In-basket Messages Research Articles

From Headache to Handled: Advanced In-Basket Management System in Primary Care Clinics Reduces Provider Workload Burden and Self-Reported Burnout.

The electronic health record (EHR) has been associated with provider burnout, exacerbated by increasing In-Basket burden. We sought to study the impact of implementing a team-based approach to In-Basket management on a series of primary care ambulatory sites. We performed a workflow analysis of the transition to the Advanced In-Basket Management (AIM) nurse team triage for six family medicine clinic locations in a large health system. We abstracted and analyzed associated provider workflow metrics from our EHR. We conducted a postintervention provider survey on satisfaction with the AIM project and provider burnout. The AIM project was implemented in six family medicine clinics after provider townhalls and workgroup development. A nurse team curriculum was created using the principles of "maturing the message" before sending it to a provider and "only handle it once" to improve response efficiency. Provider workload metrics abstracted from the EHR demonstrated 12.2 fewer In-Basket messages per provider per day (p < 0.05), 6.3 fewer minutes per provider per day worked outside scheduled hours (p < 0.05), 3.5 fewer minutes spent in the In-Basket per provider per day (p < 0.05), but 13.7 more seconds spent per completed message per provider (p = 0.017), likely attributable to increased message complexity. Sixty-four percent of providers reported no burnout symptoms in a postintervention survey, 56% agreed that the AIM project reduced their burnout, and approximately 70% of providers agreed that the AIM project was acceptable and appropriate for their clinic. The AIM project demonstrates team-based nurse In-Basket triage is possible to implement across multiple primary care sites, is an acceptable intervention for providers, can reduce provider workload burden and self-reported provider burnout.

The “Inboxologist” — A Novel Approach to In-Basket Management in Primary Care

Primary care providers are spending more time responding to electronic patient requests and completing other tasks in their digital in-basket that do not require their level of expertise. This contributes to job dissatisfaction and burnout. Last year, at two large primary care practice sites, Corewell Health West implemented a pilot program that introduced a new in-basket triage system involving a registered nurse completing initial triage and an advanced practice provider serving as an “inboxologist.” Under this model, primary care providers handled 41% fewer in-basket messages than before the pilot and spent an average of 47% less time on the messages they did receive. Self-reported time spent on the in-basket during office hours and after office hours decreased by an average of 37% and 32%, respectively. The number of messages that were resolved by the nonphysician staff member who initially triaged the message increased by 14.2% from pre-pilot to the end of the pilot period. Time to resolution of patient-generated messages decreased by 93%, with subsequent improvement in patient satisfaction scores. Given the favorable reception by physicians, patients, and staff, Corewell plans to expand the inboxologist program to all of its primary care sites and to selected specialty practices.

Assessing Artificial Intelligence-Generated Responses to Urology Patient In-Basket Messages.

Electronic patient messaging utilization has increased in recent years and has been associated with physician burnout. ChatGPT is a language model that has shown the ability to generate near-human level text responses. This study evaluated the quality of ChatGPT responses to real-world urology patient messages. One hundred electronic patient messages were collected from a practicing urologist's inbox and categorized based on the question content. Individual responses were generated by entering each message into ChatGPT. The questions and responses were independently evaluated by 5 urologists and graded on a 5-point Likert scale. Questions were graded based on difficulty, and responses were graded based on accuracy, completeness, harmfulness, helpfulness, and intelligibleness. Whether or not the response could be sent to a patient was also assessed. Overall, 47% of responses were deemed acceptable to send to patients. ChatGPT performed better on easy questions with 56% of responses to easy questions being acceptable to send as compared to 34% of difficult questions (P = .03). Responses to easy questions were more accurate, complete, helpful, and intelligible than responses to difficult questions. There was no difference in response quality based on question content. ChatGPT generated acceptable responses to nearly 50% of patient messages with better performance for easy questions compared to difficult questions. Use of ChatGPT to help respond to patient messages can help to decrease the time burden for the care team and improve wellness. Artificial intelligence performance will likely continue to improve with advances in generative artificial intelligence technology.

Prompt engineering on leveraging large language models in generating response to InBasket messages.

