Bridging the gap with Pando

The Pando app is UK based and part of the NHS Clinical Communication Procurement Framework, which is designed to provide continuity of care with virtual patient management (https://www.bad.org.uk/healthcare-professionals/covid-19/re mote-dermatology-guidance), and drive tech-enabled connectivity across the National Health Service (NHS). This has also been used in the British Army to help defence medical staff connect with and seek advice from their colleagues in the UK while in the field (www.hellopando.com). Lack of on-site medical illustration, the COVID-19 pandemic and plastic surgeons operating in a NHS-funded private setting with no access to Picture Archiving and Communication System (PACS) in our Trust prompted use of the Pando app to capture prebiopsy pictures, avoid wrong-site surgery and improve interdepartmental communication. We present our multidisciplinary quality improvement project, involving dermatology and plastic surgery, evaluating the use of the Pando app from September to December 2020, mostly from 2-week-wait skin cancer clinics. All dermatology and plastic surgery colleagues downloaded the Pando app to their mobile phones and created a group entitled 'Dermatology/Plastics' to share their patient photos with identity labels. Patient photos could also be emailed to the clinicians' NHS email addresses - all done with patient consent. We evaluated our project with pre-and post-Pando feedback questionnaires. In the pre-Pando questionnaires, the majority of 14 colleagues involved were concerned with the varying quality of photos emailed by patients, the time lag in photos being uploaded to PACS and any likelihood of compromising patient safety. With post-Pando questionnaires, the majority found the app to be user-friendly, that the photographs taken were of superior quality, that there were no reported concerns with patient consent and they preferred using the app to the previous pathway. Comments suggested the Pando app to be invaluable for site recognition in patients with cognitive impairment, multiple lesions, difficult-to-see areas, medicolegal, educational and audit purposes, and local cancer multidisciplinary discussions. The drawbacks were the lack of seamless connection between the app and PACS, the inability to search for pictures in the app with patient identification and lack of access to previously shared pictures for new users. Despite some limitations, the Pando app has immensely improved patient safety and proved to be invaluable for our joint dermatology and plastic surgery interactions. However, there is an unmet need for a system with the ability to instantly transfer pictures to PACS and patient electronic records, to improve things further.

Optimizing Precision Medicine Cancer Clinical Trials

With the advent of precision medicine, getting the right treatment for the right patient at the right time has illuminated a variety of challenges and opportunities for innovation in trial design and conduct. Although there is no onesize- fits-all approach to precision medicine, a number of approaches, particularly basket and umbrella trial platforms that permit simultaneous evaluation of multiple treatments in multiple patient cohorts, have evolved to improve trial efficiency. The novel coronavirus pandemic has illuminated the need for, and feasibility of, conducting trials with fewer requirements, greater flexibility, and more decentralization. Through the lens of precision medicine cancer clinical trials, successes and challenges will be discussed to share practical solutions of how to improve evidence generation in the era of precision medicine. Through discussion of precision medicine cancer clinical trials within the United States, this session will provide an overview of how best to optimize these clinical trials. This session comprises the following three main topic areas: matching treatments to patients, including the use of novel patient identification strategies, genomic matching rules, molecular tumor boards (MTBs), decision-support tools, incorporating precision medicine trials into a research portfolio, and how to overcome challenges;accelerating evidence development, including the use of adaptive trial designs, cohort management strategies and data sharing plans;and improving diversity of trial participants and increasing generalizability of study results through expanded eligibility criteria and site selection strategies. This topic will be explored through 90 min of invited talks and a panel discussion with Q&A. Moderator and session chair, Richard L Schilsky, MD, FACP, FSCT, FASCO, will introduce the session and speakers and provide an introduction on the basics of precision oncology. Timothy Cannon, MD, a medical oncologist and clinical trial researcher, will present a case study of two patients with the same alteration and discuss how the care for each differed based on access to a precision medicine cancer clinical trial. Dr. Cannon will also discuss how to implement precision medicine trials within a research portfolio and how to identify patients at a site. Christine Walko, PharmD, BCOP, FCCP, a pharmacist and researcher, will then discuss the basics of matching therapies to genomics, how to use decision support tools, the role of an MTB, and identifying therapeutic options for patients. Edward S Kim, MD, MBA, FACP, FASCO, a medical oncologist, will talk about MTBs from a clinician perspective and precision medicine cancer clinical trials from a community practice perspective. Dr. Kim will also speak about how to extend the research team to create an adequate community research portfolio and about using broader eligibility criteria that might facilitate enrollment of diverse populations. Jane Perlmutter, PhD, MBA, FASCO, a cancer survivor and patient advocate, will discuss what pragmatic trials are, how they increase the opportunity for diversity and generalizability, and patient perspectives regarding pragmatic and precision cancer clinical trials. Susan Halabi, PhD, FASCO, FSCT, a researcher and professor of biostatistics and bioinformatics, will talk about the best ways to design an adaptive trial, especially its role in real-world settings, how these types of designs can be used for efficient signal finding in rare populations, and how these trials can create opportunities for data sharing and collaboration. Pam K Mangat, MS, a research scientist and the Director of Clinical Research for American Society of Clinical Oncology's (ASCO) Targeted Agent and Profiling Utilization Registry (TAPUR) Study, will discuss the operations behind a precision medicine study, advantages and disadvantages of a pragmatic trial, cohort closing and collapsing rules to help with management of large numbers of small cohorts, and contributing knowledge to other trials. The session will conclude with a panel di cussion moderated by session chair, Richard L Schilsky, MD, FACP, FSCT, FASCO, for approximately 10 min, with 5 min for a Q&A session.

