Comparing virtual consults to traditional consults using an electronic health record: an observational case–control study

Traditionally, patients receive health care primarily by face-to-face encounters. Some authors suggest that 25% to 70% of all patients seeking care do not need a face-to-face appointment with a physician [1, 2]. “Virtual medicine” using telecommunication and computerized means is being studied in various venues in efforts to provide coordinated care [3–5]. Virtual outreach may reduce the need for face-to-face appointments, reduce unneeded contacts with the health care system, increase patient satisfaction and improve patient health status. Secure, internet-based consultations may complement regular healthcare by providing “better, faster, cheaper” processes for diagnosing, treating, and monitoring health conditions [6–8]. Virtual consults may allow clinicians to receive guidance without added patient cost and inconvenience of physically going to the specialist, and may improve access to sub-specialists by “freeing up” some appointments for use by other patients needing in-person visits (unpublished, Price DW, 2009). Physicians have on average up to two questions about patient care per patient encounter [9–12], but search for answers less than half the time [9, 10, 12, 13], with only a 50-75% success rate in finding answers [10, 12, 14–18]. Electronic “virtual” consultations may help clinicians obtain answers. However, even with increasing availability of electronic communication, less than 5% of consultations occur by email [14]. In at least one study, electronic communication between family medicine physicians and specialists was described as an efficient tool [19]. Electronic communication and increasing use of electronic health records (EHRs) make virtual consultation between physicians a promising method of streamlining care, potentially saving patients the time and expense of added visits. Younger generations of patients, who increasingly use electronic communication and social networking may come to expect virtual models of healthcare [20].

Shershneva and colleagues described a model of how physicians learn from specialty consultations [21]. Earlier pilot work [22] suggests that answers obtained from consultations have the potential to affect care that physicians provide for subsequent patients. Other than this work, to our knowledge, the extent to which information provided during physician-to-physician consultation is used in the care of the index patient or subsequent patients has not been well described. A few studies have investigated virtual consultation using the electronic messaging capabilities of an EHR system within an integrated healthcare environment [23–25]. Researchers have evaluated the use of telemedicine for remote consultations [26–29]. A recent meta-analysis showed a consistent pattern of improved primary care–specialist collaboration when interactive communication methods were employed [30]. A recent demonstration project showed how the use of an EHR has the potential to improve coordination of care between members of a patient’s care team but using an EHR to facilitate inter-office communication is still a barrier [31]. Detailed descriptions are lacking which examine the use of an EHR with an embedded secure messaging system between primary care and specialty care physicians in the context of an integrated health care delivery system.

We conducted an observational study comparing primary care referring physician perceptions on the value of virtual consultations (VCs) compared to traditional consultations (TCs). We quantified and classified the types of VCs and TCs requested, by whom, and for what purpose. We also surveyed patients who had either type of consult to gauge their impression of the care they received. We hypothesized that there would be no difference in physician satisfaction with the consultation process, the immediate clinical utility of the information from the consultant, the use of received information in subsequent patients, and overall satisfaction with the consultation by the physician or by the patient. We also hypothesized that referring physicians would receive information from a VC sooner than from a TC.

Little is known about the use of virtual consultations in an electronic health record. Our study demonstrated that using virtual consults via secure messaging within an EHR did not adversely affect patient’s perceived satisfaction with their care. While less than 7% of all the consults were virtual during the study time, when a provider opted to use a virtual consult to ask a question regarding patient care, 40% of the time a face-to-face consultation with the specialist was not needed. As providers (both primary care and specialists) and our patients gain more experience with this type of consultation, if the rate of virtual consults increases, we may experience improved access to specialists for patients who do require a face-to-face consultation. Patient satisfaction with care may increase if their questions are addressed by specialists without the burden of additional office visits. We do not know the optimal proportion between VC and TCs. With increasing experience, providers may learn which conditions do not need TCs and can be addressed with VCs. By incorporating this knowledge into caring for their other patients, they may improve the efficiency and quality of care they provide for their entire patient population.

