Effectiveness of motivational interviewing for improving physical activity self-management for adults with type 2 diabetes: A review

Abstract

Objectives

This review examines the effectiveness of motivational interviewing for physical activity self-management for adults diagnosed with diabetes mellitus type 2. Motivational interviewing is a patient centered individually tailored counseling intervention that aims to elicit a patient’s own motivation for health behavior change. Review questions include (a) How have motivational interviewing methods been applied to physical activity interventions for adults with diabetes mellitus type 2? (b) What motivational interviewing approaches are associated with successful physical activity outcomes with diabetes mellitus 2?

Methods

Database searches used PubMed, CINAHL, and PsycINFO for the years 2000 to 2016. Criteria for inclusion was motivational interviewing used as the principal intervention in the tradition of Miller and Rollnick, measurement of physical activity, statistical significance reported for physical activity outcomes, quantitative research, and articles written in English.

Results

A total of nine studies met review criteria and four included motivational interviewing interventions associated with significant physical activity outcomes.

Discussion

Findings suggest motivational interviewing sessions should target a minimal number of self-management behaviors, be delivered by counselors proficient in motivational interviewing, and use motivational interviewing protocols with an emphasis placed either on duration or frequency of sessions.

Keywords

Diabetes type 2 diabetes physical activity exercise motivational interviewing

Diabetes mellitus is the seventh leading cause of death in the United States.¹ Around 29 million American adults have diabetes, and more than 90% suffer from diabetes mellitus type 2 (T2DM).² The clinical marker for T2DM is hyperglycemia. Hyperglycemia can lead to serious complications including heart disease, stroke, retinopathy, kidney failure, and lower-limb amputation.² According to the 2014 Disability and Health Data System, around 20% of Americans with T2DM suffer from a mobility-related disability, 19% from a vision disability, and 16.5% from a cognitive disability.³ Mortality risk is 50% higher for adults with diabetes than for those without the disease.⁴

T2DM and related complications are also associated with increased body mass index.^5–7 However, regular physical activity (PA) can help maintain healthy blood glucose levels and prevent or delay complications related to the disease.² PA can lead to weight loss, help maintain a healthy weight, improve the body’s ability to use insulin, and reduce elevated cholesterol.^8,9 For optimal PA health benefits, the Centers for Disease Control and Prevention¹⁰ recommends 150 min of moderate-intensity aerobic activity, and 2 days of muscle strengthening activities per week. Moderate-intensity aerobic activity increases the heart rate for sustained periods of time, and includes activities like brisk walking, water aerobics, running, and bicycling. Common muscle-strengthening exercises include push-ups, sit-ups, yoga, or lifting weights.¹⁰ Although adults with T2DM may have activity restrictions, only 60 min of moderate-intensity aerobics per week can have health benefits.¹¹

Despite PA benefits, less than 40% of US adults with diabetes engage in regular PA.^12,13 Thus, there is a need for evidence-based self-management interventions, as outlined in the National Standards for Diabetes Self-Management Education and Support.¹⁴ Patient-centered, individually tailored interventions are associated with improved lifestyle change. As a result, T2DM self-management interventions should facilitate action-oriented goal setting and should adapt to meet needs based on an individual’s age, comorbidities, cultural factors, health literacy, and psychosocial circumstances.¹⁴ One example of a patient-centered intervention is motivational interviewing (MI).

MI is an individually tailored counseling intervention that aims to elicit a patient’s own motivation for health behavior change. Reviews of the literature provide evidence for the efficacy of MI in the healthcare setting. For instance, MI sessions have led to increased PA for those with chronic health conditions,¹⁵ beneficial lifestyle behavior change for different types of cancer,¹⁶ reduced sexual risk behavior for those with HIV,¹⁷ decreased substance abuse,¹⁸ and smoking cessation.¹⁹ However, few studies on MI have specifically addressed T2DM and PA self-management. Therefore, the purpose of this review was to examine the effectiveness of MI as an intervention to help improve PA self-management for adults diagnosed with T2DM.

MI

MI counselors use certain methods to guide conversations toward behavior change. In general, counselors facilitate movement by helping clients resolve ambivalent feelings about behavior change.²⁰ In MI, there are four processes that help facilitate behavior change: (a) engagement, both the counselor and client establish a helpful connection and working relationship; (b) focusing, the counselor maintains the conversation in a specific direction; (c) evoking, helps to elicit the client’s own motivation for change; and (d) planning, this requires a commitment for change along with a specific plan of action.²¹ The underlying “spirit” of MI requires providers to engage in a collaborative partnership, to connect behavior change with each patient’s own values and concerns, and ultimately to respect autonomy by accepting the choices patients make about their lives. MI was initially developed for substance abuse, but now is widely used and examined in the health care setting.²⁰ The purpose of this review was to examine the effectiveness of MI as an intervention to help improve PA self-management for adults diagnosed with T2DM.

