Health Coaching by Medical Assistants to Improve Control of Diabetes, Hypertension, and Hyperlipidemia in Low-Income Patients: A Randomized Controlled Trial

Rachel Willard-Grace; Ellen H. Chen; Danielle Hessler; Denise DeVore; Camille Prado; Thomas Bodenheimer; David H. Thom

doi:10.1370/afm.1768

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1
CONSORT diagram.

appt = appointment; CONSORT = Consolidated Standards of Reporting Trials.

View popup

Table 1

Training of Health Coaches: Topics and Skills

Collaborative communication

Eliciting patient knowledge and motivation

Closing the loop (teach back)

Setting the agenda

Reviewing laboratory numbers

Action plans

Assessing patient motivations

Exploring ambivalence

Disease-specific knowledge

Basics of study conditions

Monitoring control

Medication management

Medication adherence

Assessing patient knowledge and motivation

Addressing barriers to adherence

Working with primary care clinicians

Taking notes during a visit

Checking understanding

Offering help

Advocating for a patient’s agenda

Community and clinic resources

Behavior health

Physical activity and healthy eating

Smoking cessation

Social determinants of health

View popup

Table 2

Characteristics of Participants, Overall and by Study Arm

Characteristic	All Participants (N = 441)	Study Arm
Characteristic	All Participants (N = 441)	Usual Care (n = 217)	Health Coaching (n = 224)
Demographic characteristics
Clinic site, % (No.)
Clinic site A	75.3 (332)	76.0 (165)	74.6 (167)
Clinic site B	24.7 (109)	24.0 (52)	25.4 (57)
Age, mean (SD), y	52.7 (11.1)	52.9 (11.5)	52.6 (10.7)
Sex (female), % (No.)	55.3 (244)	52.2 (127)	58.5 (117)
Married/long-term relationship, % (No.)	53.1 (234)	57.1 (124)	49.1 (110)
Born in the United States, % (No.)	25.6 (113)	24.9 (54)	26.3 (59)
Years in United States, mean (SD)^a	18.2 (11.2)	17.9 (11.9)	18.5 (10.4)
Primary language, % (No.)
English	27.7 (122)	26.7 (58)	28.6 (64)
Spanish	68.7 (303)	69.5 (151)	67.9 (152)
Other	3.6 (16)	3.7 (8)	3.6 (8)
Race/ethnicity, % (No.)
Asian	4.1 (18)	5.1 (11)	3.1 (7)
African American	19.0 (84)	18.4 (40)	19.6 (44)
Latino or Hispanic	70.1 (309)	71.0 (154)	69.2 (155)
White	2.5 (11)	2.3 (5)	2.7 (6)
Other	4.3 (19)	3.2 (7)	5.4 (12)
Working status, % (No.)
Full time	18.6 (82)	16.1 (35)	21.0 (47)
Part time	25.6 (113)	26.3 (57)	25.0 (56)
Homemaker	13.8 (61)	17.1 (37)	10.7 (24)
Unemployed	16.1 (71)	16.1 (35)	16.1 (36)
Retired/disabled/SSI/other	25.9 (114)	24.3 (53)	27.2 (61)
Education, % (No.)
≤5th grade	22.7 (100)	23.1 (50)	22.3 (50)
6th to 8th grade	21.1 (93)	20.7 (45)	21.4 (48)
Some high school	13.4 (59)	12.4 (27)	14.3 (32)
High school graduate or GED	17.7 (78)	16.6 (36)	18.8 (42)
Some college	15.6 (69)	19.4 (42)	12.1 (27)
College graduate	9.5 (42)	7.8 (17)	11.2 (25)
Income, % (No.)
≤$5,000	34.0 (150)	31.3 (68)	36.6 (82)
$5,000–$10,000	24.3 (107)	25.3 (55)	23.2 (52)
$10,000–$20,000	29.5 (130)	29.0 (63)	29.9 (67)
≥$20,000	12.2 (54)	14.2 (31)	10.2 (23)
Number of PCP visits in year before study, mean (SD)	5.4 (3.9)	5.5 (4.3)	5.4 (4.1)
Clinical characteristics
BMI, mean (SD), kg/m²	31.4 (6.7)	31.4 (6.3)	31.5 (7.0)
HbA_1c, mean (SD), %^b	9.9 (1.5)	10.0 (1.4)	9.8 (1.5)
LDL cholesterol, mean (SD), mg/dL^b	147.0 (35.6)	147.8 (34.1)	146.3 (36.9)
SBP, mean (SD), mm Hg^b	159.4 (15.4)	160.9 (16.8)	157.7 (13.5)
Uncontrolled at baseline, % (No.)
For 1 condition	72.6 (320)	73.3 (159)	71.9 (161)
For 2 conditions	23.6 (104)	23.5 (51)	23.7 (53)
For 3 conditions	3.9 (17)	3.2 (7)	4.5 (10)
For HbA_1c	35.8 (158)	33.6 (73)	37.9 (85)
For SBP	43.5 (192)	46.5 (101)	40.6 (91)
For cholesterol	51.9 (229)	49.8 (108)	54.0 (121)

