Mikaela Coady, a physician assistant with Priority Urgent Care of Ellington, gives Jane McCarthy, of Manchester, a vaccination record card before giving her the first dose of the Moderna COVID-19 vaccine at a clinic at St. Bernard Church in Vernon Thursday, Jan. 28. The one-day clinic was set up by Priority Urgent Care in partnership with the town of Vernon in an effort to reach out to the community to make it easier for seniors age 75 and over to get the vaccine. They dispensed 100 doses. Credit: Cloe Poisson / CTMirror.org

At press conference after press conference, Gov. Ned Lamont has set the stage how he wants Connecticut to be evaluated on its COVID-19 vaccine rollout by touting the percentage of people vaccinated as a key measure of its success. By that metric, Connecticut has been a national leader, consistently in the top five states, according to federal data collected by the Centers for Disease Control and Prevention.

As of last week, 20% of the state’s population had received a first dose. That put Connecticut about four or five percentage points above the national rate. As of Sunday, Connecticut’s figure was 24%.

But the state has its terms for success defined backward, said Saad Omer, Yale School of Public Health epidemiologist and the director of the Yale Institute for Global Health. “That’s a process metric,” he said. “It’s not an outcome metric.”

How important is speed in the COVID-19 vaccine rollout? To Connecticut, it’s an important enough consideration to partially justify bucking CDC guidance on prioritizing people with co-morbidities, though experts suggest that it is the best way to prevent deaths in younger populations.

But by rolling out vaccine through an age-based process, the state will effectively de-prioritize younger adults with co-morbidities that put them at higher risk of dying from COVID-19, Omer said, because in those younger age groups, those with existing health issues will be part of a much larger crowd of eligible residents.

Credit: CTMirror

Omer serves on the National Academy of Medicine’s committee for the equitable allocation of COVID-19 vaccines and the World Health Organization’s global COVID-19 allocation committee. His guiding principle in making allocation recommendations has been simple: “You can focus treatments on reversing the long-term effects of COVID. You can have an economic stimulus,” he said. ” But when you have three or four bad outcomes and one of them is death, you always focus on death.”

“The governor has said from the beginning that the goals of this vaccine rollout are to prevent severe illness and death and do it equitably,” said Deidre Gifford, the Commissioner of the Department of Public of Health. “We have an obligation to both administer vaccine quickly and protect the people at highest risk, and do that in a way that balances the two the most carefully, and doesn’t disadvantage certain populations along the way.”

The original plan to prioritize those with co-morbidities would have both hampered the speed of vaccine distribution and hurt the state’s equity goals, the administration has said. It has sold its new, age-based plan to Connecticut residents in part on the premise that the rollout will be faster this way.

“We absolutely are in a race against the virus and these variants that are emerging in certain parts of the country and the world,” said Josh Geballe, the state’s chief operating officer. “The quicker we can get as many people vaccinated in the state as possible, the more protection we have against COVID-19 generally and variants in particular.”

But in a situation of limited vaccine supply, getting doses to the most vulnerable should be the first priority, Omer argued. Doses in arms should be a longer-term, secondary goal when the amount of vaccines available and being administered is sufficient to reduce transmission.

In other words: “Is vaccination an end, or a means to reduce morbidity and mortality and suffering?”

Age and mortality

The Lamont administration has argued that an age-based approach does take into account COVID mortality. Last Monday, Lamont presented a chart showing the age breakdowns of COVID-19 deaths in various 10-year age brackets, showing that the majority of deaths under 65 have been individuals aged 55-64.

But a strict age-based rollout has limitations in being able to identify those most at risk of dying of the disease among younger populations, Omer said.

Generally speaking, age is a strong factor in COVID mortality across all racial groups, though the risk of dying does not increase uniformly across racial groups with age, a CT Mirror analysis found — the death rate for Black and Hispanic COVID residents is higher at younger ages than for whites.

Omer’s recommendations for whom to prioritize for vaccinations, like the CDC’s, also factored in age in initial stages, though Omer and his colleagues recommended starting with individuals aged 65 and older — as opposed to the 75+ age bracket the CDC started with — to account for the the fact that Black and Hispanic people who die of COVID-19 are younger on average than their white counterparts. In Connecticut, the median age of death for a white person with COVID is 85, while for Black and Hispanic residents it was 76 and 73 respectively, a CT Mirror analysis of death data through November 2020 found.