Review Question 1: How have MI counseling methods been applied to PA interventions for adults with T2DM?

Review Question 2: What MI approaches are associated with successful PA outcomes for adults with DM2?

Method

Search strategy

This review identified primary research studies that used MI as the principal intervention to improve levels of PA for individuals diagnosed with T2DM. Healthcare databases used for the search included CINAHL, PubMed, and PsycINFO. Reference lists of retrieved articles were reviewed to complete the search. The following key search terms for published articles between 2000 and 2016 were included: MI, health coaching, wellness coaching, diabetes type 2, diabetes, obesity, PA, and exercise. Articles in this review met the following inclusion criteria: (a) MI was used in the tradition of Miller and Rollnick as the principal intervention, (b) measurements of PA were included, (c) statistical significance was reported for PA outcomes, (d) quantitative experimental research was used, and (e) articles were written in English. Studies either reported using MI in the tradition of Miller and Rollnick, or the type of MI used was verified by the author. Use of MI in the tradition of Miller and Rollnick was verified through written correspondence with R. Whittemore, PhD (May 2016) and R.J. Sigal, MD (August 2016). Studies were excluded if sample populations were 10 or less, and if MI was used in conjunction with another principal intervention strategy or counseling method. However, MI used in conjunction with face-to-face or phone T2DM education met criteria for inclusion. This search retrieved a total of 826 articles, 114 abstracts for review, 32 articles for eligibility, and concluded with 9 studies that met criteria for this review. For 1 of the 9 studies, only the abstract was available.²² Based on written correspondence with M.J. Armstrong, PhD (July 2016), the full article is pending publication. Figure 1 illustrates the article selection process. Content reported in this review was guided by the Preferred Reporting items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.²³

Figure 1.

PRISMA flow diagram.

Study characteristics

Studies used in this review were conducted in the United States, Canada, Europe, Asia, and South Africa. Research designs included seven randomized controlled trials,^22,24–29 one non-randomized controlled trial,³⁰ and one quasi-experimental design.³¹ Among these results, two were pilot studies.^29,31 Most studies examined middle-aged adults with a mean age between 50 and 60 years old.^{22,24,26,27,29–31} One study examined female participants only.²⁹ The majority of studies reporting the average length of T2DM diagnosis recruited participants who had T2DM for 5 years or less.^24,29,31 Clark²⁶ examined participants who had a diagnosis of T2DM for an average of 8 years, and Rubak et al.²⁸ recruited participants with screen detected T2DM. All studies examined MI to help improve two or more T2DM self-care behaviors. This review provides study characteristics for studies with significant PA outcomes in Table 1, and characteristics of studies lacking significant PA outcomes in Table 2.

Table 1.

Characteristics of studies with significant PA findings.

