Page 24 - Layout 1

COVID-19
PANDEMIC

Loosening up the In-Home Sheltering Policy.

What will happen next?

By Bob Leverence, MD

As of this writing, the nation is focused on concerns over a Let me try to explain. We certainly have good reason to hope the
COVID-19 resurgence in the face of many states reopening despite next surge will not be so bad. Warm weather has been shown to
not meeting CDC criteria. Economic pressures for reopening are slow the spread of this virus, people are now behaving more pru-
certainly valid and compelling, however they come at the risk of dently regardless of easing community restrictions, testing is more
one or many resurgences. This would not only lead to more loss of available, we have at least one effective treatment (Remdesivir) and
life but more harm to the economy as well. there is at least a low level of population immunity. All these factors
One way to help guide this reopening is by using predictive mod- weigh in favor of a lighter resurgence. So, let us hope for the best.
els, and so there are many in use today. 1,2 These models quantitate Alternatively, government officials, civic leaders, scientists, and the
and incorporate the many factors which influence the spread of press all have the obligation to plan for the worst. Unfortunately, it is
COVID-19 — climate, urban density, age distribution in the com- hard to listen to them night-after-night and now month-after-month,
munity, social distancing mitigations used, population immunity, etc. prophesying devastation and gloom. But again, that is their job and I
Since the results or predictions from these models depend on a host am thankful for them. The factors which most dial up the likelihood
of assumptions related to these factors, it is not surprising we see a of a bad surge are the lessening of social distancing and a reduction
broad range of outcomes. In other words, despite the fact these in infection control practices. If people congregate closely in public
models are data driven, forecasting the future will always be plagued places and shops, if they do not regularly sanitize their hands, or if
with uncertainty. With that said, higher quality data leads to more they do not wear masks and stay home when ill, then we are likely to
accurate assumptions which reduces that uncertainty. see a repeat filling of our hospital beds with COVID-19 patients and
Many critics state these models have not been very useful thus another round of in-home sheltering measures.
far. One reason being that early in the pandemic good data just was We are still in the thick of this pandemic and will be until a highly
not available — after all, a pandemic like this has never happened effective treatment or vaccine becomes available. Until then, patients
before. So, the models have been learning with us. Then again, one will continue to delay needed care out of fear. Many will also lose
might argue this is not our first pandemic, and so should not models their insurance due to unemployment. All of these factors mean the
already be sitting on the shelf just needing calibration for this virus? health of our communities will certainly suffer. As physicians, we
Well, we could use that same argument for greater stockpiles of need to reach out to our patients to assure them our clinics and pro-
PPE, more reliable supply chains for swabs, and better public health cedure centers are safe so they will feel comfortable seeking the care
infrastructure for crisis management and contact tracing. But that they need. In the spirit of social distancing, we also need to use tele-
is a whole other subject. If we learn anything from this pandemic, health wherever possible. Since COVID-19 will likely be here for
I hope it is the need for better pandemic preparedness. the next year or so, this will be our new normal.
So, getting back to the predictive models, ideally, they should be But COVID-19 will someday pass, and we will eventually start
applied to local environments since that is where outbreaks occur thinking about other things. So, what will happen next? Maybe we
and much of the mitigation happens. Stated differently, predictions will see sustained reductions in greenhouse emissions. I hope we
which are modeled at the state or national level can mislead a com- celebrate the new breed of hero we have found in our healthcare
munity since micro-environments can be so different. Likewise, workers and civic leaders. I myself have certainly come to a new ap-
models should include an analysis of supply and demand for re- preciation of what is important.
sources such as hospital bed capacity. Finally, results should be dis-
played in a way that non-scientists can readily understand them. So Bob Leverence MD, is Professor of Medicine and CMO UTHSA and is
that is a tall order few models have achieved and why multiple mod- a member of the Bexar County Medical Society.
els are used.
In my view, the predictive models help most when we report them References
in a balanced way. I like to call this the “Hope for the best, but plan 1. Jewel, Nicholas P., Predictive Mathematical Models of the
for the worst” approach. I say that because if each half of the state- COVID-19 Pandemic. jamanetwork, April 16,2020. https://ja-
ment were to stand alone, we would be in danger. If you only hope manetwork.com/journals/jama/fullarticle/2764824
and report on the best-case scenario, this denial of the worst case
would leave you ill-prepared. Alternatively, if you are the kind of 2. Centers for Disease Control and Prevention. COVID-19 Fore-
person who only prepares for the worst without a vision for how casts. Updated May 6,2020. https://www.cdc.gov/coron-
good it could be, then people will avoid or not even listen to you. avirus/2019-ncov/covid-data/forecasting-us.html.

24 San Antonio Medicine • June 2020