A brief, regular, proactive telephone “coaching” intervention for diabetes

Abstract

Telephone-delivered interventions (TDIs) represent a potentially cost-effective method to increase medical adherence. TDIs for diabetes patients have typically been delivered by nurses or computerized telephone messaging. Psychology undergraduates, however, are less costly than nurses, have a strong background in behavioral science, and provide the personal relationship missing with computerized contact. This paper presents the rationale for and description of a brief, regular, proactive telephone intervention designed to be delivered by psychology undergraduates (i.e., paraprofessionals). “Coaches” administer a 15-min telephone intervention weekly for 3 months and biweekly for 3 additional months. Guided by a semistructured protocol that focuses on behavioral goals, coaches provides support, collaborative problem-solving, and apply basic cognitive-behavioral techniques. Results from a pilot study on type 1 diabetes patients are presented. This preliminary evidence suggests that the program is feasible, acceptable to a large majority of patients, and effective in reducing HbA1c levels.

Introduction

The adverse effects of chronic hyperglycemia include damage, dysfunction, and failure of various organs, especially the eyes, kidneys, nerves, heart, and blood vessels (Expert Committee on the Diagnosis and Classification of Diabetes Mellitus, 2002). To maintain healthy blood glucose levels, diabetes patients must adhere continually to a complex and often burdensome daily regimen that includes monitoring and self-regulation of diet, blood sugar levels, exercise, and medication. Patients can also reduce complications by engaging in daily foot care, obtaining biannual dental and annual eye exams, and receiving annual influenza vaccinations. Finally, diabetes self-care also entails coping with psychological distress, which can adversely influence blood sugar levels and diabetes control (Rubin & Napora, 2001). Although adherence to this regimen has positive short- and long-term effects on physical and psychological health, many patients fall short of the prescribed self-care goals (Fisher et al., 1997). Successful diabetes treatment thus requires the development of psychosocial interventions that promote and maintain effective self-management.

Despite the grave consequences of uncontrolled diabetes and evidence that self-management training can improve patient health, only 35% of people with diabetes in the US have attended a diabetes education class or course (Fertig, Simons, & Martin, 1995). Self-management interventions must, therefore, be acceptable and accessible to a large majority of patients. Most insurance plans do not adequately cover psychosocial and behavioral interventions, and many diabetes patients are uninsured. Even when affordable, many patients will not attend weekly self-management programs. Patients are often too busy to allocate time or modify their work schedules to attend formal programs. Patients in rural areas find the distance to their healthcare facility an obstacle to attending classes. Medical complications (e.g., eye damage, amputations) also make travel difficult for many patients, and economically disadvantaged patients do not have access to inexpensive and convenient transportation. Thus, self-management programs with proven efficacy will have limited utility if most diabetes patients will not or cannot participate (Glasgow, McKay, Piette, & Reynolds, 2001).

Due to strains on medical resources, self-management programs must also be cost-effective (Glasgow & Eakin, 1998). Physicians and nurses today are too overburdened to address fully the psychosocial and behavioral aspects of patient adherence. Interventions must take little or no additional time, or alternatively save time for the medical staff. Desirable interventions should therefore be inexpensive and easily integrated into the existing health care delivery system. Optimally, self-management programs can address the unique barriers faced by each individual patient (i.e., tailored for the patient). Self-management interventions must also address a wide range of mental health problems associated with diabetes, such as stress and high rates of depression (Anderson, Freedland, Clouse, & Lustman, 2001). Finally, it is widely accepted that single or irregular interventions are insufficient to promote stable change. Frequent interventions and regular follow-ups over an extended period of time appear necessary to promote enduring change Norris et al., 2001, Rubin & Napora, 2001.

To achieve these objectives, we have designed a brief, regular, proactive, telephone-delivered self-management intervention that can be delivered by undergraduate psychology students (i.e., paraprofessionals). The program provides “telephone coaching” to patients for 15–20 min weekly for 3 months and biweekly for 3 additional months. Guided by a semistructured protocol that focuses on behavioral goals, the coach provides informational and emotional support, collaborative problem-solving, and applies basic cognitive-behavioral techniques. This paper describes the program, its rationale, and preliminary evidence of its reach and effectiveness in a sample of type 1 diabetes patients.

The seminal Diabetes Control and Complications Research Group study (Diabetes Control and Complications Trial Research Group, 1993) showed that intensive treatment of type 1 diabetes produced a significant and enduring reduction in blood glucose. Glycemic control, in turn, substantially reduced the frequency and severity of long-term health complications. A salient feature of the DCCT was its unusually high level of patient adherence to the medical regimen (e.g., <1% dropout rate over a 9-year period). Many have speculated that the adherence rate was due largely to the psychosocial and educational components of the treatment regimen, especially regular contact with the medical team. Intensive therapy patients visited their study center each month and were telephoned more frequently to review and adjust their regimens. Although telephone care was designed primarily to extend patient education and address medical questions, often it provided encouragement and support in dealing with adherence difficulties and more general life issues. DCCT patients identified the staffs' accessibility, encouragement, and support as particularly helpful (Fisher et al., 1997).