Author and study design	Location and setting	Sample	MI group, control group, and attrition		Length of T2DM diagnosis	Self-care measures	Self-care results	Review findings: for MI approaches
Armstrong et al.²² M. J. Armstrong, PhD (written communication, September, 2016) RCT	Canada Exercise site	Inclusion: Adults with T2DM, completed exercise program Sample (n = 55), female (55%), Mean age (60.3) Mean baseline A1C (6.9%), SES (N/S)	MI sessions: Face to face delivered by researcher over 6 months MI dose (2), length (30 to 45 min) Study arms: MI plus standard care group; standard care without MI group Attrition: 44/55 completed study		N/S	Self-report: PA	Baseline and 6-month assessment times PA results: At 6 months MI mean MET-min/week group significantly more than mean UC at 6 months (p = 0.0361).	1. Number of SM behaviors: 1 2. Emphasis on MI frequency or Duration: Duration 3. MI proficiency: Satisfactory
Chlebowy et al.³⁰ Non-randomized controlled trial	South Eastern USA Outpatient clinics	Inclusion: age ≥ 18, African American ethnicity, T2DM taking oral hypoglycemic or insulin, English speaking Sample (n = 62), female (64.5%), mean age MI (55.8), UC (53), BMI MI (36.3), UC (33), mean baseline A1C MI (7.8), UC (8.1), low-income	MI sessions: Face to face delivered by Baccalaureate nurses (not employees of clinic intervention site) delivered sessions over 3 months MI dose (4 to 6), intervals (every 2 weeks), length (45 to 60 min) Study arms: MI intervention group, and T2DM clinic care group Attrition: MI (53.8%), UC (50%)		N/S	Objective: PA accelerometer, blood glucose monitor Self-report: medication usage	Baseline and 3-month assessment times PA results: Increased odds of adhering to PA recommendations (p = 0.018) Self-care Results: Medication usage or glucose monitoring (p > 0.05)	1. Number of SM behaviors: 3 2. Emphasis on MI frequency or Duration: Frequency and duration 3. MI proficiency: Satisfactory
Cinar and Schou²⁵ RCT	Turkey, Istanbul Outpatient clinics	Inclusion: T2DM, age 30 to 65, minimum of four functional teeth, no psychological treatment or hospitalization Sample (n = 186), Mean age (N/S), age 30 to 65, women (N/S) BMI (MI 14% < 30) HE (17% < 30) Mean baseline A1C (7.5%) SES (N/S)	MI sessions: Face to face delivered by dental professional over 6 months MI dose face to face (5 to 6), phone (3 to 4), length (N/S) Study arms: MI group and HE group (standard oral healthcare advice, diet, PA) Attrition: 7% drop out; 13% did not attend all sessions.	N/S	T2DM diagnosis: Mean MI (13.1) HE (10.8)	Self-report: PA and tooth brushing frequency	Baseline and 16-month follow-up. PA results: MI group who brushed teeth at least once per day were more likely to be physically active, 95% CI, odds ratio 2.1 (1.5–3.2), p = 0.001	1. Number of SM behaviors: 2 2. Emphasis on MI frequency or duration: Frequency 3. MI proficiency: N/S
Clark et al.,²⁶ Clark and Hampson³² RCT	UK, National Health Service Diabetes center	Inclusion: Age 40 to 70, T2DM suboptimal condition, BMI > 25, able to engage in walking program Sample (n = 100), mean age (59.5), female (42%) Mean BMI (31) Mean baseline A1C (8.42), SES (N/S)	MI sessions: Face to face and phone delivered by researcher over 3 months MI dose face to face (1), phone (3), length 30 min (face to face), 10 min (phone) Study arms: MI and UC group Attrition: 6% at 12-month follow-up	N/S	Approximately 8 years	Self-report: PA and diet	Baseline, 3-, and 12-month long-term follow-up PA results: MI group (SDSCA questionnaire) significantly more PA than control group at 3 months, t (98) = 2.118, p = 0.037, and 12 months t (98) = 2.348 p = 0.021; PA not significant (PASE questionnaire) Diet results: MI group significantly improved over UC for	1. Number of SM behaviors: 2 2. MI frequency or duration: Frequency 3. MI proficiency: N/S
							substituting high-fat foods for low-fat at 3 months, but not 12; and decreasing daily fat intake at 12 months, but not 3 months. Significant findings between groups were not found on other dietary measures

N/S: not specified; UC: usual care; HE: health education; RCT: randomized controlled trial; CI: confidence interval; SM: self-management.

Definitions: Frequency equals > two sessions that average > 1 per month; Duration equals 30 to 45 min sessions.

Table 2.

Characteristics of studies without significant PA findings.

