Vol. 57 / No. 44
Self-Reported Prediabetes and Risk-Reduction Activities — United States, 2006
At least one fourth of U.S. adults are known to have predia- betes, a condition defined as having impaired fasting glucose (plasma glucose level of 100 to <126 mg/dL after an overnight fast), impaired glucose tolerance (plasma glucose level of 140 to <200 mg/dL after a 2-hour oral glucose tolerance test), or both (1–3). Persons with prediabetes are at increased risk for devel- oping type 2 diabetes, heart disease, and stroke (3–5). However, lifestyle changes can prevent or delay development of diabetes and its complications among persons with prediabetes (3,5).To assess the prevalence of self-reported prediabetes among U.S. adults and the prevalence of activities that can reduce the risk for diabetes, CDC analyzed responses to questions regarding prediabetes asked for the first time in the 2006 National Health Interview Survey. is report summarizes the results of that analysis, which determined that, although at least one fourth of U.S. adults are known to have prediabetes through surveys that included laboratory testing, in 2006, only an estimated 4% of U.S. adults had been told they had prediabetes. Among those who had been told they had prediabetes, 68% had tried to lose or control weight, 55% had increased physical activity or exercise, 60% had reduced dietary fat or calories, and 42% had engaged in all three activities. Persons at greater risk for diabetes should be tested according to published recommen- dations, and persons with prediabetes should lose or control their weight and increase their physical activity to reduce their risk for developing diabetes.
NHIS is an annual, nationally representative, household probability survey of the noninstitutionalized, U.S. civilian population, conducted by face-to-face interview (6). In 2006, the survey included 24,275 adults aged >18 years. NHIS sample weights were calibrated to 2000 U.S. census-based totals for sex, age, and race/ethnicity. e total household response rate was 87%, and 81% of persons identified as sample adults completed the interview, yielding a final response rate of 71% for the adult sample (6).
All participants who did not answer “yes” to the question “Other than during pregnancy, have you ever been told by a doctor or health professional that you have diabetes or sugar diabetes?” were asked five additional questions: “Have you ever been told by a doctor or health-care professional that you have 1) prediabetes? 2) impaired fasting glucose? 3) impaired glucose tolerance? 4) borderline diabetes? or 5) high blood sugar?” Prediabetes was defined as answering “yes” to at least one of these five questions. Under this definition, 984 participants reported having been told they had prediabetes.
NHIS participants also were asked three questions about their risk-reduction activities: “People often engage in activi- ties to lower their risk for health problems or certain diseases. During the past 12 months have you 1) been trying to control or lose weight? 2) increased your physical activity or exercise? or 3) reduced the amount of fat or calories in your diet?”
Prevalence of self-reported prediabetes was estimated by sex, age group, race/ethnicit , and weight status.*To make compari- sons among the levels of each covariate while simultaneously adjusting for differences in the distributions of all the other covariates, adjusted percentages were estimated from a logistic regression model predicting prediabetes as a function of sex, age group, race/ethnicity, and weight status (7). Among those adults with self-reported prediabetes, estimates were made of the percentages told they had each of five conditions indicative of prediabetes (i.e., prediabetes itself, impaired fasting glucose, impaired glucose tolerance, borderline diabetes, or high blood sugar). Estimates then were made of the percentages with prediabetes who engaged in each of the three risk-reduction activities and the percentages who engaged in zero, one, two, or all three activities.
In 2006, an estimated 4.0% of U.S. adults had self-reported prediabetes (Table). Among them, 64.4% (95% confidence interval [CI] = 59.9%–68.7%) had been told they had bor- derline diabetes. Percentages for the other four conditions indicative of prediabetes were as follows: high blood sugar,
% (CI = 34.3%–42.5%); prediabetes itsel , 33.7% (CI =
%–37.3%); impaired glucose tolerance, 15.5% (CI =
%–19.2%); and impaired fasting glucose, 15.2% (CI =
%–18.6%). In addition, 43.3% (CI = 39.5%–47.3%) had
been told that they had two or more of the five conditions.
The estimated prevalence of self-reported prediabetes increased with age, ranging from 2.7% among those aged 18–44 years to 6.0% among those aged >65 years (Table). Prevalence also increased with weight status: 2.3% among those with normal weight, 3.9% among those categorized as overweight, and 6.3% among those categorized as obese. Prediabetes was more prevalent among women (4.8%) than men (3.2%). No significant differences were observed in prevalence by race/ethnicity.
Among adults with self-reported prediabetes, 67.6% (CI = 63.7%–71.4%) had been trying to lose or control weight dur- ing the preceding 12 months, 59.8% (CI = 55.6%–64.0%) had reduced dietary fat or calories, and 54.7% (CI = 50.7%–
%) had increased physical activity or exercise. Although
% (CI = 38.0%–46.3%) reported engaging in all three
Categorized as normal weight (body mass index [BMI] of <25 kg/m2),
overweight (25 to <30 kg/m2), or obese (>30 kg/m2), using height and weight reported by participant.