Experiences of Patients Who Were Early Adopters of Electronic Communication With Their Physician: Satisfaction, Benefits, and Concerns

September 1, 2004
Thomas K. Houston, MD, MPH

,
Daniel Z. Sands, MD, MPH

,
Mollie W. Jenckes, MHS

,
Daniel E. Ford, MD, MPH

The American Journal of Managed Care, September 2004, Volume 10, Issue 9

Objective: To explore the experiences of patients who wereearly adopters of e-mail communication with their physicians.

Methods: Patients' experiences were assessed with an Internetbasedsurvey of 1881 individuals and in-depth telephone follow-upinterviews with 56 individuals who used e-mail to communicatewith providers. Two investigators qualitatively coded interviewcomments independently, with differences adjudicated by groupconsensus.

Results: A total of 311 (16.5%) of the 1881 individuals reportedusing electronic mail to communicate with their physicians.Compared with the population-based Behavioral Risk FactorSurveillance Survey, users of e-mail with physicians were twice aslikely to have a college education, were younger, were less frequentlyethnic minorities, and more frequently reported fair/poorhealth. Among the 311 patients who used e-mail with their physicians,the most frequent topics were results of laboratory testingand prescription renewals. However, many of the 311 users (21%)also reported using asynchronous e-mail inappropriately to conveyurgent or sensitive issues (suicidality, chest pain, etc). Almost all(95%) perceived that e-mail was more efficient than the telephone.Important benefits uncovered from the interviews were that somepatients felt more emboldened to ask questions in e-mail comparedwith face-to-face communication with doctors, and liked the abilityto save the e-mail messages. Users also expressed concernsabout privacy.

Conclusion: Patients that use electronic communication withtheir physicians find the communication efficient for disease management.Further patient education about inappropriate use of emailfor urgent issues is needed.

(Am J Manag Care. 2004;10:601-608)

Use of the Internet by patients is increasing.1-6 Inaddition to seeking health information, individualsalso can use the Internet to communicatewith each other in online support groups,7-9 and canpotentially use the Internet to communicate with theirhealthcare providers.1,10,11 In a survey of primary carepatients who had Internet access, more than 80% wantedto use e-mail to communicate with their physician.12However, only 6% of patients have used e-mail to communicatewith their healthcare provider.6

Crossing the Quality Chasm

Effective physician-patient communication is importantto patient satisfaction, treatment adherence, andhealth outcomes.13-16 The recent report of the Instituteof Medicine, , states that"patients should receive care whenever they need it andin many forms, not just face-to-face visits...access tocare should be provided over the Internet, by telephone,and by other means."17 Electronic patient-centeredcommunication, using e-mail or Web-based technology,has the potential to enhance physician-patient interactionsby providing asynchronous, self-documentingcommunication of patient questions and physicianadvice.10 However, the utilization of physician-patientelectronic communication is low, and physicians reportconcerns including excessive demands on their timeand medicolegal risks.11,18-22

Patient perceptions of the relative utility of electroniccommunication also may be important for incorporatingthis technology into clinical practice. To betterunderstand patient factors that may influence the possiblefuture of patient-physician electronic communication,we chose to explore the experiences of the earlyadopters of electronic communication, patients whoalready communicate with their physicians by electronicmail. Our research questions included the following:

  • What are the characteristics of these earlyadopters of patient-physician e-mail?
  • What topics do these patients address in e-mail totheir providers?
  • What do these patients perceive to be the benefitsof and problems with e-mail communications?

To conduct this hypothesis-generating research, weused a combination of quantitative surveys and qualitativein-depth interviews with patients who currentlycommunicate with their providers through the Internet.

METHODS

Study Design and Subjects

We conducted a cross-sectional study with data collectedfrom 2 sources: (1) an Internet-based survey ofpatients who use the Internet to look for health informationand (2) follow-up in-depth telephone interviewswith patients. We targeted early adopters to search forwhat is currently working and not working with electronicpatient-centered communication. The data collectionfor this study was part of a larger researchproject funded by the Bayer Institute for HealthcareCommunication to explore the potential of patient-physiciane-mail communication from the perspectiveof physicians and patients.1

Recruitment Procedures

Use of electronic mail to communicate with physiciansis still uncommon in the general population;therefore, recruiting patients for this study was a challenge.Because early adopters of e-mail with their physiciansalso would be likely to use the Internet to searchfor health information, we chose to recruit and surveypatients who were using the Internet for this purpose.