Author and study design	Location and setting	Sample	MI group, control group, and attrition	Length of DM2 diagnosis	Self-care measures	Self-care results	Review findings: for MI approaches
Heinrich et al.²⁴ RCT	Europe, Southern Netherlands Outpatient clinics	Inclusion: T2DM ≤ 5 years, between 40 and 70 years old. Sample (n = 584), mean age (59), female (44.9%), BMI (41.3% below 28; 42.2% above 30), mean baseline A1C (6.49) Lower education (61.7%), medium education (23.6%), higher education (14.7%)	MI sessions: Face to face delivered by nurses recruited from general practices over 24 months during regular T2DM visits MI dose (3 or 4), length (20 min) Study arms: MI plus UC and UC alone Attrition 77% completed 12-month follow-up, 72% completed 24-month follow-up, 67% completed both follow-up times, 5% filled out none.	T2DM diagnosis: ≤1 year (26.4%), 2 to 3 years (47%), 4 to 5 years (26.6%)	Self-report: PA, diet, smoking	Baseline, 12 month, and 24-month follow-up No significant differences between MI and UC for PA or smoking. MI group significantly higher fat intake at 12- and 24-month follow-up times (p < 0.05).	1. Number of SM behaviors: 3 2. Emphasis on MI frequency or duration: No 3. MI proficiency: N/S
Rubak et al.²⁸ RCT	Europe, Denmark Outpatient clinics	Inclusion: Newly diagnosed T2DM, age 40 to 69 Sample (n = 628), mean age (61), female (42%), mean BMI MI (30.5), CG (30.8), mean baseline A1C at screening (6.9%), SES (N/S)	MI sessions: Face to face delivered by General Practitioners (GP) over 1 year MI dose (3), length (45 min) Study arms: MI group GP’s used MI during regular T2DM consultations, UG group GP’s delivered consultations without MI Attrition: On average, MI group received < 2 of 3 consultations	T2DM diagnosis: Detected during screening for research project	Self-report: PA, smoking Objective: Medication usage determined by pharmacy purchases	Baseline and 1-year follow-up No significant differences between MI and UC found	1. Number of SM behaviors: 3 2. Emphasis on MI frequency or duration: High attrition rate MI proficiency: N/S
Mash et al.²⁷ RCT	South Africa, Cape Town Community health centers	Inclusion criteria: T2DM patient in community health center; exclusion: Type 1 diabetes, acute mental or physical illness Sample: n = 1570; mean age MI = 55.8 (11.5); UC = 56.4 (11.6), Women MI = 71.5%, UC = 75.7%. Mean weight kg MI = 84.4 (18.7), UC = 83.9 (18.5) Mean baseline A1C MI = 8.9 (2.3) UC = 9.3 (2.3). SES: N/S	MI sessions: Group delivered by community health promoters trained in MI over 6 months MI dose (4), length (N/S) Study arms: MI group received MI by health promoters, UC group included education, and individual counseling by providers Attrition: MI group 59.4% no sessions; 28% ≤ 2 sessions	N/S	Self-report: PA, diet, foot care, smoking, medication use	Baseline and 12-month follow-up No significant differences between MI and UC for PA, diet, foot care, smoking, or medication use	1. Number of SM behaviors 2. Emphasis placed on MI frequency or duration: Duration not specified and high attrition rate 3. MI proficiency: N/S
Whittemore et al.²⁹ RCT (pilot)	USA, Northeastern Connecticut Diabetes education center	Inclusion: T2DM, female, age 30 to 70, cleared for exercise by doctor, A1C > 7% Sample (n = 53), mean age (57.6), overweight or obese (96%) Mean baseline A1C (7.7%), low to low-middle income	MI Sessions: Face to face and phone delivered by nurses over 6 months MI dose face to face (6), phone (2), length (N/S) Study Arms: MI group received MI plus UC, UC group received UC alone Attrition: Low	T2DM diagnosis: Mean 2.7 years	Self-report: PA and diet	Baseline, 3 months, and 12 months No significant differences between MI and UC for PA; however, there was a trend toward increased PA MI significantly better diet than UC (p = 0.02)	1. Number of SM behaviors: 2 Emphasis on MI frequency or duration: Frequency, however trend toward improved PA MI proficiency: N/S
Calhoun et al.³¹ One-group pretest posttest design (pilot)	USA Montana Indian health services clinic	Inclusion: Tribal members on Northern Plains Reservation ≥ 18 years old, T2DM, receiving T2DM treatment at local Indian Health Service clinic Sample (n = 26), mean age (54), female (53.8%), mean BMI (31.8), mean baseline A1C (8.8)	MI sessions: Face to face by researcher over 3 weeks MI dose (2), length (30 min) Attrition: 77% completed both sessions	T2DM diagnosis: Mean 4.4 years	Self-report: PA, and Diet Note: MI focused on any self-management behaviors addressed by participants (smoking, blood glucose monitoring, medication, and mood)	Baseline and 3 months No significant findings for PA or diet	1. Number of SM behaviors: 2, along with additional behaviors the client wanted to focus on 2. Emphasis on MI frequency or duration: No 3. MI proficiency: Not measured

N/S: not specified; UC: usual care; RCT: randomized controlled trial; SM: self-management.

Definitions: Frequency equals > two sessions that average > 1 per month; duration equals 30 to 45 min sessions.

MI outcomes

Three MI approaches related to significant PA outcomes for adults with T2DM were found in this review: (a) target a minimal number of self-management behaviors; (b) MI protocols with an emphasis placed on either duration or frequency of sessions; and (c) delivery of MI by health professionals likely proficient in MI.

T2DM self-management measures and results

Four of the nine studies included MI interventions that were associated with significant PA outcomes. Chlebowy et al.³⁰ objectively measured PA using an accelerometer at two time points. Participants assigned to MI or usual care (UC) wore the accelerometer for 1 week at baseline and at the 3-month follow-up time. A logistical regression model found that MI significantly increased the odds that a participant followed recommended levels of PA. Regression models also were used to evaluate medication usage and glucose monitoring; however, MI did not have a significant effect on other T2DM self-management behaviors.³⁰ Cinar and Shou²⁵ used self-report questionnaires to measure PA and tooth brushing frequency at baseline and at the 16-month follow-up time. Data from PA multiple-choice questionnaires were reclassified into dichotomous answers, “Favorable PA” and “Unfavorable PA.” Patients who brushed their teeth at least once per day were significantly more likely to be physically active in the MI group verses those in the health education group.²⁵ Armstrong et al.²² measured PA using a self-report questionnaire to determine PA maintenance results after an 8-week exercise program. Differences in PA maintenance between the MI and control group were examined using linear mixed modeling. The MI group was significantly more active at 6 months than the control group.²²

