You are currently browsing the tag archive for the ‘measures’ tag.

You Cannot Manage What You Cannot Measure:

How to Implement Evidence-Based COVID-19 Strategy

By Shlomo Maital

  As a management educator, I stress a simple principle: Management begins with measurement. What you do not, cannot, measure, you cannot, will not, manage. But you have to measure wisely, correctly, accurately, and promptly.

   Man, does this ever apply to the chaos we find ourselves now, in my country Israel and in other hotspot countries like the US!

   So, what should we be measuring?   Here is a thorough, reasoned proposal by Tom Frieden and Cyrus Shahpar.   Frieden is a former director of the Centers for Disease Control and Prevention, runs the nonprofit group Resolve to Save Lives, and Dr. Shahpar is the director of a team devoted to preventing epidemics.   * https://www.nytimes.com/2020/07/21/opinion/coronavirus-state-data.html     Here is the URL for the list of 15 (if it is hard to read the jpeg’s below: https://preventepidemics.org/covid19/resources/indicators/

     Resolve to Save Lives, a coalition of national, state and academic partners including the American Public Health Association and the Johns Hopkins Center for Health Security, has developed a list of 15 indicators. Their report argues: Every state and county should be able to collect and publish nine of these immediately and the other six within a few weeks.

       If we had these measures, our leaders, policymakers and the general public would know far better where we stand, what the goals are, how we are doing, and what lies ahead.

       The mass media continue to report on new cases, total cases, and deaths. The result is misleading, because, for instance, “very ill” cases lag behind new cases by two weeks or more.

       Here are the 15 measures that we really need, to manage the pandemic. CLI is COVID-like illness; ILI is influenza-like illness. PCR is polymerase chain reaction, widely used to rapidly make millions to billions of copies of a specific DNA sample, to detect, e.g., coronavirus. No country, I believe, has the full set of 15 – and very few countries have even a partial set of the 15.

       So ideally:   A team of epidemiologists, virologists and statisticians join together, and put in place a system for collecting data for the 15 indicators. In the US, this should be the Center for Disease Control and Prevention (notice: the acronym CDC omits the all-important P, Prevention!). The results are shown on a dashboard, simple, clear and understandable. And all of us can see for ourselves where we are, what is happening, what’s good, what’s bad, and how close we are to our goals. And, derived from the dashboard, what each of us needs to do to help reach the measured goals.

 

   

Unpacking the COVID-19 Stats:  Four Measures

By Shlomo Maital

 

 The COVID-19 statistics can be very confusing, and the press has not been great at dispelling the fog. Here is how the New York Times explains the four key measures:    

  • By Nate CohnJosh KatzMargot Sanger-Katz and Kevin Quealy March 27, 2020
  •  1. Cases per 1,000 people. This is a measure of the prevalence of coronavirus in the community. Confirmed Cases per 1,000 residents.   An imperfect measure, because of spotty testing – are there few cases because of limited testing, or few cases because testing has shown this to be true? Hard to know.

 

  • Confirmed cases per 1,000 by metro area
  •    New York City, at 2.15, is just below #1, Wuhan, China, 4.59, and Lombardy Region, Italy, 3.48, but Albany Ga. And New Orleans are surprisingly high, too, as ‘hot spots’, at 1.35, and 1.32, respectively.

 

  • * HOW WIDESPREAD IS THE CORONAVIRUS IN THE COMMUNITY OR CITY?
  • Each measure answers a different question.
METRO AREA POPULATION CASES PER THOUSAND
Wuhan, China 11.1 mil. 50,821 4.59
Lombardy region, Italy 10 mil. 34,889 3.48
New York 20 mil. 43,016 2.15

 

*HOW DEADLY IS THE VIRUS? WHAT IS THE DEATH RATE?

  1. Deaths per 1,000 people. This measure is likely to be more accurate, alas, than other measures. But this measure lags the number of infections by several weeks…

Deaths per capita are currently higher in the New York City area than in most other places.

METRO AREA POPULATION DEATHS PER THOUSAND
Lombardy region, Italy 10 mil. 4,861 0.48
Wuhan, China 11.1 mil. 2,535 0.23
Albany, Ga. 153,000 10 0.07
New Orleans 1.3 mil. 65 0.05
Seattle 3.9 mil. 133 0.03
Burlington, Vt. 221,000 6 0.03
New York 20 mil. 500 0.03

 

HOW FAST IS THE CORONAVIRUS SPREADING?

  1. Growth Rates Over Time.

Measure: the rate of cumulative cases over time, averaged over the previous week. This helps us learn, is the epidemic getting better or worse? Are we at the ‘apex’ (worst is behind us) or before it?

A growth rate of 40 percent on this chart means the cumulative number of cases is growing by 40 percent every day. A rate of 100 percent would mean that the number of cases was doubling daily.

March 1         March 8                         March 22

 

HOW WELL HAS THE COMMUNITY OR CITY DONE, IN SLOWING DONE THE RATE AT WHICH THE CORONAVIRUS IS GROWING?

  1. Growth Rates by Case Count.

This measure is the rate of growth of the number of cases in a given place – it measures whether a community has slowed the rate of growth, before there is many cases. I.e., is the community flattening the curve?

   Seattle and San Francisco succeeded in flattening the curve. How come? These measures help us ask the right questions

AVERAGE DAILY CHANGE IN TOTAL CASES, OVER THE PREVIOUS 7 DAYS

CONFIRMED CASES PER THOUSAND PEOPLE

= = = = = =

We need to be cautious when interpreting coronavirus statistics. And a wide variety of stats are being tossed at us, often by those who do not fully understand them.

 

Blog entries written by Prof. Shlomo Maital

Shlomo Maital

Pages