A patient survey was developed and implementedonline. After receiving approval from the Johns HopkinsCommittee on Clinical Investigation, InteliHealth.com,a health media company owned by Aetna US Healthcareand affiliated with Harvard Medical School, agreed topost a volunteer recruitment request. InteliHealth has apopular consumer health information Web site availableto any consumer searching for health informationonline. By collaborating with InteliHealth, we hoped tohave the widest possible visibility of our request for volunteersamong our target population. A link to the surveyalso was posted on the consumer Web site ofCareGroup Healthcare System, a Harvard-affiliatedintegrated health delivery network in the Boston area.The survey was available from May through October2001. We confirmed that each response was from aunique IP address, thus lessening the chance that wehad repeat users. After recruiting approximately 300individuals who reported that they had used electronicmail to communicate with their physicians, we closedenrollment.

After the questionnaire portion of the study wascompleted, we asked participants who had used e-mailwith their physicians whether they would agree to participatein an in-depth telephone interview. Those whoagreed entered their e-mail address onto the secureserver. Initially, 6 patients were contacted by a researchassistant through e-mail to schedule an in-depth telephoneinterview. These interviews were audiotaped andreviewed to identify themes. Based on the results of initialinterviews, additional open-ended questions wereadded to the telephone interview. Additional audiotapedinterviews were completed until preliminaryreview of tapes suggested that themes had been saturated.All interviews were transcribed verbatim.

Survey and Interview Content

The Internet-based survey included demographicquestions (age, sex, ethnicity, level of education);health-related questions (general health status, numberof medications, and number of visits to healthcareprovider); and a screening question: "In a typicalmonth, please estimate the average number of e-mailsyou send to your physician(s) and your physician'sstaff." To be inclusive, this screening question consideredany e-mail to the physician practice. In our interviews,we asked those who responded to the Internetsurvey whether e-mail communication was with thephysician or the physician's staff. Those who reportedmore than "none" were considered users of e-mail withtheir physician. Individuals who had used electronicmail with their physician were asked additional questionsrelated to the physician's specialty; the clinicaltopics discussed through e-mail; the perceived benefitsof e-mail and problems; and their overall satisfactionwith e-mail communication with their physician.Clinical topics, benefits, and problems were picked froma list that was derived through an iterative process withinput from all the authors (2 of whom have considerableclinical experience with patient e-mail) and then inpilot-testing with students at the Johns Hopkins Schoolof Public Health.

We triangulated from the quantitative surveys usingthe in-depth telephone interviews. Interviewers asked aseries of open-ended questions to guide the interviews,including: "How has e-mail with your physician helpedyou and your doctor manage your health conditions?""Tell me about the most recent e-mail exchange youhave had with your physician" "How is e-mail differentthan telephone contact?" We taught interviewers toprobe for specific examples. Interviews averaged 15 to20 minutes.

Analysis

We used Mantel-Haenszel χ2 trend statistics to comparedemographic characteristics and health statusvariables of individuals participating in our Internet-based survey who had used electronic mail with theirphysicians with the characteristics and health statusvariables of those who reported not using e-mail withtheir healthcare providers. To further characterize thisselect sample of patients who used electronic mail withtheir physicians, we compared their demographic characteristicswith those of participants in the 2000Behavioral Risk Factor Surveillance Survey (BRFSS).The Centers for Disease Control and PreventionBehavioral Risk Factor Surveillance System is an ongoing,cross-sectional, state-based telephone survey conductedamong a representative sample of each state'sadult population. The purpose of this surveillance systemis to collect uniform, population-based data on preventivehealth practices and risk behaviors that arelinked to chronic diseases and injuries in the US population.Most states use either a disproportionate stratifiedsample or a Mitofsky-Waksberg—type sample designto draw a random sample from the set of all possibletelephone numbers based on area codes and prefixes(www.cdc.gov/brfss/pdf/userguide.pdf). Thus, comparisondata from the 2000 BRFSS survey were adjusted forthe complex survey design.