BMI = body mass index; GED = general equivalency diploma; HbA_1c = hemoglobin A_1c; LDL = low-density lipoprotein; PCP = primary care provider; SBP = systolic blood pressure; SSI = supplemental security income.
↵a For the 328 participants born outside the United States.
↵b Includes only patients qualifying for the study on this measure (158 for HbA_1c, 218 for LDL, and 192 for SBP).

View popup

Table 3

Characteristics of Participants, Overall and by Clinic Site

Characteristic	All Participants (N = 441)	Clinic Site
Characteristic	All Participants (N = 441)	Clinic Site A (n = 332)	Clinic Site B (n = 109)	P Value
Demographic characteristics
Age, mean (SD), y	52.7 (11.1)	52.3 (11.3)	54.0 (10.5)	.18
Sex (female), % (No.)	55.3 (244)	53.3 (177)	61.5 (67)	.14
Born in the United States, % (No.)	25.6 (113)	6.0 (20)	85.3 (93)	<.001
Primary language, % (No.)				<.001
English	27.7 (122)	7.8 (26)	88.1 (96)
Spanish	68.7 (303)	89.5 (297)	5.5 (6)
Other	3.6 (16)	2.7 (9)	6.4 (7)
Race/ethnicity, % (No.)				<.001
Asian	4.1 (18)	3.3 (11)	6.4 (7)
African American	19.0 (84)	1.8 (6)	71.6 (78)
Latino or Hispanic	70.1 (309)	90.7 (30)	7.3 (8)
White	2.5 (11)	1.5 (5)	5.5 (6)
Other	4.3 (19)	2.7 (9)	9.2 (10)
Working status, % (No.)				<.001
Full time	18.6 (82)	22.0 (73)	8.3 (9)
Part time	25.6 (113)	28.0 (93)	18.3 (20)
Homemaker	13.8 (61)	16.9 (56)	4.6 (5)
Unemployed	16.1 (71)	13.6 (45)	23.9 (26)
Retired/disabled/SSI/other	25.9 (114)	19.5 (65)	44.9 (49)
Education, % (No.)				<.001
≤5th grade	22.7 (100)	29.6 (98)	1.8 (2)
6th to 8th grade	21.1 (93)	27.1 (90)	2.8 (3)
Some high school	13.4 (59)	11.1 (37)	20.2 (22)
High school graduate or GED	17.7 (78)	14.2 (47)	28.4 (31)
Some college	15.6 (69)	9.9 (33)	33.0 (36)
College graduate	9.5 (42)	8.1 (27)	13.8 (15)
Income, % (No.)				.52
≤$5,000	34.0 (150)	33.7 (112)	34.9 (38)
$5,000–$10,000	24.3 (107)	23.8 (79)	25.7 (28)
$10,000–$20,000	29.5 (130)	31.0 (103)	24.8 (27)
≥$20,000	12.2 (54)	11.4 (38)	14.7 (16)
Clinical characteristics
BMI, mean (SD), kg/m²	31.4 (6.7)	30.6 (4.7)	34.0 (10.2)	<.001
HbA_1c, mean (SD), %^a	9.9 (1.5)	9.8 (1.4)	10.3 (1.9)	.17
LDL cholesterol, mean (SD), mg/dL^a	147.0 (35.6)	147.3 (36.7)	145.2 (28.9)	.75
SBP, mean (SD), mm Hg^a	159.4 (15.4)	158.3 (13.5)	160.9 (17.5)	.25
Number of PCP visits in year before study, mean (SD)	5.4 (4.1)	5.2 (3.5)	6.1 (5.5)	<.05
Number of coach interactions, mean (SD)	12.4 (7.4)	14.1 (6.7)	7.6 (7.2)	<.001
Total time interaction, mean (SD), min	540.8 (307.6)	621.1 (281.1)	305.5 (258.7)	<.001
Number of interactions by topic/activity addressed, mean (SD)
Medications	9.3 (6.0)	11.0 (5.4)	4.1 (4.6)	<.001
Reviewing clinical values and goals	7.4 (4.6)	8.4 (3.9)	4.5 (5.2)	<.001
Discussing lifestyle changes	7.4 (5.0)	9.1 (4.3)	2.1 (2.2)	<.001
Agenda setting	5.7 (4.3)	7.1 (4.1)	1.7 (1.9)	<.001
Navigational support	5.1 (5.1)	6.4 (5.3)	1.5 (1.8)	<.001
Action plans for behavior change	4.2 (3.2)	5.1 (2.9)	1.6 (2.3)	<.001
Facilitating communication with PCP	3.3 (3.8)	4.0 (4.1)	1.1 (1.5)	<.001
Closing the loop (teach back)	5.3 (3.6)	6.1 (3.3)	2.8 (3.4)	<.001
Patient-reported quality of health-coaching interactions, mean (SD)^b	3.3 (0.7)	3.4 (0.5)	2.8 (0.9)	<.001
Patient-reported trust in health coach, mean (SD)^c	4.1 (0.5)	4.1 (0.5)	3.9 (0.7)	<.005

