• Analytics
  • Healthcare organisations need several approaches to analytics

    Data needed for business and healthcare analytics can be complex. A white paper from Sisense Business Analytics and the Data Complexity Matrix shows four components:

    Data size

     

     

    Large    

    BIG

    COMPLEX

    Small

    SIMPLE 

    DIVERSIFIED

     

    Few        

    Many

    Number of Sources as Data Tables

    Analytics for each of the four types have different requirements. Using simple data can rely on access to databases and visualisation tools. Big data needs data warehouse and appliance tools to step up analytics to this next level. It requires resources for data preparation and aggregation, and to configure, maintain, and support the tools and their integration with other parts of the analytics map.

    Diversified data adds to the sophistication. Extract, Transform Load (ETL) become more complicated and may need ICT teams input to set up and structure data and analytical tools before generating analytic output. The workload may be so laborious that specialised ETL tools are needed to automate data preparation.

    Complex data needs specialised skills and resources. Costs are high, comprising manpower costs that increase with each data preparation or ETL activity and the need for sophisticated tools.

    As Africa’s health systems move towards more analytics, they’ll need to set up increasingly sophisticated methodologies and develop people with specialised skills. Sisense has a free test drive of its analytics tool that provides an insight to a health analytics journey.

  • Smart dashboards are essential for eHealth benefits

    Realising benefits depends extensively on maximising the number of users. It also depends on them using the data effectively. This, in turn depends on meeting their requirements. There are two main parts to this, the information they need and having in a format, style and presentation that they can use for faster decision taking.

    Tableau, a dashboard supplier, has a white paper saying there are four main ways to use data to improve healthcare:

    • Using analytics for better  population health management
    • Using real-time analytics to increase productivity
    • Aggregating and blending data to reveal and fix supply chain inefficiencies
    • Automating ad hoc visual analysis for better revenue cycle management.

    Providing more data doesn’t always help. The first step’s to simplify data that’s already available. It might easier said than done. In a hospital organisation, there can be a thousand or more health workers. Common themes for simplification include: 

    • Use data visualisation so users can quickly automate processes rapidly
    • Enable users to visualise and assimilate data the way their minds work
    • Helps users see and understand their healthcare data no matter how big it is, or how many systems it is stored in
    • Connect quickly to any data, analyse it and share insights to reveal opportunities to benefit patients, health workers and healthcare organisations.

    As Africa’s eHealth expands, it’s vital that these concepts are in place too. Maximising eHealth’s benefits depend on it. 

  • Keep health analytics simple, for now

    Claims for Big Data’s and analytics’ benefits are considerable for commercial enterprises. For health and healthcare, it seems a bit more complicated, and sometimes bewildering. Where and how to start is a challenge.

    In an article in Health Analytics, Dr Danyal Ibrahim, emergency physician and Chief Data and Analytics Officer at Saint Francis Care in Connecticut, invokes the wisdom of the ancient Chinese philosopher Lao Tzu, who famously said “A journey of a thousand miles begins with a single step.”

    He sees the first step towards health analytics as a daunting leap of faith. Succeeding depends on:

    1. Clinical leadership and champions
    2. The right partnerships
    3. A strong, diverse Big Data analytics team
    4. Create an analytics infrastructure that delivers actionable, meaningful data to points of care
    5. Identify and break down data silos and develop a Big Data roadmap.

    The last point holds the key to creating a streamlined data analytics infrastructure that reveals comprehensive and meaningful patients’ stories. These lead on to help make better and timely decisions. While healthcare data’s supposed to be connected around individual patients, it isn’t. It ends up in different siloes, creating a big barrier to using data to improve care.

    From the streamlined data, the next steps are:

    1. Bring all the data stewards together
    2. Redesign the analytics team to focus on value for better care, more cost effective care and better patients’ experiences for patients

    Effective analytics teams comprise a wide range of skills. They include, clinical, financial analytics, SQL development, data warehousing and data science, including statistical modelling and natural language processing. For Africa’s health systems, all these skills and knowledge may not be readily available. Analytics teams will need structured, integrated and technical development, so a sustainable training budget.

    Lao Tzu’s proposed first step of analytics costs has a second step of producing valuable outputs that healthcare teams can use for measurable, proven patient benefits. Analytics will then justify itself.