Consistent with these observations, evidence indicates that telephone-delivered interventions (TDIs) offer a promising, cost-effective method to improve health service delivery across a wide range of medical problems (McBride & Rimer, 1999), including diabetes Estey et al., 1990, Piette et al., 2000, Weinberger et al., 1995, Whitlock et al., 2000. TDIs for diabetes patients have typically been delivered by nurses or computerized telephone messaging. However, there is a shortage of nurses, nurses are too costly to be widely assigned to telephone care delivery, their training does not emphasize behavior change interventions, and they are often not fully prepared to deal with co-morbid mental disorders (e.g., depression, anxiety) that often accompany diabetes (Anderson et al., 2001). Although less costly, computerized automated telephone contact fails to provide one-on-one, patient–provider interactions that many consider essential.

To address competing concerns over cost and the need for one-on-one contact, our program utilizes psychology undergraduates, i.e., advanced undergraduates or recent graduates with Bachelor degrees. Psychology undergraduates have a strong background in behavioral science, are oriented toward the helping professions, are less costly than nurses, and provide the personal relationship missing with computerized contact. With specialized training and under close supervision, we hypothesize that psychology students can effectively leverage the expertise of relevant healthcare professionals to improve the quality of care for diabetes patients.

The literature on diabetes self-management training suggests a pressing need to investigate the potential role of nontraditional healthcare providers and, in particular, paraprofessionals. A recent ADA report on National Standards for Diabetes Self-Management notes the potential value of paraprofessionals and the need for research on their effectiveness (Mensing et al., 2002). A PubMed and PsychInfo literature search using the terms “diabetes and paraprofessional” yielded no publications. In contrast, meta-analytic and qualitative reviews of psychological interventions show that, for many psychological problems, paraprofessional therapists are generally as effective as professionals (Christensen & Jacobson, 1994). To the extent that delivery of diabetes self-management training involves many of the same skills required to deliver psychological interventions, these data suggest that paraprofessionals may also be effective in promoting diabetes self-management.

The telephone coaching intervention is designed to ensure patient understanding of the treatment protocol, provide informational and emotional support, monitor adherence, enhance motivation, and foster problem-solving skills to carry out the treatment plan established by the medical team. Coaches are taught to be nonjudgmental, to reinforce adherence, offer empathy, encourage effort, and be prepared to engage the patient in discussions aimed at identifying ways to follow the prescribed regimen better. Often nominated by faculty members, coaches are selected for maturity, conscientiousness, and academic standing. They are entitled “coach” to convey to patients that the calls are intended to facilitate the patient's personal attempt to manage their diabetes, i.e., the calls are task-oriented. “Coach” was chosen instead of “counselor” also to avoid invoking a mental health set or implying that the calls are open-ended “counseling” sessions.

The intervention is highly structured. Before the weekly telephone sessions, patients are asked to complete a battery of self-report measures that assess adherence patterns, quality of life, psychological well-being, and other relevant psychological variables (e.g., social support, self-efficacy). Data from the self-report measures and medical record are reviewed to identify areas of self-care that warrant attention. This information is summarized on a Data Summary Form. The coach then conducts a semistructured telephone interview similar to The Diabetes Self-Management Profile (Harris et al., 2000), which we modified for adult types 1 and 2 patients. The purposes of the interview are to obtain more detailed information about adherence patterns, learn about each patient's unique obstacles and barriers to adherence, and, perhaps most important, to establish rapport. During the interview, the coach and patient collaboratively identify behavioral goals related to the major facets of diabetes care (e.g., blood testing, medication, exercise, diet). Participants are then sent a packet of educational materials and self-management guides (e.g., weekly goal sheets, a 24-h food diary).

The 6-month telephone intervention involves one call per week for 3 months and biweekly calls for the following 3 months. Participants are informed that telephone sessions must end in 15 min so that other patients can be called. However, coaches can occasionally allow the session to continue a bit longer if the 15-min mark occurs at an inopportune time for ending the conversation. Any topic is open for discussion if it pertains to the patient's management of diabetes (e.g., interpersonal, familial, financial).

