Center for Health Information and Decision Systems
Volume 1, Issue 2A
RESEARCH BRIEFING
EARLY FINDINGS ON PERSONAL HEALTH RECORDS
AND INDIVIDUAL USE
Ritu Agarwal, Professor and Robert H. Smith Dean’s
Chair of Information Systems and Director of CHIDS
Corey M. Angst, Research Assistant Professor and
Associate Director of CHIDS
The transformational power of information technology in
altering the nature of competition in an industry and
creating value for both firms and consumers has long
been acknowledged in diverse industry sectors such as
airlines, financial services, and retailing. A common
characteristic among industries that have experienced
such transformations is that they are information
intensive – i.e., a significant proportion of their valuecreation activities occurs through the storage,
processing, and analysis of data. The transformation
has typically been attributed to specific software
applications – e.g., the Sabre System in the airline
industry and Merrill Lynch’s Cash Management System
in the brokerage industry – that provide the trigger for
far-reaching changes.
Healthcare cost account for nearly 10% of GDP in most
industrialized nations and over 15% in the US.
Healthcare also represents a sector with significant
consequential outcomes – the quality of care delivered
often spells the difference between life and death. Thus,
it is not surprising that considerable emphasis by
governments, policy makers, and other stakeholders in
this sector is being given to trying to better understand
how the delivery of care can be improved. Human
capital in the form of knowledge and skills is doubtless a
critical input factor for the quality of healthcare; however,
to the extent that healthcare is an information intensive,
knowledge-based activity that requires high reliability in
operations, another important input is technology that
can aid in the “movement” of critical information.
In this briefing we explore an emergent IT application –
the electronic personal health record (PHR) that
arguably, offers the same potential for revolutionary,
discontinuous change in the healthcare sector. In very
general terms, a PHR is an electronic, universally
available, lifelong resource of health information that an
individual uses to make health decisions. It can be
Winter 2006
made available through the use of a web-based, ASP
(application service provider) model, a stand-alone PCbased platform, or a portable storage device such as a
USB-flash drive. The information in the PHR is owned
and managed by the patient. PHR features typically
include self-tracking and monitoring of health information
and self-entry or downloading of information related to
diagnoses,
medications,
laboratory
tests,
and
immunizations. The PHR sometimes has the ability to
receive and store information from a doctor's electronic
medical record (EMR) or other electronic data source.
Some PHRs include features that notify the user of drugto-drug interactions and dosage warnings.
We conducted an exploratory empirical study to
understand the characteristics of individuals who are
early adopters of PHR software. We also sought to
understand what type of value individuals perceive in the
PHR, and the drivers of their usage intentions. We
found three distinct perceived components of value: (1)
compliance, (2) connectedness, and (3) convenience.
We also found that different demographic and medical
condition variables influence the type of value a user
sees in a PHR. Finally, our results show that future
intentions to use a PHR are driven by perceptions that
the PHR provides compliance and improves a patient’s
relationship and connectedness with their healthcare
provider.
Defining the PHR
Figure 1 shows a screen shot of a typical PHR.
Figure 1. Screenshot of a PHR
CHIDS ▫ Van Munching Hall ▫ College Park, MD ▫ University of Maryland ▫ www.smith.umd.edu/chids ▫ chids@rhsmith.umd.edu ▫ ph. 301.405.0702
© 2006 Center for Health Information and Decision Systems CHIDS. CHIDS Research Briefings are published on a quarterly basis to update CHIDS members.
CHIDS Research Briefing, vol.1, issue 2A
A PHR is fundamentally a tool for end-users. It is helpful
to draw a connection between a PHR and a personal
financial management software package such as
®
Quicken . With Quicken, an individual can track her
finances by monitoring personal checks, logging ATM
transactions, and tracking deposits, to name but a few
features. Then, at the end of the month, the individual
can compare the previous-month’s transactions with
those on file at the defined bank. This process could be
labeled as dissociative, since both ‘systems’ operate in
isolation. On the other hand, Quicken offers a feature
that allows the user to continue to maintain her private
banking information on her personal computer, but also
gives the option of downloading data directly from the
participating bank as a means of reconciling the account.
