COVID-19 data accuracy would be unacceptable in the business world

If IBM managed data as ineptly as many state-run medical operations are currently doing during this COVID-19 panic-demic, it would have gone out of business years ago.

IBM is a company that deals with a lot of data, and it works hard to get those data right, harder than any company I've ever observed.  If I've learned anything from my years with IBM, getting the data right is a hard job.  IBM has spent years perfecting methods for getting data right, from validating the source data to guaranteeing it processes those data with no corrupting influences.  Why?  It affects every business element of the company, and more importantly, it ensures that when IBM publishes its business results, those results are accurate and believable.

How does IBM do this?  As a business, IBM has a large number of processes that make the company hum.  These business processes all have designated business process owners who manage them and are responsible for their compliance with standards.  Those not familiar with how a business process is constructed, think of it as a series of steps to get work done.  These business process steps are enabled by computing technology that collects and processes data.  Every business process within IBM has to pass an information processing audit.  If changes are introduced to any process, it has to be re-audited.  I have been through a lot of these audits, and they are rigorous and brutally honest.  IBM calls this data compliance program "Application Systems Control and Auditability" (ASCA).  Its role is to ascertain and certify that the IBM Corporation's information assets are protected, and that’s why you never hear about IBM's information systems being compromised.

As an observer, a veteran of rigorous data collection and management methods, I'm watching the COVID-19 goat rodeo being executed by many different states' health organizations.  From my perspective, this is what I see: no auditing of the source data, no protection of the data from corrupting influences, no named owners of any of the processes involved.  There's no guarantee that any of the data the public and policymakers are using is accurate.  I would submit when news media and social media folks turn up one problem after another, from miscategorization of what qualifies as a COVID-19 death to over-counting, the numbers are worthless.  In the IBM world, this would be unacceptable, but I guess in the world of government, it's good enough for government work.

It would seem that many states weren't ready for processing data needed for making critical medical and health policy decisions during an epidemic.  I also suspect they have rudimentary analytical programs for crunching this data.  I'm guessing once this panic-demic is over, the data collected will hit the circular file if manually compiled, and the bit-bucket if digitally stored.

J.H. Capron is an author and writer living in the Hudson Valley of N.Y. and has worked with information processing systems since the punch card days.

If IBM managed data as ineptly as many state-run medical operations are currently doing during this COVID-19 panic-demic, it would have gone out of business years ago.

IBM is a company that deals with a lot of data, and it works hard to get those data right, harder than any company I've ever observed.  If I've learned anything from my years with IBM, getting the data right is a hard job.  IBM has spent years perfecting methods for getting data right, from validating the source data to guaranteeing it processes those data with no corrupting influences.  Why?  It affects every business element of the company, and more importantly, it ensures that when IBM publishes its business results, those results are accurate and believable.

How does IBM do this?  As a business, IBM has a large number of processes that make the company hum.  These business processes all have designated business process owners who manage them and are responsible for their compliance with standards.  Those not familiar with how a business process is constructed, think of it as a series of steps to get work done.  These business process steps are enabled by computing technology that collects and processes data.  Every business process within IBM has to pass an information processing audit.  If changes are introduced to any process, it has to be re-audited.  I have been through a lot of these audits, and they are rigorous and brutally honest.  IBM calls this data compliance program "Application Systems Control and Auditability" (ASCA).  Its role is to ascertain and certify that the IBM Corporation's information assets are protected, and that’s why you never hear about IBM's information systems being compromised.

As an observer, a veteran of rigorous data collection and management methods, I'm watching the COVID-19 goat rodeo being executed by many different states' health organizations.  From my perspective, this is what I see: no auditing of the source data, no protection of the data from corrupting influences, no named owners of any of the processes involved.  There's no guarantee that any of the data the public and policymakers are using is accurate.  I would submit when news media and social media folks turn up one problem after another, from miscategorization of what qualifies as a COVID-19 death to over-counting, the numbers are worthless.  In the IBM world, this would be unacceptable, but I guess in the world of government, it's good enough for government work.

It would seem that many states weren't ready for processing data needed for making critical medical and health policy decisions during an epidemic.  I also suspect they have rudimentary analytical programs for crunching this data.  I'm guessing once this panic-demic is over, the data collected will hit the circular file if manually compiled, and the bit-bucket if digitally stored.

J.H. Capron is an author and writer living in the Hudson Valley of N.Y. and has worked with information processing systems since the punch card days.