Large Language Models (LLMs) have been proposed as a solution to address high volumes of Patient Medical Advice Requests (PMARs). This study addresses whether LLMs can generate high quality draft responses to PMARs that satisfies both patients and clinicians with prompt engineering. We designed a novel human-involved iterative processes to train and validate prompts to LLM in creating appropriate responses to PMARs. GPT-4 was used to generate response to the messages. We updated the prompts, and evaluated both clinician and patient acceptance of LLM-generated draft responses at each iteration, and tested the optimized prompt on independent validation data sets. The optimized prompt was implemented in the electronic health record production environment and tested by 69 primary care clinicians. After 3 iterations of prompt engineering, physician acceptance of draft suitability increased from 62% to 84% (P <.001) in the validation dataset (N = 200), and 74% of drafts in the test dataset were rated as "helpful." Patients also noted significantly increased favorability of message tone (78%) and overall quality (80%) for the optimized prompt compared to the original prompt in the training dataset, patients were unable to differentiate human and LLM-generated draft PMAR responses for 76% of the messages, in contrast to the earlier preference for human-generated responses. Majority (72%) of clinicians believed it can reduce cognitive load in dealing with InBasket messages. Informed by clinician and patient feedback synergistically, tuning in LLM prompt alone can be effective in creating clinically relevant and useful draft responses to PMARs.

Large Language Model–Based Responses to Patients’ In-Basket Messages

Virtual patient-physician communications have increased since 2020 and negatively impacted primary care physician (PCP) well-being. Generative artificial intelligence (GenAI) drafts of patient messages could potentially reduce health care professional (HCP) workload and improve communication quality, but only if the drafts are considered useful. To assess PCPs' perceptions of GenAI drafts and to examine linguistic characteristics associated with equity and perceived empathy. This cross-sectional quality improvement study tested the hypothesis that PCPs' ratings of GenAI drafts (created using the electronic health record [EHR] standard prompts) would be equivalent to HCP-generated responses on 3 dimensions. The study was conducted at NYU Langone Health using private patient-HCP communications at 3 internal medicine practices piloting GenAI. Randomly assigned patient messages coupled with either an HCP message or the draft GenAI response. PCPs rated responses' information content quality (eg, relevance), using a Likert scale, communication quality (eg, verbosity), using a Likert scale, and whether they would use the draft or start anew (usable vs unusable). Branching logic further probed for empathy, personalization, and professionalism of responses. Computational linguistics methods assessed content differences in HCP vs GenAI responses, focusing on equity and empathy. A total of 16 PCPs (8 [50.0%] female) reviewed 344 messages (175 GenAI drafted; 169 HCP drafted). Both GenAI and HCP responses were rated favorably. GenAI responses were rated higher for communication style than HCP responses (mean [SD], 3.70 [1.15] vs 3.38 [1.20]; P = .01, U = 12 568.5) but were similar to HCPs on information content (mean [SD], 3.53 [1.26] vs 3.41 [1.27]; P = .37; U = 13 981.0) and usable draft proportion (mean [SD], 0.69 [0.48] vs 0.65 [0.47], P = .49, t = -0.6842). Usable GenAI responses were considered more empathetic than usable HCP responses (32 of 86 [37.2%] vs 13 of 79 [16.5%]; difference, 125.5%), possibly attributable to more subjective (mean [SD], 0.54 [0.16] vs 0.31 [0.23]; P < .001; difference, 74.2%) and positive (mean [SD] polarity, 0.21 [0.14] vs 0.13 [0.25]; P = .02; difference, 61.5%) language; they were also numerically longer (mean [SD] word count, 90.5 [32.0] vs 65.4 [62.6]; difference, 38.4%), but the difference was not statistically significant (P = .07) and more linguistically complex (mean [SD] score, 125.2 [47.8] vs 95.4 [58.8]; P = .002; difference, 31.2%). In this cross-sectional study of PCP perceptions of an EHR-integrated GenAI chatbot, GenAI was found to communicate information better and with more empathy than HCPs, highlighting its potential to enhance patient-HCP communication. However, GenAI drafts were less readable than HCPs', a significant concern for patients with low health or English literacy.

