Cognitive Behavioral Group Training for Poorly Controlled Type 1 Diabetes Patients: A Psychoeducational Approach

Frank J. Snoek, PhD, Nicole C.W. van der Ven, MA, and Caroline Lubach, CDE

People with diabetes are faced with the challenge to adequately self-manage their disease on a day-to-day basis while simultaneously living a "normal," full life. Self-management of type 1 diabetes includes a variety of self-care activities (e.g., multiple self-injections of insulin, checking and recording one's blood glucose levels frequently, eating carefully) that must be carried out daily in many different situations and over a long period of time. In addition, patients have to be watchful for symptoms that may indicate hyperglycemia or hypoglycemia and must take action if necessary to correct blood glucose fluctuations.

Diabetes care is rightfully considered one of the most psychologically and behaviorally demanding of the medical chronic illnesses.¹ Research findings confirm that many patients perceive diabetes as burdensome, particularly the demands of the daily treatment regimen and worries about the risk of developing chronic complications.²

Yet the majority of patients seem to cope reasonably well with their diabetes and manage to keep their blood glucose in fair control. However, only a few patients, if any, achieve optimal regulation of glycemia. Even in the Diabetes Control and Complications Trial (DCCT), only a minority of the highly motivated and monitored group of type 1 patients achieved the target value (HbA_1c <6.05%), of which only 5% was able to maintain that level through the trial.³

Normoglycemia, then, seems not a realistic goal in diabetes management. Striving for acceptable blood glucose control, usually set at HbA_1c values <7.5%,⁴ certainly is worthwhile. In fact, any improvement in glycemic control is worth pursuing, since the DCCT showed that even a modest improvement in HbA_1c can contribute to the delay of onset and progression of chronic microvascular complications.

Unfortunately, clinical data suggest that a substantial group of type 1 patients, estimated to be 25­30% of the total type 1 diabetic outpatient population, is in persistent poor glycemic control (HbA_1c > 8.5%), mainly due to poor regimen adherence. As Rubin and associates⁵ have pointed out, several aspects of diabetes treatment should make it no surprise that people with diabetes generally would find it hard to adhere. These include a) the regimen is demanding; b) the regimen is unpleasant; c) improved control may result in more hypoglycemic episodes; d) "going by the book" does not guarantee results; and e) feeling different and isolated as a person with diabetes may sap motivation. Moreover, efforts to self-manage diabetes do not provide many direct positive rewards, but rather pay off over the longer term, which also makes it difficult to stay on track. For some patients, effective diabetes self-management may be hampered by serious psychological problems (e.g., depression, anxiety), but this is not necessarily the case for all patients with poor adherence.

Research shows that patients in poor control do not share a psychological profile of distinct personality characteristics. Most, at least in the Netherlands, do not lack diabetes education, nor are they unaware of the potential benefits of intensive insulin therapy. As heterogeous as this group may be, they do share a strong, negative attitude toward diabetes, along with feelings of hopelessness and low self-efficacy, i.e., not believing that they can make the necessary behavior changes to control their diabetes. Some patients may even suffer from "diabetes burnout," a psychological condition characterized by chronic frustration and feelings of failure, which may negatively affect glycemic control via the effects of stress and, indirectly, via the effects of distress on self-care behaviors.⁶

How then can these patients be helped? This question was the starting point for developing the short cognitive behavioral group training (CBGT) at our center that is the focus of this paper. Before describing the aims, content, and effects of CBGT, we will look more closely at the role of beliefs in determining how patients feel about their diabetes and cope with the treatment regimen.

Cognitive Behavioral Model of Diabetes
It has long been known that what people believe strongly influences their feelings and behaviors. The ancient Greek philospher Epictetus stated, "People are not disturbed by the things that happen to them, but by the views they take of them."⁷

More recently, the effects of people's cognitions and attitudes regarding their self-care behaviors have drawn increasing attention from clinicians and researchers. Cognitive models of health behavior have been introduced to help us appreciate the complexity of the health behavior change process and to better understand how behaviors are determined by attitudes and beliefs. From clinical practice, we know that educating patients and helping them acquire the necessary technical skills (e.g., injecting insulin, checking and interpreting blood glucose) is a prerequisite for self-care behavior but by no means guarantees success. Nor does informing patients about the benefits of intensive insulin therapy and raising their awareness about the potential risks of nonadherence automatically prompt a change in self-care behavior.

