Americans spend more per capita on medical care than just about any other country and, yet they often have little to show for it. Americans have worse access to care than people in other countries, and are often less likely to receive primary care services, like preventive therapies and screening tests. Determined to address these problems, Medicare leaders have been testing out new models of primary care, hoping to find win-win situations — reimbursement schemes that improve quality while maintaining or lowering the cost of care.

So far, many of those efforts have failed.

Near the end of 2012, Medicare began giving extra money to almost 500 primary care practices across the U.S., money the practices used to try to improve the care they offered to their patients. The goal of this Comprehensive Primary Care Initiative was to prod primary care practices to make it easier for patients to: contact providers quickly, coordinate care with other specialists, provide care management to patients with complex chronic illnesses and better engage with patients and their caregivers. The extra Medicare payments were decently sized, almost $60,000 per physician per year. The practices could use this money to hire extra nurse practitioners, or to reimburse those who were working odd hours to give patients more access to care, or other efforts.

Medicare not only gave practices these upfront payments, but also offered to give practices extra money if they reduced overall spending for their Medicare population, an incentive known as shared savings.

I am a primary care physician, and for around 20 years I worked in VA medical centers, a system that, during my time there, did a great job of coordinating care between primary care physicians and sub-specialists, and of offering care management for patients with complex illnesses. When I practiced in the VA, I often worked closely with pharmacists and nurse practitioners, for example, to address the need of patients with uncontrolled diabetes. So I am very excited that Medicare is trying to invest in and test ways of improving primary care.

Medicare administrators hoped that better primary care would lead to lower costs. Coordinating care with specialists, for example, should reduce unnecessary testing. Better care management should reduce the need for hospital care.

Unfortunately, after two years, the initiative looks like a failure. Spending didn’t decline at all:

ER visits? Unchanged.

Hospitalization rates? Nada.

Chance of seeing a provider within two weeks of leaving the hospital? Not. One. Iota. Better.

There was a small decline in the number of primary care office visits, presumably because people could more easily access telephone assistance or their patient portal. There was also a tiny improvement in the care of people with diabetes, but we are really squeezing the juice out of this lemon at this point. After two years of the program, it doesn’t look like the cost or quality of care changed much at all.

Why the failure? No one knows, but there are a few possibilities worth considering.

1. The upfront incentive was too small. My mom always told me, “Every $60,000 helps, Peter.” But expanding patient access to primary care, and giving sick patients care coordination, is expensive.

2. The shared savings incentive was insanely small. Medicare refused to financially reward practices that saved money on their patients unless, get this, there entire region of the country also reduced how much it spent on Medicare patients. That’s like telling your kid: “I will only praise you for getting A’s if all your classmates get A’s too.” Crazy!

3. EHRs don’t talk to each other well. It is hard to coordinate with specialists when, say, the cardiology notes don’t show up in the primary care clinics’ record system.

4. Change takes time. Perhaps these primary care practices need more time to get better at these new initiatives. Maybe in the first couple of years, they have flailed away at their care management activities or other activities. Maybe the physicians need time to figure out how to better coordinate care with nurse practitioners and specialists. If that’s so, than two years from now, we should see a change in spending or quality, because Medicare plans to continue the experiment for that duration.

So far, this experiment is largely a failure. But you know what would have been worse? Not running this experiment at all! That’s the thing about science — it’s unpredictable. Researchers run experiments because they don’t know how things will turn out. Medicare should be lauded for trying to improve primary care. And while they wait for the final results of this particular initiative, they should continue running other such experiments.

If we want to know how to improve primary care in the United States, we have to keep testing promising ideas.

Peter Ubel is a physician and behavioral scientist who blogs at his self-titled site, Peter Ubel and can be reached on Twitter @PeterUbel. He is the author of Critical Decisions: How You and Your Doctor Can Make the Right Medical Choices Together. This article originally appeared in Forbes.

Image credit: Shutterstock.com