The International Classification of Functioning, Disability and Health (ICF) in Electronic Health Records

Maritz, Roxanne; Aronsky, Dominik; Prodinger, Birgit

doi:10.4338/aci-2017050078

Cited by 32 publications

(37 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We used 10-fold cross validation over the full Physical Therapy dataset for all of our machine learning-based experiments 12 . For parameter selection, including (i) selection of static and contextualized embedding models; (ii) selection of input features for classification models; (iii) selection of ICF code definition sources; and (iv) selection of highest-performing classification and candidate selection approaches; we held out 10% of data of each fold as development data (leaving 80% of the data for training and 10% for testing) and chose the settings that yielded best performance on this held-out data.…”

Section: Training and Evaluation Proceduresmentioning

confidence: 99%

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

Newman-Griffis¹,

Fosler‐Lussier²

2021

Front. Digit. Health

View full text Add to dashboard Cite

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.

show abstract

Section: Training and Evaluation Proceduresmentioning

confidence: 99%

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

Newman-Griffis¹,

Fosler‐Lussier²

2021

Front. Digit. Health

View full text Add to dashboard Cite

show abstract

“…To implement an accommodation, knowing the patient's diagnosis is less useful than knowing the limitation in function. A related thread of research has studied how functional status can be inferred from the medical narrative or the six ACS questions included in EHRs to enable analyses of health equity and the social determinants of health [42][43][44][45]. Whether for immediate use to facilitate a medical visit, or for broader understanding of disability and healthcare outcomes, the EHR wording to identify patients' disability-related functional limitations and needs requires careful consideration.…”

Section: Use Of Ehr Accommodation Needs Information At the Medical Visitmentioning

confidence: 99%

Can disability accommodation needs stored in electronic health records help providers prepare for patient visits? A qualitative study

Mudrick

Breslin

Nielsen

et al. 2020

BMC Health Serv Res

View full text Add to dashboard Cite

Background Embedding patient accommodation need in the electronic health record (EHR) has been proposed as one means to improve health care delivery to patients with disabilities. Accommodation need is not a standard field in commercial EHR software. However, some medical practices ask about accommodation need and store it in the EHR. Little is known about how the information is used, or barriers to its use. This exploratory-descriptive study examines whether and how information about patients’ disability-related accommodation needs stored in patient records is used in a primary health care center to plan for care. Methods Four focus groups (n = 35) were conducted with staff of a Federally Qualified Health Center that asks four accommodation questions at intake for the EHR. Respondents were asked how they learned about patient accommodation need, whether and how they used the information in the EHR, barriers to its use, and recommendations for where accommodation information should reside. A brief semi-structured interview was conducted with patients who had indicated an accommodation need (n = 12) to learn their experience at their most recent appointment. The qualitative data were coded using structural coding and themes extracted. Results Five themes were identified from the focus groups: (1) staff often do not know accommodation needs before the patient’s arrival; (2) electronic patient information systems offer helpful information, but their structure creates challenges and information gaps; (3) accommodations for a patient’s disability occur, but are developed at the time of visit; (4) provider knowledge of a regular patient is often the basis for accommodation preparation; and (5) staff recognize benefits to advance knowledge of accommodation needs and are supportive of methods to enable it. Most patients did not recall indicating accommodation need on the intake form. However, they expected to be accommodated based upon the medical practice’s knowledge of them. Conclusions Patient accommodation information in the EHR can be useful for visit planning. However, the structure must enable transfer of information between scheduling and direct care and be updatable as needs change. Flexibility to record a variety of needs, visibility to differentiate accommodation need from other alerts, and staff education about needs were recommended.

show abstract

“…All domains are considered to be potentially interlinked and might even affect the disease pathophysiology itself. 11 A core set of the ICF has been developed for MS and can be applied to characterize the functional domains where limitations can occur in MS. 12 For example, ambulatory dysfunction, cognitive dysfunction and fatigue are very common throughout the disease course and have great impact on physical activity, mood, quality of life and social participation. 13 Similarly, lifestyle factors affect physical and cognitive functions with active lifestyles being hypothesized to be protective.…”

Section: Four Reasons Why Consistent Longitudinal Multidisciplinary Smentioning

confidence: 99%

Multidisciplinary data infrastructures in multiple sclerosis: Why they are needed and can be done!

et al. 2018

View full text Add to dashboard Cite

Personalized treatment is highly desirable in multiple sclerosis (MS). We believe that multidisciplinary measurements including clinical, functional and patient-reported outcome measures in combination with extensive patient profiling can enhance personalized treatment and rehabilitation strategies. We elaborate on four reasons behind this statement: (1) MS disease activity and progression are complex and multidimensional concepts in nature and thereby defy a one-size-fits-all description, (2) functioning, progression, treatment, and rehabilitation effects are interdependent and should be investigated together, (3) personalized healthcare is based on the dynamics of system biology and on technology that confirms a patient’s fundamental biology and (4) inclusion of patient-reported outcome measures can facilitate patient-relevant healthcare. We discuss currently available multidisciplinary MS data initiatives and introduce joint actions to further increase the overall success. With this topical review, we hope to drive the MS community to invest in expanding towards more multidisciplinary and longitudinal data collection.

show abstract

The International Classification of Functioning, Disability and Health (ICF) in Electronic Health Records

Cited by 32 publications

References 36 publications

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

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

Can disability accommodation needs stored in electronic health records help providers prepare for patient visits? A qualitative study

Multidisciplinary data infrastructures in multiple sclerosis: Why they are needed and can be done!

Contact Info

Product

Resources

About