Revisiting COVIDeracy: What’s in a Number?
Over the past couple of months, CTData has been scraping local and national COVID-19 data, staying up to date on resources, and following policy developments to investigate a broad but important question: what is the data telling us? We developed Connecticut COVID-19 dashboards that track cases, rates by county and town, racial disparities, and economic impacts. We have also written blog posts about finding trusted data sources, flattening the curve, COVIDeracy (COVID-19 data literacy), Connecticut unemployment claims, drops in business registrations, and COVID-related survey results.
Like many of our partners, with every new data point, news article, and piece of information, more questions continue to arise. This week, we dove into the COVID-19 death data to highlight how a simple difference in data definitions can influence key takeaways. To do this, we compared day by day deaths to cumulative deaths.
As of May 20, 2020, Connecticut has reported 3,529 deaths, which represents 9% of all cases and 99 per 100,000 residents in the state. According to daily reports from the Connecticut Department of Public Health (CT DPH), “for public health surveillance, COVID-19-associated deaths include persons who tested positive for COVID-19 around the time of death (confirmed) and persons whose death certificate lists COVID-19 disease as a cause of death or a significant condition contributing to death (probable).” It's possible that individuals who were not tested but had COVID-19 listed as their cause of death were not included in the original case counts.
Daily Deaths
As shown in the chart below, daily death counts (defined as the number of deaths reported each day), as reported by CT DPH, vary by day. However, these numbers should be viewed with caution since daily death counts are likely the result of staffing and administrative capacity rather than actual severity. This discrepancy is because death data comes from official death certificates. In Connecticut, death certificates still rely on a paper-based processing system. Gathering information to complete the death certificates takes an average of five days, which results in a lag between when a death actually occurs and when the death officially appears in the data. CT DPH is in the process of moving to an online reporting system for death certificates.
Source: CT DPH, Chart created by: Connecticut Data Collaborative
Furthermore, the source of data matters as reporting lags between states and the Centers for Disease Control and Prevention (CDC) could impact the numbers, depending on what data source you are using: official records from the National Center for Health Statistics (NCHS) or data from local health departments. According to an April 2020 New York Times article, “The speed of that data reporting varies considerably by state. In Connecticut, for example, where reported coronavirus deaths are high, the CDC statistics include zero reported deaths from any cause since Feb. 1, because of reporting lags.” As of May 21, 2020, NCHS only included official death data through February 2020. In contrast, as of the same date, the CDC COVID Data Tracker appears to match CT DPH numbers for both total cases and deaths.
Cumulative Deaths
Deaths can also be analyzed cumulatively, which is defined as the total number of deaths that have happened over time. In this analytic approach, deaths are added each day so that today’s death count sums all of the deaths reported up to today. This process allows readers to see a general trend in the data and is less prone to discrepancies due to administrative lags in reporting. A curve that begins to flatten indicates deaths slowing down in the state.
As shown in the chart below, the cumulative death count continues to increase across the state, with the 14-day average of approximately 54 new deaths.
Source: CT DPH, Chart created by: Connecticut Data Collaborative
The two charts above highlight different ways to look at COVID-19 death data. Both are acceptable ways of analyzing the information and have different strengths and challenges. It is important to understand what data and data definitions are being used when looking at charts, especially if you realize two figures look different from each other.
Something that sounds simple—counting COVID-related deaths—is actually quite complex. In addition to differences in analysis and administrative reporting lags, what and who gets counted in the COVID-death toll is also rife with challenges. Some states have been scrutinized for suppressing their death counts while others have been criticized for having too wide of criteria. These discussions highlight the critical need to have transparent, clear, and consistent data definitions. For more tips on understanding and interpreting COVID-19 data, check out our COVIDeracy blog post.
You can always count on CTData to provide accurate, updated information, so make sure to check out our website, blog, and newsletter. You can also follow us on Twitter, LinkedIn, and Facebook for real-time updates and feel free to share any tips with us about how you critically consume data!