Keywords
Data technology - data-driven culture - data analytics - data governance
Introduction
In July 2013, FierceBigData[1] published a very interesting paper entitled “The datafication of everything - even
toilets”. The attractiveness of the paper does not lie in the catchy title but in
the increasing pervasiveness of the Internet of Things (IOT) and how in recent years,
IOT has contributed to our increasing appreciation of data as a valued asset and the
understanding of data science in shaping the way we live, work, and play. The whole
world is collecting data - lots and lots of data. In the example given in the FierceBigData
paper, we are beginning to see how even the toilet seat in our daily milieu of living
has the potential of becoming a connected device. If you sit on the toilet seat, your
weight can be measured. Insert your arms through the cuffs of the armrest and your
blood pressures can be taken. The intelligent toilet seat of the future will have
the ability to analyse your liquid and solid discharges for various disease markers.
Smart toilets are also making debuts all over the world. In these smart toilets, toilet
roll holders can notify the cleaners using WiFi technology when toilet papers are
about to run out so that toilet users will not be caught in an embarrassing situation
of not having toilet papers to clean up after a session. Smart toilets can also count
and collect data on the number of flushes to produce a KPI (Key Performance Indictor)
on usage. Sensors (thankfully not cameras) can detect non-motion or falls especially
for toilets designated for the handicap and the elderly.
Impact of Information Technology
Impact of Information Technology
The International Medical Informatics Association (IMIA) and health informatics have
come a long way over the past fifty years. At MEDINFO 2017 held in August 21 -25 2017
in Hangzhou, the People's Republic of China, IMIA celebrated the 50th anniversary of its founding in 1967. The occasion provided a poignant reminder of
how information technology has helped to develop and propel the development of health
informatics over the last half century. At the time of IMIA's founding, the healthcare
community was just making early attempts to use computers to process large amounts
of data collected from hospital billings and health insurance claims. One of these
earliest systems was TMIS (Technicon Management Information Systems), a collaboration
between Lockheed and the El Camino Hospital in California in the 1960s[2]. The early attempts of using information technology in healthcare to process data
were purely administrative. These were followed some years later by attempts to evolve
from these administrative, hospital-centric systems, to patient-centric systems integrating
clinical components into administrative and financial systems. Some of the early efforts
included the work of Larry Weed on the Problem-Oriented Medical Record (POMR) at the
University of Vermont in the early 1970s[3] and the work of Homer Warner on the HELP System at the LDS Hospital in Utah at around
the same time[4].
Since then, health informatics has made tremendous progress, largely riding on the
wave of advances in information and communications technology (ICT). Progress is evident
from the current state-of-the-art in the development of electronic medical and health
records, use of standards for messaging, terminology, disease coding and knowledge
representation, remote delivery of healthcare through the use of telehealth and telemedicine,
development of autonomous and non-autonomous decision support systems in patient care,
advances in digital imaging, storage and retrieval within and across the healthcare
enterprise, and, more recently, use of mobile devices for patient-centric care.
The Era of Data Technology
The Era of Data Technology
The increasingly pervasive use of ICT in healthcare has now started to spur advances
in information technology as hospital information systems, electronic health records,
mobile devices, social media and IOT are making an impact on the generation of data
- big data. The Chinese saying, 
(“the rear waves of the Yangtze River push the waves in front forward”) aptly describes
how the success of the ICT era has ushered and propelled the development of the Data
Technology (DT) era.
No other industry probably fits the definition of Big Data better than the healthcare
industry in terms of volume, variety, and velocity[5]. Every day, the world generates about 2.5 quintillion (1018) bytes of data of which healthcare accounts for a sizeable share. In terms of variety,
the code set in the base classification of ICD-10 allows for more than 14,400 different
codes. ICD-10 CM ramps up the code set to more than 14,400 different codes. SNOMED
CT contains more than 311,000 active concepts and provides the core general terminology
for the electronic health record. There are an estimated 30,000 human diseases known
to medicine. The number of investigation procedures, laboratory investigations, and
drug-use in healthcare are just mind-boggling. Anyone, even lacking an intimate knowledge
of healthcare, will not fail to realize that the velocity of data in the field is
phenomenal. Hospitals and clinics all over the world have to attend to millions of
patients requiring care one way or another.
Jack Ma, the founder of the Ali Baba Group, and an “evangelist” for DT, gave a very
interesting opening address at CeBit in Hannover, Germany, in March 2015. In his landmark
speech, Jack Ma declared that “the era of Data Technology is here and in the society
of tomorrow, data will be the most important means of production, innovation, social
development to improve lives, productivity and even to increase profit”[6].
Today, we have no shortage of examples to show how very successful businesses have
leveraged on data to succeed. Facebook has the world's biggest and most up-to-date
registration office (and this is done without legal regulation). Pixar has won 16
Oscars (and has never employed an actor/actress). Airbnb is the world's biggest hotel
chain (and does not own a single room). Uber is the world's largest cab operator (and
it does not operate a single car). Amazon is the world's biggest book store (and does
not have a single bookshelf). All of these companies have one success factor in common
- data[7].