Using descriptive statistics, we summarized the contentof the patients' e-mails, perceived benefits, andpotential concerns related to patient-physician e-mail.We measured the strength of the association betweenthe level of satisfaction with electronic mail and the perceivedimpact on patients' medical care, concerns, anddemographic characteristics using χ2 tests and multivariablelogistic regression.

Two authors independently identified distinct commentsfrom the transcripts of the audiotaped interviewsand together with a third author (MWJ), who has expertisein qualitative methodology, reviewed comments anddeveloped domains and subdomains. Repeated orreworded comments representing the same thought bythe same participant were counted only once. Any disagreementon whether a particular segment representeda unique thought or concept was adjudicated. Domainsand subdomains were agreed on by consensus. Taxonomyof all comments was then sent to the remaining authorsto be reviewed for relevance and consistency.

RESULTS

P

P

A total of 1881 individuals completed the Internet-basedsurvey. Most were recruited from Intelihealth.com, with 88 (5%) recruited from the CareGroup site.Among the respondents, 311 (17%) reported using electronicmail with their physician(s) and/or physician'sstaff. Of these, 79 (25%) reported an average of morethan 1 e-mail per month to their healthcare provider, 69(22%) reported at least 1 e-mail per month, and 164(53%) reported less than 1 e-mail per month. We targetedindividuals who reported communicating with theirhealthcare provider via e-mail 1 or more times permonth to complete an in-depth telephone interview.A total of 56 telephone follow-up interviews were completed,containing 694 unique comments. Intervieweeshad similar education and were similar in ageto the other 311 users, but were more frequently white(95% vs 82%; = .017) and somewhat less frequentlyfemale (63% vs 78%; = .023).

P

Demographic characteristics of survey participantsand participants in the BRFSS are shown in Table 1.Respondents to our survey were twice as likely to havea college education, were younger, and were less frequentlyethnic minorities than those who responded tothe BRFSS. Similar to previous surveys of online healthinformation seekers,23 more respondents were femalecompared with the BRFSS participants. Also, participantsin our survey more frequently reported fair/poorhealth (30% vs 16%). Compared with Internet surveyrespondents who reported not using e-mail with physicians,those respondents who had used e-mail with theirphysicians had higher levels of education (68% vs 52%were college educated) and more frequently reportedtaking medications (66% vs 58% took more than 1 medication; = .04). Physicians to whom these patientssent e-mail messages included general internists (n =154; 50%) and subspecialty internal medicine physicians(n = 77; 25%), family medicine physicians (9%),surgeons (9%), obstetricians/gynecologists (9%), pediatricians(3%), and others (23%).

Common Topics in E-mail With Physicians

Based on survey responses, a variety of topics werediscussed in e-mail messages (Table 2). Among the 311patients who used e-mail with their physicians, themost frequently reported topics were results of laboratorytesting and prescription renewals (both reported by85%). Messages focusing on new, nonurgent symptoms(73%) and nonurgent advice on medical conditions(49%) also were common. Topics reported infrequentlyby these 311 patients included urgent issues (chestpain, shortness of breath, suicidal thoughts; 21%) andsensitive issues (17%). The common topics were echoedin the comments collected in our patient interviews.Some representative topics are shown in Table 3.

Benefits of and Problems WithPatient-physician E-mail

Overwhelmingly, among these 311 patients, the mostcommon (95%) perceived benefit to e-mail communication was that it was more efficient than the telephone(Table 2). Least common, but still reported by 40% ofpatients, was that it was a less intimidating forum to askquestions. In interviews, patients echoed this theme ofe-mail being a less intimidating mode of communicationthrough comments such as "[it's] easier to talk about itwith one step removed." The overall taxonomy of benefitsand problems, with example comments identifiedfrom the 56 patient interviews, is summarized in Table3. Other major identified benefits included the ability tosave e-mails to re-read instructions and improved communicationbecause patients were better able to composetheir questions, and physicians were more"articulate."

