SmartAlert - Implementing Machine Learning-Driven Clinical Decision Support for Inpatient Laboratory Utilization Reduction
Open accessThe DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index; 10 of the 14 sources its claims cite were checked for support.
Abstract
Repetitive laboratory testing that is unlikely to yield clinically useful information is a common practice that burdens patients and increases health care costs. Education and feedback interventions have limited success, while general test ordering restrictions and electronic alerts impede appropriate clinical care. We introduce and evaluate SmartAlert, a machine learning-driven clinical decision support (CDS) system integrated into the electronic health record that predicts stable laboratory results to reduce unnecessary repeat testing. This Case Study describes the implementation process, challenges, and lessons learned from deploying SmartAlert targeting complete blood count (CBC) utilization in a randomized controlled pilot across 9270 admissions in eight acute care units across two hospitals between August 15, 2024, and March 15, 2025. The results show a significant decrease in the number of CBC results within 52 hours of SmartAlert display (1.54 vs. 1.82; P<0.01) without adverse effect on secondary safety outcomes, representing a 15% relative reduction in repetitive testing. Implementation lessons learned include interpretation of probabilistic model predictions in clinical contexts, stakeholder engagement to define acceptable model behavior, governance processes for deploying a complex model in a clinical environment, user interface design considerations, alignment with clinical operational priorities, and the value of qualitative feedback from end users. In conclusion, a machine learning-driven CDS system backed by a deliberate implementation and governance process can provide precision guidance on inpatient laboratory testing to safely reduce unnecessary repetitive testing. (Funded by the Agency for Science, Technology, and Research and others.).
Overview
CoChat Summary
TL;DR
SmartAlert, an EHR-integrated machine learning clinical decision support tool, was deployed in a randomized pilot to reduce repetitive inpatient complete blood count (CBC) testing. It significantly reduced CBC results within 52 hours of alert display (1.54 vs. 1.82; P<0.01) without adverse effects on secondary safety outcomes.
Key findings
- In a 7-month randomized pilot across 9270 admissions in eight acute care units at two hospitals, the treatment group had 1.54 CBC results within 52 hours of SmartAlert display versus 1.82 in the control group (P<0.01), a 15.4% relative reduction.
- A significant reduction in CBC results was also seen 28 hours after an alert; no significant differences were detected in ICU transfers within 52 hours, encounter-level ICU transfer rate, length of stay, readmission rate, or mortality.
- Of 486 displayed SmartAlerts, 30.1% resulted in order adjustments, 49.6% were overridden, and 20.3% were dismissed; among overrides, “medical necessity” (40.9%) and “more review needed” (38.7%) were most common.
- The model’s positive predictive value remained near the 90% goal in silent prospective evaluation (88.1%, 95.4%, and 93% for white blood cell counts, hemoglobin, and platelet counts, respectively); retrospective recall was 29.3%, 59.5%, and 100%, respectively.
Methods
- Randomized controlled pilot conducted August 15, 2024, to March 15, 2025, in eight acute care inpatient units across two hospitals; each admission was randomly assigned via an Epic SmartDataElement to treatment (alert displayed when workflow criteria met) or control (alert hidden), excluding primary bone marrow transplant and hematology admissions.
- Primary outcome was mean number of CBC results within 52 hours of the alert trigger; secondary outcomes included CBC results within 28 hours, ICU transfers within 52 hours, encounter-level ICU transfer rate, length of stay, readmission rate, and encounter-level mortality.
- Poisson regression compared mean CBC counts; Mann–Whitney U test compared length of stay; Fisher’s exact test compared ICU transfers, readmission, and mortality; 95% confidence intervals were obtained through bootstrapping. Governance included CDS committee review, user codesign interviews with 18 clinicians, silent prospective evaluation, and a FURM assessment.
Limitations
- The paper states the work was conducted at a single academic medical center with established governance committees that may not exist elsewhere, though the architecture and governance process are described as replicable.
- The paper states the pilot focused on general medical and surgical settings, and additional evaluation would be necessary in other settings such as intensive care units and hematology wards.
