Review of Motivational Interviewing in promoting health behaviors,☆☆

https://doi.org/10.1016/j.cpr.2009.02.001Get rights and content

Abstract

There is considerable evidence for the effectiveness of Motivational Interviewing (MI) in the treatment of substance abuse, as well as a number of other health behavior areas. The present paper summarizes and critically reviews the research in three emerging areas in which (MI) is being applied: diet and exercise, diabetes, and oral health. Although 10 prior reviews focused in part on MI studies in the areas of diet, exercise, or diabetes, the present paper provides an up-to-date review, and includes oral health as another emerging area of MI research. Overall, 37 articles were reviewed: 24 in the areas of diet and exercise, 9 in the area of diabetes, and 4 in the oral health area. Research in these areas suggests that (MI) is effective in all these health domains, although additional research is needed, particularly in the oral health arena. Specifically, future research in the areas of diet and exercise should examine the clinical utility of MI by health care professionals (other than dietitians), studies in the area of diabetes should continue to examine long-term effects of MI on glycemic control, and research in the area of oral health should focus on developing additional trials in this field. Further, future studies should demonstrate improved research methodology, and investigate the effects of possible outcome mediators, such as client change talk, on behavior change.

Introduction

Motivational Interviewing (MI) has been applied to a number of areas of behavioral change, and is best known for applications in the realm of substance abuse. MI has been utilized in several health arenas with promising results, for example, among individuals with HIV, to reduce risky behaviors, and to reduce substance abuse and improve HIV medication adherence (e.g., Parsons, Rosof, Punzalan, & Di Maria, 2005). This paper will provide a brief overview of MI and will evaluate empirical studies testing the effectiveness and clinical utility of MI in three health areas: diet and exercise, diabetes, and oral health. Further, suggestions to guide future research are provided.

Miller and Rollnick (2002) defined MI as a client-centered, “directive method for enhancing intrinsic motivation to change by exploring and resolving ambivalence” (p. 25). MI is a directive psychosocial intervention used to identify and resolve discrepancies between desired behaviors and actual behaviors, and to increase motivation to facilitate behavior change (Miller & Rollnick, 2002). Concepts such as reflective listening are balanced with a directive approach. The “spirit” (Miller & Rollnick, 2002; p. 34) of MI includes principles of collaboration, evocation (of motivation from the client herself), and client autonomy.

MI consists of two phases. During phase one, intrinsic motivation for change is enhanced whereas in phase two, commitment to change is strengthened (Miller & Rollnick, 2002). The goal of MI is to strengthen the importance of change from the patient's perspective (Burke, Arkowitz, & Menchola, 2003), using four basic principles to enhance motivation: (a) expression of empathy, (b) development of discrepancy, (c) rolling with resistance, and (d) the support of self-efficacy (Miller & Rollnick, 2002). Discrepancy is developed between desired behaviors and actual behaviors. The patient presents reasons for change while the facilitator provides support. Alternately, if the patient is resistant to change, the facilitator “rolls” with it instead of fighting against it. If and when the patient is ready to initiate a change, the facilitator supports that decision (Miller & Rollnick, 2002).

MI involves a client-centered approach to consultation. MI adheres to the concept that behavior change is not the sole responsibility of the patient, but is a shared endeavor. Practitioners are in a unique position to either enhance the client's motivation to change or to contribute to resistance (Rollnick, Mason, & Butler, 2002). Traditionally, in health care settings, recommendations for behavior change are delivered through brief advice-giving in which overt recommendations are provided (Hertiage and Sefi, 1992, Rollnick et al., 2002). MI challenges traditional intervention delivery methods by suggesting that patients know what is best for themselves, and that professionals should work with them to determine what behavior change strategies will work best, while acknowledging freedom of choice. A menu of choices is one way in which recommendations can be given while maintaining the patient's freedom of choice (Rollnick et al., 2002).

Ten prior reviews, which have focused at least in part on diet, exercise, or diabetes, have been published, and are presented chronologically. A review of studies focused on MI in the oral health arena has not been published. Each presented review includes studies that met certain requirements of methodological rigor, individually administered MI, and adaptations of MI in a variety of health areas.

Brief MI interventions in the areas of substance abuse, smoking, HIV risk behaviors, and diet and exercise were reviewed by Dunn, DeRoo, and Rivara (2001). There were 29 studies reviewed that delivered MI face-to-face, included MI monitoring, random assignment, inclusion of a control group, and measurement of behavioral and/or health outcomes. Dunn and colleagues reported support for MI in the areas of substance abuse, and diet and exercise, smoking cessation and HIV risk reduction, with effect sizes (i.e., Hedge's g) ranging from .23 to 2.17. Among the areas reviewed, diet and exercise demonstrated the most significant effects whereas the most modest effects were found in the area of smoking cessation.

Burke, Arkowitz, and Dunn (2002) examined 26 studies in which interventions based on MI principles were delivered individually and face-to-face. The meta-analysis found small to moderate effect sizes (i.e., d= .25 to .57) when MI was compared to no treatment or treatment as usual for substance abuse, as well as diet and exercise; support for MI for smoking cessation and reduction of HIV-risk behaviors, however, was more modest.