Only 2 concerns reported on our survey approached50%: "My physician may not answer" and "I am worriedabout bugging my physician too much" (Table 2). Confidentialityconcerns, especially concerns about work orfamily, were reported by a minority of participants inthe survey, but were elicitedas a major theme in theinterviews, as exemplifiedby the comment: "[I] almostfelt reluctant to do it [e-mailwith physicians] because…what if the e-mail gets intercepted?"(Table 3). Anotherlimitation to electronic mailreported by patients was arelative lack of empathy inphysician e-mail responses.

Satisfaction With Patient-physicianE-mail

P

Of the 311 patients whoreported using e-mail withtheir physicians, 272 (87%)provided information ontheir satisfaction with e-mailcommunication. Of these272 respondents, 82% (n =222) were satisfied with thecommunication. In fact,more than 40% (n = 117)reported they would be willingto pay a fee per e-mail tohave the service, with 60patients reporting theywould be willing to pay morethan $3.00 per e-mail. Wecould not identify any significantdifferences in sex, age,education, ethnicity, numberof medications, or health status between those whowere willing to pay and those who were not. However,those who were willing to pay were more frequent usersof e-mail with their physician than those who used emailwith their physician but were not willing to pay(35% vs 18% sent 2 or more e-mails per month; < .01).

df

P

The majority of patients (78%, n = 213) reported thattheir physician had responded to between 75% and 100%of their e-mail messages, but 13% reported that a physicianresponded to fewer than 25% of their e-mails. Thosewho reported that their physician always responded totheir e-mail messages were more frequently satisfied(93% vs 52%; &#967;2 = 64 [ = 1]; < .001). After adjustmentfor patient age, sex, race, education level, self-ratedhealth, their physician's specialty, and number of e-mailmessages per month by using logistic regression,patients who reported that their physician alwaysresponded were more likely to be satisfied (odds ratio=15.9; 95% confidence interval = 7.0, 36.2) than thosewho reported that theirphysician responded lessthan 100% of the time.Satisfaction was not associatedwith demographiccharacteristics, healthstatus, or average numberof e-mail messageswith physician permonth.

DISCUSSION

The experiences ofthese patients who wereearly adopters of electroniccommunicationwith their physician werequite positive. The mostfrequent benefits fromthese patients' perspectivewere related to theefficiency of communication,as evidenced by acomment obtained duringour interviews: "Ithink it is good, ratherthan playing telephonetag or dealing with frontdesk people who don'tnecessarily get the messagesrelayed." The frequentlyreported topicsrelated to administrativeissues (refills, appointments,laboratory results)are easily handled in anasynchronous manner and may be ideal for electroniccommunication.

Another clear, although less frequent, benefit reportedwas that e-mail was a less intimidating venue forcommunication. Previous research in understandingpeer-to-peer communication on Internet disease-relatedmessage boards supports the theory that communicatingonline creates a "relative anonymity" and thisallows some level of disinhibition for patients to askquestions they may not have otherwise.7

Our study significantly adds to the literature byspecifically targeting the experiences of early adoptersof electronic communication, and by triangulatingquantitative survey data with qualitative data. Previousresearch has clearly documented patients' desire forelectronic communication with providers.12 However,these studies have surveyed patients who have not yetbegun to use e-mail with their providers.12,24

This study identified 1 specific risk of electroniccommunication between physician and patient. Anoticeable subgroup of our participants who had used emailto communicate with their physician (21%) reportedusing e-mail for urgent matters such as chest painor suicidality. The American Medical Informatics Association(AMIA) guidelines suggest that e-mail should notbe used for emergencies or other time-sensitive issues.10Some physicians have expressed concern that patientsmay not be able to distinguish between emergencies androutine issues. Because these serious concerns were self-reported,however, patients did seem to understand thatthey were emergencies. Either way, it would seem thatpatients are not being educated properly about theappropriate use of electronic messaging. The AMIA andother guidelines state that patients need to be educatedand frequently reminded about the rules of use, and amention of this education be documented in the medicalrecord.10 The asynchronous nature of electronicmail is not amenable to communications of an urgentnature. Providersmight not checktheir e-mail forhours, or even days.Thus, if electronicmethods of communicationbetweenphysician andpatient do continueto increase, it iscritically importantthat physicians educatepatients aboutappropriate use ofthis medium.