Maximizing efficiency and patient outcomes to meet the demands of a rapidly growing cancer institute.

e23203 Background: In 2023, the US is projected to have 1.96 million new cancer diagnoses with a 33% decline in cancer-related deaths. The prevalence of cancer is increasing at a rate that is difficult for oncology practices across the country to manage. Improvement processes must be evaluated to address these demands and create a practice with a foundation for long-term growth, scalability, and structure. The main target of this research was to standardize practice procedures to improve communication leading to reduced variation in order entry practices that resulted in confusion for staff, interruptions for providers, and chair time delays for patients. Methods: A QI project was undertaken at a rapidly expanding community oncology practice to meet patient volume demands. Three separate areas were identified to create standardization of procedures to improve care including referral guidelines, infusion center orders, and handoffs on patient care. New patient referral guidelines were created to determine if patients met criteria for urgent referrals or needed a referral at all. In addition, infusion center guidelines were standardized to require providers to sign all orders prior to an infusion clinic appointment time. The last area addressed was creating a means of safe and effective handoff between staff members at the end of a patient appointment through required utilization of the "Wrap-Up" section in Epic. Data was collected both before and after implementation of routine practice standards in all three areas mentioned. Utilizing the data gathered over a 6-month period, comparisons were made to identify if the changes had any measurable benefit. Results: Standardization of practice procedures had a significant positive impact. Through standardization of referral guidelines, 500 patients being referred to as urgent were determined not to meet criteria for an urgent evaluation in addition to over 300 patients that did not even need to be seen. By identifying these unnecessary referrals, patients that needed to be seen could be seen sooner. In addition, improved infusion center guidelines requiring providers to sign orders prior to appointment times increased infusion volumes that had a direct margin improvement upwards of $3 million dollars over a 6-month period. The last area addressed was streamlining patient handoff that directly correlated to an average reduction of 100 monthly in-basket messages across all providers. Conclusions: By standardizing daily practices across multiple areas within the practice, there were direct measurable benefits related to each intervention. These were simple, manageable changes that most practices could easily implement to increase productivity, improve patient care, and decrease the work burden on the staff. Ultimately, these improvements led to increased patient and employee satisfaction with decreased staff turnover as observed in this study.

A Closed-Loop Digital Health Tool to Improve Depression Care in Multiple Sclerosis: Iterative Design and Cross-Sectional Pilot Randomized Controlled Trial and its Impact on Depression Care.

People living with multiple sclerosis (MS) face a higher likelihood of being diagnosed with a depressive disorder than the general population. Although many low-cost screening tools and evidence-based interventions exist, depression in people living with MS is underreported, underascertained by clinicians, and undertreated. This study aims to design a closed-loop tool to improve depression care for these patients. It would support regular depression screening, tie into the point of care, and support shared decision-making and comprehensive follow-up. After an initial development phase, this study involved a proof-of-concept pilot randomized controlled trial (RCT) validation phase and a detailed human-centered design (HCD) phase. During the initial development phase, the technological infrastructure of a clinician-facing point-of-care clinical dashboard for MS management (BRIDGE) was leveraged to incorporate features that would support depression screening and comprehensive care (Care Technology to Ascertain, Treat, and Engage the Community to Heal Depression in people living with MS [MS CATCH]). This linked a patient survey, in-basket messages, and a clinician dashboard. During the pilot RCT phase, a convenience sample of 50 adults with MS was recruited from a single MS center with 9-item Patient Health Questionnaire scores of 5-19 (mild to moderately severe depression). During the routine MS visit, their clinicians were either asked or not to use MS CATCH to review their scores and care outcomes were collected. During the HCD phase, the MS CATCH components were iteratively modified based on feedback from stakeholders: people living with MS, MS clinicians, and interprofessional experts. MS CATCH links 3 features designed to support mood reporting and ascertainment, comprehensive evidence-based management, and clinician and patient self-management behaviors likely to lead to sustained depression relief. In the pilot RCT (n=50 visits), visits in which the clinician was randomized to use MS CATCH had more notes documenting a discussion of depressive symptoms than those in which MS CATCH was not used (75% vs 34.6%; χ21=8.2; P=.004). During the HCD phase, 45 people living with MS, clinicians, and other experts participated in the design and refinement. The final testing round included 20 people living with MS and 10 clinicians including 5 not affiliated with our health system. Most scoring targets for likeability and usability, including perceived ease of use and perceived effectiveness, were met. Net Promoter Scale was 50 for patients and 40 for clinicians. Created with extensive stakeholder feedback, MS CATCH is a closed-loop system aimed to increase communication about depression between people living with MS and their clinicians, and ultimately improve depression care. The pilot findings showed evidence of enhanced communication. Stakeholders also advised on trial design features of a full year long Department of Defense-funded feasibility and efficacy trial, which is now underway. ClinicalTrials.gov NCT05865405; http://tinyurl.com/4zkvru9x.

