SAN FRANCISCO — A population-based screening program for type 2 diabetes does not decrease all-cause, cardiovascular, or cancer-related mortality over a 5-year period, according to a large randomized controlled trial presented at the annual scientific sessions of the American Diabetes Association.
Moreover, screened patients offered intensive diabetes treatment did no better in terms of mortality than did screened patients offered routine diabetes treatment, said Dr. Justin Basile Echouffo-Tcheugui of the University of Cambridge (England).
The results call into question the value of large-scale screening for type 2 diabetes and of intensive diabetes management. And they conflict with mathematical models that predicted such screening would result in a 26%–40% reduction in diabetes-specific mortality, he said.
The 79,085 people who participated in the study were patients at 32 primary care practices in England. Using data from medical records, researchers calculated patients' Cambridge Risk Score (CRS), which reflects an individual's risk of developing diabetes. Of the original cohort, 19,881 people with CRS scores above 0.17 were included in the study. According to an earlier study, using a CRS score of 0.17 as a cut-off point results in a 70% sensitivity and a 64% specificity in identifying patients at high risk of type 2 diabetes (Diabetes Care 2002;25:984-8).
In five of the practices, having a total of 4,137 high-risk patients, no further screening was done. These patients constituted the control group. In the remaining 27 practices, patients were offered stepwise screening for type 2 diabetes.
In the first step, they were tested for capillary blood glucose and hemoglobin A1c. Those with suspicious results went to the second step: capillary fasting blood glucose. Those with suspicious results from that test underwent a glucose tolerance test for a definitive diagnosis of diabetes.
In 13 practices, with 7,462 high-risk patients, those with diabetes were offered routine diabetes treatment. In 14 of the practices, with 8,282 high-risk patients, those with diabetes were offered intensive treatment. In all, 78% of the high-risk patients in the routine care and intensive care practices attended the stepwise screening, and 407 received a diagnosis of diabetes. During 104,218 person-years of observation (a mean of 5.5 years of follow-up per patient) there were 743 deaths in the screening practices and 192 deaths in the control practices.
After adjusting for the practice, age, gender, and steroid and antihypertensive drugs, the researchers found no significant differences between patients in screening practices and those in control practices in overall mortality, cardiovascular mortality, or cancer mortality. There were also no significant differences between the intensive care and routine care patients in overall, cardiovascular, or cancer mortality.
There were two positive results in the trial. In screening practices, high-risk patients who agreed to be screened had 27% lower overall mortality than did those in the control practices. Those who were offered but declined screening had an 86% higher overall mortality than did controls. But these positive results are less persuasive than the others because the researchers don't know if there were important differences between patients who chose to be screened and those who chose not to be screened.
In response to a question, Dr. Echouffo- Tcheugui acknowledged that one would not expect many deaths in 5.5 years of follow-up in newly diagnosed diabetics. He said he and his colleagues would continue to follow the patients. Dr. Echouffo-Tcheugui said he had no conflicts of interest related to his presentation.