Ten Steps to Establishing an e-Consultation Service to Improve Access to Specialist Care

A project of this nature requires (1) clinical champions, consisting of family physicians and specialists to advocate the need for and efficacy of the system, (2) health region buy-in to provide infrastructure and human resource support, and (3) technical/e-health support to facilitate system development, maintenance, and technical support. Much of the infrastructure and assistance required may already be established through a hospital or a local/regional health network. Look to partner with other organizations in order to minimize the financial and technical burdens of planning and deploying the infrastructure/platform to host the e-consultation application. Our team originally comprised two physicians who had identified a need and developed the original proposal upon which the e-consultation system was based. The core team now consists of a specialist physician (at Ottawa Hospital), a research coordinator and an academic PCP (at Bruyère Research Institute), the Chief Information Officer at one of the regional hospitals (Winchester District Memorial Hospital) where the infrastructure is hosted, and a project manager from the Champlain LHIN based in Ottawa, ON.

Technology designed to facilitate communication and collaboration between physicians is advancing. The choice of system should be based on local existing infrastructure in order to enable sustainability and scalability, and this is facilitated by ensuring key partners are part of the initial project team. Many different types of computer systems are used in healthcare practices. In Canada, most hospitals and community organizations are separate enterprises, with each typically using internal systems and tools that do not easily lend themselves to collaboration with external partner organizations or individuals. To find an application able to integrate seamlessly with diverse systems is often challenging. More integrated business solutions across multiple healthcare agencies are not expected to come for some time.

It is also important to acknowledge the existing challenges when attempting to implement technology to facilitate communication within the healthcare setting. As physicians have limited time and may not use e-consultation on a daily basis, the system should be user friendly, simple, and to the point, making it easy to learn and re-learn.

We based our e-consultation service, including associated forms, automated workflows, and interface engines, on SharePoint¹³ extranet. This system was widely available and had been established by the Champlain LHIN as a secure regional collaboration space, which provided a unique solution in the region that addresses the issues of usability and deployment of electronic/automated processes and interactions across organizations. The registration of users is controlled through the LHIN but is not limited to specific organizations, thereby facilitating collaboration with individuals or organizations outside the health network. The platforms deployed for the extranet provide a secure, robust, and scalable environment that supports a wide range of capabilities, including:

We started with a small core group representing the key partners, including both primary care providers and specialists. Starting with a small group will allow you to identify many of the obstacles before the system is sufficiently large that they prove extremely difficult to manage. We used a participatory-type approach to develop the system in an iterative, localized manner. We met monthly with our initial primary care providers as part of their regular monthly meetings. Starting small and engaging users early on in the development process were critical to our success. We conducted an in-depth qualitative evaluation (focus groups and interviews) with our initial users.¹¹ This approach enabled us to develop the system based on the user's needs and test several versions of the form, assess workflows, and identify and address the user's concerns related to privacy, liability, and remuneration.

We started with 14 PCPs and 11 specialists who represented four specialty areas: endocrinology, nephrology, neurology, and rheumatology. Other specialties were gradually added to the system to a total of 28 specialists representing 12 specialties in April 2011. The specialties were generally added based on recommendations from PCPs and included dermatology, cardiology, internal medicine, thrombosis, ear/nose/throat, general pediatrics, and obstetrics/gynecology; one specialty, pediatric hematology/oncology, was added based on the core team being approached by the specialty to get added.

An online form is the simplest solution as it can be used by participants anywhere with Internet access (Appendix, Figs. A1 and A2). The use of specialized forms that enable data to be sent securely over the Internet is an asynchronous technique, which suits the busy schedules of both the PCPs and the specialists using the service. Three different form formats were explored with our initial user group: free text, directed form (mandatory fields) including prompts for additional information, and existing EMR forms. There are many varieties of consultation forms from which the e-consultation system can be based.¹⁴ Examples of consultations forms currently used by PCPs were drawn upon when designing e-consultation forms. System users (both primary care and specialists) were consulted to determine a format preference. Most users preferred forms that most closely aligned with the way providers were completing current consultations and thus would allow for a free text question, the ability to attach files, and the option of free text for additional patient information. For users who already do consults within a computer system, this also allows for them to either attach usual consult form or do a “cut and paste” from an existing Word document.

We also included three tips for both primary care providers (reason for consultation, specific treatments already prescribed, suggestions for possible treatments) and specialists (reasons for recommendation, suggestions for possible treatments, suggestions for tests and recommendations according to the results of the test) to “remind” them of the key components of quality consult-referral discussion.

