PRE-DIABETES
An Informatics Research Agenda
Barbara M. Hayes
School of Informatics, Indiana University, 535 W. Michigan Street, Indianapolis, IN, USA
William Aspray
School of Informatics, Indiana University, 901 E. 10
th
Street, Bloomington, IN, USA
Keywords: Pre-diabetes, informatics research, public health and community informatics, knowledge dissemination,
management alerts, information alerts, decision support, clinical guidelines, health literacy, technological
literacy, patient feedback systems, pharmacy feedback systems, laboratory feedback systems, interface
design, reminder systems, information quality, consumer informatics, security, privacy.
Abstract: The paper sets out a research agenda for practitioners of the relatively new, interdisciplinary field of
informatics who wish to improve the health experience for people who have susceptibility to diabetes – a
condition known as pre-diabetes. Using information technology tools and methods, but with sensitivity to
the social and organizational complexities of the health care system, the article suggests addressing a set of
problems that will improve the lives of patients and their friends and families, as well as making the
provision of pre-diabetes care more effective and cost-efficient. Topics include public health and
community informatics, knowledge dissemination, information alerts, decision support, clinical guidelines,
health literacy, patient, pharmacy, and laboratory feedback systems, interface design, reminder systems,
consumer informatics, and privacy and security issues.
1 INTRODUCTION
The healthcare industry is one of the last to heed the
call of the information revolution. Its relationship
with information technology is deeply conflicted.
Healthcare leaders recognize that transformational
IT will decrease the risk of many common errors,
streamline workflow and, in some instances, save
money. Advanced computational techniques may
uncover genetic predispositions to disease and
identify new and more targeted pharmacologic
agents. Despite these promises, the healthcare
industry faces tremendous challenges as it integrates
information technology into healthcare delivery. The
cost of the transformation is high. Patient privacy is
in jeopardy. And, by its very nature, IT threatens to
disrupt the treasured, traditional relationship
between physician and patient.
Healthcare is shaped strongly by the interaction
of human need, economics, social demographics,
and the complex organization of the healthcare
community. These interacting variables make
healthcare a subject well suited to study by the
relatively new academic discipline of social and
organizational informatics. This field of study,
which is pursued in approximately twenty
universities in the United States and a few other
universities (including the University of Edinburgh,
City University in London, National University of
Singapore, Singapore Management University, and
the IT University of Copenhagen) is not to be
confused with the similarly named programs in
Europe that are focused primarily on computer
hardware and software. The practitioners of this new
informatics discipline are faculty members with an
interdisciplinary mindset, a social science
methodology, and a familiarity with IT and its
applications.
It is difficult for a healthcare practitioner to find
out about research in social and organizational
informatics not only because the literature is
scattered, but also because it often appears in places
where the healthcare community might not typically
look. Nevertheless, healthcare information system
17
Hayes B. and Aspray W. (2008).
PRE-DIABETES - An Informatics Research Agenda.
In Proceedings of the First International Conference on Health Informatics, pages 17-27
Copyright
c
SciTePress
designers, healthcare policy makers, Medical and
Nursing Informatics researchers, and teachers need
to know about the discipline because it can improve
technological solutions in healthcare and quality of
life for patients. Since IT development and
deployment are becoming important adjuncts in the
treatment of chronic diseases such as diabetes,
awareness of social and organizational informatics
research is especially important to those who
develop, deploy, and assess those technologies.
Examples of these technologies include e-mail and
Internet-based support, consumer-centered personal
electronic health records, home monitoring systems,
telemedicine, decision support aids, and online
interventions.
Healthcare insurers and providers have made
substantial investments in IT in order to make their
care more effective and cost-efficient. Less attention
has been given to using IT to improve the lives of
patients and their families. This paper uses one stage
of a widespread and expensive illness, Type 2
diabetes, as a means of examining ways in which IT
can be used to improve the lives of patients when
social and organizational factors are considered in
the design and delivery of care.
Pre-diabetes lends itself to social and
organizational informatics study because tools for
managing that syndrome may be found in several
areas of IT: public health and community
informatics, knowledge dissemination and
management, decision support, health literacy and
technological literacy, feedback systems, interface
design, information quality, consumer informatics,
and security. There is a particular need to
disseminate research on IT design and management
that takes into full consideration the way IT affects
individuals and organizations. IT designers often
lack an understanding of the environments in which
their work will be deployed, particularly in the
multifaceted world of healthcare.
In an ongoing research program, the authors are
looking at the social and organizational informatics
issues related to every stage of diabetes, from the
public health issues, to the diagnosis of the disease,
to the self-care issues that face most diabetes
patients as they live with the disease, to
complications of the disease such as loss of eyesight
or heart or kidney problems, to end-of life issues for
the diabetes patient.
We have chosen pre-diabetes, a syndrome
associated with Type 2 diabetes, as the focus of this
paper. Individuals with pre-diabetes have blood
glucose levels that are higher than normal, but not
high enough to qualify for a diagnosis of diabetes.
The American Diabetes Association (ADA) now
estimates that there are 54 million people in the
United States who have pre-diabetes (American
Diabetes Association, 2007c) We will examine
several information and IT challenges associated
with identifying pre-diabetes to allow informatics
researchers who are unfamiliar with health care to
“witness” the social and organizational factors in the
ebb and flow of information around this syndrome.