The vast majority of study physicians relied on TCs, perhaps out of habit, comfort with traditional means of consultation, or slow adaptation of a new workflow. They seemed to somewhat selectively apply VCs, specifically by opting for TCs for confirmation of diagnosis, initiating a new treatment, or assuming care of the patient. We found that physicians received information from consultants more quickly using VCs than TCs. Otherwise, physicians utilized VCs and TCs for similar types of clinical questions and conditions. Most information from both TCs and VCs was used in care of index patients; many physicians had already or planned to use information from the consultation in the care of subsequent patients. TCs and VCs were minimally disruptive to physician workflow, and there were no differences in referring physician satisfaction between the two types of consultations. Most physicians did not determine patient preference for consult type, and physicians reported that most patients did not volunteer a preference for type of consultation.

Also, most patients did not express a preference for type of consult ordered, perhaps due to being unaware of the option of a VC (for those patients receiving traditional consultation), trusting their physician to determine the most appropriate type of consult, previous experience with consultation, lack of incentive to choose one form of consult over another, or truly having no preference.

Our results suggest that information from consultations is often applied not only in the care of the index patient, but also in caring for subsequent patients. Future application of new knowledge learned from a patient encounter is a desirable outcome of “point of care” learning; helping physicians document, process, and reflect on knowledge learned through practice-based consultation is a potential area for future outcomes-focused continuing medical education (CME) efforts [35]. Planning to use this information in the care of subsequent patients can be considered a type of commitment to change statement; these statements have been shown to be a reasonable predictor of future practice change [36–38].

If additional studies confirm these findings, increased use of VCs may decrease the need for face-to-face specialty encounters without a decrease in the patient’s perception in the quality of care, patient or referring physician satisfaction, with a quicker delivery of useful information to physicians requesting consultation. Using VCs for non-urgent consultation requests could potentially improve access to specialty care visits for patients with more urgent problems.

We believe the results of this investigation provide information useful for other specialties and health systems seeking to design virtual consultations to help streamline patient care. The results show the potential for using novel methods of health information technologies to explore new workflows to improve the affordability of health care, provide decision support to clinicians at the point of care, and inform clinicians how to provide alternative forms of medical care. This study also supports the anecdotal findings of the eHealth Initiative demonstration project which concluded EHR communications improved the referral process between primary care and the specialists [31].

Our study is a beginning step in gaining understanding if novel care delivery models can lead to “better, faster, cheaper” processes of care. In future studies we will need to perform a cost/benefit analysis between traditional and virtual consults to evaluate if virtual consults may lead to improved processes of healthcare at lower costs. If the value of the information a physician obtains via a virtual consult is equivalent to that obtained by a traditional consult, but results in saving the patient time (and potentially cost), then it may benefit both the patient and the physician. Therefore a virtual consult may be another way to potentially and conveniently meet patient needs and give physicians another tool for improving their practices.

In future work we plan to include additional incentives to increase percent of surveys returned, examine reasons physicians choose virtual consults for individual patients, assess the consistency of consultation advice with evidence, explore the applicability of our findings to other integrated health care systems as well as systems without electronic medical records or with lesser degrees of EHR implementation, and examine the patient outcomes of virtual consultations.

TEP is the physician manager for clinical reporting for the Colorado Permanente Medical Group (Denver CO) and a clinician researcher in the Institute for Health Research. He is the chair of the committee for integrating new technology with benefits and operations for Kaiser Permanente Colorado. Dr. Palen’s research focus is on the use of health information technology for enhancing models of care delivery.

DP is Director of Medical Education for the Colorado Permanente Medical Group (Denver CO) and the Permanente Federation (Oakland CA). He is also co-Director of the Kaiser Permanente Colorado Institute for Health Research's Center for Health Education, Dissemination and Research, and a member of the Board of Directors of the Accreditation Council on Continuing Medical Education. Among Dr. Price's areas of focus are examination of methods of effective physician learning, newer methods of learning for health professionals, and improving the translation of continuing education learnings into practice.