Comparing clinician descriptions of frailty and geriatric syndromes using electronic health records: a retrospective cohort study

Assessing the Benefits and Harms of Pharmacotherapy in Older Adults with Frailty: Insights from Pharmacoepidemiologic Studies of Routine Health Care Data

The objective of this review is to summarize and appraise the research methodology, emerging findings, and future directions in pharmacoepidemiologic studies assessing the benefits and harms of pharmacotherapies in older adults with different levels of frailty. Older adults living with frailty are at elevated risk for poor health outcomes and adverse effects from pharmacotherapy. However, current evidence is limited due to the under-enrollment of frail older adults and the lack of validated frailty assessments in clinical trials. Recent advancements in measuring frailty in administrative claims and electronic health records (database-derived frailty scores) have enabled researchers to identify patients with frailty and to evaluate the heterogeneity of treatment effects by patients' frailty levels using routine health care data. When selecting a database-derived frailty score, researchers must consider the type of data (e.g., different coding systems), the length of the predictor assessment period, the extent of validation against clinically validated frailty measures, and the possibility of surveillance bias arising from unequal access to care. We reviewed 13 pharmacoepidemiologic studies published on PubMed from 2013 to 2023 that evaluated the benefits and harms of cardiovascular medications, diabetes medications, anti-neoplastic agents, antipsychotic medications, and vaccines by frailty levels. These studies suggest that, while greater frailty is positively associated with adverse treatment outcomes, older adults with frailty can still benefit from pharmacotherapy. Therefore, we recommend routine frailty subgroup analyses in pharmacoepidemiologic studies. Despite data and design limitations, the findings from such studies may be informative to tailor pharmacotherapy for older adults across the frailty spectrum.

Natural language processing systems for extracting information from electronic health records about activities of daily living. A systematic review

Objective

Natural language processing (NLP) can enhance research on activities of daily living (ADL) by extracting structured information from unstructured electronic health records (EHRs) notes. This review aims to give insight into the state-of-the-art, usability, and performance of NLP systems to extract information on ADL from EHRs.

Materials and Methods

A systematic review was conducted based on searches in Pubmed, Embase, Cinahl, Web of Science, and Scopus. Studies published between 2017 and 2022 were selected based on predefined eligibility criteria.

Results

The review identified 22 studies. Most studies (65%) used NLP for classifying unstructured EHR data on 1 or 2 ADL. Deep learning, combined with a ruled-based method or machine learning, was the approach most commonly used. NLP systems varied widely in terms of the pre-processing and algorithms. Common performance evaluation methods were cross-validation and train/test datasets, with F1, precision, and sensitivity as the most frequently reported evaluation metrics. Most studies reported relativity high overall scores on the evaluation metrics.

Discussion

NLP systems are valuable for the extraction of unstructured EHR data on ADL. However, comparing the performance of NLP systems is difficult due to the diversity of the studies and challenges related to the dataset, including restricted access to EHR data, inadequate documentation, lack of granularity, and small datasets.

Conclusion

This systematic review indicates that NLP is promising for deriving information on ADL from unstructured EHR notes. However, what the best-performing NLP system is, depends on characteristics of the dataset, research question, and type of ADL.

New Onset Geriatric Syndromes and One-year Outcomes Following Elective Gastrointestinal Cancer Surgery

OBJECTIVE

To assess whether older adults who develop geriatric syndromes following elective gastrointestinal surgery have poorer 1-year outcomes.

BACKGROUND

Within 10 years, 70% of all cancers will occur in older adults ≥65 years old. The rise in older adults requiring major surgery has brought attention to age-related complications termed geriatric syndromes. However, whether postoperative geriatric syndromes are associated with long-term outcomes is unclear.

METHODS

A population-based retrospective cohort study using the New York State Cancer Registry and the Statewide Planning and Research Cooperative System was performed including patients >55 years with pathologic stage I-III esophageal, gastric, pancreatic, colon, or rectal cancer who underwent elective resection between 2004 and 2018. Those aged 55 to 64 served as the reference group. The exposure of interest was a geriatric syndrome [fracture, fall, delirium, pressure ulcer, depression, malnutrition, failure to thrive, dehydration, or incontinence (urinary/fecal)] during the surgical admission. Patients with any geriatric syndrome within 1 year of surgery were excluded. Outcomes included incident geriatric syndrome, 1-year days alive and out of the hospital, and 1-year all-cause mortality.

RESULTS

In this study, 37,998 patients with a median age of 71 years without a prior geriatric syndrome were included. Of those 65 years or more, 6.4% developed a geriatric syndrome. Factors associated with an incident geriatric syndrome were age, alcohol/tobacco use, comorbidities, neoadjuvant therapy, ostomies, open surgery, and upper gastrointestinal cancers. An incident geriatric syndrome was associated with a 43% higher risk of 1-year mortality (hazard ratio, 1.43; 95% confidence interval, 1.27-1.60). For those aged 65+ discharged alive and not to hospice, a geriatric syndrome was associated with significantly fewer days alive and out of hospital (322 vs 346 days, P < 0.0001). There was an indirect relationship between the number of geriatric syndromes and 1-year mortality and days alive and out of the hospital after adjusting for surgical complications.

CONCLUSIONS

Given the increase in older adults requiring major surgical intervention, and the establishment of geriatric surgery accreditation programs, these data suggest that morbidity and mortality metrics should be adjusted to accommodate the independent relationship between geriatric syndromes and long-term outcomes.

Investigating the association of anticholinergic burden with depression in older adults: a cross-sectional study

BACKGROUND

Although depression and anticholinergic drug use are common comorbidities that impair health status in later life, there are insufficient data on their relationship. This study aimed to investigate the relationship between depressive symptoms and anticholinergic use in older individuals.

METHODS

Community-dwelling older adults (≥65 years) admitted to the tertiary referral geriatric outpatient clinic were included. Participants were evaluated for depressive symptoms using the Geriatric Depression Scale (GDS) with a cut-off score of ≥6 for depression. Exposure to anticholinergic drugs was assessed using the anticholinergic cognitive burden (ACB) scale and three subgroups were created: ACB = 0, ACB = 1, and ACB ≥ 2. The relationship between these two parameters was assessed using multivariate logistic regression analysis considering other potential variables.

RESULTS

The study included 1232 participants (mean age 78.4 ± 7.2 years and 65.2% female) and the prevalence of depression was 24%. After adjusting for potential confounders, compared to ACB = 0, having ACB ≥ 2 was related to depression symptoms (odds ratio (OR): 1.56, 95% CI: 1.04-2.35, P = 0.034), whereas having ACB = 1 did not increase the risk (OR: 1.27, 95% CI: 0.88-1.83, P = 0.205).

CONCLUSION

Our findings indicate that special attention should be paid to drug therapy in preventing depression in older adults, as exposure to a high anticholinergic load is negatively associated with psychological status.

Identifying Functional Status Impairment in People Living With Dementia Through Natural Language Processing of Clinical Documents: Cross-Sectional Study

BACKGROUND

Assessment of activities of daily living (ADLs) and instrumental ADLs (iADLs) is key to determining the severity of dementia and care needs among older adults. However, such information is often only documented in free-text clinical notes within the electronic health record and can be challenging to find.

OBJECTIVE

This study aims to develop and validate machine learning models to determine the status of ADL and iADL impairments based on clinical notes.

METHODS

This cross-sectional study leveraged electronic health record clinical notes from Mass General Brigham's Research Patient Data Repository linked with Medicare fee-for-service claims data from 2007 to 2017 to identify individuals aged 65 years or older with at least 1 diagnosis of dementia. Notes for encounters both 180 days before and after the first date of dementia diagnosis were randomly sampled. Models were trained and validated using note sentences filtered by expert-curated keywords (filtered cohort) and further evaluated using unfiltered sentences (unfiltered cohort). The model's performance was compared using area under the receiver operating characteristic curve and area under the precision-recall curve (AUPRC).

RESULTS

The study included 10,000 key-term-filtered sentences representing 441 people (n=283, 64.2% women; mean age 82.7, SD 7.9 years) and 1000 unfiltered sentences representing 80 people (n=56, 70% women; mean age 82.8, SD 7.5 years). Area under the receiver operating characteristic curve was high for the best-performing ADL and iADL models on both cohorts (>0.97). For ADL impairment identification, the random forest model achieved the best AUPRC (0.89, 95% CI 0.86-0.91) on the filtered cohort; the support vector machine model achieved the highest AUPRC (0.82, 95% CI 0.75-0.89) for the unfiltered cohort. For iADL impairment, the Bio+Clinical bidirectional encoder representations from transformers (BERT) model had the highest AUPRC (filtered: 0.76, 95% CI 0.68-0.82; unfiltered: 0.58, 95% CI 0.001-1.0). Compared with a keyword-search approach on the unfiltered cohort, machine learning reduced false-positive rates from 4.5% to 0.2% for ADL and 1.8% to 0.1% for iADL.

