Tom Jones

eHealth strategist, planner and evaluator

  • eHealth Group offers a telehealth paradigm shift

    About a third of African countries invest in telehealth, identified by the WHO and Global Observatory for eHealth in their 2015 survey. Since then, eHealth Group, based in South Africa, has leapfrogged telehealth’s technology and its supply side. 

    Its lineage goes back to the Da Vinci Robotic Surgery. This enabled eHealth Group to incorporate the ideas, principles and techniques into its robotic telehealth services. Instead of a link between a patient and a healthcare professional, eHealth Group’s robots and products enable healthcare professionals to deal directly and easily with several patients in wards, ICUs and communities.

    At its core are:

    • Direct access to a wide range of clinical data about patients
    • Explicit, precise and clear images of patients
    • Its own excellent, global telecommunication network. An example is where doctors can have two screens.

    The picture shows a doctor accessing a patient’s view and PACS side by side. It’s an efficient and effective use of his time and benefits patients directly.


    South Africa’s eHealth Group provides services across Africa. It’s part of a global network, operating in over 30 countries, across 20 specialties in over a thousand hospitals and over 4,000 medical specialists available online. About 70% of US telestroke hospitals in the USA use eHealth Group’s services, up from 30% four years ago. 

    There’s a wide product range extending from tablets to sophisticated robots that provide high quality telehealth. Alongside the products, eHealth Group provides a service range including access to advice from a network of medical specialists, specialists who can provide direct patient care, and training. 

    Elliot Sack and some of his robots from eHealth Group will be at eHealthAfro 2017 on 2 to 4 October in Johannesburg. His presentation’ll reveal the paradigm shift that can make a difference to Africa’s health and healthcare. eHNA has more to report on it.

     

  • Informatics and EHRs can prevent strokes and improve monitoring

    Increasing responses to strokes and their after effects are important health priorities. A report in the US National Library of Medicine has estimated that in 2015, strokes were the second-leading cause of death worldwide after ischaemic heart disease. In 2010, strokes caused 5.3 million deaths globally, 10% of all deaths. Trends include increasing stroke mortality and lost Disability Adjusted Life Years (DALYs) in low- and middle-income countries and a dire estimate of the global economic impact unless effective preventive measures are implemented.

    Another study identified the aged-standardised incidence of stroke in Africa as 316 per 100,000, 0.3% population, and age-standardised prevalence rates of up to 981 per 100,000, almost 1%. Stroke incidence’s increasing, but the study said the “peculiar factors responsible for the substantial disparities in incidence velocity, ischaemic stroke proportion, mean age and case fatality compared to high-income countries remain unknown.” This is despite the incidence being lower than higher-income countries. A study in Sage Journals estimated the incidence of stroke, adjusted to the WHO World standard population, in 51 countries. It ranges from 76 to 199 per 100,000 population.

    Atrial fibrillation (AF) an irregular and often very fast heart rate may cause symptoms like heart palpitations, fatigue and shortness of breath. Treating it’s important because it may cause a stroke, with resulting adverse DALYs. After a stroke, AF needs monitoring. A study in Cardiology, published by Karger, aimed to identify the characteristics of atrial fibrillation (AF) in post-cryptogenic stroke. It’s a stroke with an unknown origin.

    The US Stroke Association has an estimate that cryptogenic strokes (CS) may be between 25% and 45% of ischemic strokes, so about 30%. They are where blood supply to part of the brain is interrupted or severely reduced, depriving brain tissue of oxygen and nutrients. Within minutes, brain cells begin to die. Ischemic strokes are about 87% of all types.

    The team’s project included Transient Ischemic Attacks (TIA). Mayo Clinic has a simple description of a TIA. It produces similar symptoms to a stroke, but usually lasting only a few minutes and causing no permanent damage. Often called a mini-stroke, a TIA may be a warning of worse to come.

    The team of cardiologists and informatics researchers from the Department of Medicine and Division of Cardiology at Santa Clara Valley Medical Center, the Biomedical Informatics Training Program, Stanford University, the Center for Biomedical Informatics Research, Stanford University School of Medicine and the University of California San Francisco, stratified a cohort of stroke patients by risk factors. It used data from EHRs.

    These included obesity, congestive heart failure, hypertension, coronary artery disease, peripheral vascular disease and valve disease. A risk-scoring model applied seven clinical variables that assigned patients into three groups. The risk-score’s measures of AF risk and may be used to select patients who need extended AF monitoring, especially home monitoring.

    The study’s an example of the value of doctors, informaticians and analysts working together to exploit the value of data in EHRs. It’s a model for Africa’s health systems and universities to work towards.

