Strategies to Curb the Epidemic of Diabetes and Obesity in Primary Care Settings

Type 2 diabetes affects over 10 million Americans, accounts for the bulk of new cases of kidney failure, blindness, and lower extremity amputations, and consumes over $100 billion per year in direct medical costs. Obesity, the single strongest risk factor for type 2 diabetes, has also reached epidemic proportions, with rates of overweight and obesity (body mass index of 25 kg/m²or more) and obesity (30 kg/m² or more) soaring over 60% and 30%, respectively, in recent national surveys.¹ Strategies to combat these twin epidemics are urgently needed. Six papers in this issue of Journal of General Internal Medicine suggest specific lines of attack, all relevant to primary care.

Two papers identify psychiatric disorders as novel aggravating factors: depression, via its effects on cardiovascular risk factors²; and schizophrenia, via the newer psychotropic medications used to treat it.³

Two papers explore improved means of screening. For primary prevention, Edelman et al. present an approach to screening for persons at high risk for diabetes in the general population.⁴ For secondary prevention, Hayashino et al. present an approach to evaluate the cost-effectiveness of screening for persons at high risk for coronary heart disease in adults who already have diabetes.⁵

One paper focuses squarely on obesity and attempts to determine the value placed by patients on modest weight loss—a key bit of information for primary care physicians looking for psychological leverage as they consider promoting healthier lifestyles.⁶

Each of the aforementioned papers points to a new line of attack. Attack diabetes and its complications by attacking depression, by modifying drug therapy, by finding it earlier, by rooting out coronary heart disease, or by promoting weight loss in primary care settings. Each approach sounds promising and plausible; however, evidence that will enable these factors to be immediately incorporated into practice is less than definitive.

In contrast, large-scale randomized trials such as the Diabetes Control and Complications Trial⁷ and the United Kingdom Prospective Diabetes Study^8,9 have shown that tight control of hemoglobin A1c and blood pressure can substantially reduce the risk of diabetes-related complications. How these results should be translated into clinical practice, however, remains uncertain. In the absence of proven translational strategies, an alarming gap has remained between diabetes care guidelines and actual diabetes care.¹⁰ This gap has sparked interest in translational diabetes research,^11–13 typified by a recent conference sponsored by the National Institute of Diabetes and Digestive and Kidney Diseases on the design and conduct of translational diabetes research.¹⁴

In this issue, the practical randomized trial by Glasgow et al. introduces strategies to improve diabetes care by using a computer-assisted patient-centered intervention conducted among primary practices and a large insurer in Colorado.¹⁵ Interventions included a patient self-report of how long it had been since several items from the National Committee for Quality Assurance and American Diabetes Association Diabetes Physician Recognition Program measures had been assessed: checking blood pressure, measuring serum cholesterol, examining the feet, testing for microalbuminuria, and having a dilated eye exam. Several patient-centered outcomes that involved counseling or assistance for the patient with lifestyle were also assessed (setting a self-management goal, receiving nutrition therapy, self-monitoring of glucose, and patient satisfaction), along with an interactive computer session where patients received feedback about their health behaviors and strategies to implement their self-management action plan. Computerized printouts resulting from the session were given to the physician, the patient, and to a clinic care manager (nurse or medical assistant) to assist with implementation of the action plan. After 6 months, the study showed that the intervention significantly improved the number of recommended laboratory assays and patient-centered aspects of diabetes care patients received.

The trial's results were analyzed within the “RE-AIM” framework (reach, effectiveness, adoption, and implementation), which provides a good summary of its strengths and limitations.¹⁶ The trial had good reach; of the 886 patients, 74.6% of those eligible participated. Effectiveness of the trial was demonstrated by improvements in process outcomes. Adoption by physicians was challenging, as only 4.9% of the total sample participated. Nonetheless, the project was consistently implemented, as 92% discussed the computer printouts with their physician. An important logistical advantage of the intervention was its brevity (30 minutes)—short enough to conduct as an adjunct to a routine office visit. It also used computer-assisted technology to facilitate use of the program and minimize the burden on clinic staff.

Future trials should incorporate measured clinical parameters (i.e., hemoglobin A1c and serum lipid levels) as part of standard trial outcomes. Moreover, given the significant disparities that are present in diabetes and its complications for ethnic minorities compared to whites,¹⁷ programs are needed that include these populations and implement culturally tailored interventions with the same scientific rigor and practicality. Several studies have developed diabetes interventions for African-American and Latino populations,^18–22 yet more are necessary to curb the epidemic of type 2 diabetes.