CONCLUSIONS

In this study, we demonstrated the ability of machine learning models to accurately identify ADL and iADL impairment based on free-text clinical notes, which could be useful in determining the severity of dementia.

Use of Artificial Intelligence in the Identification and Management of Frailty: A Scoping Review Protocol

INTRODUCTION

Rapid population ageing and associated health issues such as frailty are a growing public health concern. While early identification and management of frailty may limit adverse health outcomes, the complex presentations of frailty pose challenges for clinicians. Artificial intelligence (AI) has emerged as a potential solution to support the early identification and management of frailty. In order to provide a comprehensive overview of current evidence regarding the development and use of AI technologies including machine learning and deep learning for the identification and management of frailty, this protocol outlines a scoping review aiming to identify and present available information in this area. Specifically, this protocol describes a review that will focus on the clinical tools and frameworks used to assess frailty, the outcomes that have been evaluated and the involvement of knowledge users in the development, implementation and evaluation of AI methods and tools for frailty care in clinical settings.

METHODS AND ANALYSIS

This scoping review protocol details a systematic search of eight major academic databases, including Medline, Embase, PsycInfo, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Ageline, Web of Science, Scopus and Institute of Electrical and Electronics Engineers (IEEE) Xplore using the framework developed by Arksey and O'Malley and enhanced by Levac et al and the Joanna Briggs Institute. The search strategy has been designed in consultation with a librarian. Two independent reviewers will screen titles and abstracts, followed by full texts, for eligibility and then chart the data using a piloted data charting form. Results will be collated and presented through a narrative summary, tables and figures.

ETHICS AND DISSEMINATION

Since this study is based on publicly available information, ethics approval is not required. Findings will be communicated with healthcare providers, caregivers, patients and research and health programme funders through peer-reviewed publications, presentations and an infographic.

REGISTRATION DETAILS

OSF Registries (https://doi.org/10.17605/OSF.IO/T54G8).

Application of natural language processing to identify social needs from patient medical notes: development and assessment of a scalable, performant, and rule-based model in an integrated healthcare delivery system

Objectives

To develop and test a scalable, performant, and rule-based model for identifying 3 major domains of social needs (residential instability, food insecurity, and transportation issues) from the unstructured data in electronic health records (EHRs).

Materials and Methods

We included patients aged 18 years or older who received care at the Johns Hopkins Health System (JHHS) between July 2016 and June 2021 and had at least 1 unstructured (free-text) note in their EHR during the study period. We used a combination of manual lexicon curation and semiautomated lexicon creation for feature development. We developed an initial rules-based pipeline (Match Pipeline) using 2 keyword sets for each social needs domain. We performed rule-based keyword matching for distinct lexicons and tested the algorithm using an annotated dataset comprising 192 patients. Starting with a set of expert-identified keywords, we tested the adjustments by evaluating false positives and negatives identified in the labeled dataset. We assessed the performance of the algorithm using measures of precision, recall, and F1 score.

Results

The algorithm for identifying residential instability had the best overall performance, with a weighted average for precision, recall, and F1 score of 0.92, 0.84, and 0.92 for identifying patients with homelessness and 0.84, 0.82, and 0.79 for identifying patients with housing insecurity. Metrics for the food insecurity algorithm were high but the transportation issues algorithm was the lowest overall performing metric.

Discussion

The NLP algorithm in identifying social needs at JHHS performed relatively well and would provide the opportunity for implementation in a healthcare system.

Conclusion

The NLP approach developed in this project could be adapted and potentially operationalized in the routine data processes of a healthcare system.

The use of natural language processing in detecting and predicting falls within the healthcare setting: a systematic review

Falls are a common problem associated with significant morbidity, mortality, and economic costs. Current fall prevention policies in local healthcare settings are often guided by information provided by fall risk assessment tools, incident reporting, and coding data. This review was conducted with the aim of identifying studies which utilized natural language processing (NLP) for the automated detection and prediction of falls in the healthcare setting. The databases Ovid Medline, Ovid Embase, Ovid Emcare, PubMed, CINAHL, IEEE Xplore, and Ei Compendex were searched from 2012 until April 2023. Retrospective derivation, validation, and implementation studies wherein patients experienced falls within a healthcare setting were identified for inclusion. The initial search yielded 2611 publications for title and abstract screening. Full-text screening was conducted on 105 publications, resulting in 26 unique studies that underwent qualitative analyses. Studies applied NLP towards falls risk factor identification, known falls detection, future falls prediction, and falls severity stratification with reasonable success. The NLP pipeline was reviewed in detail between studies and models utilizing rule-based, machine learning (ML), deep learning (DL), and hybrid approaches were examined. With a growing literature surrounding falls prediction in both inpatient and outpatient environments, the absence of studies examining the impact of these models on patient and system outcomes highlights the need for further implementation studies. Through an exploration of the application of NLP techniques, it may be possible to develop models with higher performance in automated falls prediction and detection.

Characterizing Fall Circumstances in Community-Dwelling Older Adults: A Mixed Methods Approach

BACKGROUND

Understanding fall circumstances can help researchers better identify causes of falls and develop effective and tailored fall prevention programs. This study aims to describe fall circumstances among older adults from quantitative data using conventional statistical approaches and qualitative analyses using a machine learning approach.

METHODS

The MOBILIZE Boston Study enrolled 765 community-dwelling adults aged 70 years and older in Boston, MA. Occurrence and circumstances of falls (ie, locations, activities, and self-reported causes of falls) were recorded using monthly fall calendar postcards and fall follow-up interviews with open- and close-ended questions during a 4-year period. Descriptive analyses were used to summarize circumstances of falls. Natural language processing was used to analyze narrative responses from open-ended questions.

RESULTS

During the 4-year follow-up, 490 participants (64%) had at least 1 fall. Among 1 829 falls, 965 falls occurred indoors and 804 falls occurred outdoors. Commonly reported activities when the fall occurred were walking (915, 50.0%), standing (175, 9.6%), and going down stairs (125, 6.8%). The most commonly reported causes of falls were slip or trip (943, 51.6%) and inappropriate footwear (444, 24.3%). Using qualitative data, we extracted more detailed information on locations and activities, and additional information on obstacles related to falls and commonly reported scenarios such as "lost my balance and fell."

CONCLUSIONS

Self-reported fall circumstances provide important information on both intrinsic and extrinsic factors contributing to falls. Future studies are warranted to replicate our findings and optimize approaches to analyzing narrative data on fall circumstances in older adults.

Evaluating the comparability of patient-level social risk data extracted from electronic health records: A systematic scoping review

Objective: To evaluate how and from where social risk data are extracted from EHRs for research purposes, and how observed differences may impact study generalizability. Methods: Systematic scoping review of peer-reviewed literature that used patient-level EHR data to assess 1 ± 6 social risk domains: housing, transportation, food, utilities, safety, social support/isolation. Results: 111/9022 identified articles met inclusion criteria. By domain, social support/isolation was most often included (N = 68/111), predominantly defined by marital/partner status (N = 48/68) and extracted from structured sociodemographic data (N = 45/48). Housing risk was defined primarily by homelessness (N = 39/49). Structured housing data was extracted most from billing codes and screening tools (N = 15/30, 13/30, respectively). Across domains, data were predominantly sourced from structured fields (N = 89/111) versus unstructured free text (N = 32/111). Conclusion: We identified wide variability in how social domains are defined and extracted from EHRs for research. More consistency, particularly in how domains are operationalized, would enable greater insights across studies.

Using electronic consultation (eConsult) to identify frailty in provider-to-provider communication: a feasibility and validation study

BACKGROUND

Frailty is a complex age-related clinical condition that increases vulnerability to stressors. Early recognition of frailty is challenging. While primary care providers (PCPs) serve as the first point of contact for most older adults, convenient tools for identifying frailty in primary care are lacking. Electronic consultation (eConsult), a platform connecting PCPs to specialists, is a rich source of provider-to-provider communication data. Text-based patient descriptions on eConsult may provide opportunities for earlier identification of frailty. We sought to explore the feasibility and validity of identifying frailty status using eConsult data.

METHODS

eConsult cases closed in 2019 and submitted on behalf of long-term care (LTC) residents or community-dwelling older adults were sampled. A list of frailty-related terms was compiled through a review of the literature and consultation with experts. To identify frailty, eConsult text was parsed to measure the frequency of frailty-related terms. Feasibility of this approach was assessed by examining the availability of frailty-related terms in eConsult communication logs, and by asking clinicians to indicate whether they can assess likelihood of frailty by reviewing the cases. Construct validity was assessed by comparing the number of frailty-related terms in cases about LTC residents with those about community-dwelling older adults. Criterion validity was assessed by comparing clinicians' ratings of frailty to the frequency of frailty-related terms.

RESULTS

One hundred thirteen LTC and 112 community cases were included. Frailty-related terms identified per case averaged 4.55 ± 3.95 in LTC and 1.96 ± 2.68 in the community (p < .001). Clinicians consistently rated cases with ≥ 5 frailty-related terms as highly likely of living with frailty.

