Unnecessary Blindness: Hospitals Preparing for Pandemics

When the inevitable next pandemic influenza virus emerges, hospitals will be challenged to meet the requirements of a large cohort of individuals with varying degrees of illness. These patients will likely strain all the resources of hospitals including personnel, medical supplies, pharmaceutical supplies, and medical equipment. Because of the uncertainty regarding the magnitude and the nuances inherent in such events, it is a difficult task for a hospital to right-size its planning. Several tools exist, however, that have been developed to help provide estimates of supply needs including one developed by my colleagues: Panalysis.

To provide a real-world test of Panalysis, a team of us from the Center, Interdisciplinary Solutions, the University of Pennsylvania’s Wharton School, and the Mayo Clinic performed a stress test of the Mayo Clinic’s emergency pandemic supplies using various modeled scenarios. The result of that exercise was just published in the American Journal of Infection Control.

In this paper, my colleagues and I developed several different pandemic influenza scenarios of varying severity and, using Monte Carlo simulation, juxtaposed it against the specific features of Mayo Clinic and its patient catchment region in multiple iterations. Through the simulations, we could generate demand curves for certain supplies such as oseltamivir, gloves, and ventilators allowing insight into what types of demand would be expected for each of these items during various pandemic scenarios.

Using these demand curves, a facility like the Mayo Clinic could determine what level of preparedness they determined it prudent to invest in and compare current stockpiles to desired levels. For example, ventilator inventories could be maintained to be sufficient to meet the demands expected for 75% of the pandemic scenarios generated and an attendant cost generated. Similar cost-benefit analysis could be applied to N-95 respirators, courses of oseltamivir, or any other relevant item.

Every hospital will face unique challenges based on their location, services offered, catchment demographics, and size. Each will also have a differing risk calculus for preparedness and, instead of approaching this vital issue in an off-the-cuff/back-of-the-envelope manner tools such as Panalysis could be implemented to help bring rigor and quantification to these decisions allowing them to be evaluated in a manner much more fitting to their importance.

Please Refrain…

You never meant to cause us any sorrow.

From: eonline.com

From: eonline.com

You never meant to cause us any pain.

You only wanted one time to raise awareness.

But when you think on speculation,

Pause, and please refrain.

Please refrain, please refrain…

Be First.  Be Right.  Be Credible.  It is advice so central to crisis communication that it graces the cover of the CDC Crisis & Emergency Risk Communication (CERC) manual.  These are the tried-and-true basics of emergency and crisis communication.  I have lost track of how many after action reports have lauded those who heeded that advice, and pilloried those who forgot.

And then Prince died.

In the days following Prince’s death, speculation ran wild that he may have died from influenza or some complication thereof—prompted by reports from Prince’s spokespeople that he was recently hospitalized for the flu.  Some of these were written by public health and medical professionals, experts in their respective fields.  These pieces were, in turn, referenced by dozens of news articles and fed into the broader discourse surrounding the artist’s untimely demise.  While Prince having the flu and passing away in short succession may have seemed, superficially, like the perfect opportunity to raise awareness of flu fatalities (and I certainly appreciate this desire), I am left questioning the impact on public opinion of public health now that his death may have been linked to prescription painkillers.

In all fairness, all of the articles that I read readily acknowledged that the cause of Prince’s death was unknown.  But when recognized experts volunteer their opinion, regardless of whether or not they acknowledge the uncertainty, it lends validity to the speculation.  My major concern is that this was an unforced error.  During the response to an actual public health incident, health authorities are responsible for communicating what is known as well as what unknown about a given scenario, and they may be forced to speculate in order to take appropriate actions.  This was not a public health incident.  This was a celebrity death that was leveraged into an opportunity to discuss public health.  None of this speculation was necessary.  I fear that by trying to raise awareness about the severity of influenza, the experts—on whom we rely for clear, accurate knowledge during emergencies—may have damaged their credibility in the eyes of the public.  I fear that this speculation may be viewed by the public as yet another case of public health simply overreacting.