Although wewere successful inidentifying a groupof patients who hadexperience using emailwith theirphysicians, a limitationof our study isthat it was based ona convenience sampleof patientsresponding to anInternet-based survey.We have attemptedto compare our sample withdata from the population-based BRFSSto characterize the biases of selectionas much as possible.Unfortunately,1 limitation of using the BRFSS as acomparison group is that the number ofvariables available for comparison waslimited. Thus, we were not able tocompare factors such as amount of Internet or e-mailuse. However, patients who responded to our survey didreport a higher level of education, were more frequentlyfemale, and were of younger age compared with theBRFSS respondents. In fact, patients who reportedusing e-mail with physicians were even more educatedthan other Internet users who did not use e-mail withphysicians. Our respondents also had higher rates offair/poor health status compared with respondents tothe BRFSS.

These differences echo the disparities in Internetaccess termed the digital divide.25,26 If, in the vision of theInstitute of Medicine, the Internet is co-opted to increasepatient-provider communication in the hopes of improvingthe quality of care, and access is not available formany, the result may be an increase in health disparities.In fact, the importance of the digital divide has beenacknowledged in Healthy People 2010, where "toincrease Internet access to 80% of the US population" isincluded as a measurable health objective.27 Although wehave not achieved this goal, there is evidence that thedigital divide is narrowing. A recent Forrester Researchstudy indicated that the majority of the medically uninsurednow have a computer and Internet access athome and go online at least once a month (www.forrester.com/ER/Research/DataSnapshot/Excerpt/0,1317,32374,00.html).

Our participants also were different from respondentsto previous surveys of online health informationseekers. Specifically, fewer than half (41%) of our usersof e-mail with physicians reported being concernedabout the privacy of their e-mails. In a previous surveyby the Pew Internet and American Life Project, morethan 70% of Internet health information seekers were"very concerned" about online privacy.23 This differencemay represent a misperception that e-mail ismore private than information provided online.However, it also may suggest another differencebetween these early adopters and the general populationof online health information seekers.

Using the perspective of diffusion of innovation,28patient desire for electronic communication withphysicians and the perceived advantage over telephonecommunication, especially for chronic disease managementissues, may facilitate the eventual adoption of thistechnology. Thus, patient demand, combined withchanges in policy derived from the recent Institute ofMedicine reports, may increase use of electronic communicationbetween physicians and patients, especiallyif confidentiality concerns can be overcome by usingmore secure, Web-based technology.11,29

Overall, patient satisfaction with primary care clinicalservices, especially related to physician-patientcommunication, has declined in recent years.30Electronic communication is potentially a valuableservice physicians could use to increase patient satisfactionwith care. Certainly these early adopters ofelectronic patient-centered communication seemed tohave derived benefit from their e-mail exchanges. Earlyadopters were overwhelmingly satisfied and probablywill communicate this satisfaction to their peers. Infact, a subgroup of these users, who were actually themost frequent users of e-mail with their physicians,would be willing to pay a fee of up to $3.00 per messageto send e-mail to their providers. These data are valuablewithin the context of the growing number of insurerswho are or are considering reimbursing physiciansfor electronic patient-centered communication.31

As we noted, patient satisfaction with electroniccommunication is intimately linked with the experiencethat physicians will respond. Patients who reportedthat their physician did not always respond wereless satisfied. Previous reports also have indicated thatpatients have concerns about lost messages or delayedresponses.32 As younger patients who are even more emailsavvy develop more chronic diseases, many morepatients are likely to want to use electronic communicationwith their physician, potentially increasing thevolume of messages. Currently, physicians who areearly adopters of e-mail with patients report that theyuse e-mail only with a small percentage of theirpatients.1 As volume increases, physicians may havedifficulty responding to all patient e-mail messages.Our patient participants are concerned about this, asevidenced by concerns about "bugging the physiciantoo much."

