Novel tech throws knock-out punch to ECG improving GP referral decisions to cardiology.

PURPOSE: In a comparator study, designed with assistance from the Food and Drug Administration, a State-of-the-Art (SOTA) ECG device augmented with automated analysis, the comparator, was compared with a breakthrough technology, Cardio-HART (CHART). METHODS: The referral decision defined by physician reading biosignal-based ECG or CHART report were compared for 550 patients, where its performance is calculated against the ground truth referral decision. The ground truth was established by cardiologist consensus based on all the available measurements and findings including echocardiography (ECHO). RESULTS: The results confirmed that CHART analysis was far more effective than ECG only analysis: CHART reduced false negative rates 15.8% and false positive (FP) rates by 5%, when compared with SOTA ECG devices. General physicians (GP's) using CHART saw their positive diagnosis rate significantly increased, from ~10% to ~26% (260% increase), and the uncertainty rate significantly decreased, from ~31% to ~1.9% (94% decrease). For cardiology, the study showed that in 98% of the cases, the CHART report was found to be a good indicator as to what kind of heart problems can be expected (the 'start-point') in the ECHO examination. CONCLUSIONS: The study revealed that GP use of CHART resulted in more accurate referrals for cardiology, resulting in fewer true negative or FP-healthy or mildly abnormal patients not in need of ECHO confirmation. The indirect benefit is the reduction in wait-times and in unnecessary and costly testing in secondary care. Moreover, when used as a start-point, CHART can shorten the echocardiograph examination time.

The effect of a decision-support mHealth application on maternal and neonatal outcomes in two district hospitals in Rwanda: pre - post intervention study.

BACKGROUND: Globally, mobile health (mHealth) applications are known for their potential to improve healthcare providers' access to relevant and reliable health information. Besides, electronic decision support tools, such as the Safe Delivery mHealth Application (SDA), may help to reduce clinical errors and to ensure quality care at the point of service delivery. The current study investigated the use of the SDA and its relationship to basic emergency obstetric and newborn care (BEmONC) outcomes for the most frequent complications in Rwanda; post-partum haemorrhage (PPH) and newborn asphyxia. METHODS: The study adopted a pre-post intervention design. A pre-intervention record review of BEmONC outcomes: Apgar score and PPH progressions, was conducted for 6 months' period (February 2019 - July 2019). The intervention took place in two district hospitals in Rwanda and entails the implementation of the SDA for 6 months (October 2019- March 2020), and included 54 nurses and midwives using the SDA to manage PPH and neonatal resuscitation. Six months' post-SDA intervention, the effect of the SDA on BEmONC outcomes was evaluated. The study included 327 participants (114 cases of PPH and 213 cases of neonatal complications). The analysis compared the outcome variables between the baseline and the endline data. Fisher's exact test was used to compare the proportions and test between-group differences and significance level set at p < 0.05. RESULTS: Unstable newborn outcomes following neonatal resuscitation were recorded in 62% newborns cases at baseline and 28% newborns cases at endline, P-value = 0.000. Unstable maternal outcomes following PPH management were recorded in 19% maternal cases at baseline and 6% maternal cases at endline, P-value = 0.048. There was a significant association between the SDA intervention and newborns' and maternal' outcomes following neonatal resuscitation and PPH management, 6 months after baseline. CONCLUSION: The use of the SDA supported nurses and midwives in the management of PPH and neonatal resuscitation which may have contributed to improved maternal and neonatal outcomes during 6 months of the SDA intervention. The findings of this study are promising as they contribute to a broader knowledge about the effectiveness of SDA in low and middle income hospital settings.

Feasibility and usability of a mobile health tool on anticoagulation management for patients with atrial fibrillation: a pilot study.

