdetermination, and sample size on the quality of factor solutions extracted. In this study, researchers had 3 levels of communalities (high: mean = .7, wide: mean = .5, with values ranging from .2 to .8, low: mean = .3), 3 levels of factor determination (p:r ratios = 10:3, 20:3, and 20:7), and four levels of sample size: (60, 100, 200, 400). Results from this study suggest that sample size has less impact on the quality of factor solutions in samples in which communalities are high and factors are highly overdetermined (i.e. a p:r ratio of 20:3 or 10:3). In fact, MacCallum and colleagues (1999) found that conducting a factor analysis using a sample size of only 60 produced â€œconvergent and admissibleâ€ solutions 100% of the time when communalities were high and there was a p:r ratio of 20:3 or 10:3. Furthermore, with a sample size of 60, overdetermined factors, and wide communality, an acceptable solutions was also produced in 100% of the cases. However, with a sample size of 60, overdetermined factors, and low communalities, admissible solutions were produced in only 87% of the cases. In summary, MacCallum and colleaguesâ€™ (1999) findings suggest that synopsis admissible solutions can be obtained even with a sample size of 60 when communalities are high and factors are highly overdetermined. In the current study with 62 participants, communalities were high for the T2FFQ parent form (mean = .7) and relatively high for the child questionnaire (mean = .6). In addition, the factors were highly overdetermined with a p:r ratio of 17:3 for both the parent and child questionnaires. Given these results and the findings of MacCallum and
32 colleagues (1999), it is realistic to assume that the factor solution obtained is of high quality. However, the addition of additional participants would provide for more compelling results. Other limitations of the current study include the correlational nature of the relationships between metabolic control and family variables, the sample of convenience, and the biased nature of self-report data. In addition, the current study used metabolic control as an outcome measure and did not include a behavioral measure of patient adherence to physician recommendations. Although the inclusion of a behavioral adherence measure might strengthen these findings, previous research has been inconclusive about the relationship between adherence and metabolic control in the pediatric diabetes population. Given that medical research has shown that metabolic control, not a behavioral measure of adherence, is related to decreased diabetes related complications, the current study used HgbA1c as an outcome measure. Future Research Directions Future research should focus on the replication of these findings with the goal of using the Type 2 Family Factors Questionnaire in a medical setting. To accomplish this goal, further psychometric testing of the questionnaire should be conducted. More specifically, data on test-retest reliability would be helpful. In addition, additional items should be constructed that tap into the same constructs as items found to correlate with metabolic control. More items that correlate with metabolic control will help to improve the practical validity of the T2FFQ. Furthermore, it would be useful to examine the predictive power of the T2FFQ. For example, if the T2FFQ can predict patients at risk for poor metabolic control, this information could be used to target those families with psychological interventions aimed at improving family functioning.
33 Another intriguing avenue of research would be collecting information about type of treatment the patient is prescribed (e.g. oral diabetes medication, insulin injections, or diet and exercise alone). Given the wide range of treatment options used with pediatric patients with type 2 diabetes, there may be differences in the self-report of family factors between patients prescribed insulin injections compared to those prescribed diet and exercise. For example, patients who inject insulin may report more conflict surrounding diabetes tasks than those who manage their condition with less complex regimens such as diet and exercise alone.