Security is of utmost concern when using Internet-based information-sharing programs in the healthcare system. Electronic transfer of patient information can jeopardize a patient's right to privacy if the proper precautions are not taken.^15–17 The e-consultation platform described in this article utilized a secure region-wide information-sharing network. Privacy impact and threat risk assessments were performed in compliance with the Personal Health Information Protection Act.¹⁸ The process took about 3 months to obtain the necessary approvals and roughly 6 months for all documentation to be finalized. This work was facilitated by working collaboratively with the health region and partnered hospitals. Most health regions and healthcare organizations will have a privacy office/officer who can assist with the process of conducting a privacy impact assessment, whereas threat risk assessment of the technology platform is usually performed by independent third-party experts.

The workflow of the e-consultation system must fit as seamlessly as possible into the physician's usual workflow to ensure participation. It is important to minimize system usage time. The physician should be able to enter, or complete, a consult request in under 10 min. Again, early consultation with potential system users was a key enabler to understanding workflow issues. Below (and shown in Appendix, Fig. A3) is a description of the workflow used in our e-consultation system.

The PCP accesses the e-consultation service through a computer that is connected to the Internet, using his or her unique username and password. After logging into the system, the PCP will open and complete the e-consultation request form (Appendix, Fig. A1), which includes the date of request, specialty needed for consultation, patient's demographic information, and the body of question. On the same page, PCPs have the option to attach additional files such as pictures, test results, etc. Following submission, two automatic notification e-mails are generated: (1) to the requestor to confirm successful submission and (2) to the “Assigner” for the specialty to let him or her know of a new request pending assignment to the appropriate specialist—for the pilot, this role was primarily filled by the LHIN Project Manager. It is important to note that the e-mails do not contain any information related to the case itself, other than identifying the requesting PCP and the target specialty. The project manager logs into the system and assigns the request to a specialist based on the specialty required and specialist availability. The availability of the specialists varied and the scheduling depended on size of the group and personal preference. In some instances, there is only one specialist, and generally this specialist can receive notifications and access the cases anytime/anywhere, so availability is not an issue—in the rare situation where they may be unreachable, PCPs are notified of a potential delay in receiving a response. Where there are only two or three specialists, assignment rotates between the specialists, and specialists contact the manager if they plan to be away for any extended period of time. Where a larger group is involved, the manager is provided with updated e-consultation coverage schedules of availability from the relevant departments. This assignment prompts an automatic e-mail to the assigned specialist to let him or her know of the consultation request pending his or her review and response within 1 week—once more, the e-mail does not contain any other information about the case. The specialist logs into the e-consultation system using his or her own username and password and answers the consultation request (Appendix, Fig. A2). At this point, the PCP receives an automatic e-mail message notifying of the consultation response now being available. The PCP would then log back into the e-consultation system to view the response from the specialist, following which the PCP could ask a follow-up question or close the case. Closing the case includes completion of a short survey that is presented at the bottom of the e-consultation form to the PCP about the specific consultation. Once the survey response is complete, a .pdf transcript of the case is generated by the system and made accessible to both the PCP and the specialist, with both receiving an automated notification e-mail regarding the existence of the file.

Early in the process, two main issues were identified with regard to fitting the e-consultation forms in the physician's workflow. First, PCPs wanted the option of delegating preparation, closure, and filing of the e-consultation request to an assistant or office administrator. This was addressed by adding the possibility for “delegates” to take any necessary action in the e-consultation system on behalf of the PCP. Second, PCPs using EMRs wanted to be able to generate the e-consultation request directly from the EMR. We invited any interested clinics to participate in a proof-of-concept phase by engaging their EMR vendors and working with the project team to develop the necessary interfaces for this purpose. Two clinics and their vendors were subsequently involved in the implementation of the proof-of-concept, and the ability to initiate e-consultations directly from the EMR was successfully completed and tested. However, it became apparent that many physicians found the initiation of the requests directly from the e-consultation Web site more convenient, and therefore the integrated EMR solution was not further pursued or enhanced at that point. Most clinicians have been generating an electronic file from their EMR and attaching it to the Web-based form when needed/appropriate.

Physician participation will ultimately determine the success of e-consultation. Keeping the physician engagement aspects—joining, training, and ongoing support—simple is important.

Recruitment of physicians into projects of this kind is often challenging. Solidifying clinical champions within the project team is the first important step in recruiting physicians, as it is their connections and willingness to campaign for the system that will have the most significant influence on physician participation.

