Primary Care Research

Health Coaches Help FPs Improve Chronic Disease Management

April 06, 2015 10:47 am Sheri Porter

Most family physicians would acknowledge that they derive great satisfaction from helping patients with chronic conditions such as diabetes, hypertension and hyperlipidemia better manage their disease.

FP Lydia Leung, M.D., center, works with a medical assistant trained as a health coach to create a care plan during a patient's visit to the University of California, San Francisco, Center for Excellence in Primary Care.

However, because caring for these patients can be time-intensive, another set of hands and a few more hours a day would no doubt be helpful.

Now, new research conducted by the Department of Family and Community Medicine at the University of California, San Francisco, (UCSF) could offer a partial solution.

In an article(annfammed.org) titled "Health Coaching by Medical Assistants to Improve Control of Diabetes, Hypertension, and Hyperlipidemia in Low-Income Patients: A Randomized Controlled Trial," authors explore whether health coaching performed by medical assistants in two safety-net primary care clinics in San Francisco improved patients' control of cardiovascular and metabolic risk factors.

"Prior studies have shown that additional patient support or case management by registered nurses, pharmacists or other licensed professionals can have positive impacts on patients' health," said corresponding author Rachel Willard-Grace, M.P.H., research manager at UCSF's Center for Excellence in Primary Care.

Story Highlights
  • Researchers from the University of California, San Francisco, trained three medical assistants as health coaches.
  • The health coaches worked with patients with diabetes, hypertension and hyperlipidemia to help them better manage their conditions.
  • Patients who had access to the health coaches fared significantly better in achieving clinical goals than did patients who received usual care from one of two safety-net clinics involved in the study.

"We wanted to understand if medical assistants -- who are relatively affordable members of the care team and are ubiquitous across primary care practices -- could be trained as health coaches and help patients improve their health," she told AAFP News.

The research findings are published in the March/April Annals of Family Medicine.

Authors Lay Out Study Methods and Highlights

A total of 441 patients between the ages of 18 and 75 were assigned to the study's coaching/intervention arm or to usual care.

The three female health coaches all were younger than age 40, self-identified as Latina and spoke both English and Spanish. None had graduated from a four-year college, but all had earned diplomas from medical assistant educational programs varying in duration from three months to 12 months.

The health coaches were established as part of the care team -- two coaches at one clinic and one coach at the second, smaller site.

Health coaches met at the clinic with patients before their visits, remained in the exam room during visits, reviewed care plans with patients immediately after their visits, and followed up with them between appointments, noted the authors.

Each coach was assigned as many as 100 patients and contacted those patients at least once each month. Face-to-face visits were scheduled at least every three months.

Patients assigned to receive usual care had full access to all other clinic resources, including scheduled time with their clinician, diabetes educators and chronic care nurses, as well as participation ineducational classes.

The primary outcome was a composite measure of patients in each study arm being at or below goal at 12 months for at least one of three uncontrolled conditions at baseline as defined by hemoglobin A1c, systolic blood pressure and LDL cholesterol levels. Secondary outcomes were defined as meeting all three goals, as well as individual goals.

Researchers reported that coached patients were significantly more likely than patients receiving usual care to achieve both the primary composite measure of meeting one of the clinical goals (46.4 percent versus 34.3 percent) and the secondary composite measure of reaching all clinical goals (34.0 percent versus 24.7 percent.)

In fact, nearly twice as many coached patients achieved the hemoglobin A1c goal as did patients who did not have a health coach (48.6 percent versus 27.6 percent). And coached patients at the larger of the two clinic sites achieved the LDL cholesterol goal at a significantly higher rate than did patients receiving usual care (41.8 percent versus 25.4 percent).

The proportion of patients meeting the systolic blood pressure goal did not differ significantly.

Family Physician Says Patients Benefit

David Thom, M.D., Ph.D., research director for the UCSF Department of Family and Community Medicine, served as principal investigator for the study. He described for AAFP News his personal experience with health coaching outside of the research realm.