APPENDIX A T2FFQ CHILD We would like to know how often one of your parents does each of the following things that have to do with your diabetes. There are no right or wrong answers and your parent will not see your answers. Circle the answer that best describes how much these things go on in your family. Who is the parent you will be answering these questions about? For example, is he/she your mother, father, step-parent, grandparent, or another type of parent? ____________ Never Once A Once A Several Times At Least Once Month Week A Week A Day 1. My parent praises me 1 2 3 4 5 for taking care of my diabetes. 2. My parent chooses 1 2 3 4 5 what I eat for meals and snacks 3. My family eats foods 1 2 3 4 5 that arenâ€™t a part of my diet in front of me. 4. My parent reminds 1 2 3 4 5 me to check my blood sugar. 5. My parent listens 1 2 3 4 5 when I talk about my diabetes. 6. My parent reminds me 1 2 3 4 5 to take my diabetes medicine. 34
35 Never Once A Once A Several Times At Least Once Month Week A Week A Day 7. I feel like I have no 1 2 3 4 5 one in my family to talk to about diabetes. 8. My parent argues with 1 2 3 4 5 me about how well Iâ€™m taking care of my diabetes. 9. My parent bugs me 1 2 3 4 5 about exercising. 10. I take responsibility for 1 2 3 4 5 my diabetes care. 11. My parent encourages 1 2 3 4 5 me to get some exercise. 12. My parent takes 1 2 3 4 5 responsibility for taking care of my diabetes. 13. My parent gives me my 1 2 3 4 5 diabetes medicine. 14. My parent helps me 1 2 3 4 5 choose foods that are a part of my diet. 15. My parent writes down 1 2 3 4 5 my blood sugar tests for me.
36 Pilot Questions Child Form In order to better understand how families deal with type 2 diabetes, we would like to see which issues are most important to you and your parents. The following questions are trial questions that will help us improve our questionnaire for kids with type 2 diabetes and their parents. Again, there are no right or wrong answers and your parent will not see your answers. Please circle the best answer that describes how much these things go on in your family. Who is the parent you will be answering these questions about? For example, is he/she your mother, father, step-parent, grandparent, or another type of parent? _____________ Never Once A Once A Several Times At Least Once Month Week A Week A Day 1. My parent exercises 1 2 3 4 5 with me. 2. My parent watches while 1 2 3 4 5 I check my blood sugar. 3. My parent gives me a 1 2 3 4 5 reward for taking care of my diabetes. 4. My parent takes care 1 2 3 4 5 of my diabetes for me. 5. My parent reads books, 1 2 3 4 5 magazines, or other information about diabetes. 6. My parent talks with 1 2 3 4 5 me about my diabetes. 7. My parent tests my 1 2 3 4 5 blood sugar for me. 8. When we go out to eat, 1 2 3 4 5 my parent helps me choose foods that are a part of my diet.
37 Never Once A Once A Several Times At Least Once Month Week A Week A Day 9. My parent pays attention 1 2 3 4 5 to how well I take care of my diabetes. 10. My parent wants me to 1 2 3 4 5 do things like other kids. 11. My parent thinks that I 1 2 3 4 5 should check my own blood sugar. 12. My parent is embarrassed 1 2 3 4 5 that I have diabetes. 13. My parent gets mad when 1 2 3 4 5 I donâ€™t take care of my diabetes. 14. My family plans activities 1 2 3 4 5 to fit in with my diabetes tasks (like taking pills or checking my blood sugar). 15. My parent bugs me about 1 2 3 4 5 not following my diet. We would like to know who takes responsibility for the following tasks related to your diabetes care. For each task, please indicate whether you take responsibility, your parent takes responsibility, or both of you share responsibility. Who takes responsibility? Me My Both Parent Of Us 16. Explaining your diabetes to friends and relatives. 1 2 3 17. Carrying a snack in case of a low blood sugar. 1 2 3 18. Remembering clinic appointments. 1 2 3 19. Watching for signs of low blood sugar. 1 2 3 20. Talking to teachers about your diabetes. 1 2 3
APPENDIX B T2FFQ PARENT We would like to know how often you do each of the following things that have to do with your childâ€™s diabetes. There are no right or wrong answers and your child will not see your answers. Circle the answer that best describes how much these things go on in your family. What is your relationship to the child you will be answering these questions about? For example, is you his/her mother, father, step-parent, grandparent, or another type of parent? ____________ Never Once A Once A Several Times At Least Once Month Week A Week A Day 1. I praises my child 1 2 3 4 5 for taking care of his/ her diabetes. 2. I choose what my child 1 2 3 4 5 eats for meals and snacks. 3. Our family eats foods 1 2 3 4 5 that arenâ€™t a part of my childâ€™s diet in front of him/her. 4. I remind my child to 1 2 3 4 5 check his/her blood sugar. 5. I listen when my child 1 2 3 4 5 talks about his/her diabetes. 6. I remind my child to 1 2 3 4 5 take his/her diabetes medicine. 38
39 Never Once A Once A Several Times At Least Once Month Week A Week A Day 7. My child feels like he/ 1 2 3 4 5 she has no one in my family to talk to about diabetes. 8. I argue with my child 1 2 3 4 5 about how well he/she is taking care of his/her diabetes. 9. I bugs my child 1 2 3 4 5 about exercising. 10. My child takes 1 2 3 4 5 responsibility for his/ her own diabetes care. 11. I encourage my child 1 2 3 4 5 to get some exercise. 12. I take responsibility 1 2 3 4 5 for taking care of my childâ€™s diabetes. 13. I give my child his/her 1 2 3 4 5 diabetes medicine. 14. I help my child choose 1 2 3 4 5 foods that are a part of his/her diet. 15. I write down my childâ€™s 1 2 3 4 5 blood sugar tests for him/her.