PURPOSE: Appropriate prescription of oral anticoagulants (OACs) and good patient adherence are essential to ensure optimal anticoagulation in patients with atrial fibrillation (AF). The aim of this study is to develop a mobile health tool to aid both clinicians and patients with AF in anticoagulation therapy. METHODS: In this study, a novel anticoagulation management model integrating decision support and patient follow-up, the I-Anticoagulation, was developed based on a WeChat Mini Program. With this tool, the risks of stroke and bleeding in AF patients can automatically be calculated according to their characteristics. Anticoagulation regimens were recommended based on a trade-off analysis that balances stroke and bleeding risks according to recent clinical guidelines. A shared decision can be made with full communication between medical professionals and patients. Moreover, follow-up was also conducted using I-Anticoagulation. RESULTS: A total of 120 AF patients receiving anticoagulants (40 received warfarin and 80 received non-vitamin K antagonist oral anticoagulants [NOACs]) were included in the pilot study. The incidence of thromboembolic events was 2.5% and 1.3%, and the rates of bleeding events were 22.5% and 13.8% in the warfarin and NOAC groups, respectively. Generally, self-reported adherence was high, and the satisfaction with anticoagulation was good in all patients with AF. CONCLUSION: Overall, the anticoagulation management model developed in this study could be involved in the full process of anticoagulation therapy in AF patients to improve rationality, adherence, and satisfaction in both medical professionals and patients. However, the usability, feasibility, and acceptability of the I-Anticoagulant-based anticoagulation management model need to be further assessed through well-designed random clinical trials.

Design of a novel multifunction decision support/alerting system for in-patient acute care, ICU and floor (AlertWatch AC).

BACKGROUND: Multifunction surveillance alerting systems have been found to be beneficial for the operating room and labor and delivery. This paper describes a similar system developed for in-hospital acute care environments, AlertWatch Acute Care (AWAC). RESULTS: A decision support surveillance system has been developed which extracts comprehensive electronic health record (EHR) data including live data from physiologic monitors and ventilators and incorporates them into an integrated organ icon-based patient display. Live data retrieved from the hospitals network are processed by presenting scrolling median values to reduce artifacts. A total of 48 possible alerts are generated covering a broad range of critical patient care concerns. Notification is achieved by paging or texting the appropriated member of the critical care team. Alerts range from simple out of range values to more complex programing of impending Ventilator Associated Events, SOFA, qSOFA, SIRS scores and process of care reminders for the management of glucose and sepsis. As with similar systems developed for the operating room and labor and delivery, there are green, yellow, and red configurable ranges for all parameters. A census view allows surveillance of an entire unit with flashing or text to voice alerting and enables detailed information by windowing into an individual patient view including live physiologic waveforms. The system runs via web interface on desktop as well as mobile devices, with iOS native app available, for ease of communication from any location. The goal is to improve safety and adherence to standard management protocols. CONCLUSIONS: AWAC is designed to provide a high level surveillance view for multi-bed hospital units with varying acuity from standard floor patients to complex ICU care. Alerts are generated by algorithms running in the background and automatically notify the selected member of the patients care team. Its value has been demonstrated for low acuity patients, further study is required to determine its effectiveness in high acuity patients.

Artificial intelligence-enabled electrocardiograms for identification of patients with low ejection fraction: a pragmatic, randomized clinical trial.