Professional Fulfillment: A Pilot Study to Decrease Burnout in Clinical Academic Neurology Faculty

A pilot project, NeuroMessage, was implemented to improve professional fulfillment for full-time academic clinical neurologists through restructuring administrative non–patient-facing support. It was a targeted intervention to improve the number and quality of clinical messages received through an appropriate triage system. A messaging triage algorithm was developed to decrease physician time spent on in-basket messages. The administrative team and a research assistant were trained to manage nonclinical messages. Although the program did decrease total participant in-basket messages, the neurologist subjects did not find the intervention decreased time spent on responding to in-basket messages or improved physician job satisfaction. Improved triaging and additional administrative support for in-basket messages did not notably decrease time spent on managing in-basket messages. Further research may help understand what types of clinical support may be helpful to improve burnout. Clinical support to triage messages, so that only rare urgent messages get through to the physician, should be studied to see if aligning duties that are uniquely suited for a physician improves personal achievement and decreases burnout.

In-Basket Reduction: A Multiyear Pragmatic Approach to Lessen the Work Burden of Primary Care Physicians

Summary The in-basket of the primary care physician (PCP) has become unmanageable. A growing volume of messages from system notifications, communications from colleagues, and especially patient medical advice requests has resulted in several hours of such additional indirect care for physicians each day — and is a leading contributor to burnout. Atrius Health embraced this in-basket challenge in 2017, and the authors share their experience here. They determined that they could not effectively create efficiencies in practice for Atrius Health’s PCPs without tackling the in-basket. Atrius had used a team-based model of primary care for decades, but the in-basket processes did not reflect its care model of teamwork; in-basket messages defaulted to the PCP to manage and resolve. Feedback from PCPs, assessments of professional burnout, and measurement of PCP in-basket volume presented a clear need to address this issue. Atrius Health focused on reducing volume but considered performance in quality, safety, and patient experience in all work. The organization launched a multiyear initiative to reduce the burden of what should be a beneficial tool by applying the principles of eliminate, automate, delegate, and collaborate. By 2022, Atrius Health had reduced the in-basket volumes by category as follows: media manager (which refers to a folder in which scanned documents are filed), 98% reduction; carbon copy charts (which refers to notes from other physicians and contains a range of critical new diagnoses, prognoses, or treatment plans along with routine follow-up visits), 35% reduction; ED/hospital admissions/discharge/transfer messages, 100% reduction; prescription renewals, 50% reduction; laboratory results, 30% reduction; and patient medical advice requests, 40% reduction. Through a multipronged approach, the total daily in-basket volume for PCPs was reduced by approximately 25% from 100 messages in 2017 to 75 messages in 2022.

Association of Electronic Health Record Inbasket Message Characteristics With Physician Burnout

Physician burnout is an ongoing epidemic; electronic health record (EHR) use has been associated with burnout, and the burden of EHR inbasket messages has grown in the context of the COVID-19 pandemic. Understanding how EHR inbasket messages are associated with physician burnout may uncover new insights for intervention strategies. To evaluate associations between EHR inbasket message characteristics and physician burnout. Cross-sectional study in a single academic medical center involving physicians from multiple specialties. Data collection took place April to September 2020, and data were analyzed September to December 2020. Physicians responded to a survey including the validated Mini-Z 5-point burnout scale. Physician burnout according to the self-reported burnout scale. A sentiment analysis model was used to calculate sentiment scores for EHR inbasket messages extracted for participating physicians. Multivariable modeling was used to model risk of physician burnout using factors such as message characteristics, physician demographics, and clinical practice characteristics. Of 609 physicians who responded to the survey, 297 (48.8%) were women, 343 (56.3%) were White, 391 (64.2%) practiced in outpatient settings, and 428 (70.28%) had been in medical practice for 15 years or less. Half (307 [50.4%]) reported burnout (score of 3 or higher). A total of 1 453 245 inbasket messages were extracted, of which 630 828 (43.4%) were patient messages. Among negative messages, common words included medical conditions, expletives and/or profanity, and words related to violence. There were no significant associations between message characteristics (including sentiment scores) and burnout. Odds of burnout were significantly higher among Hispanic/Latino physicians (odds ratio [OR], 3.44; 95% CI, 1.18-10.61; P = .03) and women (OR, 1.60; 95% CI, 1.13-2.27; P = .01), and significantly lower among physicians in clinical practice for more than 15 years (OR, 0.46; 95% CI, 0.30-0.68; P < .001). In this cross-sectional study, message characteristics were not associated with physician burnout, but the presence of expletives and violent words represents an opportunity for improving patient engagement, EHR portal design, or filters. Natural language processing represents a novel approach to understanding potential associations between EHR inbasket messages and physician burnout and may also help inform quality improvement initiatives aimed at improving patient experience.

