Having quality data is paramount to how we make decisions to combat COVID-19. The decisions made now at an institutional level have life-and-death consequences, as do the choices we make around individual actions like social distancing.
In the midst of today’s pandemic, I wish we still had Dr. Bill Bicknell, one of my professors from the Boston University School of Public Health, who passed in 2012. He worked in public health policy, program implementation, and clinical medicine around the world, and served as Massachusetts Commissioner of Public Health. He spoke truth to power, he didn’t shy away from challenging authority, and he wasn’t afraid to call “horseshit.”
As a teacher drawing on decades of his own experiences, he forced us to think about how public health data and decisions impact human lives, and to remember the people not at the table. What I learned from him shaped my worldview around the ethics of the decisions we make in healthcare and how we, as data visualization designers, inform those decisions with our charts, graphs, and maps.
Here are three lessons from his last lecture that we should all revisit today.
1. Our decisions (and the data that inform them) have life-or-death consequences.
I remember my first day in Dr. Bicknell’s introduction to global health policy class where he challenged us with a bold statement:
“Public health is the art and science of deciding who dies, when, and with what degree of misery. OR the art and science of deciding who lives a longer, less miserable, and happier life. If you start with the second one, people forget that when they screw up they’ve killed people.”
Dr. Bicknell believed “public health is much more dangerous than medicine. We deal with populations, doctors deal mainly with individuals. So bad public health professionals are far more dangerous than bad doctors. We can kill more people with less accountability.”
The data we communicate can impact individual actions and institutional decisions, and in turn contribute to these life and death consequences. Dr. Bicknell described himself as “not big on lots of data, but big on looking selectively and really figuring what you know.” He believed having accurate denominators to calculate public health rates and measures were particularly important. This rings true as we grapple with a pandemic where we don’t have great certainty in the total number of infected persons — just confirmed cases, in part due to limited testing. An uncertain denominator adds extra complexity to calculating metrics like the case fatality rate for COVID-19.
In the spirit of calling “horseshit,” he questioned the sources of aggregate figures, benchmarks, and recommendations. In his last lecture, he recalled probing a recommendation that 15% of a country’s budget should be allocated to healthcare in order effectively meet population health needs (defined in the Abuja Declaration) and calling colleagues to try to track down the rationale behind that number. He wouldn’t blindly accept the idea that anything below 15% was “horrible” and recognized that incremental progress toward that figure was still good.
When working with data, we should be just as willing to probe and understand the sources, math, assumptions, and limitations behind calculations, and recognize where there is uncertainty. When visualizing data, we should remember that plotting numbers as marks on a graph implies a degree of certainty in the figures, which is why the discussion on visualizing uncertainty is so critical to our field.
Dr. Bicknell reinforced for each of us that the choices we make in our work have consequences. Whether in a public health emergency like we have today, or the slow slog of strengthening health systems, we need to hold ourselves accountable for those choices.
2. The Realities of Rationing
Health systems always face resource constraints.
Dr. Bicknell taught us that in some health systems “very scarce resources mean some get [healthcare services] and some don’t. Those who don’t die sooner, or they live in more misery. This is rationing [of healthcare] — we don’t like it. So rather than face reality, we spend a lot of time convincing ourselves that promotion, prevention, sound policies, and good management can avoid the pain of explicit rationing. It is a dangerous myth that you can avoid rationing.”
We don’t like to talk about rationing healthcare in the US. It’s a common argument for why we should keep our fragmented health system, with its patchwork of public and private pieces instead of moving to a single-payer system.
Americans often seem more comfortable talking about healthcare “triage” than rationing. Triage implies assigning a judgment on the severity of a case in the midst of an emergency department, where every person who needs care gets seen by the doctor but we treat the most urgent cases first. Rationing implies someone doesn’t get seen. Both healthcare triage and rationing imply the same choice though: Who receives health care in a system with limited resources?
Dr. Bicknell made us confront the reality that rationing care is part of our health system under normal circumstances. The urgency and challenges around rationing are amplified in the face of a pandemic.