Optimising COVID-19 Vaccination and Reducing Health Inequalities in Patients on Renal Replacement Therapy

Background: COVID-19 vaccine hesitancy has been asscociated with social deprivation and selected minority ethnic groups who are over-represented in the Renal Replacement Therapy (RRT) population. We designed a COVID-19 vaccination programme for our RRT population with the aim to increase vaccination uptake and decrease health inequalities. Method(s): Key interventions included addressing vaccine hesitancy by deploying the respective clinical teams as trusted messengers, prompt eligible patient identification and notification, deployment of resources to optimise vaccine administration in a manner convenient to patients and timely collection and analysis of local safety and efficacy data. First COVID-19 vaccination uptake data in relation to ethnicity and social deprivation, measured by the multiple deprivation index, in our RRT population were analysed and compared with uptake data in the regional total adult clinically extremely vulnerable (CEV) population in Greater Manchester (GM). Univariate logistic regression analysis was used to explore the factors associated with not receiving a vaccine. Result(s): Out of 1156 RRT patients included in this analysis (Table) 96.7% received the first dose vaccination compared to 93% in the cohort of CEV patients in the GM. Age, sex, ethnicity and index of multiple deprivation were not associated with first dose vaccine uptake. Vaccine uptake in Asian and Black RRT patients was 94.9% and 92.3% respectively compared to 93% and 76.2% for the same ethic groups in the reference CEV GM. Vaccine uptake was 96.1% of RRT patients in lowest quartile of multiple deprivation index compared to 90.5% in the GM reference population. Conclusion(s): Bespoke COVID-19 vaccination programme based on local clinical teams as trusted messengers can address vaccine hesitancy and reduce health inequalities.

Identifying Patients With Inflammatory Bowel Disease on Twitter and Learning From Their Personal Experience: Retrospective Cohort Study.

