Beyond the hype of big data and artificial intelligence: building foundations for knowledge and wisdom.

Big data, coupled with the use of advanced analytical approaches, such as artificial intelligence (AI), have the potential to improve medical outcomes and population health. Data that are routinely generated from, for example, electronic medical records and smart devices have become progressively easier and cheaper to collect, process, and analyze. In recent decades, this has prompted a substantial increase in biomedical research efforts outside traditional clinical trial settings. Despite the apparent enthusiasm of researchers, funders, and the media, evidence is scarce for successful implementation of products, algorithms, and services arising that make a real difference to clinical care. This article collection provides concrete examples of how "big data" can be used to advance healthcare and discusses some of the limitations and challenges encountered with this type of research. It primarily focuses on real-world data, such as electronic medical records and genomic medicine, considers new developments in AI and digital health, and discusses ethical considerations and issues related to data sharing. Overall, we remain positive that big data studies and associated new technologies will continue to guide novel, exciting research that will ultimately improve healthcare and medicine-but we are also realistic that concerns remain about privacy, equity, security, and benefit to all.

Validating the use of registries and claims data to support randomized trials: Rationale and design of the Extending Trial-Based Evaluations of Medical Therapies Using Novel Sources of Data (EXTEND) Study.

BACKGROUND: Randomized controlled trials are the "gold standard" for comparing the safety and efficacy of therapies but may be limited due to high costs, lack of feasibility, and difficulty enrolling "real-world" patient populations. The Extending Trial-Based Evaluations of Medical Therapies Using Novel Sources of Data (EXTEND) Study seeks to evaluate whether data collected within procedural registries and claims databases can reproduce trial results by substituting surrogate non-trial-based variables for exposures and outcomes. METHODS AND RESULTS: Patient-level data from 2 clinical trial programs-the Dual Antiplatelet Therapy Study and the United States CoreValve Studies-will be linked to a combination of national registry, administrative claims, and health system data. The concordance between baseline and outcomes data collected within nontrial data sets and trial information, including adjudicated end point events, will be assessed. We will compare the study results obtained using these alternative data sources to those derived using trial-ascertained variables and end points using trial-adjudicated end points and covariates. CONCLUSIONS: Linkage of trials to registries and claims data represents an opportunity to use alternative data sources in place of and as adjuncts to randomized clinical trial data but requires further validation. The results of this research will help determine how these data sources can be used to improve our present and future understanding of new medical treatments.

["MonDossierMedical.ch": an efficient tool for sharing medical data between patients and doctors]. / L'aventure "MonDossierMedical.ch". En route vers le déploiement a large échelle.

"MonDossierMedical.ch" is a project led by the canton of Geneva, making it possible for every patient to access his own electronic health record (EHR) and to share the medical files with his doctors. It was introduced across the canton in mid-2013, and provided to all patients free of charge. it is based on the first Swiss-wide e-health-compliant pilot project "e-toile". The canton of Geneva developed "e-toile" as a public-private partnership together with Swiss Post and it was launched in 2009 in some of the canton's municipalities. Back then, Geneva's EHR represented the first Swiss attempt to link all healthcare professionals in the treatment chain. Today, it serves more than 4,600 patients and 380 physicians. This number is growing regularly, as well as the health care institutions (physicians, private hospitals, labs) joining the community. Added value tools, such as a shared care plan and a shared medication list will make the platform even more attractive.

The use of a kiosk-model bilingual self-triage system in the pediatric emergency department.

BACKGROUND: Streamlining the triage process is the key in improving emergency department (ED) workflow. Our objective was to determine if parents of pediatric ED patients in, low-literacy, inner-city hospital, who used the audio-assisted bilingual (English/Spanish) self-triage kiosk, were able to enter their child's medical history data using a touch screen panel with greater speed and accuracy than routine nurse-initiated triage. METHODS: Parent/child dyads visiting the pediatric ED for nonurgent conditions (February to April 2012) were randomized prospectively to self-triage kiosk group (n = 200) and standard nurse triage group (n = 200). Both groups underwent routine nurse-initiated triage that included verbal elicitation of basic medical history and manual entry into patients' electronic medical records. RESULTS: The kiosk user was a parent in 88.5% of the cases, a patient (range, 11-17 years) in 9.5% of the cases, and a proxy user (sibling or friend) in 2% of the cases. Language choice for kiosk use was equally distributed (English vs Spanish, 50.5% vs 49.5%). The mean (SD) time to enter medical history data by the kiosk group was significantly shorter than the standard nurse triage group (94.38 [38.61] vs 126.72 [62.61] seconds; P < 0.001). Significant inverse relationship was observed between parent education level and kiosk usage time (r = -0.26; P < 0.001). The mean inaccuracies were significantly lower for kiosk group (P < 0.05) in areas of medical, medication and immunization histories, and total discrepancy score. CONCLUSIONS: Kiosk triage enabled users to enter basic medical triage history data quickly and accurately in an ED setting with future potential for its wider use in improving ED workflow efficiency.

