I am continually reminded of the human tendency to ignore past errors and follow the latest shiny new thing. And sometimes we are blind to what is already working quite well.

The latest new thing in Connecticut’s health care landscape is the State Innovation Model (SIM), which has been deeply troubled from the start. But the latest bad idea from the group risks our state’s successful Medicaid program, both for the 770,000 state residents that rely on it and for our strained state budget.

No one supports intelligent, thoughtful reform more than consumers. Connecticut has received a small federal grant meant to support thoughtful reforms. All of the surrounding states are being collaborative and thoughtful with their SIM grants. Unfortunately, our state is using that grant to rush our Medicaid program into shared savings, a very new and unproven payment model.

Our state Medicaid program is just climbing out of a train wreck caused by the last shiny new, but untested payment model. Since we shifted three years ago from using capitated insurers to a model focused on care coordination, things are far better. Since the shift, per person costs in Connecticut Medicaid are actually down, saving the state hundreds of millions of tax dollars each year. Quality is up, more providers are participating, hospital admissions are down, fewer people are forced to get regular care at an ER, and more people are getting preventive care. Connecticut’s Medicaid program is busting the myth that improving care has to drive up costs; in fact, it can be the key to controlling costs.

But SIM is ignoring that progress and forcing Medicaid into shared savings, a very new and untested payment arrangement. The plan is to encourage providers to assemble into large health care networks to coordinate care, and to keep people well. In return, those large health systems share half (or more) of the resulting savings. The only problem is that the savings aren’t there. Medicare has been testing this model for three years across the country and, rather than saving money, the program has actually increased spending on health care. And Connecticut’s shared savings networks have performed even worse than the national average. If Connecticut’s health systems perform the same way in the new Medicaid experiment as they have in Medicare, the state will spend about $100 million more each year on Medicaid than if we continue with the current system that is working.

New initiatives in Medicaid are complex and planning for even small policy shifts takes years. But our state’s SIM office is pushing 200,000 Medicaid members into the untested shared savings model by July 2016 with only a few short months of planning. Just a few of the issues we haven’t yet considered include how shared savings will be fairly adjusted, who is eligible for payments, what we will require of the networks, how the networks will be monitored and governed, how resources will be devoted to quality improvement, how we will monitor for inappropriate under-service, and the impact on the rest of our state’s health system. We haven’t even touched on excluding conflicts of interest, a perennial SIM problem. The Department of Social Services and other stakeholders have been working overtime, but the scope of the task is immense.

Connecticut can’t afford to rush into a risky experiment that could significantly increase costs. The Medicaid program alone is facing $64 million in cuts announced last week and 20,000 working HUSKY parents were cut from the program in the last budget. We have accomplished a lot by engaging all stakeholders in building success and healing from past mistakes. Connecticut should be proud of our Medicaid success; we can’t risk an untested experiment driven by an arbitrary deadline.

Ellen Andrews, PhD, is the executive director of the CT Health Policy Project. Follow her on Twitter @CTHealthNotes.

DISCLAIMER: The views, opinions, positions, or strategies expressed by the author are theirs alone, and do not necessarily reflect the views, opinions, or positions of CTNewsJunkie.com.