The functionality offered by a typical PHR is summarized
in Table 1 below.
Information
• General personal information
• Family medical history
• Emergency contacts
• Personal medical contacts
• Insurance coverage(s)
• Reminders
Visits and Hospitalizations
• Conditions
• Tests
• Treatments
• Medicines
• Immunizations
• Miscellaneous data
Medical Details
• Active conditions
• Vitals and profiles (incl.
graphing)
• Test results
• Treatments received
• Drugs administered
• Immunizations
• Miscellaneous data
Communications
• Records of letters, calls,
faxes, emails
• Printable emergency card
• All screens generate into
reports
• All reports can be printed or
emailed
Medical Summary
• Conditions, problems
• Tests
• Immunizations
• Medications
• Treatments and therapies
• Other orders, recommendations
Optional: Sponsor
customization
• Introductory branded screens
and reports
• Custom pick-lists of PHR
information
• Dedicated web links
• Dedicated field-level links
• Customized profiles, guidelines
• Custom “Community of Care”
page that auto-builds to user’s
needs based on sponsorselected information
Page 2
Winter 2006
record systems. In most cases, information is only made
available for download to the patient’s personal record,
but in some systems, such as one run by Brigham and
Women’s Hospital in Boston, MA, patients are also given
the ability to upload notes and data into their permanent
medical record, schedule appointments, and email
doctors, to name a few features.
Early Adopters of PHRs
Our goal was to study what types of perceptions
individuals have about the use of PHRs, and what the
determinants of their intentions to use the PHR are.
Following prior theory, the conceptual model underlying
the study, shown in Figure 2, suggests that individual
characteristics of two varieties: demographics, and those
related specifically to the individual’s medical condition,
predict the value cognitions individuals possess in
regard to using the software. These cognitions, in turn,
drive their usage intentions. Additionally, we allow for
the possibility that individual characteristics exhibit direct
effects on usage intentions, over and above their
mediated influence via perceived value.
The demographic variables included in the study were
age, education, and income. The medical conditionrelated variables were, the existence of a chronic health
condition in the individual’s circle of care, whether or not
multiple medications are required for this chronic
condition, and the total number of doctor visits.
Demographics
o Age
o Income
o Education
Perceived Value of
Using a PHR (PV)
Behavioral
Intention to
Use a PHR
(BI)
Medical Condition
o Chronic Illness (ILLNESS)
o Multiple Medications (MED)
o Number of doctor visits (DOC)
Table 1. Typical PHR Functionality
A PHR is similar in this respect. A user can maintain all
health information on her personal computer, entering
data from doctor visits as they occur and tracking
medications and dosages. In this case, two separate
databases (in actuality, several databases are
maintained, as it is very common for people to see
multiple health providers and it is typical that each
maintains its own isolated medical database record) are
managed independently. There is, however, technology
available such that a user/patient can have a direct link
through the Internet or a portable device such as a USBflash drive to participating providers’ electronic health
Figure 2. Conceptual Model
Study Context and Sample
We mailed a survey to 813 purchasers of an electronic
personal health record. These 813 users represent
people that had purchased the software through the
company’s website, ordered it over the telephone, or
through a 3rd party distributor in the 3-month period just
prior to our study.
There were 47 unusable or
undeliverable surveys and 190 complete surveys,
representing a 24.8% response rate. In addition to
questions measuring the research variables in Figure 2,
CHIDS ▫ Van Munching Hall ▫ College Park, MD ▫ University of Maryland ▫ www.smith.umd.edu/chids ▫ chids@rhsmith.umd.edu ▫ ph. 301.405.0702
CHIDS Research Briefing, vol.1, issue 2A
Page 3
we also asked users what their most important
healthcare tasks were, and what the primary uses for a
PHR might be. A description of the sample is provided
in Table 2.
