Approximately one-third of participants aged 40 years or older with diabetes had LED. A similar prevalence of risk factor control at the ADA-recommended level was observed in participants with and without LED for 4 of 6 measures (systolic blood pressure, diastolic blood pressure, BMI, and smoking status). People with LED had a significantly lower prevalence of risk factor control at the ADA recommended level compared with those without LED in 2 of 6 measures (HbA1c and HDL cholesterol).
In our study, people with LED were less likely to have tight glycemic and lipid control than were those without LED. Although incomplete success controlling these risk factors was observed in both groups, the implications may be worse for people with LED because they are at increased risk for cardiovascular events and amputation. Poor glycemic control has been associated with increased susceptibility to foot infection and poor wound healing among people with diabetes (18
), and poor lipid control and poor glycemic control have been associated with cardiovascular mortality (20
). Therefore, efforts to improve glycemic control of blood lipids among diabetic people with early stage LED may be an effective strategy in reducing subsequent amputations and cardiovascular events.
Among participants with LED, we found racial/ethnic differences in the level of a few risk factors that could influence progression to amputation and other cardiovascular outcomes. Non-Hispanic blacks were less likely to have adequate glycemic control and blood pressure control than were non-Hispanic whites, but no differences were observed for HDL control. Unlike non-Hispanic blacks, Mexican Americans had no differences in level of risk factor control among people with LED compared with non-Hispanic whites.
Racial/ethnic differences in late-stage disease have been observed. Data from the National Hospital Discharge Survey have shown that whites with diabetes have a lower rate of nontraumatic amputations than do blacks with diabetes (21
). Additionally, higher rates of lower limb amputation have been shown to be directly attributable to peripheral vascular disease among blacks compared with whites (22
). Differences in disease control may play a role in the racial/ethnic differences in amputation rates (23
). Racial/ethnic differences in cardiovascular disease have also been documented. Compared with non-Hispanic whites with diabetes, non-Hispanic blacks with diabetes have higher rates of hospitalizations for cardiovascular events (25
). Racial/ethnic differences observed in risk factor control may be associated with subsequent racial/ethnic differences in late-stage disease. Although the findings from our study do not provide direct support for that association, they do suggest that future studies aimed at clarifying potential racial/ethnic differences in diabetes control may be warranted.
Low education level was associated with an increased risk for poor control of modifiable risk factors among people with and without LED. This finding is similar to those of previous reports that have documented a higher prevalence of cardiovascular disease risk factors in people with less than high school education compared with people with some college education or greater (26
Our findings have implications for national diabetes objectives. Healthy People 2010
has established objectives to reduce the rate of lower extremity amputations in people with diabetes and to reduce deaths from cardiovascular disease in people with diabetes (28
). Interventions focused on risk factor reduction among diabetic people, especially those with LED, may contribute to reducing lower extremity amputations and cardiovascular disease and help achieve Healthy People 2010
objectives. Our findings also reinforce the importance of the Control your Diabetes for Life national education message promoted by the National Diabetes Education Program, American Diabetes Association, and American College of Cardiology, which encourages diabetes patients to maintain recommended levels of HbA1c, blood pressure, and cholesterol (29
Our study has several limitations. First, only noninvasive tests were used to measure PAD and PN. Although more comprehensive tests may provide more accurate diagnostic testing, both the ABI test for PAD and the monofilament test for PN have been shown to have high degrees of both sensitivity and specificity (13
). Additionally, the ADA recommends annual foot examinations that include PN testing with the monofilament and PAD testing with the ABI, and cutpoints for PAD and PN determination were based on clinical practice guidelines and epidemiologic studies (13
). Second, we used HDL cholesterol to assess blood lipid control, but LDL cholesterol is the current preferred measure for assessment of blood lipid control and treatment. Third, physical activity level is associated with cardiovascular disease, but we did not analyze it as a risk factor because clinical recommendations for physical activity by diabetic people with LED have not been established (15
). Furthermore, it is not known whether physical activity (especially weight-bearing activity) is contraindicated and therefore not promoted by health care providers for diabetic people with severe LED. Fourth, medication data were not used in these analyses. It was difficult to assess whether medications were appropriate for respondents' diabetic management needs, so interpreting these findings would be problematic. Fifth, data on LED were missing for approximately 12% of the sampled people aged 40 or older with diabetes. People with missing PAD data were more likely to be female, older, and have less than a high school education. These differences may have biased our results because increasing age and lower socioeconomic status have been linked to worse diabetes outcomes. Finally, NHANES does not include institutionalized people (eg, nursing home residents), and findings that include institutionalized people may differ from ours.
Our study provides the first estimates of modifiable risk factors among diabetic people with and without LED using nationally representative data. Our data show a higher prevalence of 2 of 6 modifiable risk factors studied that may affect progression to late-stage disease among diabetic adults with LED compared with those without LED. Our findings also show different levels of risk factor control for some risk factors among racial/ethnic groups with diabetes and LED. Exploring the potential role of risk factor control among people with LED may reduce amputations and cardiovascular events and may help explain and possibly reduce the observed racial/ethnic disparity in amputations.