-
digital.ahrq.gov/ahrq-funded-projects/identification-patients-low-life-expectancy
January 01, 2023 - Publications
Events
The inclusion of information from free-text notes into risk of death … prediction models significantly improves the ability to predict the probability of death, with the potential … With a goal of identifying patients at high risk of death, this research integrated the artificial intelligence … This model of risk of death was created, tested, and analyzed against benchmark statistical and machine … The researchers found that inclusion of NLP-generated data improved estimates of the probability of death
-
digital.ahrq.gov/sites/default/files/docs/citation/r01hs024090-turchin-final-report-2019.pdf
January 01, 2019 - conjunction
with natural language processing can achieve higher prediction accuracy for the risk of death … Results: Dynamic Logic had a small but consistent advantage in estimating the probability
of death over … Death Master File. … Specifically, we evaluated the risk of death once a year, on January 1st. … Benchmark Methods
We utilized several benchmark methods analytical for prediction of risk of death.
-
digital.ahrq.gov/2020-year-review/research-summary/creating-health-information-exchange-application-provide-fast-access-patient-data-emergency
January 01, 2020 - about a patient’s health history, especially in the emergency department (ED) setting, where life and death … repeated or unnecessary diagnostic tests and to expedite care and treatment in potentially life or death … You really can't spend minutes looking up stuff on the computer when somebody is bleeding to death right
-
digital.ahrq.gov/2020-year-review/research-dissemination/reaching-research-community-through-web-conferences
January 01, 2020 - Turchin highlighted research that showed inclusion of information from free-text notes into risk-of-death … prediction models significantly improves the ability to predict the probability of death.
-
digital.ahrq.gov/sites/default/files/docs/clinical-care-slides-10142020.pdf
October 14, 2020 - in conjunction with natural language
processing can be used to identify patients at high risk for death … counseling
while patient’s HbA1c > 7%
• Primary outcome: MI, CVA, hospitalization for angina or
death … months between 2000 and 2014
• Data for every patient were re-analyzed every 12 months
to predict death … neural network (MPCNN) that can directly handle a heterogeneous dataset
Predicting Postoperative Death … Our Approach
Architecture of the Multipath Convolutional Neural Network �
Predicting Postoperative Death
-
digital.ahrq.gov/ahrq-funded-projects/improving-healthcare-processes-and-outcomes-directly-integrating-health
September 29, 2024 - about a patient’s health history, especially in the emergency department (ED) setting, where life and death … repeated or unnecessary diagnostic tests and to expedite care and treatment in potentially life or death … You really can't spend minutes looking up stuff on the computer when somebody is bleeding to death right
-
digital.ahrq.gov/sites/default/files/docs/page/2006Sundwall_051111comp.pdf
June 01, 2006 - Commissioner for Health 11
8th Leading Cause of Death … Leading Causes of Death and Estimated Medical Injuries That Contributed to
Death: Utah, 2001
Cause … of Death Deaths
1 Diseases of heart 2,875
2 Malignant neoplasms 2,304
3 Cerebrovascular diseases … Pilot Projects� (beginning now)
Moving to ‘Clinical’ Exchanges
Where We Are
8th Leading Cause of Death
-
digital.ahrq.gov/sites/default/files/docs/citation/r18hs026188-gong-final-report-2022.pdf
January 01, 2022 - ) 40 (37.0)
-20 (-115.5,
0.3)
81
(75.0)
27
(25.0)
-314 (-588.5,
-127.8)
ARDS or death … 652
308
(47.2)
-58 (-112, -
20)
ARDS and
Death 255
237
(92.9) 4 (-1, 18) 86 (36.3) 156 ( … 163
128
(78.5) -37 (-87, -11)
ARDS and
Death 255
237
(92.9) 4 (-1, 18) 86 (36.3) 156 (65.8 … ) 0 (-12, 10)
125
(52.7)
112
(47.3)
- 112 (-211.3,
-52)
ARDS or death
746
555
(74.4) … The model identified patients at risk for ARDS or death at a median of 10 hours and an
interquartile
-
digital.ahrq.gov/sites/default/files/docs/page/2006FriedmanSantinonFormica_052411comp.pdf
June 16, 2021 - NEJM 1999:341:1725
Death Rate
All Dialysis 16.1/100 pt yrs
Waitlist 6.3/100 pt yrs
PROJECTED YEARS … NEJM 1999:341:1725
Death Rate
All Dialysis 16.1/100 pt yrs
Waitlist 6.3/100 pt yrs
Transplant Recipients
-
digital.ahrq.gov/2020-year-review/research-summary/anesthesiology-control-tower-air-traffic-control-operating-rooms
January 01, 2020 - patients who are at risk for specific complications, including respiratory failure, kidney failure, and death … ACTFAST into operating rooms improves care
Machine learning algorithms for predicting postoperative death
-
digital.ahrq.gov/sites/default/files/docs/page/2006Wang_051611comp.pdf
January 01, 2006 - Recurrent theme:
Paramedic ETI associated with increased risk of death.