Clark et al.²⁶ used self-report questionnaires to measure PA and dietary habits. PA was measured using two different self-report questionnaires and results were mixed. The researchers used a PA Scale for the Elderly (PASE) to measure leisure time, household, and work-related PA over 7 days, and no significant differences occurred between the MI and UC group. However, the Summary of Diabetes Self-Care Activities (SDSCA) questionnaire was used to measure PA frequency over 7 days. The researchers found that the MI group, when compared to UC, had significantly increased PA at 3 and 12 months after baseline. The intervention group improved significantly more than the UC group for two items on the dietary scales.²⁶ Overall, studies with significant PA outcomes targeted a limited number of T2DM self-management behaviors; studies targeted either 1,²² 2,^25,26 and 3³⁰ self-management behaviors.

Five MI studies were not associated with significant PA outcomes, and most targeted between 3 and 6 self-management behaviors. Heinrich et al.²⁴ used self-report questionnaires to measure PA, dietary intake, sedentary behavior, and smoking at baseline, 12 months, and 24 months. The SDSCA measured participant engagement in more than 30 min of PA 5 days per week, and the International Physical Activity Questionnaire (IPAQ) was used to recall the amount and intensity of PA in the previous 7 days. No significant outcomes for self-management behaviors were found in the MI group. However, the UC group had significantly lower fat intake at 12 and 24 months.²⁴ Rubak et al.²⁸ used the self-report IPAQ to measure PA, and the SDSCA to report smoking habits at baseline and 12 months. Medication use was determined by pharmacy purchases. No significant differences were found between the MI and control groups for PA or smoking; however, both groups purchased almost 100% of recommended T2DM medications.²⁸ Mash et al.²⁷ used the SDSCA self-report questionnaire to measure PA, diet, exercise, foot care, medication use, and smoking at baseline and 12 months. No significant differences were found between the MI and the UC group for T2DM self-care activities.²⁷ Whittemore et al.²⁹ used a self-report PA questionnaire that differentiated between structured exercise and PAs of daily living in the previous month. Dietary behavior was measured with the self-report SDSCA questionnaire. There was a trend toward increased PA for women in the MI group, but results were not significant. Women in the MI group reported significantly better dietary behaviors.²⁹ Calhoun et al.³¹ used self-report questionnaires at baseline and 3 months to estimate the amount of PA for the last 7 days and to estimate the consumption of healthy and unhealthy foods. PA and diet were the two self-management behaviors measured. However, the MI counselor targeted any self-management behavior the participant wanted to discuss, such as smoking, blood glucose monitoring, medication management, and mood. Measures of health-related behaviors were not significant.³¹

MI counseling methods

MI approaches associated with significant PA outcomes either implemented frequent MI sessions (greater than two sessions that average more than once per month), placed an emphasis on session duration (last a minimum of 30 to 45 min each), or emphasized both frequency and duration. Chlebowy et al.³⁰ used baccalaureate prepared nurses to deliver six MI sessions over 3 months in an outpatient medical clinic. Individual sessions were scheduled every 2 weeks and lasted around 45 min to an hour. Around 50% were lost to attrition in both groups, and participants actually received four to six sessions.³⁰ Cinar and Schou^25,33 used a dental professional to deliver five to six individual MI sessions in the outpatient clinical setting, as well as three to four phone sessions. Questionnaires were filled out at baseline and at 16 months post-intervention. Details pertaining to length of MI sessions were not specified. Participant dropout rate was 7%, and 13% did not take part in all the sessions.²⁵ Armstrong et al.²² evaluated the impact of MI for PA maintenance following an 8-week exercise program. The principal investigator conducted two individual 30- to 45-min MI sessions. The first session was conducted at the end of the 8-week exercise program, and the last session was 3 months later (M.J. Armstrong, PhD., written communication, August, 2016).

