Mining for equitable health: Assessing the impact of missing data in electronic health records

Getzen, Emily J.; Ungar, Lyle H.; Mowery, Danielle L.; Jiang, Xiaoqian; Long, Qi

doi:10.1016/j.jbi.2022.104269

Cited by 43 publications

(30 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This highlighted incomplete recording by clinical staff within the electronic patient record, indicating that other individuals meeting the study criteria may not have been identified. Other studies have found missing data in electronic patient records of individuals with intellectual disability, and missing data has been found to lead to poorer access to health care for those individuals (Getzen et al , 2023; Soper et al , 2022). A possible explanation for the lack of ICD 10 codes entered on the electronic patient record is that diagnostic codes are often recorded in psychiatry clinic letters, rather than officially uploaded to the electronic patient record as is the expected procedure.…”

Section: Discussionmentioning

confidence: 99%

Assessing adherence to National Institute for Health and Care Excellence dementia assessment and diagnosis guidelines in adults with intellectual disability: a retrospective cohort study

Duncan,

Wilkinson,

Jaydeokar

et al. 2023

AMHID

View full text Add to dashboard Cite

Purpose This study aims to evaluate the dementia assessment and diagnosis care provided to adults with intellectual disability. The authors selected recommendations from the National Institute for Health and Care Excellence (NICE) standards which could be evidenced in clinical notes and aimed to identify characteristics which may be associated with improved adherence to these recommendations. Design/methodology/approach The study population was adults with an intellectual disability who were diagnosed with dementia between January 2019 and December 2022 by a UK-based intellectual disability service. Data to demonstrate adherence to selected recommendations and demographic and clinical characteristics were extracted from electronic patient records. Findings The authors identified 41 individuals. A mean of six of the eight recommendations were adhered to. There was low adherence with structural imaging to support dementia subtype diagnosis (9 individuals, 22%). This may be linked with the low percentage of people diagnosed with vascular dementia (1 individual, 2%) despite a national figure of 20%. No demographic or clinical characteristics were associated with level of adherence recorded. The authors found incomplete recording of diagnostic clinical coding in electronic patient records. This may disadvantage this population, as they cannot be readily identified for post diagnostic support or resource allocation. Originality/value To the best of the authors’ knowledge, this is the first study to examine adherence to these NICE guidelines in this population.

show abstract

Section: Discussionmentioning

confidence: 99%

Assessing adherence to National Institute for Health and Care Excellence dementia assessment and diagnosis guidelines in adults with intellectual disability: a retrospective cohort study

Duncan,

Wilkinson,

Jaydeokar

et al. 2023

AMHID

View full text Add to dashboard Cite

show abstract

“…A recent study using the Medical Information Mart for Intensive Care III (MIMIC-III) database demonstrates that the addition of missing data tends to more negatively impact the performance of disease prediction models for groups that have less access to healthcare [38]. Specifically, when stratified by insurance status, patients insured by Medicare/Medicaid experienced a sharper drop in model performance as the amount of missing data increased compared to those with private insurance [38].…”

Section: Phases Of Sample Selection and Evolutionmentioning

confidence: 99%

“…Accordingly, these patients' data may be spread across multiple health facilities, increasing the likelihood of missing observations in a single-center machine learning dataset. Individuals with low SES and racially minoritized individuals may also tend to use the healthcare system less often due to structural barriers resulting in less access, including receiving less testing, so studies examining an electronic health record would report more missing data for these patients [38].…”

Section: Input-level Exclusionmentioning

confidence: 99%

Participant Flow Diagrams for Health Equity in AI Research

Ellen,

Matos,

Viola

et al. 2023

Preprint

View full text Add to dashboard Cite

Biases in sample creation can arise at any study phase, including initial patient recruitment, exclusion criteria, input-level exclusion and outcome-level exclusion, and often reflect the underrepresentation or exclusion of demographic groups historically disadvantaged in medical research. The use of non-representative samples to construct clinical algorithms in artificial intelligence (AI) and machine learning (ML) applications may further amplify this selection bias. Building on the “Data Cards” initiative for transparency in AI research, we advocate for the addition of a detailed participant flow diagram for AI studies, emphasizing the need to detail excluded participant demographic characteristics at every study phase. This tracking of excluded participants enhances understanding of potential algorithmic biases before their clinical implementation, and thus deserves to be detailed in any medical AI study. We include both a model for this flow diagram as well as a brief case study explaining how it could be implemented in practice. Through standardized reporting of participant flow diagrams, we can better gauge the potential inequity embedded in AI applications, facilitating more reliable and equitable clinical algorithms.

