Paying doctors to better coordinate care for Medicare beneficiaries in Arkansas and seven other states helped hold down the cost of patients' medical care over a two-year period, although the savings didn't fully offset the cost of the extra payments, a report found.
Still, the authors of the report by Mathematica Policy Research said the Comprehensive Primary Care Initiative's effect on medical expenses was bigger than they expected.
"We did not expect to find effects of [comprehensive primary care] in the first two years," the authors wrote.
Released this month, the report is the second annual evaluation of the pilot project, which began in October 2012 and is to close out at the end of this year.
The program is led by the federal Centers for Medicare and Medicaid Services' Innovation Center, which was created by the 2010 Patient Protection and Affordable Care Act.
Under the pilot project, Medicare, the federal health insurance program for the elderly and disabled, makes upfront payments to physician practices that agree to take steps such as developing care plans for patients with chronic health problems and providing around-the-clock access to a medical professional.
The payments averaged $20 per patient per month for the first two years of the program and fell to an average of $15 per patient for the final two years.
The payments are higher for patients who are older or have a history of medical problems and lower for younger, healthier patients.
In Arkansas, the Medicaid program also contributes upfront payments along with Arkansas Blue Cross and Blue Shield, QualChoice Health Insurance and Humana.
Several employers, including Wal-Mart Stores Inc., also make payments on behalf of their employees. The health plans for public school and state employees also make payments.
The money goes to 58 clinics in the state on behalf of about 338,000 patients, including about 93,000 who are enrolled in Medicare or Medicaid.
In Crossett, the Family Clinic of Ashley County has used the payments to assign two nurses to coordinate patient care, including one who handles those duties full time.
The nurses develop care plans for patients with chronic health conditions such as diabetes or high blood pressure and check on them regularly, Ben Walsh, one of the clinic's four physicians, said.
The nurses also contact patients who visit the emergency room or are discharged from a hospital and arrange for them to visit the clinic within two weeks.
Those visits help fill information gaps that were created as doctors employed by hospitals began taking over duties once performed by primary-care doctors, Walsh said.
"If you go back five years ago, we were seeing our own patients in the hospital, so when they came out of the hospital and came to see us, there was no transition, because we knew what we did, and we knew what medicines they came out on," Walsh said.
The pilot project also includes practices in Colorado, New Jersey, Oregon and parts of New York, Ohio, Kentucky and Oklahoma.
In Arkansas, 179 clinics are enrolled in a similar program started by the state Medicaid program in 2014 with help from a $42 million grant from the innovation center. The Medicaid program contributes payments associated with 330,000 patients under that program, known as the state's patient-centered medical home initiative.
The evaluation report compared practices in the federal pilot program in all eight states to similar practices that are not participating. It found that the medical costs were about 1 percent lower for Medicare beneficiaries in the pilot project -- a difference of about $11 per patient, per month.
Lower costs for hospitalizations and nursing-home expenses accounted for much of the difference, the report found.
Medicare beneficiaries whose doctors are enrolled in the pilot project also visited the clinics less frequently, indicating that phone calls and emails may have replaced the need for visits in some instances.
The estimates "suggest a near certainty that Medicare [fee-for-service] expenditures have been reduced relative to what they would have been in the absence of [comprehensive primary care], but only a 4 percent likelihood that those reductions exceed" the care coordination fees, the authors wrote.
In Arkansas, Medicare beneficiaries' medical care costs were slightly higher than those of patients in a comparison group, although the difference wasn't considered to be statistically significant.
The U.S. Department of Health and Human Services' Centers for Medicare and Medicaid Services announced a similar finding in October.
Comparing the practices' costs in 2014 with projections based on the growth of spending the previous year, the agency found that the practices had reduced monthly costs by about $6.13 per Medicare beneficiary before accounting for the care-management fees, which averaged $20 per patient, per month in each region.
In contrast, the savings generated during the first year of Arkansas' Medicaid-led program were enough to offset the program's fees, state officials said.
According to a report by the Arkansas Center for Health Improvement, the Medicaid program paid $12.4 million in care-coordination fees, which average $4 per patient, per month, in 2014. The same year, the medical costs for Medicaid recipients assigned to the practices came in $34.3 million below the target.
Per-patient costs for the practices fell by 1.2 percent. The state Department of Human Services had set a target increase of 2.6 percent based on increases in previous years. For practices that were not enrolled in the program, average patient costs increased 0.6 percent in 2014, according to the Center for Health Improvement report.
The state Department of Human Services announced in October that 19 of the provider groups had per-patient costs in 2014 low enough to qualify for bonuses totaling $5.3 million.
Joe Thompson, director of the Center for Health Improvement, noted that the Medicaid program allowed practices serving at least 5,000 Medicaid patients to qualify for bonuses based on their average costs.
That may have provided a greater incentive to lower costs than the Medicare pilot program did, he said. In the pilot program, bonuses were based on the costs for the entire region.
Thompson also pointed to findings from the evaluation indicating the pilot project has produced good results. For instance, from 2013 to 2014, slightly more Medicare beneficiaries whose doctors are in the pilot project reported discussing their health goals with a clinic staff member, being able to quickly schedule appointments when needed and having input in their treatment.
Meanwhile, the comparison group's scores in those areas fell slightly.
Compared with the comparison group, practices in the pilot project also showed improvement in the number of Medicare beneficiaries with diabetes who received at least one of four types of tests to monitor their condition.
The pilot project is an example of "the health care system moving in the right direction," Thompson said.
The program also has helped Arkansas Blue Cross and Blue Shield's customers get better care, company spokesman Max Greenwood.
Customers whose doctors are in the program visit the emergency room less frequently, use cheaper generic drugs instead of brand-name drugs more often and have fewer hospitalizations and readmissions, she said.
The results have been similar for patients whose doctors are enrolled in the Medicaid-led program, she said.
William Golden, the state Medicaid program's medical director, said the program is "reinvesting in primary care, which is frankly something that is long overdue."
Walsh said he's reserving judgment. The approach "sounds good," but hasn't yet been fully tested, he said.
"I think it's just going to take time to see, well, was it worth it? Did it save money? Did it save lives?" Walsh said. "I don't think we know the answer to that yet."
Metro on 04/24/2016