Information Latency: Why Don’t We Change?

I have had this note to self for a while so I am finally going to put a quick entry out here on the topic.

The issue is data latency or more appropriately information latency.  The data often exists right away, but the challenge is how to you get the data into a usable form, with context, and with enough data to make decisions.

In communications, this manifests itself in healthcare in two ways that immediately jump to mind:

1. In a traditional letter program:
   • You send a letter to a patient (7-10 days from data targeting to mailbox)
   • Patient opens the letter and has to contact their physician (if they choose to do anything)
   • Patient trades messages with physician and/or has to schedule an appointment
   • Patient meets with physician who (for example) writes them a new prescription
   • Patient waits for medication to run out then refills with new drug (e.g., generic, on-formulary drug)
   • Claims get aggregated and reports run
   • Best case – 30+ days to see if program had any effect (most likely 6 months)
2. In a traditional survey:
   • Company prints a survey and mails it to 10,000 people hoping for a 10% response rate to get a statistically valid sample size of 1,000
   • Patients fill out the survey over the next month and mail them to a data entry company
   • Data entry company manually enters them, aggregates the data, and creates a report
   • 45-60 days later the company has information from the survey

Of course, the issue with both of these is that you have lost a huge window of time especially if you need to make changes to your program or the survey tells you that you need to gather more information.

Why don’t more companies talk about on-the-fly program changes and how to use modern technology to get real-time feedback for programs where they can pause the program, make change (e.g., change the message, add a new question), and then continue the program?