BACKGROUND: Patients use social media as an alternative information source, where they share information and provide social support. Although large amounts of health-related data are posted on Twitter and other social networking platforms each day, research using social media data to understand chronic conditions and patients' lifestyles is limited. OBJECTIVE: In this study, we contributed to closing this gap by providing a framework for identifying patients with inflammatory bowel disease (IBD) on Twitter and learning from their personal experiences. We enabled the analysis of patients' tweets by building a classifier of Twitter users that distinguishes patients from other entities. This study aimed to uncover the potential of using Twitter data to promote the well-being of patients with IBD by relying on the wisdom of the crowd to identify healthy lifestyles. We sought to leverage posts describing patients' daily activities and their influence on their well-being to characterize lifestyle-related treatments. METHODS: In the first stage of the study, a machine learning method combining social network analysis and natural language processing was used to automatically classify users as patients or not. We considered 3 types of features: the user's behavior on Twitter, the content of the user's tweets, and the social structure of the user's network. We compared the performances of several classification algorithms within 2 classification approaches. One classified each tweet and deduced the user's class from their tweet-level classification. The other aggregated tweet-level features to user-level features and classified the users themselves. Different classification algorithms were examined and compared using 4 measures: precision, recall, F1 score, and the area under the receiver operating characteristic curve. In the second stage, a classifier from the first stage was used to collect patients' tweets describing the different lifestyles patients adopt to deal with their disease. Using IBM Watson Service for entity sentiment analysis, we calculated the average sentiment of 420 lifestyle-related words that patients with IBD use when describing their daily routine. RESULTS: Both classification approaches showed promising results. Although the precision rates were slightly higher for the tweet-level approach, the recall and area under the receiver operating characteristic curve of the user-level approach were significantly better. Sentiment analysis of tweets written by patients with IBD identified frequently mentioned lifestyles and their influence on patients' well-being. The findings reinforced what is known about suitable nutrition for IBD as several foods known to cause inflammation were pointed out in negative sentiment, whereas relaxing activities and anti-inflammatory foods surfaced in a positive context. CONCLUSIONS: This study suggests a pipeline for identifying patients with IBD on Twitter and collecting their tweets to analyze the experimental knowledge they share. These methods can be adapted to other diseases and enhance medical research on chronic conditions.

Extra-analytical Point of Care Testing errors in the pandemic era

Introduction: during the current SARS-CoV-2 pandemic phase, the use of rapid diagnostic devices outside the laboratory has expanded enormously, creating great opportunities but also new risks. Methods: the present observational study evaluated the type and frequency of errors of the extra-analytical phases through an active search on all unclear or ambiguous cases. 252 241 rapid antigenic tests performed outside the laboratory in different health facilities over a 132-day period were considered. The requests, the patient demographics and the results were later entered manually onto the Laboratory Information System (LIS). Results: through a number of data checks and internal reports, 2 556 cases of errors in the pre-examination phase were recorded, with a relative frequency of 12,274 parts per million (ppm). The vast majority of errors were observed in this phase;these were due mainly to computer communication problems induced by human errors that made the loading of results or the issuing of the reports difficult. The remaining cases involving erroneous personal data or patient identification amounted to 16 (64 ppm), confirming the relative safety of this phase in decentralized analysis. The errors identified in the post-examination phase were 540, with a relative frequency of 2140 ppm. The assessment of the severity of the errors with Failure Mode and Effect Analysis (FMEA) allowed us to identify in particular, the attribution of the report to the wrong person (20 ppm) and the manual transcription of an incorrect result (20 ppm). Discussion: this study contributes to the comprehension of the critical issues connected to the Point of Care Testing and made it possible to establish corrective actions: improving staff training, choice of instruments with reading devices and establishing direct computer connection for the entering of the requests and results to the LIS.

LATVIA'S DOCTORS EXPERIENCE ON PROVIDING REMOTE CONSULTATIONS DURING COVID-19 PANDEMICS

Objective: Major challenge for health care service became a situation of a pandemic caused by COVID-19 infection, when doctors continued to provide care switching to remote consultations when possible. This provided better accessibility to receive consultation form doctor, and those consultations where sometimes the only way to be consulted for the patient. Aims of the study: To explore and analyze Latvia's doctors experience on providing remote consultations during COVID-19 pandemics. Objective: 1) To detect the volume and type of work physicians are able to manage remotely. 2) To find out what technical, GDPR-associated and other difficulties face doctors during the process. 3) To define the general attractiveness of remote consulting. Design and method: Two groups of doctors filled in the questionnaires in year 2021. Results: In one group, which consisted of 200 general practitioners, most respondents consider they could provide remote medical consultation about healthy lifestyle - 89,0% (n = 178);regarding check-ups, screening results - 86,5% (n = 173);to manage recommendations of other healthcare specialists - 66% (n = 132). Difficulties faced: 54,0% (n = 108) were not satisfied with payment, 22,5% (n = 45) admitted lack of specific skills, 45,5% (n = 91) found it problematic to make e-referrals. 61,0% (n = 122) like the idea of implementing services in practice on a regular basis. Another group, which consisted of 62 doctors of different specialties working in Clinical University Hospital. All doctors (62) were consulting by phone, 18 doctors only by phone, however 44 were consulting via e-mails, social network on video-conference platforms as well. Research revealed that the process of patient identification, search of patient's data and sensitive data transfer is complicated, time consuming and requires different approaches to manage. Conclusions: Latvian doctors have experience in remote consulting. When consulting remotely Latvian doctors were facing technical, legal, social and other difficulties. Some Latvian doctors are interested in introducing remote consulting services in their practice, if the restrictive moments for their provision would be eliminated. As well as there is need for amendments in normative regulation for providing remote consultations, and challenges relates to payment system implementations especially for state paid services.