show abstract

“…increased (2,3). In ICU cohort studies, missing data may introduce biased results as well as a loss of statistical power and precision (4).…”

mentioning

confidence: 99%

“…Using large cohorts in ICU research has become an increasingly common approach to answering clinically relevant questions (1). As the use of big data has increased, where data are often aggregated from multiple sources including electronic health records (EHRs), insurance claim databases, and government networks databases, concerns about data missingness have increased (2, 3). In ICU cohort studies, missing data may introduce biased results as well as a loss of statistical power and precision (4).…”

mentioning

confidence: 99%

Data Missingness Reporting and Use of Methods to Address It in Critical Care Cohort Studies

Wu,

Smith,

Vernooij

et al. 2023

Critical Care Explorations

View full text Add to dashboard Cite

IMPORTANCE: Failure to recognize and address data missingness in cohort studies may lead to biased results. Although Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines advocate data missingness reporting, the degree to which missingness is reported and addressed in the critical care literature remains unclear. OBJECTIVES: To review published ICU cohort studies to characterize data missingness reporting and the use of methods to address it. DESIGN, SETTING, AND PARTICIPANTS: We searched the 2022 table of contents of 29 critical care/critical care subspecialty journals having a 2021 impact factor greater than or equal to 3 to identify published prospective clinical or retrospective database cohort studies enrolling greater than or equal to 100 patients. MAIN OUTCOMES AND MEASURES: In duplicate, two trained researchers conducted a manuscript/supplemental material PDF word search for “missing*” and extracted study type, patient age, ICU type, sample size, missingness reporting, and the use of methods to address it. RESULTS: A total of 656 studies were reviewed. Of the 334 of 656 (50.9%) studies mentioning missingness, missingness was reported for greater than or equal to 1 variable in 234 (70.1%) and it exceeded 5% for at least one variable in 160 (47.9%). Among the 334 studies mentioning missingness, 88 (26.3%) used exclusion criteria, 36 (10.8%) used complete-case analysis, and 164 (49.1%) used a formal method to avoid missingness. In these 164 studies, imputation only was used in 100 (61.0%), an analytic strategy only in 24 (14.6%), and both in 40 (24.4%). Only missingness greater than 5% (in ≥ 1 variable) was independently associated with greater use of a missingness method (adjusted odds ratio 2.91; 95% CI, 1.85–4.60). Among 140 studies using imputation, multiple imputation was used in 87 studies (62.1%) and simple imputation in 49 studies (35.0%). For the 64 studies using an analytic method, 12 studies (18.8%) assigned missingness as an unknown category, whereas sensitivity analysis was used in 47 studies (73.4%). CONCLUSIONS AND RELEVANCE: Among published critical care cohort studies, only half mentioned result missingness, one-third reported actual missingness and only one-quarter used a method to manage missingness. Educational strategies to promote missingness reporting and resolution methods are required.

show abstract

Mining for equitable health: Assessing the impact of missing data in electronic health records

Cited by 43 publications

References 20 publications

Assessing adherence to National Institute for Health and Care Excellence dementia assessment and diagnosis guidelines in adults with intellectual disability: a retrospective cohort study

Assessing adherence to National Institute for Health and Care Excellence dementia assessment and diagnosis guidelines in adults with intellectual disability: a retrospective cohort study

Participant Flow Diagrams for Health Equity in AI Research

Data Missingness Reporting and Use of Methods to Address It in Critical Care Cohort Studies

Contact Info

Product

Resources

About