40 Pilot Questions Child Form In order to better understand how families deal with type 2 diabetes, we would like to see which issues are most important to you and your child. The following questions are trial questions that will help us improve our questionnaire for kids with type 2 diabetes and their parents. Again, there are no right or wrong answers and your child will not see your answers. Please circle the best answer that describes how much these things go on in your family. What is your relationship to the child you will be answering these questions about? For example, are you his/her mother, father, step-parent, grandparent, or another type of parent? _____________ Never Once A Once A Several Times At Least Once Month Week A Week A Day 1. I exercise with my 1 2 3 4 5 child. 2. I watch while my child 1 2 3 4 5 checks his/her blood sugar. 3. I give my child a 1 2 3 4 5 reward for taking care of his/her diabetes. 4. I take care of my childâ€™s 1 2 3 4 5 diabetes for him/her. 5. I read books, 1 2 3 4 5 magazines, or other information about diabetes. 6. I talk with my child 1 2 3 4 5 about his/her diabetes. 7. I test my childâ€™s 1 2 3 4 5 blood sugar for him/her. 8. When we go out to eat, 1 2 3 4 5 I help my child choose foods that are a part of his/her diet.
41 Never Once A Once A Several Times At Least Once Month Week A Week A Day 9. I pay attention to how 1 2 3 4 5 well my child takes care of his/her diabetes. 10. I want my child to 1 2 3 4 5 do things like other kids. 11. I think that my child 1 2 3 4 5 should check his/her own blood sugar. 12. I am embarrassed that 1 2 3 4 5 my child has diabetes. 13. I get mad when my child 1 2 3 4 5 doesnâ€™t take care of his/her diabetes. 14. Our family plans activities 1 2 3 4 5 to fit in with my childâ€™s diabetes tasks (like taking pills or checking my blood sugar). 15. I bug my child about not 1 2 3 4 5 following his/her diet. We would like to know who takes responsibility for the following tasks related to your childâ€™s diabetes care. For each task, please indicate whether you take responsibility, your child takes responsibility, or both of you share responsibility. Who takes responsibility? Me My Both Child Of Us 16. Explaining diabetes to friends and relatives. 1 2 3 17. Carrying a snack in case of a low blood sugar. 1 2 3 18. Remembering clinic appointments. 1 2 3 19. Watching for signs of low blood sugar. 1 2 3 20. Talking to teachers about diabetes. 1 2 3
REFERENCES American Diabetes Association. (2001). Standards of medical care for patients with diabetes mellitus. Diabetes Care , 24, S33-S43. Anderson, B.A., Auslander, W.F., Jung, K.C., & Miller, J.P. (1990). Assessing family sharing of diabetes responsibilities. Journal of Pediatric Psychology, 15 , 477-492. Brownlee-Duffeck, M., Peterson, L., Simonds, J.F., Goldstein, D., Kilo, C., & Hoette, S. (1987). The role of health beliefs in the regimen adherence and metabolic control of adolescents and adults with diabetes mellitus. Journal of Consulting and Clinical Psychology, 55 , 139-144. Cattell, R.B. (1978). The scientific use of factor analysis . New York: Plenum. Cliff, N. (1988). The eigenvalues-greater-than-one rule and the reliability of components. Psychological Bulletin, 103 , p 276-279. Dean, H. (1998). Diagnostic criteria for non-insulin dependent diabetes in youth. Clinical Pediatrics , 67-71. Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The New England Journal of Medicine, 329 , 977-986. Glasgow, R.E., McCaul, K.D., & Schafer, L.C. (1987). Self-care behaviors and glycemic control in type 1 diabetes. Journal of Chronic Diseases, 40 , 399-412. Gorsuch, R.L. (1988). Exploratory factor analysis. In J.R. Nesselroade & R.B. Cattell (Eds.), Handbook of multivariate experimental psychology: Perspectives on individual differences (2 nd ed.). New York: Plenum Press. Hauser, S. T., Jacobson, A. M., Lavori, P., Wolfsdorf, J. I., Herskowitz, R. D., Milley, J. E., Bliss, R., Wertlieb, D., & Stein, J. (1990). Adherence among children and adolescents with insulin-dependent diabetes mellitus over a four-year longitudinal follow-up: Immediate and long-term linkages with the family milieu. Journal of Pediatric Psychology, 15, 527-542. 42
43 Haynes, R.B. (1979). Introduction. In R.B. Haynes, D.W. Taylor, & D.L. Sackett (Eds.), Compliance in Health Care (pp. 1-10). Baltimore: Johns Hopkins University Press. Jacobson, A.M., Hauser, S.T., Lavori, P., Willett, J.B., Cole, C.F., Wolfsdorf, J.I., Dumont, R.H., & Wertlieb, D. (1994). Family environment and glycemic control: A four-year longitudinal follow-up: The influence of patient coping and adjustment. Journal of Pediatric Psychology, 15 , 511-526. Johnson, S.B. (1992). Methodological issues in diabetes research. Diabetes Care, 15 , 1658-1667. Johnson, S.B. (1993). Chronic diseases of childhood: Assessing compliance with complex medical regimens. In N.A. Krasnegor, L. Epstein, S.B. Johnson, & S.J. Yaffer (Eds.), Developmental aspects of health compliance behavior (pp.77-89). Hillsdale, NJ: Lawrence Erlbaum Associates. Johnson, S.B. (1994). Health behavior and health status: Concepts, methods, and applications. Journal of Pediatric Psychology, 19 , 199-141. Johnson, S.B., Silverstein J.H., Rosenbloom A., Carter R., Cunningham W. (1986). Assessing daily management in childhood diabetes. Health Psychology, 5 , 545-564. Jones, K.L. (1998). Non-insulin depedent diabetes in children and adolescents: The therapeutic challenge. Clinical Pediatrics , 103-110. Kaufman, F.R. (1997). Diabetes mellitus. Pediatrics in Review, 18 , 383-392. Kovacs, M., Goldston, D., Obrosky, S., & Iyengar, S. (1992). Prevalence and predictors of pervasive noncompliance with medical treatment among youths with insulin-dependent diabetes mellitus. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 1112-1119. Liss, D. S., Waller, D. A., Kennard, B. D., McIntire, D., Capra, P., & Stephens, J. (1998). Psychiatric illness and family support in children and adolescents with diabetic ketoacidosis: A controlled study. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 536-544. MacCallum, R.C., Widaman, K.F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological Methods, 4 , 84-99. Matsushima M., Tajima N., LaPorte R.E., Orchard T.J., Tull E.S., Gower I.F., & Kitagawa T. (1995). Markedly increased renal disease mortality and incidence of renal replacement therapy among IDDM patients in Japan in contrast to Allegheny County, Pennsylvania, USA. Diabetes Epidemiology Research International (DERI) U.S.-Japan Mortality Study Group. Diabetologia, 38 , 236-43.