- Evident from the text: the intervention targeted only CBCs, alerts were displayed only between 7 a.m. and 6 p.m., and data retrieval delays led to batched pre-calculation rather than fully real-time prediction, which may limit immediate generalizability to other tests or workflows.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
5 Valid · 2 Not valid · 0 Not checked · 22 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“Existing mitigation strategies, including clinician education, audit and feedback, financial incentives, and broad ordering restrictions, have limited precision and efficacy.11,12”
§ SmartAlert — Implementing Machine Learning–Driven Clinical Decision Support for Inpatient Laboratory Utilization Reduction > Introduction
What the sources say
- Disputedcites [11] Yeshoua B, Bowman C, Dullea J, Interventions to reduce repetitive ordering of low-value inpatient laboratory tests: a systematic review. BMJ Open Qual 2023; 12: e002128. DOI: 10.1136/bmjoq-2022-002128. PMC10040017 36958791 sourceExperimental — accuracy not yet measured
The Discussion (page not available) says all five strategies reviewed were effective in most studies, contrary to the claim of generally limited efficacy. The Results report that audit and feedback was effective in 11 of 12 studies and policy change in 9 of 9. The review notes that strict ordering limits were unpopular with providers, but does not present that as evidence of limited efficacy; it does not assess financial incentives or intervention precision. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
- Not checked — no full text of the sourcecites [12] Eaton KP, Levy K, Soong C, Evidence-based guidelines to eliminate repetitive laboratory testing. JAMA Intern Med 2017; 177: 1833– 1839. DOI: 10.1001/jamainternmed.2017.5152. 29049500 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“During model development, performance on retrospective test data drawn from 7 years of admissions in our health system was tuned to achieve 90% precision (positive predictive value [PPV]) previously identified by clinicians as meaningful14 with recall (sensitivity) for white blood cell counts, hemoglobin, and platelet counts of 29.3%, 59.5%, and 100%, respectively.19”
§ SmartAlert — Implementing Machine Learning–Driven Clinical Decision Support for Inpatient Laboratory Utilization Reduction > Governance > RETROSPECTIVE MODEL VALIDATION
What the sources say
- Disputedcites [19] Jiang Y, Ma SP, Liang AS, SmartAlert: integrating machine learning and alert triggers into live electronic medical record systems, targeting low-yield inpatient lab tests. May 6, 2026 (https://www.medrxiv.org/content/10.64898/2026.04.29.26351965v1). Preprint. sourceExperimental — accuracy not yet measured
Split into 2 atoms — 0 confirmed, 1 partially supported, 1 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“20%–50% of inpatient laboratory tests are medically unnecessary,2–4”
§ SmartAlert — Implementing Machine Learning–Driven Clinical Decision Support for Inpatient Laboratory Utilization Reduction > Introduction
What the sources say
- Partly supportedcites [2] Zhi M, Ding EL, Theisen-Toupal J, Whelan J, Arnaout R. The landscape of inappropriate laboratory testing: a 15-year meta-analysis. PLoS One 2013; 8: e78962. DOI: 10.1371/journal.pone.0078962. 24260139 PMC3829815 sourceExperimental — accuracy not yet measured
The Results section, “Overall rates of inappropriate laboratory testing, 1997–2012,” reports 20.6% inappropriate overutilization overall; “Overutilization: initial vs. repeat testing” reports 43.9% for initial tests but 7.4% for repeat tests. These figures support substantial unnecessary testing, but they are not an estimate that 20%–50% of inpatient laboratory tests are unnecessary. (LLM, conf=0.87)
- Supportedcites [3] Koch C, Roberts K, Petruccelli C, Morgan DJ. The frequency of unnecessary testing in hospitalized patients. Am J Med 2018; 131: 500– 503. DOI: 10.1016/j.amjmed.2017.11.025. 29224739 PMC8628817 source
The Results report that 31.5% of tests ordered in the first 24 hours of inpatient care were judged unnecessary. The Results table identifies laboratory tests as the most frequent unnecessary test type; its listed laboratory-test counts also yield a proportion within the claimed 20%–50% range. The study concerns the first day of care at two academic medical centers, rather than every day of an inpatient stay. (LLM, conf=0.85)
- Partly supportedcites [4] Xu S, Hom J, Balasubramanian S, Prevalence and predictability of low-yield inpatient laboratory diagnostic tests. JAMA Netw Open 2019; 2: e1910967. DOI: 10.1001/jamanetworkopen.2019.10967. 31509205 PMC6739729 sourceExperimental — accuracy not yet measured
The Introduction reports estimates that 25%–50% of inpatient laboratory testing is medically unnecessary. This supports the substance of the claim, but its stated lower bound is 25%, not 20%. (LLM, conf=0.85)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“This confirmed that the model’s PPV remained near the goal of 90% (e.g., 88.1%, 95.4%, and 93% for white blood cell counts, hemoglobin, and platelet counts, respectively).”
