Instead of Relying on Stories, We All Need Access to Better COVID-19 Data
by Nick Redding, Hawaii Data Collaborative
In late June, daily new cases of COVID-19 were slowly but steadily increasing on Oahu — just as earlier forecasts indicated would happen. Even when it seemed that the number of new cases was plateauing during the first two weeks of July, we were warned against complacency. This was not going to be the same as our previous peak at the beginning of April — the rapidly increasing transmission rates signaled more alarming exponential increases on the horizon.
This was not a guess or speculation based on pessimistic thinking. This forecast was based on reliable data modeled by local experts. Thanks to the robust data sharing systems implemented by the State, combined with extensive input from the broader community of health experts and data scientists, we were able to detect important signals early:
Disproportional increases in cases among certain populations,
Patterns in clusters of transmission,
Mobility data pointing to more activity in key places, and
Decreasing compliance with PPE and distancing guidelines.
In addition, daily data from the Department of Health showed that contact tracing and testing capacities would soon be compromised, making it clear that we would not be able to mitigate further spread through detection and isolation alone.
Fortunately, leadership and the broader community knew these things early, enabling us to act accordingly without the need to again close businesses and restrict access to beaches and parks. As a consequence, today we are back to a manageable number of new cases.
Because we had the data, analysis and expertise we needed when we needed it, our schools are reopening safely, hospitals are nowhere near capacity, and (pre-screened) visitors are beginning to travel to the islands again. We continue to be a national and international example of how to manage the new-normal of COVID effectively with robust data and reporting systems, and a well-informed public.
That is one scenario of where we could be today — if only we had done all of the above. It’s a nice story, but as we all know, our reality today is much different, as we again stay at home keeping a close watch on the daily numbers of new cases. We feel helpless not knowing why we are where we are, or what lies ahead, because we do not have the information we need. We find no shortage of stories attempting to explain our situation, but those stories are too inconsistent, too vague, too detached from what we are observing to address our basic questions:
Should we expect the increases in cases to continue? While there are admirable efforts to develop robust forecast models for the state, so far an authoritative public resource for this has not emerged. While Hurricane Douglas reminded us that forecasting is challenging, it also showed us how invaluable weather forecasters are in helping us prepare for disasters.
Do we have enough contact tracers to manage the current number of cases? There have been a couple stories that address this question, essentially boiling down to “yes” and, well, “no.” We should not have been left to guess which story is true. More importantly, this is symptomatic of the fact that we lack good data for monitoring contract tracing capacity. In addition to the number of active contact tracers and contacts being monitored (metrics currently updated weekly), we need indicators of how well our contact tracing system is performing, such as the average time between positive test result and contact tracing interview and the percent of identified contacts that are traced—all of which would tell a more complete story.
Will Oahu’s hospitals soon be over capacity with COVID-19 patients? This data is collected daily from the hospitals, but it is not shared publicly – even at the state or county level. For now, we are left to rely upon alarming media accounts and statements from those closer to the data to keep us informed. How many ICU beds do we have? How many ICU beds are occupied by COVID patients? How many ICU beds are occupied by non-COVID patients? If we were able observe these numbers daily, we could tell our own story about whether hospitals have sufficient capacity.
In short, we are not taking the potential of sound data and analysis seriously in this state in responding to COVID-19.
The potential of data lies not in its ability to reinforce stories — that leaves us reliant upon the storytellers. Instead, comprehensive and readily available data facilitate conversations that allow the broader community to see exactly what is happening and, if necessary, question assumptions about our current situation and what lies ahead. Over the last two weeks, it has become clear that state and county leadership do not have full access to the data and expertise needed to manage the disease. And neither does the public.
Government and our health institutions are responsible for managing this crisis, but the responsibility does not sit only with the State. The consequences loom too large for that to be the case. It is incumbent on all of us — as policymakers, business and nonprofit leaders, educators, health professionals, and concerned citizens — to act based on the best information we have available to us.
But we can only play our part if we are all looking at the same numbers to guide our questions and scrutinize prevailing assumptions, engage where we are needed, and hold those responsible to account.
Enough stories, we need to be looking at the data.
