Doctors who share patients may provide lower cost care

Patients with diabetes or congestive heart failure who receive care from doctors with high levels of patient overlap have lower total health care costs and lower rates of hospitalization, according to a new study in the Journal of General Internal Medicine.

The study represents a first step in measuring relationships among providers, which may facilitate better coordination of care as patients move through the health care system.

Previous research has viewed having multiple doctors for one patient as a risk factor for poor care coordination, overlooking the potential for beneficial interactions among providers, suggest the researchers.

“It’s not just how many different doctors you see, but which doctors you see and how well the doctors coordinate care with one another,” says lead author Craig Pollack, M.D., MHS, of the Johns Hopkins School of Medicine. “The way we looked at that was by measuring how often doctors shared patients with one another.”

The study is based on insurance claims data for more than 60,000 patients with either diabetes or congestive heart failure (CHF). Using innovative techniques adapted from social network analysis, the authors looked at claims from all providers for each patient to determine what they called “care density,” a measure of patient-sharing among doctors. For patients with CHF, high care density was associated with significantly lower inpatient costs than patients with lower care density. Patients with diabetes and high care density had lower inpatient and outpatient costs.

“Using network science to understand more about these complex interactions in patient care is a new and very exciting field of research,” says Bruce Landon, M.D., MBA, of Harvard Medical School’s Department of Health Care Policy. Understanding the interconnections among health care providers may also present opportunities to speed up diffusion of medical innovations across networks, he adds.

The authors note that their approach does not analyze how continuity of care may change over time, nor were they able to examine how patient sharing varies according to patient race/ethnicity or socioeconomic status.

“It’s our hope that these types of metrics can one day be used to identify patients at risk for poorly coordinated care,” says Pollack. “This study is an important step towards that goal.”