CONCLUSIONS

The availability of frailty-related terms establishes the feasibility of using provider-to-provider communication on eConsult to identify patients with high likelihood of living with this condition. The higher average of frailty-related terms in LTC (versus community) cases, and agreement between clinician-provided frailty ratings and the frequency of frailty-related terms, support the validity of an eConsult-based approach to identifying frailty. There is potential for eConsult to be used as a case-finding tool in primary care for early recognition and proactive initiation of care processes for older patients living with frailty.

Can Patients with Dementia Be Identified in Primary Care Electronic Medical Records Using Natural Language Processing?

Dementia and mild cognitive impairment can be underrecognized in primary care practice and research. Free-text fields in electronic medical records (EMRs) are a rich source of information which might support increased detection and enable a better understanding of populations at risk of dementia. We used natural language processing (NLP) to identify dementia-related features in EMRs and compared the performance of supervised machine learning models to classify patients with dementia. We assembled a cohort of primary care patients aged 66 + years in Ontario, Canada, from EMR notes collected until December 2016: 526 with dementia and 44,148 without dementia. We identified dementia-related features by applying published lists, clinician input, and NLP with word embeddings to free-text progress and consult notes and organized features into thematic groups. Using machine learning models, we compared the performance of features to detect dementia, overall and during time periods relative to dementia case ascertainment in health administrative databases. Over 900 dementia-related features were identified and grouped into eight themes (including symptoms, social, function, cognition). Using notes from all time periods, LASSO had the best performance (F1 score: 77.2%, sensitivity: 71.5%, specificity: 99.8%). Model performance was poor when notes written before case ascertainment were included (F1 score: 14.4%, sensitivity: 8.3%, specificity 99.9%) but improved as later notes were added. While similar models may eventually improve recognition of cognitive issues and dementia in primary care EMRs, our findings suggest that further research is needed to identify which additional EMR components might be useful to promote early detection of dementia.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41666-023-00125-6.

Natural Language Processing in Electronic Health Records in relation to healthcare decision-making: A systematic review

BACKGROUND

Natural Language Processing (NLP) is widely used to extract clinical insights from Electronic Health Records (EHRs). However, the lack of annotated data, automated tools, and other challenges hinder the full utilisation of NLP for EHRs. Various Machine Learning (ML), Deep Learning (DL) and NLP techniques are studied and compared to understand the limitations and opportunities in this space comprehensively.

METHODOLOGY

After screening 261 articles from 11 databases, we included 127 papers for full-text review covering seven categories of articles: (1) medical note classification, (2) clinical entity recognition, (3) text summarisation, (4) deep learning (DL) and transfer learning architecture, (5) information extraction, (6) Medical language translation and (7) other NLP applications. This study follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

RESULT AND DISCUSSION

EHR was the most commonly used data type among the selected articles, and the datasets were primarily unstructured. Various ML and DL methods were used, with prediction or classification being the most common application of ML or DL. The most common use cases were: the International Classification of Diseases, Ninth Revision (ICD-9) classification, clinical note analysis, and named entity recognition (NER) for clinical descriptions and research on psychiatric disorders.

CONCLUSION

We find that the adopted ML models were not adequately assessed. In addition, the data imbalance problem is quite important, yet we must find techniques to address this underlining problem. Future studies should address key limitations in studies, primarily identifying Lupus Nephritis, Suicide Attempts, perinatal self-harmed and ICD-9 classification.

Recommended practices and ethical considerations for natural language processing-assisted observational research: A scoping review

An increasing number of studies have reported using natural language processing (NLP) to assist observational research by extracting clinical information from electronic health records (EHRs). Currently, no standardized reporting guidelines for NLP-assisted observational studies exist. The absence of detailed reporting guidelines may create ambiguity in the use of NLP-derived content, knowledge gaps in the current research reporting practices, and reproducibility challenges. To address these issues, we conducted a scoping review of NLP-assisted observational clinical studies and examined their reporting practices, focusing on NLP methodology and evaluation. Through our investigation, we discovered a high variation regarding the reporting practices, such as inconsistent use of references for measurement studies, variation in the reporting location (reference, appendix, and manuscript), and different granularity of NLP methodology and evaluation details. To promote the wide adoption and utilization of NLP solutions in clinical research, we outline several perspectives that align with the six principles released by the World Health Organization (WHO) that guide the ethical use of artificial intelligence for health.

Use of a nursing-led geriatrics consult service to deliver age-friendly care

The Elder Veteran Program (EVP) is a nursing-led approach to deliver inpatient consultative geriatrics care at our academic Midwestern Veterans Hospital. From April to December of 2021, EVP modified its workflow using a Plan-Do-Study-Act approach to include previously under-addressed components of the IHI's "4M's" of Age-Friendly Care (Medication, Mobility, Mentation, and What Matters), with three months of retrospective data review as a Plan phase, three months of monthly Do and Study phases, and a three month Act phase to analyze post-intervention care. We found improvements in frequency of documentation of Medication, Mentation, and What Matters in EVP notes, and maintenance of Mobility documentation. Next steps include translating these documentation and workflow changes into other relevant outcome measures and outreach to other departments. Overall, our project demonstrates a novel way to integrate these Pillars into a hospital system, by leveraging an existing nursing-led geriatric consult service focused on prevention and education.

Malnutrition and its contributing factors for older people living in residential aged care facilities: Insights from natural language processing of aged care records

BACKGROUND

Malnutrition is a serious health risk facing older people living in residential aged care facilities. Aged care staff record observations and concerns about older people in electronic health records (EHR), including free-text progress notes. These insights are yet to be unleashed.

OBJECTIVE

This study explored the risk factors for malnutrition in structured and unstructured electronic health data.

METHODS

Data of weight loss and malnutrition were extracted from the de-identified EHR records of a large aged care organization in Australia. A literature review was conducted to identify causative factors for malnutrition. Natural language processing (NLP) techniques were applied to progress notes to extract these causative factors. The NLP performance was evaluated by the parameters of sensitivity, specificity and F1-Score.

RESULTS

The NLP methods were highly accurate in extracting the key data, values for 46 causative variables, from the free-text client progress notes. Thirty three percent (1,469 out of 4,405) of the clients were malnourished. The structured, tabulated data only recorded 48% of these malnourished clients, far less than that (82%) identified from the progress notes, suggesting the importance of using NLP technology to uncover the information from nursing notes to fully understand the health status of the vulnerable older people in residential aged care.

CONCLUSION

This study identified 33% of older people suffered from malnutrition, lower than those reported in the similar setting in previous studies. Our study demonstrates that NLP technology is important for uncovering the key information about health risks for older people in residential aged care. Future research can apply NLP to predict other health risks for older people in this setting.

Using machine learning to detect sarcopenia from electronic health records

Introduction

Sarcopenia (low muscle mass and strength) causes dysmobility and loss of independence. Sarcopenia is often not directly coded or described in electronic health records (EHR). The objective was to improve sarcopenia detection using structured data from EHR.

Methods

Adults undergoing musculoskeletal testing (December 2017-March 2020) were classified as meeting sarcopenia thresholds for 0 (controls), ≥1 (Sarcopenia-1), or ≥2 (Sarcopenia-2) tests. Electronic health record diagnoses, medications, and laboratory testing were extracted from the Indiana Network for Patient Care. Five machine learning models were applied to EHR data for predicting sarcopenia.

Results

Of 1304 participants, 1055 were controls, 249 met Sarcopenia-1 and 76 met Sarcopenia-2. Sarcopenic participants were older, with higher fat mass, Charlson Comorbidity Index, and more chronic diseases. All models performed better for Sarcopenia-2 than Sarcopenia-1. The top performing models for Sarcopenia-1 were Logistic Regression [area under the curve (AUC) 71.59 (95% confidence interval [CI], 71.51-71.66)] and Multi-Layer Perceptron [AUC 71.48 (95%CI, 71.00-71.97)]. The top performing models for Sarcopenia-2 were Logistic Regression [AUC 91.44 (95%CI, 91.28-91.60)] and Support Vector Machine [AUC 90.81 (95%CI, 88.41-93.20)]. For the best Logistic Regression Model, important sarcopenia predictors included diabetes mellitus, digestive system complaints, signs and symptoms involving the nervous, musculoskeletal and respiratory systems, metabolic disorders, and kidney or urinary tract disorders. Opioids, corticosteroids, and antihyperlipidemic drugs were also more common among sarcopenic participants.

Conclusions

Applying machine learning models, sarcopenia can be predicted from structured data in EHR, which may be developed through future studies to facilitate large-scale early detection and intervention in clinical populations.

Development of a machine learning model to predict mild cognitive impairment using natural language processing in the absence of screening

BACKGROUND

Patients and their loved ones often report symptoms or complaints of cognitive decline that clinicians note in free clinical text, but no structured screening or diagnostic data are recorded. These symptoms/complaints may be signals that predict who will go on to be diagnosed with mild cognitive impairment (MCI) and ultimately develop Alzheimer's Disease or related dementias. Our objective was to develop a natural language processing system and prediction model for identification of MCI from clinical text in the absence of screening or other structured diagnostic information.