We have conducted a pragmatic clinical trial aimed to assess whether an electrocardiogram (ECG)-based, artificial intelligence (AI)-powered clinical decision support tool enables early diagnosis of low ejection fraction (EF), a condition that is underdiagnosed but treatable. In this trial ( NCT04000087 ), 120 primary care teams from 45 clinics or hospitals were cluster-randomized to either the intervention arm (access to AI results; 181 clinicians) or the control arm (usual care; 177 clinicians). ECGs were obtained as part of routine care from a total of 22,641 adults (N = 11,573 intervention; N = 11,068 control) without prior heart failure. The primary outcome was a new diagnosis of low EF (≤50%) within 90 days of the ECG. The trial met the prespecified primary endpoint, demonstrating that the intervention increased the diagnosis of low EF in the overall cohort (1.6% in the control arm versus 2.1% in the intervention arm, odds ratio (OR) 1.32 (1.01-1.61), P = 0.007) and among those who were identified as having a high likelihood of low EF (that is, positive AI-ECG, 6% of the overall cohort) (14.5% in the control arm versus 19.5% in the intervention arm, OR 1.43 (1.08-1.91), P = 0.01). In the overall cohort, echocardiogram utilization was similar between the two arms (18.2% control versus 19.2% intervention, P = 0.17); for patients with positive AI-ECGs, more echocardiograms were obtained in the intervention compared to the control arm (38.1% control versus 49.6% intervention, P < 0.001). These results indicate that use of an AI algorithm based on ECGs can enable the early diagnosis of low EF in patients in the setting of routine primary care.

Serial use of existing clinical decisions aids can reduce computed tomography pulmonary angiography for pulmonary embolism.

Pulmonary embolism (PE) remains a diagnostic challenge in emergency medicine. Clinical decision aids (CDAs) like the Pulmonary Embolism Rule-Out Criteria (PERC) are sensitive but poorly specific; serial CDA use may improve specificity. The goal of this before-and-after study was to determine if serial use of existing CDAs in a novel diagnostic algorithm safely decreases the use of CT pulmonary angiograms (CTPA). This was a retrospective before-and-after study conducted at an urban ED with 105,000 annual visits. Our algorithm uses PERC, Wells' score, and D-dimer in series, before moving to CTPA. The algorithm was introduced in January, 2017. Use of CDAs and D-dimer in the 24 months pre- and 12 months post-intervention were obtained by chart review. The algorithm's effect on CTPA ordering was assessed by comparing volume 5 years pre- and 3 years post-intervention, adjusted for ED volume. Mean CTPAs per 1000 adult ED visits was 11.1 in the 5 pre-intervention years and 9.9 in the 3 post-intervention years (p < 0.0001). Use of PERC, Wells' score and D-dimer increased from 1.1%, 1.1%, and 28% to 8.8% (p = 0.0002) 8.1% (p = 0.0005), and 35% (p = 0.0066), respectively. Pre-intervention, there were six potentially missed PEs compared to three in the post-intervention period. Introduction of our serial CDA diagnostic algorithm was associated with increased use of CDAs and D-dimer and reduced CTPA rate without an apparent increase in the number of missed PEs. Prospective validation is needed to confirm these results.

Impact of a computerized decision support tool deployed in two intensive care units on acute kidney injury progression and guideline compliance: a prospective observational study.

BACKGROUND: Acute kidney injury (AKI) affects a large proportion of the critically ill and is associated with worse patient outcomes. Early identification of AKI can lead to earlier initiation of supportive therapy and better management. In this study, we evaluate the impact of computerized AKI decision support tool integrated with the critical care clinical information system (CCIS) on patient outcomes. Specifically, we hypothesize that integration of AKI guidelines into CCIS will decrease the proportion of patients with Stage 1 AKI deteriorating into higher stages of AKI. METHODS: The study was conducted in two intensive care units (ICUs) at University Hospitals Bristol, UK, in a before (control) and after (intervention) format. The intervention consisted of the AKIN guidelines and AKI care bundle which included guidance for medication usage, AKI advisory and dashboard with AKI score. Clinical data and patient outcomes were collected from all patients admitted to the units. AKI stage was calculated using the Acute Kidney Injury Network (AKIN) guidelines. Maximum AKI stage per admission, change in AKI stage and other metrics were calculated for the cohort. Adherence to eGFR-based enoxaparin dosing guidelines was evaluated as a proxy for clinician awareness of AKI. RESULTS: Each phase of the study lasted a year, and a total of 5044 admissions were included for analysis with equal numbers of patients for the control and intervention stages. The proportion of patients worsening from Stage 1 AKI decreased from 42% (control) to 33.5% (intervention), p = 0.002. The proportion of incorrect enoxaparin doses decreased from 1.72% (control) to 0.6% (intervention), p < 0.001. The prevalence of any AKI decreased from 43.1% (control) to 37.5% (intervention), p < 0.05. CONCLUSIONS: This observational study demonstrated a significant reduction in AKI progression from Stage 1 and a reduction in overall development of AKI. In addition, a reduction in incorrect enoxaparin dosing was also observed, indicating increased clinical awareness. This study demonstrates that AKI guidelines coupled with a newly designed AKI care bundle integrated into CCIS can impact patient outcomes positively.

