Derivation and validation of the automated search algorithms to identify cognitive impairment and dementia in electronic health records

Sakusic, Amra; O’Horo, John C.; Singh, Thakar; Wilson, Gregory; Kashyap, Rahul; Petersen, Ronald C.; Roberts, Rosebud O.; Fryer, John Denis; Rabinstein, Alejandro A.; Gajic, Ognjen

doi:10.1016/j.jcrc.2016.09.026

Cited by 47 publications

(55 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several other tools were recently developed that use text‐based EHR data to differentiate more accurately between patients with and without dementia . For example, Amra et al developed an algorithm that uses EHR text words (such as “cognitive impairment,” “impaired memory,” or “difficulty concentrating”) to discriminate between patients with and without dementia. A limitation is that many clinical settings may not have the resources or technical capabilities to extract information from free text fields such as clinic notes.…”

Section: Discussionmentioning

confidence: 99%

Development and Validation of eRADAR: A Tool Using EHR Data to Detect Unrecognized Dementia

Barnes

Zhou

Walker

et al. 2019

J American Geriatrics Society

View full text Add to dashboard Cite

OBJECTIVES Early recognition of dementia would allow patients and their families to receive care earlier in the disease process, potentially improving care management and patient outcomes, yet nearly half of patients with dementia are undiagnosed. Our aim was to develop and validate an electronic health record (EHR)‐based tool to help detect patients with unrecognized dementia (EHR Risk of Alzheimer's and Dementia Assessment Rule [eRADAR]). DESIGN Retrospective cohort study. SETTING Kaiser Permanente Washington (KPWA), an integrated healthcare delivery system. PARTICIPANTS A total of 16 665 visits among 4330 participants in the Adult Changes in Thought (ACT) study, who undergo a comprehensive process to detect and diagnose dementia every 2 years and have linked KPWA EHR data, divided into development (70%) and validation (30%) samples. MEASUREMENTS EHR predictors included demographics, medical diagnoses, vital signs, healthcare utilization, and medications within the previous 2 years. Unrecognized dementia was defined as detection in ACT before documentation in the KPWA EHR (ie, lack of dementia or memory loss diagnosis codes or dementia medication fills). RESULTS Overall, 1015 ACT visits resulted in a diagnosis of incident dementia, of which 498 (49%) were unrecognized in the KPWA EHR. The final 31‐predictor model included markers of dementia‐related symptoms (eg, psychosis diagnoses, antidepressant fills), healthcare utilization pattern (eg, emergency department visits), and dementia risk factors (eg, cerebrovascular disease, diabetes). Discrimination was good in the development (C statistic = .78; 95% confidence interval [CI] = .76‐.81) and validation (C statistic = .81; 95% CI = .78‐.84) samples, and calibration was good based on plots of predicted vs observed risk. If patients with scores in the top 5% were flagged for additional evaluation, we estimate that 1 in 6 would have dementia. CONCLUSION The eRADAR tool uses existing EHR data to detect patients with good accuracy who may have unrecognized dementia. J Am Geriatr Soc 68:103–111, 2019

show abstract

Section: Discussionmentioning

confidence: 99%

Development and Validation of eRADAR: A Tool Using EHR Data to Detect Unrecognized Dementia

Barnes

Zhou

Walker

et al. 2019

J American Geriatrics Society

View full text Add to dashboard Cite

show abstract

“…Others have explored the use of natural language processing or machine learning to discriminate between patients with and without dementia . For example, Amra et al developed electronic search algorithms with high sensitivity and specificity for identifying ADRD from EMR notes.…”

Section: Discussionmentioning

confidence: 99%

Passive Digital Signature for Early Identification of Alzheimer's Disease and Related Dementia