METHODS

There were two populations of patients: 1794 participants in the Adult Changes in Thought (ACT) study and 2391 patients in the general population of Kaiser Permanente Washington. All individuals had standardized cognitive assessment scores. We excluded patients with a diagnosis of Alzheimer's Disease, Dementia or use of donepezil. We manually annotated 10,391 clinic notes to train the NLP model. Standard Python code was used to extract phrases from notes and map each phrase to a cognitive functioning concept. Concepts derived from the NLP system were used to predict future MCI. The prediction model was trained on the ACT cohort and 60% of the general population cohort with 40% withheld for validation. We used a least absolute shrinkage and selection operator logistic regression approach (LASSO) to fit a prediction model with MCI as the prediction target. Using the predicted case status from the LASSO model and known MCI from standardized scores, we constructed receiver operating curves to measure model performance.

RESULTS

Chart abstraction identified 42 MCI concepts. Prediction model performance in the validation data set was modest with an area under the curve of 0.67. Setting the cutoff for correct classification at 0.60, the classifier yielded sensitivity of 1.7%, specificity of 99.7%, PPV of 70% and NPV of 70.5% in the validation cohort.

DISCUSSION AND CONCLUSION

Although the sensitivity of the machine learning model was poor, negative predictive value was high, an important characteristic of models used for population-based screening. While an AUC of 0.67 is generally considered moderate performance, it is also comparable to several tests that are widely used in clinical practice.

Development and assessment of a natural language processing model to identify residential instability in electronic health records’ unstructured data: a comparison of 3 integrated healthcare delivery systems

Objective

To evaluate whether a natural language processing (NLP) algorithm could be adapted to extract, with acceptable validity, markers of residential instability (ie, homelessness and housing insecurity) from electronic health records (EHRs) of 3 healthcare systems.

Materials and methods

We included patients 18 years and older who received care at 1 of 3 healthcare systems from 2016 through 2020 and had at least 1 free-text note in the EHR during this period. We conducted the study independently; the NLP algorithm logic and method of validity assessment were identical across sites. The approach to the development of the gold standard for assessment of validity differed across sites. Using the EntityRuler module of spaCy 2.3 Python toolkit, we created a rule-based NLP system made up of expert-developed patterns indicating residential instability at the lead site and enriched the NLP system using insight gained from its application at the other 2 sites. We adapted the algorithm at each site then validated the algorithm using a split-sample approach. We assessed the performance of the algorithm by measures of positive predictive value (precision), sensitivity (recall), and specificity.

Results

The NLP algorithm performed with moderate precision (0.45, 0.73, and 1.0) at 3 sites. The sensitivity and specificity of the NLP algorithm varied across 3 sites (sensitivity: 0.68, 0.85, and 0.96; specificity: 0.69, 0.89, and 1.0).

Discussion

The performance of this NLP algorithm to identify residential instability in 3 different healthcare systems suggests the algorithm is generally valid and applicable in other healthcare systems with similar EHRs.

Conclusion

The NLP approach developed in this project is adaptable and can be modified to extract types of social needs other than residential instability from EHRs across different healthcare systems.

The utilization of specialized healthcare services among frail older adults in the Helsinki Birth Cohort Study

BACKGROUND

The association between frailty and specialized healthcare utilization is not well studied. We, therefore, examined the utilization of specialized healthcare services among frail Finnish older adults.

METHODS

A sub-sample of 1060 participants of the Helsinki Birth Cohort Study were followed prospectively for specialized healthcare utilization from nationwide registers between the years 2013 and 2017. The participants' frailty status was assessed according to Fried's criteria at a mean age of 71.0 (2.7 SD) years between the years 2011 and 2013. A negative binomial regression model was used to examine the association between frailty and the total number of visits, emergency visits, outpatient appointments separating the first outpatient appointments and the follow-up appointments, inpatient care including elective and non-elective hospital admissions and the total number of hospital days. We also calculated average length of stay (ALOS) and used the Kruskal-Wallis test to examine the differences between the groups.

RESULTS

After adjusting for covariates, frailty was significantly associated with the number of specialized healthcare visits (IRR 1.50, 95% CI = 1.04-2.15) and all subgroups of visits apart from follow-up outpatient appointments. Frailty was particularly strongly associated with the number of hospital days (IRR 5.24, 95% CI = 2.35-11.7) and notably with emergency visits (IRR = 2.26, 95% CI = 1.45-3.51) and hospital admissions (IRR 2.23, 95% CI = 1.39-3.56). Frail older adults had also higher ALOS compared to non-frail participants (p = .009).

CONCLUSIONS

Frailty increases the use of most specialized healthcare services. Preventative interventions against frailty are needed to decrease the burden on specialized healthcare systems.KEY MESSAGEFrailty is associated with the utilization of most specialized healthcare services, the most expensive part of the healthcare in most high-income countries.The association of frailty with inpatient care is particularly strong.Preventative interventions against frailty are needed to decrease the burden on specialized healthcare systems.

Development and validation of a prediction model for actionable aspects of frailty in the text of clinicians' encounter notes

OBJECTIVE

Frailty is a prevalent risk factor for adverse outcomes among patients with chronic lung disease. However, identifying frail patients who may benefit from interventions is challenging using standard data sources. We therefore sought to identify phrases in clinical notes in the electronic health record (EHR) that describe actionable frailty syndromes.

MATERIALS AND METHODS

We used an active learning strategy to select notes from the EHR and annotated each sentence for 4 actionable aspects of frailty: respiratory impairment, musculoskeletal problems, fall risk, and nutritional deficiencies. We compared the performance of regression, tree-based, and neural network models to predict the labels for each sentence. We evaluated performance with the scaled Brier score (SBS), where 1 is perfect and 0 is uninformative, and the positive predictive value (PPV).

RESULTS

We manually annotated 155 952 sentences from 326 patients. Elastic net regression had the best performance across all 4 frailty aspects (SBS 0.52, 95% confidence interval [CI] 0.49-0.54) followed by random forests (SBS 0.49, 95% CI 0.47-0.51), and multi-task neural networks (SBS 0.39, 95% CI 0.37-0.42). For the elastic net model, the PPV for identifying the presence of respiratory impairment was 54.8% (95% CI 53.3%-56.6%) at a sensitivity of 80%.

DISCUSSION

Classification models using EHR notes can effectively identify actionable aspects of frailty among patients living with chronic lung disease. Regression performed better than random forest and neural network models.

CONCLUSIONS

NLP-based models offer promising support to population health management programs that seek to identify and refer community-dwelling patients with frailty for evidence-based interventions.

Thyroid stimulating hormone levels and geriatric syndromes: secondary nested case-control study of the Mexican Health and Aging Study

PURPOSE

To determine the incidence of geriatric syndromes (GS) in community dwelling older adults with subclinical hypothyroidism.

METHODS

This is an analysis from the Mexican Health and Aging Study, of a subsample of 2089 subjects with TSH determination. From this last subsample, we included 1628 individuals with TSH levels in the subclinical range (4.5-10 µU/ml).

RESULTS

The multivariate analysis showed that when comparing data obtained from the 2012 wave with the 2015 wave results, there was a significant incidence of some GS such as falls (OR 1.79, CI 1.16-2.77, p = 0.0116), fatigue (OR 2.17, CI 1.40-3.38, p = 0.0348) and depression (OR 1.70, CI 1.06-2.71, p = 0.0246) among the subclinical hypothyroidism group.

CONCLUSION

This study showed a greater incidence of GS in subjects 50 years and older with sub-clinical hypothyroidism, when compared to those with normal thyroid function.

Use of Machine Learning and Artificial Intelligence Methods in Geriatric Mental Health Research Involving Electronic Health Record or Administrative Claims Data: A Systematic Review

Introduction: Electronic health records (EHR) and administrative healthcare data (AHD) are frequently used in geriatric mental health research to answer various health research questions. However, there is an increasing amount and complexity of data available that may lend itself to alternative analytic approaches using machine learning (ML) or artificial intelligence (AI) methods. We performed a systematic review of the current application of ML or AI approaches to the analysis of EHR and AHD in geriatric mental health. Methods: We searched MEDLINE, Embase, and PsycINFO to identify potential studies. We included all articles that used ML or AI methods on topics related to geriatric mental health utilizing EHR or AHD data. We assessed study quality either by Prediction model Risk OF Bias ASsessment Tool (PROBAST) or Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) checklist. Results: We initially identified 391 articles through an electronic database and reference search, and 21 articles met inclusion criteria. Among the selected studies, EHR was the most used data type, and the datasets were mainly structured. A variety of ML and AI methods were used, with prediction or classification being the main application of ML or AI with the random forest as the most common ML technique. Dementia was the most common mental health condition observed. The relative advantages of ML or AI techniques compared to biostatistical methods were generally not assessed. Only in three studies, low risk of bias (ROB) was observed according to all the PROBAST domains but in none according to QUADAS-2 domains. The quality of study reporting could be further improved. Conclusion: There are currently relatively few studies using ML and AI in geriatric mental health research using EHR and AHD methods, although this field is expanding. Aside from dementia, there are few studies of other geriatric mental health conditions. The lack of consistent information in the selected studies precludes precise comparisons between them. Improving the quality of reporting of ML and AI work in the future would help improve research in the field. Other courses of improvement include using common data models to collect/organize data, and common datasets for ML model validation.

