Data Quality in Electronic Health Record Research: An Approach for Validation and Quantitative Bias Analysis for Imperfectly Ascertained Health Outcomes Via Diagnostic Codes.

It is incumbent upon all researchers who use the electronic health record (EHR), including data scientists, to understand the quality of such data. EHR data may be subject to measurement error or misclassification that have the potential to bias results, unless one applies the available computational techniques specifically created for this problem. In this article, we begin with a discussion of data-quality issues in the EHR focusing on health outcomes. We review the concepts of sensitivity, specificity, positive and negative predictive values, and demonstrate how the imperfect classification of a dichotomous outcome variable can bias an analysis, both in terms of prevalence of the outcome, and relative risk of the outcome under one treatment regime (aka exposure) compared to another. This is then followed by a description of a generalizable approach to probabilistic (quantitative) bias analysis using a combination of regression estimation of the parameters that relate the true and observed data and application of these estimates to adjust the prevalence and relative risk that may have existed if there was no misclassification. We describe bias analysis that accounts for both random and systematic errors and highlight its limitations. We then motivate a case study with the goal of validating the accuracy of a health outcome, chronic infection with hepatitis C virus, derived from a diagnostic code in the EHR. Finally, we demonstrate our approaches on the case study and conclude by summarizing the literature on outcome misclassification and quantitative bias analysis.

A Qualitative Study of Implementing Universal Hepatitis C Screening Among Adults at an Urban Community-Based Health Provider in Delaware.

OBJECTIVES: We conducted a qualitative study of primary care providers to assess the challenges and opportunities in implementing a universal screening program for Hepatitis C Virus (HCV) at an urban community-based health center serving a largely disadvantaged population. METHODS: Qualitative semi-structured interviews of prescribing providers took place pre- and post-educational intervention, at a single federally qualified health center in Wilmington, Delaware, between September 2018 and July 2019. The intervention included a two-day didactic session and shadowing specialist providers. Data captured provider perspectives on universal screening and treatment. The interviews were transcribed verbatim, then grouped into codes, then finally, themes. RESULTS: Emergent themes included hesitancy in managing universal screening programs in the primary care environment, positive attitudes surrounding treatment, fewer HCV cases than expected, and concern with both patient-level barriers and practice-level barriers. Pre-intervention and post-intervention themes were similar. CONCLUSIONS: Implementation programs exploring universal HCV screening in the primary care environment should include educational opportunities that are available to all individuals in the practice, sustained organizational support, and available patient literature targeted to patients with varying health literacy and in languages other than English. In short, universal HCV screening and treatment is feasible in the primary medical environment but requires ongoing support and education for providers to ensure success.

Inverse probability weighting for selection bias in a Delaware community health center electronic medical record study of community deprivation and hepatitis C prevalence.

PURPOSE: To demonstrate how selection into a healthcare facility can induce bias in an electronic medical record-based study of community deprivation and chronic hepatitis C virus infection, in order to more accurately identify local risk factors and prevalence. METHODS: We created a catchment model that attempted to define the probability of selection into a retrospective cohort. Then using the inverse of this probability, we compared naïve unweighted and weighted models to demonstrate the impact of selection bias. RESULTS: ZIP code-level ecological plots of the cohort demonstrated that there was a pattern of the community deprivation, hepatitis C outcome, and distance to the health center (an intuitive proxy for being within catchments). The naïve multilevel analysis found that living in an area with greater deprivation resulted in 1.25 times greater odds of HCV (95% CI: 1.06, 1.48), whereas the weighted analysis found less certainty of this effect due to a selection bias. CONCLUSIONS: We observed that selection into the catchment area of the studied healthcare facility may bias the association of community deprivation and hepatitis C. This may be mitigated through inverse probability weighting.

Expanding care for patients infected with Hepatitis C through community partnership in Delaware.

BACKGROUND: Hepatitis C virus (HCV) is the most common blood-borne pathogen in the U.S., and Delaware has one of the highest sero-prevalence rates in the country. As a cause of significant morbidity and mortality, it is a public health priority to identify and link individuals with HCV to care. The demand of patients with HCV far exceeds the current availability of providers in Delaware that offer HCV management. . OBJECTIVE: To propose a framework for enabling non-specialist providers within Westside Family Healthcare to manage patients with HCV. METHODS: As a recipient of a Harrington Value Institute Community Partnership grant, the HIV Community Program of Christiana Care Health System (CCHS) started working together with the NE Wilmington pilot site of Westside in July 2018 to: 1) provide HCV education to Westside, 2) implement an HCV care model at Westside, and 3) conduct programmatic evaluation of this framework's effectiveness. Our goal is to improve Westside rates of HCV knowledge amongst patients and staff, as well as to improve the HCV care continuum, starting with universal HCV screening. RESULTS AND CONCLUSIONS: Results from the first year of collaboration will be available in August 2019.  Implementation of this partnership will enable future expansion and continuation of HCV management amongst Westside sites.

Modulating the strength of cadherin adhesion: evidence for a novel adhesion complex.

Adherens junctions and desmosomes are critical for embryogenesis and the integrity of adult tissues. To form these junctions, classical cadherins interact via alpha- and beta-catenin with the actin cytoskeleton, whereas desmosomal cadherins interact with the intermediate filament system. Here, we used a hormone-activated mutant N-cadherin expressed in fibroblasts to show the existence of a novel classical cadherin adhesion system. N-cadherin was fused at its C-terminus to a modified estrogen receptor ligand-binding domain (NcadER) that binds 4-hydroxytamoxifen (4OHT) and expressed in L cells, which lack an endogenous cadherin. Cells with the mutant cadherin (LNER cells) aggregated in the absence of 4OHT, but only in its presence formed tightly compacted aggregates like those formed by L cells expressing wild-type N-cadherin (LN cells). Compaction of LNER cells treated with 4OHT was accompanied by elevated levels of p120ctn in NcadER immunoprecipitates, compared to immunoprecipitates of non-treated cells, but without changes in alpha- and beta-catenin, or actin. Compaction induced by 4OHT was also accompanied by increased interaction of the NcadER with the cytoskeleton and increased vimentin organization. Vimentin co-immunoprecipitated with the NcadER/catenin complex, suggesting an interaction between cadherin and vimentin. The mechanism by which vimentin interacts with the cadherin appears to involve p120ctn as it co-immunoprecipitates and colocalizes with vimentin in the parent L cells, which lack a cadherin and alpha- and beta-catenins. Disrupting the actin cytoskeleton with cytochalasin B inhibited aggregation, whereas knocking down vimentin with specific siRNAs inhibited compaction. Based on our results we propose that a vimentin-based classical cadherin complex functions together with the actin-based complex to promote strong cell-cell adhesion in fibroblasts.