  • Algorithms can carry rare, but big risks

    As algorithms expand and replace human decisions, they make life easier and can make better decisions. They also have another, less valuable effect. They reduce users’ skill and decision-taking levels. So, when users have to intervene for the rare decisions that are beyond algorithms’ capabilities, users may not be sharp enough.

    This phenomenon’s set out in Messy How to be Creative and Resilient in a Tidy-minded World by Tim Harford, published by Little, Brown. He describes several examples where it’s happened, and they’re often catastrophic. This’s the territory that Big Data and analytics are taking eHealth.

    Gary Klein, a psychologist, has researched decision taking and supports Harford’s view. In Streetlights and Shadows Searching for the Keys to Adaptive Decision Making published by MIT Press, he says when algorithms take decisions, users tend to stop improving their skills and performance. Algorithm dependency’s associated with people’s eroded judgement, increasing their algorithm dependency in a vicious cycle. Eventually users become passive and less vigilant. In healthcare, it can be catastrophic.

    A solution’s to use algorithms to confirm healthcare professionals’ decisions. Where it’s been tried in aviation and meteorology, human decisions are usually supported by algorithms. This creates a role for algorithms of ensuring people haven’t overlooked something significant in a critical decision. It also keeps people in control and their decision-taking prowess sharp.

    As Africa’s health systems adopt algorithms, it’s important they don’t become replacements for healthcare professionals. If they do, on the rare occasions when algorithms can’t do it, people who intervene might not have skills that are too rusty to be able to do it either.

  • How can Africa start health analytics?

    Healthcare organisations face a wide range of tough objectives. They include improving outcomes, better patient satisfaction, complying with regulations, responding to government mandates and controlling operational costs. Pyramid Analytics, in a post on Health IT Analytics links to its white paper Healthcare Analytics: Key Drivers and Opportunities, says business analytics can help. The approach may help Africa’s health systems make a start on using analytics.

    It proposes three main ways business analytics can help:

    1. Reduce costs
    • Fully use collected clinical data
    • Increase visibility into finances to identify profitable and underutilised services
    • Identify providers that perform the costliest procedures
    • Evaluate cost-effective treatment options
    • Streamline claims processes to control costs and improve operational efficiency
    • Optimise resources such as consolidating block scheduling and buying supplies in bulk
    1. Improve medical outcomes and enhance patient satisfaction
    • Prevent readmissions caused by poor-quality care
    • Use patient data from numerous sources to improve diagnoses, fine-tune treatments and reduce waiting times
    • Use performance benchmarking, retrospective reports and multi-dimensional analysis to translate clinical data into meaningful information
    • Apply key performance metrics to detect potential risks and make informed decisions
    1. Comply with regulations and government mandates to improve patient health
    • Switch to regulated EHRs
    • Use data governance to restrict access to sensitive protected health information
    • Improve healthcare quality to satisfy performance initiatives
    • Reduce readmissions and time spent at hospitals
    • Improve claims and reimbursement management.

    Taken together, these three initiatives have considerable resource requirements. For Africa’s health systems to apply them, they’ll need to set them into an achievable, affordable sequence over at least a medium term horizon.

  • Healthcare analytics can be affordable

    Healthcare analytics promises massive rewards by improving hospitals efficiency. For Africa, the drawback’s that countries don’t have millions of dollars to spend on ICT upgrades and developments. An article in Healthcare IT News shows that hospitals don’t need a staff of 50 and a US$3 million budget to achieve valuable investment returns from healthcare analytics.

    Karen Reff, manager of decision support at Union General Hospital, a small facility in Georgia with 45 beds, invested US$50,000 and launched an analytics programme that freed staff from time-consuming data-crunching. It enabled clinicians to spend more time on delivering care and uncovering best practices to help reduce readmissions and pinpoint the best location for an outpatient clinic for Chronic Obstructive Pulimnary Disease COPD and Congestive Heart Failure (CHF) patients.

    “It all started with our CEO finally crossing his pain threshold for what he could tolerate when it came to trying to manage and mine data in Excel, having everyone working from the same page,” Reff explained. “From there it was a very organic process. I read tons of literature, including Gartner’s magic quadrant reports, and other information online to see what other organizations were doing. I investigated quite a few vendors, narrowed it down to six, and asked for RFPs.”