Insulin dose optimization using an automated artificial intelligence-based decision support system in youths with type 1 diabetes.

Despite the increasing adoption of insulin pumps and continuous glucose monitoring devices, most people with type 1 diabetes do not achieve their glycemic goals1. This could be related to a lack of expertise or inadequate time for clinicians to analyze complex sensor-augmented pump data. We tested whether frequent insulin dose adjustments guided by an automated artificial intelligence-based decision support system (AI-DSS) is as effective and safe as those guided by physicians in controlling glucose levels. ADVICE4U was a six-month, multicenter, multinational, parallel, randomized controlled, non-inferiority trial in 108 participants with type 1 diabetes, aged 10-21 years and using insulin pump therapy (ClinicalTrials.gov no. NCT03003806). Participants were randomized 1:1 to receive remote insulin dose adjustment every three weeks guided by either an AI-DSS, (AI-DSS arm, n = 54) or by physicians (physician arm, n = 54). The results for the primary efficacy measure-the percentage of time spent within the target glucose range (70-180 mg dl-1 (3.9-10.0 mmol l-1))-in the AI-DSS arm were statistically non-inferior to those in the physician arm (50.2 ± 11.1% versus 51.6 ± 11.3%, respectively, P < 1 × 10-7). The percentage of readings below 54 mg dl-1 (<3.0 mmol l-1) within the AI-DSS arm was statistically non-inferior to that in the physician arm (1.3 ± 1.4% versus 1.0 ± 0.9%, respectively, P < 0.0001). Three severe adverse events related to diabetes (two severe hypoglycemia, one diabetic ketoacidosis) were reported in the physician arm and none in the AI-DSS arm. In conclusion, use of an automated decision support tool for optimizing insulin pump settings was non-inferior to intensive insulin titration provided by physicians from specialized academic diabetes centers.

Intelligent, Autonomous Machines in Surgery.

Surgeons perform two primary tasks: operating and engaging patients and caregivers in shared decision-making. Human dexterity and decision-making are biologically limited. Intelligent, autonomous machines have the potential to augment or replace surgeons. Rather than regarding this possibility with denial, ire, or indifference, surgeons should understand and steer these technologies. Closer examination of surgical innovations and lessons learned from the automotive industry can inform this process. Innovations in minimally invasive surgery and surgical decision-making follow classic S-shaped curves with three phases: (1) introduction of a new technology, (2) achievement of a performance advantage relative to existing standards, and (3) arrival at a performance plateau, followed by replacement with an innovation featuring greater machine autonomy and less human influence. There is currently no level I evidence demonstrating improved patient outcomes using intelligent, autonomous machines for performing operations or surgical decision-making tasks. History suggests that if such evidence emerges and if the machines are cost effective, then they will augment or replace humans, initially for simple, common, rote tasks under close human supervision and later for complex tasks with minimal human supervision. This process poses ethical challenges in assigning liability for errors, matching decisions to patient values, and displacing human workers, but may allow surgeons to spend less time gathering and analyzing data and more time interacting with patients and tending to urgent, critical-and potentially more valuable-aspects of patient care. Surgeons should steer these technologies toward optimal patient care and net social benefit using the uniquely human traits of creativity, altruism, and moral deliberation.

Measures of self-efficacy among doctors in conducting an online search for clinical decision making.

