Use of email in a family practice setting: opportunities and challenges in patient- and physician-initiated communication

With the rise of health care consumerism, patients have increasing expectations regarding the health care they receive. These expectations, coupled with the rising complexity of medical care and the increasing number of recommended preventive and chronic disease management services, have created a demand for more of physicians' time than is available [1, 2]. Electronic communication technology could increase communication between the health care system and patients in general, enhance patient-physician interaction in particular, and provide opportunities to improve both the quality of care and the efficiency of clinical time use [3].

The use of electronic mail (email) among Americans has increased from 9% in 1995 to 74% in 2005 [4]. Consequently, email communication between patients and physicians is becoming an increasingly relevant possibility. In studies of primary care practices, more than half of surveyed patients had access to email or described themselves as email users, and a large proportion (70–90%) were interested in using email to communicate with their physicians [5‐8]. Patients note speed, convenience, utility for managing simple problems, and avoidance of 'telephone tag' – the parties alternately leaving messages for each other – as clearly positive aspects of email [9]. However, patient-physician email use has not been widely adopted: only 5–10% of patients actually communicate with their doctors using this medium [6, 10, 11].

Physicians have been wary of adopting email as a major mode of communication with patients. They are concerned that lack of reimbursement, inundation with email, and dealing with trivial issues or topics inappropriate to email will only increase the time demands that they currently face [6, 7, 11]. Even among physicians who regularly use email, 25% are unhappy with using it, citing 'patient request' as the main reason for engaging in email contact [12]. These physicians, like those in other studies, report concerns about time demands, medicolegal risks, and the ability of patients to use email appropriately. Even the asynchronous nature of email, often cited as a benefit since it is available at the convenience of the user and does not require the arrangement of a scheduled appointment, can be seen as a potential detriment if used to replace appropriate face-to-face communication between patient and provider [13].

Reimbursement remains a problem; however, some studies indicate that the medium shows promise. Controlled trials have shown that physicians using email systems to communicate with patients spend only 5–10 minutes a day dealing with on average 12–13 emails a week [11, 14, 15]. Patients also tend to use the format appropriately by avoiding emergent issues, limiting the content to medical and business-related topics (e.g. appointment setting), and including only one request per email [16‐18]. Also, while an early study showed no effect of an email system on physicians' efficiency [15], later studies using web-messaging demonstrated an increase in the number of patient visits and services provided per workday per physician, as well as a reduction in telephone calls from patients, resulting in a 10% increase in physicians' productivity [19, 20]. Efficiency might be further enhanced by the use of non-physician staff to triage emails [21].

As physicians begin to adopt email communication with patients, it is of interest to learn the extent to which email can be used to improve medical care and to expand the possibilities email presents. For instance, while a number of studies have focused on patient-initiated communication, only one published report that we could identify has addressed the possibility of disseminating health-related information to established patients using email [22]. The use of electronic media to provide health education to patients could reduce the time needed during face-to-face patient visits for counseling and preventive education, which can be time-consuming [1, 23]. To assess the possibilities for both patient-initiated and physician-initiated email communication, we conducted two research studies in a family practice setting to determine 1) the proportion of patients who would be receptive to communicating via email with their physician, and 2) the feasibility of providing preventive health education via email to patients.

Our survey results look promising for the future utilization of patient-initiated electronic communication in family practice clinics. Sixty-eight percent of respondents had email access, and 80% were interested in using email to communicate with their doctors. Forty-two percent of the patients with email would be willing to pay a small annual fee to have electronic access to their physician, similar to what has been reported in the general population (37%) [8]. In the future, the introduction of a clinical procedure code for email communication could make it possible for time spent using email to be reimbursed, which, coupled with appropriate billing systems, could spur increased utilization.

Respondents also reported that their email addresses were fairly stable, and in many cases less likely to change than their other contact information. This finding suggests that it may be in the interest of clinics to not only develop email capabilities but to collect email addresses from patients as part of their standard personal contact information.