To that end, the first part of this paper is
organized in sections, each of which is preceded by
a question. The body of the section then provides
information about the topic, some examples of how
the question has been addressed, and some examples
of current challenges to date. The authors hope that
this format will stimulate informatics researchers to
create innovative research agendas that can provide
ever-improving answers to these critical questions
relating information to pre-diabetes. In the last
portion of the paper, we offer our own suggestions
for new research.
2 PUBLIC HEALTH AND
COMMUNITY INFORMATICS
How do researchers educate and persuade the public
to act on important new information about a syndrome
-- in this case, pre-diabetes, in which higher than
normal glucose levels and insulin resistance are
present but do not qualify for a diagnosis of diabetes?
New information about diabetes is frequently
incorporated into the medical literature. The general
public may read about new medical studies in the
newspaper, hear about them on the evening news or
encounter them on web sites, but the vast majority of
those studies are of interest only to the provider
community and then only as background
information. Occasionally, however, a major shift in
thinking occurs. In 2002, the Department of Health
and Human Services and the ADA issued position
statements to the press on two conditions linked to
an increased risk for developing diabetes. The term
pre-diabetes was used to describe these conditions.
Patients with pre-diabetes have either impaired
fasting glucose (IFG) or impaired glucose tolerance
(IGT) (American Diabetes Association, 2007b).
Research shows that some long term cardiac and
circulatory damage may already be occurring during
pre-diabetes (American Diabetes Association,
2007c). While healthcare providers had known about
IFG and IGT for some time, the results of three
major randomized controlled studies in different
countries with different populations had concluded
HEALTHINF 2008 - International Conference on Health Informatics
18
at nearly the same time that individuals could
prevent or delay diabetes with changes in diet and
exercise (Narayan, Imperatore, Benjamin, &
Engelgau, 2002). On that basis, the ADA
recommended screening overweight people 45 years
of age or older to detect those with impaired glucose
tolerance or impaired fasting glucose (American
Diabetes Association & National Institute of
Diabetes and Digestive and Kidney Diseases, 2003).
Those with pre-diabetes became candidates for
diabetes prevention interventions.
A risk test for pre-diabetes is available on the
Association’s web site at
http://www.diabetes.org/risk-test.jsp (American
Diabetes Association, 2007a). People with pre-
diabetes are slowly becoming insulin resistant.
Medications exist to reduce insulin resistance, but
more emphasis is put on weight loss, healthy diet,
and exercise, which also reduce insulin resistance
(Diabetes Prevention Research Group, 2002).
The identification of a new condition,
syndrome or infectious agent triggers an effort to
educate the public about (1) the existence of the
condition (2) its symptoms (3) screening tools (4)
treatments, if they exist and (5) prevention, if
prevention is possible. The identification and
“naming” of pre-diabetes signalled a shift from
identifying people with diabetes to identifying
people with pre-diabetes. The public policy
ramifications of such a shift are substantial. Every
time a new syndrome is identified and a
recommendation is made for screening, new costs
are added to the nation’s healthcare bill.
The ADA and the National Institute of Diabetes
and Digestive and Kidney Diseases (NIDDK)
discussed five conditions that should be met before
attempting to prevent a disease. They are (1) that
the disease to be prevented is an important public
health problem and affects a significant population
(2) that the early history of the disease is understood
well enough to measure its progression (3) that safe,
predictable, acceptable tests exist to identify the pre-
disease state (4) that safe and reliable methods exist
to help prevent or delay the disease and (5) that it is
cost effective to find individuals at high risk and
treat them. The ADA and NIDDK argued that all
five conditions had been met by cumulative research
(Diabetes Prevention Research Group, 2002).
National and state public health agencies, along
with diabetes advocacy groups, began to
communicate the shift in emphasis and information
to the public. They employed traditional media
(television, radio and print) and websites to engage
the public in learning about this pre-cursor to
diabetes. The shift was important because it
conveyed hope. Lifestyle changes really do reduce
risk. For example, the Diabetes Prevention Program
study concluded that individuals with pre-diabetes
who lost 5% to 10% of their total body weight and
exercised could lower their risk substantially (58%)
(Diabetes Prevention Research Group, 2002)
3 KNOWLEDGE
DISSEMINATION AND
MANAGEMENT/INFORMATION
ALERTS
How do researchers undertake informing an entire
community of healthcare providers who are already in
practice about a new diagnostic entity and the
appropriate screening and treatment of that entity?
When the discussion about screening for pre-
diabetes intensified, multiple avenues were already
in place to educate providers who routinely came
into contact with diabetic patients. Providers were
asked to screen individuals (1) who are overweight
and at least 45 years of age and (2) who are under 45
and have one of several other risk factors, such as
membership in high risk ethnic groups, high blood
pressure, a close relative with diabetes, and others
factors. Lifestyle recommendations were fairly
simple: increase physical activity and achieve
weight loss.
