Automated Clinician Prompts and Referrals Facilitate Access to Counseling Services, Leading to Positive Behavior Changes Among Patients

Problem Addressed

• Unhealthy behaviors lead to high mortality: Smoking, physical inactivity, poor diet, and problem drinking account for approximately 37 percent of deaths in the U.S. population.¹
• Clinician advice considered influential, but often lacking: Behavior modification can be difficult, but individuals who succeed in modifying unhealthy behaviors cite physician recommendations as an important motivator.² However, due to time constraints, physicians often fail to provide such counseling. For example, one study found that only 40 percent of physicians assessed the physical activity status of patients.³ Another study of 12,835 obese adults found that only 42 percent were advised by their physician to lose weight.⁴ Furthermore, only 2 to 5 percent of patients are referred to community resources that offer behavioral counseling assistance.⁵
• Patients may not proactively pursue counseling: Even if physicians recommend behavioral change and external counseling services, patients often do not follow up on these recommendations on their own.⁶

Description of the Innovative Activity

• Based on "5 As" model: The program is based on the “5 As” (ask, advise, agree, assist, arrange), a decades-old clinical model that has been embraced by many organizations to encourage change in health-related behaviors.
• Introduction of counseling services: At check-in or in the examination room, clinicians and staff members distribute a one-page flyer describing the different counseling services, including three intensive options (group counseling, telephone counseling, and computer care; see bullet on automated referrals below for more details on these programs). Use of the handout is optional; some clinicians prefer to verbally describe the options instead.
• Data gathering: Once the patient is in the examination room, a staff member takes the patient's vital signs, measures height and weight, and inquires about smoking status (“ask”); the staff member enters this information into the EMR. Staff do not query patients about alcohol use/misuse; this information is gathered by reviewing previous visit documentation in the EMR (information is dependent on whether physicians happened to ask about alcohol use and documented the response).
• EMR prompt to clinician: When the physician comes in to see the patient, he or she calls up the patient’s record on the EMR. When information in the record reveals the presence of unhealthy behaviors, the EMR automatically prompts the clinician to consider providing counseling and referral services. The prompt, generated by a specially developed application (eLinkS), lists the unhealthy behaviors and asks the clinician if the Prescription for Health form should be loaded.
• Prescription for Health form: If the clinician answers “yes” to the prompt, eLinkS uploads a form that the clinician can use to document care and to initiate a patient referral to appropriate counseling services. The form has boxes that can be checked by clinicians to document the following: whether they advised the patient to change his or her behavior (“advise”); whether the patient was willing to change behavior and seemed engaged in the process (“agree”); whether the patient wanted to be referred outside of the practice to one of the intensive counseling options (“assist”); and by what means (telephone contact, office visit, email contact, or none) the patient wanted to follow up with the practice (“arrange”). Those who do not agree to outside counseling are considered to be receiving "usual care," which consists of any other alternative chosen by the patient and clinician (e.g., counseling by the clinician, pharmacotherapy, or no intervention). Clinicians may choose to click a button on the form labeled “View Patient Counseling Script” to review scripts that can be used to assist in providing advice to patients. Clinicians can also click a button labeled “Add Translation to Note” to transfer the information from the form to the encounter note, thereby automatically documenting the counseling in the medical record.
• Automated referral to community-based counseling programs: When the clinician closes the form, eLinkS automatically e-mails a patient referral form to a designated counseling center with whom a referral relationship has previously been arranged. The counselor at the center logs onto a secure Web site (using a username and password) to view the patient’s contact information. The community-based options (along with data on patient preferences for them) are described below (note: counseling options can be tailored to services available in the community):
  • Group counseling: Group counseling is offered in by a commercial weight loss program (Weight Watchers) for overweight and obesity, by the Riverside Hospital System wellness center for smoking, and by local Alcoholics Anonymous meetings for risky drinking behavior. (Referrals to Alcoholics Anonymous are not automated; the clinician prints a list of available meetings to hand to the patient.)
  • Telephone counseling: Trained counselors from the University of Kentucky Health & Wellness Behavior Health Improvement Program (BeHIP) provide telephone counseling for smoking cessation and weight loss.
  • Computer care: Computer care provides patients with access to the My Healthy Living Web site(developed through the Prescription for Health program; see the Context section for more information), which includes about 200 local and national resources for patient self-management; computer care also includes an electronic counseling service, developed by University of Kentucky counselors, which follows BeHIP’s telephone counseling protocol but allows online communication via a secure Web portal.
  • Patient preferences: Participants indicated varying preferences for counseling services based on the behavior addressed; in general, group care and telephone counseling were preferred (computer care was rarely selected). For weight loss, roughly 70 percent of patients preferred group care, while approximately 30 percent preferred telephone counseling. For smoking cessation, approximately 70 percent of patients preferred telephone counseling, while approximately 30 percent preferred group classes.
• Proactive contact to encourage participation: With the exception of referrals to Alcoholics Anonymous, representatives of the community-based programs proactively contact the referred patient to encourage participation and arrange for counseling services.
• Counselor updates to patient record: At any time, the counselor can log onto the Web site to enter followup information; the Web site automatically sends information regarding the patient’s progress back to the EMR so that it can be accessed by the patient’s clinician.
• Cost of counseling services: Grant funding allowed the program to provide up to 9 months of intensive counseling services free of charge to those patients who enrolled during a 5-week period (April 16, 2006 to May 22, 2006), after which the demand for referrals exceeded the project budget. Since May 22, 2006, anyone referred to a counseling program must pay for those services out of their own pocket or through their own insurance coverage, although the program's developers are currently working with insurers to arrange for expanded coverage of services.