Timely Short-Term Specialized Palliative Care For Older People With Frailty and Their Family in Primary Care: Easier Said Than Done

Background/aims: Short-term specialized palliative homecare that is initiated timely based on complex needs has been advocated for older people with frailty. But there is insufficient evidence on the feasibility of such an intervention. To assess the feasibility and preliminary effectiveness of a timely shortterm specialized palliative homecare intervention (i.e., Frailty+) for older people with frailty and family. Methods: Pilot randomised controlled trial and process evaluation. Eligible patients were 70 years or older, had a clinical frailty score 5-7, complex needs and about to be discharged home from hospital. They were randomly assigned (1:1) to Frailty+ or standard care. The Frailty+ group received specialized palliative care by a palliative homecare nurse over 8 weeks, who followed guides for home visits. The primary endpoint was change on a sum score based on five key symptoms, i.e., breathlessness, pain, anxiety, constipation, drowsiness, over 8 weeks. Feasibility, implementation, mechanisms and context were assessed post-intervention using mixed methods. We used thematic and descriptive statistical analyses. Results: 37 patients were randomly assigned to Frailty+ group (n=19, mean age 84y) or standard care (n=18, mean age 84y) and 26 family carers. The Frailty+ group received at least one home visit, as intended. Hospital care staff reported difficulties in patient identification. Patients and family reported positive views on the home visits, nurses that the guides were often not useful. Most important contextual factors were related to the COVID-19 crisis e.g., less continuity of care. Mean sum scores on primary outcome at baseline was 6.0 in Frailty+ and 5.6 in the control group, at 8-weeks was 4.5 in Frailty+ and 4.1 in the control group (adjusted ratio 1.0). Conclusions: Frailty+ was well received by patients and family and to a lesser extent healthcare providers. Based on these results, further refinement of Frailty+ and RCT design is needed to optimize the intervention and evaluation.

Transfusion errors in the COVID-19 ward resulting in changes in standard operating procedures (SOP)

Background: The pandemic of COVID-19 has led to alterations in SOP across the transfusion process, including administration of blood in COVID-19 wards. COVID-19 patients who present with symptomatic anaemia and have multiple risk factors will need blood transfusions. ABO-incompatible blood transfusions leading to acute haemolytic transfusion reaction is a rare but potentially fatal complication. The National Haemovigilance Coordinating Centre of the National Blood Centre, Malaysia reported the national incidence of incorrect blood components transfused (IBCT) in relation to total blood products transfused in 2019 to be 75 per 10,000 units. Five IBCTs reported were related to administration errors. Aims: We reported two IBCTs involving two patients who required blood transfusions in a COVID-19 ward. Patient 1 was a 74 year old man who complained of chest pain, with a haemoglobin (Hb) of 5.7 g/ dl. Patient 2 was a 50 year old woman with a Hb of 6.4 g/dl. When the two units of blood arrived on the ward, one doctor completed the pre-administration checklist in the 'clean' zone, which Nurse 1 then counter-checked. Nurse 1 put the two blood bags into separate transparent plastic bags and labelled them with the wrong patient's identity sticker. She then handed both blood bags to Nurse 2 in the 'dirty' zone, without the patients' blood compatibility labels, blood request forms and bedside checklist forms. Positive patient identification was not done by Nurse 2 and the transfusions commenced. Patient 1 complained of chills around 10 min into the transfusion. The error was only realized 35 min into the transfusion when the symptoms persisted and the temperature taken was 38°C. Patient 2, who was given a unit of group O blood that belonged to Patient 1, did not report any adverse reactions. Our aim is to identify the root causes of these IBCTs and to execute the necessary changes in pre-transfusion SOP to minimize future recurrences. Methods: Samples from both patients were investigated for transfusion reactions. Rechecking of blood groups was done manually with the test tube method. Direct and indirect anti human globulin tests (DAT/IAT) and recheck of cross-matching were performed with the column agglutination technology (CAT) method at 37/AHG phase. Urine samples were tested using urine dipsticks. Isohaemagglutinin (anti-A/B titre) was performed using the CAT method at 37/AHG phase. Plasma Hb was measured with a photometer and a microcuvette. Results: Both patient 1's post-transfusion samples (immediate and post-24 h) were O-RhD positive. Blood group of the donor's bag was B-RhD positive. DAT for both samples was positive with IgG(3+) and C3D (1+). IAT for both samples was negative. Recheck of crossmatching with both samples was incompatible (4+). Urine tests were negative for haemoglobin. A low anti-B titre of 1:16 was detected. Plasma Hb was measured twice at a low level below the reference range. Patient 2's workup was unremarkable. Summary/Conclusions: Two IBCTs occurred in the COVID-19 ward, with one major ABO-mismatched IBCT due to human errors and deviation from standard SOP. Pre-transfusion SOP was still unclear in the COVID-19 ward setting prior to these incidences. All medical personnel in the COVID-19 ward underwent retraining on safe transfusion practices. One COVID-19 patient's blood compatibility label, blood request form, bedside checklist form and blood bag should be brought into 'dirty' zone to be checked by two medical personnel at one time.