§ SmartAlert — Implementing Machine Learning–Driven Clinical Decision Support for Inpatient Laboratory Utilization Reduction > Governance > USER CODESIGN AND SILENT PROSPECTIVE MODEL PERFORMANCE
What the sources say
- Supportedcites [19] Jiang Y, Ma SP, Liang AS, SmartAlert: integrating machine learning and alert triggers into live electronic medical record systems, targeting low-yield inpatient lab tests. May 6, 2026 (https://www.medrxiv.org/content/10.64898/2026.04.29.26351965v1). Preprint. source
Table 2 (Model performances in prospective silent deployment evaluation settings) lists Precision (%) of 88.1 for WBC, 95.4 for HGB and 93.0 for PLT, matching the three figures attributed in the claim (precision being PPV). The Results section states they 'targeted a threshold of 90% precision when training the model based on prior clinician feedback, which was maintained during prospective deployment', supporting the 'remained near the goal of 90%' framing, and the Methods confirm the evaluation was a silent prospective deployment (March 8-14, 2024, 199 patients). (LLM, conf=0.92)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Qualitative feedback from clinician interviews was inductively coded with consensus. This revealed favorable attitudes toward the use of ML in the alert with high trust and positive sentiment (Fig. 5A).”
§ SmartAlert — Implementing Machine Learning–Driven Clinical Decision Support for Inpatient Laboratory Utilization Reduction > Governance > USER CODESIGN AND SILENT PROSPECTIVE MODEL PERFORMANCE
plainn = 18- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
2 more claims
Create a free account to see all 29 claims and every source verdict.
- It may state two separate claims as one. 19
- Probably not a substantial claim. 11
Citations
Cites
The Landscape of Inappropriate Laboratory Testing: A 15-Year Meta-Analysis
Zhi, M., Ding, E. L., Theisen-Toupal, J., Whelan, J., & Arnaout, R. (2013) · PLoS ONE · doi:10.1371/journal.pone.0078962 · Cited by 535
Thavendiranathan, P., Bagai, A., Ebidia, A., Detsky, A. S., & Choudhry, N. K. (2005) · Journal of General Internal Medicine · doi:10.1111/j.1525-1497.2005.0094.x · Cited by 260
Salisbury, A. C. (2011) · Archives of Internal Medicine · doi:10.1001/archinternmed.2011.361 · Cited by 194
Evidence-Based Guidelines to Eliminate Repetitive Laboratory Testing
Eaton, K. P., Levy, K., Soong, C., Pahwa, A. K., Petrilli, C., Ziemba, J. B., Cho, H. J., Alban, R., Blanck, J. F., & Parsons, A. S. (2017) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2017.5152 · Cited by 178
Choosing wisely in adult hospital medicine: Five opportunities for improved healthcare value
Bulger, J. B., Nickel, W. K., Messler, J., Goldstein, J. D., O'Callaghan, J. P., Aurón, M., & Gulati, M. S. (2013) · Journal of Hospital Medicine · doi:10.1002/jhm.2063 · Cited by 164
Grossman, M. N., Anderson, S. L., Worku, A., Marsack, W., Desai, N., Tuvilleja, A., Ramos, J., Francisco, M. A., Lafond, C., Balachandran, J. S., et al. (2017) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.6468 · Cited by 73
Corbin, C. K., Maclay, R., Acharya, A., Mony, S., Punnathanam, S., Thapa, R., Kotecha, N., Shah, N. H., & Chen, J. H. (2023) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocad114 · Cited by 68
Self-citationCallahan, A., McElfresh, D., Banda, J. M., Bunney, G., Char, D., Chen, J., Corbin, C. K., Dash, D., Downing, N. L., Jain, S. S., et al. (2024) · NEJM Catalyst · doi:10.1056/cat.24.0131 · Cited by 61
Self-citationVrijsen, B. E. L., Naaktgeboren, C. A., Vos, L. M., van Solinge, W. W., Kaasjager, H. A. H., & ten Berg, M. J. (2020) · Annals of Medicine and Surgery · doi:10.1016/j.amsu.2020.02.002 · Cited by 55
The Frequency of Unnecessary Testing in Hospitalized Patients
Koch, C., Roberts, K., Petruccelli, C., & Morgan, D. J. (2018) · The American Journal of Medicine · doi:10.1016/j.amjmed.2017.11.025 · Cited by 50
Yeshoua, B., Bowman, C., Dullea, J., Ditkowsky, J., Shyu, M., Lam, H., Zhao, W., Shin, J. Y., Dunn, A., Tsega, S., et al. (2023) · BMJ Open Quality · doi:10.1136/bmjoq-2022-002128 · Cited by 46
Prevalence and Predictability of Low-Yield Inpatient Laboratory Diagnostic Tests
Xu, S., Hom, J., Balasubramanian, S., Schroeder, L. F., Najafi, N., Roy, S., & Chen, J. H. (2019) · JAMA Network Open · doi:10.1001/jamanetworkopen.2019.10967 · Cited by 40
