Houston data startup analyzes COVID-19 risks as companies return to work
In an effort to better help our clients, and frankly all of us, maneuver these uncertain times and to better understand what the upcoming months are likely to bring, we have applied our data science expertise to create a structural model of the spread of COVID-19. The aim of the national model is to determine specifically how mobility and weather impact the local transmission rates while controlling for population density, population immunity rate and the fact that people are taking more precautions.
When I discuss COVID19 with other Houstonians, I'm often asked "We're going back to work — there's traffic! Why haven't cases spiked?"
First, it is worth noting that cases and deaths have increased again in Harris County. But, the question is still valid. Greg Abbott started allowing things to open six weeks ago and we are only starting to see a rise now.
Fortunately, our model (being quantitative and multivariate) can explain why cases may not have 'spiked' the way that was expected. There are four main reasons why cases are only starting to tick up now:
- There has not been a 'spike' in people leaving their homes. While Greg Abbott did allow restaurants, movie theaters and malls to begin re-opening on April 30, there was not a sudden spike in people leaving their homes. Indeed, the "people staying home" index, according to Google Mobility Data, peaked on April 1 at 22 percent above normal and has gradually decreased ever since. In Harris County, the extent people are staying at home stands at 14 percent above normal as of May 29 (unfortunately, Google Mobility Data reports 7 to 10 days after the fact.) So, from the peak 'stay at home' measure, we were only a third the way back to normal last week.
- Temperatures have increased. Our model indicates that warmer temperatures decrease the transmission rate of COVID19. Our model does not posit a mechanism, but we can rule out both geographic explanations and behavioral explanations, which leaves us with the compelling reason to believe that temperatures matter.
- People's behaviors when they do go out have changed. These changes — everything from masks, to skipping hand shakes, to readily available hand sanitizer, to keeping your distance, to staying home when you're feeling sick — have an important and measurable impact on the spread.
- Kids are not back in school yet. While our model does not directly measure the impact of kids being in school, the estimate our model produces measuring the importance of staying at home (2.6) is higher than they should be (2, mathematically speaking). We suspect that's because we are missing an important cohort that started staying at home at the same time mobile phone users started staying at home - kids that don't have cell phones. So, while it may seem like we are most of the way back to normal, with regards to going out, being summer time in Houston, kids are not at school, which is likely keeping the rate of spread down.