Description
Value
Surveys sent
813
Unusable or undeliverable
47
Usable surveys
190
Response Rate
24.8%
Male/Female
72/28
% of users with chronic illness
63%
Average number visits to doctor/yr
7.1
Average years of computer experience
Table 2: Sample Description
15.3
In prior work we found that individuals’ beliefs with
regard to the value of PHRs consisted of three distinct
dimensions: “compliance (PVa),” “connectedness with
one’s healthcare provider (PVb),” and “convenience
(PVc).”
Value: Compliance (PVa)
• Using the PHR helps me to perform my healthcare
activities (by reminding me to make and keep my
appointments, etc.)
• Using the PHR helps me to stay on schedule with my
healthcare activities (such as getting my regular
checkup)
• Using the PHR helps me perform my healthcare
activities at the appropriate times (such as refilling
prescriptions)
• Using the PHR helps me remember to perform my
health care activities (like testing my blood sugar)
• Using the PHR allows me to accomplish more of my
healthcare objectives (such as losing weight)
Value: Connectedness (PVb)
• Using the PHR improves communications between
my care providers and me
• Using the PHR improves my relationship with my
care providers
• Reducing the number of forms to fill-out during
registration by having the information available on my
PHR is valuable to me
Value: Convenience (PVc)
• It would be valuable to have my health information
available at all times
• It would be valuable to have my complete medical
record with me at all times
• It is critical to have my emergency medical
information with me at all times
Winter 2006
• It would be valuable to have all of my healthcare
information located in one place
In the data we collected, we found that the typical PHR
user is male, between the ages of 51 to 60, has a
chronic illness, takes multiple medications daily, and
goes to the doctor more than 7 times annually. This
user is also well educated and falls into the upper-middle
socio-economic class.
Most Important Health Care Tasks
Track / Trend
Lab Results
4.1%
Other
20.7%
Ability to
organize data
in useful way
18.6%
Track Vitals /
Self-Monitor
5.2%
Store Tests,
Evaluations.
6.1%
Access to
doctor reports
and lab
4.1%
Manage
Info/Knowledge
8.8%
Medication
Tracking
10.9%
Medical
History Archive
21.6%
In terms of healthcare tasks, the most frequently
mentioned was maintaining a medical history archive,
followed by the ability to organize data in useful ways.
Thus, it appears that users view the PHR as an
electronic repository of critical medical information.
Primary Uses for PHR
Other
5.9%
Family Medical
History
26.4%
Track Lab
Results
14.1%
Track
Medications
18.7%
Track/Trend
my health
13.5%
Track Doctor
Visits
21.4%
This is echoed in the primary uses cited for the PHR,
where the maintenance of family medical history and
keeping track of doctor visits were the most popular uses
of the PHR.
Data Analysis, Results, and Discussion
We used structural equation modeling techniques with
the EQS computer program to perform all confirmatory
factor and structural analyses. We first tested the
relationship between the demographic and medical
condition variables and behavioral intention. The age of
the subject and the education level were found to be
significant predictors of intention to use. These results
showed that younger and less educated users have
greater intentions for future use. Our next step was to
test a mediated model with the perceived value
constructs acting as mediators. When we introduced the
mediators, both age and education became non-
CHIDS ▫ Van Munching Hall ▫ College Park, MD ▫ University of Maryland ▫ www.smith.umd.edu/chids ▫ chids@rhsmith.umd.edu ▫ ph. 301.405.0702
CHIDS Research Briefing, vol.1, issue 2A
significant and PVa and PVb emerged as significant
predictors of BI, collectively explaining over 40% of the
variance in usage intentions.
Our findings show that the effects of individual
characteristics on usage intentions for PHR software are
fully mediated by value perceptions. Interestingly, the
results suggest that value is perceived very differently
across the demographic and medical condition variables.
