Wax On, Wax Off
One of my all-time favorite films was the 1984 version of The Karate Kid. In
We have been badly served by elected and appointed government officials, public health authorities, and the media in getting an understanding of what has happened and is likely to happen relative to Covid19. Whether we are running a business operation, trying to plan for schooling for ourselves or our children, or deciding whether we trust those we elect to public office, we need the best information available.
Regrettably, we are far short of meeting that standard. Two op-ed pieces authored or co-authored by Dr. Tom Frieden, the former director of the Centers for Disease Control and Prevention help us understand why.
The first appeared in the June 10, 2020, issue of The Washington Post. Dr. Frieden points out the many misconceptions that appear in both political speeches and media reporting on public health pronouncements in Covid19 reporting. In so doing, he states that we focus too much on the wrong metrics and not enough on the right ones.
www.washingtonpost.com/opinions/2020/06/10/how-amateur-epidemiology-can-hurt-our-covid-19-response/
One misconception he cites is particularly noteworthy:
“Obsession with case counts is misleading; we estimate that only about 10 to 15 percent of U.S. infections are diagnosed.” (Italics and bold print added)
That percentage may even be lower than 10%, since we do not have a reliable way of estimating the total. We have locked ourselves into a position that the only way to confirm cases is to do Covid19 testing, but we are very far from having the ability in every part of the country to test everyone who might have been infected.
The CDC now enables healthcare providers and medical examiners to claim that a death is due to Covid19 without doing a positive test, but their guidance is highly imprecise. There are two ways to address this issue better: (i) using a surrogate and supplemental method of estimating virus outbreaks, such as a continuous analysis of untreated wastewater; or (ii) continually doing random total population sampling.
Dr. Frieden also notes that the public has been bombarded relentlessly with data on Covid19 deaths day after day. The better metric, which takes into account a much broader set of consequences from Covid19 actions and their consequences, is a statistic he calls “excess mortality.” That would include not only Covid19 deaths not discovered through testing, but deaths that occur because of delayed cancer treatments, suicides, failure to manage coronary artery, Type 2 diabetes or other chronic conditions, and deaths from opioid overdoses.
We should be reducing or eliminating “excess mortality,” not single-mindedly focusing on Covid19.
Dr. Frieden co-authored a second op-ed piece, this one in the July 21, 2020, issue of The New York Times, with Dr. Cyrus Shahpar. In it, the co-authors make another significant point:
“Doctors caring for patients track vital signs of temperature, blood pressure, breathing and pulse. Public health doctors fighting epidemics do something similar — they track the most important indicators of the spread of a disease and attempts to control it…..We aren’t tracking the public health equivalent of vital signs. That’s one big reason the United States is losing the battle against Covid-19….Over the past three weeks, researchers in our initiative, Resolve to Save Lives, searched all the data they could find on publicly available websites from all 50 states. They found it to be shockingly inconsistent, incomplete and inaccessible.” (emphasis added)
www.nytimes.com/2020/07/21/opinion/coronavirus-state-data.html
Drs. Frieden and Shahpar list 15 data sets that every public health authority should be collecting, reporting, and analyzing, citing Dr. Shahpar’s “Prevent Epidemics” web site.
preventepidemics.org/covid19/resources/indicators/
Indicator number 8 is of particular interest:
“List (to extent legally permissible in State) of long-term care and other congregate facilities (homeless shelters, correctional facilities), and essential workplace (e.g. meatpacking) outbreaks with COVID-19 cases and deaths in residents and staff.”
These are many reasons this vital data is not collected, but I believe that one of them is the focus of political leaders on publicizing data that fits a specific narrative, particularly one that diverts public attention from the troubling and fundamental public policy failings in these congregate facilities and essential workplaces.
Governors and public health authorities talk extensively about “vulnerable populations,” particularly elderly people with “underlying health conditions.” They also discuss overcrowded housing that creates virus spread risks for low-income populations, particularly immigrants.
However, the data on “congregate facilities” is more nuanced and, if it were analyzed, we would find that the problem of Covid19 deaths is more heavily concentrated in nursing homes that have a higher percentage of state Medicaid support. We also might find that inadequate regulation of nursing home processes, protocols, and staffing has contributed more to Covid19 deaths than many of the more publicized reasons.