    After several months, the hospital found a vendor that met its needs within the hospital’s budget and price range. Reff said the analytics programme has helped to reduce readmissions substantially. “In the past, 30-day readmissions were reviewed and reported quarterly due to the time-intensive process required to gather the information needed for analysis,” Reff said. “Now the information is reviewed and considered on a monthly basis by the case management team to identify opportunities to reduce readmissions. Additionally, since readmission information is now available to case managers in real time, interventions can be implemented in a more timely manner.”

    African countries with tight budgets can learn from Union General Hospital. Implementing an analytics programme can be modest investment with more chance of sustainability. Good research into what’s available and a good understanding of needs and budgets, it’s possible to set up an affordable service that delivers real value. It should find a place in Africa’s eHealth strategies.

  • Analytics can reduce ER waiting times and reduce costs

    Sometimes, waiting in overcrowded and busy ERs can be tedious and frustrating. A report in the Manufacturing & Service Operations Management from the Columbia Business School proposes an algorithm to predict demand trends, improve resource allocation, efficiency, reduce overcrowding and shorten waiting times. The team found that their simulations could reduce delays by up to 15% and may limit the need to divert patients to other services. Africa’s ERs always full, so the algorithm may help them tackle the problem.

    Because ER attenders can arrive in several modes and at various times of day, and across different weekdays, the algorithms have to be sophisticated and regularly validated. The report is seen by the team as an important first step in understanding how to reduce ER waiting times by making admission decisions that rely on predictions of arrivals.

    A big challenge’s knowing how to use analytic tools to manage ER demand and improve patient flow. Another’s knowing when and how predictive information can be used. A requirement’s that sufficient future information’s needed to maximise the algorithm’s benefits, so Africa’s health systems will have to ensure that this is in place before introducing an analytic model. Another constraint’s that the USA model includes opportunities to divert some patients to primary care. This option may be limited in many of Africa’s ERs.

    These constraints shouldn’t discourage healthcare and eHealth leaders from adopting analytics to help to manage ER demand. It can provide an effective entry into a modern predictive analytics across all healthcare.

  • Is Moneyball an analytics model for Africa?

    Michael Lewis’s book Moneyball: The Art of Winning an Unfair Game was published in 2003. It describes how Billy Beane, manager of Oakland Athletics baseball team used the performance statistics of each US baseball player to develop a transfer strategy that set the team on a rapid transition up the league.

    Now, there’s an eBook on how to do it in healthcare. Moneyball analytics Connecting and leveraging the best data across the health care continuum. Published by Optum, a clinical analytics provider, it sets out an approach to analytics and Big Data in healthcare, and includes:

    1. An introduction to healthcare’s Moneyball and Big Data to improving healthcare
    2. Big Data’s fundamentals of big data in healthcare, such as claims data. EMRs, ta he longstanding concept of  Garbage In, Garbage Out (GIGO)
    3. Analytics, both comparative and predictive
    4. Leveraging analytics, so starting with available knowledge, focusing on one major, chronic condition, creating a blueprint  and creating high-value care for high-risk patients
    5. Driving organisational change with leadership, cultural shift and aligning teams with new goals
    6. Dealing with a continuous journey of population health management , illness prevention and wellness
    7. Starting up. 

    Its starting point is that Big Data doesn’t necessarily mean good data. For Africa, it also means enough data. It worked at Oakland because baseball, like cricket and as football’s becoming with OptaPro, has a plentiful supply of statistics about players. Health and healthcare’s much more complex, and in Africa, has less data.

    Success with Big Data needs data extracting, validating, transforming and normalising. After this, once it’s in a usable form, it needs analytics so it can be scrutinised and understood. This’s an important proposition in the eBook. Africa’s not flush with these skills. They need sustained investment in data, recruitment, retention and healthcare transformation. Most of Africa’s eHealth strategy we’re prepared when Big data and analytics was hardly emerging. They need resetting to take advantage of the new opportunities

    The eBook’s realistic about this. It’s explicit that “Becoming a data-driven health care provider doesn’t happen overnight. While most large provider organizations have some background in financial analytics, clinical analytics are another story.” It’s clear that using analytics to identify and compare health and healthcare patterns care’s outside many hospital operations leaders’ comfort zones. Comparing clinical protocols and predicting hospital admission seems even more imaginative. Some leaders will need convincing to give up existing ways of working.

