Training and Interpreting Machine Learning Algorithms to Evaluate Fall Risk After Emergency Department Visits

Patterson, Brian W.; Engstrom, Collin; Sah, Varun; Smith, Maureen A.; Mendonça, Eneida A.; Pulia, Michael S.; Repplinger, Michael D.; Hamedani, Azita G.; Page, David; Shah, Manish N.

doi:10.1097/mlr.0000000000001140

Cited by 38 publications

(27 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar method of developing an AI algorithm has also been used in identifying frailty and establishing the risk for falls in older people following an emergency department visit. Thus, the evaluation of several machine learning methodologies of automated risk strati ication and referral intervention led to a predictive model with an accuracy of 78% [12]. This is very similar to the accuracy of an AI algorithm used in identifying frailty among residents aged 75 years and over (75%) [13].…”

Section: Examples Of ML and Dl Implementation In The Medical Care Of mentioning

confidence: 58%

Artificial Intelligence in the healthcare of older people

Makaetova-Ladinska¹,

Harwood²,

Maltby³

2020

Arch Psychiatr Ment Health

View full text Add to dashboard Cite

Clinical applications of Artifi cial Intelligence (AI) in healthcare are relatively rare. The high expectations in relation to data analysis infl uencing general healthcare have not materialized, with few exceptions, and then predominantly in the fi eld of rare diseases, oncology and pathology, and interpretation of laboratory results. While electronic health records, introduced over the last decade or so in the UK have increased access to medical and treatment histories of patients, diagnoses, medications, treatment plans, immunization dates, allergies, radiology images, laboratory and test results, these have potential for evidence-based tools that providers can use to make decisions about a patient's care, as well as streamline workfl ow. In the following text, we review the advances achieved using machine learning and deep learning technology, as well as robot use and telemedicine in the healthcare of older people. Key points: 1. Artifi cial Intelligence use is extensively explored in prevention, diagnosis, novel drug designs and after-care. 2. AI studies on older adults include a small number of patients and lack reproducibility needed for their wider clinical use in different clinical settings and larger populations. 3. Telemedicine and robot assisted technology are well received by older service users. 4. Ethical concerns need to be resolved prior to wider AI use in routine clinical setting. seem to face a number of barriers in contacting communitybased primary health care, such as limited access, lack of standardized information systems and care pathways [3], all necessary to address their complex health care and social care needs. Indeed, older adults have much higher prevalence of nearly all major chronic and long-term conditions. In addition, they are more likely to succumb to adverse health events, such as a falls or infections, and these can lead to dramatic changes in their physical and mental wellbeing even after an apparently minor incident [4]. However, person and familyfocused care, self-management resources, and successful collaborative practice have been all highlighted as facilitators of good health care provision both by older people and their families [3]. All the above places the importance of diagnosis, monitoring of disease risks and their prevention, as well as management and optimizing of geriatric syndromes in the community for both older people living independenly or in 24h care facilities. In particular, identi ication of health issues/ diagnosis, support/treatment needs evaluation, development

show abstract

Section: Examples Of ML and Dl Implementation In The Medical Care Of mentioning

confidence: 58%

Artificial Intelligence in the healthcare of older people

Makaetova-Ladinska¹,

Harwood²,

Maltby³

2020

Arch Psychiatr Ment Health

View full text Add to dashboard Cite

show abstract

“…These approaches use available EHR data to guide risk stratification that informs clinical testing or decision‐making. 43 , 44 , 45 , 46 Many can support clinical team screening with complex calculations occurring within fractions of a second to prompt action. 47 Using the risk factors identified in this analysis to inform risk prediction during the ED intake phase of care could capture the patients with STEMI identified during the triage phase earlier in the ED care process.…”

Section: Discussionmentioning

confidence: 99%

Fallacy of Median Door‐to‐ECG Time: Hidden Opportunities for STEMI Screening Improvement

