Design And Implementation Of An Automated E-mail Notification System For Results Of Tests Pending At Discharge

Slide 1

Design And Implementation Of An Automated E-mail Notification System For Results Of Tests Pending At Discharge

Anuj K Dalal, MD, FHM, Jeffery L Schnipper, MD, MPH, Eric G Poon, MD, MPH, Kathleen Rossi-Roh, MEd, Allison Macleay, Deborah H Williams, MHA, Catherine L Liang, MPH, Nyryan V Nolido, MA, Jonas Budris, David W Bates, MD, MSc, Christopher L Roy, MD

Brigham and Women's Hospital
Partners Healthcare, Inc

Slide 2

Case

• A 55 y/o F is hospitalized for an unprovoked DVT and PE.
• A hypercoaguable work-up is sent by the inpatient medical team but is pending at discharge.
• Results return a week later but are neither reviewed by the responsible inpatient attending nor communicated to the PCP.

Slide 3

Background: Tests Pending At Discharge (TPADs)

• Physicians are aware of approximately 40% of the final results of TPADs.*
• Few institutions have standardized systems to manage TPADs.
• Automated systems using "push" notification may improve awareness of TPADs:
  • Successful development and implementation requires integration with inpatient workflow, clinical information systems, and institutional culture.

*Roy CL, Poon EG, Karson AS, Ladak-Merchant Z, Johnson RE, Maviglia SM, Gandhi TK. Patient safety concerns arising from test results that return after hospital discharge. Ann Intern Med 2005 Jul 19;143(2):121-8

Slide 4

Purpose: Automated E-mail Notification System

• Identify patients discharged with TPADs.
• Notify responsible physicians of the finalized results of TPADs via secure, network E-mail:
  • TO Discharging Inpatient Attending.
  • CC Primary Care Physicians (network PCPs).
• 3 E-mail Notification Types:
  • Chemistry, Hematology.
  • Radiology, Pathology.
  • Microbiology (culture and non-culture).

Slide 5

Design of System: A coordinated sequence of events

E-mail Notification Types:

1. Chem/Heme.
2. Rad/Path.
3. Micro.

Discharge time entered by unit clerk
Identifies providers from administrative databases
Files TPADs in a queue
Updates status of TPADs at 12:01 AM
Any TPAD finalized?
E-mail sent: TO Attending CC PCP (network)

1. Triggers E-mails until all non-suppressed TPADs finalized.
2. Limits one E-mail to provider per notification type per day.
3. Sends E-mails to inpatient attendings and network PCPs.

Slide 6

Design Considerations: Alert Fatigue

Goal: Maximize utility of system by timely notification of relevant results
Important Questions:
Should we notify providers on only abnormal results?
Should we notify providers on negative results, and if so which ones?
Should we exclude only commonly ordered inpatient results with fast turn-around (i.e., all basic metabolic panels, CBC's, coags etc.)?

Slide 7

Design Considerations: Alert Fatigue

• Incorporated logic to suppress certain, inpatient-specific, non-essential TPADs, modifiable "on-the-fly".
• Kept to a minimum during pilot period to see what is coming thru (i.e., kept sensitivity high):
  • Chemistry: ABG, VBG.
  • Hematology: RBC, MCV, MCH, MCHC, Diff Count.
  • Radiology:
    • Fluoroscopy use.
    • Uploaded outside hospital images (no reports generated).
  • Pathology/Microbiology: none.
• Limited notification volume to no more than one E-mail per notification type per day, until all TPADs finalized:
  • Micro alerts: after initial notification, sent subsequent notifications only on abnormal results

Slide 8

Chemistry/Hematology Notification

Image: A sample Chemistry/Hematology Notification is shown.

Slide 9

Image: A sample Chemistry/Hematology Notification is shown.

Slide 10

Image: A sample Chemistry/Hematology Notification is shown.

Slide 11

Measures

• Background Performance:
  • What's happening "under-the-hood"?
    • TPAD processing: volume, % flagged abnormal, % suppressed by rules.
    • Reliability: discharge time, provider identification.
• E-mail Notification Performance:
  • What did physicians see?
    • Volume of notifications by discharged patient, provider, and test type?
• User Satisfaction:
  • Surveys to inpatient attending physicians receiving E-mail notifications.