  • Kenya’s mHealth standards are strong on IOp

    Kenya’s Ministry of Health has set a solid foundation for its next step in eHealth regulation and good practices. The second main section in Kenya Standards and Guidelines for mHealth Systems deals with information exchange and Interoperability (IOp). It has a seven stage model of IOp maturity, including level 0 for no maturity and three conventional IOp classifications of technical, syntactic and semantic. They’re:

    • Conceptual, enabling other engineers to understand documentation and evaluation
    • Dynamic, to recognise and comprehend data changes in systems over time
    • Pragmatic, including modest AI
    • Semantic
    • Syntactic and workflow integration
    • Technical and integrated
    • None, so can be ignored.

    They combine into three categories, integration, IOp and composability for maximum interoperation. It’s a requirement that all Kenya’s mHealth complies with its IOp standards. These include Health Level (HL)7 version 3 for clinical messaging and International Classification of Diseases (ICD) 10, Systematized Nomenclature of Medicine (SNOMED) for coding, Logical Observation Identifiers Names and Codes (LOINC) and Rx Norm for pharmacies.

    Developers have to provide Standards for Applications Programming Interfaces (API) to define how their mHealth interacts with other systems. It fits into a Fast Health Interoperability Resources (FHIR) architecture. It complies with Integrating the Healthcare Enterprise (IHE) and HL7 standards

    While these apply to health and healthcare data, Kenya’s standards apply to social health determinants too. It’s an indicator of the breadth of its approach.

  • IBM’s cyber-threat index shows it’s increasing

    eHNA’s covered many cyber-threats reports. The IBM X-Force Threat Intelligence Index 2017 puts them into perspective. It’s not good. It sees 2016 as a defining year for cyber-security.

    Three themes are Distributed Denial of Service (DDoS) attacks, loads of records leaked through data breaches and a step up by organised cyber-crime on businesses. Over four billion records were leaked, more than the combined total 2014 and 2015. 

    Mega breach was redefined, with one source leaking over 1.5 billion records. The average cyber-security events of IBM® X-Force®’s clients exceeded 54 million, 3% up on 2015. An event’s illicit activity on a system or network detected by a security device or application. 

    Attacks are security events classified by correlation and analytics tools as malicious activity attempting to collect, disrupt, deny, degrade or destroy information system resources or its Information. X-Force found an average of 1,019 attacks, a 12% decrease compared to 1,157 attacks in 2015.

    Incidents are security events worthy of further investigation by IBM security analysts. It’s better news. Average incidents were down to 94 in 2016, a 48% drop from 2015’s 178. This doesn’t automatically mean cyber-security’s safer. It may be that cyber-attackers rely more on proven attacks that need fewer attempts. It’s also inconsistent with the combination of huge record leaks and a record year of vulnerability disclosures.

    There were many notable leaks in 2016 involving hundreds of gigabytes of email archives, documents, intellectual property and source code. They exposed organisations’ digital footprints. Previous data breaches were often fixed sets of structured information. Examples are credit card data, passwords, national ID numbers and Personal Health Information (PHI) data. This’s a paradigm shift.

    X-Force’s report profiles a range of cyber-attack methods. They’re:

    • Cross-Site Scripting (XSS)
    • Physical access
    • Brute force
    • Misconfiguration
    • Malvertising
    • Watering hole
    • Phishing
    • Structured Query Language Injection (SQLi)
    • DDoS
    • Malware
    • Heartbleed.

    The analysis and overview are extremely valuable for Africa’s heath executive and ICT teams. Web application vulnerability disclosures made up 22% of all vulnerability disclosures in 2016. Injecting unexpected items manipulating data structures comprised 74% of all cyber-attacks. These are a few of the priorities.

  • Technology’s setting timescales for Africa’s eHealth strategies

    A few years ago, eHealth strategies were much simpler. They set and captured relatively straightforward information architectures and standards, and bounded decisions, such as which structured systems were needed, which vendor should provide them, which ones were affordable and how to realise their benefits. These all sat in a temporal setting, rarely more than five years. Relativity in this setting doesn’t mean easy, it’s compared to the immense range eHealth opportunities and decisions now.

    eHealth’s increased and continuing sophistication means that Africa’s health systems and eHealth strategies need to be explicit about how far into the future they want to look. Their settings now are determined more by eHealth’s technology than calendars. Two commentaries can help to set these.

    Machine, Platform, Crowd: Harnessing our Digital Future, a book by Andrew McAffee and Erik Brynjolfsson, both working at MIT Sloan School of Management and on the MIT Initiative on the Digital Economy, sets out changes led by new ICT. They identify three.

    • AI that moves more control from people to computers
    • A shift from products to platforms that can scale investment
    • A move away from centralised institutions to users, the core to the crowd.