Public health struggles under the best of circumstances to maintain the public’s attention and trust.  Under the worst of circumstances, the burden of too little information and the demand to act and speak quickly can quickly derail a response.

On occasion, a well-meaning but misinformed public demands the impossible of us.

But there’s no study disproving a link between vaccines and autism.

Sometimes we do get it wrong, and recovering from that can be an immense challenge.

Yeah, remember when you guys said any hospital could handle an Ebola case?

Sometimes even when we are technically or factually correct, public perception may still disagree.

Remember in 2009 when you guys said that H1N1 was going to be a pandemic?

And sometimes, the public only reads the headlines, not appreciating the context or nuances of a given issue.

Remember when you guys said there would be a million cases of Ebola?

Now, I am just imagining—and fearing—the next conversation.

Flu isn’t that bad.  Remember when you guys said Prince died from the flu?



Too Many Patients, Too Few Resources

From: Chipotle

How do you decide who gets access to life saving medical resources like a ventilator when there are not enough to go around in a catastrophic disaster? My colleagues and I, along with collaborators at the Johns Hopkins Office of Emergency Management, School of Medicine and Berman Institute of Bioethics and Resolve, have been wrestling with this question for the last several years. Other researchers and experts have convened working groups and task forces to draft expert guidelines. We think their efforts have been very valuable, but we decided to approach the problem from a different angle. We decided to ask the public first.

For the last 5 years, we have been developing and running a community engagement research project asking lay people and healthcare workers in Maryland what ethical principles they think should best be used in allocating scarce resources. Specifically, we have conducted 15 day-long community engagement forums across the state over the last 2 years, asking Marylanders about what values they think should be applied when deciding who should get priority for a ventilator in a severe pandemic when there may be many times more patients with respiratory failure then there are ventilators to go around. I presented some findings from this work at the 3rd annual National Healthcare Coalition Preparedness Conference in Denver recently, and we will be presenting even more findings at the 2015 Preparedness Summit in Atlanta in April. The findings of this work will inform recommendations to be made to the State of Maryland for inclusion in a potential future Crisis Standards of Care plan.

For this project we used a “deliberative democracy” approach designed by the Carnegie Mellon University Program for Deliberative Democracy. In this approach, community members engage in small group discussions with trained facilitators. The goal is not to create consensus but rather to capture the many different ways that people think about such difficult decisions. The magic in this process is to see how people’s strongly held opinions evolve over the course of a thoughtful discussion.

We found that, universally, the participants were able to understand the ethical principles involved and work in constructive and civil discourse with others who hold different views. They were able to think creatively about the issues and apply their own personal values in thoughtful and nuanced ways. This was true whether the forum was held in a blighted inner city neighborhood, a wealthy suburb or a rural community. We did, however, also find that the community in which someone lives does play a role in how one think as about such things, especially when it comes to aspects such as social justice and access to care.

I used to think that a measure of success was getting on the cover of Rolling Stone (in the ‘70s) or Time (in the ‘90s) but now it appears that the measure of success is being on the cover of a Chipotle bag! Chipotle asked prominent authors to write short essays that they have printed on their bags, and Sheri Fink is one of the authors. Sheri, who won a Pulitzer Prize for her reporting on Hurricane Katrina, has been following our project from its inception and decided to make her essay on the bag about this project. She also wrote about the project in the epilogue of her bestselling book Five Days in Memorial.

What’s really important about this work, I think, is not which ethical principle (or combination of principles) is found to be most popular but that it shows that community members can constructively participate in such deliberations. It is important that a plan to allocate scarce resources reflects the values of the community as much as possible. But, as we have found, there is not one set of values, but many. To be able to reflect these values, we first need to understand them. We also found that you cannot get at these values through a simple poll or survey. As our project has shown, people’s opinions change as they engage in deep discussion—this is when their true values come out.