Brief CommunicationElectronic Reminders to Patients Within an Interactive Patient Health Record

Gary S. Fischer, MD,1 Rachel Hess, MD, MS,1

Babette M. Landeen, BS,2 Melissa Weimer, MS,1

Caroline R. Zieth,1 Xinxin Dong, MS,1 Sunday Clark, ScD,3

and Mark S. Roberts, MD, MPP1,4

1Department of Medicine, University of Pittsburgh School ofMedicine, Pittsburgh, Pennsylvania.

2Information Services Division, University of Pittsburgh MedicalCenter, Pittsburgh, Pennsylvania.

3Department of Medicine, Weill Cornell Medical College,New York, New York.

4Department of Health Policy and Management, Universityof Pittsburgh Graduate School of Public Health, Pittsburgh,Pennsylvania.

AbstractKeeping patients with complex medical illnesses up to date with their

preventive care and chronic disease management services, such as

lipid testing and retinal exam in patients with diabetes, is chal-

lenging. Within a commercially available electronic health record

(EHR) with a secure personal health record (PHR), we developed a

system that sends up to three weekly reminders to patients who will

soon be due for preventive care services. The reminder messages

reside within the secure PHR, which is linked to the EHR, and are

displayed on a screen where patients can also send to the physician’s

office an electronic message to request appointments for the needed

services. The reminder messages stop when the patient logs on to

review the reminders. The system, designed with patient input,

groups together all services that will be due in the next 3 months to

avoid repeatedly messaging the patient. After 2 months, the cycle of

reminders begins again. This system, which is feasible and eco-

nomical to build, has the potential to improve care and compliance

with quality measures.

Key words: e-health, cardiology/cardiovascular disease, medical

records, technology

Introduction

Patients with multiple complex diseases often do not receive

appropriate preventive care, nor do they receive appropriate

monitoring and therapies for chronic conditions such as

diabetes, heart disease, and congestive heart failure.1–8

There are many reasons for this, including fragmentation of care

among specialists,9 the lack of patients’ participation in their own

care,4 and the inherent complexity of caring for patients with mul-

tiple diseases and risks.1,2

The appropriate application of health information technology

meant to be used by providers has been shown to improve process

and clinical outcomes for many individual diseases.10,11 For pro-

viders, active reminders at the time of decision-making have been

found to be more successful than passive access to protocols,12–15

and simple mechanisms that allow the reminder function to facilitate

ordering of needed services have improved the performance of such

systems. However, similar reminders have rarely been used to directly

engage patients in their own care.

The effectiveness of the use of personal health records (PHRs) by

patients has been mixed. Although patients will use and view in-

formation on their own PHR,16,17 there is little evidence that current

PHRs improve the receipt of prevention services.18 The purpose of

the Self-Management and Reminders in Technology (SMART)

project is to test whether an interactive PHR that engages the patient

with active reminders linked to mechanisms for reducing preven-

tion gaps can improve adherence to prevention guidelines and

improve cardiovascular risk in patients with coronary artery dis-

ease, diabetes, or either hypertension or high cholesterol requiring

medication that must be monitored. Data collection is ongoing. This

report describes the design of the system and its early utilization by

participants.

Existing InfrastructureThis project was conducted at the University of Pittsburgh Medical

Center (Pittsburgh, PA), which uses an electronic health record (EHR)

(EpicCare; Epic Systems Corp., Verona, WI) in its outpatient facilities.

The EHR contains a decision-support regimen, including a health

maintenance (HM) module, which tracks services, procedures, and

tests for which a patient may be due. The system has defaults set for

age- and sex-appropriate general preventive health services. For

example, for patients between the ages of 50 and 85 years, the

module shows that a colonoscopy is due every 10 years. Members of

the healthcare team can add, delete, or change the interval of a topic

for a patient by applying modifiers, so that the physician can change

the colonoscopy interval to every 5 years for a patient who requires

more frequent screening.

The list of HM topics is modified for chronic disease management

or medication monitoring. Practitioners receive alerts recommending

that they add the appropriate modifiers to HM for certain conditions

or medications. For example, if a patient has diabetes, physicians are

prompted to add the corresponding modifier, which places biannual

A1c testing and annual urine albumin to creatinine ratios, lipids, and

eye exams on the HM module.

The existing infrastructure includes a PHR linked to the EHR.19

Through the PHR, patients can view a substantial amount of their EHR

information, including their HM. From the PHR HM screen, patients

DOI: 10.1089/tmj.2012.0116 ª M A R Y A N N L I E B E R T , I N C . � VOL. 19 NO. 6 � JUNE 2013 TELEMEDICINE and e-HEALTH 497

can send a request to their provider’s office to schedule needed ser-

vices. SMART was designed to further improve the PHR by adding a

more active component that alerts patients to needed services.

Design of SMARTIn order to assure the application was patient-centered, we dis-

cussed the design of the reminders with two small groups of patients

who were active users of the PHR. The following design decisions

were made (Fig. 1):

1. Participants receive an electronic reminder message once a

targeted service is due or about to be due (see Item 3 for time

frame).

