Studies of the Informational Value of Clinical Call Center Data

Espino et al. (2003) used correlation analysis to test a hypothesis that call center data could provide more timely detection of influenza than reference data collected by the Centers for Disease Control and Prevention (CDC). They studied data from two types of call centers—emergency room telephone triage and after hours telephone triage—and one outbreak of influenza. In emergency telephone triage, an insured person calls a telephone triage center to obtain authorization to visit an emergency room. In after-hours telephone triage, a person calls a triage center instead of his or her physician.

The data sets obtained from the call centers included the following information:

1. Start date/time of initial call.

2. Acuity. Acute or nonacute.

3. Inclination of the caller. There were 12 unique reasons why a call could be made (e.g., call physician, research).

4. Disposition assigned to the call by the nurse. There were

23 unique dispositions (e.g., home care, see physician within

24 hours, call primary care physician within 24 hours, emergency room immediately).

5. Call outcome. This element used the same 23 possible values used in the disposition.

6. Primary symptoms expressed in free text. Examples are "stomach bug—six loose bowel movements throughout the day'' and "feels hot and some pain in throat.''

7. Five-digit home zip code.

9. Gender.

10. Treatment guideline used. There were 380 unique guidelines used in the data sets. A single guideline is recorded for each call. Examples include diarrhea, earache, breathing difficulty, vomiting, and asthma attack.

They studied the treatment guideline (item 10) used to handle the call. They assigned 21 of the specific guidelines used for the calls to respiratory syndrome and four to constitutional syndrome.They combined the calls for respiratory and for constitutional and studied the weekly counts of the combined categories.

Reference standards for influenza included three types of influenza surveillance data from the CDC:

Weekly state epidemiologist reports of influenza activity (CDC-SI).

TABLE 28.3 Maximum Correlation and Time Lag (in Weeks) Between Influenza Calls to Two Call Centers and Reference Influenza Time Series Data

Was this article helpful?

0 0

Post a comment