The Health Belief Model (HBM)⁸ proposes that, in addition to the level of knowledge, we need to take into account the beliefs people hold about their vulnerability to medical complications, the benefits and barriers they perceive, and their self-efficacy expectations, i.e., their confidence in being able to perform and maintain the self-care behaviors required to attain glycemic control. Related to this is the concept of "readiness to change" from the transtheoretical model that conceptualizes behavior change as a staged process and distinguishes between different motivational phases ranging from "precontemplation" to "action," when patients finally make the behavior change.⁹ Obviously, a person needs to have a certain readiness to change to gain any benefit from an educational intervention aimed at behavior change.

Nonadherence, then, may occur for several reasons, which are not mutually exclusive.¹⁰

• Patients do not know what to do.
• Patients do not have the skills to carry out the self-care activities.
• Patients do not think they are able to carry out self-care activities.
• Patients do not think carrying out self-care activities will make a difference.
• Patients perceive the costs as greater than the benefits.
• Patients do not receive adequate reinforcement (pay-off) for meeting short-term goals.
• Patients experience a lack of social support.

When cognitive attitudinal barriers, rather than lack of knowledge or skill, are the main obstacles to self-management, a cognitive behavioral intervention is required. Such an approach may be particularly beneficial for patients who, on the basis of experiencing mutiple failures in attempting to manage their diabetes, have little or no belief anymore in their ability to cope adequately with the demands of the treatment regimen, i.e., those who have developed strong negative beliefs regarding their diabetes self-efficacy. Such negative thoughts (e.g., "My diabetes is beyond my control.") result in negative feelings (e.g., sadness, anger) and maladaptive health behaviors (e.g., skipping injections, overeating, not checking blood glucose), thereby reinforcing this negative cycle. The basic premise of the cognitive behavioral model is that there is a constant interplay among the cognitions or beliefs that patients hold about their diabetes, their feelings, their behaviors, and other people (Figure 1).

Challenging patients' dysfunctional beliefs with more realistic, self-helping cognitions will enable them to feel less overwhelmed by the treatment regimen as well as by other stressors in life and will consequently help them cope more effectively with the demands imposed on them. This approach originates from Albert Ellis's Rational Emotive Therapy (RET) and Aaron T. Beck's Cognitive Behavior Therapy (CBT), two similar therapeutic approaches that have been successfully applied to a wide variety of emotional and behavioral problems, including depression and anxiety disorders.^11,12

Literature on RET and CBT in diabetes is still surprisingly limited. Zettler and associates¹³ successfully used a cognitive-behavioral group program to help type 1 diabetic patients cope more effectively with their fears of long-term complications. Recently, a study by Lustman and associates¹⁴ demonstrated significant favorable effects of CBT both on psychological well-being and on glycemic control in depressed type 2 diabetic patients.

RET and CBT are based on the observation that people's emotional problems are founded in a system of dysfunctional or distorted beliefs about themselves and the world surrounding them. These core beliefs generate automatic negative thoughts that are easily triggered and often contain some cognitive distortion, such as mindreading, exaggeration (magnifying), overgeneralization, making unvalidated assumptions, all-or-nothing thinking, or personalization.

According to Ellis, core beliefs fall into three categories: 1) being demanding toward self (e.g., "I must adhere to my diabetes regimen in all circumstances, otherwise I am not a good patient."), 2) being demanding toward others (e.g., "My doctor must always listen to me, or else he is a bad doctor."), and 3) being demanding toward the world (e.g., "My life must always be enjoyable, or else it is not worth living."). Holding these irrational beliefs can easily cause feelings of frustration and result in dysfunctional behaviors, leading to poor diabetes control. This negative experience, in turn, reinforces a person's already existing negative attitude toward diabetes.

From this perspective, it may be extremely helpful for patients with poorly controlled diabetes to learn to explore their automatic thoughts, discover how these thoughts affect their mood and behavior, and challenge these beliefs with more constructive, self-helping cognitions. This will result in reduced negative feelings and more adaptive coping. Negative thoughts, however, are not per se always distorted, and it is important to keep that distinction in mind. There is much that is negative and unpleasant about diabetes, and negative thoughts should therefore be acknowledged as perfectly normal and realistic.