[GP practice in Belgium: strengths and weaknesses. Potential improvement by electronic patient record]. / Médecine générale en Belgique: forces et faiblesses. Apport potentiel du dossier patient informatisé.

In July 2012, NIHDI (National Institute for Health and Disease Insurance) has published a GP's performance report under the umbrella of the quality promotion Council, addressed to policymakers, stake holders and general practitioners (GP). This article is summarizing the performance measurement, strengths and weaknesses. Main findings are a pretty high level of population satisfaction, a good accessibility and the central role of GP's within the health organization. The report points out weaknesses in quality variation and sustainability. Electronic medical file could be helpful to improve quality.

Code Status Documentation Availability and Accuracy Among Emergency Patients with End-stage Disease.

INTRODUCTION: Some patients with end-stage disease who may neither want nor benefit from aggressive resuscitation receive such treatment if they cannot communicate in an emergency. Timely access to patients' current resuscitation wishes, or "code status," should be a key metric of electronic health records (EHR). We sought to determine what percentage of a cohort of patients with end-stage disease who present to the emergency department (ED) have accessible, code status documents, and for those who do, how quickly can this documentation be retrieved. METHODS: In this cross-sectional study of ED patients with end-stage disease (eg, palliative care, metastatic malignancy, home oxygen, dialysis) conducted during purposefully sampled random accrual times we performed a standardized, timed review of available health records, including accompanying transfer documents. We also interviewed consenting patients and substitute decision makers to compare available code status documents to their current wishes. RESULTS: Code status documentation was unavailable within 15 minutes of ED arrival in most cases (54/85, or 63%). Retrieval time was under five minutes in the rest, especially when "one click deep" in the EHR. When interviewed, 20/32 (63%) expressed "do not resuscitate" wishes, 10 of whom had no supporting documentation. Patients from assisted-living (odds ratio [OR] 6.7; 95% confidence interval [CI], 1.7-26) and long-term care facilities (OR 13; 95% CI, 2.5-65) were more likely to have a documented code status available compared to those living in the community. CONCLUSION: The majority of patients with end-stage disease, including half of those who would not wish resuscitation from cardiorespiratory arrest, did not have code status documents readily available upon arrival to our tertiary care ED. Patients living in the community with advanced disease may be at higher risk for unwanted resuscitative efforts should they present to hospital in extremis. While easily retrievable code status documentation within the EHR shows promise, its accuracy and validity remain important considerations.

Impact on child vaccination completion rates of short message services (SMS) reminders in developing countries.

INTRODUCTION: The Expanded Programme on Immunization has, since its inception, struggled to achieve high completion rates for child immunizations. The introduction of 2YL (second year of life) immunizations presents the programme with fresh challenges to assuring high completion rates. METHODS: Using the same procedures as those employed in the 2017 article on SMS reminders, of which this is an update, I searched the NLM database for all recent articles from developing countries on SMS reminders for reduction of vaccination dropout rates. I summarized these and earlier articles in tabular form. RESULTS: The freshly reviewed articles are confirmatory of earlier studies which show an improvement in vaccination completion rates when SMS reminders are sent to mothers and other caregivers. CONCLUSION: All of the studies reviewed were based on pilot projects. It is time, and past time, to go to scale with SMS reminders, perhaps stand alone, or as part of a larger system of electronic immunization registers. There may be potential for use of WhatsApp in dropout reduction, thus far documented only in other public health applications.

Predictors of student use of an electronic record.