    Alongside these, the Institute of Electrical and Electronics Engineers (IEEE) identified nine top 2016 ICT trends. Its performance was reviewed in Computing Now. Some came to fruition in 2016. Others may reach critical development points this year. They were:

    1.     5G, promising unimaginable speeds

    2.     Virtual Reality (VA) and Augmented Reality (AR), with VR now available and AR expanding

    3.     Nonvolatile memory that can store more data at less cost and power

    4.     Cyber Physical Systems (CPS), used as the Internet of Things (IoT), its deeply embedded hardware and software in smart medical technologies

    5.     Data science, processes and systems to extract knowledge or insights from data in various forms, either structured or unstructured, and a continuation of some data analysis fields such as statistics, data mining, and predictive analytics

    6.     Capability-based security to provide finer grain protection and defences

    7.     Advanced machine learning, used in medical diagnosis, and exploring the construction of algorithms that can learn from and make predictions using data

    8.     Network Function Virtualization (NFV), an emerging technology providing virtualised infrastructure for the next-generation of cloud services

    9.     Containers that can deliver app faster and more efficiently.

    As the year progresses, how will Africa’s health systems evaluate and take decisions on these combined twelve ICT and eHealth trends? Ignoring them means falling behind. Assessing and adopting some of them means highly complex eHealth strategies. Maybe AI, 5G and analytics will be enough for now.

  • Will China expand its role in Africa’s eHealth?

    China already has a longstanding role in Africa’s economy. A report in TechCrunch by Lux Capital, a venture capital firm for science and technology, suggests it’ll expand and diversify.

    Josh Wolfe, Lux Capital managing partner and co-founder, says China’s access and influence throughout the African continent keeps growing. Some trends he explores include:

    • In 35 years, Africans will be 25% of the global population
    • 50% of Africa’s population’s under 20
    • The working-age population will expand by 20 to 30 million a year over the next 20 years, up from 530 million in 2015 to 920 million in 2035, then 1.4 billion by 2055
    • By 2050, Africa will make up 25% the world’s workforce.
    • By 2055, most of the net growth on the world’s workforce will be in Africa
    • Trade between Africa and China has mushroomed over the last 20 years, up from US$10 billion in 2000 to US$220 in 2014
    • China’s loans to African countries totalled US$5 billion a year from 2006-2010, US$10 billion a year from 2011-2012, then US$15 billion a year from 2013-2014
    • China maybe accounting for one-sixth of all lending across Africa.

    China’s been investing in training across Africa. Huawei, its technology heavyweight, earns 15% of its global revenue in Africa. It also trains 12,000 students in telecommunications and technology in Kenya, Congo, Angola and Nigeria. 

    It’s popular too. Wolfe says in Senegal, 86% of people said China’s helped to improve their country. About 56% said America had. He sees a significant expanding market for software, services, telecommunications, data centres, technological infrastructure and application layers. China’s technology companies will play a big role. It’ll include US start-ups partnering with big Chinese companies instead of their US equivalents. 

    China’s expanded role in Africa’s an extension of a trend started some 15 years ago. It may offer an opportunity for Africa’s health leaders to attract Chinese investment in scaling-up and widening their eHealth strategies and plans. A healthier population can enhance Africa’s workforce.

  • Kenya’s mHealth standards for documentation add clarity

    Covering a wide range of mHealth standards, Kenya’s Ministry of Health has set a firm foundation to step up its wide eHealth regulation and good practices. The first main section in Kenya Standards and Guidelines for mHealth Systems deals with development and functions. It’s comprehensive.

    Software development has to comply with a set of phases: 

    • Requirement gathering
    • Systems analysis
    • Systems design
    • Development and implementation
    • Systems testing
    • Operations and maintenance
    • Support
    • Post-implementation M&E.

    Documentation needed for these includes:

    ·       Systems Requirement Specification (SRS)

    ·       Software design documents, depending on the mHealth software design methodology, will include some of:

    o   Unified Modelling Language (UML) diagrams

    o   Data Flow Diagrams (DFD)

    o   Flow charts

    o   Entity relationship diagrams

    ·       Implementation plan, including:

    o   Implementation manual

    o   Training and capacity building manual

    ·       Test plans

    ·       Deployment procedures

    ·       M&E criteria.

    Three other required documents are:

    • Technical manual
    • Developer’s guide
    • User manual.

    Four requirements for data validation are included:

    • First order, ensure valid data formats and values and prevent obvious data entry errors
    • Second order, historical data comparisons for alerts for changes
    • Third order assess data for consistency in specific forms and indicator sets
    • Fourth order, assess statistical outliers for validity. 

    These examples show the range and rigour of Kenya’s mHealth standards. They fit all types of eHealth too. It’s a considerable benchmark for all Africa’s health systems.