Slide 12

Background Performance

Discharge time entered by unit clerk:

• 83 discharged patients (~ 1 month, general medicine, cardiology):
  • 1 incorrect discharge time stamp.

Identifies providers from administrative databases
Files TPADs in a queue:

• Detected 405 chem/heme TPADs:
  • 4.9 per patient.
  • 264 chemistry (65%), 141 hematology (35%).
  • 73 flagged abnormal (18%).

Updates status of TPADs at 12:01 AM
Any TPAD finalized?
E-mail sent: TO Attending CC PCP (network)
136 E-mails triggered

• ~1.7 alerts per discharged patient (136/82).
• 2 or more E-mails on 34% (28/82).

Slide 13

Background Performance

• One incorrectly entered discharge time stamp (1.2%, 1/83):
  • Unit clerk inadvertently "discharged" patient on Day 4 of 10-day hospitalization:
    • Detected 510 TPADs (249 chem, 261 heme)!
  • Triggered 9 E-mails!.
  • A rare event.
• 3 responses from physicians (on 3 distinct patients) claiming E-mail sent to incorrect provider (3.6%, 3/83):
  • 2 from Inpatient Attending.
  • 1 from PCP.

Slide 14

E-mail Notifications By Discharged Patient

Image: A chart of E-mail Notifications By Discharged Patient is shown.

E-mail notifications sent to providers of 95 discharged patients with all TPADs finalized.

Slide 15

E-mail Notifications By Inpatient Provider

A chart showing the E-mail Notifications By Inpatient Provider is shown.

E-mail notifications received by 35 distinct inpatient physicians from 3/15/11 thru 4/15/11

Slide 16

User Satisfaction

• Inpatient physicians surveyed 3 days after receiving the final notification E-mail.
• Asked to rate satisfaction on 5-point Likert scale.
• 70 survey responses (29 physicians):
  • 84% satisfied or very satisfied.
  • 6% neutral.
  • 10% dissatisfied or very dissatisfied.

Slide 17

Selected Comments

"I find this extremely useful, knowing the final results of tests, both test results that are positive as well as negative."

"Was unaware of this test even being ordered—had it not been for auto-notification, would never have known about test or result. No call to PCP as test is in normal range and will not affect management."

"The concept is great. All the notifications I have received are for negative results. Might be more worthwhile for blood tests if it was only for abnormal results."

"Test was not needed and was not ordered by me..."

"It is best to send these pathology results not just to the ordering physician but also the GI physician performing the biopsy."

Slide 18

Conclusions

• Automated E-mail notification is a feasible and reliable strategy for managing results of TPADs, and compatible with inpatient workflow.
• Successful implementation is dependent on accuracy and reliability of:
  • Discharge time stamp.
  • Provider identification.

"Garbage in, garbage out" phenomena

Slide 19

Conclusions

• The high volume of TPADs and notifications to providers are challenging to negotiate:
  • Logic to limit volume and frequency of notifications to minimize alert fatigue.
  • A user-configurable system to modify suppression rules is desirable.
• Users are highly satisfied:
  • Physicians vary with regard to type of results they wish to receive.

Slide 20

Thank You!

• Questions?
• Comments?

Slide 21

Slide 22

Acknowledgements: TPAD Team

• BWH Information Systems Project Team:
  • Kathleen Ross-Roh, MEd.
  • Allison Macleay.
  • Margo Daphnis.
  • Eric Poon, MD, MPH.
• BWH DGM Research Team:
  • Catherine Liang, MPH.
  • Nyryan V Nolido, MA.
  • Deborah H Williams, MHA.
  • Jonas Budris.
  • Catherine Yoon, MSc.
  • David Bates, MD, MSc.
  • Christopher Roy, MD.
  • Jeffrey Schnipper, MD, MPH.

Slide 23

Funding

• Brigham and Women's Healthcare Information Technology Innovation Program.
• Agency for Healthcare Research and Quality, grant number R21HS018229.