BACKGROUND: Little is known on ambulatory clerkship students' use of an electronic medical record (EMR). We investigated students' use of recommended EMR tasks across different types of sites and studied the predictors of these recommended tasks. METHODS: Students documented how often they performed recommended EMR tasks and suggested improvements to enhance EMR use. We compared student performance of recommended tasks across different types of sites using χ2 tests and the Fisher's exact test. We performed regression analyses to investigate factors predicting students' performance of EMR tasks. Two faculty members read all of the suggested improvements and agreed on themes. RESULTS: From January 2014 to June 2015, 263 of 295 Family and Community Medicine Clerkship (FCMC) students (89.2%) were at sites that used an EMR. Of the 263 students, 68.4% typed their own note into the EMR, but only 31.2% entered orders and 27.8% entered prescriptions for their teacher to sign. Students' rating of the orientation to the EMR predicted their use of all EMR tasks. The number of years that the teaching site used an EMR predicted the students' use of some tasks. Suggested improvements included a better orientation to the EMR, more use of the EMR, and access to a computer and the EMR. Little is known on ambulatory clerkship students' use of an electronic medical record DISCUSSION: Many students did not perform recommended EMR tasks. To help more students learn EMR tasks, clinical teachers can offer students a detailed orientation to their EMR, provide them with access to a computer and the EMR, and give them the opportunity to perform recommended EMR tasks, including typing their own note and entering orders and prescriptions.

Identification of Overweight and Obesity in Low-Income Minority Children by Pediatric Providers and Child Characteristics Associated With Underrecognition.

INTRODUCTION: Childhood overweight and obesity remains prevalent. We sought to identify characteristics associated with correct coding of obese and overweight children ages 2-9 years by pediatric primary care providers (PCPs) in an urban, minority clinic. METHOD: Samples included electronic health records pre- (n = 262) and post- (n = 244) provider training to implement an intervention aiming to improve consistency in obesity prevention and management guidelines from 15 providers. Analysis included descriptive statistics and simple logistic regression. RESULTS: Child characteristics consistently associated with correct coding in both pre- and post-intervention included children with obese body mass index percentiles (vs. overweight) and older-aged children (vs. toddlers). Provider characteristics were not associated. DISCUSSION: Improvements in coding overweight and obese children occurred post-intervention, yet many were missed. Knowledge of characteristics associated with missed coding can inform targeted training for PCPs to implement evidence-based obesity guidelines with full fidelity for all families.

Differences in Access to and Use of Electronic Personal Health Information Between Rural and Urban Residents in the United States.

PURPOSE: The increase in use of health information technologies (HIT) presents new opportunities for patient engagement and self-management. Patients in rural areas stand to benefit especially from increased access to health care tools and electronic communication with providers. We assessed the adoption of 4 HIT tools over time by rural or urban residency. METHODS: Analyses were conducted using data from 7 iterations of the National Cancer Institute's Health Information National Trends Survey (HINTS; 2003-2014). Rural/urban residency was based on the USDA's 2003 Rural-Urban Continuum Codes. Outcomes of interest included managing personal health information online; whether providers maintain electronic health records (EHRs); e-mailing health care providers; and purchasing medicine online. Bivariate analyses and logistic regression were used to assess relationships between geography and outcomes, controlling for sociodemographic characteristics. FINDINGS: In total, 6,043 (17.6%, weighted) of the 33,749 respondents across the 7 administrations of HINTS lived in rural areas. Rural participants were less likely to report regular access to Internet (OR = 0.70, 95% CI = 0.61-0.80). Rural respondents were neither more nor less likely to report that their health care providers maintained EHRs than were urban respondents; however, they had decreased odds of managing personal health information online (OR = 0.59, 95% CI = 0.40-0.78) and e-mailing health care providers (OR = 0.62, 95% CI = 0.49-0.77). CONCLUSIONS: The digital divide between rural and urban residents extends to HIT. Additional investigation is needed to determine whether the decreased use of HIT may be due to lack of Internet connectivity or awareness of these tools.

[Medicine 4.0, the importance of electronics, information technology and microsystems in modern medicine - the case of customized chemotherapy]. / « Médecine 4.0 ¼ ou de l'importance des nouvelles technologies dans la médecine moderne - Le cas de la chimiothérapie personnalisée.

A paradigm shift seems to emerge, not only in industrial engineering ("Industry 4.0") but also in medicine: we are on the threshold to "Medicine 4.0". For many years, molecular biology had a leading position in life sciences, but today scientists start realizing that microelectronic systems, due to an increasing miniaturization, are reaching the scale of human cells and consequently can be used for therapeutic approaches. This article shows how microelectronics can play a major role in modern medicine, through the example of customized chemotherapy. This consists in determining, before the beginning of the treatment, what kind of chemotherapy or drug combination will be most effective for a given patient, and at which dose. This of course allows the lessening of a patient burden during treatment, but also to be more efficient and, in the long run, to save money. In order to do this, we have developed the Intelligent Microplate Reader (IMR), which allows us to accurately test different drugs on living cells by mimicking part of their usual environment.

