Telephonic Nurse-Led Disease Management Saves Money .... Again

They did it again!

Before too long, the Disease Management Care Blog will start getting bored with the accumulating evidence (for example, here, here and here) that telephonic disease management (DM) reduces medical claims expense in excess of program costs.  There's only one thing that's more tiring, and that is the lingering ideological assumption among many academics that the Medicare program's inability to save money with DM means there's a problem with DM and not with Medicare.

Enter this study on diabetes telephonic care management that was just published in the American Journal of Managed Care.  This was about as real world as you can get.

The Denver Health's Westside Family Health Center, a federally funded community health center, randomly assigned 381 patients with diabetes to a 20-month telephone outreach program and another 381 to usual physician care.  In the intervention arm, patients not only saw their physicians, but had the involvement of 3 nurses sharing a ".75 full time equivalent" position.  The nurses used motivational interviewing and clinical protocols to guide these patients in self-care.  The nurses also independently ordered labs, initiated and titrated medications (with the use of pre-signed prescriptions as necessary), arranged vaccinations and facilitated physician appointments.  The nurses focused on control of lipids, blood glucose control and treatment of high blood pressure.

 Since many of these patients had no insurance, linear regression models were used to evaluate costs "from the perspective of the health care system" with "cost to charge ratios."  Patients with no primary care visit in the year preceding the end of the study were excluded from the analysis and the use of health care services outside of the Center was not counted. 

Aside from a slight gender imbalance (64% vs. 57% female in the intervention and control groups, respectively) as well as different rates of cerebrovascular disease (14% vs. 9%) and use of insulin (24% vs. 38%), the two patient populations were mostly similar in multiple baseline labs and co-morbid conditions.  In terms of the baseline costs, the control group had higher baseline hospitalization rates and total costs.

At the end of the study, 58% vs. 47% had achieved a target cholesterol LDL, but there was no difference in blood glucose or blood pressure control. 65 patients were lost to follow-up. The remainder were contacted an average of three times over the course of the study.

There was a significant drop in costs in the intervention group.  Over 18 months, their costs decreased from a baseline of $6390 per patient to $6217, while the control group went up from $7694 to $9033.
When the cost of the program was rolled in, the intervention resulted in $60,902 in total savings, which included the costs of $134,750 for the nurse-based intervention.  The savings appeared to be due to fewer hospitalizations.

While this study didn't use remotely positioned telephonic disease management and instead used nurses that were on-site, the DMCB points out that many telephonic programs use the same playbook. If the nurses have a good working relationship with the clinic physicians, they can do great things by telephone.  Last but not least, while the nurses pursued lipids, blood sugar and blood pressure, the DMCB suspects the real secret sauce was the countless and immeasurable other pieces of advice and help that were provided to the intervention patients.