Insurance payment or capitation arrangementsmight increase physician willingness to respond. Triagemethods (ie, using staff to read e-mails and thus incorporatingelectronic communication more efficientlyinto the work flow of the practice) also may be necessary.32 However, our early-adopter patients seem tovalue being able to communicate directly with theirdoctors. Using staff to read patient e-mails first mayrequire more discussion of privacy issues with patientsand create a significant barrier to use of e-mail. Futureresearch might compare the perceived value and relativeuse of physician-directed and triage-based electroniccommunication from the patient and providerperspectives.

In conclusion, patients in our study preferred theefficiency of using electronic communication foradministrative issues. Thus, we might speculate that emailbetween physicians and patients may be more frequentlyused and more valuable in chronic diseasepopulations for whom prescription refills, appointments,and laboratory tests are most frequent, ratherthan more general, healthy populations with intermittentillnesses requiring diagnostic evaluation. In fact,physicians who are early adopters of e-mail withpatients frequently report using e-mail with theirpatients who have chronic diseases.1 Further researchis needed to understand the dynamics involved in integratingelectronic communication with patients intoproviders' clinical practice. Future interventionsdesigned to increase use of electronic communicationbetween physician and patient should consider the targetpatient population, investigate reimbursementstrategies for providers, and explore whether thepatients are willing to pay for this service.

From the Department of Medicine, University of Alabama at Birmingham, Ala (TKH);the Divisions of General Medicine and Clinical Computing, Department of Medicine, BethIsrael Deaconess Medical Center and Harvard Medical School, Boston, Mass (DZS); and theDepartments of Medicine (MWJ, DEF), Epidemiology (DEF), and Health Policy andManagement (DEF), The Johns Hopkins University School of Medicine and the JohnsHopkins University Bloomberg School of Public Health, Baltimore, Md.

This work was supported by grant 00577 from the Bayer Institute for HealthCommunication, grant ZT32PE010025 from the Academy for Health Services Research andHealth Policy (now AcademyHealth), and by grant 5T32PE010025-04 from Health Resourcesand Services Administration.

Address correspondence to: Thomas K. Houston, MD, MPH, Assistant Professor ofMedicine, Director, Health Informatics Unit, Division of General Internal Medicine,University of Alabama at Birmingham, 510 20th Street South, FOT 720, Birmingham, AL35294. E-mail: tkhouston@uabmc.edu.

J Med Internet Res

1. Patt MR, Houston TK, Jenckes MW, Sands DZ, Ford DE. Doctors who are usinge-mail with their patients: a qualitative exploration. . April-June2003;5(2):E9.

J Med Internet Res

2. Sciamanna CN, Clark MA, Houston TK, Diaz JA. Unmet needs of primary carepatients in using the Internet for health-related activities. .December 2002;4(3):E19.

J Med Internet Res

3. Houston TK, Allison JJ. Users of Internet health information: differences byhealth status. . Apr-Nov 2002;4(2):E7.

West J Med

4. Pennbridge J, Moya R, Rodrigues L. Questionnaire survey of California consumers'use and rating of sources of health care information including the Internet.. 1999;171(5-6):302-305.

BMJ

5. Eysenbach G, Kohler C. How do consumers search for and appraise healthinformation on the World Wide Web? Qualitative study using focus groups, usabilitytests, and in-depth interviews. . 2002;324(7337):573-577.

JAMA

6. Baker L, Wagner TH, Singer S, Bundorf MK. Use of the Internet and e-mail forhealth care information: results from a national survey. . 2003;289(18):2400-2406.

Am J Psychiatry

7. Houston TK, Cooper LA, Ford DE. Internet support groups for depression: a 1-year prospective cohort study. . 2002;159(12):2062-2068.

Arch Intern Med

8. Lorig KR, Laurent DD, Deyo RA, Marnell ME, Minor MA, Ritter PL. Can a backpain e-mail discussion group improve health status and lower health care costs? Arandomized study. . 2002;162(7):792-796.

Am Psychol

9. Davidson KP, Pennebaker JW, Dickerson SS. Who talks? The social psychologyof illness support groups. . 2000;55(2):205-217.

