BACKGROUND: Management of diabetes mellitus is difficult for both the patient and the clinician, and poor communication resulting from different conceptions of the disease and different treatment goals can exacerbate the problem.
METHODS: We used qualitative methods, including semistructured interviews with patients and clinicians and direct observation of clinical precepting sessions by trained observers at an inner-city family practice training site. Transcripts from these sessions were analyzed for content (with the help of Nud.ist software) to identify themes that highlight problem areas in clinician-patient communication.
RESULTS: The most important themes revolved around the concept of control, which was used in at least 3 different ways, and the concept of frustration. Frustration resulted from different understandings of the biology, physiology, and natural history of the disease, and approaches to treatment. In the interviews, clinicians often spontaneously acknowledged the multiple physical, psychological, and social obstacles to treatment confronted by their patients, but in observed practice they almost always focused entirely on managing blood sugar numbers.
CONCLUSIONS: Different conceptions of the term “control” affect the ability of patients and clinicians to communicate effectively. The tendencies of clinicians to view their own management strategy as scientific truth, and their focus on managing numbers rather than attempting to understand their patient’s conception of disease and their treatment goals, lead to frustration and are serious obstacles to effective collaboration.
Diabetes, especially type II (non-insulin-dependent diabetes mellitus [NIDDM]) is extremely common among primary care patients. It accounts for a large percentage of office visits, and treatments for it are well described. However, caring for patients with diabetes is often a frustrating problem for clinicians.1 A biopsychosocial understanding2,3 of diabetes recognizes that medical outcomes, while occurring at the organ and cellular level, are affected by patient behaviors, and that these behaviors are largely determined by the patient’s world-view.4-9 Physicians and other health professionals may have different world-views than their patients,10-13 but must be able to communicate their ideas effectively and listen to what their patients are saying.
Ineffective communication has been identified as a barrier to effective treatment of diabetes.6,14,15 Differing interpretations of medical terms, differences between the clinician’s personal feelings and professional ideas conveyed to the patient, and discrepancies between what the clinician emphasizes and what the patient thinks is important can obstruct effective communication and successful management of the disease.9
The authors of several recent articles16-19 have pointed out the importance of clinician attitude or patient perception of physician behavior (eg, Dietrich16 argues that “the reaction and attitude of physicians displayed toward the patients at the point of diagnosis were crucial in influencing attitudes toward…the disease [diabetes] and consequently compliance.”), and some research even suggests that the physician’s attitudes toward diabetes management may be more important than actual knowledge of the disease.20 Published literature, however, continues to focus almost exclusively on patient, rather than clinician, beliefs and practices.
The Journal of Family Practice o june 2000 o Vol. 49, No. 6 n 507y medicine, university of texas health science center at san antonio, 7703 floyd curl drive, san antonio, tx 78229-3900.
Methods
Setting
Our project focuses on clinicians and their patients in a busy family practice training site adjacent to a large public hospital in Chicago. The majority of the patient population at this site is poor, predominantly middle-aged or elderly, and sick. Approximately 70% of the patients are African American, 15% are Latino, and 15% are of Asian or Eastern European descent. Two other residency training sites, one a predominantly Latino clinic and the other an African American clinic, were also used to recruit patients and to observe precepting encounters and support groups.
We employed multiple qualitative methods (ie, interviews, informal conversation, and participant observation) to investigate the attitudes and practices of clinicians involved in the care of patients with diabetes. These enabled us to relate statements and actions to their ethnographic or clinical context and to enhance the validity of the data.21,22 Although qualitative research has low reliability compared with experimental research, it has very strong face validity when it includes an observation component that enables the researcher to compare oral statements with actual practice. The presence of this element makes our study unique. The second methodologic objective was to collect data in multiple venues. This is important because physicians and patients are likely to say different things in different conversational settings; to make valid statements about attitudes and beliefs, therefore, it is necessary to sample broadly.
Semistructured Interviews with Clinicians and Patients
A semistructured interview guide for clinicians and a slightly modified guide for patients were developed to insure general consistency across interviews and to facilitate comparisons within and across groups. However, during the interviewing process, issues raised in earlier interviews were formulated as questions for subsequent interviewees (eg, Is fear an effective impetus to behavior change?), thus using the constant comparative method of Glaser and Strauss.23 Interviewers were instructed to elicit stories from patients or particularly significant cases from clinicians.