In the United States, primary care physicians
(usually internal medicine specialists or family
physicians) see the majority of patients before a
diagnosis of diabetes is established. All physicians
have a sizeable information burden, but primary care
physicians bear the additional burden of having to
stay aware of medical research in nearly every area
of medicine since they are the first line
diagnosticians. They need tools with which to
screen, organize, absorb, and implement the
substantial amounts of new medical information
created each year.
Physicians receive information from a variety
of resources, including colleagues, conferences,
medical journals available in print or on the Internet,
online texts and repositories such as UpToDate,
handheld decision support tools such as Epocrates,
web-based decision support tools such as Isabel,
online databases such as PubMed, and alerts from
Federal and state public health agencies. Many
physicians receive information from pharmaceutical
representatives who come to discuss products,
although there is concern about this practice in the
PRE-DIABETES - An Informatics Research Agenda
19
United States on the presumption that such
information is biased toward increasing sales.
State licensure boards and most professional
societies require physicians to complete a certain
number of mandatory continuing medical education
hours each year. In addition, most medical
specialties require re-certification. For example,
family practitioners recertifying through the
American Board of Family Medicine (ABFM)
interact with online clinical simulations that may
include content about diabetes.
There is often considerable lag time between
the introduction of new information or
recommendations and the subsequent formation of
clinical guidelines, the necessary adaptation of paper
or electronic record systems, and the adoption of
new practice patterns in the office and clinic. While
some physicians see patients in large group practices
or hospital settings with information infrastructures
that support the rapid transmission of information,
others are solo or small group practitioners with
severe time constraints and paper record keeping
systems that are not easily updated.
4 DECISION SUPPORT AND
CLINICAL GUIDELINES
How can new information be reviewed and inserted
into existing information systems to help providers
identify those patients who are at risk for developing
pre-diabetes
?
With the identification of a new diagnostic entity or
clinical recommendation, existing information
systems must be revised to prompt providers to
screen and treat those patients at risk. Procedures
must exist to vet new clinical information and
determine whether and how it will be integrated into
routine data collection, treatment activity, and
quality measures. As in the case of screening people
who may pre-diabetic, a general consensus about
what should occur emerged after the results of
several large clinical trials became public. In the
United States, how to implement those
recommendations is usually left to individual health
systems and individual physicians.
While all physicians make claim to expertise in
medical care, there will be “experts among experts”
(in this case, those with recognized expertise in
diabetes) who step forward to provide authoritative
guidance in implementing new information. Many
healthcare information systems, both paper and
electronic, include some component of expert
guidance. Clinical advisory committees assemble to
work with technical staff to approve changes for the
paper or electronic health record. These groups may
adopt existing clinical guidelines or develop their
own guidelines. Such guidelines always have cost
implications, so administrators and financial staff
may also be found on advisory committees. The
National Guideline Clearinghouse of the Agency for
Healthcare Research and Quality (AHRQ) provides
a searchable database of evidence-based guidelines
at http://www.guideline.gov/.
After new guidelines and procedures are agreed
upon, the designated experts must educate their
colleagues about new forms or procedures and
persuade them that new behavior is in order. That
often presents a challenge, since practicing
physicians are very busy and their practice patterns
are difficult to change. Large groups can provide
incentives to change by capturing compliance data
and feeding it back to providers. Solo practitioners
may find it easier to change their own behavior, but
face a greater burden of staying up to date on their
own.
5 HEALTH LITERACY AND
TECHNOLOGICAL LITERACY
Patients have different degrees of health literacy and
computer and technological literacy. If digital tools
become important ways to communicate with patient
and prompt them to action, what factors must be
considered in the design and delivery of those health
messages?
Health literacy has been defined in various ways.
Practically speaking, it is the ability of a patient to
take in health information, comprehend it, and take
appropriate action to protect and preserve his or her
health. In 2007, the Joint Commission published a
white paper on improving health literacy in order to
protect patient safety. It recommended raising
awareness across organizations of the impact of
health literacy and English proficiency on patient
safety and emphasizing patient-centered
communication across the entire span of care (Joint
Commission, 2007). The report noted that many
individuals with chronic medical conditions also
have low health literacy. It recommended specific
techniques, such as “teach back” and “show back”
techniques to assess patient understanding; limiting
information provided to two or three important
points at one time; using drawings, models and
devices; and giving patients information about all of
HEALTHINF 2008 - International Conference on Health Informatics
20
their diagnoses, medications, test result, and plans
for follow-up care. Studies have shown that diabetic
patients with low health literacy were less likely to
achieve good control over their blood sugars and
more likely to have vision impairment (Schillinger et
al., 2002).
Computer literacy or technological literacy is a
topic more often discussed in traditional academic
circles. There are active, ongoing debates in
education about what students need to learn about
computing, technology, and strategies for acquiring
information in the digital world in order to become
informed and effective adults. Providers, insurers,
and hospitals are interested in using digital tools to
communicate with patients because they present
opportunities to target information, reinforce it, and
reduce costs. However, patients have very different
degrees of computer or technological literacy as well
as health literacy. There will be no “one-size-fits-
all” disease information. It is far more likely that
patients’ comfort with technology will need to be
assessed in the same way that their health literacy
will need to be assessed, before they can simply be
referred, for example, to web-based education
materials or cell phone reminder systems.