Defending the state’s decisions to open up eligibility to people 55 and older on the CT Mirror’s Steady Habits podcast, Geballe pointed out that 70% of the deaths of Black and white individuals aged 16-64 occurred among individuals aged 55-64. For Hispanic residents, that number drops to 59%.

But there’s a point at which age becomes less useful in identifying at-risk individuals, as there are relatively fewer individuals who die of COVID-19 in younger age groups. Omer argues that that cutoff is under 60 — though he conceded that it might be 55. “Beyond that, focusing on co-morbidities would have been an evidence-based approach rather than jumping by age-bracket.”

Slicing the data solely by age doesn’t consider the underlying factors for why most younger individuals are dying at all. A study of a few hundred individuals aged 18-35 hospitalized in New York City found that 57% of them had at least one comorbidity; 97% of those who went on to die had at least one.

“I am very sympathetic to those people who are in younger age groups who need to wait longer,” Gifford said. “But the question is, when you’re implementing a population-wide strategy, how do you identify as a group those that are at highest risk?” Connecticut data and national data show that COVID mortality increases with age brackets at all levels, including under 55, Gifford said — though the scale of the increase is lower with younger age groups. The prevalence of some co-morbidities — such as heart disease — also increases with age, she said.

A JAMA Internal Medicine study found that 18- to 34-year-olds with multiple underlying conditions — obesity, hypertension and diabetes — had similar risk to 35- to 64-year-olds without those co-morbidities.

While Black and Hispanic residents comprise 11% and 17% of the state’s 18-64 population, they comprised 27% and 26% of COVID-19 deaths in that age bracket, according to a CT Mirror analysis of Census data and death records through November.

If the state had stuck to its original plan, younger individuals at risk of dying would have already been eligible. Now they will have to wait weeks — in the case of those under 35, until eligibility is opened upto everyone over 16.

“You’re going to have people who really need it at the back of the line,” Omer said. “By opening up so wide the eligibility groups, you’re decreasing the probability of the most highly vulnerable to get that vaccine.”

Behind the state’s U-turn

Originally, the state’s plan for the next phase would have had “between 1.1 and 1.4 million people” eligible for a vaccine, including people with co-morbidities and essential workers. That was an impossible job, Geballe said.

The main limiting factor in vaccine administration is not the state’s capacity so much as federal supply, Geballe said. In order to have staggered demand to meet supply, Geballe says the state would have had to stratify the original phase — and considered using age to do so, which would have added an additional layer of complexity.

At a press conference two weeks ago, Gifford said that an age-based allocation would reach at-risk individuals faster than the original plan. In addition, if the state had proceeded with the original plan, “we would have been down a path of vaccinating a lot of perfectly healthy 25, 30-year-olds who, statistically speaking, regardless of what race and ethnicity they are, were highly unlikely to be severely ill or die from COVID if they were infected,” Geballe said.

As the governor put it: “Keep it simple, stupid.”

Is equity served with an age-based rollout?

In privileging simplicity, the state is also defining the terms it wants to be judged on when it comes to equity.

Equity in the rollout process can be measured two ways: allocation and administration. Equity in allocation is about who can get in line for the vaccine and when. Equity in administration is whether the doses actually end up in the arms of people who have been disproportionately impacted by the pandemic. They are separate, if related, goals; one does not necessarily follow from the other. A system that is equitable on paper is not necessarily equitable in practice.

The Lamont administration is largely focused on administrative equity, not just because officials believe that it would be logistically easier, but because they believe that detailed eligibility criteria hurt equity aims in practice, they have said.

Age-based rollout “avoids the confusion and complexity that so many other states are facing, which impede equity,” Gifford said.

Who has qualifying co-morbidities?

When it came to the question of identifying people with co-morbidities, the state had two problems: defining what would count as a co-morbidity, and figuring out how they would enforce it. Lamont has expressed frustration with the CDC lists for co-morbidities, deeming them confusing, arbitrary and non-comprehensive. And when it came to  enforcing criteria, providers pushed back.

“Yes, I understand that they can’t do everyone who has a comorbidity,” said 29-year-old Will Hermann, who has muscular dystrophy and can’t wear a mask for prolonged periods. “But I do think they should try to find the people who can’t protect themselves.”

The state is painting the situation as a “false choice between two extremes,” Omer said. “The sweet spot at the moment is not to focus on the so-called essential workers but people with high risk factors at this time.”