    Monyeball in healthcare’s like the classic journey of a thousand miles envisaged by Laozi, the ancient Chinese philosopher and writer. He also said “A tower nine stories high is built from a small heap of earth.”  Both are onerous and achievable.

  • Need for analytics to improve patient care

    Transforming and improving healthcare is always challenging. A core cause of difficulty is patients accessing healthcare at several locations, and sometime from different providers. This normal healthcare behaviour isn’t matched by patients eHealth data moving easily with them. Limitations to data mobility are due to factors such as poor, or no, connectivity and security weaknesses. A guide by Axway, available from HealthIT Analytics, describes good practices that can help.

    It proposes five information and healthcare steps to follow in sequence:

    1. Collect: create digital footprints in EHRs to establish a single information source of patients and communities and ensure delivery of high quality healthcare
    2. Unlock: enable data to move information seamlessly and effortlessly across disparate systems so patients’ data available across the healthcare system where it’s needed most
    3. Exchange: provide information about healthcare to all online and mobile communication channels to improve patient engagement to support value-based healthcare and improve the health of populations
    4. Engage: leverage tools for patient engagement for convenient and meaningful engagement anytime and anywhere
    5. Optimise: leverage health data value using analytics and actionable intelligence to drive efficiency and better patient outcomes.

    There’s also a set of enablers needed for the five steps to succeed:

    1. Robust health ICT infrastructure including secure and agile data integration functions enabling interoperability across devices, departments and types of data and timely access by users
    2. Collaboration for healthcare to derive meaningful value from the volumes of data organisations are collecting, so secure, compliant ways to expose data locked in internal systems for patients’ benefits
    3. Extant and emerging health ICT standards and Application Program Interfaces (API) so internal and external systems can integrate
    4. Flexible and robust health IT infrastructure that can connect numerous external parties and internal systems and provide visibility, alerting, and issue-resolution capabilities that support high efficiency across the entire healthcare organisations
    5. Removal of technical, regulatory and financial barriers so the right information’s available to the right people at the right time.

    As Africa’s health systems update their eHealth strategies, these steps and enablers can be incorporated to advance the role of health analytics. Affordability limitations will determine the pace and scale of change, so a long-term context’s needed to create an analytic environment that can move Africa’s eHealth along.






  • Prevention’s better than cure at Kaiser

    Most health systems allocate a tiny proportion of their resources on prevention compared to curative services. A small team from Kaiser Permanente Colorado Institute for Health Research has shown that analytics can add considerable value to that small resource. Its report, Applying Sequential Analytic Methods to Self-Reported Information to Anticipate Care Needs, published in Generating Evidence & Methods to improve patient outcomes (eGEMs) describes how it set out to identify care needs for newly enrolled and insured members in heterogeneous groups.

    Using a retrospective cohort investigation of 6,047 people who answered ten questions as a Brief Health Questionnaire (BHQ), the team developed a predictive model for the top 25% services by cost, then applied cluster analytic techniques to identify different high cost sub-populations. Each member’s monthly cost from six to twelve months following BHQ response was then quantified. The result was that BHQ responses are significantly predictive of high cost care, including self-reported health status, functional limitations, medication use, presence of up to four chronic conditions, self-reported Emergency Department (ED) attendance in the previous year. Age, gender, and deductible-based insurance product were predictive too.

    The largest possible range of predicted probabilities of being in the top 25% of cost was 3.5% to 96.4%. In the top cost quartile, potentially actionable patient clusters included those with:

    1. High morbidity
    2. Prior utilisation
    3. Depression risk
    4. Financial constraints
    5. High morbidity
    6. Previously uninsured with few financial constraints
    7. Relatively healthy, previously insured individuals with medication needs.

    Developing an approach like this by using sequential predictive modelling and cluster analytics to patients’ reported information may offer Africa’s health systems a step forward in its population health management endeavours. To succeed, health systems need to know more about their patients and health services, such specialty and sub-specialty costs and some patient costs. Adopting the BHQ and expanding human capacities in costing and analytics are where to start.