Education is the only variable significantly related to the
desire for compliance (PVa), and this relationship is
negative. Thus, less educated users believe that the
software will assist them in becoming more organized in
managing their medical information. The presence of a
chronic medical condition is associated with a perception
that the PHR can yield a closer relationship and greater
connectedness with healthcare providers (PVb), while
age and income are negatively related to such value.
Not surprisingly, younger individuals have a lower desire
for close relationships because their health is more likely
to be in good condition. High income users possibly
have other mechanisms for ensuring high-quality
healthcare (e.g., using private physicians). Finally, the
convenience aspects of using a PHR (PVc) are
negatively associated with age, and positively
associated with the severity of the health condition as
assessed by the need for multiple medications on a daily
basis and more frequent doctor visits. Overall, the
medical condition variables predict PVb and PVc, but not
PVa. We also find that convenience (PVc) is not a
significant predictor of usage intentions.
In summary, the pattern of results reveals that individual
profiles in regard to demographic and medical condition
factors yield varying levels and types of value
perceptions. The lack of a relationship between the
medical condition variables and perceived value in the
form of compliance suggests that such value is likely to
be salient for most users, independent of whether they
have a need to manage their health proactively. As
might be expected, value perceptions related to closer
interaction with a healthcare professional and
empowerment are amplified in the presence of severe
medical conditions. Economically disadvantaged users
that may otherwise be challenged in regard to receiving
medical attention, view the PHR as an important means
for staying more connected with their doctors. To the
degree that behavioral intentions drive actual adoption
and use, overall the findings indicate that the early
adopters of PHRs are likely to be individuals who are
less educated, older, less wealthy, and suffering from a
chronic illness.
Conclusion
The motivation for this study is the pressing need to
contain rising healthcare costs while simultaneously
ensuring that medical errors are reduced and patient
Page 4
Winter 2006
safety is enhanced. Technologies such as the PHR can
assist in these endeavors by both providing a repository
of critical data for use by clinicians, and aiding patients in
becoming more proactive in the management of their
health. Ongoing research that can aid in developing
adoption strategies is critical.
Suggested Citation:
Agarwal, R. and Angst, C.M., "Early Findings on
Personal Health Records and Individual Use" CHIDS
Research Briefings (1:2A), Center for Health Information
and Decision Systems, Robert H. Smith School of
Business, University of Maryland, Winter 2006, pp 1-4.
For more information about this research study or to
request the complete working paper, please send an
email to cangst@rhsmith.umd.edu.
CHIDS CONTACT INFORMATION
Director – Ritu Agarwal, Professor and Robert H. Smith
Dean's Chair of Information Systems
Associate Director – Corey Angst, Research Assistant
Professor
Center for Health Information and Decision Systems
Robert H. Smith School of Business
University of Maryland
Van Munching Hall
College Park, Maryland 20742
Ph: 301.405.0702
chids@rhsmith.umd.edu
www.smith.umd.edu/chids
CHIDS MEMBERSHIP INFORMATION
Become a corporate member of CHIDS and realize the
benefits of having access to a world-class team of
researchers. With three levels of membership, it is easy
to find a membership that suits the needs of your
organization. All levels of membership provide your
organization with the following benefits:
• Early insights from research
• Access to top scholars
• Quarterly updates on key HIT activities
• Participation in bi-annual conferences
• Opportunity to influence CHIDS’ research agenda
• Priority access to research projects
CHIDS CORPORATE MEMBERS
Johnson & Johnson Health Care Systems, Inc.
Pfizer Inc
CapMed, a Division of Bio-Imaging Technologies, Inc.
CTIS Inc.
CHIDS ▫ Van Munching Hall ▫ College Park, MD ▫ University of Maryland ▫ www.smith.umd.edu/chids ▫ chids@rhsmith.umd.edu ▫ ph. 301.405.0702