OBJECTIVES: To measure the perceived ability and level of confidence among doctors in performing the different tasks involved in conducting an online search for clinical decision making. METHODS: A large-scale cross-sectional survey was conducted in 36 District Headquarter Hospitals (DHQs), 89 Tehsil Headquarter Hospitals (THQs), 293 Rural Health Centers (RHCs) and 2455 Basic Health Units (BHUs) in Punjab, Pakistan. Using a quota sampling, data were collected from 517 doctors on a set of 11 statements. The collected data were analysed statistically. RESULTS: Of the 517 doctors, 73 (14.1%) had 'never accessed health care information online' for clinical decision making. Mean values of the doctors' response to the 11 statements ranged from 1.66 to 2.30 indicating that most of the doctors were 'not confident' in their ability to perform the tasks. CONCLUSION: The majority of doctors perceived themselves able to perform the different tasks involved in conducting an online search. Age and working experience were significant factors in the perception of their ability in performing the tasks. The study recommends promotional and educational activities to motivate interest, increase awareness, develop knowledge and skills for doctors to access information that would help in their clinical decision making.

Evolution of Diabetes Technology.

Technological innovations have fundamentally changed diabetes care. Insulin pump use and continuous glucose monitoring are associated with improved glycemic control along with a better quality of life; automated insulin-dosing advisors facilitate and improve decision making. Glucose-responsive automated insulin delivery enables the highest targets for time in range, lowest rate and duration of hypoglycemia, and favorable quality of life. Clear targets for time in ranges and a standard visualization of the data will help the diabetes technology to be used more efficiently. Decision support systems within and integrated cloud environment will further simplify, unify, and improve modern routine diabetes care.

Design and implementation of a clinical decision support tool for primary palliative Care for Emergency Medicine (PRIM-ER).

BACKGROUND: The emergency department is a critical juncture in the trajectory of care of patients with serious, life-limiting illness. Implementation of a clinical decision support (CDS) tool automates identification of older adults who may benefit from palliative care instead of relying upon providers to identify such patients, thus improving quality of care by assisting providers with adhering to guidelines. The Primary Palliative Care for Emergency Medicine (PRIM-ER) study aims to optimize the use of the electronic health record by creating a CDS tool to identify high risk patients most likely to benefit from primary palliative care and provide point-of-care clinical recommendations. METHODS: A clinical decision support tool entitled Emergency Department Supportive Care Clinical Decision Support (Support-ED) was developed as part of an institutionally-sponsored value based medicine initiative at the Ronald O. Perelman Department of Emergency Medicine at NYU Langone Health. A multidisciplinary approach was used to develop Support-ED including: a scoping review of ED palliative care screening tools; launch of a workgroup to identify patient screening criteria and appropriate referral services; initial design and usability testing via the standard System Usability Scale questionnaire, education of the ED workforce on the Support-ED background, purpose and use, and; creation of a dashboard for monitoring and feedback. RESULTS: The scoping review identified the Palliative Care and Rapid Emergency Screening (P-CaRES) survey as a validated instrument in which to adapt and apply for the creation of the CDS tool. The multidisciplinary workshops identified two primary objectives of the CDS: to identify patients with indicators of serious life limiting illness, and to assist with referrals to services such as palliative care or social work. Additionally, the iterative design process yielded three specific patient scenarios that trigger a clinical alert to fire, including: 1) when an advance care planning document was present, 2) when a patient had a previous disposition to hospice, and 3) when historical and/or current clinical data points identify a serious life-limiting illness without an advance care planning document present. Monitoring and feedback indicated a need for several modifications to improve CDS functionality. CONCLUSIONS: CDS can be an effective tool in the implementation of primary palliative care quality improvement best practices. Health systems should thoughtfully consider tailoring their CDSs in order to adapt to their unique workflows and environments. The findings of this research can assist health systems in effectively integrating a primary palliative care CDS system seamlessly into their processes of care. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03424109. Registered 6 February 2018, Grant Number: AT009844-01.

