Responsible Engineering in Healthcare
Transcript of Responsible Engineering in Healthcare
![Page 1: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/1.jpg)
Responsible Engineering in Healthcare
John Wood, PhD Thom Walsh, PhD November 1, 2014
Copyright © 2014 by GCorp Consulting. Permission granted to INCOSE to publish and use.
![Page 2: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/2.jpg)
First, do no harm. Healthcare has a need for Systems Engineering;
however, solutions based upon common but faulty assumptions will cause harm.
2
![Page 3: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/3.jpg)
3
!
NEW HEALTCHCARE
PARADIGMS MAY
LEAVE YOU DAZED
AND CONFUSED
![Page 4: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/4.jpg)
Theoretical Tasks
1. Minimizing costs by reducing Length of Stay
2. Reducing 30-day readmissions with technology
3. Increasing patient satisfaction by reducing wait times
4
![Page 5: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/5.jpg)
Task #1: Minimizing costs by reducing
Length of Stay
5
![Page 6: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/6.jpg)
Background
• Average Length of Stay (LOS) for certain procedures varies among hospitals in same system.
• 1,600 bed-days and associated costs could be avoided if all hospitals achieved the minimum LOS.
• Bonus: Those beds would be available for other procedures.
6
![Page 7: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/7.jpg)
“Solution”
Implement strict rules on LOS for certain heart conditions and require justification
for anything over the minimum LOS.
7
![Page 8: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/8.jpg)
Results…as expected
After 6 weeks, average LOS dropped by 15%.
8
![Page 9: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/9.jpg)
Surprise!
After 12 weeks, extra capacity filled with
new (unique) heart patients.
9
![Page 10: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/10.jpg)
Why?
Supply sensitive care.
10
![Page 11: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/11.jpg)
The Three Types of Care
• Necessary and Effective (~12%) – “One right answer” – Example: Beta-blockers for heart attacks victims
• Preference Sensitive (~25%) – Multiple viable options – Example: Choice between mastectomy and lumpectomy
• Supply Sensitive (~63%) – Influenced by regional supply – Example: Invasive interventions for cardiac disease – Note: More care is not correlated with better outcomes!
• Additional Reading: Tracking Medicine by Jack Wennberg
11
![Page 12: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/12.jpg)
Task #2: Reducing 30-day readmissions
with technology
12
Spillover: Patients released too soon are returning with complications.
![Page 13: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/13.jpg)
Background
Finance team notices high cost of 30-day readmissions for patients returning
with heart conditions.
13
![Page 14: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/14.jpg)
“Solution”
Technology director has a plan! Provide patients with a take-home sensor kit.
14
![Page 15: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/15.jpg)
Results…as expected
• Nearly 20% of patients asked to return for screening after “anomalies” are found.
• Some of these patients were in need of treatment.
15
![Page 16: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/16.jpg)
Surprise!
• Screening team is overloaded.
• Patient wait times increase dramatically.
16
![Page 17: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/17.jpg)
Why?
• Tech team did not coordinate with the lab.
• In-home monitors identify numerous “anomalies.”
17
![Page 18: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/18.jpg)
The Problem of Inference
• No statistical analysis on “new” home data
• Reliance on humans for inference (turning data into knowledge) is error-prone
• Accuracy of inferences is hard to judge
• Resulting actions may be ineffectual or harmful
• Additional Reading: Frontiers in Massive Data Analysis from the National Research Council
18
![Page 19: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/19.jpg)
Task #3: Increasing patient satisfaction by
reducing wait times
19
Spillover: Patient satisfaction is decreasing due to increased screening.
![Page 20: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/20.jpg)
Background
• Patient satisfaction scores recently dropped.
• Lab director “knows” the problem is due to the long lines.
20
![Page 21: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/21.jpg)
“Solution”
Lean team helps re-map the screening process.
21
![Page 22: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/22.jpg)
Results…as expected
• Wait times are nearly eliminated.
• Throughput increases 30%.
22
![Page 23: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/23.jpg)
Surprise!
Patient satisfaction scores fall even lower.
23
![Page 24: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/24.jpg)
Why?
There are 6 possible outcomes to any screening, but only 1 outcome benefits the health of the patient.
24
![Page 25: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/25.jpg)
The 6 Possible Outcomes of any Screening (part 1)
• True negative. Patient is subject to costs, risks, and anxiety of testing but receives no health benefit.
• False negative. Patient is wrongly reassured.
• False Positive. Patient is subject to risks, costs, and anxiety of further testing.
25
![Page 26: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/26.jpg)
The 6 Possible Outcomes of any Screening (part 2)
• True Positive but patient is not destined to be affected by the disease. Patient is subject to risks, costs, and anxiety of further treatment.
• True positive but early detection will not affect the outcomes of treatment. Patient is subject to risks, costs, and anxiety of further treatment.
26
![Page 27: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/27.jpg)
The 6 Possible Outcomes of any Screening (part 3)
• True positive and early treatment prevents complications. Patient receives timely care.
• Additional Reading: Overdiagnosed by Gil Welch
27
![Page 28: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/28.jpg)
In the end…
• Cost saving initiative drove up costs
– More readmissions (induced by inappropriate LOS)
– More screenings (from 30-day readmission “fix”)
• Cost saving initiative reduced reimbursement
– Reimbursements are tied to patient satisfaction scores which decreased (due to “overdiagnosis”)
28
![Page 29: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/29.jpg)
How a Systems Engineer could have stopped the cascade of bad events
• Task 1 – Requirements elicitation would focus solutions on affecting total patient costs.
• Task 2 – Root-cause analysis would have identified new LOS rules as the problem.
• Task 3 – Root-cause analysis would have identified true patient concerns.
• Spillovers – Risk analysis/management would have identified/mitigated second order affects.
29
![Page 30: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/30.jpg)
Time for Q & A
30
![Page 31: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/31.jpg)
Wrap up: Responsible Engineering
• Be sure to understand the overarching goal – Proper requirements elicitation!
• Take time to identify root cause before forming solutions
• Perform thorough risk analysis – Anticipate potential second and third order effects
– Know the peculiarities of healthcare
• And remember:
31
First, do no harm!
![Page 32: Responsible Engineering in Healthcare](https://reader033.fdocuments.in/reader033/viewer/2022052013/6286043f1b51f81583547b05/html5/thumbnails/32.jpg)
John Wood, PhD [email protected]
(760) 456-9431
http://www.gcorpconsultingservices.com/hs
Thom Walsh, PhD [email protected]
(626) 817-3337