Use of email for patient communication in student health care: a cross-sectional study

Citizens increasingly use email in personal communication [1]. It is not however clear to what extent physicians utilize it for patient communication. In reports from the 1990s 1–14% of doctors in the USA and Norway used email in patient work [2, 3]. In recent studies up to 73% of physicians had used email for patient communication [4‐6].

International and national recommendations and guidelines have been published on email use between doctor and patient; the contents of the Finnish guidelines are well in line with European guidelines [7, 8]. These guidelines emphasize the suitability of email for only certain limited purposes and stress the risks to information security.

Use of email between doctor and patient has been studied in one controlled and randomized and three cross-sectional studies [4, 5, 9, 10]. In a study by Katz and colleagues the number of email contacts of physicians in their study group (46 email messages/100 scheduled visits) was greater than that in the control group (9/100) [5]. Houston and associates found that the majority of doctors received daily 1–5 email messages from their patients [9]. According to Sittig, physicians received daily in average 2.6 messages, and monthly an average of 40 per 140 visits [10]. Gaster and colleagues noted that physicians on average received 7.7 email messages in a month from their patients. Physicians in university clinics were most active in email use, while those in municipal primary health care were least active. Of physicians 58% reported in the questionnaire that the email contacts with patients were for the most part not registered in patient records [4].

Among Finnish citizens of working age young adults are the most active users of email and Internet [1, 11]. University students use these electronic net services even more actively than the young adult population as a whole. In a study from 2002 99% of students reported using email and Internet at least weekly [12]. All students have an email address at the university and their health providers at the FSHS can be reached by email. The student health care system can be seen as an appropriate setting to use email for patient communication [13]. The students represent a young, well educated, relatively healthy part of population which has been identified to be the most active to use email in patient-doctor communication [2, 5, 14].

All physicians (n = 82) in the FSHS' functionary register in April 2003 received a questionnaire by email. We excluded six physicians, who were not any more working for the FSHS and took exception to the two authors. The actual number of survey population was 74. The questionnaire (see Additional file 1) included background factors and a registration (in form of daily tally) of numbers of patient contacts, phone calls and email messages over one working week. Respondents were also asked to assess the number of visits and calls replaceable by email, and the number of email messages including a request, which could not be fulfilled without face-to-face contact. Also doctors' attitudes toward email use for patient communication were asked.

The first mailing of questionnaire took place 28.4.2003 and a reminder was sent 5.5.2003. Recipients were asked to print the survey form, fill it in by hand, and return it by internal mail. Overall 52 out of 74 (70%) physicians returned a completed survey.

Respondents were grouped according to age, location, speciality licence, and type of employment (Table 1). Facts on years of birth were collected from the register book "Finland's Doctors 2002" and other background factors from the FSHS' functionary register.

All physicians at the FSHS utilize EPR (Medicus^®). We collected the numbers of patient contacts documented at Medicus^® for the study period using the statistical software Cognos^®.

We entered data using Microsoft Excel^® software and performed statistical analysis using StatsDirect Statistical^® (3,2,7 -version) software. Statistical analyses were conducted using the proportion test for two independent groups, the χ² test, Fisher's exact test, the unpaired t-test, the Mann-Whitney test and the Kruskal-Wallis test. All tests were made two-sided and p-values below .05 were regarded as statistically significant.

Even if university students do not represent the whole population, they can act as "pilot population" representing adults of working age of a future information society. Our study population – doctors taking care of students – was small with only 52 respondents. Thus the results of the study cannot be indiscriminately generalized. Because of the small study population comparison of the subgroups may not be reliable.

Although the study group was small, it well represented all the doctors at the FSHS, and the response rate was high. A further strength was that we compared the number of patient contacts documented in the questionnaires during the study week to the statistical data of contact numbers from the EPR at the same time. In other studies no corresponding comparison has been made.