Abstract 10401: Development and Pilot Implementation of Two Targeted Approaches to Ambulatory Clinical Decision Support for Cardiologists to Optimize Medical Therapy for Heart Failure

Introduction: Shortfalls in prescribing of proven therapies for patients with heart failure and reduced ejection fraction (HFrEF), particularly for mineralocorticoid receptor antagonists (MRA), account for several thousand preventable deaths per year nationwide. Electronic clinical decision support (CDS) is a potential low-cost and scalable solution to improve prescribing, but optimal timing and format of CDS tools are unknown. Objective: We developed two CDS interventions to inform cardiologists of gaps in MRA therapy for patients with HFrEF and without contraindication to MRA therapy: 1) a best practice alert (BPA) that notifies cardiologists at the time of patient visit, and 2) an automated in-basket message that allows for review between visits. Methods: We designed CDS tools using an established CDS framework and findings from semi-structured interviews with cardiologists. We then pilot-tested both CDS tools at two cardiology practices for 12 weeks in a cross-over design, and further enhanced them based on additional semi-structured interviews. Results: Based on pilot testing (n=596 patients) and semi-structured interviews (n=11 cardiologists), in-basket messages were modified to reduce the number of patients listed, include future visits, and list date of next visit. The BPA was modified to improve noticeability, reduce extraneous information on guidelines, and include key information on contraindications. Final CDS interventions after incorporation of feedback are depicted in the Figure. Conclusions: Pilot testing and semi-structured interviews were instrumental in the development of CDS interventions and should be employed for other similar studies. We will compare the effectiveness of the BPA vs the in-basket message vs usual care on the primary outcome of MRA prescribing in The BETTER CARE-HF trial (Building Electronic Tools to Enhance and Reinforce CArdiovascular REcommendations for Heart Failure) (Clinicaltrials.gov NCT05275920)

Abstract 11295: Design and Rationale for the BETTER CARE-HF Trial: A Pragmatic Cluster-Randomized Controlled Trial Comparing Two Targeted Approaches to Ambulatory Clinical Decision Support for Cardiologists

Introduction: Heart failure with reduced ejection fraction (HFrEF) is a leading cause of morbidity and mortality. However, shortfalls in prescribing of proven therapies, particularly mineralocorticoid receptor antagonist (MRA) therapy, account for several thousand preventable deaths per year nationwide. Electronic clinical decision support (CDS) is a potential low-cost and scalable solution to improve prescribing of therapies. However, the optimal timing and format of CDS tools is unknown. We developed two CDS tools to inform cardiologists of gaps in MRA therapy for patients with HFrEF and without contraindication to MRA therapy: 1) a best practice alert (BPA) that notifies cardiologists at the time of patient visit, and 2) an automated in-basket message that allows for review between visits (Figure). Objective: The BETTER CARE-HF (Building Electronic Tools to Enhance and Reinforce CArdiovascular REcommendations for Heart Failure) trial aims to compare the effectiveness of a BPA vs. in-basket message vs. usual care on the primary outcome of MRA prescribing. Methods: This is a pragmatic, cluster-randomized, three-arm intervention trial. Patients are randomized as a cluster at the level of their cardiologist in a 1:1:1 fashion (180 cardiologists included). The primary outcome is MRA prescription at the end of the study period, and secondary outcomes include prescribing of other GDMT, time to prescription, and provider-level changes in prescribing rates. We will require a total of 1,503 patients (501 per arm) to detect 10% improvement in prescribing with 80% power, and estimate a trial duration of 6 months. Implications: To our knowledge, no study has directly compared the efficacy of a BPA vs. in-basket message. If effective, our findings can be rapidly disseminated to improve morbidity and mortality for patients with HFrEF, and can also inform the development of future CDS interventions for other disease states. Trial registration: NCT05275920

Assessing the impact of the COVID-19 pandemic on clinician ambulatory electronic health record use.

