As the COVID-19 pandemic ravages communities, the federal agency responsible for outbreak control, the Centers for Disease Control and Prevention, is conspicuously absent from leading the U.S. response. There's now a cottage industry of plans for opening up America in the midst of the pandemic. But any plan that will keep us safe in the long run will have to include reforming the CDC.

The men and women at the CDC are some of the most well-trained and dedicated civil servants in the country. But even those of us who have been cheerleaders for the CDC's role in keeping America and the world safe recognize the need for fixing this venerable institution.

It is reasonable to ask why the CDC was unable to quickly develop and scale up testing for SARS-CoV-2. Why was there no backup plan? How come SARS-CoV-2 circulation in many parts of the United States remained undetected by the CDC's early warning systems for at least several weeks after January? Why did the CDC not take the lead in developing, or at least systematically collating, projection models for the pandemic's spread, which would have decreased confusion among decisionmakers and the public?

If we want to avoid asking similar questions again the next time a new virus appears, the CDC needs some changes.

Over the years, there has been a substantial expansion in the types of activities the CDC is involved with. The federal agency whose original name was the Communicable Disease Center now deals with issues ranging from birth defects to injury prevention. Critics have charged that this expanded focus has resulted in dilution of the CDC's outbreak response mission. The CDC should not ignore other health issues such as obesity and noncommunicable diseases. But it cannot afford to lose focus as the nation's insurance policy against infectious disease threats.

Reform efforts should focus on increasing the CDC's infectious disease laboratory capacity and enhancing its expertise in scaling-up testing during public health emergencies. One of the most effective ways to get ahead of outbreaks is to have a highly sensitive epidemiological surveillance system. The CDC currently maintains several surveillance systems, but they were of limited utility as early warning systems for COVID-19. The CDC will have to reassess the data it acquires, how fast these data get transmitted and what analytical tools it uses to detect signals. The CDC will also have to modernize the types of data it routinely uses. For example, rapidly sequencing and analyzing genomes of circulating viruses can provide important information, such as where the virus was imported from. To better prepare for fast-spreading outbreaks, the CDC will have to expand the use of genomic epidemiology and other modern tools for surveillance.

The CDC also badly needs more money, and Congress has to change the way it funds the agency. Public health investments yield very high returns: For every dollar spent on prevention, there is a five times return on investment within five years. Despite this, Congress has tried to fund public health on the cheap. As a result, the CDC has been chronically underresourced. For example, CDC's Public Health Emergency Preparedness (PHEP) program that supports states and local areas in their preparations for pandemics and other emergencies has had its funding shrunk from $940 million in 2002 to $675 million in 2020. PHEP funds can be used to develop laboratory and contact tracing capacity — which could have come in handy in this pandemic.

The CDC's programs are micromanaged by Congress through detailed line-item budgeting, which means lawmakers and their aides have undue influence over CDC priorities. A stroll through the CDC campus can be illustrative: You arrive at the (Rep. Edward R.) Roybal campus to check in at the (Sen.) Tom Harkin communication center to walk to the (Sen.) Arlen Specter Emergency Operations Center. The instinct to appease political leadership was criticized by the National Academy of Sciences during another infectious disease emergency — the smallpox bioterrorism threat in 2005. A nimble, evidence-driven CDC would require flexibility in resource allocation based on scientific acumen and experience of public health professionals, rather than ideological leanings of vested interests.

The Epidemiological Intelligence Service (EIS), CDC's flagship training program for its staff, needs to be modernized. This postdoctoral program trains CDC scientists in epidemiologic field work through a combination of classroom training and experiential learning. Career CDC leaders usually come from this cadre of staff; it is rare for a non-EIS trained CDC staff to rise to the senior echelons.

EIS officers are trained in conventional methods for investigating and responding to outbreaks, such as fast-paced studies that compare exposures among those with illness labeled as "cases" and those without disease labeled as "controls."

But the science of disease control has evolved substantially and now includes tools such as advanced mathematical modeling, genomic epidemiology, and high-end laboratory methods. The CDC does employ scientists with expertise in these and other emerging subfields. These skills are increasingly so seminal to modern disease control approaches, though, that they should be a major part of the core EIS training requirements. Moreover, EIS officers should be well-versed in the science of behavioral interventions — designing a smart social-distancing strategy is as much about human behavior as it is about biological characteristics of the virus.

Just over a year ago in a U.S. Senate committee hearing, I was asked about the value of the CDC. Without hesitation, I described the CDC as a national treasure. I continue to believe this. And it is precisely because I have such faith in the abilities of the men and women of the CDC that I know we must reform this American institution of global significance. We need it to be as excellent as it can be.

Saad B. Omer is director of the Yale Institute for Global Health and a professor at the Yale University schools of medicine and public health. He wrote this article for the Washington Post.