Such thoughts become dysfunctional when, for example, unvalidated assumptions are made (e.g., "I'll lose my eyesight no matter what I do."), or when exaggeration takes place (e.g., "I can never do anything pleasant due to my diabetes."). Patients can benefit from looking critically at negative thoughts that are overwhelming and paralyzing and replacing them with more positive cognitions. A good example is given by Albert Ellis, the founder of RET and a diabetes patient himself, who experienced quite a few problems in adjusting to the demands of the diabetes treatment regimen.
". . . So it is a pain! If I don't keep up with this annoying regimen, I'll soon create much greater pain--and not even be around long enough to be able to cope with it. It's really rough to persist at this regimen. But it'll be much rougher if I don't!"⁵

Cognitive Behavioral Group Training (CBGT)
CBGT was developed by Snoek and colleagues in 1997 as part of a collaborative research project with the Joslin Diabetes Clinic in Boston. The aim was to develop a feasible, short, psychoeducational training program to be delivered in small groups (6­8 patients) by a psychologist and diabetes nurse specialist or diabetes educator, in a diabetes outpatient setting. The target group for CBGT is type 1 diabetic patients in persistent poor glycemic control (HbA_1c >8.5%), in the age range of 18­50 years, with no serious medical or psychiatric comorbidity. Obviously, patients need to have a certain readiness to change in order to make participating in CBGT useful, because CBGT depends highly on the attendence and active participation of the patients during, as well as in between, classes (homework).

The goal of CBGT is to help the patients improve their self-care behaviors and hence their glycemic control without compromising, and if possible while enhancing, their psychological well-being. The assumption is that the content of the program and the subsequent experience of actually improving diabetes regulation will contribute to an improvement of emotional well-being and quality of life.

To evaluate the feasibility and effectiveness of this new intervention, pre- and post-CBGT assessments were scheduled, including HbA_1c and various psychosocial parameters. Feasibility was determined on the basis of different indices, such as response rate, attendance, logistic aspects and costs, trainer evaluations, and patient satisfaction. The effectiveness of CBGT was assessed by changes at 3 and 6 months after completing the intervention in a) glycemic control (HbA_1c, Biorad, ref. 4.3-6.1%); b) psychological well-being, both diabetes-specific (Problem Areas in Diabetes [PAID] scale¹⁵) and generic well-being (Well-Being Questionnaire [WBQ]¹⁶); and c) self-care activities.¹⁷

CBGT was piloted in four groups (n = 24) between October 1997 and November 1998, of which 16 patients now have completed the 3-month follow-up and 9 have completed the 6-month follow-up. The program is "psychoeducational" in the sense that it combines a psychotherapeutic approach (CBT/RET) to diabetes with educational elements, for example regarding the impact of stress on blood glucose. The basic three ingredients of the program are 1) group sharing, 2) cognitive restructuring, and 3) stress management. Before entering CBGT, patients are interviewed by the psychologist about their specific barriers to diabetes self-management and their expectations with regard to the group training. Classes are scheduled from 4:00­5:30 p.m., with a 20-minute break. Juices, coffee, tea, snacks, and sandwiches are provided at no charge.

In four weekly classes, the following topics are addressed sequentially:

1. Learning about CBT/RET and diabetes
2. Stress and diabetes
3. Complications of diabetes/future outlook
4. Interpersonal relationships/social support and diabetes

A trainer's manual has been developed containing a program outline, a detailed schedule, specific instructions, and a list of materials needed (e.g., flip-overs, worksheets). For the patients, four short papers on the respective topics were written and are handed out the week before the relevant class. These handouts contain information on the subject and suggestions for further reading.

Throughout the training, participants work on exploring, challenging, and restructuring their cognitive distortions, with specific focus on the aforementioned topics. For this purpose, participants read the written information before class and do homework assignments that are discussed in the group at the beginning of the meeting. The assignments consist of ABC worksheets, based on the ABC paradigm of RET, where A stands for the Activating event, B for the Beliefs, and C for the Consequences. An example of such a worksheet is shown in Figure 2.

Figure 2. ABC worksheet

Patients are instructed to fill out this sheet in between classes, with regard to situations where they felt distressed or discouraged by their diabetes. They note, as precisely as possible, the Activating event, the unwanted Consequences and the specific Beliefs that gave rise to these emotions and behaviors. They should then ask themselves how they would have wanted to feel and act, and discuss or challenge the rationality of their cognitions. By doing so, patients gain insight into their own cognitive distortions and how these thoughts affect their feelings and behaviors. To explore the rationality of a belief, patients can ask themselves several questions, such as: "Is what I believe true?," "What is the evidence for what I am thinking?," "Does this belief actually help me achieve my goal?," and "Is there another way of looking at it?" The ABC worksheets are discussed in the group meetings among the participants themselves and with the group trainers.  

As a strategy to help patients manage their stress more adequately, a relaxation exercise is introduced, based on Jacobson's widely used progressive muscle relaxation technique.¹⁸ Participants are asked to practice this exercise daily at home and to record their stress levels on a scale of 0 to 100. They are also asked to record blood glucose fluctuations in stressful situations.