"I came into it like a lot of doctors -- a little skeptical. But there are services that would benefit our patients that we just don't have the time or focus to do," said Thom.

Health coaching fell into that category.

"It's not for all patients, but when health coaching has been offered, many of our patients here have tried it and liked it," said Thom. He added that the clinical team still had much to learn about how to best utilize a health coach.

"One reason the model works so well here is because we choose people (for the health coaching role) who have certain characteristics and we train them well. People who step into this role are very committed to working with the physicians and the primary care team. Our health coaches are good at forming relationships and they make themselves available to patients," said Thom.

"When patients are managed by the doctor, we often do so in four or five 15-minute visits a year, and in between those visits we have little or no contact with the patient.

"To think that is optimal is naïve. It's amazing that we've gotten along for so long without using health coaches, and we haven't even realized all the benefits yet," he said.

Thom encouraged family physicians to consider using health coaches, but warned of one familiar sticky obstacle.

"It can be difficult to figure out how to finance this component of health care," he said.

Corresponding Author Adds Context

AAFP News asked Willard-Grace more about the study and health coaching in general.

Q. What findings stood out as particularly interesting or surprising?

A. We asked primary care providers to allow medical assistant health coaches to come into the exam room with them during patient visits. We weren't sure how they or the patients would feel about allowing one more person in the room. It turned out that most providers and patients loved it.

Providers reported that patient visits conducted with coaches in the room were "on a higher level" and rated them as less demanding than visits without a health coach. Coaches also encouraged the patients to speak up with their concerns and questions before the visit ended to avoid leaving confused. The coaches felt that their patients grew in confidence during the course of the study as a result.

Q. How can a physician know that his or her medical assistant coach is adequately trained for the job?

A. There are no formal standards for the training or certification of health coaches. Our training program included 40 hours of instruction and practice provided over a period of several weeks. Coaches had to demonstrate coaching skills in simulated visits and observed real visits, as well as relevant basic medical knowledge through a series of written tests for which we required a score above 90 percent to pass.

When we introduced the health coaches to the clinicians and medical staff at the clinic, we explained the coaches training' and their skills and knowledge. Physicians and other providers were able to observe the coaches' level of training from working with them. To learn more about our training or to access our tools for skills checks, see our website.(cepc.ucsf.edu)

Q. What impact did the practice setting and demographics of the patient population have on the study results, and could your findings be replicated in other settings?

A. We conducted this study at two safety-net clinics serving low-income patients who were largely uninsured or publically insured. As a result, the needs of the patients were greater and resources more limited compared to other settings. More research can help explore how this intervention applies to patients from different practice settings. However, we know from the broader literature that the principles of self-management support and patient-centered communication are effective across a wide variety of settings, so we are optimistic about the translation of this intervention to other places and populations.

Q. What would you like family physicians to take away from this research?

A. Medical assistants are one of the fastest-growing allied health professions and are available and affordable in many settings. Health coaching by medical assistants could provide an answer to the barriers of time, resources and cultural concordance faced by many primary care practices seeking to provide chronic disease support for their patients.

Q. With these research findings in hand, what are the next steps for researchers and policymakers?

A. There is tremendous interest in implementing health coaching programs, but practices struggle to find the funding to make those programs sustainable. Policymakers can help by supporting funding mechanisms that make health coaching financially viable. Additional research can help us understand who benefits most from coaching and what parts of coaching are most important to improve health outcomes.

Related AAFP News Coverage
Diabetes Research
Studies Tout Benefits of Peer Coaching, Care Management Teams

(4/2/2013)

Health Coaching Dramatically Lowers Patients' Systolic Blood Pressure
Expanding Health Care Team Saves Physicians Time

(7/4/2012)

More From AAFP
Patient Self-Management: Frequently Asked Questions(2 page PDF)