Diabetes or Impaired Glucose Tolerance: Does the label matter?

Participants were adults attending a primary care clinic, and inclusion criteria were fasting glucose level 100–140 mg/dl and glucose level >140 but <200 mg/dl 2-h post–75-g glucose. Glucose tests were obtained from routine evaluation. Patients had no previous diagnosis of impaired glucose tolerance or diabetes or any additional comorbidity; they also had not ingested medicines that affect glucose metabolism. All patients volunteered to participate in the study. Written consent was not requested because the maneuver being tested involved only what was being said to the patient. The study was approved by the local ethics committee.

A sample size of 52 patients was calculated to detect a difference in compliance between the groups (1:1 ratio) of 30% (from 2.4 to 3.2 points in the scale), assuming 20% loss to follow-up, with a two-tailed significance test and α = 0.05 with 80% power. Diagnostic labeling was standardized via a leaflet containing general information on problems of glucose metabolism and how to manage them. Half of the leaflets (randomly) used the label “diabetes” and the other half used “impaired glucose tolerance;” the rest of the information was identical. The leaflets advised readers that problems associated with glucose are dynamic and could result in changes in diagnosis, depending on the phase of the illness. The specific variables studied were knowledge about the disease (2), compliance with treatment (2), quality of life (3), emotional functioning (3,4), coping mechanisms (4), and glucose control.

The study was presented as a “Program for Glucose Problems.” Randomization was carried out in blocks of 13 subjects. The informational leaflets were in envelopes numbered from 1 to 52, corresponding to participants’ sequence in signing up for the study. One of the authors, not involved with clinical care of patients, did the “labeling” and was also responsible for interviewing the participants and giving them the appropriate leaflet. No recommendations for drug therapy were given. Family physicians continued with patients’ treatment and were trained not to emphasize either diagnostic label.

At the first postlabeling visit, 8 weeks after the baseline visit, the researcher reinforced the “label” as described in the leaflet. The final evaluation was carried out 16 weeks after baseline. Both postlabeling evaluations were done by an evaluator blind to group assignment. Differences between groups were analyzed with Student's t test and changes within groups with paired t tests.

We recruited 52 participants: 50 patients (25 in each group) remained until the end of the study, of which 42 were women. The mean ± SD level of glycemia postload was 164 ± 15.4 mg/dl (range 141–193), and the fasting glucose level at the end of the study was 107 ± 8.6 mg/dl. No statistically significant differences (data not shown) were found between the groups compared at the baseline and follow-up evaluations. In this context, our primary hypothesis—to detect enhanced compliance in the diabetes-labeled group—was not confirmed.

In comparisons within each group, however, patients who were labeled with “impaired glucose tolerance” significantly increased their knowledge about their disease on the composite scale (see Table 1). Patients labeled with “diabetes” also increased their knowledge of the disease, as measured by both of the evaluations used. In addition, patients labeled with “diabetes” had decreased scores on emotional impact, avoidance distraction, and the integration subscale (with lower ratings showing that a patient is more likely to accept the prospect of living with diabetes). For example, the mean score for avoidance distraction decreased from 0.32 to 0.21 (P = 0.002) among the diabetes-labeled group.

Telling patients they had diabetes had a greater impact in changing some of their views about the disease than telling them they had impaired glucose tolerance. When treating patients, physicians should recognize that diagnosis (labeling) is a critically important component of the therapeutic process. In fact, patients can identify with a diagnostic label despite a lack of understanding regarding the details of their illness. Much of the tendency to adhere to treatment depends on whether patients accept the label placed on their illness (5).

This study has several strengths, including its randomized trial design, blinded evaluations, and measurements using previously validated instruments. Study limitations included a short time interval and relatively small sample size, and although information provided to patients was standardized, it was not possible to control all information patients may have received. Overall, the results highlight the potential impact of how diagnostic information is provided to patients.

In summary, diagnosis, like measurement, assigns subjects to a category according to defined rules. Feinstein (6) has pointed out that the measurement of clinical phenomena must be done after answering key questions, e.g., what is the objective of the measurement? Does treatment differ depending on the diagnosis assigned? More recently, the Standards for Reporting of Diagnostic Accuracy (STARD) Initiative (7) provided criteria for evaluation diagnostic accuracy. The current study emphasizes the additional consideration of how patients react to diagnostic information.