The Importance of Leading and Lagging Indicators for Ongoing Monitoring of COVID-19 in Hawaii

by Nick Redding, Hawaii Data Collaborative

In a recent post, we highlighted a framework from Resolve to Save Lives that proposes “15 essential indictors” that should be publicly available in every State COVID-19 dashboard. The report for Hawaii suggests that there are multiple data points, currently not reported, that would help us better understand how well we are mitigating the spread of COVID-19 across our communities, and whether there are strains across our systems from how well we are able to monitor new infections, to the extent to which our hospitals have remaining capacity for increasing caseloads.

While it likely makes sense to most of us that these indicators measure different components of COVID-19 management – i.e. new cases, daily test rates, hospitalizations, etc. – it might be more difficult to understand how they all fit together, and why it is important to have a balance of indicators that represent the range of relevant factors for managing COVID-19 locally. One important frame to consider is distinguishing where data points fall along the continuum of leading to lagging indicators.

On one end, leading COVID-19 indicators suggest what is likely to happen in the coming days and weeks. They are forward-looking, providing early insights into shifts that lie ahead. Lagging indicators, on the other hand, are the result of the phenomena we are interested in. They are backward-looking, providing details about the outcomes that resulted from what happened in prior days or weeks. To help us understand this better, let’s take a look at a selection of COVID-19 indicators based on where they fall on the leading versus lagging continuum:

Note: Indicators highlighted in blue are those that are publicly available in Hawaii, to some extent.

There is a lot to unpack here, but it is helpful to start right in the middle with the indicator we are probably most familiar: the 7-day average of new daily cases. We track new cases daily because it is the best indication of how much COVID-19 transmission is occurring, and to what extent transmission is increasing or decreasing. However, focusing solely on this metric misses the opportunity to observe now what we might expect to see in terms of new cases in the coming days or weeks. Leading indicators of new cases could include use of personal protective equipment in public places, the extent to which individuals are following social distancing recommendations, and how well facilities and businesses are implementing COVID-19 guidelines. If these indicators are decreasing, we can expect increases in new cases in the near future.

Conversely, with increases in new cases inevitably comes increases in hospitalizations – including ICU bed and ventilator capacity devoted to COVID-19 patients – and, unfortunately fatalities. The fact that these indicators follow from new cases means they are lagging – they tell us what happened as a result of conditions weeks earlier. These indicators are less helpful for managing COVID-19 for this reason. Once we observe increases in hospitalizations, any interventions to mitigate this would take at least 2-3 weeks to effect hospitalization numbers. That is why it is critical for decision makers to act well before hospitals near capacity.

While we have discussed the ends of the continuum from leading to lagging, there are other indicators that fall somewhere in between. For example, indicators of contact tracing and testing capacity are slightly leading. When these capacities are low – for example, in terms of quantity of COVID-19 tests conducted, or delays in receiving test results – that indicates likely increases in new cases because we are not able to identity and isolate infectious individuals before they infect others.

Lastly, Hawaii benefits from being able to track travel into the State with high precision. Compared to mainland States with porous borders, we are only accessible by plane (cruise ships are not currently active) making it possible to measure the number of incoming individuals on a daily basis. This certainly can be a leading indicator of increases in cases – although more or less so depending upon departure origin travelers, measures in place to prevent infectious individuals from traveling here (e.g. pre-travel testing) or preventing individuals from spreading the virus if not screened (i.e. our current 14-day quarantine).

In recent days, there have been calls for relevant data that speaks to the broader set of considerations needed to manage the ongoing spread of COVID-19. Frameworks like this can help us make a strong case for expanding the availability of timely data in Hawaii.