  • EHRs aren’t enough, they need communications

    Communicating isn’t easy. In “Is Anybody Listening?” a 1950s article in Fortune Magazine, William H Whyte, a journalist and author, suggested that “The single biggest problem in communication is the illusion that it has taken place.” Across health systems and within healthcare organisations, it can be at the demanding end of the spectrum.

    An eBrief from Spok says EHRs don’t do enough for clinical communication and collaboration. It’s particularly prevalent for information needing acknowledgements and prompt action. Spok’s proposal’s a complementary system for messaging and collaboration among all healthcare team members and across whole healthcare organisations. In Picking Up Where EHRs Leave Off: 6 Ways to Bolster the Benefits of Your EHR by Improving Communications in Your Hospital, six requirements are:

    • Support all clinical and other healthcare team members
    • Provide an enterprise-wide directory to serve  as the source of truth
    • On-call schedule integration and clinician status
    • Integration with third-party systems
    • Capability to support many devices
    • Deliver emergency notification rapidly.

    These will combine to enhance EHRs’ benefits. It reflects the changing nature of EHRs since the 1990s when they were seen as a database for healthcare professionals. Now, they’re a vital data source for health analytics, and Spok’s communicating needs.

    The perspective offers Africa’s health systems a broader approach to moving towards their goals for EHRs. Anton Chekov, the Russian story-teller, had an illuminating view of the wider world when he wrote in his Note Book “Don’t tell me the moon is shining; show me the glint of light on broken glass.”

  • US eHealth IOp should focus on big impacts

    Interoperability (IOp) in eHealth isn’t an absolute state. Measuring it isn’t either. Sir William Osler, a Canadian doctor and one of four founding professors of Johns Hopkins Hospital, didn’t need to bother with eHealth IOp, but hinted at it in strategy when he said “In seeking absolute truth we aim at the unattainable and must be content with broken portions.”

    Challenges are how much, and which IOp measurement will achieve contentment without breaking it. An article in Fierce Health provide some indication. It sets out responses from five US organisations to the Office of the National Coordinator (ONC) report in the Proposed Interoperability Standards Measurement Framework, reported by eHNA in May.

    It has two main themes.  One’s measuring standards implementation. The other’s how end users can refine and customise standards to meet their needs. Most groups expressed some trepidation that new standards would result in an undue burden for providers. They want the ONC to focus on measurement areas with the biggest impact. Their advice is directly relevant for Africa’s IOp plans.

    The American Medical Informatics Association (AIMIA) supported the ONC’s framework in its response. It also asked the ONC to target “high-value standards” that offer the biggest impact. Specific requirements are functionalities for accessing drug databases and transmitting laboratory data. It wants the measurement framework to be automated too, so easier reporting mechanisms translate to higher participation.

    A combination of support and caution was part of the Health Information Management and Systems Society (HIMSS) contribution. It offers three main themes: 

    A sub group of HIMSS, the Electronic Health Record Association (EHRA) suggested a combination of standardised approaches with non-standard methods. It’s a focus on use cases with the biggest impact. It emphasised the potential burdens too.

    The College for Health Information Management Executives (CHIME) highlighted patient matching as one of the biggest IOp barriers. It says measurement standards are premature, and wants the ONC to:

    • Develop standards for seamless communication between ICT systems
    • Ensure that data exchange identifies patients with 100% certainty
    • Make data exchange usable for clinicians before tackling IOp standards.

    CHIME proposed that if the measurement framework’s implemented, the ONC should work with stakeholders to prioritise cases and develop a granular set of standards.

    Health IT Now, a coalition of patient groups, healthcare organisations, employers and payers, recognises that measuring IOP’s necessary, and said a narrow focus on successful data transmissions devalue improvements in using data to improve care and defer the capability of health systems to exchange information. It wants collaboration with patients and patient advocates and private sector organisations that can contribute to identifying, developing, and deploying IOp standards for better information systems.

    These perspectives can inform Africa’s eHealth development. IOp and its choices are seldom off the eHealth agenda.

  • Microsoft's joined the AI race

    The Artificial Intelligence (AI) supply side’s heating up. An article on the BBC says Microsoft is promoting its AI credibility and visibility. It brought its top scientists from across the world to London demonstrate their ideas, vision, research, initiatives and AI’s direction that over the next few years.

    One project’s Seeing AI. It does several clever things. It helps people with visual impairment access information using a smartphone’s camera. Pointing a phone at documents enables it to read them aloud. A set of bleeps guide people to barcoded on drinks cans to tell users what it is. 

    Creating Seeing AI took several years. Microsoft’s AI research programme’s been underway for 25 years. It has three main parts, speech, language and vision. Some of it’s now coming to fruition and seeing, recognising and understanding our world in a similar way we do. 

    Microsoft’s AI progress adds extra opportunities for Africa’s consumers and health systems. When will they become part of Africa’s eHealth strategies?