Back

Governing Decisions of Probability Cutoffs in Clinical AI Deployment: A Case Study of Asthma Exacerbation Prediction

Zheng, L.; Agnikula Kshatriya, B. S.; Ohde, J.; Rost, L.; Malik, M.; Peterson, K.; Brereton, T.; Loufek, B.; Pereira, T.; Gai, C.; Park, M.; Hartz, M.; Fladager-Muth, J.; Wi, C.-I.; Tao, C. J.; Garovic, V.; Juhn, Y. J.; Overgaard, S. M.

2026-03-22 health informatics
10.64898/2026.03.18.26348562 medRxiv
Show abstract

Models that estimate the probability of an adverse clinical outcome require an operational cutoff to translate continuous estimated probabilities into discrete labels that can trigger clinical action. Although statistical methods identify optimal cut-offs, threshold selection ultimately reflects value judgments regarding harm tolerance, resource allocation, and workflow feasibility. We describe a governance-informed approach to selecting a deployment threshold for an asthma exacerbation (AE) prediction model integrated into clinical workflows. Using prevalence-adjusted performance metrics and real-world provider capacity modeling, we evaluated multiple candidate thresholds and quantified downstream workload and missed-event trade-offs. We demonstrate that statistically optimal thresholds may produce operationally infeasible alert volumes or unacceptable miss rates. We propose a structured threshold governance framework integrating statistical performance, clinical utility, stakeholder input, and human oversight safeguards. This case illustrates how threshold decisions should be treated as organizational governance processes rather than purely technical optimizations.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

1
Journal of the American Medical Informatics Association
61 papers in training set
Top 0.1%
26.1%
2
npj Digital Medicine
97 papers in training set
Top 0.2%
18.7%
3
JAMIA Open
37 papers in training set
Top 0.2%
6.4%
50% of probability mass above
4
PLOS Computational Biology
1633 papers in training set
Top 7%
4.9%
5
PLOS ONE
4510 papers in training set
Top 34%
4.3%
6
PLOS Digital Health
91 papers in training set
Top 0.6%
4.0%
7
Scientific Reports
3102 papers in training set
Top 31%
4.0%
8
Nature Communications
4913 papers in training set
Top 39%
3.6%
9
Journal of Medical Internet Research
85 papers in training set
Top 2%
1.8%
10
Proceedings of the National Academy of Sciences
2130 papers in training set
Top 32%
1.7%
11
eLife
5422 papers in training set
Top 43%
1.7%
12
JCO Clinical Cancer Informatics
18 papers in training set
Top 0.5%
1.5%
13
BMC Medical Informatics and Decision Making
39 papers in training set
Top 2%
1.1%
14
JMIR Medical Informatics
17 papers in training set
Top 1%
0.9%
15
BMJ Health & Care Informatics
13 papers in training set
Top 0.8%
0.8%
16
Frontiers in Digital Health
20 papers in training set
Top 1%
0.8%
17
Informatics in Medicine Unlocked
21 papers in training set
Top 1%
0.8%
18
International Journal of Medical Informatics
25 papers in training set
Top 2%
0.8%
19
Medical Decision Making
10 papers in training set
Top 0.3%
0.8%
20
Nature Medicine
117 papers in training set
Top 5%
0.8%
21
Physical Biology
43 papers in training set
Top 2%
0.8%
22
Journal of Biomedical Informatics
45 papers in training set
Top 1%
0.8%
23
Physical Review X
23 papers in training set
Top 0.6%
0.7%
24
Scientific Data
174 papers in training set
Top 3%
0.7%
25
The Lancet Digital Health
25 papers in training set
Top 2%
0.5%
26
Critical Care Explorations
15 papers in training set
Top 0.6%
0.5%
27
Annals of Internal Medicine
27 papers in training set
Top 1%
0.5%