Other states are grappling with similar problems. Maine also announced an age-based rollout last month, and has allowed providers to prioritize those with co-morbidities in appointments within each stratum of eligibility. In California, efforts to transition to an age-based rollout that didn’t consider underlying conditions were challenged by disability advocates and successfully reversed. Two federal complaints have been filed against Connecticut so far for not prioritizing people with disabilities; the second also claims the state’s policy is racially discriminatory.

The logistical complexities of considering co-morbidities hold little water with Omer.

“What were we doing in the fall as a country and as a state when the first allocation schemes came out?” he asked. “Why do we continue to react instead of being proactive and thoughtful?”

While it is difficult to identify everyone who might have a co-morbidity, people with multiple risk factors do show up in the health care system one way or another because of the severity of their symptoms and their need for hospital care, Omer said.

“While you may have inequities in identifying people with just diabetes, it’s rare to miss someone with multiple co-morbidities, because they are so severely ill that even in an imperfect health care system they get identified.”

Even if that were the case, that still disadvantages those who have lost their health insurance during the pandemic, Gifford said.

There’s a precedent for using co-morbidities to administer vaccines; the flu shot in the United States, before it was expanded to the whole population, prioritized people with co-morbidities, though coverage was admittedly poor, Omer said.

“I refuse to believe that we don’t have any muscle memory from that,” he said.

Vaccine ‘open season’ and challenges ahead

Anthony Fauci, the chief medical advisor to the president, described April as “vaccine open season.” President Joe Biden has announced that he expects to have secured enough doses to vaccinate every adult in the country by May.

When vaccine supply is no longer the main constraint, doses in arms is a reasonable metric to try to optimize, Omer said. When the state reaches a point where 30%-50% of the population has been covered by two doses, there will be population-level impacts on transmission.

“We’re not there,” he said. “But then it will get faster. Seventy to 80%, or 60 to 70% — depending on which variant is circulating — you will then have a substantial drop.” That’s what is known as herd immunity — which is not a set threshold — and doesn’t preclude outbreaks, but it does make them manageable, Omer said.

Connecticut’s own projections from two weeks ago suggest the state will have enough first doses on hand to vaccinate 45% of the population by mid- to late April, according to data from the governor’s office. The United States is projected to vaccinate 50% of the population with first doses late May, according to projections by the New York Times.

The state could potentially vaccinate a high percentage of the population by the summer, Gifford said, though vaccine uptake among younger populations will be a deciding factor. “Whether we reach herd immunity in May or July depends on how much vaccine we get and how quickly we administer it,” Gifford said.

Countering vaccine hesitancy among people of color will be a key challenge for the state, said Andrew Heinrich, lecturer at the Yale School of Public Health. Data also suggest that the state will have to address vaccine hesitancy along partisan lines as well. Connecticut’s current vaccination timeline assumes that the state can push out 105,000-110,000 first doses a week and vaccinate 60% of people in each age bracket as they become eligible; of course, once rendered eligible, residents can continue to receive vaccines in successive phases.

The question from an equity perspective is which 60% of the eligible population that will be initially.

The state has committed to sending 25% of vaccines to residents in 50 ZIP codes that have a “highly vulnerable” Census tract within them. Vulnerability is determined by a CDC index that takes into account 15 different factors, including race, income and living situation. The state has consistently found that individuals who live in highly vulnerable towns are less likely to be vaccinated than those who don’t. The state’s data on vaccinations by race suggest disparities in vaccinations by race, but missing data — nearly a third of vaccinations went to individuals whose race was not properly recorded — makes it difficult to understand the scale of the problem.

Last week, Keith Grant, senior system director for infection prevention at Hartford HealthCare, addressed legislators, advocates and others at the Ministerial Health Fellowship town hall on vaccine hesitancy among people of color — and the need to continue to combat health inequity well past the pandemic. “The mortality that’s attached to this disease is not because of the COVID-19 pandemic, it’s because of disproportionate access — decades of disproportionate access — to health systems, disproportionate access in many very disadvantaged communities,” he said. “This is forcing us to raise critical questions in specific communities that we should use to administer and adjudicate what we need to fix in health care in general, and so much more work needs to be done, much more communication needs to be done.”

Staff writer Jenna Carlesso contributed to this report.

Kasturi was CT Mirror’s data reporter. She is a May 2020 graduate of the Columbia Journalism School’s master’s program in data journalism and holds a degree in comparative literature from Brown University, where she was editor-in-chief of the student newspaper. Prior to joining CT Mirror, Kasturi interned for publications in India.

Leave a comment