Measuring the Value of a Practical Text Mining Approach to Identify Patients With Housing Issues in the Free-Text Notes in Electronic Health Record: Findings of a Retrospective Cohort Study

Introduction: Despite the growing efforts to standardize coding for social determinants of health (SDOH), they are infrequently captured in electronic health records (EHRs). Most SDOH variables are still captured in the unstructured fields (i.e., free-text) of EHRs. In this study we attempt to evaluate a practical text mining approach (i.e., advanced pattern matching techniques) in identifying phrases referring to housing issues, an important SDOH domain affecting value-based healthcare providers, using EHR of a large multispecialty medical group in the New England region, United States. To present how this approach would help the health systems to address the SDOH challenges of their patients we assess the demographic and clinical characteristics of patients with and without housing issues and briefly look into the patterns of healthcare utilization among the study population and for those with and without housing challenges.

Methods: We identified five categories of housing issues [i.e., homelessness current (HC), homelessness history (HH), homelessness addressed (HA), housing instability (HI), and building quality (BQ)] and developed several phrases addressing each one through collaboration with SDOH experts, consulting the literature, and reviewing existing coding standards. We developed pattern-matching algorithms (i.e., advanced regular expressions), and then applied them in the selected EHR. We assessed the text mining approach for recall (sensitivity) and precision (positive predictive value) after comparing the identified phrases with manually annotated free-text for different housing issues.

Results: The study dataset included EHR structured data for a total of 20,342 patients and 2,564,344 free-text clinical notes. The mean (SD) age in the study population was 75.96 (7.51). Additionally, 58.78% of the cohort were female. BQ and HI were the most frequent housing issues documented in EHR free-text notes and HH was the least frequent one. The regular expression methodology, when compared to manual annotation, had a high level of precision (positive predictive value) at phrase, note, and patient levels (96.36, 95.00, and 94.44%, respectively) across different categories of housing issues, but the recall (sensitivity) rate was relatively low (30.11, 32.20, and 41.46%, respectively).

Conclusion: Results of this study can be used to advance the research in this domain, to assess the potential value of EHR's free-text in identifying patients with a high risk of housing issues, to improve patient care and outcomes, and to eventually mitigate socioeconomic disparities across individuals and communities.

Social and Behavioral Determinants of Health in the Era of Artificial Intelligence with Electronic Health Records: A Scoping Review

Background. There is growing evidence that social and behavioral determinants of health (SBDH) play a substantial effect in a wide range of health outcomes. Electronic health records (EHRs) have been widely employed to conduct observational studies in the age of artificial intelligence (AI). However, there has been limited review into how to make the most of SBDH information from EHRs using AI approaches.Methods. A systematic search was conducted in six databases to find relevant peer-reviewed publications that had recently been published. Relevance was determined by screening and evaluating the articles. Based on selected relevant studies, a methodological analysis of AI algorithms leveraging SBDH information in EHR data was provided.Results. Our synthesis was driven by an analysis of SBDH categories, the relationship between SBDH and healthcare-related statuses, natural language processing (NLP) approaches for extracting SBDH from clinical notes, and predictive models using SBDH for health outcomes.Discussion. The associations between SBDH and health outcomes are complicated and diverse; several pathways may be involved. Using NLP technology to support the extraction of SBDH and other clinical ideas simplifies the identification and extraction of essential concepts from clinical data, efficiently unlocks unstructured data, and aids in the resolution of unstructured data-related issues.Conclusion. Despite known associations between SBDH and diseases, SBDH factors are rarely investigated as interventions to improve patient outcomes. Gaining knowledge about SBDH and how SBDH data can be collected from EHRs using NLP approaches and predictive models improves the chances of influencing health policy change for patient wellness, ultimately promoting health and health equity.

Clinical validation of nursing diagnosis Fragile Elderly Syndrome

OBJECTIVE

to clinically validate the nursing diagnosis of NANDA-I Frail Elderly Syndrome in hospitalized elderly.

METHOD

a methodological study, guided by the STROBE instrument, composed of 40 elderly people admitted to a teaching hospital in Paraíba, Brazil. The last phase of Hoskins' Nursing Diagnostic Validation Model: clinical validation was adopted. Data collection took place from August to December 2018. The data were analyzed using univariate descriptive statistics. It was approved by the hospital's ethics and research committee.

RESULTS

nine defining characteristics were validated; seven risk factors; six populations at risk and two associated conditions.

CONCLUSION

the validation of the nursing diagnosis of the Frail Elderly Syndrome in our socio-cultural context was considered appropriate, being an important step for critical thinking that underlies the decision-making of nurses in the care of the frail elderly, as well as professional practice.

Translational NLP: A New Paradigm and General Principles for Natural Language Processing Research

Natural language processing (NLP) research combines the study of universal principles, through basic science, with applied science targeting specific use cases and settings. However, the process of exchange between basic NLP and applications is often assumed to emerge naturally, resulting in many innovations going unapplied and many important questions left unstudied. We describe a new paradigm of Translational NLP, which aims to structure and facilitate the processes by which basic and applied NLP research inform one another. Translational NLP thus presents a third research paradigm, focused on understanding the challenges posed by application needs and how these challenges can drive innovation in basic science and technology design. We show that many significant advances in NLP research have emerged from the intersection of basic principles with application needs, and present a conceptual framework outlining the stakeholders and key questions in translational research. Our framework provides a roadmap for developing Translational NLP as a dedicated research area, and identifies general translational principles to facilitate exchange between basic and applied research.

Mortality, falls and fracture risk are positively associated with frailty: a SIDIAP cohort study of 890,000 patients

Background

Frail subjects are at increased risk of adverse outcomes. We aimed to assess their risk of falls, all-cause mortality, and fractures.

Method

We used a retrospective cohort study using the Sistema d’Informació per al Desenvolupament de l’Investigació en Atenció Primària database (>6 million residents). Subjects aged 75 years and older with ≥1 year of valid data (2007–2015) were included. Follow-up was carried out from (the latest of) the date of cohort entry up to migration, end of the study period or outcome (whichever came first). The eFRAGICAP classified subjects as fit, mild, moderate, or severely frail. Outcomes (10th revision of the International Classification of Diseases) were incident falls, fractures (overall/hip/vertebral), and all-cause mortality during the study period. Statistics: hazard ratios (HRs), 95% CI adjusted (per age, sex, and socioeconomic status), and unadjusted cause-specific Cox models, accounting for competing risk of death (fit group as the reference).

Results

A total of 893 211 subjects were analyzed; 54.4% were classified as fit, 34.0% as mild, 9.9% as moderate, and 1.6% as severely frail. Compared with the fit, frail had an increased risk of falls (adjusted HR [95% CI] of 1.55 [1.52–1.58], 2.74 [2.66–2.84], and 5.94 [5.52–6.40]), all-cause mortality (adjusted HR [95% CI] of 1.36 [1.35–1.37], 2.19 [2.16–2.23], and 4.29 [4.13–4.45]), and fractures (adjusted HR [95% CI] of 1.21 [1.20–1.23], 1.51 [1.47–1.55], and 2.36 [2.20–2.53]) for mild, moderate, and severe frailty, respectively. Severely frail had a high risk of vertebral (HR of 2.49 [1.99–3.11]) and hip fracture (HR [95% CI] of 1.85 [1.50–2.28]). Accounting for competing risk of death did not change results.

Conclusion

Frail subjects are at increased risk of death, fractures, and falls. The eFRAGICAP tool can easily assess frailty in electronic primary care databases in Spain.

Frailty Recognition by Clinicians and its Impact on Advance Care Planning

BACKGROUND

Frailty has important implications for the care of the elderly and how their needs are met.

OBJECTIVE

To assess clinicians' acknowledgement of frailty in the electronic medical records (EMR) and the impact of frailty recognition on advance care planning (ACP).

METHODS

We performed a retrospective study on 119 patients 65 years or older with moderate or severe frailty assessed using a validated frailty scale. We reviewed notes to determine if primary team identified frailty and obtained data regarding ACP planning. We present the characteristics and outcomes of patients who were identified as frail and compared them with patients whose frailty was unrecognized in EMR.

RESULTS

Among the 119 frail patients, one third were ≥85 years and one-year mortality was 25.4%. Most patients were taking ≥5 medications and only 14.3% rated their health as excellent or good prior to hospitalization. Only 15 patients (12.6%) were identified as frail in the EMR. The only significant differences between those recognized versus unrecognized frail were body mass index (23.4 vs 28.6, p = 0.02) and reported weight loss in the 3 months prior to admission (93.3% vs 59.6%, p = 0.009). Geriatric or palliative care consults, and changes in code status to do-not resuscitate were more frequent among those recognized vs not. (33.3% vs 11.5%; 13.3% vs 1.9% respectively).

