Meredith Taylor (2000)

*Note: This research represents work from both Psych 205 [now 315] and an Honors Summer Research Fellowship (Summer 2000)

ABSTRACT

This study examines the relationship between depression, disordered attitudes toward food and eating in a population of college-aged women with type I diabetes mellitus as compared with non-diabetic age-matched peers. Participants responded to the Beck’s Depression Indicator, the Eating Attitudes Test, and a health survey designed specifically for this study. Results indicate significantly elevated levels of both depression and disordered eating attitudes. The results from a small, exploratory study involving college-aged males, which indicate similar trends are also discussed. This project examines the rates of depression and eating disorders in college-aged participants, and regards the current, rigorous treatment of the illness as a possible cause. Previous studies, as well as the implications and limitations of this study are discussed.

INTRODUCTION

Of the approximately 16 million Americans with diabetes, it was estimated in 1999 that 500,000 – 1 million persons had a diagnosis of Type I, or Insulin-Dependent Diabetes (American Diabetes Association, p. 9). Insulin-Dependent Diabetes Mellitus (IDDM) is an illness often diagnosed in young people, hence the misnomer, "juvenile diabetes" (although a person of any age can be diagnosed with this disease). A stressor, such as a virus or exposure to certain chemicals typically triggers the onset of IDDM. The body’s immune system then turns on the insulin-producing beta cells of the pancreas. This autoimmune response either destroys the beta cells completely or leaves them unable to function properly. As this occurs, there is a period marked by a considerable weight loss, extreme thirst, frequent urination and increased hunger in the undiagnosed diabetic.

Diagnosis typically occurs during preadolescence and brings with it abrupt, often unwanted, prescribed lifestyle changes. The newly diagnosed diabetic is confronted with a restrictive diet, daily injections of medication, and a lifetime of schedules and doctors’ appointments. It may be this initial period of adjustment, or the subsequent lifestyle changes that trigger psychological struggles in many adolescents with diabetes.

Diabetes Mellitus is a disorder of the endocrine system involving the production and/or absorption of insulin (American Diabetes Association, 1999). Insulin is a metabolic hormone, produced in the pancreas, which allows glucose in the blood stream, present after food digestion, to be used by the bodily organs for nourishment. Without a sufficient amount of insulin, the levels of glucose in the blood can become extraordinarily elevated, and may reach toxic levels. There are, in all, three types of Diabetes Mellitus: Type I, or Insulin-Dependent Diabetes Mellitus (IDDM), Type II, or Non-Insulin-Dependent Diabetes Mellitus (NIDDM), and Gestational Diabetes. In each type of diabetes, the body either fails to produce sufficient insulin, or fails to use the insulin efficiently. This research project will focus on those persons with IDDM.

Depression and eating disorders in diabetics have been widely studied from both medical and psychological standpoints. Depression has been associated with a lack of adjustment to the chronic illness and poor diabetic control

Lustman, Griffith, Freedland, and Clouse (1997) have shown that depression is three to four times more prevalent in diabetics than in the general population. Lustman’s team followed up on participants from a previous study and examined the effects of an experimental drug on the regulation of blood glucose levels in diabetics as well its effects on depression. A recurrent problem with both eating disorders and depression in those with diabetes is that many symptoms of poor diabetic control (fatigue, weakness, sleep disturbances and sexual dysfunction) are also reflective of depression (Lustman, et. al, 1997). There may be a tendency among medical personnel to seek a somatic explanation for such symptoms, and to conclude that they are caused by poor control of blood sugar levels. They are hesitant to over-diagnose psychological disorders, and instead perhaps, over-diagnose somatic problems in the diabetic patient. As a consequence, a pattern develops which may perpetuate and encourage depression in diabetics; patients seek help and are treated as though their symptoms are indicative of poor control alone, and may be berated for this poor control. Depression is not diagnosed, and the patient feels worse after the visit to the doctor.

As is true with the rate of eating disorders in diabetic patients, it is accepted that diabetics have higher rates of depression than do non-diabetics, though researchers widely disagree as to the variance between the two groups (Goodnick, 1997; de Groot, Jacobson, Samson & Welch, 1999). While many trends have been indicated, few have found statistically significant differences indicating that diabetics would be more prone to depression. Lustman, et al. (1997) note that depression is often overlooked in diabetics and medical professionals may indicate that the lethargy, sleep disturbances, and sexual dysfunction would lessen with tighter control of the diabetes-care regimen. The DSM-IV explicitly states diagnosis of depression in persons with chronic illnesses (including diabetes) is more difficult because, physical symptoms such as those mentioned above can only be considered indicative of depression if they are not "clearly and fully accounted for" by a general medical condition (APA 1994, p. 321).