ObjectiveThe COVID-19 pandemic changed clinician electronic health record (EHR) work in a multitude of ways. To evaluate how, we measure ambulatory clinician EHR use in the United States throughout the COVID-19 pandemic.Materials and MethodsWe use EHR meta-data from ambulatory care clinicians in 366 health systems using the Epic EHR system in the United States from December 2019 to December 2020. We used descriptive statistics for clinician EHR use including active-use time across clinical activities, time after-hours, and messages received. Multivariable regression to evaluate total and after-hours EHR work adjusting for daily volume and organizational characteristics, and to evaluate the association between messages and EHR time.ResultsClinician time spent in the EHR per day dropped at the onset of the pandemic but had recovered to higher than prepandemic levels by July 2020. Time spent actively working in the EHR after-hours showed similar trends. These differences persisted in multivariable models. In-Basket messages received increased compared with prepandemic levels, with the largest increase coming from messages from patients, which increased to 157% of the prepandemic average. Each additional patient message was associated with a 2.32-min increase in EHR time per day (P < .001).DiscussionClinicians spent more total and after-hours time in the EHR in the latter half of 2020 compared with the prepandemic period. This was partially driven by increased time in Clinical Review and In-Basket messaging.ConclusionsReimbursement models and workflows for the post-COVID era should account for these demands on clinician time that occur outside the traditional visit.

Assessment of Electronic Health Record Use Between US and Non-US Health Systems

Understanding how the electronic health record (EHR) system changes clinician work, productivity, and well-being is critical. Little is known regarding global variation in patterns of use. To provide insights into which EHR activities clinicians spend their time doing, the EHR tools they use, the system messages they receive, and the amount of time they spend using the EHR after hours. This cross-sectional study analyzed the deidentified metadata of ambulatory care health systems in the US, Canada, Northern Europe, Western Europe, the Middle East, and Oceania from January 1, 2019, to August 31, 2019. All of these organizations used the EHR software from Epic Systems and represented most of Epic Systems's ambulatory customer base. The sample included all clinicians with scheduled patient appointments, such as physicians and advanced practice practitioners. Clinician EHR use was tracked by deidentified and aggregated metadata across a variety of clinical activities. Descriptive statistics for clinician EHR use included time spent on clinical activities, note documentation (as measured by the percentage of characters in the note generated by automated or manual data entry source), messages received, and time spent after hours. A total of 371 health systems were included in the sample, of which 348 (93.8%) were located in the US and 23 (6.2%) were located in other countries. US clinicians spent more time per day actively using the EHR compared with non-US clinicians (mean time, 90.2 minutes vs 59.1 minutes; P < .001). In addition, US clinicians vs non-US clinicians spent significantly more time performing 4 clinical activities: notes (40.7 minutes vs 30.7 minutes; P < .001), orders (19.5 minutes vs 8.75 minutes; P < .001), in-basket messages (12.5 minutes vs 4.80 minutes; P < .001), and clinical review (17.6 minutes vs 14.8 minutes; P = .01). Clinicians in the US composed more automated note text than their non-US counterparts (77.5% vs 60.8% of note text; P < .001) and received statistically significantly more messages per day (33.8 vs 12.8; P < .001). Furthermore, US clinicians used the EHR for a longer time after hours, logging in 26.5 minutes per day vs 19.5 minutes per day for non-US clinicians (P = .01). The median US clinician spent as much time actively using the EHR per day (90.1 minutes) as a non-US clinician in the 99th percentile of active EHR use time per day (90.7 minutes) in the sample. These results persisted after controlling for organizational characteristics, including structure, type, size, and daily patient volume. This study found that US clinicians compared with their non-US counterparts spent substantially more time actively using the EHR for a wide range of clinical activities or tasks. This finding suggests that US clinicians have a greater EHR burden that may be associated with nontechnical factors, which policy makers and health system leaders should consider when addressing clinician wellness.

An Implementation Science Framework to Develop a Clinical Decision Support Tool for Familial Hypercholesterolemia.