In the coming weeks, hospitals and government leaders will decide what facilities will receive the available tests for suspected cases of COVID-19, which patients receive tests, where to ship personal protective equipment for our health workers, and how to maximize access to ventilators and other medical devices for treating those with the most severe symptoms.
We need better data on where the burden of disease is greatest to inform these decisions. Better confirmed case data requires wider availability of testing for COVID-19. While the US is still in the relatively early stages of community transmission — rationing tests more than ventilators — countries like Italy further along their epidemic curve have doctors making stark decisions on who to treat.
As we think about health system capacity, we must also remember the other patients a hospital serves that aren’t suspected and confirmed cases of COVID-19. In the midst of this pandemic, people are still having babies, getting in car accidents, and requiring health services for other needs that cannot be delayed. While postponing a hip replacement is feasible, you can’t say to a woman in labor to just wait and have the baby later.
This is why the dotted line on the graphic about flatting the curve is so important. Slowing the pace of new infections helps us avoid exceeding our healthcare system’s capacity.
3. Remember the person not at the table
“What do you need for professional success?” asked Dr. Bicknell in his last lecture. “I think you need a moral compass. I think you need to know what the hell you’re in this business for. You need to listen. Really listen. Understand all the concepts — don’t do all the detail, but figure out the relevant detail.”
He reminded those of us working in health, “It’s vital to remember; we’re in the service business, whether we’re making policy or suturing a wound. We’re serving people who need help and are hurting. And respect. And the person who’s not at the table.”
“The Melon Lady” was his own anecdotal example for remembering that person not at the table. Kate Mitchell, a friend from my days at BUSPH, writes:
A story Bill often recalled from his time as Commissioner of Public Health for the Commonwealth of Massachusetts.
Bill was new on the job and working long hours. One evening as he was leaving the office, he noticed several perfectly good melons in the dumpster in the parking lot. Bill walked over, examined the melons and, after a healthy diagnosis, decided to bring them home to his kids. After a few more evenings of snatching free melons from the dumpster, a janitor approached Bill. He said, “Commissioner, you may not know this, but there are women who rely on this dumpster. They come here every evening looking for food.”
Every time I heard Bill tell this story, he concluded by opening his eyes, looking out across the auditorium, and declaring, “Let the melon lady be your guide.”
Another professor might have simply pointed out that, as global health and development decision-makers, we have a moral obligation to consider the person not at the table — and to think critically about the far-reaching implications of our actions.
Leave it to Bicknell to hammer home the point with the story of the melon lady.
And the janitor.
Right now, that might be the mom whose kids get two hot meals at school each day and is sorting out how to put food on the table thanks to closures, or the waitress losing out on tips as people work from home. Or, it may be a grandmother getting pieces of information from the news and feeling nervous about what this pandemic means for her.
This pandemic will have far-reaching impacts on people, health systems, economies, and perhaps even the climate. The weeks in March and April with widespread closures and cancellations will become an intensely studied period as economists, sociologists, and other researches try to assess the effects of our social distancing efforts. Those effects won’t all be positive, and those most impacted are likely those not at the table who don’t have the resources or flexibility to just work from home and stock up on supplies.
Dr. Bicknell shared these anecdotes in his last lecture, thankfully recorded and shared by BU. His slides weren’t fancy, and he didn’t care much for spending time on graphic design, but he cared deeply about the design of systems to be accessible and equitable for all. I’d like to think that he would find value in the ways data visualization has made data more accessible in the eight years since he delivered this lecture.
He passed from cancer at age 75 in 2012, but I’m hopeful that his legacy lives on in the generations of public health professionals he taught and impacted through his blunt, honest emphasis on the ethics of the decisions we make.
Amanda Makulec is the Senior Data Visualization Lead at Excella and holds a Masters of Public Health from the Boston University School of Public Health. She worked with data in global health programs for eight years before joining Excella, where she leads teams and develops user-centered data visualization products for federal, non-profit, and private sector clients. Amanda volunteers as the Operations Director for the Data Visualization Society and is a co-organizer for Data Visualization DC. Find her on Twitter at @abmakulec.