J Am Med Informatics Assoc

10. Kane B, Sands D. Guidelines for the clinical use of electronic mail withpatients. . 1998;5:104-111.

Health Commun

11. Baur C. Limiting factors on the transformative powers of e-mail in patient-physicianrelationships: a critical analysis. . 2000;12(3):239-259.

JOkla State Med Assoc

12. Mold JW, Cacy JR, Barton ED. Patient-physician e-mail communication. . 1998;91(6):331-334.

J Gen Intern Med

13. Ende J, Kazis L, Ash A, Moskowitz MA. Measuring patients' desire for autonomy:decision making and information-seeking preferences among medical patients.. 1989;4:23-30.

Doctors Talking with Patients, Patients Talking with Doctors:Improving Communication in Medical Visits

14. Roter DL, Hall JA. . Westport, Conn: Auburn House; 1992.

JAMA

15. Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM. Physician-patientcommunication. The relationship with malpractice claims among primary carephysicians and surgeons. . 1997;277(7):553-559.

J GenIntern Med

16. Greenfield S, Kaplan SH, Ware JE, Yano EM, Frank HJL. Patients' participationin medical care: effects on blood sugar control and quality of life in diabetes. . 1988;3:448-457.

Crossing the Quality Chasm

17. Committee on Quality of Health Care in America, Institute of Medicine.. Washington, DC: National Academy Press; 2001.

Health Commun

18. Houston TK, Sands DZ, Nash BR, Ford DE. Experiences of physicians who frequentlyuse e-mail with patients. . 2003;15(4):515-525.

JAMA

19. Spielberg AR. On call and online: sociohistorical, legal, and ethical implicationsof e-mail for the patient-physician relationship [see comments]. .1998;280(15):1353-1359.

Am J Law Med

20. Spielberg AR. Online without a net: physician-patient communication by electronicmail. . 1999;25(2-3):267-295.

Informatics Rev

21. Sittig DF, King S, Hazlehurst B. Provider attitudes toward patient-related email.. June 1, 2000. Essays on Clinical Informatics. Available at:http://www.informatics-review.com/thoughts/provider-email.html. Accessed April 7,2004.

WMJ

22. Sherman L. Patients and e-mail: technology means increased confidentialityconcerns. . 1999;98(3):66-67.

The Online Health Care Revolution: How the Web HelpsAmericans Take Better Care of Themselves

23. Fox S, Rainee L. . Washington, DC: The Pew Internet &American Life Project; 2000.

Am J Manag Care

24. Moyer CA, Stern DT, Dobias KS, Cox DT, Katz SJ. Bridging the electronicdivide: patient and provider perspectives on e-mail communication in primarycare. . 2002;8(5):427-433.

Health Aff (Millwood)

25. Brodie M, Flournoy RE, Altman DE, Blendon RJ, Benson JM, Rosenbaum MD.Health information, the Internet, and the digital divide. .2000;19(6):255-265.

Health Aff (Millwood)

26. Burstin H. Traversing the digital divide. .2000;19(6):245-249.

Healthy People 2010

27. Office of Disease Prevention and Health Promotion. Health communication.In: . Conference ed. Washington, DC; US Dept of Health andHuman Services; January 2000:11-11–11-24.

Diffusion of Innovations

28. Rogers E. . 4th ed. New York, NY: The Free Press;1995.

J Am Med Informatics Assoc

29. Liederman EM, Morefield CS. Web messaging: a new tool for patient-physiciancommunication. . 2003;10:260-270.

Ann Intern Med

30. Safran DG. Defining the future of primary care: what can we learn frompatients? . 2003;138(3):248-255.

Am MedNews

31. Cook B. Illinois PPO to pay physicians for online consultations. . 2000;43(20). Available at: http://www.ama-assn.org/amednews/2000/2005/2022/tesb0522.htm. Accessed July 7, 2000.

J GenIntern Med

32. Katz SJ, Moyer CA, Cox DT, Stern DT. Effect of a triage-based e-mail systemon clinic resource use and patient and physician satisfaction in primary care. . 2003;18(9):736-744.