The role of medical smartphone apps in clinical decision-support: A literature review.

INTRODUCTION: The now ubiquitous smartphone has huge potential to assist clinical decision-making across the globe. However, the rapid pace of digitalisation contrasts starkly with the slower rate of medical research and publication. This review explores the evidence base that exists to validate and evaluate the use of medical decision-support apps. The resultant findings will inform appropriate and pragmatic evaluation strategies for future clinical app developers and provide a scientific and cultural context for research priorities in this field. METHOD: Medline, Embase and Cochrane databases were searched for clinical trials concerning decision support and smart phones from 2007 (introduction of first smartphone iPhone) until January 2019. RESULTS: Following exclusions, 48 trials and one Cochrane review were included for final analysis. Whilst diagnostic accuracy studies are plentiful, clinical trials are scarce. App research methodology was further interrogated according to setting and decision-support modality: e.g. camera-based, guideline-based, predictive models. Description of app development pathways and regulation were highly varied. Global health emerged as an early adopter of decision-support apps and this field is leading implementation and evaluation. CONCLUSION: Clinical decision-support apps have considerable potential to enhance access to care and quality of care, but the medical community must rise to the challenge of modernising its approach if it is truly committed to capitalising on the opportunities of digitalisation.

Development and Performance of the Pulmonary Embolism Result Forecast Model (PERFORM) for Computed Tomography Clinical Decision Support.

Importance: Pulmonary embolism (PE) is a life-threatening clinical problem, and computed tomographic imaging is the standard for diagnosis. Clinical decision support rules based on PE risk-scoring models have been developed to compute pretest probability but are underused and tend to underperform in practice, leading to persistent overuse of CT imaging for PE. Objective: To develop a machine learning model to generate a patient-specific risk score for PE by analyzing longitudinal clinical data as clinical decision support for patients referred for CT imaging for PE. Design, Setting, and Participants: In this diagnostic study, the proposed workflow for the machine learning model, the Pulmonary Embolism Result Forecast Model (PERFORM), transforms raw electronic medical record (EMR) data into temporal feature vectors and develops a decision analytical model targeted toward adult patients referred for CT imaging for PE. The model was tested on holdout patient EMR data from 2 large, academic medical practices. A total of 3397 annotated CT imaging examinations for PE from 3214 unique patients seen at Stanford University hospitals and clinics were used for training and validation. The models were externally validated on 240 unique patients seen at Duke University Medical Center. The comparison with clinical scoring systems was done on randomly selected 100 outpatient samples from Stanford University hospitals and clinics and 101 outpatient samples from Duke University Medical Center. Main Outcomes and Measures: Prediction performance of diagnosing acute PE was evaluated using ElasticNet, artificial neural networks, and other machine learning approaches on holdout data sets from both institutions, and performance of models was measured by area under the receiver operating characteristic curve (AUROC). Results: Of the 3214 patients included in the study, 1704 (53.0%) were women from Stanford University hospitals and clinics; mean (SD) age was 60.53 (19.43) years. The 240 patients from Duke University Medical Center used for validation included 132 women (55.0%); mean (SD) age was 70.2 (14.2) years. In the samples for clinical scoring system comparisons, the 100 outpatients from Stanford University hospitals and clinics included 67 women (67.0%); mean (SD) age was 57.74 (19.87) years, and the 101 patients from Duke University Medical Center included 59 women (58.4%); mean (SD) age was 73.06 (15.3) years. The best-performing model achieved an AUROC performance of predicting a positive PE study of 0.90 (95% CI, 0.87-0.91) on intrainstitutional holdout data with an AUROC of 0.71 (95% CI, 0.69-0.72) on an external data set from Duke University Medical Center; superior AUROC performance and cross-institutional generalization of the model of 0.81 (95% CI, 0.77-0.87) and 0.81 (95% CI, 0.73-0.82), respectively, were noted on holdout outpatient populations from both intrainstitutional and extrainstitutional data. Conclusions and Relevance: The machine learning model, PERFORM, may consider multitudes of applicable patient-specific risk factors and dependencies to arrive at a PE risk prediction that generalizes to new population distributions. This approach might be used as an automated clinical decision-support tool for patients referred for CT PE imaging to improve CT use.