CONCLUSION

Documentation of frailty in the EMR was rare and it was associated with a lower likelihood of providing advance care planning. These findings suggest a need for consistent frailty assessment, which might promote patient-centered care.

Automated Coding of Under-Studied Medical Concept Domains: Linking Physical Activity Reports to the International Classification of Functioning, Disability, and Health

Linking clinical narratives to standardized vocabularies and coding systems is a key component of unlocking the information in medical text for analysis. However, many domains of medical concepts, such as functional outcomes and social determinants of health, lack well-developed terminologies that can support effective coding of medical text. We present a framework for developing natural language processing (NLP) technologies for automated coding of medical information in under-studied domains, and demonstrate its applicability through a case study on physical mobility function. Mobility function is a component of many health measures, from post-acute care and surgical outcomes to chronic frailty and disability, and is represented as one domain of human activity in the International Classification of Functioning, Disability, and Health (ICF). However, mobility and other types of functional activity remain under-studied in the medical informatics literature, and neither the ICF nor commonly-used medical terminologies capture functional status terminology in practice. We investigated two data-driven paradigms, classification and candidate selection, to link narrative observations of mobility status to standardized ICF codes, using a dataset of clinical narratives from physical therapy encounters. Recent advances in language modeling and word embedding were used as features for established machine learning models and a novel deep learning approach, achieving a macro-averaged F-1 score of 84% on linking mobility activity reports to ICF codes. Both classification and candidate selection approaches present distinct strengths for automated coding in under-studied domains, and we highlight that the combination of (i) a small annotated data set; (ii) expert definitions of codes of interest; and (iii) a representative text corpus is sufficient to produce high-performing automated coding systems. This research has implications for continued development of language technologies to analyze functional status information, and the ongoing growth of NLP tools for a variety of specialized applications in clinical care and research.

Examining the Interfacility Variation of Social Determinants of Health in the Veterans Health Administration

Introduction

Recently, numerous studies have linked social determinants of health (SDoH) with clinical outcomes. While this association is well known, the interfacility variability of these risk favors within the Veterans Health Administration (VHA) is not known. Such information could be useful to the VHA for resource and funding allocation. The aim of this study is to explore the interfacility variability of 5 SDoH within the VHA.

Methods

In a cohort of patients (aged ≥ 65 years) hospitalized at VHA acute care facilities with either acute myocardial infarction (AMI), heart failure (HF), or pneumonia in 2012, we assessed (1) the proportion of patients with any of the following five documented SDoH: lives alone, marginal housing, alcohol use disorder, substance use disorder, and use of substance use services, using administrative diagnosis codes and clinic stop codes; and (2) the documented facility-level variability of these SDoH. To examine whether variability was due to regional coding differences, we assessed the variation of living alone using a validated natural language processing (NLP) algorithm.

Results

The proportion of veterans admitted for AMI, HF, and pneumonia with SDoH was low. Across all 3 conditions, lives alone was the most common SDoH (2.2% [interquartile range (IQR), 0.7-4.7]), followed by substance use disorder (1.3% [IQR, 0.5-2.1]), and use of substance use services (1.2% [IQR, 0.6-1.8]). Using NLP, the proportion of hospitalized veterans with lives alone was higher for HF (14.4% vs 2.0%, P < .01), pneumonia (11% vs 1.9%, P < .01), and AMI (10.2% vs 1.4%, P < .01) compared with International Classification of Diseases, Ninth Edition codes. Interfacility variability was noted with both administrative and NLP extraction methods.

Conclusions

The presence of SDoH in administrative data among patients hospitalized for common medical issues is low and variable across VHA facilities. Significant facility-level variation of 5 SDoH was present regardless of extraction method.

Electronic Medical Record-Based Case Phenotyping for the Charlson Conditions: Scoping Review

Background

Electronic medical records (EMRs) contain large amounts of rich clinical information. Developing EMR-based case definitions, also known as EMR phenotyping, is an active area of research that has implications for epidemiology, clinical care, and health services research.

Objective

This review aims to describe and assess the present landscape of EMR-based case phenotyping for the Charlson conditions.

Methods

A scoping review of EMR-based algorithms for defining the Charlson comorbidity index conditions was completed. This study covered articles published between January 2000 and April 2020, both inclusive. Embase (Excerpta Medica database) and MEDLINE (Medical Literature Analysis and Retrieval System Online) were searched using keywords developed in the following 3 domains: terms related to EMR, terms related to case finding, and disease-specific terms. The manuscript follows the Preferred Reporting Items for Systematic reviews and Meta-analyses extension for Scoping Reviews (PRISMA) guidelines.

Results

A total of 274 articles representing 299 algorithms were assessed and summarized. Most studies were undertaken in the United States (181/299, 60.5%), followed by the United Kingdom (42/299, 14.0%) and Canada (15/299, 5.0%). These algorithms were mostly developed either in primary care (103/299, 34.4%) or inpatient (168/299, 56.2%) settings. Diabetes, congestive heart failure, myocardial infarction, and rheumatology had the highest number of developed algorithms. Data-driven and clinical rule–based approaches have been identified. EMR-based phenotype and algorithm development reflect the data access allowed by respective health systems, and algorithms vary in their performance.

Conclusions

Recognizing similarities and differences in health systems, data collection strategies, extraction, data release protocols, and existing clinical pathways is critical to algorithm development strategies. Several strategies to assist with phenotype-based case definitions have been proposed.

Identifying vulnerable older adult populations by contextualizing geriatric syndrome information in clinical notes of electronic health records

OBJECTIVE

Geriatric syndromes such as functional disability and lack of social support are often not encoded in electronic health records (EHRs), thus obscuring the identification of vulnerable older adults in need of additional medical and social services. In this study, we automatically identify vulnerable older adult patients with geriatric syndrome based on clinical notes extracted from an EHR system, and demonstrate how contextual information can improve the process.

MATERIALS AND METHODS

We propose a novel end-to-end neural architecture to identify sentences that contain geriatric syndromes. Our model learns a representation of the sentence and augments it with contextual information: surrounding sentences, the entire clinical document, and the diagnosis codes associated with the document. We trained our system on annotated notes from 85 patients, tuned the model on another 50 patients, and evaluated its performance on the rest, 50 patients.

RESULTS

Contextual information improved classification, with the most effective context coming from the surrounding sentences. At sentence level, our best performing model achieved a micro-F1 of 0.605, significantly outperforming context-free baselines. At patient level, our best model achieved a micro-F1 of 0.843.

DISCUSSION

Our solution can be used to expand the identification of vulnerable older adults with geriatric syndromes. Since functional and social factors are often not captured by diagnosis codes in EHRs, the automatic identification of the geriatric syndrome can reduce disparities by ensuring consistent care across the older adult population.

CONCLUSION

EHR free-text can be used to identify vulnerable older adults with a range of geriatric syndromes.

New Horizons in the use of routine data for ageing research

The past three decades have seen a steady increase in the availability of routinely collected health and social care data and the processing power to analyse it. These developments represent a major opportunity for ageing research, especially with the integration of different datasets across traditional boundaries of health and social care, for prognostic research and novel evaluations of interventions with representative populations of older people. However, there are considerable challenges in using routine data at the level of coding, data analysis and in the application of findings to everyday care. New Horizons in applying routine data to investigate novel questions in ageing research require a collaborative approach between clinicians, data scientists, biostatisticians, epidemiologists and trial methodologists. This requires building capacity for the next generation of research leaders in this important area. There is a need to develop consensus code lists and standardised, validated algorithms for common conditions and outcomes that are relevant for older people to maximise the potential of routine data research in this group. Lastly, we must help drive the application of routine data to improve the care of older people, through the development of novel methods for evaluation of interventions using routine data infrastructure. We believe that harnessing routine data can help address knowledge gaps for older people living with multiple conditions and frailty, and design interventions and pathways of care to address the complex health issues we face in caring for older people.

Potential value of electronic health records in capturing data on geriatric frailty for population health

OBJECTIVES

Despite the availability of many frailty measures to identify older adults at risk, frailty instruments are not routinely used for risk assessment in population health management. Here, we assessed the potential value of electronic health records (EHRs) and administrative claims in providing the necessary data for variables used across various frailty instruments.

SETTING AND PARTICIPANTS

The review focused on studies conducted worldwide. Participants included older people aged 50 and older.

DESIGN

We identified frailty instruments published between 2011 and 2018. Frailty variables used in each of the frailty instruments were extracted, grouped, and categorized across health determinants and various clinical factors.

MEASURES

The availability of the extracted frailty variables across various data sources (e.g., EHRs, administrative claims, and surveys) was evaluated by experts.