BMI = body mass index; GED = general equivalency diploma; HbA_1c = hemoglobin A_1c; LDL = low-density lipoprotein; PCP = primary care physician; SBP = systolic blood pressure; SSI = supplemental security income.
↵a Includes only patients qualifying for the study on this measure (158 for HbA_1c, 218 for LDL cholesterol, and 192 for SBP).
↵b On a scale of 1 to 5, where 5 = best.
↵c On a scale of 1 to 5, where 5 = high.

View popup

Table 4

Primary Composite and Condition-Specific Outcomes by Study Arm, for Total Sample and by Clinic Site

Outcome	Health Coaching, % (No./n)	Usual Care, % (No./n)	Difference, % (95% CI)	P Value
Total sample
Composite (primary)^a	46.4 (90/194)	34.3 (57/166)	12.1 (2 to 23)	.02
Composite (secondary)^b	34.0 (66/194)	24.7 (41/166)	9.3 (1 to 19)	.05
HbA_1c goal achieved^c	48.6 (36/74)	27.6 (16/58)	21.0 (5 to 39)	.01
Cholesterol goal achieved^d	42.7 (41/96)	32.0 (24/75)	10.7 (−4 to 25)	.15
SBP goal achieved^e	23.8 (19/80)	28.9 (22/76)	−5.1 (−9 to 19)	.46
Clinic site A
Composite (primary)^a	49.7 (73/147)	32.8 (42/128)	16.9 (5 to 29)	.01
Composite (secondary)^b	36.1 (53/147)	24.2 (31/128)	11.9 (1 to 23)	.03
HbA_1c goal achieved^c	52.3 (34/65)	29.4 (15/51)	22.9 (5 to 41)	.01
Cholesterol goal achieved^d	41.8 (33/79)	25.4 (16/63)	16.4 (1 to 32)	.04
SBP goal achieved^e	25.5 (12/47)	31.9 (15/47)	−6.4 (−11 to 25)	.49
Clinic site B
Composite (primary)^a	36.2 (17/47)	39.5 (15/38)	−3.3 (−17 to 24)	.76
Composite (secondary)^b	27.7 (13/47)	26.3 (10/38)	1.4 (−18 to 20)	.89
HbA_1c goal achieved^c	22.2 (2/9)	14.3 (1/7)	7.9 (−31 to 46)	.69
Cholesterol goal achieved^d	47.1 (8/17)	66.7 (8/12)	−19.6 (−18 to 56)	.29
SBP goal achieved^e	21.2 (7/33)	24.1 (7/29)	−2.9 (−18 to 24)	.78