Self-citationFrom Bad to Worse: Anemia on Admission and Hospital-Acquired Anemia
Koch, C. G., Li, L., Sun, Z., Hixson, E. D., Tang, A. S., Phillips, S. C., Blackstone, E. H., & Henderson, J. M. (2017) · Journal of Patient Safety · doi:10.1097/pts.0000000000000142 · Cited by 39
Guidance on the critical shortage of sodium citrate coagulation tubes for hemostasis testing
Gosselin, R. C., Bowyer, A., Favaloro, E. J., Johnsen, J. M., Lippi, G., Marlar, R. A., Neeves, K., & Rollins‐Raval, M. A. (2021) · Journal of Thrombosis and Haemostasis · doi:10.1111/jth.15514 · Cited by 25
Variation in charges for 10 common blood tests in California hospitals: a cross-sectional analysis
Hsia, R. Y., Akosa Antwi, Y., & Nath, J. P. (2014) · BMJ Open · doi:10.1136/bmjopen-2014-005482 · Cited by 23
Lab testing overload: a comprehensive analysis of overutilization in hospital-based settings
Shaik, T., Mahmood, R., Kanagala, S. G., Kaur, H., Mendpara, V., Gupta, V., Aggarwal, P., Anamika, F., Garg, N., & Jain, R. (2024) · Baylor University Medical Center Proceedings · doi:10.1080/08998280.2023.2288788 · Cited by 22
Rabbani, N., Ma, S. P., Li, R. C., Winget, M., Weber, S., Boosi, S., Pham, T. D., Svec, D., Shieh, L., & Chen, J. H. (2023) · Clinical Biochemistry · doi:10.1016/j.clinbiochem.2023.01.002 · Cited by 20
Self-citationProbabilistic Prediction of Laboratory Test Information Yield.
Jiang, Y., Lee, A. H., Ni, X., Corbin, C. K., Irvin, J., Ng, A. Y., & Chen, J. H. (2023) · PubMed Central · Cited by 4
Self-citationJiang, Y., Ma, S. P., Liang, A. S., Kim, G., Acharya, A., Mony, S., Punnathanam, S., Makeown, J., Jose, J., Shieh, L., et al. (2026) · doi:10.64898/2026.04.29.26351965 · Cited by 1
Self-citation
Research works printed but not linked: 1
- Committee on the Learning Health Care System in America, Institute of Medicine. Best care at lower cost: the path to continuously learning health care in America. (Smith M, Saunders R, Stuckhardt L, McGinnis JM, eds.). Washington, DC: National Academies Press (US), 2013 (http://www.ncbi.nlm.nih.gov/books/NBK207225/).
Metadata
- Authors
- Liang, A. S., Amrollahi, F., Jiang, Y., Corbin, C. K., Kim, G. Y. E., Mui, D., Crowell, T., Acharya, A., Mony, S., Punnathanam, S., McKeown, J., Smith, M., Lin, S., Milstein, A., Schulman, K., Hom, J., Pfeffer, M. A., Pham, T. D., Svec, D., Chu, W., Shieh, L., Sharp, C., Ma, S. P., & Chen, J. H.
- Affiliations
- Stanford Medicine (US)
- Stanford University (US)
- SmarterDx, New York, NY, USA
- Johns Hopkins University (US)
- Stanford Health Care (US)
- Technology and Digital Solutions, Stanford Health Care, Palo Alto, CA
- ORCID
- Liang, April S.
- Amrollahi, Fatemeh
- Jiang, Yixing
- Corbin, Conor K.
- Kim, Grace Y. E.
- Mui, David
- Crowell, Trevor
- Acharya, Aakash
- Mony, Sreedevi
- Punnathanam, Soumya
- McKeown, Jack
- Smith, Margaret
- Lin, Steven
- Milstein, Arnold
- Schulman, Kevin
- Hom, Jason
- Pfeffer, Michael A.
- Pham, Tho D.
- Svec, David
- Chu, Weihan
- Shieh, Lisa
- Sharp, Christopher
- Ma, Stephen P.
- Chen, Jonathan H.
- Venue
- NEJM AI
- Publisher
- Massachusetts Medical Society
- Published
- 2026-06-25
- Volume
- 3
- Issue
- 7
- ISSN
- 2836-9386 (ISSN-L)
- PMC
- PMC13367503
- OpenAlex
- W7165639712
- Cited by
- 0
- References
- 18
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- Locations
- NEJM AI (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (submittedVersion)
- Funders
- Agency for Science, Technology and Research
- National Institutes of Health (UM1TR004921)
- Stanford Bio-X
- National Institute of Allergy and Infectious Diseases (R01AI178121)
- National Institute of Neurological Disorders and Stroke (U01NS134358)
- National Center for Advancing Translational Sciences
- Topics
- Health Sciences › Health Professions › Health Information Management › Electronic Health Records Systems
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Keywords
- clinical decision support, laboratory test utilization, complete blood count, electronic health record alerts, machine learning prediction, inpatient laboratory testing, implementation lessons learned, model governance, stakeholder engagement, user interface design, health care costs
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.