Intentional insulin omission (undertreating elevated blood glucose levels) by diabetics can be reflective of disordered eating patterns. These patterns can manifest in two ways: the first occurs when diabetics (as mentioned above), intentionally omit their prescribed doses of insulin. Insulin can, in fact, make Type I diabetics gain weight, if they should increase their dosages to compensate for an increase in food intake. Omitting insulin injections can lead to a condition called, ketoacidosis, in which there is not enough insulin in the body to control the blood sugar level by transporting the blood glucose to the bodily systems that are in need of that glucose. The excess blood glucose, instead, is filtered out through the kidneys, while ketone acids break down fat, and then muscle tissue, into a substance easily used by the body without insulin. In essence, the omission of insulin causes weight loss in two ways, first by starving the body though the patient may still be eating, and then by the ketones breaking down body mass. A second disordered eating pattern found in diabetics is occurs when low blood sugar levels are overtreated with the rapid intake of large quantities of food, a behavior often seen in those who engage in binge-eating. Low blood sugar reactions, if untreated, can lead to a comatose state during which the patient may incur brain damage, or may even die. Often, in the midst of an insulin reaction (low blood sugar), the diabetic may eat until he/she feels his/her blood sugar return to a safe level; this may in fact be over eating.

Crow, Keel & Kendall (1998) examined how diabetic women manipulate their diabetic control as a means of regulating weight. Crow et al. (1998) noted that the nutrition education often given to diabetic patients could either frustrate or scare patients into further reliance on such manipulation of their diabetes in order to manage their weight. The same study refutes the idea that diabetics administer less insulin than is necessary simply because they do not understand the complications. Crow, et al. (1998) indicate that diabetic patients are, perhaps all too aware of the effects of insulin that are not directly related to blood sugar regulation (i.e., weight gain). They indicate that these effects might motivate diabetics to take less insulin than is needed for healthy control. The rates of eating disorders in diabetics and those suffering from eating disorders in the absence of diabetes are comparable. As indicated by Crow, et al. (1998), results from the Eating Attitudes Test (EAT) have showed that diabetic women have higher scores (indicating less healthy attitudes toward eating) than do non-diabetics, while their rates for eating disorder diagnosis are the same. It may be that the extreme focus on diet and nutrition associated with diabetes education affects adolescents (typical age of diagnosis for Type I Diabetes Mellitus) at a time that they are also most susceptible to pressures from peers and society.

Within the period of adolescence, Marcia recognized four distinct phases of identity development that a person may progress through in any order, at any pace. The first such phase is referred to as identity-diffusion. Subjects in this phase have not yet experienced any identity exploration, and have not made any personal commitment to a set of beliefs. Marcia’s second phase, identity foreclosed, sees the subject with a clear set of beliefs, though he/she has not yet experienced any identity exploration. Instead these beliefs have been dictated to them by parents or other authority figures such as clergy. Phase number three, moratorium, involves an intense struggle to forging a new identity during which the subject is actively exploring new beliefs, and struggling to develop an identity. The subject in the fourth phase, identity-achieved, has gone through a crisis of identity formation, and has come out with a strong personal commitment to a set of beliefs and values.

Schur, Gamsu and Barley (1999) postulate that the presence of a chronic illness during adolescence may create such a disruption to the development of self that it takes chronically ill teens longer to develop their sense of self than it takes their peers. Marcia might counter that there is no set timeline for the progression to identity achievement. However, diabetic adolescents may take longer to reach moratorium, having so completely incorporated others’ beliefs regarding diabetes as their own. Some diabetic youths may be completely foreclosed in regard to their identity as a diabetic, and may well remain this way until their health is completely in their own hands. Diabetic adolescents may also stay in a period of moratorium longer, struggling to develop secure beliefs regarding their illness. This extended period of moratorium might well be detrimental to the health of the young diabetic as struggling to incorporate diabetes into an identity schema may involve a period of denial and poor control.

In 1993, the results from a longitudinal, multiple-site study were published, which indicated for the first time that intensive regulation of IDDM can prevent the progression of long-term complications (Diabetes Control and Complications Trial Research Group). Though it had long been believed that tighter control over diabetes would prevent many complications, it had never before been empirically studied. Prior to the publication of the results from the Diabetes Control and Complications Trial (DCCT), IDDM was typically managed by two injections of insulin per day, daily monitoring of blood glucose levels, and visits to a physician every three months.

The shift in diabetes care, from traditional to intensive treatment, may have led to a marked increase in the presence of eating disorders and depression in persons with IDDM. This causes some concern that, in the rush to embrace the results of the DCCT, and to prevent medical complications down the road, the immediate psychological needs of young diabetic adults are not being addressed. In fact, it is likely that the shift in medical treatment has increased the rates of psychological disturbances in diabetics.

HYPOTHESES

1. Diabetic females will exhibit higher rates of disordered attitudes toward food and eating than will their non-diabetic peers.
2. Diabetic females will exhibit higher rates of depression than will their non-diabetic peers.
3. Diabetic males will exhibit higher rates of depression than will non-diabetic females.
4. Diabetic males will exhibit higher rates of disordered attitudes toward food and eating than will non-diabetic females.