The doctors were asked to keep a daily tally of visits, phone calls and email messages, and to evaluate how many visits or phone calls could have been replaced by email. Many doctors undoubtedly did this simultaneously with patient work. Some doctors might have been in a hurry, they probably supplemented the questionnaire at the end of the day. To achieve a more accurate evaluation of visits and phone calls replaceable by email, a continuous assessment (visit by visit, phone call by phone call) could have been stressed even more in the instructions.

The doctors at the FSHS do not have a specific electronic communication system focused on patient communication. They use their general, unprotected email system also to communicate with patients. A specific communication system used only for patient communication would enable an automatic collection of the patient communication data and create a more accurate database than our data collection method.

Katz and colleagues have made the only controlled and randomized study concerning physicians' use of email [5]. Our own results on the average number of doctors' email contacts and email usage are of the same magnitude as those referred to above and in other recent studies in the USA [4, 9, 10]. In the present study 79% of respondents used email for patient communication. The proportion of those who had used email is clearly larger than in older studies, and at the same level as reported in recent international studies [2‐6].

Our study revealed individual differences in the use of email in patient work. Differences in physicians' activity in using email have previously been reported in only one study [4]. Deriving of our small study group only especially glaring association between subgroups of respondents could be verified. Our findings still support the results published by Gaster and associates. Physicians working in the capital area were more active email users than their colleagues elsewhere in Finland.

Physicians reckoned that email could replace only 2% of visits. This confirms Sittig's evaluation in 2003 that email could possibly cover a small percentage of visits [10]. Increasing the use of email can thus not considerably reduce the number of patient visits. On the other hand it could make physicians' crowded telephone hours easier [15].

When we compared contacts in the EPR with contacts registered daily on the questionnaires we found that the majority of email contacts were not registered in the EPR. This finding is supported by Gaster and colleagues who asked physicians themselves to describe how often they usually registered email contacts in patient records [4].

FSHS provides specific electronic services for focused issues (email service for cancellation of appointments, health counseling service on the Internet, and email service for feedback). Principles of recommended issues to use email between health providers and patients are available for students at FSHS' websites. We have had a presumption that email messages between FSHS' physicians and their patients mainly handle patient care. Nyström's congress report from 2004 supports our presumption. He explored 139 email messages from 103 individual patients at his GP practice at the FSHS and noticed that 77 % of email messages handled medical tests, and 16 % handled follow-ups of symptoms or illnesses [16]. Thus the information in email communication should be entered in patient records.

All university students in Finland have access to Internet and email at their universities. Use of email as communication method in health care does not in their case cause inequalities in health. A general tendency in the societies to provide also health services widely in electronic form (in Internet or by email) can contribute to inequalities for those who are not able to use modern technologies [13, 17].

Doctors at the FSHS had an average of 8.6 email contacts with their patients during one week. The proportion of email contacts per visit was 20%. Physicians estimated that email contacts could substitute 2% of patients' visits and 21% of phone calls. Of email contacts 73% were not documented in the EPR.

Our study indicates that email communication really constitutes a part of patient work. This should be taken into account in planning working time and daily timetables. Use of software not integrated with the EPR increases the physician's registering load and currently involves extra work. It is not possible to confirm the patient's identity reliably using two separate systems. A system allowing retrieval of patient's identity safely and with no need to register separately the email communication in the EPR would promote the patient's adequate treatment and reduce the physician's medico-legal risks.

There is a need for a larger study on email utilization between patient and physician which better covers the medical profession. The consumer point of view should also be better taken into account. A content analysis of email messages for patient communication combined with assessments of email documentation in the EPR could have strengthened present study regarding the importance of email registration in patient records.

In Finland the Ministry of Social Affairs and Health has instigated a major project to safeguard health care [18]. One part of this programme demands that the whole public health care field should be using EPR by 2007 [19]. More research data are needed on electronic communication with patients and on users' experiences. The future EPRs should include a purposeful, safe and secure means of patient communication.