6 PATIENT, PHARMACY, AND
LABORATORY FEEDBACK
SYSTEMS
How can IT systems develop patient, pharmacy and
laboratory feedback systems that might assist in the
treatment of patients with pre-diabetes?
Healthcare providers may prescribe smoking
cessation, diet, exercise, medications, and laboratory
tests for pre-diabetic patients. However, there is a
significant body of literature on non-adherence to
medical recommendations. Relatively few patients
are successful at making substantive change,
particularly when those changes involve ingrained
habits. In the case of weight loss, adherence to
recommendations can be measured by a decrease in
the patient’s weight, but it is still difficult to
determine if the patient is following a healthy diet,
exercising regularly, and using medication
appropriately.
There is great interest in extending
communication and a sense of connection between
the diabetes provider and the patient with pre-
diabetes or diabetes outside the walls of the exam
room. Web-based tools such as interactive risk
assessments, exercise diaries, and diet planners are
widely available for patients. Virtual health coaches
are being developed to help patients adhere to
medication and exercise recommendations. A few
providers and systems have established two-way
communication through the Internet on password-
protected sites so that risk scores and patients
exercise and food plans can be integrated into
reporting mechanisms that give feedback to
providers on patients’ actions. This type of
communication has not been universally
implemented for two reasons: (1) concerns about
protecting patient privacy and complying with
Health Insurance Portability and Accountability Act
(HIPAA) regulations and (2) concerns about
overwhelming the provider with information that he
or she will not have time to read. If a physician has
information and doesn’t act upon it, his or her legal
liability may increase.
There is also interest in mobile communication
devices for extending the relationship. Patients and
providers, who move from room to room, are both
mobile. The portability of cell phones and Personal
Digital Assistants (PDAs) offer many advantages.
For example, if the patient is recording diet choices,
he or she may record and transmit them via cell
phone application while at the dinner table rather
than waiting to access a desktop or laptop computer.
The assumption is that such information will be
more accurate since it is reported so close to the
event.
Compliance with medication is a particular
concern. In a recent study of heart attack patients,
researchers were surprised to find that one out of
eight patients quit taking important medications –
aspirin, beta blockers, and statins – within one
month of discharge from the hospital. Those same
patients were three times more likely to die during
the next year than those who adhered to
recommended medications (Ho et al., 2006).
Pharmacy benefits manager Medco Health Solutions
predicts that spending on diabetes therapies will
increase up to 20% each year between 2007 and
2009 (Medco Health Solutions, 2007). One recent
study found that physicians discussed cost,
coverage, or purchase logistics of medications in just
one-third of interviews when prescribing new
medications (Tarn et al., 2006). Many patients are
embarrassed to admit to their physician that they
cannot afford to pay for medication.
Physicians need to know whether patients are
actually filling prescriptions as directed or showing
up for laboratory tests. If the patient is filling
prescriptions at an in-house pharmacy (one owned
by the clinic, hospital, or health plan) that
information may be captured and fed back into the
PRE-DIABETES - An Informatics Research Agenda
21
record fairly easily. It will not be added to the record
easily if the patient fills the prescription at an
“outside” pharmacy. The physician’s only recourse
is asking the patient directly. Self-report is not the
best way to gather that information, since many
patients have difficulty facing a physician if they
have not complied entirely with his or her
recommendations. At present, the capacity to give
providers feedback on whether a patient actually
filled or refilled a prescription exists, but it exists
with some risk to the patient’s privacy.
It is somewhat easier to know if the patient has
completed a laboratory test. The convention is that
the results are sent directly to the physician. The
patient must ask the physician to interpret the test
results. The fact that a patient actually reported for a
laboratory test is most easily added to the record if
the patient is sent to a laboratory owned by the
clinic, hospital, or health system.
7 INTERFACE DESIGN ISSUES
AND REMINDER SYSTEMS
Can easy-to-use, inexpensive, reliable reminder
systems be devised that will assist patients with a
variety of lifestyles to remember office visits and
medication?
Patients who are pre-diabetic need regular visits to
assess their progress. If the initial treatment
interventions do not produce the desired weight loss
and increase in exercise, other recommendations
may be made. In most cases, results from routine
laboratory tests, which are private medical
information, will stay in the medical record until the
patient’s next visit. If the patient wants the results,
the patient needs to return to the office.
Physician offices employ a number of strategies
for reminding patients about the time of their next
visit. Most patients leave the office with a printed
reminder of the date of their next visit. Some offices
provide a telephone prompt a few days before the
visit, although such calls add to office overhead.
Some offices charge a fee for missed appointments
to motivate patients to keep appointments.
The Veterans Administration and a few private
insurers are beginning to “push” information out to
patients on their appointment times and lab tests
(Ferris, 2007). Some private medical providers and
facilities also see the Internet as a way for patients to
view their information on line. Authentication
technologies will make it easier to assure that only
the patient can see his or her own personal medical
information.
Information may be captured about whether a
patient filled a prescription, but it will still be
difficult for physicians to assess if patients are
actually taking medication or using it as prescribed.
One company has produced a pill bottle that uses
Short Message Service (SMS) to track how often
pills are taken and send a reminder to the patient’s
phone if a dose is missed (www.simpill.com).