RESULTS

We identified 135 frailty instruments, which contained 593 unique variables. Clinical determinants of health were the best represented variables across frailty instruments (n = 516; 87 %), unlike social and health services factors (n = 33; ∼5% and n = 32; ∼5%). Most frailty instruments require at least one variable that is not routinely available in EHRs or claims (n = 113; ∼83 %). Only 22 frailty instruments have the potential to completely rely on EHR (structured or free-text data) and/or claims data, and possibly be operationalized on a population-level.

CONCLUSIONS AND IMPLICATIONS

Frailty instruments continue to be highly survey-based. More research is therefore needed to develop EHR-based frailty instruments for population health management. This will permit organizations and societies to stratify risk and better allocate resources among different older adult populations.

Frailty status and increased risk for falls: The role of anticholinergic burden

PURPOSE OF THE STUDY

Frailty leads to serious adverse outcomes including falls. The relationship between frailty and falls has not been evaluated in the context of the side effects of drugs with anticholinergic properties. The aim of this study was to examine the potential association of anticholinergic burden (ACB) with the risk of falls among frail older adults.

DESIGN AND METHODS

Community-dwelling older adults were consecutively selected from the geriatrics outpatient clinic. Based on a fall history in the last 12 months, the participants were grouped as fallers and non-fallers. Frailty status was assessed by Fried's phenotype method. Exposure to anticholinergic medications was estimated using the ACB scale, and the participants were classified into ACB_0 (none), ACB_1 (possible) and ACB_2+ (definite).

RESULTS

The study included 520 older adults (mean age 77.7 years, 62.7 % female), with a fall prevalence of 25.8 % 12 months past. The proportions of frailty and pre-frailty were 33.1 % and 57.4 %, respectively. After adjustment for study confounders, receiving at least 1 drug with either possible or definite anticholinergic properties was independently associated to falls in frail [OR = 3.84 (1.48-9.93), p = 0.006] and pre-frail participants [OR = 2.71 (1.25-5.89); p = 0.012], but not in robust subjects. Moreover, ACB was significantly associated with the frailty components on adjusted analysis (p's<0.05).

IMPLICATIONS

Current study showed that the use of any drugs with possible or definite anticholinergic properties was associated with an increased risk of falls in frail older adults. The results emphasize the importance of medication management with respect to fall prevention in these patients.

Measuring Frailty in Health Care Databases for Clinical Care and Research

Considering the increasing burden and serious consequences of frailty in aging populations, there is increasing interest in measuring frailty in health care databases for clinical care and research. This review synthesizes the latest research on the development and application of 21 frailty measures for health care databases. Frailty measures varied widely in terms of target population (16 ambulatory, 1 long-term care, and 4 inpatient), data source (16 claims-based and 5 electronic health records [EHR]-based measures), assessment period (6 months to 36 months), data types (diagnosis codes required for 17 measures, health service codes for 7 measures, pharmacy data for 4 measures, and other information for 9 measures), and outcomes for validation (clinical frailty for 7 measures, disability for 7 measures, and mortality for 16 measures). These frailty measures may be useful to facilitate frailty screening in clinical care and quantify frailty for large database research in which clinical assessment is not feasible.

The application of artificial intelligence (AI) techniques to identify frailty within a residential aged care administrative data set

INTRODUCTION

Research has shown that frailty, a geriatric syndrome associated with an increased risk of negative outcomes for older people, is highly prevalent among residents of residential aged care facilities (also called long term care facilities or nursing homes). However, progress on effective identification of frailty within residential care remains at an early stage, necessitating the development of new methods for accurate and efficient screening.

OBJECTIVES

We aimed to determine the effectiveness of artificial intelligence (AI) algorithms in accurately identifying frailty among residents aged 75 years and over in comparison with a calculated electronic Frailty Index (eFI) based on a routinely-collected residential aged care administrative data set drawn from 10 residential care facilities located in Queensland, Australia. A secondary objective included the identification of best-performing candidate algorithms.

METHODS

We designed a frailty prediction system based on the eFI identification of frailty, allocating 84.5 % and 15.5 % of the data to training and test data sets respectively. We compared the performance of 18 specific scenarios to predict frailty against eFI based on unique combinations of three ML algorithms (support vector machines [SVM], decision trees [DT] and K-nearest neighbours [KNN]) and six cases (6, 10, 11, 14, 39 and 70 input variables). We calculated accuracy, percentage positive and negative agreement, sensitivity, specificity, Cohen's kappa and Prevalence- and Bias- Adjusted Kappa (PABAK), table frequencies and positive and negative predictive values.

RESULTS

Of 592 eligible resident records, 500 were allocated to the training set and 92 to the test set. Three scenarios (10, 11 and 70 input variables), all based on SVM algorithm, returned overall accuracy above 75 %.

CONCLUSIONS

There is some potential for AI techniques to contribute towards better frailty identification within residential care. However, potential benefits will need to be weighed against administrative burden, data quality concerns and presence of potential bias.

Burden and impact of multifactorial geriatric syndromes in allogeneic hematopoietic cell transplantation for older adults

Multifactorial geriatric syndromes are highly prevalent in older patients with cancer. Because an increasing number of older patients undergo allogeneic hematopoietic stem cell transplantation (allo-HCT), we examined the incidence and impact of transplant-related geriatric syndromes using our institutional database and electronic medical records. We identified 527 patients age 60 years or older who had undergone first allo-HCT from 2001 to 2016 for hematologic malignancies. From the initiation of conditioning to 100 days posttransplant, new geriatric syndromes were predominantly delirium with a cumulative incidence of 21% (95% confidence interval [CI], 18%-25%) at day 100 followed by fall at 7% (95% CI, 5%-9%). In multivariable analyses of available pretransplant variables, fall within the last year, potentially inappropriate use of medication, thrombocytopenia, and reduced creatinine clearance were significantly associated with delirium; age older than 70 years and impaired activities of daily living were significantly associated with fall. In the 100-day landmark analysis, both delirium (hazard ratio [HR], 1.66; 95% CI, 1.09-2.52; P = .023) and fall (HR, 2.14; 95% CI, 1.16-3.95; P = .026) were significantly associated with increased nonrelapse mortality; moreover, fall (HR, 1.93; 95% CI, 1.18-3.14; P = .016), but not delirium, was significantly associated with reduced overall survival. Here, we establish baseline incidences and risk factors of common transplant-related geriatric syndromes. Importantly, we demonstrate significant associations of delirium and fall with inferior transplant outcomes. The burden and impact of transplant-related geriatric syndromes warrant the institution of patient-centered, preemptive, longitudinal, and multidisciplinary interventions to improve outcomes for older allo-HCT patients.

Modified frailty index and hypoalbuminemia as predictors of adverse outcomes in older adults presenting to acute general surgical unit

INTRODUCTION

Health professionals are progressively drawing on the concept of frailty as a determinant of adverse surgical outcomes in of older adults. We aimed to determine the prevalence of frailty and the correlation between frailty and mortality among older adults admitted to the acute surgical unit.

MATERIALS AND METHODS

This prospective cohort study was conducted in the acute general surgical unit over a two month period. We recruited 150 consecutive patients aged 65yrs and above. The modified frailty index was employed to measure frailty and the albumin levels on admission were obtained from electronic medical records. The patients were followed up for a period of thirty days.

RESULTS

We found that more than 40% of the older adults admitted to the acute general surgical unit were frail and frailty was associated with higher rate of mortality at 30 days. Hypoalbuminemia was associated with a longer length of stay, higher rate of complications, and an increased likelihood of discharge to a rehabilitation facility. There was also a significant univariate correlation between frailty and the presence of hypoalbuminemia on admission.

CONCLUSION

Frailty and hypoalbuminemia are common in older general surgical patients and predict the likelihood of some of the adverse outcomes relevant to older adults and health economy such as mortality, increased length of stay, rate of complications, and likelihood of discharge to a rehabilitation facility. Further studies should investigate a possible causal association between frailty and low albumin levels in an acute surgical setting.

[Characteristics of pharmacotherapy in older patients with rheumatism]

Due to medical advances and the availability of efficient immunosuppressive therapies, the life-expectancy of people suffering from inflammatory rheumatic diseases is continuously increasing. In Germany, geriatric patients (definition: age older than 70 years combined with geriatric multimorbidity) affected, e. g. by rheumatoid arthritis (RA) frequently receive corticosteroids and less often biologic disease-modifying antirheumatic drugs (bDMARDs) and conventional DMARDs (cDMARDs), which is justified by additionally existing comorbidities and polypharmacy. Using geriatric typical assessments as well as detailed medication regimens the treatment risk of bDMARD and cDMARD administration can be properly evaluated. Current data on biological therapy in older patients with rheumatism support this recommendation. Following the "choosing wisely" initiative of the German Association of Internal Medicine the authors listed 5 positive and 5 negative recommendations concerning the pharmacotherapy of older patients suffering from rheumatism (e. g. RA) as practical guidance towards safer bDMARD and cDMARD treatment for geriatric RA patients.

