Temporal trends in the treatment of pediatric type 1 diabetes and impact on acute outcomes

Abstract

Objective: To evaluate temporal trends in pediatric type 1 diabetes (T1DM) management and resultant effects on outcomes.

Study design: Two pediatric T1DM cohorts were followed prospectively for 2 years and compared; Cohort 1 (N = 299) was enrolled in 1997 and Cohort 2 (N = 152) was enrolled in 2002. In both cohorts, eligible participants were identified and sequentially approached at regularly scheduled clinic visits until the target number of participants was reached. Main outcome measures were hemoglobin A1c (A1c), body mass index Z score (Z-BMI), and incidence rate (IR; per 100 patient-years) of hypoglycemia, hospitalizations, and emergency room (ER) visits.

Results: At baseline, Cohort 2 monitored blood glucose more frequently than Cohort 1 (> or = 4 times/day: 72% vs 39%, P < .001) and was prescribed more intensive therapy than Cohort 1 (> or = 3 injections/day or pump: 85% vs 65%, P < .001). A1c was lower in Cohort 2 than Cohort 1 at baseline (8.4% vs 8.7%, P = .03) and study's end (8.7% vs 9.0%, P = .04). The cohorts did not differ in Z-BMI (0.83 vs 0.79, P = .57) or IR of hospitalizations (11.2 vs 12.9, P = .38). Cohort 2 had lower IR of total severe hypoglycemic events (29.4 vs 55.4, P < .001) and ER visits (22.0 vs 29.3, P = .02).

Conclusions: T1DM management intensified during the 5 years between cohorts and was accompanied by improved A1c and stable Z-BMI. Along with improved control, IR of severe hypoglycemia and ER visits decreased by almost 50% and 25%, respectively.

Figure 1

a, Baseline daily SMBG. Frequency of SMBG was higher in Cohort 2 than Cohort 1 at baseline (χ²=48.0, df=3, p<.0001). b, Baseline daily injection frequency. There was greater use of intensive treatment (≥ 3 injections/day or CSII) in Cohort 2 than Cohort 1 at baseline (χ²=90.2, df=3, p<.0001). Gray bar=cohort 1 (1997), black bar=cohort 2 (2002).

Figure 2

a, Incidence rate of severe hypoglycemia. Severe hypoglycemia was divided into two mutually exclusive categories. Dark gray area=events requiring the help of another person for oral treatment. Light gray area=events such as seizure or coma requiring emergency medical response or treatment with glucagon and/or intravenous dextrose. The total number of events is shown above each bar. Rates of severe hypoglycemia of adolescents in the DCCT are shown for comparison (10). b, Integrated A1c. An integrated A1c value was calculated for each patient by averaging all A1c values obtained during follow-up and compared with the adolescent cohort of the DCCT (10). Cohorts 1 and 2 had significantly lower A1c values than the conventionally treated adolescent DCCT cohort (p<.001) and significantly higher A1c values than the intensively treated adolescent DCCT cohort (p<.001) (10). The x-axis labels below Figure 2b apply to both Figures 2a and 2b.