Boustani

Perkins

Khandker

et al. 2019

J American Geriatrics Society

View full text Add to dashboard Cite

OBJECTIVES Developing scalable strategies for the early identification of Alzheimer's disease and related dementia (ADRD) is important. We aimed to develop a passive digital signature for early identification of ADRD using electronic medical record (EMR) data. DESIGN A case‐control study. SETTING The Indiana Network for Patient Care (INPC), a regional health information exchange in Indiana. PARTICIPANTS Patients identified with ADRD and matched controls. MEASUREMENTS We used data from the INPC that includes structured and unstructured (visit notes, progress notes, medication notes) EMR data. Cases and controls were matched on age, race, and sex. The derivation sample consisted of 10 504 cases and 39 510 controls; the validation sample included 4500 cases and 16 952 controls. We constructed models to identify early 1‐ to 10‐year, 3‐ to 10‐year, and 5‐ to 10‐year ADRD signatures. The analyses included 14 diagnostic risk variables and 10 drug classes in addition to new variables produced from unstructured data (eg, disorientation, confusion, wandering, apraxia, etc). The area under the receiver operating characteristics (AUROC) curve was used to determine the best models. RESULTS The AUROC curves for the validation samples for the 1‐ to 10‐year, 3‐ to 10‐year, and 5‐ to 10‐year models that used only structured data were .689, .649, and .633, respectively. For the same samples and years, models that used both structured and unstructured data produced AUROC curves of .798, .748, and .704, respectively. Using a cutoff to maximize sensitivity and specificity, the 1‐ to 10‐year, 3‐ to 10‐year, and 5‐ to 10‐year models had sensitivity that ranged from 51% to 62% and specificity that ranged from 80% to 89%. CONCLUSION EMR‐based data provide a targeted and scalable process for early identification of risk of ADRD as an alternative to traditional population screening. J Am Geriatr Soc 68:511–518, 2020

show abstract

“…The course of cognitive impairment is long and involves many interfering factors, which would be difficult to detect from questionnaire results alone. Fifth, in accordance with the recommendations in (Amra et al, 2016), we stipulated that instances of cognitive impairment required at least two relevant ICD diagnoses. Thus, it is very likely that the risk of cognitive impairment was underestimated.…”

Section: Methodological Considerationsmentioning

confidence: 99%

A population‐based study of the association between hemodialysis and cognitive impairment

Chien¹,

Lin

Huang

et al. 2020

Asia-Pacific Psychiatry

View full text Add to dashboard Cite

Introduction: End-stage renal disease is a serious public health issue. The objective of this retrospective cohort study was to assess the association between hemodialysis and cognitive impairment, while controlling for age, sex, residence, and comorbidities. Methods: This study assesses the risk of cognitive impairment among a nationwide cohort of new hemodialysis patients derived from the NHIRD. Results: A total of 4330 patients were assigned to the dialysis group and 17 320 patients were assigned to the control group. A total of 2103 of the patients developed cognitive impairment within 2 years after the date of dialysis initiation. Patients who developed cognitive impairment were older (69.85 ± 11.56) than their counterparts who did not develop cognitive impairment (58.58 ± 14.77; P < .001). The logrank test of Kaplan-Meier analysis revealed a higher risk of cognitive impairment in the hemodialysis group than in the non-hemodialysis group (P < .001). The interval between dialysis initiation and the onset of cognitive impairment was 98.66 ± 46.39 months among non-dialysis subjects and 53.45 ± 41.90 months among dialysis subjects, and the between-group difference was significant (P < .001). The Cox Proportional Hazard Model revealed that after controlling for gender, age, residence, and comorbidities, hemodialysis was shown to have a significant impact on cognitive impairment (Hazard Ratio [HR]: 1.44; 95% confidence interval [CI]: 1.29-1.60). Furthermore, the risk of developing cognitive impairment increased with age (HR: 1.07; 95% CI: 1.06-1.08). Discussion: Hemodialysis was associated with cognitive impairment. There was a significant association between age and cognitive impairment, regardless of the comorbidities prior to hemodialysis. There was no evidence of an association between comorbidities and cognitive impairment after beginning hemodialysis.

show abstract

Derivation and validation of the automated search algorithms to identify cognitive impairment and dementia in electronic health records

Cited by 47 publications

References 9 publications

Development and Validation of eRADAR: A Tool Using EHR Data to Detect Unrecognized Dementia

Development and Validation of eRADAR: A Tool Using EHR Data to Detect Unrecognized Dementia

Passive Digital Signature for Early Identification of Alzheimer's Disease and Related Dementia

A population‐based study of the association between hemodialysis and cognitive impairment

Contact Info

Product

Resources

About