The strategic importance of a data-driven culture has been clearly expounded in a
paper in the Economist entitled “Fostering a data-driven culture”[8]. The paper was based on the results of a survey done by the Economic Intelligence
Unit (EIU) in 2012 on 530 senior executives from around the world. The results of
the survey clearly demonstrated a clear link between the use of data and financial
performance of the surveyed companies. Financially successful companies have found
that the promotion of data-sharing across their rank-and-file has helped to generate
a data-driven culture and recognize the importance of data as a corporate asset.
It is therefore timely for healthcare organizations and institutions to value the
data that they collect and leverage on their data to gain business competitive advantage,
improve operational workflow and efficiency, and provide better quality and safer
patient care. For example, Geisinger Health System[9], one of the largest health care systems in Pennsylvania, USA, has, in recent years,
moved beyond one-off projects to recognize and adopt data-driven care as an enterprise-wide
strategy.
In order to leverage on data to achieve business competitiveness, organizations and
institutions must have access to good quality data and not just big data. Checks must
be in place to ensure that data quality is good. There have been numerous cases in
which the analyses of poor quality big data have resulted in spurious results. Indeed,
papers have been published questioning the robustness and quality of medical data
collected from IT systems. The paper by Wagner and Hogan[10] questioned the accuracy of medication data from the electronic medical record (EMR)
system of a geriatric centre in the USA. Analyses of just the medication records showed
significant levels of data error.
Not long ago, Scientific American published an article on “Why we must keep track of errors in Electronic Medical Records”[11]. It noted that “for all their promise, however, electronic medical records have
their own flaws. In dozens of known cases, caregivers have entered information into
the wrong chart or listed important details — such as drug dosages — incorrectly.
Data sometimes disappear. In one case, a patient's allergy to penicillin was improperly
entered into an electronic record. The patient later received ampicillin and nearly
died of shock”. The Scientific American article advocated the setting up of a national database for electronic record errors
as learning from mistakes can help to save lives. Proper tracking can only help. Robust
data, especially in healthcare, are obviously more important than big data because
it is the value and not the size of the data that counts.
Demand for Data Analytic Skills
Demand for Data Analytic Skills
With the increasing availability of data, it is important that healthcare organizations
and institutions know how to leverage on the data collected. The adoption of a strategic
data-driven culture within the organization or institution calls for, among other
things, the possession of in-house data analytic skill sets for the rank-and-file
- from the C-Suite (CEO, CIO, CMIO, CFO). Everyone within the organization or institution
must be mindful of the importance of data - the instruments for data collection, the
kind of data collected, whether the collected data are properly defined and documented,
guidelines for inventory, classification, storage, and utilization etc. These are
part and parcel of good data governance within the organization which should identify,
define, and enforce processes, policies, standards, and technologies that are required
to manage and ensure the availability, accessibility, quality, consistency, auditability,
and security of data within the organization.
As not everyone is comfortable with looking at data, there has been rapid development
of interactive data visualization techniques in recent years. Commercial dashboard
fabrication to aid decision-making has become a profitable industry. Actually, the
great Chinese sage Confucius saw the importance of data visualization as early as
504 B.C. “A picture is worth a thousand words” is an adage accredited to Confucius.
Today, a spatial map can readily be visualized by rendering several million data points.
One such study is demonstrated by the project to visualize a million geocodes from
taxicab pickup and drop-off locations in New York[12]. The study was intended as a stress test on the R package ggplot2 to render one
million data points. The analytic value of rendering and analysing the taxicab traffic
data is to be able to identify busy hours of traffic in the city and make necessary
plans for maintenance schedule, traffic engineering, and urban traffic planning.
The classic work of Dr. John Snow in describing the massive cholera outbreak that
ravaged the Soho area of London in the 1850s amply demonstrated the value of data
visualization in healthcare. By mapping the outbreak sites (data points), Snow was
able to trace the source of the outbreak to the Broad Street pump. On ordering the
removal of the pump handle so that Londoners could no longer draw water contaminated
with Vibrio cholerae from the River Thames, the outbreak was promptly arrested[13].
The Time is Now
So, as the era of data technology dawns on us, there will be, increasingly, a whole
lot of data out there in the healthcare ecosystem but there is not a whole lot of
people who knows how to deal with them! There is already a worldwide shortfall of
data scientists, leading to the proclamation by some that “the data scientist is the
sexiest job of the 21st century”[14].
It is therefore timely to recognize the strategic importance of leveraging on data
to advance healthcare and the time is now. And the time is now for IMIA as an international
organization in health informatics to take that lead. Already, IMIA has a Working
Group on Data Mining and Big Data Analysis. The time is now for IMIA to broaden the
terms of reference of this Working Group to incorporate other aspects of data-driven
healthcare and rename it as IMIA Working Group on Data-Driven Healthcare. The re-organized
Working Group should focus on:
-
Guidelines on data governance in healthcare;
-
Data analytics with special reference to healthcare;
-
Big data tools and solutions (the old Big Data and Data-Mining)
-
Bridging information technology and data technology;
-
Data quality issues and challenges.
With a timely revamp of the Working Group on Data Mining and Big Data Analysis to
a Working Group on Data-Driven Healthcare, IMIA should be well poised to provide the
international leadership in the era of data technology in the field of healthcare.