2. The reminder message includes information about every ser-

vice that is due within 3 months, along with a link to the HM

module in the PHR and information about how to request the

tests from the doctor’s office.

3. A message is triggered 2 weeks before tests that are easily sched-

uled, like blood tests (category A services), and 2 months before

tests requiring more time, like colonoscopies (category B services).

4. Participants who do not log into the portal get up to three

weekly electronic messages. A postal letter accompanies

the last reminder. The letter contains the same information as

the electronic message. It also informs the patient about the

PHR message and gives troubleshooting information in case

the patient is having trouble accessing the PHR.

5. Once a participant has logged into the portal or has received

three weekly messages, he or she will not receive another re-

minder message for 2 months. After 2 months, when health

services are due, the sequence begins again.

We used the EHR’s capacity for user-run real-time reporting to

allow designated personnel using a special account to run reports

that identify participants who should receive the first, second, or

third reminder in each cycle. A staff person would run the reports

on a weekly basis and send the participants on each list the correct

electronic reminder message or letter. The EHR allowed the person

running the reports to send the reminder messages or letters to all

of the participants on the reports directly from the screen showing

the report results.

To implement this work flow, four separate reports were written.

The design of the reports is outlined in Table 1. Each of the reports

includes those SMART participants who meet specific criteria re-

garding whether they have targeted services that are about to be

due, how many reminder messages they have received in the last 2

months, and whether they have logged in since receiving a re-

minder message. Because all reminder messages are sent from a

specific, dedicated account, the reports identify messages sent from

that account as reminder messages.

Initial ResultsWe are evaluating the effectiveness of the active PHR under a

demonstration grant from the Agency for Healthcare Research and

Quality. Recruitment to evaluate the effectiveness of SMART began

on July 26, 2010, and was completed September 2011. We re-

cruited individuals who had (1) coronary artery disease or (2)

congestive heart failure or (3) who had hypertension or hyperlip-

idemia and were taking a medication that required routine labo-

ratory monitoring (e.g., lipid-lowering drug). They were excluded if

they did not speak English or were not willing to use the PHR.

Subjects were randomized to receive the modified, active version of

the PHR described here or to continue to receive the standard

version of the PHR. In total, 1,169 subjects have been enrolled, with

584 in the active arm receiving reminder messages. The interven-

tion and control groups are demographically similar (Table 2), and

the study population is similar to our clinic population (data not

shown). The participants in the active reminder arm have received a

total of 3,524 messages as of March 2012, and nearly 65% of pa-

tients logged in to the portal upon receiving the first message of a

cycle. In a few cases, subjects noted that they had had procedures,Fig. 1. Flowchart of reminder messages and letters. SMART, Self-Management and Reminders in Technology.

FISCHER ET AL.

498 TELEMEDICINE and e-HEALTH JU NE 2013

the completion of which was not reflected on the HM module. The

analytic and programming work effort involved in configuring

SMART was approximately 60 h.

ConclusionsBy using simple tools available in a commonly used EHR, we

constructed a system of electronic reminders that are sent directly to

patients to increase patient awareness of tests and procedures that are

recommended for preventive health care, medication management,

or chronic disease management. Our development efforts demon-

strate that patients can be active contributors to the design of such a

system, and our early results demonstrate that a wide variety of

patients can and will use the PHR and that patients can be actively

engaged, with significant numbers logging in to check the reminder

messages. At the conclusion of the study, further analysis will in-

vestigate whether this translates into improvements in care.

Disclosure StatementNo competing financial interests exist.

R E F E R E N C E S

1. Coleman MT, Newton KS. Supporting self-management in patients with chronicillness. Am Fam Physician 2005;72:1503–1510.

2. Redelmeier DA, Tan SH, Booth GL. The treatment of unrelated disordersin patients with chronic medical diseases. N Engl J Med 1998;338:1516–1520.

3. Boyd CM, Darer J, Boult C, Fried LP, Boult L, Wu AW. Clinical practice guidelinesand quality of care for older patients with multiple comorbid diseases:Implications for pay for performance. JAMA 2005;294:716–724.

4. Heisler M, Cole I, Weir D, Kerr EA, Hayward RA. Does physician communicationinfluence older patients’ diabetes self-management and glycemic control?Results from the Health and Retirement Study (HRS). J Gerontol A Biol Sci MedSci 2007;62:1435–1442.

5. Chojnowski D. Protecting patients from harm: Taking aim at heart failure.Nursing 2007;37:50–55; quiz 55–56.

6. Schmittdiel J, Mosen DM, Glasgow RE, Hibbard J, Remmers C, Bellows J. PatientAssessment of Chronic Illness Care (PACIC) and improved patient-centeredoutcomes for chronic conditions. J Gen Intern Med 2008;23:77–80.

7. Smith S, Mitchell C, Bowler S. Patient-centered education: Applying learner-centered concepts to asthma education. J Asthma 2007;44:799–804.

