Objective Telemedicine practice has been shown to vary from clinical guidelines. Variations in practice patterns may be caused by disruptions in the continuity of care between traditional and telemedicine providers. This study compares virtual and in-person visits in Stanford’s ClickWell Care (CWC) – where patients see the same provider for both visit modalities. Methods Clinical data for two years of patient encounters at CWC from January 2015–2017 (5772 visits) were obtained through Stanford STRIDE. For the 20 most common visit categories, including 17 specific diagnoses, we compared the frequency of prescriptions, labs, procedures, and images ordered, as well as rates of repeat visits. Results For the 17 specific diagnoses, there are no differences in labs ordered. Two diagnoses show differences in images ordered, and four differences in prescriptions. Overall, there are more labs (0.16 virtual, 0.33 in-person p < 0.0001) and images ordered (0.07 virtual, 0.16 in-person, p < 0.0001) for in-person visits – due mainly to general medical exam visits. Repeat visits were more likely after in-person visits (19% virtual, 38% in-person, p < 0.0001), 10 out of 17 specific diagnoses showed differences in visit frequency between visit modalities. Visits for both anxiety (5.3x, p < 0.0001) and depression (5.1x, p < 0.0001) were much more frequent in the virtual setting. Conclusions Prescriptions, labs, and images ordered were similar between in-person and virtual visits for most diagnoses. Overall however, for in-person visits we find increased orders for labs and images, primarily from general medical exams. Finally, for anxiety and depression patients show clear preferences for virtual visits.

Our findings suggest that a "bricks-and-clicks" care model where telemedicine is supported by a brick-and-mortar location may be an effective way to leverage telemedicine to deliver primary care.

Background Though artificial intelligence (AI) has the potential to augment the patient-physician relationship in primary care, bias in intelligent health care systems has the potential to differentially impact vulnerable patient populations. Objective The purpose of this scoping review is to summarize the extent to which AI systems in primary care examine the inherent bias toward or against vulnerable populations and appraise how these systems have mitigated the impact of such biases during their development. Methods We will conduct a search update from an existing scoping review to identify studies on AI and primary care in the following databases: Medline-OVID, Embase, CINAHL, Cochrane Library, Web of Science, Scopus, IEEE Xplore, ACM Digital Library, MathSciNet, AAAI, and arXiv. Two screeners will independently review all abstracts, titles, and full-text articles. The team will extract data using a structured data extraction form and synthesize the results in accordance with PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. Results This review will provide an assessment of the current state of health care equity within AI for primary care. Specifically, we will identify the degree to which vulnerable patients have been included, assess how bias is interpreted and documented, and understand the extent to which harmful biases are addressed. As of October 2020, the scoping review is in the title- and abstract-screening stage. The results are expected to be submitted for publication in fall 2021. Conclusions AI applications in primary care are becoming an increasingly common tool in health care delivery and in preventative care efforts for underserved populations. This scoping review would potentially show the extent to which studies on AI in primary care employ a health equity lens and take steps to mitigate bias. International Registered Report Identifier (IRRID) PRR1-10.2196/27799

Objective This study assesses whether neural networks trained on electronic health record (EHR) data can anticipate what individual clinical orders and existing institutional order set templates clinicians will use more accurately than existing decision support tools. Materials and Methods We process 57 624 patients worth of clinical event EHR data from 2008 to 2014. We train a feed-forward neural network (ClinicNet) and logistic regression applied to the traditional problem structure of predicting individual clinical items as well as our proposed workflow of predicting existing institutional order set template usage. Results ClinicNet predicts individual clinical orders (precision = 0.32, recall = 0.47) better than existing institutional order sets (precision = 0.15, recall = 0.46). The ClinicNet model predicts clinician usage of existing institutional order sets (avg. precision = 0.31) with higher average precision than a baseline of order set usage frequencies (avg. precision = 0.20) or a logistic regression model (avg. precision = 0.12). Discussion Machine learning methods can predict clinical decision-making patterns with greater accuracy and less manual effort than existing static order set templates. This can streamline existing clinical workflows, but may not fit if historical clinical ordering practices are incorrect. For this reason, manually authored content such as order set templates remain valuable for the purposeful design of care pathways. ClinicNet’s capability of predicting such personalized order set templates illustrates the potential of combining both top-down and bottom-up approaches to delivering clinical decision support content. Conclusion ClinicNet illustrates the capability for machine learning methods applied to the EHR to anticipate both individual clinical orders and existing order set templates, which has the potential to improve upon current standards of practice in clinical order entry.

BACKGROUND Though artificial intelligence (AI) has potential to augment the patient-physician relationship in primary care, bias in intelligent healthcare systems has the potential to differentially impact vulnerable patient populations. OBJECTIVE The purpose of this scoping review is to summarize the extent to which AI systems in primary care examine the inherent bias towards or against vulnerable populations and appraise how these systems have mitigated the impact of such biases during their development. METHODS We will conduct a search update from an existing scoping review to identify AI and primary care articles in the following databases: Medline-OVID,Embase,CINAHL, Cochrane Library, Web of Science, Scopus, IEEE Xplore, ACM Digital Library, MathSciNet, AAAI, and arXiv. Two screeners will independently review all abstracts, titles and full-texts. The team will extract data using structured data extraction form and synthesize the results according to PRISMA-Scr guidelines. RESULTS This review will provide an assessment of the current state of healthcare equity within AI for primary care. Specifically, we will identify the degree to which vulnerable patients have been included, assess how bias is interpreted and documented, and understand the extent harmful biases are addressed. As of October 2020, the scoping review is in the title and abstract screening stage. The results are expected to be submitted for publication in fall of 2021. CONCLUSIONS AI applications in primary care are becoming an increasingly common tool in health care delivery, including in preventative care efforts for underserved populations. This scoping review aims to understand to what extent AI-primary care studies employ a health equity lens and take steps to mitigate bias.