A Letter From Your Epidemiologist

Demystifying Infectious Disease Jargon

8 min readApr 15, 2020

By Kate Mallula, MPH MSW and Sarah Martin, MPP MPH PhD

Getting a degree in Public Health means that many of our holiday dinners with family were spent saying things like:

“No I am not a medical doctor”
“I track diseases, but I don’t treat them”
“You probably didn’t get food poisoning from that restaurant, you probably gave it to yourself”

But now, nobody asks what it is we studied. A global pandemic brings the field of Public Health into the public view in a way we’d never want, but demands our guidance.

Yep, the world is a little scary. Numbers are in constant flux. Epidemiological terms and statistics that come from carefully constructed methodologies are turned into sound bites in our news cycle. And all the while, the public is seeking simple answers: What is happening? Do we know what to do? How can I help? With an outbreak trajectory driven by complex biology and social factors — people want and need to understand where we go from here. In the midst of the COVID-19 pandemic, this is the new data and communication reality our public health colleagues are grappling with.

We here at mySidewalk want to use this space to empower our partners to understand some of the real-time epi statistics you’re likely encountering. To help you break through the jargon and navigate the evolving landscape of outbreak measures, we’ve pulled 5 of the most common terms with basic explanations to boost your knowledge (or refresh your memory).

Case count
Secondary attack rate (SAR)
Basic reproduction number (R0)
Case hospitalization proportion
Case fatality rate

There are two key considerations that hold true for each of these measures. First, all of these measures warrant careful interpretation and messaging. Second, the impact of COVID-19 is significantly influenced by the social determinants of health. Where we live, the resources available to us, how health is already distributed throughout communities, and how our interventions take these SDoH into account will all contribute to the way the measures below unfold in our communities.

In order to understand these measurements best, it’s helpful to know about a margin of error. We usually use something called a 95% Confidence Interval around a number to show the range that is generally accepted as probable. When we show numbers here, we may abbreviate this as “CI”, with a lower and upper bound estimate. We do this because we lack complete and accurate information about actual stats, so we do the best we can to provide a good estimate while quantifying uncertainty around it.

Case count

This basic aspect of public health surveillance seems to be the most straightforward, but should be interpreted carefully during outbreaks. First, it’s easy to throw around the term “case,” and lose sight of the fact that each case represents a person, and, by extension, their family, friends, and community. Though we refer to cases in the measures below, we are aware of the profound implications that each case — and person — holds for our communities.

In considering case counts, it’s important to remember that people with more severe symptoms are more likely to seek treatment and thus be counted as a case. Because cases identified early in outbreaks tend to be more severe, they can skew case hospitalization and fatality rates, which we discuss below (case counts are components of both these measures). As many of us are aware, the availability of and access to testing also affects a community’s capacity to accurately identify and count people with COVID-19 infections.

Regularly updated case counts for the US can be found at this site created by Johns Hopkins School of Public Health. The CDC’s case count is available here, though each state is also tabulating cases, sometimes more quickly than CDC can track. Additional considerations and strategies for COVID-19 surveillance can be found here.

Secondary attack rate (SAR)

This rate is used to express the rate of transmission in specific settings and is usually expressed as a percentage. In relation to the current outbreak, SAR can be thought of as the probability that COVID-19 will be transmitted to people with no immunity in specified settings, such as households. This measure is useful in describing how different social environments contribute to transmission¹ SAR is the following fraction:

# of new cases among contacts after exposure to primary case for the relevant incubation period / # of contacts².

Estimates released by the CDC for the COVID-19 SAR in the United States are .45% (95% confidence interval [CI] = .12% — 1.6%) among all close contacts and 10.5% (95% CI = 2.9% — 31.4%) among household members.³

Basic reproduction number (R0 )

This is the most controversial statistic here. R0 (“R naught”) expresses the average or median number of infections resulting from exposure to a person with an illness when there is no immunity present in the community at risk.⁴ In other words, a R0 of 3 means that on average, one person will infect 3 other people.

R0 describes the number of people one sick person may infect on average. Graphic Source: Adam Cole/NPR

R0 is calculated using statistical models that are based on a variety of assumptions. It is important to remember — and convey when utilizing this statistic — that R0 is not a fixed value.⁴ Rather, R0 is influenced by an array of factors, including individual-level characteristics, like how contagious a person is once they are infected and how susceptible their close contacts are to infection.⁵ Community-level factors also affect R0 and include population density, climate, and public health interventions that reduce transmission, such as isolation and hand-washing.⁵

Studies from China estimated that the R0 for COVID-19 is approximately 2⁶.⁷ Studies in both Wuhan and on the Diamond Princess Cruise ship both found that public health interventions significantly reduced the R0 in both settings. For example, initial estimates on the Diamond Princess suggested the R0 was 14.8 prior to isolation and quarantine measures. After these interventions, however, the R0 decreased to 1.48.⁸

Case hospitalization rate

This statistic describes the proportion of people with COVID-19 who are hospitalized over a specified period of time per specified unit of population. Because of COVID-19’s differential effect on older populations, this number is most frequently presented by age group.³ The rate consists of the following:

[# of COVID-19 cases who are admitted to hospital / # of total cases over a specified period of time] * 100,000

For COVID-19, the CDC estimates the case hospitalization rate to be 4.6 per 100,000 people. The rate increases with age to 12.2 cases per 100,000 for people 65–74 and 17.2 per 100,00 for people 85 years and older.⁹ Regularly updated hospitalization rates from CDC surveillance sites across the country are available here.

Case fatality rate

This measure describes the proportion of people with COVID-19 who lose their lives as a result of the infection and expresses the severity of the illness.¹⁰ This measure is usually calculated using a specified period of time (ie: the number of cases identified between March and April 2020) and is expressed as a percentage. Note that the numerator includes only the deaths that occur among cases included in the denominator.

# deaths caused by COVID-19 among cases / total # of cases with COVID-19 over a specified time interval.10

Based on available data, the CDC estimates that the case fatality rate for COVID-19 ranges between 1.8–3.4%.¹¹ However, this rate varies by population (age group, comorbidities, gender, etc.). Case fatality rates for COVID-19, SARS, and influenza are presented in the table below, which presents WHO and CDC estimates compiled by Ruan, et al. in this recent article.¹²

Table 1: Case fatality rates across overall age group for SARS, COVID-19, and influenza from Ruan et al. Full table with age-group specific rates available here.

Bringing public health to the public.

We believe that an informed society is a healthy society. Our motto around the office — that our job is to democratize data — holds even in times of pandemic. We think that understanding the numbers, and how the public may correctly or incorrectly interpret them, is key to balancing the need for calm with the need for action. The world isn’t going to be the same, and it shouldn’t be. Throughout our response, recovery, and resilience planning — public health must take a front seat. As we move through this together, we must continue to democratize the practice.

As always, we are here to help. Visit our COVID-19 Resources pages to see the latest resources we are making available, and connect with us if you have any questions, are curious about how we can work together, or just to say hello.