8 INFORMATION QUALITY AND
CONSUMER INFORMATICS
What tools can be developed to assure consumers that
they are accessing the highest quality health
information as they interact with the Internet and other
information resources? How can search engine results
lead patients to reputable information? How can
patients avoid “health mythology” propagated by
participatory tools such as chat rooms and blogs that
may transmit information with little basis in fact?
The Internet affords patients with access to a
computer an incredible number of tools with which
to research their risk factors and conditions. The
amount of information can be overwhelming. On a
single day in April 2007, a Google search using the
word “diabetes” returned 92,500,000 “hits.” Yahoo
returned 78,900,000 results. Microsoft’s search
engine, MSN, returned 18,462,447 results.
An increasing number of U.S. citizens research
their medical conditions online. The major search
companies are well aware of that fact. Steve Case,
the founder of AOL, has launched Revolution
Health, a health web site that will coach subscribers
on their health, store their health information, match
them with doctors, and help them with insurance
claims (Freudenheim, 2007). Google introduced a
health information subscription service in 2006
(Modern Healthcare, 2006). Microsoft bought a
health information search engine in 2007 (Lohr,
2007). Most of the major search engines are actively
engaged in a race to produce more relevant, focused
results. Google, Microsoft, and Yahoo all have test
sites that collect and display large amounts of
information in intuitive ways. Some return
definitions first and then categorize results. For the
search term “diabetes,” one search engine grouped
clusters of information into these categories: care,
research, management, control, risk, centers, types,
drugs, and supplies. Video search engine
HEALTHINF 2008 - International Conference on Health Informatics
22
Blinkx.com retrieved over 14,000 videos with
diabetes content.
Unfortunately, there is no consensus on how to
evaluate the quality of the information cited
(Eysenbach, Powell, Kuss, & Sa, 2002). Online
health information varies in quality; patients are
vulnerable to misinformation and fraud if they are
unable to evaluate the quality of the material
accessed. Information gained through participatory
tools such as chat rooms or blogs may be inaccurate
information, giving patients false hope or diverting
them from evidence based treatment. Websites may
expose them to worthless or even harmful diet pills
and exercise equipment for which unreasonable
claims are made.
Operational definitions of quality are still
needed, although rating tools are beginning to
emerge. The Health on the Net Foundation offers the
HONcode designation for health web sites that
follow its standards of quality. The Foundation has a
policing system that is designed to help developers
monitor their own compliance to the code, as well as
remain responsive to user concerns. The policing
procedures can be initiated by individual site users
or by the Foundation itself.
Reputable organizations try to provide quality
information on the Internet. Some of these include
the American Diabetes Association, The National
Diabetes Educational Program, The National
Diabetes Information Clearinghouse, and the
National Institute of Diabetes and Digestive and
Kidney Diseases.
There is a growing body of research on
information seeking behavior. Researchers are
beginning to build a set of tools and techniques with
which to examine patient interaction with healthcare
materials available on the Internet. Several studies
have shown that online health information has a
positive influence on patients’ ability to cope with
serious illness (Mills & Davidson, 2002; Ziebland et
al., 2004). Access to disease information online has
also been linked to reduced anxiety and increased
perceptions of self-efficacy (Ybarra & Suman,
2006). There is still much to be learned about search
strategies, information retrieval, demographic
differences, and subsequent actions over the course
of a chronic illness.
While information available on the Internet
has helped to equalize the power in physician-patient
relationships, it has two consequences that are less
positive. One is the annoyance some physicians feel
when patients question their judgment and
recommendations. The other is the time that must be
spent evaluating and responding to the patient’s
attempt to gather information and participate in his
or her healthcare.
9 SECURITY AND PRIVACY
How can systems be designed to secure patients’
confidential information? How can stigmatizing
information be kept private so that patients feel they
can confide in their physicians? How can patients be
persuaded to allow their treatment information to be
collected and be assured it will not be used to
penalize them at a later date, for example, by cutting
off care for those with pre-existing conditions?
In the past two years, there have been literally
millions of accidental and intentional breaches of
patient privacy through lost laptops, inadequate
storage procedures, and outright fraud. Some recent
news stories provide examples: hackers accessed
personal data for 14,000 Pentagon employees
through health insurance records (Pulliam, 2006);
the loss of 130,000 Aetna records by Aetna when
backup tapes were stolen in a burglary (Zeller Jr,
2006).
Few incentives exist to encourage insurers,
hospitals, and providers to tighten their security.
Although mandatory notification of data loss and
financial penalties are being considered by several
states, there are few real penalties for compromising
or losing medical information. Further, the Health
Information Portability and Accountability Act
(HIPAA), which was the impetus for spending
millions of dollars to redesign systems to achieve
compliance with governmental privacy directives,
has had little impact. “In the three years since
Americans gained federal protection for their
private medical information (through HIPAA), the
Bush administration has received thousands of
complaints alleging violations yet not imposed a
single civil fine and has prosecuted just two criminal
case”(Stein, 2006).