44 McKelvey, J., Waller, D. A., North, A. J., Marks, J. F., Schreiner, B., Travis, L. B., & Murphy, J. N. , III. (1993). Reliability and validity of the diabetes family behavior scale. The Diabetes Educator, 19, 125-132. Miller-Johnson S., Emery, R.E., Marvin, R.S., Clarke, W., Lovinger, R., & Martin, M. (1994). Parent-child relationships and the management of insulin-dependent diabetes mellitus. Journal of Consulting and Clinical Psychology, 62 , 603-610. Patterson, G.R. (1974). The aggressive child: Victim and architect of a coercive system. In E. J. Mash, L. A. Hamerlynck, & L. C. Handy (Eds.), Behavior Modification and Families (pp. 267-316). New York: Brunner/ Mazel. Pinhas-Hamiel, O., Standiford, D., Hamiel, D., Dolan, L. M., Cohen, R., & Zeitler, P. S. (1999). The type 2 family: a setting for development and treatment of adolescent type 2 diabetes mellitus. Archives of Pediatric and Adolescent Medicine, 153, 1063-1067. Reise, S.P., Waller, N.G., & Comrey, A.L. (2000). Factor analysis and scale revision. Psychological Assessment, 12 , 287-297. Rosenbloom, A. L., Young, R. S., Joe, J. R., & Winter, W. E. (1999). Emerging epidemic of type 2 diabetes in youth. Diabetes Care, 22, 345-354. Schafer, L. C., McCaul, K. D., & Glasgow, R. E. (1986). Supportive and nonsupportive family behaviors: Relationships to adherence and metabolic control in persons with type 1 diabetes. Diabetes Care, 9, 179-185. Susman-Stillman, A., Hyson, D.M., Anderson, F.S., Collins, W.A. (1997). Adolescent psychosocial development and adherence to treatment for insulin-dependent diabetes mellitus. In J.A. McNamara & C. Trotman (Eds.), Creating the compliant patient. Craniofacial growth series (Vol. 33, pp. 72-101). Ann Arbor, MI: Center for Human Growth and Development. Thompson, S.J., Auslander, W. F., White, N.H. (2001). Comparison of single-mother and two-parent families on metabolic control of children with diabetes. Diabetes Care, 24 (2), 234-8. Turner, R., Cull, C., & Holman, R. (1996). United Kingdom prospective diabetes study 17: A 9-year update of a randomized, controlled trial on the effect of improved metabolic control on complications in non-insulin-dependent diabetes mellitus. Annals of Internal Medicine, 124, 136-145. Waller, D. A., Chipman, J. J., Hardy, B. W., Hightower, M. S., North, A. J., Williams, S. B., & Babick, A. J. (1986). Measuring diabetes-specific family support and its relation to metabolic control: A preliminary report. Journal of the American Academy of Child Psychiatry, 25, 415-418.
45 Zwick, W.R., & Velicer, W.F. (1986). Comparison of five rules for determining the number of components to retain. Psychological Bulletin, 99 , 432-442.
BIOGRAPHICAL SKETCH Laura M. Bimbo is a native of Asheboro, North Carolina, where she graduated from Asheboro High School in 1997 as a North Carolina Scholar and a Distinguished Graduate. She subsequently attended the University of North Carolina at Chapel Hill where she majored in psychology. While at UNC-CH, Laura participated in research involving risk and resilience for adolescent substance use with Dr. Andrea Hussong. She also participated in extracurricular activities such as providing respite care for children with autism, participating in the Smoke-Free Kids program, and serving as a summer camp counselor at a camp for children and adults with autism and a camp for children with diabetes. In addition, Laura conducted her senior honors thesis entitled â€œThe Effect of Siblings on the Symbolic Play of Children with Autismâ€ under the guidance of Dr. Gary Mesibov, Director of Division TEACCH. In 2001, Laura graduated from UNC-CH with a Bachelor of Arts degree in psychology with highest honors and highest distinction. She currently resides in Gainesville, Florida, where she is pursuing her doctorate of philosophy in clinical and health psychology with emphasis in pediatric psychology at the University of Florida. 46