Researchers in Baltimore, Maryland, set out a few years ago to find a successful approach for managing type 2 diabetes mellitus in U.S. blacks, a group disproportionately afflicted with the disease and its complications.

The randomized interventional study followed 542 adult blacks (defined in the study as African American) with type 2 diabetes at a community-based, managed care organization affiliated with Johns Hopkins University in Baltimore. Patients in a "minimal intervention" group were given a phone call every six months to remind them to come in for preventive care, as well as diabetes-specific information in the mail; a report on their use of health care services was sent to their primary care providers. An "intensive intervention" group was scheduled to visit a nurse case manager at least once a year and to receive at least three visits at home per year from a community health worker. The community health workers were trained in the use of culturally tailored "intervention action plans" developed "to address traditional cardiovascular risk factors" and "nontraditional obstacles to optimal [diabetes] care and self-management," such as depression, poverty, and "household problems interfering with medication adherence." Visits focused on health education, nutrition, foot care, medication adherence, and assistance accessing the health care system. The community health workers performed home-based disease assessments, and the use of interventions was based on evidence-based clinical algorithms.

At two years" follow-up, the patients receiving intensive interventions had 23% fewer ED visits, and those with a higher number of interventions and visits at home and with the nurse case manager had 34% fewer ED visits, than those in the minimal intervention group. Hospitalization rates decreased in the intensive group by 9% overall and by 14% among those with more visits, although those differences weren't statistically significant. At three years, patients with more community health worker visits had 47% fewer ED trips and a 56% lower rate of hospitalizations (both results were statistically significant). To the authors' surprise, the number of visits to the nurse case manager was not as strongly associated with improvement as were the home visits from the community health workers. The authors write that it may be because the community health workers were "part of [the] participants' cultural group and . . . received extensive training to enhance communication of shared experiences and personalization of interventions in their roles as educators and problem solvers."

Clinical results were less dramatic, showing an increase in high-density lipoprotein cholesterol of 1.2 mg/dL and a reduction in diastolic blood pressure of 3.5 mmHg. Study coauthor Martha N. Hill, dean of the Johns Hopkins University School of Nursing and professor of nursing, medicine, and public health at the Johns Hopkins Bloomberg School of Public Health, said that the clinical outcomes were less impressive because the patients had numerous comorbidities and were already receiving decent care from their physicians. But the interventions, particularly those offered by the community health workers, were effective and gave patients someone to call instead of visiting an ED.

"The study created an alternative way to get problems addressed," she said. "Patients could see the nurse the next day instead of waiting long periods for physician visits or going to the ED."

Tammy Worth