Computer-Assisted Planning for Stereoelectroencephalography (SEEG).

Stereoelectroencephalography (SEEG) is a diagnostic procedure in which multiple electrodes are stereotactically implanted within predefined areas of the brain to identify the seizure onset zone, which needs to be removed to achieve remission of focal epilepsy. Computer-assisted planning (CAP) has been shown to improve trajectory safety metrics and generate clinically feasible trajectories in a fraction of the time needed for manual planning. We report a prospective validation study of the use of EpiNav (UCL, London, UK) as a clinical decision support software for SEEG. Thirteen consecutive patients (125 electrodes) undergoing SEEG were prospectively recruited. EpiNav was used to generate 3D models of critical structures (including vasculature) and other important regions of interest. Manual planning utilizing the same 3D models was performed in advance of CAP. CAP was subsequently employed to automatically generate a plan for each patient. The treating neurosurgeon was able to modify CAP generated plans based on their preference. The plan with the lowest risk score metric was stereotactically implanted. In all cases (13/13), the final CAP generated plan returned a lower mean risk score and was stereotactically implanted. No complication or adverse event occurred. CAP trajectories were generated in 30% of the time with significantly lower risk scores compared to manually generated. EpiNav has successfully been integrated as a clinical decision support software (CDSS) into the clinical pathway for SEEG implantations at our institution. To our knowledge, this is the first prospective study of a complex CDSS in stereotactic neurosurgery and provides the highest level of evidence to date.

Using Mobile Health to Support Clinical Decision-Making to Improve Maternal and Neonatal Health Outcomes in Ghana: Insights of Frontline Health Worker Information Needs.

BACKGROUND: Developing and maintaining resilient health systems in low-resource settings like Ghana requires innovative approaches that adapt technology to context to improve health outcomes. One such innovation was a mobile health (mHealth) clinical decision-making support system (mCDMSS) that utilized text messaging (short message service, SMS) of standard emergency maternal and neonatal protocols via an unstructured supplementary service data (USSD) on request of the health care providers. This mCDMSS was implemented in a cluster randomized controlled trial (CRCT) in the Eastern Region of Ghana. OBJECTIVE: This study aimed to analyze the pattern of requests made to the USSD by health workers (HWs). We assessed the relationship between requests made to the USSD and types of maternal and neonatal morbidities reported in health facilities (HFs). METHODS: For clusters in the intervention arm of the CRCT, all requests to the USSD during the 18-month intervention period were extracted from a remote server, and maternal and neonatal health outcomes of interest were obtained from the District Health Information System of Ghana. Chi-square and Fisher exact tests were used to compare the proportion and type of requests made to the USSD by cluster, facility type, and location; whether phones accessing the intervention were shared facility phones or individual-use phones (type-of-phone); or whether protocols were accessed during the day or at night (time-of-day). Trends in requests made were analyzed over 3 6-month periods. The relationship between requests made and the number of cases reported in HFs was assessed using Spearman correlation. RESULTS: In total, 5329 requests from 72 (97%) participating HFs were made to the intervention. The average number of requests made per cluster was 667. Requests declined from the first to the third 6-month period (44.96% [2396/5329], 39.82% [2122/5329], and 15.22% [811/5329], respectively). Maternal conditions accounted for the majority of requests made (66.35% [3536/5329]). The most frequently accessed maternal conditions were postpartum hemorrhage (25.23% [892/3536]), other conditions (17.82% [630/3536]), and hypertension (16.49% [583/3536]), whereas the most frequently accessed neonatal conditions were prematurity (20.08% [360/1793]), sepsis (15.45% [277/1793]), and resuscitation (13.78% [247/1793]). Requests made to the mCDMSS varied significantly by cluster, type of request (maternal or neonatal), facility type and its location, type-of-phone, and time-of-day at 6-month interval (P<.001 for each variable). Trends in maternal and neonatal requests showed varying significance over each 6-month interval. Only asphyxia and sepsis cases showed significant correlations with the number of requests made (r=0.44 and r=0.79; P<.001 and P=.03, respectively). CONCLUSIONS: There were variations in the pattern of requests made to the mCDMSS over time. Detailed information regarding the use of the mCDMSS provides insight into the information needs of HWs for decision-making and an opportunity to focus support for HW training and ultimately improved maternal and neonatal health.