In Connecticut, an estimated 65-74% of all Covid19 deaths occurred in nursing homes. This article cites 65% as of July 24; The New York Times estimates 74% as of July 30, 2020. In any event, the nursing home deaths around the country are a story that has not received the prominence they should have.
A task force led by Carol Raphael of Manatt Phelps made the same findings relative to New Jersey’s nursing homes, which constituted over 50% of New Jersey’s Covid19 deaths.
www.manatt.com/insights/press-releases/2020/manatt-provides-recommendations-to-state-of-new-je
In certain middle-income and upper-middle income communities with sizable nursing home populations, the death rates have been horrendous. Fairfield, CT, a town with a population of 62,000 people and an average household income between 2014-2018 of over $134,000, has had 145 Covid19 deaths. 92.4% of which have occurred in eldercare facilities, according to the data released by Fairfield Town Government.
The highest death rate of any town in Connecticut has been in Shelton, 135 deaths (as of July 30, 2020) in a town of 40,000 people, also a result of deaths in nursing homes, of which 121 occurred in four nursing homes.
portal.ct.gov/-/media/Office-of-the-Governor/News/20200730-Nursing-Homes-with-COVID-19.pdf
As the chart shown on the above-linked report indicates, many nursing homes had no or very low death rates, whereas others had many. Atria, a nursing home in Darien CT with no Medicaid recipients and not listed on the chart, has had no deaths.
The question we must ask is why there is such variation among nursing homes in the number of cases and deaths they experienced from Covid19. The Manatt report makes it clear that nursing homes with a heavy concentration of residents paid for by Medicaid and, as a result, with the lowest-income residents are those with the highest percentage of cases and deaths in New Jersey.
This article in Health Affairs presents some damning data about Medicaid-supported long-term care residents.
www.healthaffairs.org/do/10.1377/hblog20200729.267815/full/
The summary is very clear as to both the causes of lousy results for Medicaid recipients and the results themselves:
“With no federal floor on Medicaid rates or minimum staffing and wage levels, state-set Medicaid rates for nursing homes at about $200 per day are lower than they are for other payers. Low reimbursement levels have translated into outdated facilities, with often two or more residents sharing a room and a workforce, which is the backbone of the system, that is poorly paid and lacks benefits such as paid sick leave as well as the training needed to care for people with multiple chronic illnesses that can include dementia and other mental health issues. This training gap is compounded by the lack of strong connections to the broader health care system that could provide needed clinical support.
Despite all that is at stake, LTSS oversight is weak at both the federal and state levels. For example, 82 percent of nursing homes were found to have a deficiency in infection prevention and control at some point between 2013 and 2017, and, most striking, 48 percent had such deficiencies cited in multiple years during the same period.”
Someone looking at the political, media and public health dialogue in Connecticut and other states for the last several months has to work hard to find this data on nursing home deaths. The political and media narrative about the spillover effect from people from out of state or even from New York does not come close to diagnosing what really happened and what we need to do differently going forward.
The narrative that health and healthcare disparities result in greater numbers of deaths among African Americans is accurate, but incomplete. The narrative that unhealthy older people die from the virus, regardless of where they are living or securing care is false. Some nursing homes with “vulnerable populations” that have “underlying health conditions” have had no deaths and others have had far too many.
Drs. Frieden and Shahpar are absolutely right. Public health authorities need to be empowered to do the job of getting the right information to all of us as soon as it can be made available, and we need to step back from political and media spin to understand the challenge in front of us.
Final observations
We need to hold our leaders accountable to tell us the “inconvenient truth”of the Covid19 disease crisis: it has exposed the many failings of how we manage health, healthcare, housing for low-income populations, long-term care, public health governance, and many other governmental processes. As Ed Yong, a staff writer at The Atlantic, stated in a recent article in that publication:
"The coronavirus found, exploited, and widened every inequity that the U.S. had to offer.”
//www.theatlantic.com/magazine/archive/2020/09/coronavirus-american-failure/614191/
The sooner we face up to these failings and correct them, the more we can draw some benefits from this avoidable tragedy.