References/Related Articles

Contact the Innovator

Results

• Patient referrals: During a 5-week recruitment period, the practices saw 5,679 patients, 4,030 of whom had unhealthy behaviors. Clinicians decided to use the prompt system for 576 encounters, referring 400 patients for external counseling services. Reasons why providers chose not to use the system for certain encounters included that the patient didn't want to change their health behaviors; it was inappropriate for the clinician to address health behaviors at the encounter; and the clinician didn't have time to address the health behaviors. While no pre-implementation data on referral rates at the practice sites is available, this referral rate is higher than published estimates of general referral rates for smoking cessation and weight loss (typically 2 to 5 percent or lower).
• Improvements in diet: Compared to baseline survey results, 4-month survey results found that more patients reported eating five daily servings of fruits or vegetables (11 percent vs. 9 percent), limiting sweets (33 percent vs. 29 percent), and limiting fatty foods (47 percent vs. 43 percent).
• Improvements in physical activity: Compared to baseline, 4-month survey results show that  patients reported engaging in more vigorous activity (168 vs. 141 metabolic units of activity or METs), moderate activity (305 vs. 253 METs), and walking (366 vs. 332 METs).
• Weight loss: Individuals referred for weight management counseling reported losing an average of 7 pounds at 4 months post-referral, compared to no weight loss for patients receiving usual care.
• Higher quit rates for smokers: Depending on the type of smoking cessation intervention, 50 to 75 percent of patients reported no longer smoking at 4 months after referral, compared to 25 percent of patients receiving usual care.
• No impact on alcohol abuse/misuse: The system had no effect on the rate of alcohol use or misuse. The program developer believes that this could be due to several factors:
  • The documentation of alcohol misuse in the EMR might be limited, leading to incomplete identification of patients abusing alcohol.
  • No automated prompt was sent to the local program offering counseling services, and thus patients were not contacted proactively about participating.
  • Counseling services offered for alcohol use were not as extensive as for weight loss and smoking cessation.
  • Alcohol abuse may be a more difficult behavior to change. Many alcoholics do not want to stop drinking, while smokers and people with obesity are often more motivated to change their behaviors.
• Improvements sustained over time: Survey data from 9 months after program implementation show gains that are similar to those achieved at 4 months.

Evidence Rating (What is this?)

Context of the Innovation

Planning and Development Process

• Application design: The eLinkS application was designed by researchers at Virginia Commonwealth University based on consultation with clinicians of various types at the practice. It is important to understand what would be acceptable and helpful to clinicians in order to design electronic tools that make the job easier.
• Identifying and partnering with community-based organizations: The research team identified potential community-based partners by asking Riverside Medical Group physicians about their current use of such services. The research team then contacted these organizations to introduce the program, discuss service offerings, and set up a system for patient referrals.
• Programming the EMR: The eLinkS application was programmed into the EMR by the Riverside Medical Group's informatics staff; no outside vendor assistance was needed. Programming was relatively simple, involving pulling basic information from the EMR, designing the user interfaces (e.g., screens with prompts and forms), and automating a process to send that information to a secure Web site.
• Training: The eLinkS application is very simple, with basic prompts for clinicians; a 1-hour training session was held at each of the nine practices to ensure that clinicians knew how to use the system and understood the services offered by the counseling organizations. The process was also explained to counselors, who were provided with user names to allow them to access the secure Web site.

Resources Used and Skills Needed

• Staffing: No new staff were hired for the program. Existing practice-based informatics staff modified the EMR to include the eLinkS application. A small amount of research staff time was needed to identify community resources and train physicians and counselors.
• Technology: The practices already had an EMR in place.
• Costs: Program costs were minimal. Counseling services were provided by external organizations; Riverside used the RWJF grant to cover the cost of the counseling services for a 5-month period and an evaluation of the program's impact.

Funding Sources

Agency for Healthcare Research and Quality; Robert Wood Johnson Foundation

Getting Started with This Innovation

• Identify the health behaviors to be targeted: Adopters can review their patient populations to determine which health risks are most prevalent.
• Identify a variety of local resources: A selection of services should be offered to accommodate patient preferences; some patients prefer self-management through computer-prompted care, whereas others might like the convenience of telephone counseling and still others might prefer the more supportive environment offered by group visits.
• Recognize that patient cost is a barrier to participation: More patients are willing to undergo counseling services if those services do not require them to pay out of their own pockets. Pursuing grant funding or expanded insurer coverage of services will enhance program participation. In addition, adopters can attempt to identify free services to recommend to patients. For example, the Virginia State Department of Health offers free smoking cessation telephone counseling (funded by tobacco lawsuit settlement funds) through the state’s “Quit Line.”
• Sell the idea by emphasizing the benefits of linkages between physicians and community organizations: Because they typically work individually and are overburdened with existing cases, physician practices and community counseling organizations may view it as burdensome to have to work more closely together. Nevertheless, such a partnership can offer a “win-win-win” to participants: clinicians gain access to tools and resources to help their patients achieve behavior change; community counseling centers can gain access to new patients; and patients can improve their health while clinicians are kept informed of their progress.
• Obtain physician buy-in by emphasizing patient care benefits: Physicians are busy and often resist new care processes. Physicians should be assured that an automated counseling referral system will add only a few minutes to each visit but can have a meaningful impact on patient health, thereby representing a value-added service enhancement for patients.
• Determine physician comfort with EMR use: Implementation is easier in practices in which an EMR is in place and physicians are very comfortable using it.

Sustaining This Innovation

• Monitor and adapt to organizational and service changes: Over time, new clinicians and counseling organizations may enter the market, while others may exit it. In addition, service offerings in community counseling centers frequently change. Would-be adopters need to develop a system to monitor and adapt to these changes.
• Match the referral process to meet existing systems: While eLinkS is an example of an automated referral communication system, practices can develop other counseling referral tools that fit their own resources and systems.