Clark et al.²⁶ measured PA outcomes using two self-report questionnaires and found mixed results. Research interventionists delivered an individual 30-min MI session at a diabetes center along with three subsequent 10-min MI phone sessions over 3 months.³² Attrition was low, 6% of participants were lost at the final 12-month follow-up assessment time.²⁶

Four out of five studies lacking significant PA outcomes implemented infrequent MI sessions of short duration, or had a high attrition rate. Heinrich et al.²⁴ planned to implement quarterly MI sessions during routine T2DM visits each year for 24 months. Nurses recruited from primary care clinics delivered individual 20-min MI sessions during regular T2DM consultations. Participants attended three or four sessions each year.²⁴ Rubak et al.²⁸ used general practitioners (GPs) to deliver three 45-min MI sessions over 1 year; however, participants attended less than two of the three consultations.²⁸ Mash et al.²⁷ recruited health promoters from district health services and planned to deliver four 20- to 60-min group sessions with 10 to 15 participants. The intervention took place over 6 months, and post-intervention follow-up data were collected at 12 months. Close to 60% of participants did not attend any MI educational sessions. The remaining participants attended two sessions or less (28%).²⁷ A researcher in the Calhoun et al.³¹ study delivered two 30-min MI sessions for each participant over 3 weeks. Approximately 77% of participants completed the study.³¹ Whittemore et al.²⁹ used nurses to deliver six individualized MI sessions and two brief follow-up phone calls over 6 months. Five MI sessions were delivered within the first 3 months. Subsequent phone calls and the last MI session were completed over the last 3 months; attrition rate was low.²⁹

MI training and fidelity

Two of the four studies with significant PA outcomes measured MI fidelity, and rated counselors as MI proficient. In the study by Chlebowy et al.,³⁰ bachelor degree prepared registered nurses (RNs) with T2DM care experience received extensive MI training by a motivational interviewer network trainer. After training, fidelity assessments for adherence to MI methods were conducted throughout intervention delivery. All MI sessions were audio recorded and assessed using the MI Treatment Integrity (MITI) scale. Assessment scores indicated satisfactory MI proficiency.³⁰ Armstrong, PhD (written correspondence, September 2016) recorded all MI sessions, and MI fidelity was determined by randomly sampling 25% of the recordings. A psychologist used the MITI scale and rated the MI counselor as MI proficient (M.J. Armstrong, written correspondence, September 2016). Cinar²⁵ and Clark²⁶ did not report information pertaining to MI training or fidelity assessment.

One of the five studies that lacked significant PA outcomes measured MI fidelity; however, the counselors were not rated as MI proficient. In the study by Heinrich et al.,²⁴ nursing staff were taught MI methods by a certified MI trainer during two 5-h sessions, and received MI counseling charts to guide them through T2DM consultations. Nurses received written feedback at 3 months from two audiotaped consultations by a MI-trained researcher, and direct feedback during consultations at 9, 11, and 18 months after training by a MI-trained teaching nurse. However, nurses proficiency using MI counseling methods was not reported.²⁴ In the study by Rubak et al.,²⁸ a trained teacher provided 1½-day MI training sessions, and two half-day follow-up sessions for GPs over the first year. Fidelity measures were not reported.²⁸ Mash et al.³⁴ initially trained health promoters during a 4-day workshop. The workshop focused on the first two sessions and included communication style, structure of the sessions, skills, and diabetes knowledge. Two months later, a 2-day workshop was delivered to reinforce previous training and to focus on the previous two sessions. MI fidelity was evaluated by a researcher at the health center at least twice, and 36 randomly selected audio-recorded sessions were assessed for MI proficiency using the MITI scale. Findings indicated achievement of an overall collaborative style of communication, but health promoters lacked sufficient listening skills and the guiding communication style.³⁴ In the study by Calhoun et al.,³¹ the research interventionist was trained and supervised by a MI Network Trainer. MI sessions were not coded to ensure fidelity.³¹ In the study conducted by Whittemore et al.,²⁹ MI training and fidelity were not discussed.

Discussion

Close to half the studies in this review reported significant improvements in PA outcomes for adults diagnosed with T2DM. All studies with significant PA findings either conducted face-to-face sessions or used a combination of face-to-face and phone sessions. Despite mixed results, insights gained about effective counseling approaches show promise for MI and T2DM PA self-management. This review identified three MI approaches associated with significant PA outcomes for adults diagnosed with T2DM.

The first MI approach associated with significant PA outcomes found that sessions were more successful when counselors focused on a minimal number of T2DM self-management behaviors. For example, most studies with significant PA outcomes either targeted PA alone, or with one other self-management behavior. One study was an exception to this finding, Chlebowy et al.³⁰ targeted three self-management behaviors. However, the 3-month intervention placed an emphasis on both frequency (4 to 6) and duration (30 to 45 min) of MI sessions. In contrast, most studies lacking PA significance targeted three or more T2DM self-management behaviors over the course of the intervention. Although Whittemore et al.,²⁹ only targeted two self-management behaviors and still lacked significant PA outcomes, a trend toward improved PA was found and behavior change for dietary self-management was significant. Even though a large conceptual category like T2DM self-management does qualify as a target behavior for a MI session (T.B. Moyers, PhD, written correspondence, September 2016),³⁵ findings from this review indicate MI may be more effective when counselors prioritize a minimal number of target behaviors over the course of a few sessions.

