An evaluation of clinical decision support tools for Patient Health Questionnaire-9 administration

Introduction Many health care institutions are working to improve depression screening and management with the use of the Patient Health Questionnaire 9 (PHQ-9). Clinical decision support (CDS) within the EHR is one strategy, but little is known about effective approaches to design or implement such CDS. The purpose of this study is to compare implementation outcomes of two versions of a CDS tool to improve PHQ-9 administration for patients with depression. Methods This was a retrospective, observational study comparing two versions of a CDS. Version 1 interrupted clinician workflow, and version 2 did not interrupt workflow. Outcomes of interest included reach, adoption, and effectiveness. PHQ-9 administration was determined by chart review. Chi-square tests were used to evaluate associations between PHQ-9 administration with versions 1 and 2. Results Version 1 resulted in PHQ-9 administration 77 times (15.3% of 504 unique encounters) compared with 49 times (9.8% of 502 unique encounters) with version 2 (P = .011). Discussion An interruptive CDS tool may be more effective at increasing PHQ-9 administration, but a noninterruptive CDS tool may be preferred to minimize alert fatigue despite a decrease in effectiveness.

A framework synthesis of stakeholder perspectives on clinical decision support tools to inform clinical AI implementation: a qualitative evidence synthesis protocol (Preprint)

BACKGROUND Quantitative systematic reviews have identified clinical artificial intelligence (AI) enabled tools with adequate performance for real-world implementation. To our knowledge, no published report or protocol synthesizes the full breadth of stakeholder perspectives. The absence of such a rigorous foundation perpetuates the ‘AI chasm’ which continues to delay patient benefit. OBJECTIVE To synthesize stakeholder perspectives of computerized clinical decision support tools (CCDST) in any healthcare setting. Synthesized findings will inform future research and the implementation of AI into healthcare services. METHODS The search strategy will use MEDLINE (Ovid), Scopus, CINAHL (EBSCO), ACM Digital Library and Science Citation Index (Web of Science). Following deduplication, title, abstract and full text screening will be performed by two independent reviewers with a third topic expert arbitrating. The quality of included studies will be appraised to support interpretation. Best-fit framework synthesis will be performed, with line-by-line coding completed by two independent reviewers. Where appropriate, these findings will be assigned to one of 22 a-priori themes defined by the Non-Adoption, Abandonment, Scale-Up, Spread and Sustainability (NASSS) framework. New domains will be inductively generated for outlying findings. The placement of findings within themes will be reviewed iteratively by a study advisory group including patient and lay representatives. RESULTS Study registration was obtained from PROSPERO (ID 248025) in May 2021. Final searches were executed in April and screening is ongoing at the time of writing. Full text data analysis is due to be completed in October 2021. We anticipate that the study will be submitted for open-access publication in late 2021 . CONCLUSIONS This paper describes the protocol for a qualitative evidence synthesis aiming to define barriers and facilitators to the implementation of CCDSTs from all relevant stakeholders. The results of this study are intended to expedite the delivery of patient benefit from AI enabled clinical tools. CLINICALTRIAL PROSPERO ID 248025

Does real-time artificial intelligence-based visual pathology enhancement of three-dimensional optical coherence tomography scans optimise treatment decision in patients with nAMD? Rationale and design of the RAZORBILL study

Background/rationaleArtificial intelligence (AI)-based clinical decision support tools, being developed across multiple fields in medicine, need to be evaluated for their impact on the treatment and outcomes of patients as well as optimisation of the clinical workflow. The RAZORBILL study will investigate the impact of advanced AI segmentation algorithms on the disease activity assessment in patients with neovascular age-related macular degeneration (nAMD) by enriching three-dimensional (3D) retinal optical coherence tomography (OCT) scans with automated fluid and layer quantification measurements.MethodsRAZORBILL is an observational, multicentre, multinational, open-label study, comprising two phases: (a) clinical data collection (phase I): an observational study design, which enforces neither strict visit schedule nor mandated treatment regimen was chosen as an appropriate design to collect data in a real-world clinical setting to enable evaluation in phase II and (b) OCT enrichment analysis (phase II): de-identified 3D OCT scans will be evaluated for disease activity. Within this evaluation, investigators will review the scans once enriched with segmentation results (i.e., highlighted and quantified pathological fluid volumes) and once in its original (i.e., non-enriched) state. This review will be performed using an integrated crossover design, where investigators are used as their own controls allowing the analysis to account for differences in expertise and individual disease activity definitions.ConclusionsIn order to apply novel AI tools to routine clinical care, their benefit as well as operational feasibility need to be carefully investigated. RAZORBILL will inform on the value of AI-based clinical decision support tools. It will clarify if these can be implemented in clinical treatment of patients with nAMD and whether it allows for optimisation of individualised treatment in routine clinical care.

Injuries Associated With Subsequent Child Maltreatment Diagnosis: By Age, Race, Gender, and Medicaid Status

This study examined injuries that may precede a child maltreatment (CM) diagnosis, by age, race/ethnicity, gender, and Medicaid status using a retrospective case–control design among child members of a large integrated healthcare system ( N = 9152 participants, n = 4576 case). Injury categories based on diagnosis codes from medical visits were bruising, fractures, lacerations, head injury, burns, falls, and unspecified injury. Results showed that all injury categories were significant predictors of a subsequent CM diagnosis, but only for children < 3 years old. Specifically, fracture and head injury were the highest risk for a subsequent CM diagnosis. All injury types were significant predictors of maltreatment diagnosis for Hispanic children < 3 years, which was not the case for the other race/ethnicities. Overall, these findings suggest that all types of injury within these specific categories should have a more thorough assessment for possible abuse for children under 3 years. This work can inform the development of clinical decision support tools to aid healthcare providers in detecting abusive injuries.

Evaluating Adoption, Impact, and Factors Driving Adoption for TREWS, a Machine Learning-Based Sepsis Alerting System

Machine learning-based clinical decision support tools for sepsis create opportunities to identify at-risk patients and initiate treatments earlier, an important step in improving sepsis outcomes. Increasing use of such systems means quantifying and understanding provider adoption is critical. Using real-time provider interactions with a sepsis early detection tool (Targeted Real-time Early Warning System) deployed at five hospitals over a two-year period (469,419 screened patient encounters, 9,805 (2.1%) of which were retrospectively identified as having sepsis), we found high adoption rates (89% of alerts were evaluated by a physician or advanced practice provider) and an association between use of the tool and earlier treatment of sepsis patients (1.85 (95% CI: 1.66 - 2.00) hour reduction in median time to first antibiotics order). Further, we found that provider-related factors had the strongest association with alert adoption and that case complexity and atypical presentation were associated with dismissal of alerts on sepsis patients. Beyond improving the performance of the system, efforts to improve adoption should focus on provider knowledge, experience, and perceptions of the system.