Smart Trauma: Improving the Delivery of Evidence-Based Trauma Care.

BACKGROUND: Mobile technology can aid in healthcare decision-making at the point of care. We created a Web-based trauma-specific smartphone application containing links to local protocols and national organization guidelines for trauma providers. We hypothesized that smartphone access to these guidelines would facilitate application of knowledge in a timely fashion. MATERIALS AND METHODS: Trauma providers were randomized to have or not have access to their smartphone during a timed, 10-question examination of trauma scenarios based on Eastern Association for the Surgery of Trauma, Western Trauma Association, and local protocols. Participants were then surveyed regarding their experience with the application. Groups were compared based on time with completion and percentage of correct answers. Subgroup analyses were completed to assess the utility of the application. RESULTS: Of 30 participants, 16 were randomized to smartphone use. Smartphone users took longer to complete the examination than nonusers (9:18 versus 6:36, P = 0.007) but answered a greater proportion of questions correctly (50% versus 40%, P = 0.159). Smartphone users had a higher percentage correct for Eastern Association for the Surgery of Trauma and Western Trauma Association protocol-based questions (78% versus 52%, P = 0.027; 70% versus 39%, P = 0.011), but no difference for local protocol-based questions (29% versus 37%, P = 0.48). Smartphone users who reported recent application use had the longest time to completion (11:44, P = 0.023) but the highest percentage correct (60%, P = 0.03). CONCLUSIONS: Smartphone use among those familiar with our trauma application resulted in the highest percentage correct but increased times to completion. The application interface should be streamlined, and providers educated to improve usage and reduce time to access information.

Using machine learning models to predict oxygen saturation following ventilator support adjustment in critically ill children: A single center pilot study.

BACKGROUND: In an intensive care units, experts in mechanical ventilation are not continuously at patient's bedside to adjust ventilation settings and to analyze the impact of these adjustments on gas exchange. The development of clinical decision support systems analyzing patients' data in real time offers an opportunity to fill this gap. OBJECTIVE: The objective of this study was to determine whether a machine learning predictive model could be trained on a set of clinical data and used to predict transcutaneous hemoglobin oxygen saturation 5 min (5min SpO2) after a ventilator setting change. DATA SOURCES: Data of mechanically ventilated children admitted between May 2015 and April 2017 were included and extracted from a high-resolution research database. More than 776,727 data rows were obtained from 610 patients, discretized into 3 class labels (< 84%, 85% to 91% and c92% to 100%). PERFORMANCE METRICS OF PREDICTIVE MODELS: Due to data imbalance, four different data balancing processes were applied. Then, two machine learning models (artificial neural network and Bootstrap aggregation of complex decision trees) were trained and tested on these four different balanced datasets. The best model predicted SpO2 with area under the curves < 0.75. CONCLUSION: This single center pilot study using machine learning predictive model resulted in an algorithm with poor accuracy. The comparison of machine learning models showed that bagged complex trees was a promising approach. However, there is a need to improve these models before incorporating them into a clinical decision support systems. One potentially solution for improving predictive model, would be to increase the amount of data available to limit over-fitting that is potentially one of the cause for poor classification performances for 2 of the three class labels.