The second MI approach associated with significant PA outcomes indicates MI counselors should emphasize either the frequency or duration of MI sessions. When frequency is emphasized, counselors may improve PA outcomes by implementing more than two sessions that average more than once per month. When duration is emphasized, improved PA outcomes may be associated with infrequent sessions if they last more than 30 min. Most studies lacking PA significance either implemented short and infrequent MI sessions, or reported high attrition rates. Whittemore et al.²⁹ was the only study that lacked significant PA outcomes despite the emphasis placed on session frequency. However, a trend toward improved PA was found.²⁹

The third MI approach associated with significant PA outcomes included the use of MI proficient counselors. Three studies measured MI fidelity, and MI proficiency was only reported in two studies^22,30 that reported significant PA outcomes. According to Rollnick et al.²⁰ lectures and workshops help introduce counselors to MI methods, but MI proficiency can only be attained during sessions that include feedback and coaching from expert trainers. Ultimately, significant PA outcomes may be contingent upon MI counselor proficiency. However, the degree of training and counselor proficiency was unclear in most studies.

Limitations and future research

Limitations of this review must be taken into consideration. Studies included in this review were all written in English. Although the majority of studies were randomized controlled trial designs, one controlled trial did not include randomization, two were pilot studies, and one of the pilot studies was a pretest posttest design. Only one study measured PA objectively, making data related to PA measurement subject to social desirability and recall bias.

Information pertaining to MI approaches was missing. Some studies did not include a description of UC, length of MI sessions, and most did not address MI intervention fidelity. Most interventions targeted two or more T2DM self-management behaviors, however, the degree MI sessions focused on PA self-management is unknown. In addition, since a limited number of studies have examined the effectiveness of MI for T2DM PA self-management, one abstract was included in this review. Although information and data reported in the abstract through written correspondence with the author helped strengthen the findings in this review, the full article is pending publication.

Recommendations for future research include more thorough reports pertaining to MI approaches used, and stronger intervention designs that include long-term follow-up outcomes. All MI studies should include details pertaining to dose, length, and frequency of sessions. In addition, since MI studies typically target multiple self-management behaviors, investigators should report the amount of focus placed on the different behaviors. To better understand the efficacy of MI for PA self-management, future studies must use objective measures for PA. To help address barriers related to time within the healthcare setting, further investigation into the usefulness of group and phone MI sessions is needed. Most importantly, few studies measured MI fidelity. MI proficiency may be highly correlated with significant PA outcomes.

Conclusion

Although results for the effectiveness of MI were mixed, MI shows promise for T2DM PA self-management. Studies lacking significance either used MI to target multiple self-management behaviors, or counselor MI proficiency was not reported. Findings from this review suggest MI proficient counselors who emphasize PA self-management may help foster PA behavior change.

Footnotes

Acknowledgements

Dr. Gail Stuart, Dr. Martina Mueller, and Dr. Janet York critically reviewed the manuscript, and Dr. John Dinolfo and Michael Madson provided writing assistance.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Centers for Disease Control and Prevention. National vital statistics reports, 2016.

Centers for Disease Control and Prevention. National diabetes statistics report, 2014.

Centers for Disease Control and Prevention. Have diabetes by disability status and types, 2014.

Centers for Disease Control and Prevention. A snapshot: Diabetes in the United States, 2014.

Nordstrom

Pedersen

Gustafson

et al.

Risks of myocardial infarction, death, and diabetes in identical twin pairs with different body mass indexes. JAMA Int Med 2016; 176: 1522–1529.

Ganz

Wintfeld

et al.

The association of body mass index with the risk of type 2 diabetes: a case-control study nested in an electronic health records system in the United States. Diabetol Metab Syndr 2014; 6: 50–50.

Gray

Picone

Sloan

et al.

Relation between BMI and diabetes mellitus and its complications among US older adults. South Med J 2015; 108: 29–36.

National Institute of Diabetes and Digestive and Kidney Diseases. Diabetes diet, Eating, & Physical Activity. NIDDK (National Institute of Diabetes and Digestive and Kidney Diseases), https://www.niddk.nih.gov/health-information/diabetes/overview/diet-eating-physical-activity (2014).

Centers for Disease Control and Prevention. Know the facts about high blood cholesterol, 2015.

10.

Centers for Disease Control and Prevention. How much physical activity do adults need? 2015.

11.

Centers for Disease Control and Prevention. The benefits of physical activity, 2015.

12.

