Call center–based programs put into question
A telephonic DM program for HF patients in South Central Texas was not cost-effective, but the study’s authors say the program could have saved money if it had been targeted to the right patients.
Published in the February Academy Journal of Managed Care, the study, “Cost-Effectiveness of Telephonic Disease Management in Heart Failure,” was “one of the first studies to assess the cost-effectiveness of DM in HF in a large sample with a follow-up period exceeding 12 months,” wrote the authors.
The authors concluded that “the intervention was effective, but costly to implement and did not reduce utilization. It may not be cost-effective in other broadly representative samples of patients.”
“I think in our sample, we learned that careful targeting is important,” says Brad Smith, PhD, senior analyst at the Altarum Institute in San Antonio, who coauthored the study. “You really want to pick the patients who most likely will benefit based on the evidence. From our study, that would suggest the patients who are the sickest.”
Conducted by the University of Texas Health Science Center in San Antonio and funded through the U.S. Department of Defense, the randomized controlled trial’s disappointing results mirrored other reviews of call center–based HF programs. One such study, “Effect of Moderate or Intensive Disease Management Program on Outcome in Patients With Heart Failure,” published in the February 11 Archives of Internal Medicine, failed to show that a DM intervention reduced mortality or HF rehospitalizations.
In the Texas study, the authors evaluated the cost-effectiveness of a telephonic DM intervention with HF patients. The randomized controlled trial of 1,069 community-dwelling patients in South Central Texas focused on patients with systolic and diastolic HF. The enrollment period was 18 months per subject, and stretched from 1999–2003. The study randomized participants into one of three study groups: usual care, DM, and augmented DM. Those in the intervention arms were assigned a disease manager, an RN who educated the patient and provided medication management with the patient’s PCP. In addition, patients in the augmented group were given in-home devices for enhanced self-monitoring, including those for measuring blood pressure.
Because there were no differences in outcomes between the DM and augmented DM group, when performing the cost analysis, researchers pooled the two groups into one intervention group.
To gauge costs, the study authors created a list of utilization events, including patient self-report data, reviews of electronic hospital records, and physician and clinical medical records. They also analyzed the cost of care for all causes and took into account intervention costs, which were based on fees paid to the DM subcontractor. The researchers did not calculate work-related costs, such as absenteeism and presenteeism, because participants were largely of retirement age.
The study found that the intervention failed to save money, and the difference in total costs observed among the sample was almost entirely attributable to the costs of the intervention.
It also found that DM showed “statistically significant survival advantages among all patients,” although “analyses of direct medical and intervention costs showed no cost savings associated with the intervention.”
The findings do not surprise Randall Williams, MD, CEO of Pharos Innovations in Northfield, IL, which sells technology that helps payers and providers manage their chronic disease populations. Pharos is involved with two physician groups totaling about 40,000 Medicare beneficiaries in CMS’ Physician Group Practice (PGP) demonstration project.
Pharos’ Tel-Assurance system enrolls HF patients in telephony and Web-based daily interaction, which requires patients to log in or call each day and report their test results, behavior, and medication compliance. The system captures the data, and the algorithms identify which patients need a care management nurse, either in the physician’s or payer’s office, depending on the client.
Through Pharos’ program, Park Nicollet Health Services in St. Louis Park, MN, and Billings (MT) Clinic averted one HF admission per year per enrollee in the first two years of the PGP. Pharos’ Tel-Assurance program enjoys a 3:1 ROI, Williams says.
He explains that the problem with the call center–based DM program in the Texas study is that it didn’t reduce hospitalizations enough to outweigh the program’s costs and didn’t allow frequent enough communications with patients.
Williams says DM is successful with HF patients when the program is designed properly.
“I wasn’t surprised that it didn’t demonstrate hospitalization reduction. That said, there are several well-done analyses where they do and where the model is different, which I think is the explanation [for the study’s negative results],” he says.
Williams says the keys to creating a cost-effective HF DM program that averts hospitalizations are:
Provider-level connection with the patient, whether through a nurse, pharmacist, or physician.
Frequent patient intervention. The call center model of contacting patients every month or two is not effective because it doesn’t track patients between calls. On the other hand, technology is allowing DM programs to track patients daily.
A low enough intervention unit cost that allows for a positive ROI.
Williams likened the program costs for the Texas project to CMS’ Medicare Health Support demonstration project, which CMS is ending this year because it allegedly has not been cost-effective. “I think the model is wrong,” he says.
Both programs also had per-patient-per-month charges of more than $150. If a program is created with that kind of cost, Williams says the program must avert many hospital days to see a benefit. “You can avert hospitalizations, but at what cost?” he says.
The DM programs have created a parallel process to the physician’s office, says Emad Rizk, MD, president of McKesson Health Solutions in Broomfield, CO, who agrees with the study that targeted interventions are important. Companies must focus on the most costly HF patients, he says.
DM should learn from the study and create leaner programs that utilize technology, Williams says.
“I think it should tell DM that episodic interactions with heart failure patients are not frequent enough and they need to figure out how to do that at or below the cost structure they are working at,” he says.
Smith says his study differs from others that have been published in academic HF journals (and those conducted, but not published, by DM companies) because the earlier studies focused on HF patients recently discharged from the hospital. This created a regression to the mean, skewing the data, because those who are tested when they are sickest will most likely improve during the study, he says.
Another problem is the duration of many studies. Reviewing HF patients’ clinical improvements and healthcare costs for a few months does not provide a review that is long enough, says Smith.
A third problem is that many of the other studies did not provide a control group, so they were, in fact, demonstration projects rather than studies.
Smith says a benefit to his study is that the program encompassed community-dwelling patients rather than those in facilities. Although there are limitations to strictly focusing on HF patients in South Central Texas, Smith says having community-dwelling patients gave researchers a more representative picture of HF patients.