Electronic health record (EHR)-based clinical decision support (CDS) can address the low awareness and undertreatment of familial hypercholesterolemia (FH), a disorder associated with a markedly increased risk of coronary heart disease. We aimed to incorporate provider perspectives into the development and implementation of a CDS tool for FH. An implementation science framework and a user-centered design process were used to create a CDS tool for FH. Primary care physicians and specialist physicians participated in qualitative interviews, usability testing and an implementation survey. The CDS was configured in two formats—a best practice alert (BPA) and an in-basket message and subsequently deployed in the EHR in silent mode. The key themes that emerged from the analysis of interview transcripts included understanding and awareness of FH, clinical workflow, physician preferences and value of CDS tools, perspectives on patient needs and values and dissemination and implementation. Recommendations related to usability included preferred CDS format and placement, content, timing and frequency, and level of alert urgency/prioritization. In response to the survey, 84.6% of physicians agreed that the CDS would improve early FH diagnosis and 92.3% agreed that it would help them identify and manage FH patients. Physician feedback led to iterative CDS refinement. In summary, we developed a CDS tool for FH using an implementation science framework and physician feedback. Initial deployment revealed a significant burden of FH and the potential for the CDS tool to have a large impact.

SUPPORT-AF II: Supporting Use of Anticoagulants Through Provider Profiling of Oral Anticoagulant Therapy for Atrial Fibrillation

Previous provider-directed electronic messaging interventions have not by themselves improved anticoagulation use in patients with atrial fibrillation. Direct engagement with providers using academic detailing coupled with electronic messaging may overcome the limitations of the prior interventions. We randomized outpatient providers affiliated with our health system in a 2.5:1 ratio to our electronic profiling/messaging combined with academic detailing intervention. In the intervention, we emailed providers monthly reports of their anticoagulation percentage relative to peers for atrial fibrillation patients with elevated stroke risk (CHA2DS2-VASc ≥2). We also sent electronic medical record-based messages shortly before an appointment with an anticoagulation-eligible but untreated atrial fibrillation patient. Providers had the option to send responses with explanations for prescribing decisions. We also offered to meet with intervention providers using an academic detailing approach developed based on knowledge gaps discussed in provider focus groups. To assess feasibility, we tracked provider review of our messages. To assess effectiveness, we measured the change in anticoagulation for patients of intervention providers relative to controls. We identified 85 intervention and 34 control providers taking care of 3591 and 1908 patients, respectively; 33 intervention providers participated in academic detailing. More than 80% of intervention providers read our emails, and 98% of the time a provider reviewed our in-basket messages. Replies to messages identified patient refusal as the most common reason for patients not being on anticoagulation (11.2%). For the group of patients not on anticoagulation at baseline assigned to an intervention versus control provider, the adjusted percent increase in the use of anticoagulation over 6 months was 5.2% versus 7.4%, respectively (P=0.21). Our electronic messaging and academic detailing intervention was feasible but did not increase anticoagulation use. Patient-directed interventions or provider interventions targeting patients declining anticoagulation may be necessary to raise the rate of anticoagulation. Clinical Trial Registration URL: http://www.clinicaltrials.gov. Unique identifier: NCT03583008.

Re: Physicians' Well-Being Linked to In-Basket Messages Generated by Algorithms in Electronic Health Records.

Re: Physicians' Well-Being Linked to In-Basket Messages Generated by Algorithms in Electronic Health Records.

Physicians' Well-Being Linked To In-Basket Messages Generated By Algorithms In Electronic Health Records.

Despite concerns about physicians' workload associated with electronic health records (EHRs), little attention has been paid to the relationship between physicians' well-being and the in-basket messages physicians receive-specifically, their volume and sources. Analyses of EHR work performed by physicians in a multispecialty practice found that in-basket messages generated by the EHR system accounted for almost half (114) of the 243 weekly in-basket messages received per physician, on average-far exceeding the numbers received from their colleagues (53) and patients (30). In a survey, 36percent of the physicians reported burnout symptoms, and 29percent intended to reduce their clinical work time in the upcoming year. Receiving more than the average number of system-generated in-basket messages was associated with 40percent higher probability of burnout and 38percent higher probability of intending to reduce clinical work time. Physicians' perceptions of a positive work environment were associated with lower odds of burnout and intention to reduce clinical work time and with greater satisfaction with life. Female physicians had a higher risk of burnout and lower satisfaction with life, compared to males. Meaningful redesign of EHR in-basket workflow and a wellness-enhancing work environment are necessary to effectively improve physicians' well-being.

Understanding CancelRx: Results of End-to-End Functional Testing, Proactive Risk Assessment, and Pilot Implementation.