Slide 24

Configurable System: Lab Selection

Image: A screen shot of the BICS Terminal Emulator is shown.

Slide 25

Microbiology Notification

1. April 17, 2011
2. Dear Dr. HOSPITALIST, M.D.:
   
   DISCHARGED PATIENT (BWH# 12345678), for whom you were the attending of record, was discharged from Brigham and Women's Hospital on 03/27/2011. Some tests from this hospitalization were still pending at the time of discharge. We have listed below 1) tests whose results have been finalized after discharge, and 2) tests whose results are still pending. Chemistry and Hematology test types are included in this service. Radiology, Pathology, and Microbiology test types are available in separate notifications
3. 
   The patient's PCP, NON-NETWORK PROVIDER, did not receive this notification because s/he does not have a Partners E-mail address listed.
4. 
   This is a new service we are piloting that we hope you will find to be helpful. Note: Any corrections or changes made after tests are finalized are not captured by this service but are reported per current lab protocol.
5. 
   Inpatient Attending: HOSPITALIST, M.D. Work Phone: 111-111-1111
   Primary Care Physician: NON-NETWORK PROVIDER, M.D. Work Phone: 222-222-2222
6. 
   
   Microbology Results FINALIZED
7. Specimen            Test              Result           Date Collected            Date Resulted
8. B1052302 BLOOD      BLOOD CULTURE      NO GROWTH             04/10/2011           04/16/2011
9. B1052303 BLOOD      BLOOD CULTURE      NO GROWTH             04/10/2011           04/16/2011
10. B1052164 BLOOD      BLOOD CULTURE      KLEBSIELLA PNEUMONIAE    04/09/2011           04/15/2011
11. B1052165 BLOOD      BLOOD CULTURE      KLEBSIELLA PNEUMONIAE    04/09/2011           04/15/2011

Slide 26

Radiology/Pathology Notification

March 21, 2011
Dear Dr. HOSPITALIST, M.D.:

DISCHARGED PATIENT (BWH# 12345678), for whom you were the attending of record, was discharged from Brigham and Women's Hospital on 03/27/2011. Some tests from this hospitalization were still pending at the time of discharge. We have listed below 1) tests whose results have been finalized after discharge, and 2) tests whose results are still pending. Chemistry and Hematology test types are included in this service. Radiology, Pathology, and Microbiology test types are available in separate notifications

The patient's PCP, NON-NETWORK PROVIDER, did not receive this notification because s/he does not have a Partners E-mail address listed.

This is a new service we are piloting that we hope you will find to be helpful. Note: Any corrections or changes made after tests are finalized are not captured by this service but are reported per current lab protocol.

Inpatient Attending: HOSPITALIST, M.D. Work Phone: 111-111-1111
Primary Care Physician: NON-NETWORK PROVIDER, M.D. Work Phone: 222-222-2222

Pathology Results FINALIZED

Specimen: SURG,Gastric biopsy  Date Resulted: 03/20/2011

PATHOLOGIC DIAGNOSIS:

A. DUODENUM: Mild chronic active duodenitis.

B. GASTRIC ANTRUM: Antral mucosa with moderate chronic active gastritis. HELICOBACTER PYLORI ARE SEEN ON H&E STAIN. No intestinal metaplasia identified.

C. GASTRIC BODY: Corpus mucosa with mild chronic active gastritis. NUMEROUS HELICOBACTER PYLORI ARE SEEN ON H&E STAIN. No intestinal metaplasia identified.

D. DISTAL ESOPHAGUS: Squamous mucosa with active esophagitis with erosion. A PAS-D stain is negative for fungal forms.

Slide 27

What's in the E-mails?

Slide 28

Limitations

• Single institution within an large integrated healthcare network.
• A robust culture of E-mail utilization by inpatient physicians.
• Variable reliability of processes to enter discharge time and maintain administrative databases by clinical service and institution → limit adoption.

Slide 29

Implications

• "Push" notification is an acceptable strategy to manage results of TPADs.
• Future studies:
  • Evaluate impact on physician awareness.
  • Analyze downstream actions taken.
  • Elucidate desired features to maximize utility for physicians.