To foster uniform and valid implementation, each telephone session is guided by a Weekly Coaching Checklist (WCC), which provides a structured, step-by-step format for conducting the session. Sessions begin with an inquiry about items the patient would like to discuss during the session (agenda setting). The coach then reviews weekly blood glucose levels. Particular attention is given to “out of range” readings, with the goal of helping the patient identify possible causes and consequences. The rest of the session focuses on goal-attainment and goal-setting for each area of diabetes self-care that is an area of concern: blood sugar testing, medication, diet/nutrition, exercise/physical activity, foot care, stress, and other areas of self-care that require less frequent attention (e.g., eye and dental appointments, vaccinations). Goal attainment is reviewed and new goals are established for the upcoming week. Weekly goals are stated as an implementation intention (Gollwitzer, 1999), i.e., an explicit statement of when, where, and how the patient will achieve the goal. Initially, coaches encourage limited, highly attainable goals and objectives, with the expectation of gradually increasing the level, complexity, and number of goals. The WCC directs the coach to congratulate positive changes and to encourage and support further change. When patients have difficulty achieving intentions, coaches query the patient gently to help identify obstacles that interfered with goal attainment. They then attempt to engage the patient in collaborative problem-solving to overcome the obstacle. When deemed appropriate, goals are modified or reduced to increase the probability of success. Note that before recommending an increase in exercise or physical activity, coaches administer the Physical Activity Readiness Questionnaire (Thomas, Reading, & Shepard, 1992), a structured interview designed to screen for physical conditions that might contraindicate exercise. If warning signs are apparent, the patient is instructed to consult with their physician, or the physician is contacted directly by the coach (or coach's supervisor) for guidance about exercise recommendations.

In more recent iterations of the coaching intervention, a Weekly Session Review (WSR) checklist has been added. Developed for this intervention, the WSR lists 26 possible coaching responses (e.g., plan a behavioral experiment, activity scheduling, congratulate patient). The WSR is completed by the coach after each session to record the intervention components employed. The WSR allows for process evaluation and also reminds coaches of recommended coaching techniques.

Coaches are trained to avoid offering any medical advice, a feature of the program emphasized to patients and coaches. Coaches and the supervising psychologist confer with members of the diabetes healthcare team on an “as needed” basis. When practical or necessary, however, patients are encouraged to confer with the physician directly. In this way, coaches foster information exchange between the patient and the medical team.

Coaches must successfully complete the University of South Florida's Diabetes Self-Management Education Program, an American Diabetes Association recognized program taught by Certified Diabetes Educators and physicians. Besides imparting knowledge about diabetes self-care, the program gives coaches exposure to the concerns and knowledge level of a sample of diabetes patients. Each coach also receives a set of coaching intervention modules. The modules concisely describe the rationale, goals, and intervention techniques for major components of diabetes self-management: diet, blood sugar testing, modifications, exercise, foot care, eye examinations, dental care, stress management, hypertension control, influenza and pneumonia vaccinations, and aspirin therapy. Supplemental references are provided with each module. Finally, on a weekly basis, the PI sends “coaching tips” by email to coaches. These tips contain brief articles that reinforce or expand upon the information contained in the intervention modules.

A licensed clinical psychologist trains the coaches to administer the assessment interview, the telephone intervention, and provides weekly supervision. A subset of telephone sessions is tape-recorded (coach's responses only) to monitor for accuracy, professionalism and adherence to the intervention protocol. If evidence of clinically significant psychological disturbance appears (e.g., suicidal ideation or intent), the psychologist contacts the patient directly for further assessment and, when appropriate, makes a referral for mental health treatment. Milder levels of psychological disturbance (anxiety, depression) are addressed by the coaching intervention under the category of “stress management.”

Research points to the effectiveness of cognitive-behavioral techniques in facilitating medication adherence, diet, and exercise Chambless et al., 1998, Van Der Ven et al., 2000. Coaches are trained to use, when appropriate, several cognitive-behavioral techniques that are relatively easy to learn and administer. One useful technique is activity scheduling, which involves making specific plans to engage in prescribed tasks while attempting to anticipate and prepare for obstacles that might interfere with successful completion of the task. Cognitive science research has recently demonstrated that the likelihood of following-through to obtain a goal can be enhanced by establishing implementation intentions (Gollwitzer, 1999). Implementation intentions are specific plans for when, where, and how desired behaviors will be carried out within the context of an individual's established routine. Patients translate broader goals (e.g., exercise more often) to specific intentions tied to specific situational cues (e.g., upon returning from work on Mondays, Wednesdays, and Fridays, walk around the block for at least 15 min). Research has found that tying intentions to specific situational cues “helps people to effectively meet their goals in the face of problems with initiating goal-directed actions, tempting distractions, bad habits, and competing goals” (Gollwitzer, 1999, p. 501).

Conducting “behavioral experiments” is another cognitive-behavioral technique that can be readily applied by coaches. Experiments are devised collaboratively by the coach and patient to test negative expectations about engaging in prescribed activities (e.g., predictions about how painful blood glucose testing will be). By framing behavior change as a data-gathering mission, patients are usually more willing to attempt previously avoided behaviors. Equally important, patients are more likely to reconsider staunchly held negative beliefs when the data from behavioral experiments contradicts those negative beliefs, more so than from simply hearing persuasive arguments that contradict those beliefs (Sacco & Beck, 1995).

A pilot test was conducted to investigate the program's feasibility, effectiveness at reducing HbA1c levels, and the program's “reach.” “Reach” refers to the percent and representativeness of patients willing to participate, a component of the RE-AIM framework for evaluating health promotion programs (Glasgow et al., 2001).