The human cost when policy implementation lacks human-centered design

Rebecca Bruno
April 20, 2022

As the world appears to be turning a corner with the COVID-19 pandemic (fingers crossed) and things start to return to “normal” (whatever that means now), millions of Medicaid beneficiaries may soon lose the coverage that has sustained them over the past few years. Under the COVID-19 Public Health Emergency (PHE) that went into effect on January 31, 2020¹, and subsequent legislation passed by Congress, additional flexibility and federal funds were available to states, allowing more people access to Medicaid coverage and benefits. To receive the additional federal funds, states are not allowed to disenroll anyone who was enrolled in the Medicaid program as of or after March 18, 2020 — otherwise known as continuous enrollment during the PHE.

Once the COVID-19 PHE ends, the additional Medicaid funding and flexibilities will start to end. This process of reverting back to previous funding levels and policies is being referred to as “COVID unwinding,” and will be a tremendous effort for states and the Centers for Medicare and Medicaid Services (CMS). States will have a significant amount of paperwork and reporting to submit to CMS through various systems, tools, and processes. CMS will have to review and adjudicate all these state submissions. In addition, states must conduct outreach, send notices to beneficiaries, and update IT systems to reverse flexibilities implemented during the PHE.

The Human Cost

The biggest risk to current Medicaid beneficiaries is that when the continuous enrollment requirement ends, they may lose coverage. Some beneficiaries enrolled during the PHE will no longer be eligible for Medicaid because of changes in income or household size, or they moved, for example. However, many beneficiaries will still meet the eligibility criteria but are at risk of losing coverage because of administrative hurdles. Pre-COVID, many eligible individuals fell through the cracks and lost coverage because they did not receive mail, didn’t have access to requested documents, or did not have enough time to respond to requests for documents (usually 10 days). This risk will be amplified when the PHE ends, and millions of beneficiaries go through the redetermination process.

Without thoughtful consideration of how to design and implement system changes and processes, as well as craft clear communications that have input from all stakeholders, including beneficiaries and providers, the potential harm to individuals from COVID unwinding could be significant.

To illustrate the potential harm when Human-Centered Design (HCD) is not applied, I’ll use a story about another major initiative I worked on that had a major impact on access to health coverage — the Affordable Care Act (ACA).

Noble intentions

As CMCS and states prepare for COVID unwinding, I am reminded of my time working at CMCS during the ACA implementation, which was the largest change to the Medicaid program since its creation in 1965. I was tasked with drafting regulations to implement ACA requirements related to the Medicaid enrollment process. This was no small task! We had a team full of dedicated individuals who cared deeply about creating policy that would make it easier and faster for eligible individuals to enroll in Medicaid and access needed benefits.

It took years of thoughtful consideration to craft the regulations. We had to draft requirements that would simultaneously reduce burden for individuals applying for Medicaid and minimize burden on states. We also had to balance making it easier to access Medicaid coverage while maintaining program integrity (e.g., preventing fraud, waste, abuse) and fiscal responsibility. Every word we wrote had implications for people who need health care. You wouldn’t think that whether you used “and” or “or” would be such a big deal, but it could drastically change the meaning of the law.

Implementation Breakdown

After spending a few years of writing, re-writing and refining the regulations, which involved reviewing and considering thousands of public comments, they were finally in effect. Now it was time to implement all the carefully thought-out written policies. This is where the unintended consequences began.

The issues with the launch of were well-publicized. What’s not as well known is that implementing the Medicaid changes at the state level did not go so great either. State systems were not implemented correctly or on time — some online applications were not ready to go live, back-end rules were not programmed correctly, and numerous workarounds and contingencies had to be put in place. This led to application processing backlogs in a number of states which meant delayed coverage for thousands of people. There were also significant issues with many states’ ability to conduct renewals using the new ACA rules. This took years to smooth out. This was not all on the states; there were issues at the federal level that also led to these breakdowns.

As a policy person, I hadn’t fully considered the Information Technology (IT) systems and the HCD development processes needed to implement policy requirements. When I was writing the ACA-related regulations I knew how I wanted them to be implemented and what the desired outcomes were but didn’t really understand at a granular level how what I wrote would be translated into an eligibility system in a state.

As I started working more closely with the system design and development side of ACA implementation, the more I realized “Houston, we have a problem!” It didn’t take long to realize there was a big disconnect between the people who write or interpret the policy and the people that oversee system design and implementation, which ultimately had a real-life impact on people who needed health coverage.

It was this eye-opening discovery that led me to leave the government for the private sector, hoping to bridge the gap between policy creation and implementation. I knew how important it was to bring the Medicaid program insight to the design and engineering process and advocate for those who rely on the program. The design and development folks didn’t always get why I, a Medicaid policy subject matter expert (SME), would be part of the team developing tools and products, but I fought to have a seat at the table to provide that perspective. Including a dedicated SME on the team should be considered as a major asset to any design or development project in government.

The Value of Human-Centered Design During Covid Unwinding

These days I work for a company where an HCD approach is embedded in our daily work. HCD is a framework to ensure we understand the people for whom the policies, programs and services are intended. We provide strategic support to the federal government, developing products and tools to assist CMS in communicating with and supporting state Medicaid agencies. Looking back now, I see how valuable HCD would have been for the ACA implementation. With any large-scale implementation there are going to be glitches, but I firmly believe that the outcomes would have been better if HCD had been applied at the state and federal level.

The scope and impact of COVID unwinding is going to be much like ACA implementation. Though there are many aspects of the Medicaid program that will be impacted when the PHE ends (e.g., provider enrollment, home and community-based services, certain benefits), eligibility and enrollment systems and processes will require special care so that millions of people who are still eligible for Medicaid do not get disenrolled. The risk is great, as described in recent articles in the Washington Post and local news coverage.

It is crucial that designers and engineers tasked with making system or process changes keep in mind the human impact of their work: the children, families, low-income adults, elderly individuals, people with disabilities or chronic conditions, people experiencing homelessness, and individuals with limited English proficiency served by the Medicaid program. Medicaid beneficiaries may not be first-hand users of all the affected products or services, but they will truly feel the end results.

It will be a challenge to bring HCD approaches into the unwinding process due to lack of clarity around when it will actually begin, the limited time (6–12 months) to complete, staffing constraints, and many other considerations. CMS and states will need to determine the level of HCD work they can conduct to meet the unwinding timelines, including conducting research with stakeholders and testing designs and content. At the very least, design and development teams working on COVID unwinding efforts should educate themselves about the Medicaid program, relevant stakeholders and users, directly and indirectly affected beneficiaries, and do as much as possible to mitigate unintended consequences.


[1] The COVID-19 PHE has been renewed 8 times since January 2020.

Want to keep reading?

What we offer

We are dedicated to creating simple, sustainable solutions. Learn more about our Services.

Our Services
Get to know the people of A1M

Hi, I'm


Britta is a product manager with a background in technical writing and community management.

Meet Our Team