8. Schott-Baer D, Christensen M. A pilot program to increase self-care of adultasthma patients. Medsurg Nurs 1999;8:178–183.

9. Marchibroda JM. The impact of health information technology on collaborativechronic care management. J Manag Care Pharm 2008;14(2 Suppl):S3–S11.

10. Chaudhry B, Wang J, Wu S, Maglione M, Mojica W, Roth E, Morton SC,Shekelle PG. Systematic review: Impact of health information technologyon quality, efficiency, and costs of medical care. Ann Intern Med 2006;144:742–752.

Table 1. Report Logic for the Self-Management and Reminders in Technology Program

REPORT DEFINITION

Report 1a (identifies patients for first reminder message) Patient is a participant AND patient has a targeted medical service is due or about to be duea AND

patient has never received a secure reminder message.

Report 1b (identifies patients for first message 2 months

or longer after completing a message cycle)

Patient is a participant AND patient has a targeted medical service that is due or about to be duea

AND the patient has received at least one secure reminder message AND the most recent secure

reminder message was more than 2 months ago.

Report 2 (identifies patients who need a second

reminder message in a cycle)

Patient is a participant AND patient has a targeted medical service that is due or about to be duea

AND patient has received exactly one secure reminder message within the last 2 month AND has

not logged on since receiving that message.

Report 3 (identifies patients who need a third

reminder message and letter)

Patient is a participant AND patient has a targeted medical service that is due or about to be duea

AND patient has received exactly two secure reminder messages within the last 2 months AND has

not logged on since receiving those messages.

aBlood and urine tests are considered ‘‘about to be due’’ when they will be due within s weeks. Tests that take time to schedule (e.g., colonoscopies, eye exams,

mammograms, Pap test) are considered ‘‘about to be due’’ when they will be due within 2 months.

Table 2. Demographic Data of Participants

PHR + ACTIVEREMINDERS REGULAR PHR

Number 584 585

Age [mean (SD)] (years) 58 (10.57) 58 (10.70)

Sex

Male 265 (45.38%) 280 (47.86%)

Female 319 (54.62%) 305 (52.14%)

Racea

Black 68 (11.64%) 88 (15.04%)

White 511 (87.50%) 484 (82.74%)

Asian 4 (0.68%) 10 (1.71%)

Other 5 (0.86%) 2 (0.34%)

aParticipants were able to select more than one race.

PHR, personal health record; SD, standard deviation.

REMINDERS IN PHR

ª M A R Y A N N L I E B E R T , I N C . � VOL. 19 NO. 6 � JU NE 2013 TELEMEDICINE and e-HEALTH 499

11. Young AS, Chaney E, Shoai R, Bonner L, Cohen AN, Doebbeling B, Dorr D,Goldstein MK, Kerr E, Nichol P, Perrin R. Information technology tosupport improved care for chronic illness. J Gen Intern Med 2007;22(Suppl3):425–430.

12. Garg AX, Adhikari NK, McDonald H, Rosas-Arellano MP, Devereaux PJ, Beyene J,Sam J, Haynes RB. Effects of computerized clinical decision support systems onpractitioner performance and patient outcomes: A systematic review. JAMA2005;293:1223–1238.

13. Gans D, Kralewski J, Hammons T, Dowd B. Medical groups’ adoption ofelectronic health records and information systems. Health Aff (Millwood)2005;24:1323–1333.

14. Hersh W. Health care information technology: Progress and barriers. JAMA2004;292:2273–2274.

15. Shortliffe EH. Strategic action in health information technology: Whythe obvious has taken so long. Health Aff (Millwood) 2005;24:1222–1233.

16. Hess R, Bryce CL, McTigue K, Fitzgerald K, Zickmund S, Olshansky E, Fischer G.The diabetes patient portal: Patient perspectives on structure and delivery.Diabetes Spectrum 2006;92:106–110.

17. Tang PC, Lansky D. The missing link: Bridging the patient-provider healthinformation gap. Health Aff (Millwood) 2005;24:1290–1295.

18. Hess R, McTigue K, Bryce CL, Fitzgerald K, Zickmund S, Sacco D, Olshansky E,Fischer GS. A pamphet’s just a pamphlet. J Gen Intern Med 2006;21(Suppl 4):283.

19. Hess R, Bryce C, Paone S, Fischer G, McTigue K, Olshansky E, Zickmund S,Fitzgerald K, Siminerio L. Exploring challenges and potentials of personal healthrecords in diabetes self-management: Implementation and initial assessment.Telemed J E Health 2007;13:509–517.

Address correspondence to:

Gary S. Fischer, MD

Department of Medicine

University of Pittsburgh School of Medicine

200 Lothrop Street

Pittsburgh, PA 15213

E-mail: fischerg@upmc.edu

Received: May 7, 2012

Revised: August 28, 2012

Accepted: August 28, 2012

FISCHER ET AL.

500 TELEMEDICINE and e-HEALTH JUNE 2013