The Healthcare Information Management and
Systems Society and Phoenix Health Systems
published a survey on HIPAA compliance in the
summer of 2006. The survey noted that only 56% of
providers had implemented the security standards
and that a substantial portion of providers (22%) and
payers (13%) remained non-compliant with the
privacy regulations. The report further suggested
that even those who were compliant had significant
implementation gaps and that there may be a core
group of entities covered by the law that cannot or
will not implement the privacy standards at all
PRE-DIABETES - An Informatics Research Agenda
23
(Healthcare Information and Management Systems
and Society 2006).
Many people disagree about whether business
or government presents a bigger threat to privacy.
While it might be assumed that government-
sponsored healthcare programs have stricter privacy
standards, when data from millions of U.S. veterans
were contained on a stolen disk in May of 2006, the
VA waited two weeks before reporting the loss.
Several weeks went by before all the details of the
situation came to light (Stout & Zeller Jr., 2006).
Medical privacy is also important because some
diseases are stigmatizing. At different times in
history, different diseases have had greater or lesser
amounts of stigma attached to them. At one time,
cancer was a stigmatizing illness. Some people
thought cancer was contagious and that people who
had it were to be avoided. Others thought death was
inevitable and stopped visiting those afflicted
because it was “too depressing.” Even today,
individuals with lung cancer may be blamed for their
disease on the assumption that they must have been
smokers.
Diabetes has been relatively free of stigma,
although that is starting to change as the association
between obesity and diabetes becomes clearer. Many
people in the United States see the condition of
being overweight or obese as a sign of lack of
discipline or laziness. Corporate wellness programs
are beginning to reward people for losing weight and
maintaining the reduced weight.
Patients with pre-diabetes or diabetes have to
provide information to their physician in order to
receive treatment. Courts in the United States have
guarded doctor-patient privilege as essential to the
greater societal good. Many people find it difficult to
confide in a physician. In order to encourage open
and honest exchanges, patients have been assured
that information about their medical records would
be kept confidential. The assumption is that society
benefits when patients are treated, because the
potential spread of the disease and cost of its
treatment have been kept in check.
There are times when all the information and
power in the relationship resides with the physician.
An example is the treatment of an unconscious
patient in the emergency room. Society accords the
physician the responsibility of gathering information
and making decisions on the patient’s behalf. There
are also situations of shared information and power.
An example is an acute illness during which the
patient consults the physician but is unlikely to
argue or negotiate about the prescribed treatment,
such as a dosage of an antibiotic over a certain
number of days. In pre-diabetes, the power resides
with the patient and the physician is in a consulting
role. The physician, diabetes nurse, or dietician
recommends lifestyle management techniques, but
the patient has to implement them on a daily basis.
Unfortunately, healthcare providers are no
longer able to guarantee privacy when they act for
the unconscious patient, advise the acutely ill
patient, or consult with the chronically ill patient.
The worst case scenario is that the patient’s own
medical data cause his or her insurance company to
drop insurance coverage. Patients who have chronic
illness fear losing their healthcare coverage. Some
stay at jobs they dislike because they would lose
coverage for pre-existing conditions if they changed
jobs and had to obtain new insurance. If self-
employed, they run the risk that their insurance
company declares them “uninsurable” and
terminates their policy. If insured by government
program, they may lose access to the latest
treatments or experimental treatments.
In a very real sense, the information patients
give to their providers for treatment and their
insurance company for reimbursement may well be
used against them. In the United States, the courts do
not compel a defendant to testify against himself, but
healthcare information systems are used to do
exactly that. If patients choose to lie about their
conditions, they also run the risk of losing coverage.
At present, there is no way for healthcare providers
to guarantee that patients will not be penalized for
their honesty in providing information to the medical
record.
Of course, a patient is not required to use his or
her insurance benefits, but most cannot cover the
cost of treating a chronic illness without doing so. If
a patient wants to use insurance benefits, he or she
allow information about the diagnosis to be shared
with the insurance company. The insurer then
decides whether or not to reimburse for care based
on the terms of the policy. Insurance companies
battling rising healthcare costs may use data analysis
to limit access to care. They employ information
systems to control costs and increase profitability.
They review data to forestall unnecessary spending.
Some companies put burdensome procedures in
place, deny care, or insist on lesser care in the time
period before sustained evidence of efficacy can be
added to the medical research. For example, insulin
pumps became available in 1979. They are useful for
some patients, but expensive. Some insurance
companies would not cover them at first; others
required providers to provide written justification of
the need for an insulin pump.
HEALTHINF 2008 - International Conference on Health Informatics
24
10 CONCLUSIONS
Throughout the United States, medical and nursing
practitioners are rapidly becoming interested in
solving some of the information challenges
described here, as they exist specifically for pre-
diabetes and generally for every other disease and
syndrome. Healthcare professionals are integrating
technology into the everyday delivery of care. As
their familiarity and level of comfort increase, they
will seek IT support for their patients as well.
Clinicians, often led by early adopters of
technology, are commissioning IT applications from
commercial providers and academic research teams
to solve day to day medical problems. The
application developers are often unfamiliar with the
realities of healthcare. These applications tend to be
stand-alone; they rarely generalize well to wider use.
