Context: Laboratory studies have shown impairments in driving performance among subjects with type 1 diabetes mellitus when their blood glucose (BG) level is between 2.6 and 3.6 mmol/L (47-65 mg/dL). However, to our knowledge, no data exist examining subjects' decisions to drive at various BG levels during their daily routine.
Objective: To examine type 1 diabetic subjects' decisions to drive during their daily routine based on perception of BG levels compared with actual measured BG levels.
Design and setting: Two separate groups of patients were recruited 2 years apart from 4 academic medical centers.
Participants: All subjects were adults with type 1 diabetes who were drivers and who performed at least 2 BG tests per day. Group 1 (initial) subjects (n = 65) had a mean (SD) age of 38.6 (8.9) years with a mean (SD) diabetes duration of 20.5 (10.6) years, were taking 38.8 (16.8) U/d of insulin, and had a mean (SD) glycosylated hemoglobin (HbA1) level of 10.0% (1.9%). Group 2 (replication) subjects (n = 93) were 35.8 (8.0) years old with a mean diabetes duration of 17.0 (10.6) years, were taking 40.0 (15.5) U/d of insulin, and had a mean (SD) HbA1 level of 8.5% (1.6%). Each subject used a handheld computer to record data on symptoms, cognitive function, insulin dosage, food, activity, estimated and actual BG levels, and whether he/she would drive. Data were entered 3 to 6 times per day for a total of 50 to 70 collections per subject during a 3- to 4-week period.
Main outcome measures: Decisions to drive when subjects estimated their BG level to be less than 2.2 mmol/L (40 mg/dL), 2.2 to 2.8 mmol/L (40-50 mg/dL), 2.8 to 3.3 mmol/L (50-60 mg/dL), 3.3 to 3.9 mmol/L (60-70 mg/dL), 3.9 to 10 mmol/L (70-180 mg/dL), and more than 10 mmol/L (>180 mg/dL), and driving decisions when actual BG levels were in these ranges.
Results: Subjects stated they would drive 43% to 44% of the time when they estimated their BG level to be 3.3 to 3.9 mmol/L (60-70 mg/dL), and 38% to 47% of the time when their actual BG level was less than 2.2 mmol/L (40 mg/dL). Logistic regression analysis demonstrated that number of autonomic symptoms, degree of impairment on cognitive function tests, and BG level estimate predicted 76% to 80% of decisions to drive (P<.01 for all). Approximately 50% of subjects in each group decided to drive at least 50% of the time when their BG level was less than 3.9 mmol/L (70 mg/dL).
Conclusions: Our data suggest that persons with type 1 diabetes may not judge correctly when their BG level is too low to permit safe driving and may consider driving with a low BG level even when they are aware of the low level. Health care professionals should counsel their patients about the risk of driving with hypoglycemia and the importance of measuring BG level before driving.