PARTICIPANTS

Participants in the main study were all female, of various racial and ethnic groups, from the ages of 17 to 23. Male diabetics recruited for the exploratory study were also ages 17 to 23, from varied ethnic backgrounds. The control group was comprised of Sweet Briar College students. Those participants in the diabetic groups were either Sweet Briar students with diabetes whom the investigator had previously met, or persons who attended Camp Carefree summer camp for diabetic youths with the investigator during the summer of 1995. Thus, diabetic participants had lived with Type I diabetes mellitus for at least five years.

The diabetic participants were contacted by mail, as most of them reside in the greater New Hampshire area. There was an introductory letter enclosed, explaining how they were selected for the study, the nature of the project itself, the testing materials and the consent form.

PROCEDURE

Research packets were mailed to 60 potential diabetic participants. Instructions included encouraging the participants to closely look over the testing material to decide whether or not they would like to participate; explaining that they could complete the forms in any order; reminding that they were not obligated to fill out the forms in their entirety; asking that the materials be sent back to the investigator in the enclosed envelopes; and finally, encouraging that they seek professional assistance if the study caused any questions regarding their own health.

1. Beck’s Depression Indicator

As one of the better received of the normed depression indicators, Beck’s was the appropriate choice for a measure in this study. Each BDI was scored by adding the numbers associated with the statements that the participant identified as most accurately how she had been feeling over the week prior to completing the test. Scores of 0-9 are within the minimal range of depression; 10-16 indicates mild depression; scores 17-29 indicates moderate depression; and scores of 30-63 are reflective of severe depression.

2. Eating Attitudes Test

The EAT was used for two reasons: it was used by Crow, Keel & Kendall in their 1998 study, and that it measures attitudes, not behaviors, and is specifically not diagnostic. The EAT was scored by assigning values to the responses of the participant. Statements that the participant said occurred "Always" were given a value of 3, responses of "Usually" received a 2, and responses of "Often" were given a 1. Responses reflective of less food-conscious attitudes, "Sometimes," "Rarely," and "Never," were all given values of 0. Once the responses were all given a numerical value, the scoring was completed by adding up the total value for the responses to the EAT. Though not a diagnostic tool, the creators of the EAT recommend that those individuals with EAT scores greater than 20 seek professional psychological assistance.

3. Health Survey

The diabetic participants were given an additional health survey that was designed by the researcher with the intention of roughly examining such things as social support and diabetic control. There are questions regarding traditional measures of diabetic control (high and low blood sugar frequency), as well as behavior (alcohol consumption) that may be detrimental to the health of diabetic persons

RESULTS

Analysis

It was hypothesized that college-aged diabetic women would exhibit higher rates of depression than would their non-diabetic counterparts. An unpaired t-test was performed to evaluate this hypothesis. There was indeed a statistically significant difference between diabetics and non-diabetics, t (34) = -2.44, p < .019. The mean values do reflect this trend of higher scores by the diabetics on the BDI: diabetic mean = 10.89, non-diabetic mean = 4.82.

A second unpaired t-test was performed to evaluate the second hypothesis; college-aged diabetic women would exhibit higher rates of disordered eating attitudes than would their non-diabetic cohorts. Again, there was a significant difference between the two groups; t (34) = -2.64, p < .012. The mean values did reflect that the diabetics scored higher on the EAT: diabetic mean = 10.78, non-diabetic mean = 4.17.

The third and fourth hypotheses, regarding the smaller, exploratory study with the males were also evaluated with t-tests. It was found that diabetic males have slightly, though not significantly, higher rates of both depression and disordered attitudes toward eating than do non-diabetic females. While the mean depression score for diabetic males was higher than for non-diabetic females, 8.66, compared with 4.82, the P-value did not reflect significance: t (21) = -3.84, p > .181. The mean scores of diabetic males were barely higher than those of non-diabetic females on the EAT, 5.83, compared with 4.17, with P-values that were far from significant: t (21) = -.81, p > .429.

DISCUSSION
This study found a statistically significant difference between the levels of depression and disordered eating attitudes in college-aged females with diabetes when compared with their non-diabetic peers. The responses from college-aged male diabetics showed no significant difference between the levels of the two disorders when compared with non-diabetic females, or with diabetic females. Previous research (Lestman, et. al., 1997) indicates, however, that the mean depression scores of diabetics could have been 3-4 times higher than the mean scores of non-diabetics. According to Crow, Keel & Kendall (1998), the diabetics’ mean scores on the EAT might have been significantly higher than those of non-diabetics as well. It is believed that larger samples of diabetic males, and a comparison group of non-diabetic, age-matched males could show a more significant effect. As males were included in the present study for exploratory purposes only, the trend identified represents an area for certain further research.

In addition, the health surveys administered failed to accurately measure how the shift in the specific medical treatment of diabetes (in response to the results of the Diabetes Control and Complications Trial of 1993) may influence the patients’ psychological reactions to the illness. Further study in this area would also be beneficial.