Toward Real-World Reproducibility: A Systematic Review and Meta-Analysis of Common Rwd/Rwe Protocol Components

Objectives: Use of real-world data/real-world evidence (RWD/RWE) in the life sciences is accelerating. The FDA has issued draft guidance for the conduct of real-world data-driven studies in clinical development. However, RWD protocol development standards lag, leading to heterogeneity of findings and consequent unreliability of results. The need to address challenges has become urgent due to increasing importance of reliable evidence generation in the COVID-19 pandemic. We performed a systematic review of published RWD protocols to understand current practices to support improvement in standards frameworks. Methods: We extracted protocols referencing RWD from. We defined essential real-world study concepts and mapped them to standard discrete protocol components. We summarized these components, including but not limited to: objectives, operational definitions of endpoints, inclusion/exclusion criteria, patient identification algorithms, schematics, extract/transform/load (ETL) methods, common data model (CDM), safety, analysis, and machine learning (ML)/artificial intelligence (AI). We identified areas of harmonization and disagreement, as well as missing components. Results: The search identified 220 real-world protocols. Despite substantial harmonization in some areas, particularly those components typical to all research studies, there was considerable disagreement regarding the representation of RWD objectives, RWD inclusion/exclusion criteria, data management, ETL, CDM, ML/AI, study design, and statistical analysis. In many cases, studies did not include real-world-specific elements at all. Quantification and statistical attribution of heterogeneity is ongoing. Conclusions: Incorporating best practices and harmonization of protocol development methods and reporting may lead to improved quality, consistency, and reproducibility of studies. The primary limitation of this study was that “real-world” was neither sensitive nor specific as a search term as it is often used imprecisely. Follow-up surveillance studies will quantify and evaluate the impact of improved standards, encompassing all registered observational studies.

Getting the most out of digital healthcare systems: Data repurposing to improve inpatient diabetes care

Aims: At a North London hospital which lacked remote blood glucose monitoring, we sought to introduce a system that automatically alerted the Diabetes Team of inpatients with deranged blood glucose (BG) measurements. We hypothesised that this would streamline the team's workflow, and improve glycaemic control in inpatients. Methods: We developed a feature that used data from recently deployed e-vitals software to generate daily reports containing lists of inpatients who had experienced hypo-or substantial hyper-glycaemia ( < 4 or >20mmol/L) within the previous 24h. These reports were automatically sent each morning to the Diabetes Team. Results: Feedback from specialist nurses suggested improvement in the efficiency of their patient identification workflow, which previously had principally involved taking phone referrals and manually searching ward lists. Initial post-intervention data did not suggest improvement in hospital-wide deranged BG rates, however this was confounded by a sharp rise in covid-19 admissions shortly after deployment, with the majority receiving corticosteroids. After several months of use, the feature unexpectedly failed for approximately six weeks, during which time on average significantly more daily hypoglycaemic episodes occurred vs the preceding six-week period (two-sided rank sum, p < 0.001), with rates returning to baseline after it was reintroduced. Conclusions: Our intervention aided staff workflows and possibly improved inpatient glycaemic control, although worsening glycaemic control outcomes upon intervention withdrawal cannot reliably be extrapolated to infer overall benefit of the feature vs pre-intervention standard-of- care. Financial barriers often preclude deployment of gold-standard digital systems in healthcare;innovative exploitation of data generated by more affordable systems can improve productivity and patient care without additional cost.