Paramedic ETI associated … 209 pts matched with 627 historical non-intubated controls
“Prehospital RSI increased odds of death … non-intubated cases
Multivariable and propensity score adjusted
“Paramedic ETI 4x greater odds of death
-
digital.ahrq.gov/ahrq-funded-projects/longitudinal-machine-learning-approach-providing-clinicians-timely-detection
September 30, 2023 - ideation (SI), a broad term used to describe a range of contemplations, wishes, and preoccupations with death … Active duty personnel who die by suicide often maintain their duties until their death.
-
digital.ahrq.gov/ahrq-funded-projects/virtual-continuity-and-its-impact-complex-hospitalized-patients-care
January 01, 2023 - However, errors that were considered clinically significant—those leading to death, permanent or temporary … significant difference seen in PCP followup visits and ED visits at 30 days, or in 30-day readmission or death
-
digital.ahrq.gov/sites/default/files/docs/citation/r21hs024581-avidan-final-report-2020.pdf
January 01, 2020 - of surgical patients was used to develop and validate machine
learning algorithms for postoperative death … To predict death, we constructed a multi-path convolutional neural network that extracted
information … The incidence of postoperative death was 2.4%. … In the testing set, the multi-path convolutional
neural network model predicted postoperative death … The AUPRC compares favorably to the incidence of
death.
-
digital.ahrq.gov/sites/default/files/docs/publication/r18hs017786-ritchie-final-report-2012.pdf
January 01, 2012 - Eight percent of these care transitions resulted in death during the
30-day post-hospitalization time … was evident; on the other
hand, for COPD, the intervention seemed to reduce rehospitalizations or death … In the COPD group, we saw trends for
reduced rehospitalizations and death in the intervention group … 30 31 17.4 30 17.9 0.973 16 23.9 8 12.3 0.134
Rehospitalization
or Death 90 56 31.5 62 36.9 0.340 … 30 5 2.8 4 2.4 0.930 2 3.0 0 0.0 0.490
Death 90 10 5.6 13 7.7 0.565 4 6.0 1 1.5 0.380
Days out of
-
digital.ahrq.gov/2019-year-review/research-summary/improving-delivery-health-services-health-systems-level
January 01, 2019 - Dynamic Logic, a machine-learning algorithm to improve the identification of patients at high risk of death … the results indicated that Dynamic Logic had a consistent advantage in estimating the probability of death
-
digital.ahrq.gov/program-overview/research-stories/decision-precision-increasing-lung-cancer-screening-risk-patients
January 01, 2023 - Decision Precision+: Increasing Lung Cancer Screening for At-Risk Patients
Theme:
Supporting Health Systems in Advancing Care Delivery
Subtheme:
Scaling Effective Digital Healthcare Interventions Across Healthcare Systems
A shared decision-making tool to support the appropriate use of low-…
-
digital.ahrq.gov/national-webinars/applying-advanced-analytics-clinical-care
January 01, 2023 - algorithms in conjunction with natural language processing can be used to identify patients at high risk for death
-
digital.ahrq.gov/ahrq-funded-projects/engaging-patients-enable-interoperable-lung-cancer-decision-support-scale
January 01, 2024 - Engaging Patients to Enable Interoperable Lung Cancer Decision Support at Scale
Project Description
Publications
Improving shared decision making for lung cancer screening by adapting and disseminating an interoperable clinical decision support tool and patient-facing app…
-
digital.ahrq.gov/ahrq-funded-projects/e-prescribing-impact-patient-safety-use-and-cost/annual-summary/2009
January 01, 2009 - Overall, 25 percent of patients with a chronic condition had either a death or a hospitalization in the … year following their first e-prescription; however, e-prescribing had no impact on the risk of death