Centers for Disease Control and Prevention. Age-adjusted percentage of adults aged 18 years or older with diagnosed diabetes who were physically inactive, Unites States, 1994–2010, 2012.

13.

Morrato

Hill

Wyatt

et al.

Physical activity in U.S. adults with diabetes and at risk for developing diabetes, 2003. Diabetes Care 2007; 30: 203–209.

14.

Haas

Maryniuk

Beck

et al.

National standards for diabetes self-management education and support. Diabetes Care 2014; 37 Suppl 1: S144–S153.

15.

O’Halloran

Blackstock

Shields

et al.

Motivational interviewing to increase physical activity in people with chronic health conditions: a systematic review and meta-analysis. Clin Rehabil 2014; 28: 1159–1171.

16.

Spencer

Wheeler

. A systematic review of Motivational Interviewing interventions in cancer patients and survivors. Patient Educ Counsel 2016; 99: 1099–1105.

17.

Naar-King

Parsons

Johnson

. Motivational interviewing targeting risk reduction for people with HIV: a systematic review. Curr HIV/AIDS Rep 2012; 9: 335–343.

18.

Smedslund

Berg

Hammerstrom

et al.

Motivational interviewing for substance abuse. Cochr Database Syst Rev 2011. Cd008063. DOI:10.1002/14651858.CD008063.pub2.

19.

Lindson-Hawley

Thompson

Begh

. Motivational interviewing for smoking cessation. Cochrane Database Syst Rev 2015. Cd006936. DOI: 10.1002/14651858.CD006936.pub3.

20.

Rollnick

Miller

Butler

. Motivational interviewing in health care: Helping patients change behavior, New York: Guilford Press, 2008.

21.

Miller

Rollnick

. Motivational interviewing : helping people change, New York, NY: Guilford Press, 2013.

22.

Armstrong

Campbell

Lewin

et al.

Motivational interviewing-based exercise counselling promotes maintenance of physical activity in people with type 2 diabetes. Can J Diabetes 2013; 37: S3–S3.

23.

Moher D, Liberati A, Tetziaff J, et al. PRISMA 2009 checklist. The website is PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), http://prisma-statement.org/documents/PRISMA%202009%20checklist.pdf (2009).

24.

Heinrich

Candel

Schaper

et al.

Effect evaluation of a Motivational Interviewing based counselling strategy in diabetes care. Diabetes Res Clin Pract 2010; 90: 270–278.

25.

Cinar

Schou

. Impact of empowerment on toothbrushing and diabetes management. Oral Health Prevent Dentistry 2014; 12: 337–344.

26.

Clark

Hampson

Avery

et al.

Effects of a tailored lifestyle self-management intervention in patients with type 2 diabetes. Br J Health Psychol 2004; 9: 365–379.

27.

Mash

Rhode

Zwarenstein

et al.

Effectiveness of a group diabetes education programme in under-served communities in South Africa: a pragmatic cluster randomized controlled trial. Diabet Med 2014; 31: 987–993.

28.

Rubak

Sandbaek

Lauritzen

et al.

Effect of “motivational interviewing” on quality of care measures in screen detected type 2 diabetes patients: a one-year follow-up of an RCT, ADDITION Denmark. Scand J Primary Health Care 2011; 29: 92–98.

29.

Whittemore

Melkus

Sullivan

et al.

A nurse-coaching intervention for women with type 2 diabetes. Diabet Educ 2004; 30: 795–804.

30.

Chlebowy

El-Mallakh

Myers

et al.

Motivational interviewing to improve diabetes outcomes in African Americans adults with diabetes. West J Nurs Res 2015; 37: 566–580.

31.

Calhoun

Brod

Kirlin

et al.

Effectiveness of motivational interviewing for improving self-care among Northern Plains Indians with type 2 diabetes. Diabet Spectr 2010; 23: 107–114 8p.

32.

Clark

Hampson

. Implementing a psychological intervention to improve lifestyle self-management in patients with type 2 diabetes. Patient Educ Counsel 2001; 42: 247–256.

33.

Basak Cinar

Schou

. Health promotion for patients with diabetes: health coaching or formal health education? Int Dental J 2014; 64: 20–28.

34.

Mash

Levitt

Steyn

et al.

Effectiveness of a group diabetes education programme in underserved communities in South Africa: pragmatic cluster randomized control trial. BMC Fam Pract 2012; 13: 126. DOI:10.1186/1471-2296-13-126–126. DOI:10.1186/1471-2296-13-126.

35.

Moyers TB, Martin T, Manuel JK, et al. Revised global scales: Motivational interviewing treatment integrity 3.1.1. Moyers, TB, http://casaa.unm.edu/download/MITI3_1.pdf (2010).