HbA_1c = hemoglobin A_1c; HDL = high-density lipoprotein; LDL = low-density lipoprotein; SBP = systolic blood pressure.
Note: Analyses used missing-at-random (nonimputed) data. See statistical analysis in Methods section for details.
↵a Achieved target for control of 1 or more of specific conditions for which patient was uncontrolled at baseline.
↵b Achieved target for control of all conditions for which patient was uncontrolled at baseline.
↵c Goal was <8.0%.
↵d Goal for diabetic patients was LDL cholesterol <100 mg/dL or non–HDL cholesterol <130 mg/dL if triglycerides >400 mg/dL, and for nondiabetic patients was LDL cholesterol l<130 mg/dL or non–HDL cholesterol <160 mg/dL if triglycerides >400 mg/dL.
↵e Goal was <130 mm Hg for diabetic patients and <40 mm Hg for nondiabetic patients.

View popup

Table 5

Secondary Condition-Specific Outcomes by Study Arm, for Total Sample and by Clinic Site

Outcome	Health Coaching				Usual Care				Difference Between Arms (95% CI)	P Value
Outcome	n	Baseline, Mean (SD)	12 Months, Mean (SD)	DIM	n	Baseline, Mean (SD)	12 Months, Mean (SD)	DIM	Difference Between Arms (95% CI)	P Value
Total sample
HbA_1c, %	74	9.8 (1.5)	8.6 (2.0)	−1.2	58	9.9 (1.4)	9.4 (2.0)	−0.5	−0.7 (−1.4 to 0.0)	.06
LDL, mg/dL	95	147.2 (36.3)	119.3 (52.8)	−27.9	73	143.4 (33.2)	125.4 (39.1)	−18.1	−9.8 (−21.6 to 2.0)	.10
SBP, mm Hg	80	157.3 (13.8)	148.7 (16.5)	−8.6	76	160.5 (16.4)	150.3 (18.2)	−10.2	1.6 (−7.6 to 4.4)	.59
Clinic site A
HbA_1c, %	65	9.8 (1.5)	8.5 (2.0)	−1.3	51	9.8 (1.4)	9.3 (2.0)	−0.6	−0.7 (−1.4 to −0.0)	.04
LDL, mg/dL	79	148.2 (38.3)	122.6 (55.1)	−25.6	62	145.0 (35.0)	130.5 (37.6)	−14.5	−11.1 (−23.9 to 1.7)	.09
SBP, mm Hg	47	154.9 (10.5)	144.8 (12.5)	−10.0	47	160.5 (15.1)	148.5 (18.9)	−12.0	2.0 (−5.1 to 9.0)	.59
Clinic site B
HbA_1c, %	9	10.2 (1.4)	9.5 (2.1)	−0.7	7	10.5 (1.5)	10.1 (1.7)	−0.5	−0.2 (−2.9 to 2.5)	.89
LDL, mg/dL	16	142.8 (29.4)	103.6 (37.8)	−39.2	11	136.9 (29.2)	98.6 (36.9)	−38.4	−0.8 (−31.3 to 29.6)	.96
SBP, mm Hg	33	160.7 (15.8)	154.1 (19.9)	−6.6	29	160.6 (19.2)	153.2 (17.0)	−7.4	0.8 (−10.0 to 11.6)	.88

DIM = difference in means; HbA_1c = hemoglobin A_1c; LDL = low-density lipoprotein; SBP = systolic blood pressure.
Note: Analyses used missing-at-random (nonimputed) data. See statistical analysis in Methods section for details.

Additional Files

Figures
Tables

The Article in Brief

Health Coaching by Medical Assistants to Improve Control of Diabetes, Hypertension, and Hyperlipidemia in Low-Income Patients: A Randomized Controlled Trial

Rachel Willard-Grace , and colleagues

Background Health coaching can give people with chronic illnesses the knowledge, skills, and confidence to manage their conditions. This study tests the effectiveness of health coaching by medical assistants to improve indicators of health among low-income patients with uncontrolled diabetes, high blood pressure, and high cholesterol levels.

What This Study Found Health coaching by medical assistants is a promising way to improve indicators of health for people with common chronic conditions. In this study, almost twice as many patients who received health coaching achieved their diabetes goals. At the larger study site, health coached patients were more likely to achieve their cholesterol goals. There was no significant difference in the proportion of patients meeting their blood pressure goals.