CancelRx allows prescribers to send electronic cancellation messages to pharmacies when medications are discontinued. Little is known about its functionality and impact on clinical workflows. To understand CancelRx functionality, its potential impact on workflows and medication safety risks, and to develop mitigating strategies for risks introduced by implementation. We conducted direct observations and semi-structured interviews to develop CancelRx use cases and assessed CancelRx in an end-to-end test environment, proactive risk assessment, and pilot implementation from April 16 to July 15, 2018. E-cancellations were sent upon discontinuation of e-prescriptions written within the electronic health record (EHR), but not other medications (e.g., printed prescriptions) and could be initiated by nonprescribers. In our proactive risk assessment, CancelRx implementation eliminated five of seven failure modes in outpatient prescribing to Johns Hopkins pharmacies, but introduced new risks, including (1) failure to act if an e-cancellation was not sent or was unsuccessful; (2) failure to cancel all prescriptions for a medication; (3) errors in manual matching; and (4) erroneous medication cancellations. We identified potential mitigation strategies for these risks. During pilot implementation, 92.4% (428/463) of e-cancellations had confirmed approval by the receiving pharmacy, while 4.5% (21/463) were denied, and 3.0% (14/463) had no e-cancellation response. Among e-cancellations received by the pilot pharmacy, 1.7% (7/408) required manual matching by pharmacy staff. Based on performance in testing, 73.4% (340/463) of completed e-cancellations would be expected to generate an in-basket message, including 21 (6.2%) denials and 319/340 (93.8%) approvals with a note from the pharmacy. CancelRx is an important functionality with the potential to decrease adverse events due to medication errors. However, changes in implementation in our EHR and pharmacy software and enhancements in the CancelRx standard are needed to maximize safety and usability. Further studies are needed to evaluate the impact of e-cancellation on medication safety.

Use of Electronic Health Record Access and Audit Logs to Identify Physician Actions Following Noninterruptive Alert Opening: Descriptive Study.

BackgroundElectronic health record (EHR) access and audit logs record behaviors of providers as they navigate the EHR. These data can be used to better understand provider responses to EHR–based clinical decision support (CDS), shedding light on whether and why CDS is effective.ObjectiveThis study aimed to determine the feasibility of using EHR access and audit logs to track primary care physicians’ (PCPs’) opening of and response to noninterruptive alerts delivered to EHR InBaskets.MethodsWe conducted a descriptive study to assess the use of EHR log data to track provider behavior. We analyzed data recorded following opening of 799 noninterruptive alerts sent to 75 PCPs’ InBaskets through a prior randomized controlled trial. Three types of alerts highlighted new medication concerns for older patients’ posthospital discharge: information only (n=593), medication recommendations (n=37), and test recommendations (n=169). We sought log data to identify the person opening the alert and the timing and type of PCPs’ follow-up EHR actions (immediate vs by the end of the following day). We performed multivariate analyses examining associations between alert type, patient characteristics, provider characteristics, and contextual factors and likelihood of immediate or subsequent PCP action (general, medication-specific, or laboratory-specific actions). We describe challenges and strategies for log data use.ResultsWe successfully identified the required data in EHR access and audit logs. More than three-quarters of alerts (78.5%, 627/799) were opened by the PCP to whom they were directed, allowing us to assess immediate PCP action; of these, 208 alerts were followed by immediate action. Expanding on our analyses to include alerts opened by staff or covering physicians, we found that an additional 330 of the 799 alerts demonstrated PCP action by the end of the following day. The remaining 261 alerts showed no PCP action. Compared to information-only alerts, the odds ratio (OR) of immediate action was 4.03 (95% CI 1.67-9.72) for medication-recommendation and 2.14 (95% CI 1.38-3.32) for test-recommendation alerts. Compared to information-only alerts, ORs of medication-specific action by end of the following day were significantly greater for medication recommendations (5.59; 95% CI 2.42-12.94) and test recommendations (1.71; 95% CI 1.09-2.68). We found a similar pattern for OR of laboratory-specific action. We encountered 2 main challenges: (1) Capturing a historical snapshot of EHR status (number of InBasket messages at time of alert delivery) required incorporation of data generated many months prior with longitudinal follow-up. (2) Accurately interpreting data elements required iterative work by a physician/data manager team taking action within the EHR and then examining audit logs to identify corresponding documentation.ConclusionsEHR log data could inform future efforts and provide valuable information during development and refinement of CDS interventions. To address challenges, use of these data should be planned before implementing an EHR–based study.