Determinants of turnaround time in a rapid genome sequencing program

Previous studies of genome sequencing (GS) in critically ill childrenhave made use of either modified hardware or working procedureswhich would be difficult, if not impossible, to integrate into existingclinical workflows1. Our lab’s transition from exome sequencing (ES) to GS offered an opportunity to implement in-house rapid genomesequencing (rGS) in critically ill children in a manner which couldintegrate with existing clinical workflows. We conducted a feasibilityand implementation pilot by offering rGS to child-parent triosconcurrently undergoing clinical rapid ES (rES) via a reference lab.The purpose of this study was to identify and address operationalbarriers to implementation of a rGS program capable of communicatinga preliminary result within 7 days of consent. We consideredthis time span to be more reflective of clinical realities than lab-quotedturnaround times (TAT) which typically start at sample receipt andthus do not account for challenges in sample acquisition and pre-testcounseling in a critical care setting, nor the impact of shipping times.Here we present data on TAT and lessons learned from the first 27subjects enrolled.Using rapid cycle improvement methodologies, we identified fourdistinct but inter-related workflows requiring optimization:1. Pre-analytic: patient identification through acquisition ofsamples2. Wet-lab: extraction through sequencing3. Bioinformatics: secondary and tertiary analysis as well as rapididentification of causal variants4. Return of resultsFigure 1 summarizes TAT across cases, demonstrating the markedimprovements in TAT with our programmatic approach to improvement.We used our first 9 cases to determine a baseline TAT for theentire process and to delineate the 4 main workflows (above). Atbaseline, excluding cases delayed by COVID-19 restrictions, mean TATwas 17.12 days (3 sequential deviant range: 7.05–27.19 days).Following deployment of our programmatic approach to rGS, meanTAT fell to 6.19 days (3 sequential deviant range: 0.51–11.87 days).Table 1 summarizes the observations and insights, by workflow, whichimpacted upon TAT and/or implementation. The single biggest impacton TAT was optimization of bioinformatics by removing all manualsteps between starting sequencing and producing human interpretable,filtered, annotated output of high-priority variants for interpretation.The second biggest source of improvement was optimization ofthe sequencing itself as well as prioritizing sample processing for andaccess to sequencing runs. While variant ranking is helpful in identifying causal variants, in 9/10 cases with a diagnostic findingthe causal variant(s)were obvious to the study teamwithin minutes ofviewing the annotated variant list, regardless of variant rank. (Figure Presented) As time required for sequencing and analytic workflows fell, therelative contribution of other workflows to overall TAT shifted and itbecame more obvious that early identification and utilization of thisapproach is very important in lowering overall time to diagnosis(Figure 2). In 6/10 cases with a diagnostic finding, the initial approachof the clinical team was NOT rES (and thus patients were not eligiblefor rGS on a research basis). Had rGS been the initial diagnosticmodality chosen, a diagnosis could have been reached in a median 12days sooner (range 2–28 days). There were also several cases wheresequencing was delayed when one or both parents did not present tothe lab to provide a blood sample in a timely manner. Optimization ofsequencing or analytic workflows cannot meaningfully improveoutcomes either of these situations.Our findings suggest some important considerations for institutionsdeveloping or seeking to improve rapid sequencing programs for acuteand critically ill children: (Table Presented) • Optimization of computational resource utilization and phenotypecuration saves more time than improved variant filtering orprioritization.• Obtaining samples from parents is non-trivial.• Even trained geneticists may fail to recognize appropriatecandidates for rGS.