Individual public health, medical and nursing
researchers are seeking major grant funding for large
scale development of IT solutions to healthcare
conundrums. In “Toward an Informatics Research
Agenda: Key People and Organizational Issues,”
Kaplan et al. present a research agenda model that
addresses individual, institutional, trans-
organizational, and transnational concerns, aligning
them with the social science disciplines that may be
brought to bear on their exploration. Those
disciplines include cognitive psychology, social
psychology, sociology, and cultural anthropology
(Kaplan, Brennan, Dowling, Friedman, & Peel,
2001). In this context, and with these observations of
pre-diabetes, in Table 1 we suggest some areas of
productive research.
At present, many healthcare IT solutions fall
short of their intentions because patients and
providers do not respond to those solutions as
anticipated. Social and organizational factors are
often at the core of those unanticipated,
unsatisfactory responses. Money is being spent that
does not result in real human benefit. The emphasis
needs to shift from the construction of specific
technologies to human and organizational behavior
in interaction with those technologies. Informatics
researchers should lead the way toward
incorporating a respect for and expectation of social
science research in IT development.
Table 1: Pre-Diabetes Informatics Research.
Public health and community informatics
Identify effective tools, such as community dashboards, to educate individuals and communities about pre-
diabetes and the importance of life change styles to reduce the risk of diabetes; develop those tools in ways
that help communities set priorities for spending on such activities as screening.
Design and fund public health and community information systems that allow data about pre-diabetes to flow
between and among Federal, state, and local agencies, as well as advocacy groups, clinicians, and
individuals.
Explore and evaluate information gathering patterns in individual communities; design reliable, predictable
information pathways for publicizing new information in the domain of public health.
Find ways for important public health messages to rise above the “noise” of a media-saturated environment;
develop and place screening tools in the media that healthcare consumers already use, rather than trying to
train them to use new technologies; develop the ability to target individuals with personalized messages about
their specific risk for pre-diabetes.
Knowledge dissemination and management/information alerts
Identify tools for primary care physicians with which they can effectively screen, organize, absorb, and
implement the substantial amounts of new medical information created each year; consider ways to
effectively manage the information burden placed on busy clinicians.
Study search strategies and patterns of information seeking in domain experts such as physicians and nurses;
develop knowledge dissemination patterns that fit into the existing work flow rather than disrupting it.
Identify ways to reduce the lag time between the introduction of new information or recommendations about
pre-diabetes and diabetes, and the subsequent formation of clinical guidelines, the necessary adaptation of
paper or electronic record systems, and the adoption of new practice patterns in the office and clinic.
When presenting clinicians with new information, assist them with clear information on how to implement
screening and treatment recommendations at the point of care.
PRE-DIABETES - An Informatics Research Agenda
25
Decision support and clinical guidelines
Study existing processes for the institutional adoption of new standards of care; develop models for
identifying the social and organizational barriers that slow the implementation of new expert knowledge;
develop systems that will alert organizations and institutions to new evidence-based medical research and
assist them in implementing it, rather than relying solely on practitioners to come into contact with and
absorb that new knowledge on a clinician by clinician basis.
Identify effective ways to marshal current expert knowledge about diabetes and pre-diabetes, update existing
medical information systems to reflect that knowledge, and provide communication technologies, strategies,
and incentives for busy primary care physicians to pay attention to these recommended changes in practice.
Health literacy and technological literacy
Identify effective means to test the health literacy and technology literacy of pre-diabetes and diabetes
patients and match them with web sites and other communication means that are suited to their particular
health and technology literacy.
Develop rubrics for measuring whether best practices in promoting health literacy are being incorporated into
healthcare technologies.
Understand the ways in which healthcare consumers gather information about their health and the health of
their families, the ways they use technology to support that process, and the thresholds or decision points that
prompt them to take action, such as scheduling an appointment or attempting to change a habit or behavior.
Patient, pharmacy and laboratory feedback systems
Identify technologies and procedures that will extend communication and a sense of connection between the
diabetes provider and the patient with pre-diabetes or diabetes outside the walls of the exam room. These
might include, for example, web-based tools such as interactive risk assessments, exercise diaries, and diet
planners, virtual health coaches, or two-way communication systems using PDAs between patients and
providers that allow that patient to record and report data and ask questions.
Design systems to provide feedback to providers about how well patients are adhering to the prescribed
treatment (medications, diet, exercise, further tests or medical consultations); those systems should be
sensitive to patient privacy issues and avoid overwhelming providers with data that does not contribute to
decision-making or increases their legal liability.
Create and evaluate systems that optimize the capture of patient adherence data even if that data exists across
multiple organizations.
Interface design issues and reminder systems
Develop and evaluate treatment support systems that “follow” the patient into his or her work and home
environments; patients should be able to choose from among a number of support systems based on their
individual profiles and preferences.
Design inexpensive, reliable reminder systems that are patient-specific; avoid generic applications that
require the patient to wade through information or reminders that are not specific to his or her situation.
Information quality and consumer informatics
Develop tools to assure consumers that they are accessing the highest quality health information as they
interact with the Internet and other information resources.
Develop operational standards of quality of information and tools that allow both website developers and
users to rate the quality of their information on the site.