Implications

Medical assistants can successfully serve as health coaches to improve health indicators for some common chronic conditions.
The medical assistant health coaching model may help solve barriers of time, resources and cultural concordance faced by many primary care practices seeking to implement self-management support.
Annals Journal Club

Mar/Apr: Impact of Medical Assistants as Health Coaches

The Annals of Family Medicine encourages readers to develop a learning community of those seeking to improve health care and health through enhanced primary care. You can participate by conducting a RADICAL journal club and sharing the results of your discussions in the Annals online discussion for the featured articles. RADICAL is an acronym for Read, Ask, Discuss, Inquire, Collaborate, Act, and Learn. The word radical also indicates the need to engage diverse participants in thinking critically about important issues affecting primary care and then acting on those discussions.¹

HOW IT WORKS

In each issue, the Annals selects an article or articles and provides discussion tips and questions. We encourage you to take a RADICAL approach to these materials and to post a summary of your conversation in our online discussion. (Open the article online and click on "TRACK Discussion: Submit a comment.") You can find discussion questions and more information online at: http://www.AnnFamMed.org/site/AJC/.

CURRENT SELECTION

Article for Discussion

Willard-Grace R, Chen EH, Hessler D, et al. Health coaching by medical assistants to improve control of diabetes, hypertension, and hyperlipidemia in low-income patients: a randomized controlled trial. Ann Fam Med. 2015;13(2):130-138.

Discussion Tips

This article investigates the impact of in-clinic health coaching by medical assistants using data from a randomized controlled trial of low-income patients in safety net primary care clinics in San Francisco. The study considers the effectiveness of this model on control of hemoglobin A1c, systolic blood pressure, and LDL cholesterol and explores potential factors in implementation.

Discussion Questions

What question is asked by this study and why does it matter?
How does this study advance beyond previous research and clinical practice on this topic?
How strong is the study design for answering the question?
How well does the study follow the CONSORT guidelines? (www.consort-statement.org)
To what degree can the findings be accounted for by:
How patients were selected, excluded, or lost to follow-up?
How the main variables were measured?
Confounding (false attribution of causality because 2 variables discovered to be associated actually are associated with a 3rd factor)?
Chance?
How the findings were interpreted?
"Contamination" between intervention and control groups? How might this possibility be avoided in the study design?

What are the main study findings?
How comparable is the study sample to similar patients in your practice? What is your judgment about the transportability of the findings?
What contextual factors are important for interpreting the findings? How strong are the methods for assessing contextual factors affecting implementation?
How might this study change your practice? Policy? Education? Research?
What might be the financial impact of implementing a similar model? At what level do the costs accrue? At what level are the benefits likely to appear over time? (eg patient, practice, health care system, community?)
What other in-clinic interventions may lead to similar results?
Who the constituencies are for the findings, and how they might be engaged in interpreting or using the findings?
What are the next steps in interpreting or applying the findings?
What researchable questions remain?

References

Stange KC, Miller WL, McLellan LA, et al. Annals Journal Club: It's time to get RADICAL. Ann Fam Med. 2006;4(3):196-197 http://annfammed.org/content/4/3/196.full.

Main menu

User menu

Search

Health Coaching by Medical Assistants to Improve Control of Diabetes, Hypertension, and Hyperlipidemia in Low-Income Patients: A Randomized Controlled Trial

Article Figures & Data

Figures

Tables

Additional Files

The Article in Brief

Annals Journal Club

Mar/Apr: Impact of Medical Assistants as Health Coaches

HOW IT WORKS

CURRENT SELECTION

Article for Discussion

Discussion Tips

Discussion Questions

References

In this issue

Citation Manager Formats

Related Articles

Cited By...

More in this TOC Section

Similar Articles

Subjects

Keywords

Content

Info for

Engage

About

Main menu

User menu

Search

Health Coaching by Medical Assistants to Improve Control of Diabetes, Hypertension, and Hyperlipidemia in Low-Income Patients: A Randomized Controlled Trial

Article Figures & Data

Figures

Tables

Additional Files

The Article in Brief

Annals Journal Club

Mar/Apr: Impact of Medical Assistants as Health Coaches

HOW IT WORKS

CURRENT SELECTION

Article for Discussion

Discussion Tips

Discussion Questions

References

In this issue

Citation Manager Formats

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles

Subjects

Keywords

Content

Info for

Engage

About