[Value of hospital electronic health records for clinical research: contribution of the European project EHR4CR]. / Intérêt des dossiers de santé électroniques hospitaliers pour la recherche clinique - Le projet européen EHR4CR.

Electronic health records in hospitals contribute to improving the quality of care by enabling better management of clinical information. The databases thus constituted facilitate the exchange of health information with healthcare providers and optimize multidisciplinary coordination for better therapeutic results. The EHR4CR (Electronic Health Records for Clinical Research) European project has developed an innovative pilot platform enabling the reuse of this digital information for clinical research. By enhancing and speeding up clinical research procedures, this innovative approach makes it possible to conduct clinical trials more efficiently, faster, and more economically.

Validation of data from electronic data warehouse in diabetic ketoacidosis: Caution is needed.

AIMS: This study validated enterprise data warehouse (EDW) data for a cohort of hospitalized patients with a primary diagnosis of diabetic ketoacidosis (DKA). METHODS: 247 patients with 319 admissions for DKA (ICD-9 code 250.12, 250.13, or 250.xx with biochemical criteria for DKA) were admitted to Northwestern Memorial Hospital from 1/1/2010 to 9/1/2013. Validation was performed by electronic medical record (EMR) review of 10% of admissions (N = 32). Classification of diabetes type (Type 1 vs. Type 2) and DKA clinical status were compared between the EMR review and EDW data. RESULTS: Key findings included incorrect classification of diabetes type in 5 of 32 (16%) admissions and indeterminable classification in 5 admissions. DKA was not present, based on the review, in 11 of 32 (34%) admissions. DKA was not present, based on biochemical criteria, in 15 of 32 (47%) admissions. CONCLUSIONS: This study found that EDW data have substantial errors. Some discrepancies can be addressed by refining the EDW query code, while others, related to diabetes classification and DKA diagnosis, cannot be corrected without improving clinical coding accuracy, consistency of medical record documentation, or EMR design. These results support the need for comprehensive validation of data for complex clinical populations obtained through data repositories such as the EDW.

[Health data]. / Les données de santé.

Healthcare is considered as one of the most promising areas where big data can be applied to foster innovation for the benefit of patients and of the whole system. Healthcare analytics have the potential to accelerate R&D, increase knowledge on diseases and risk factors, improve treatments, develop personalised medicine, help physicians with decision support systems… The access to data is also a driving force for patients' empowerment and for the democratic debate. However, there are also concerns about the societal, economic and ethic impacts of this wave of digitization and of the growing use of data, algorithms and artificial intelligence. Given the issues at stake, collecting and analysing data generated by health care systems is a strategic challenge in all countries; in that respect the French National System of Health Data (a national data warehouse linking data from several sources and giving a vision of the care pathways for the entire population, with a ten-year history) is an asset, but it has to be completed and enriched with data from electronic health records.

Patient portal readiness among postpartum patients in a safety net setting.

BACKGROUND: Maternity patients interact with the healthcare system over an approximately ten-month interval, requiring multiple visits, acquiring pregnancy-specific education, and sharing health information among providers. Many features of a web-based patient portal could help pregnant women manage their interactions with the healthcare system; however, it is unclear whether pregnant women in safety-net settings have the resources, skills or interest required for portal adoption. OBJECTIVES: In this study of postpartum patients in a safety net hospital, we aimed to: (1) determine if patients have the technical resources and skills to access a portal, (2) gain insight into their interest in health information, and (3) identify the perceived utility of portal features and potential barriers to adoption. METHODS: We developed a structured questionnaire to collect demographics from postpartum patients and measure use of technology and the internet, self-reported literacy, interest in health information, awareness of portal functions, and perceived barriers to use. The questionnaire was administered in person to women in an inpatient setting. RESULTS: Of the 100 participants surveyed, 95% reported routine internet use and 56% used it to search for health information. Most participants had never heard of a patient portal, yet 92% believed that the portal functions were important. The two most appealing functions were to check results and manage appointments. CONCLUSIONS: Most participants in this study have the required resources such as a device and familiarity with the internet to access a patient portal including an interest in interacting with a healthcare institution via electronic means. Pregnancy is a critical episode of care where active engagement with the healthcare system can influence outcomes. Healthcare systems and portal developers should consider ways to tailor a portal to address the specific health needs of a maternity population including those in a safety net setting.