Yiadom

Gong

Patterson

et al. 2022

JAHA

Self Cite

View full text Add to dashboard Cite

Background ST‐segment elevation myocardial infarction (STEMI) guidelines recommend screening arriving emergency department (ED) patients for an early ECG in those with symptoms concerning for myocardial ischemia. Process measures target median door‐to‐ECG (D2E) time of 10 minutes. Methods and Results This 3‐year descriptive retrospective cohort study, including 676 ED‐diagnosed patients with STEMI from 10 geographically diverse facilities across the United States, examines an alternative approach to quantifying performance: proportion of patients meeting the goal of D2E≤10 minutes. We also identified characteristics associated with D2E>10 minutes and estimated the proportion of patients with screening ECG occurring during intake, triage, and main ED care periods. We found overall median D2E was 7 minutes (IQR:4–16; range: 0–1407 minutes; range of ED medians: 5–11 minutes). Proportion of patients with D2E>10 minutes was 37.9% (ED range: 21.5%–57.1%). Patients with D2E>10 minutes, compared to those with D2E≤10 minutes, were more likely female (32.8% versus 22.6%, P =0.005), Black (23.4% versus 12.4%, P =0.005), non‐English speaking (24.6% versus 19.5%, P =0.032), diabetic (40.2% versus 30.2%, P =0.010), and less frequently reported chest pain (63.3% versus 87.4%, P <0.001). ECGs were performed during ED intake in 62.1% of visits, ED triage in 25.3%, and main ED care in 12.6%. Conclusions Examining D2E>10 minutes can identify opportunities to improve care for more ED patients with STEMI. Our findings suggest sex, race, language, and diabetes are associated with STEMI diagnostic delays. Moving the acquisition of ECGs completed during triage to intake could achieve the D2E≤10 minutes goal for 87.4% of ED patients with STEMI. Sophisticated screening, accounting for differential risk and diversity in STEMI presentations, may further improve timely detection.

show abstract

“…Finally, although we considered comprehensive factors, the AUC of the final model was 0.726, which is not very high, indicating that there is a need for further research to identify other possible fall-related factors. Recently, machine learning algorithms using electronic health records have been evaluated for predicting falls among older adults making emergency department visits 36 . In the near future, developing comprehensive machine learning models to predict inpatient falls based on electronic health records is needed.…”

Section: Discussionmentioning

confidence: 99%

Fall predictors beyond fall risk assessment tool items for acute hospitalized older adults: a matched case–control study

Noh

Song

Park

et al. 2021

Sci Rep

View full text Add to dashboard Cite

We investigated whether clinical factors including comorbidities, medications, and laboratory results predict inpatient fall risk in older adults. The participants in this case–control study included hospitalized older adults with acute conditions who had falls during their hospital stay (case group) and 410 hospitalized older adults who did not experience falls (control group). Data on medical history, fall risk assessment (Morse Fall Scale; MFS), medications, and laboratory results were obtained. Conditional logistic regression analysis was performed to estimate the association between clinical factors and falls. Receiver operating characteristic curves and area under the curve (AUC) were used to determine whether clinical factors could discriminate between fallers and controls. We evaluated three models: (M1) MFS, (M2) M1 plus age, sex, ward, and polypharmacy, and (M3) M2 plus clinical factors. Patients with diabetes mellitus or MFS scores ≥ 45 had the highest risk of falls. Calcium channel blockers, diuretics, anticonvulsants, and benzodiazepines were associated with high fall risk. The AUC of the three models was 0.615, 0.646, and 0.725, respectively (M1 vs. M2, P = 0.042 and M2 vs. M3, P < .001). Examining clinical factors led to significant improvements in fall prediction beyond that of the MFS in hospitalized older adults.

show abstract

Training and Interpreting Machine Learning Algorithms to Evaluate Fall Risk After Emergency Department Visits

Cited by 38 publications

References 41 publications

Artificial Intelligence in the healthcare of older people

Artificial Intelligence in the healthcare of older people

Fallacy of Median Door‐to‐ECG Time: Hidden Opportunities for STEMI Screening Improvement

Fall predictors beyond fall risk assessment tool items for acute hospitalized older adults: a matched case–control study

Contact Info

Product

Resources

About