Within a single weekend in April 2020, the nation’s most-watched coronavirus model slashed its projections for Alabama — cutting forecast deaths by roughly 80 percent and erasing predictions of a desperate shortage of ventilators and ICU beds.
As recently as that Saturday, the model from the Institute for Health Metrics and Evaluation (IHME) at the University of Washington projected Alabama would suffer the fourth-most COVID-19 deaths of any state by the end of August: 5,516 fatalities, with critical shortages of hospital capacity. By Monday morning, two days after Gov. Kay Ivey’s statewide stay-at-home order took effect, the same model was projecting an average of 917 deaths and moved Alabama’s peak to mid-May. A further update on April 9 lowered the projection to an average of 634 deaths, within an uncertainty band of 237 to 1,837.
Why the numbers changed
Dr. Ali Mokdad, the IHME professor of health metrics sciences who built the model, said better data drove most of the improvement. Early versions leaned on CDC reports from the first U.S. cases and on outcomes from Italy and Spain; once researchers could add state-specific data in early April, Alabama’s curve looked very different.
New weight given to social distancing also mattered. “The reduction in projected deaths is the implementation of these social distancing measures, which usually have an impact about a month down the road,” Mokdad said. “This is the result of the appropriate decisions your governor has made.”
He singled out Alabama’s March 13 decision to close public schools — more than two weeks before the state’s first reported death — as likely having a significant effect. “Closing the schools and universities has had a huge impact in Alabama and has saved a lot of lives,” he said, while stressing that a stay-at-home order remained important because school closures alone only change where people gather.
Models run hot — and why that’s no reason to relax
Even the retooled projections ran well ahead of reality on the ground: the Alabama Department of Public Health reported 53 deaths as of April 6, versus an IHME projection averaging 89 for that date — about 87 percent higher. IHME also acknowledged an error in the “total deaths” figure displayed on its public dashboard, which briefly showed more deaths than the state had reported.
Mokdad attributed wide margins to live, evolving data that shifts projections day to day, with bands expected to tighten as more information arrives. But he cautioned against reading the improved outlook as an all-clear.
“We shouldn’t be letting our guard down. When something is encouraging, you keep going,” he said. “I’m happy for Alabama and what you’ve done with these measures, but it’s time to stick to them so you can reduce the number of deaths and possibly even reopen your businesses earlier than originally expected.”