The ABC worksheet can help patients identify the events and beliefs that give rise to emotional stress and learn how these beliefs may be altered to create a more effective way of coping with particular stressors. For example, a patient may read a high blood glucose level and think "Oh my God, I must have done something wrong. I am stupid!" Challenging this negative cognition may help him or her to perceive a high (or low) blood glucose reading less negatively and to develop self-helping thoughts that will not cause distress and guilt. The alternative could be, "Good that I checked, because I was not aware that I was running high. I can now take action." (or, "I can take action if the next reading is high, too.").

Classes 2, 3, and 4 start with discussing homework, after which the topic of that session is introduced by one of the trainers and discussed in the group. When addressing the topic of diabetes complications/future outlook, much attention is paid to the fear of developing chronic complications and different ways of coping with this anxiety, ranging from trying not to think about it at all (denial) to constant worrying. CBGT aims to help patients cope with their fears in a way that motivates them to self-manage their diabetes and avoid feelings of extreme distress and depression.

Following a short group discussion, participants are introduced to an anxiety exploration exercise that involves "thinking through" one's fear by using anxiety-provoking thoughts, e.g., "How realistic are my worries?," and "What can I do if my greatest fear comes true?". The results of this anxiety exploration are discussed in the group and taken home to further explore during the next week.

The topic of social support is discussed in the group, looking at individual experiences and expectations regarding the role of participants' significant others, e.g., spouse, parents, or children, in helping to manage their diabetes. Focus is on patients' beliefs about their role as a patient, other people's behaviors as being supportive or not to the patient self-managing diabetes, and what reactions they actually want to see from others around them. Suggestions are made as to how to interact positively with others regarding diabetes and to gain optimal social support from them.

Evaluation
In our pilot, recruitment took place by pre-selecting patients on the basis of a review of medical charts. Approximately one-third of the total group of eligible outpatients that was approached by the physician or another member of the diabetes health care team participated in the study. The main reason for patients to decline the group training was time restriction (work, private situation). Only two mentioned not being motivated enough to participate and foresaw that they would drop out.

The mean age of the patients included was 35.2 years (±11.1), 9 males, 15 females, with a mean duration of diabetes of 17.6 years (±9.4). Most patients were on three or four injections a day, five were treated with an insulin pump (continuous subcutaneous insulin injection), and two were on two injections a day. The mean HbA_1c at baseline was 9.2% (±1.2%).

CBGT was found to be feasible in the sense that patients who participated attended classes and completed the program. Only one patient dropped out after the first meeting. Also, patients generally did their homework, although the mean time spent on it was relatively low (30 minutes a week). On a scale of 1 (very poor) to 10 (excellent), patients overall rated CBGT 7.6 (fairly good), with highest ratings for the first class (on CBT/RET in relation to diabetes). The length of the program (4 weeks) was thought to be too short by more than half of the participants. Most of the participants thought they had invested just the right amount of time in doing their homework; two thought they had invested too little time.

When asked what patients thought was most useful, the exchange of experiences among participants and practicing CBT/RET in their daily lives were mentioned. Most patients reported to have become more actively involved in their diabetes management as a result of CBGT. As one patient put it, "I'm less passive when my blood sugar is high or low. I have come to accept that it is a part of me, rather than a series of lab results."

Because the study is ongoing, final results regarding the efficacy of CBGT cannot yet be presented. First findings are encouraging and suggest an average drop in HbA_1c of >0.5% at 3 months' follow-up, with the largest benefit found in the group in worst control, i.e., those with the highest HbA_1c at baseline. Preliminary data at the 6-month follow-up suggest that this improvement in glycemic control is maintained over time. The fact that, in addition to enhancing glycemic control, CBGT improved positive well-being (WBQ) significantly (P < 0.5), and decreased diabetes distress (PAID) significantly (P < 0.5), is of great importance, because it is known that intensifying self-care may occur at the expense of perceived life quality and is, therefore, not likely to last. Not simply good control, but an overall "good life" is the ultimate goal of diabetes treatment.

It is noteworthy that large inter-individual differences were found to exist between participants on the various psychological measures and reported self-care activities. The effects of CBGT are, therefore, bound to be differential, though a clear general trend was observed. So far, we have found no indications for excluding patients, other than lack of motivation.

Conclusions and Future Directions
Results so far suggest that CBGT is feasible and effective in a selected group of patients with a history of persistent poor glycemic control. The preliminary results need to be interpreted with some caution, because we cannot exclude the possibility that the effects found may be due to a study ("Hawthorne") effect and/or regression to the means. To substantiate the results of this trial, further research is warranted using a randomized, controlled design. This is planned for the near future in our center.