Outcomes and Predictors of Mortality in Hospitalized Frail Patients Undergoing Percutaneous Coronary Intervention

Objective To study the impact of frailty on inpatient outcomes among patients undergoing percutaneous coronary intervention (PCI). Methods The National Inpatient Sample data of all PCI-related hospitalizations throughout the United States (US) from 2010 through 2014 was utilized. Patients were divided into two groups: frailty and no-frailty. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes were used to stratify groups and outcomes. In order to address the substantial difference in the total number of valid observations between the two groups, a propensity-matched analysis was performed at a 1:1 ratio and caliper width of 0.01. Results A total of 2,612,661 PCI-related hospitalizations throughout the US from 2010 through 2014 were identified, out of which 16,517 admissions (0.6%) had coexisting frailty. Only 1:1 propensity-matched data was utilized for the study. Propensity-matched frailty group (n=14,717) as compared to no-frailty (n=14,755) was frequently older, white, and Medicare enrollee (p<0.05). The frailty group had significantly higher rates of comorbidities and complications (p<0.05). All-cause in-hospital mortality was higher in the no-frailty group (p<0.05). Age, white race, non-elective admission, urban hospitals, and comorbidities predicted in-hospital mortality in frailty group (p<0.05). Rheumatoid arthritis, depression, hypertension, obesity, dyslipidemia, and history of previous PCI decreased odds of in-hospital mortality in frailty group (p<0.05). Frailty group had prolonged hospital stay and higher hospital charges (p<0.05). Conclusions Frailty has a significant effect on PCI-related outcomes. We present a previously unknown protective effect of cardiovascular disease risk factors and other health risk factors on frail patients undergoing PCI. Frailty's inclusion in risk stratification will help in predicting the post-procedure complications and improve resource utilization.

Assessing the Availability of Data on Social and Behavioral Determinants in Structured and Unstructured Electronic Health Records: A Retrospective Analysis of a Multilevel Health Care System

Background

Most US health care providers have adopted electronic health records (EHRs) that facilitate the uniform collection of clinical information. However, standardized data formats to capture social and behavioral determinants of health (SBDH) in structured EHR fields are still evolving and not adopted widely. Consequently, at the point of care, SBDH data are often documented within unstructured EHR fields that require time-consuming and subjective methods to retrieve. Meanwhile, collecting SBDH data using traditional surveys on a large sample of patients is infeasible for health care providers attempting to rapidly incorporate SBDH data in their population health management efforts. A potential approach to facilitate targeted SBDH data collection is applying information extraction methods to EHR data to prescreen the population for identification of immediate social needs.

Objective

Our aim was to examine the availability and characteristics of SBDH data captured in the EHR of a multilevel academic health care system that provides both inpatient and outpatient care to patients with varying SBDH across Maryland.

Methods

We measured the availability of selected patient-level SBDH in both structured and unstructured EHR data. We assessed various SBDH including demographics, preferred language, alcohol use, smoking status, social connection and/or isolation, housing issues, financial resource strains, and availability of a home address. EHR’s structured data were represented by information collected between January 2003 and June 2018 from 5,401,324 patients. EHR’s unstructured data represented information captured for 1,188,202 patients between July 2016 and May 2018 (a shorter time frame because of limited availability of consistent unstructured data). We used text-mining techniques to extract a subset of SBDH factors from EHR’s unstructured data.

Results

We identified a valid address or zip code for 5.2 million (95.00%) of approximately 5.4 million patients. Ethnicity was captured for 2.7 million (50.00%), whereas race was documented for 4.9 million (90.00%) and a preferred language for 2.7 million (49.00%) patients. Information regarding alcohol use and smoking status was coded for 490,348 (9.08%) and 1,728,749 (32.01%) patients, respectively. Using the International Classification of Diseases–10th Revision diagnoses codes, we identified 35,171 (0.65%) patients with information related to social connection/isolation, 10,433 (0.19%) patients with housing issues, and 3543 (0.07%) patients with income/financial resource strain. Of approximately 1.2 million unique patients with unstructured data, 30,893 (2.60%) had at least one clinical note containing phrases referring to social connection/isolation, 35,646 (3.00%) included housing issues, and 11,882 (1.00%) had mentions of financial resource strain.

Conclusions

Apart from demographics, SBDH data are not regularly collected for patients. Health care providers should assess the availability and characteristics of SBDH data in EHRs. Evaluating the quality of SBDH data can potentially enable health care providers to modify underlying workflows to improve the documentation, collection, and extraction of SBDH data from EHRs.

Development and validation of a pragmatic natural language processing approach to identifying falls in older adults in the emergency department

BACKGROUND

Falls among older adults are both a common reason for presentation to the emergency department, and a major source of morbidity and mortality. It is critical to identify fall patients quickly and reliably during, and immediately after, emergency department encounters in order to deliver appropriate care and referrals. Unfortunately, falls are difficult to identify without manual chart review, a time intensive process infeasible for many applications including surveillance and quality reporting. Here we describe a pragmatic NLP approach to automating fall identification.

METHODS

In this single center retrospective review, 500 emergency department provider notes from older adult patients (age 65 and older) were randomly selected for analysis. A simple, rules-based NLP algorithm for fall identification was developed and evaluated on a development set of 1084 notes, then compared with identification by consensus of trained abstractors blinded to NLP results.

RESULTS

The NLP pipeline demonstrated a recall (sensitivity) of 95.8%, specificity of 97.4%, precision of 92.0%, and F1 score of 0.939 for identifying fall events within emergency physician visit notes, as compared to gold standard manual abstraction by human coders.

CONCLUSIONS

Our pragmatic NLP algorithm was able to identify falls in ED notes with excellent precision and recall, comparable to that of more labor-intensive manual abstraction. This finding offers promise not just for improving research methods, but as a potential for identifying patients for targeted interventions, quality measure development and epidemiologic surveillance.

Application of an electronic Frailty Index in Australian primary care: data quality and feasibility assessment

BACKGROUND

The primary care setting is the ideal location for identifying the condition of frailty in older adults.

AIMS

The aim of this pragmatic study was twofold: (1) to identify data items to extract the data required for an electronic Frailty Index (eFI) from electronic health records (EHRs); and (2) test the ability of an eFI to accurately and feasibly identify frailty in older adults.

METHODS

In a rural South Australian primary care clinic, we derived an eFI from routinely collected EHRs using methodology described by Clegg et al. We assessed feasibility and accuracy of the eFI, including complexities in data extraction. The reference standard for comparison was Fried's frailty phenotype.

RESULTS

The mean (SD) age of participants was 80.2 (4.8) years, with 36 (60.0%) female (n = 60). Frailty prevalence was 21.7% by Fried's frailty phenotype, and 35.0% by eFI (scores > 0.21). When deriving the eFI, 85% of EHRs were perceived as easy or neutral difficulty to extract the required data from. Complexities in data extraction were present in EHRs of patients with multiple health problems and/or where the majority of data items were located other than on the patient's summary problem list.

DISCUSSION

This study demonstrated that it is entirely feasible to extract an eFI from routinely collected Australian primary care data. We have outlined a process for extracting an eFI from EHRs without needing to modify existing infrastructure. Results from this study can inform the development of automated eFIs, including which data items to best access data from.

Extraction of Geriatric Syndromes From Electronic Health Record Clinical Notes: Assessment of Statistical Natural Language Processing Methods

Background

Geriatric syndromes in older adults are associated with adverse outcomes. However, despite being reported in clinical notes, these syndromes are often poorly captured by diagnostic codes in the structured fields of electronic health records (EHRs) or administrative records.

Objective

We aim to automatically determine if a patient has any geriatric syndromes by mining the free text of associated EHR clinical notes. We assessed which statistical natural language processing (NLP) techniques are most effective.

Methods

We applied conditional random fields (CRFs), a widely used machine learning algorithm, to identify each of 10 geriatric syndrome constructs in a clinical note. We assessed three sets of features and attributes for CRF operations: a base set, enhanced token, and contextual features. We trained the CRF on 3901 manually annotated notes from 85 patients, tuned the CRF on a validation set of 50 patients, and evaluated it on 50 held-out test patients. These notes were from a group of US Medicare patients over 65 years of age enrolled in a Medicare Advantage Health Maintenance Organization and cared for by a large group practice in Massachusetts.

Results

A final feature set was formed through comprehensive feature ablation experiments. The final CRF model performed well at patient-level determination (macroaverage F1=0.834, microaverage F1=0.851); however, performance varied by construct. For example, at phrase-partial evaluation, the CRF model worked well on constructs such as absence of fecal control (F1=0.857) and vision impairment (F1=0.798) but poorly on malnutrition (F1=0.155), weight loss (F1=0.394), and severe urinary control issues (F1=0.532). Errors were primarily due to previously unobserved words (ie, out-of-vocabulary) and a lack of context.

Conclusions

This study shows that statistical NLP can be used to identify geriatric syndromes from EHR-extracted clinical notes. This creates new opportunities to identify patients with geriatric syndromes and study their health outcomes.