Redesign search engines to lead patients to reputable information and away from “health mythology” propagated
by participatory tools such as chat rooms and blogs that may transmit information with little basis in fact.
Security and privacy
Develop legal, economic, technological, and social means to increase the privacy of patients’ confidential
information; develop technologies and protocols that protect the trust and tradition of the doctor-patient relationship.
Design systems that ensure that stigmatizing information is kept private so that patients feel they can confide
in their physicians; allow treatment information can be collected without fear it will be used to penalize
patients at a later date, for example, by cutting off care for those with pre-existing conditions.
HEALTHINF 2008 - International Conference on Health Informatics
26
REFERENCES
American Diabetes Association. (2007a). Diabetes Risk
Test [Electronic Version] from
http://www.diabetes.org/risk-test.jsp.
American Diabetes Association. (2007b). How to Tell If
You Have Pre-Diabetes [Electronic Version] from
http://www.diabetes.org/pre-diabetes/pre-diabetes-
symptoms.jsp.
American Diabetes Association. (2007c). Pre-Diabetes
[Electronic Version]. Retrieved June 14, 2007 from
http://www.diabetes.org/pre-diabetes.jsp.
American Diabetes Association, & National Institute of
Diabetes and Digestive and Kidney Diseases. (2003).
The Prevention or Delay of Type 2 Diabetes. Diabetes
Care, 26 Supplement(1), S62-S69.
Diabetes Prevention Research Group. (2002). Reduction in
the evidence of type 2 diabetes with life-style
intervention or metformin. N Engl J Med, 346(6), 393-
403.
Eysenbach, G., Powell, J., Kuss, O., & Sa, E. (2002).
Empirical studies assessing the quality of health
information for consumers on the World Wide Web: a
systematic review. JAMA: Journal of the American
Medical Association, 287(20), 2691-2700, 2723-2694.
Ferris, N. (2007, January 24). Vets can see EMR data.
Government Health IT News.
Freudenheim, M. (2007, April 17, 2007). AOL Founder
Hopes to Build New Giant Among a Bevy of Health
Care Web Sites. New York Times.
Healthcare Information and Management Systems and
Society , P. H. S. (2006). US Healthcare Industry
HIPAA Compliance Survey Results: Summer 2006.
Ho, P. M., Spertus, J. A., Masoudi, F. A., Reid, K. J.,
Peterson, E. D., Magid, D. J., et al. (2006). Impact of
Medication Therapy Discontinuation on Mortality
After Myocardial Infarction. Arch Intern Med,
166(17), 1842-1847.
Joint Commission. (2007). "What Did the Doctor Say?":
Improving Health Literacy to Protect Patient Safety
(White Paper).
Kaplan, B., Brennan, P. F., Dowling, A. F., Friedman, C.
P., & Peel, V. (2001). Toward an Informatics Research
Agenda: Key People and Organizational Issues. J Am
Med Inform Assoc, 8(3), 235-241.
Lohr, S. (2007, February 27, 2007). Microsoft to Buy
Health Information Search Engine. New York Times.
Medco Health Solutions. (2007, May 17). Unmanaged,
diabetes drug costs could surge nearly 70 percent
within three years. Drug Trends Extra.
Mills, M. E., & Davidson, R. (2002). Cancer Patients'
Sources of Information: Use and Quality Issues
Psycho-Oncology, 11, 371–378.
Modern Healthcare. (2006). Google introduces health-info
subscription service [Electronic Version]. Retrieved
May 14.
Narayan, K. M. V. a. p., Imperatore, G. m. e., Benjamin,
S. M. e., & Engelgau, M. M. c. e. a. s. b. (2002).
Targeting people with pre-diabetes: Lifestyle
interventions should also be aimed at people with pre-
diabetes. BMJ, 325(7361), 403-404.
Pulliam, D. (2006, May 1). Hackers access personal
information on TRICARE servers. Government
Executive.
Schillinger, D., Grumbach, K., Piette, J., Wang, F.,
Osmond, D., Daher, C., et al. (2002). Association of
health literacy with diabetes outcomes. JAMA: Journal
of the American Medical Association, 288(4), 475-
482.
Stein, R. (2006, June 5). Medical Privacy Law Nets No
Fines. Washington Post.
Stout, D., & Zeller Jr., T. (2006, March 24). Agency
Delayed Reporting Theft of Veterans' Data. New York
Times.
Tarn, D. M., Paterniti, D. A., Heritage, J., Hays, R. D.,
Kravitz, R. L., & Wenger, N. S. (2006). Physician
communication about the cost and acquisition of
newly prescribed medications. American Journal of
Managed Care, 12(11), 657-664.
Ybarra, M. L., & Suman, M. (2006). Help seeking
behavior and the Internet: A national survey.
International Journal of Medical Informatics, 75(1),
29-41.
Zeller Jr, T. (2006, December 18). An Ominous
Milestone: 100 Million Data Leaks. New York Times.
Ziebland, S., Chapple, A., Dumelow, C., Evans, J.,
Prinjha, S., & Rozmovits, L. (2004). How the internet
affects patients' experience of cancer: a qualitative
study. BMJ, 328(7439), 564.
PRE-DIABETES - An Informatics Research Agenda
27