Our prevention education feasibility study, however, demonstrates some of the issues involved in physician-initiated electronic communication. From a research perspective, beyond the usual difficulties in establishing initial contact and study refusals, studies based on electronic communication lose potential participants because of lack of access to email and difficulties in the paper-based consent process. One-quarter of recruited participants failed to return the consent form, indicating that a mail-based element in an otherwise electronic intervention can have a strong negative impact on recruitment. Salient to both research and to regular clinical communication, of the few women enrolled in the trial, two had problems receiving the emails sent from the clinic: multiple messages bounced back with the explanation that the 'mailbox was full'. When dealing with email from the physician to the patient, the patients may be less concerned with technological issues than if they initiate the contact themselves.

Roughly 76% of women recruited for the prevention study had access to email, a slightly higher number than we found in the general population. However, this is not unexpected, as the population attending preventive visits would be likely to have private insurance and higher levels of income and education, which are positively associated with Internet access [24]. Conversely, it is important to note that women of lower socioeconomic status, as indicated by enrollment in Medicaid, were disproportionately ineligible to take part in the study because of their lack of access to email. This issue of the 'digital divide' – decreased access to electronic technologies in lower income groups – is an ethical concern in the use of email for patient-physician communication [13, 25]. Our study showed that Medicaid patients and black patients were less likely to have access to email, and that these groups also had a higher mean number of annual visits to the clinic. They probably represent a less healthy and less affluent subgroup of our patient population, and yet they would have had access neither to patient-initiated communication nor to the health education resources initiated by the clinic.

Our patient survey study was limited by certain aspects of the design and the use of simple questions. Regarding study design, it would have been helpful to collect demographic information on those patients identified as not having an email address. This would have provided additional insight into the influence of age, income, and gender on access to email in our patient population. Also, we chose to utilize dichotomous questions for our survey instead of Likert scales. Although Likert scales, which allow the respondent to indicate levels of agreement or disagreement, would have provided more detail about 'interest level' in our patients, we used dichotomous questions to minimize the time required to complete the survey in our waiting room, and also to avoid the central tendency bias and acquiescence bias common to Likert questions [26].

Our study on the feasibility of clinic-initiated email was limited by the small number of patients completing the study and by a single clinic site. The study did not yield the results we hoped for, as recruitment was hampered by numerous issues unrelated to email access. Despite these limitations, both studies highlight the growing interest in a new method of patient-physician communication, and present potential issues for those wishing to use the internet for research and for patient education.

The ethical and legal ramifications of electronic communication with patients should be thoroughly understood prior to widespread use by individual medical physicians or practices. The American Medical Informatics Association recommends the standard use of informed consent, which should include itemized terms of communication guidelines, instructions for when to 'escalate' to phone calls and office visits, description of security mechanisms in place, and indemnity of the health care institution for information loss due to technical failure [27].

Case law regarding the use of clinical emails with patients is not yet well developed. Physicians should proceed with caution, particularly when giving medical advice or responding to unsolicited emails from patients that may initiate an 'implicit contract' [28]. In addition, many areas of the country require out-of-state physicians to be licensed in the state where the care will be administered before they may give recommendations via email [29]. The Health Insurance Portability and Accountability Act's requirements of encryption software to protect patient confidentiality should be in place prior to initiating any type of medical-based electronic communication.

Email communication has the potential to improve patient access to healthcare, reduce administrative costs, and improve patient satisfaction. Access to email is likely to increase with new technology and lower costs and may be a better way to reach patients, particularly if email addresses change less often than street addresses and phone numbers. Among those women who participated in our feasibility study and received emailed health education materials, there was a high level of interest in receiving future emails from the clinic. This approach to using email – to initiate health-related information exchange from the clinic to the patient – has not been widely reported. We believe this approach is promising but will need to be further developed and refined.