CBGT has been found to be feasible in an (academic) outpatient setting and is appreciated by the participating patients, including patients who initially claimed not to be "group people." We may assume that CBGT is transferrable to other diabetes centers and may be beneficial to various patient groups, e.g., insulin-treated type 2 diabetic patients and type 1 patients in fair-to-good control who experience diabetes as burdensome and have difficulty staying on track with their regimen.

As formative evaluations indicated, the current program was perceived as too short by the majority of patients. We have, therefore, decided to add two topics to the program to be addressed in two extra classes. These will include goal setting and being a partner in doctor-patient communication. Both topics fit in well with the CBGT program and offer good opportunities for cognitive restructuring and social skills training.

In our experience, the combination of a psychologist and diabetes educator as trainers of the group works quite well. To further implement this approach in different care settings, it seems worthwhile to offer training in cognitive behavior therapy (CBT) to diabetes educators.

References

¹Cox DJ, Gonder-Frederick L: Major developments in behavioral diabetes research. J Consult Clin Psychol 60:628-38, 1992.

²Gåfvels C, Lithner F, Börjeson B: Living with diabetes: relationship to gender, duration and complications: a survey in northern Sweden. Diabetic Med 10:768-73, 1993.

³The DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329:977-86, 1993.

⁴European Diabetes Policy Group 1998: A Desktop Guide to Type 1 (Insulin-Dependent) Diabetes Mellitus: Guidelines for Diabetes Care. Brussels, Belgium, International Diabetes Federation, European Region, 1998.

⁵Rubin R, Walen SR, Ellis A: Living with diabetes. J Rational-Emotive Cognitive Behav Ther 1:21-39, 1990.

⁶Polonsky WH: Understanding and treating patients with diabetes burnout. In Practical Psychology for Diabetes Clinicians: How to Deal With the Key Behavioral Issues Faced by Patients and Health-Care Teams. Anderson BJ, Rubin RR, Eds., 1996, p. 183-91.

⁷Epictetus: The Collected Works. Boston, Little Brown, 1890.

⁸Becker MH: The health belief model and personal health behavior. Health Educ Monographs 2:324-508, 1974.

⁹Ruggiero L, Prochaska JO (Eds.): Readiness for change: application of the transtheoretical model to diabetes. Diabetes Spectrum 6:21-56, 1993.

¹⁰Speers MA, Turk DC: Diabetes self-care: knowledge, beliefs, motivation and action. Pat Couns Health Educ 4:144-214, 1982.

¹¹Ellis A: Reason and Emotion in Psychotherapy: A Comprehensive Method of Human Disturbances. New York, Carol Publishing Group, 1994.

¹²Beck AT: Cognitive Therapy and the Emotional Disorders. New York, International Universities Press, 1976.

¹³Zettler A, Duran G, Waadt S, Herschbach P, Strian F: Coping with fear of long-term complications in diabetes mellitus: a model clinical program. Psychother Psychosom 64:178-84, 1995.

¹⁴Lustman PJ, Griffith LS, Freedland KE, Kissel SS, Clouse RE: Cognitive behavior therapy for depression in type 2 diabetes mellitus. Ann Intern Med 129:613-21, 1998.

¹⁵Welch GW, Jacobson AM, Polonsky WH: The Problem Areas In Diabetes scale: an evaluation of its clinical utility. Diabetes Care 20:760-66, 1997.

¹⁶Bradley C: The Well-Being Questionnaire. In Handbook of Psychology and Diabetes: A Guide to Psychological Measurement in Diabetes Research and Practice. Bradley C, Ed. Berkshire, England, Harwood Academic Publishers, 1994.

¹⁷Pennings-van der Eerden L: Self-Care Behavior in the Treatment of Diabetes Mellitus. Thesis, Utrecht, Netherlands, University of Utrecht, 1992.

¹⁸Jacobson E: Progressive Relaxation. Chicago, University of Chicago Press, 1938.

Acknowledgment
This project was financially supported by a grant from Roche Diagnostics (Boehringer Mannheim) and Eli Lilly and Company. We thank Alan M. Jacobson, MD, for his valuable contributions to this project.

Frank J. Snoek, PhD, is a clinical psychologist, associate professor of medical psychology in the Department of Medical Psychology, and senior researcher in the Research Institute of Endocrinology, Reproduction, and Metabolism, at the University Hospital Vrije Universiteit in Amsterdam, the Netherlands. Nicole C.W. van der Ven, MA, is a research psychologist and doctoral student at the Vrije Universiteit. Caroline Lubach, CDE, is a diabetes nurse specialist at the Diabetes Outpatient Clinic of the University Hospital Vrije Universiteit.

Return to Issue Contents

Copyright © 1999 American Diabetes Association

Last updated: 9/99
For Technical Issues contact webmaster@diabetes.org