In addition to the recruitment efforts of our PCP and specialist champions, multiple forms of advertisement, including the distribution of leaflets, booths at continuing medical education events, and word of mouth, were used to maximize reach to PCPs. We also had the opportunity to present at several local public forums (a regional health board meeting, health technology forums) and had articles published in local papers following the presentations that generated much local interest leading to participation.^19,20 PCPs are the initiators in the process of e-consultation, and their full engagement dramatically boosts the number of e-consultations.

Specialists were recruited into the project primarily by the specialist on our project team as well as through participation in continuing medical education events and in-hospital rounds. The selection of the type of specialists who were needed on the service was based on the referral needs of the primary care community; thus this may differ in each region depending on access issues. We also experienced organic growth with both specialists and PCPs who joined following recommendation from their peers.

Initially, our e-consultation project manager met with both the PCPs and specialists to give individualized training on navigating the Web site and use of the e-consultation forms. The training sessions took approximately 20 min. Physicians were also provided with a training manual for future reference and a “cheat sheet” card summarizing the steps to completing a referral shown in Figure 3. We also offered to return to the physicians' offices for additional training if they were having difficulties with any aspects of the process.

At present, most of the training sessions are conducted remotely by phone if the provider user already has a basic comfort level with computer use. System support is required at the onset of the project and in training system users and continues in the form of troubleshooting and technical assistance for the duration of time that the system is online.

As with any online service, there must be a central hub to receive and respond to questions and concerns from the users. The position requires someone who is able to respond to requests quickly and thus has availability during business hours. The types of questions posed are usually of a nontechnical nature (forgotten password, cannot remember how to upload a document, submitted request to wrong specialty, etc.) and thus requires someone with balanced proficiency in customer service practices and technical know-how.

In consultation with our initial specialist users, we established a billing rate of $200/h for their service. This rate was based on the average current remuneration of specialists in Ontario for a face-to-face consultation, which is approximately $150 for all specialties listed. The specialist e-consultation form ended by asking them to indicate how much time they spent on their response to the question (Appendix, Fig. A2). They were given time slot selections of 1–10, 10–15, 15–10, and >20 min. If >20 min was selected, the specialist was asked to provide an explanation. Time submitted by specialists across all specialties averaged at 11.7 min per case.

Funding for remunerations for specialists was originally obtained from a TOHAMO research grant and is subsequently being subsidized through the Department of Medicine at The Ottawa Hospital. Long-term compensation for this type of virtual medicine service will need to be supported through physician payment services contracts at the provincial level.

We also consulted with the Canadian Medical Protective Association to ensure that the e-consultation system did not create any novel litigation risks, as the potential for these issues was a concern among specialists during the recruitment phase of the project. The advice received from the Canadian Medical Protective Association was that the duty of care in this project is no different than any other “curbside” or telephone consultation request and that no novel litigation risks existed provided the physician had “enough information to make [his/her] recommendation.” The e-consultation forms were tailored to provide the option to specialists to ask for more information or decline to provide a recommendation.

Obtaining feedback was critical to providing a service that participants would find useful and actively engage in. We built into the e-consultation form five questions to assess impact of the service (Appendix, Fig. A4).

In addition to seeking physician feedback, system efficacy was determined by continuously evaluating usage and outcomes. Measures used to track usage included the number of participating physicians, the number completed e-consultation requests per registered PCP, and the proportion of requests per specialty. The response time for consultation requests and the self-reported amount of time spent by specialists on the response were also tracked. Outcome measures included the specialist recommendation (e.g., further investigations, treatment initiated, consultation avoided) and physician satisfaction with the process.

We are also providing regular updates to the specialists and include the number of referrals and summary feedback including comments from the PCPs. As the feedback has been very positive, being able to highlight the value of the service provided by the specialists is a way of encouraging ongoing participation.

Three key elements are needed to ensure sustainability: (1) maintaining the benefit of the service (perceived high clinical value), (2) funded infrastructure, and (3) human resource support.^21,22

The pilot phase of our e-consultation project was funded through a small research grant. These funds helped cover the start-up costs, including development of forms, system workflow, and user training, as well as everyday operational costs. The success of the pilot project enabled the project team to obtain additional bridge funding from the LHIN and hospital. The transition of e-consultation into a sustainable program comes with the realization that secured funding of the associated costs is required.

Once the system has been established, sustainability of e-consultation primarily requires human resources support. Ongoing staffing is required for project coordination and management, information technology–related tasks such as system updates, and end-user technical support, which includes assistance with system troubleshooting and recovery of login information. Funds are also required for ongoing physician engagement-related functions, such as recruiting and training.