As we move through the third year of the pandemic, the specter of long COVID or post-COVID conditions has emerged as a matter of great concern, especially since so much is still unknown. The Centers for Disease Control and Prevention estimates that the disease—characterized by an array of symptoms involving multiple organ systems and persisting at least one month post-infection—is affecting 1 in 5 adults, or up to 23 million people.

While the federal response has coalesced to a greater degree over the last few months, the U.S. has a long way to go in terms of developing uniform guidelines to diagnose, manage and track long COVID, and building structural supports for those suffering from it. Encouragingly, the Biden administration has accelerated enrollment of 40,000 individuals into the landmark National Institutes of Health study of long COVID, Researching COVID to Enhance Recovery (RECOVER), which was launched last year. The administration has also pledged to focus its efforts on underserved communities most affected by acute and long COVID. A good place to start for NIH would be to actively recruit for RECOVER at community health centers nationwide.

Community health centers serve low-income patients, two-thirds of whom are people of color. Because these groups have been disproportionately impacted by COVID, they are also more likely to suffer from post-COVID conditions. In its efforts to assist health centers to identify and treat patients who are by definition under-resourced, the National Association of Community Health Centers has accumulated some valuable lessons and recommendations on how best to centralize and coordinate a holistic response.

The first step in tackling any problem is to measure its size and identify who is most affected. This has been difficult with long COVID because comprehensive data about its prevalence—not to mention its characteristics and clinical course—is currently lacking. Primary-care providers are not incentivized to retrieve and funnel data on this disease, which could be corrected through public policy measures. To capture an accurate count and distribution of long COVID patients, the federal government could mandate provider reporting through its national surveillance and reporting system, currently used to track certain communicable infections.

In order to report and accurately count cases, diagnostic criteria must be clarified and reporting should be standardized. Because this is an emerging condition and has such variable presentation, many providers are not clear on what constitutes a formal case. How and when should it be coded in the electronic health record? What are the clinical protocols for management? As researchers work to answer these questions, information should be broadly disseminated and easy for clinicians and patients to access, without having to perpetually mine government sources such as the CDC website or search for emerging studies in academic publications.

To even identify long COVID in the first place can require significant effort, especially among patients with sporadic access to care. Patients may seek care for the first time long after chronic symptoms have developed, and may not have confirmed they ever had COVID in the first place. Once we’ve confirmed a past infection, other diseases with similar symptoms must be ruled out. Then, because long COVID manifests differently in each patient and no widely beneficial treatment yet exists, treatments must be tailored and thus can vary widely. When specialty care or therapies are needed, access to expert care teams for post-COVID conditions is currently limited to academic medical centers with busy primary-care clinics that generally require insurance.

Given the multilevel barriers to care for underserved patients and the increased use of at-home tests, additional investment in the “test to treat” model is essential to identify and track cases and to provide timely treatment and follow-up.

Models for long COVID care must be tested and built. The Department of Health and Human Services could sponsor demonstration projects to show what these might look like. Trained, multidisciplinary providers armed with appropriate knowledge of promising therapies—such as pulmonary rehab and specialized cognitive behavioral therapy for brain fog—could work in and with health centers to fill these gaps and treat patients in communities hardest hit.

Public health messaging should be revamped and reactivated so patients with long COVID are aware of available resources and how to “show up” for care in a timely and proactive way.

Clear coding and reimbursement guidelines for long COVID must be established, and reimbursement itself must be fair and adequate for a range of long COVID-related conditions treated by multidisciplinary providers, including behavioral health. This is particularly important for safety-net providers such as community health centers, which serve a large share of publicly insured patients whose care is reimbursed at lower rates. Additional funding should be allocated to health centers and other safety-net providers to ensure they can continue treating uninsured patients struggling with long COVID without exhausting limited resources.