Burke et al. (2003) conducted a meta-analysis of 30 controlled clinical trials to examine the effectiveness of adaptations of MI (AMIs) in the areas of alcohol abuse, smoking cessation, drug addiction, HIV-risk behaviors, treatment adherence, and diet and exercise. Criteria for inclusion in the meta-analysis included using an intervention based on MI principles, and MI being delivered to an individual. AMIs were determined to be as effective as other treatments, and more effective than no-treatment or placebo controls, in the areas of alcohol, drugs, and diet and exercise. As in the review the year before, however, there was contradictory evidence in the areas of smoking cessation and HIV-risk behaviors.

VanWormer and Boucher (2004) conducted a meta-analysis to examine the effectiveness of MI in diet modification. Significant effects were found in three of the four reviewed studies. Importantly, the review suggests that MI can be effective in small doses; however, more frequent contact with patients can increase behavior change.

Burke, Dunn, and Atkins (2004) conducted a meta-analysis of 39 studies examining the effectiveness of AMIs in the areas of alcohol problems, drug addiction, smoking cessation, HIV-risk behaviors, diet and exercise, treatment compliance, eating disorders, asthma management, and injury-risk behaviors. AMIs were found to be as effective as other general interventions and yielded moderate effect sizes (i.e., d= .35 to .56) in areas such as substance abuse, as well as diet and exercise.

Hettema, Steele, and Miller (2005) reviewed studies examining the effectiveness of MI in the areas of alcohol, smoking, HIV/AIDS, drug abuse, treatment compliance, gambling, intimate relationships, water purification/safety, eating disorders, and diet and exercise. There were 72 articles included in which a group or individual intervention with components of MI was delivered, a posttreatment outcome measure was included, a control or comparison group was present, and a procedure to determine the pretreatment equivalence of groups was utilized. Hettema, Steele and Miller (2005) reported a range of effect sizes (i.e., g= .30 to .77) across MI studies, which suggest medium to large effects. Interestingly, effect sizes were found to be higher when MI was used with Native Americans; however, similar findings were not found when MI was used with African Americans or Hispanic Americans. Also, higher effect sizes were noted when sessions were not guided by a manual.

Rubak, Sandbaek, Lauritzen, and Christensen (2005) performed a meta-analysis to examine the effectiveness of MI across 72 randomized, controlled studies in a variety of health areas, including diet and exercise, diabetes, and substance abuse. A significant effect of MI was demonstrated in 74% of the studies reviewed. Among studies with MI sessions that lasted 60 minutes, 81% showed an effect. Rubak, Sandbaek, Lauritzen and Christensen (2005) suggest that the likelihood of an effect was positively correlated with the number of encounters and with a prolonged follow-up period. Overall, MI was found to outperform traditional advice-giving in 75% of studies reviewed. Of the studies that targeted diet and exercise, diabetes, asthma, or smoking, 72% demonstrated an effect. Smoking studies yielded an effect in 67% of those reviewed, whereas diabetes, asthma, and diet and exercise studies reported an effect in 77% of the studies reviewed (Rubak et al., 2005).

Knight, McGowan, Dickens, and Bundy (2006) discussed eight studies in the fields of diabetes, asthma, hypertension, hyperlipidaemia, and heart disease in a systematic review of MI in physical health care settings. Three studies were in the area of diabetes (e.g., Channon, Smith, & Gregory, 2003), one in the area of asthma (e.g., Schmaling, Blume, & Afari, 2001), one in the area of hypertension (e.g., Woolard et al., 1995), one in the area of hyperlipidaemia (e.g., Mhurchú, Margetts, & Speller, 1998), and two in the area of heart disease (e.g., McHugh et al., 2001). Overall, the authors reported positive results for the effects of MI on outcomes in the areas reviewed.

Resnicow, Davis, and Rollnick (2006) focused their review on youth studies that used MI to modify diet or physical activity (e.g., Resnicow, et al., 2005), diabetes (e.g., Channon et al., 2003), and other behaviors, such as smoking (e.g., Colby et al., 1998), and included some adult studies that used MI to modify diet or physical activity (e.g., Smith, Heckemeyer, Kratt, & Mason, 1997). The authors suggest that although MI might be a feasible intervention to use with children and adolescents, additional studies with youth are needed in the areas reviewed to determine the clinical utility of MI in the prevention or treatment of pediatric obesity (Resnicow et al., 2006).

Most recently, Van Dorsten (2007) provided a summary of studies examining the effectiveness of MI for weight loss and exercise. Results of eight studies in the area of weight loss (e.g., Carels et al., 2007) and two studies in the area of exercise (e.g., Harland et al., 1999) were reported. The author concluded that MI was shown to significantly improve diet and exercise behaviors, regimen adherence, and weight loss, based on the studies reviewed.

Across the 10 reviews, the use of MI was, in general, strongly supported in affecting health behavior change and maintenance; MI typically was more effective than no treatment and when compared to traditional advice giving (Rubak et al., 2005). Nevertheless, several limitations were identified through the reviews conducted to date. Most commonly, studies have focused on adaptations of MI and not on “pure” MI as an intervention (Burke et al., 2003). The duration of MI sessions varied greatly. Additionally, high attrition rates have been identified as a common threat to internal validity among studies examining the effectiveness of MI (Burke et al., 2003). Attempts to improve the internal validity of studies investigating MI have been made by many authors, including efforts to ensure treatment fidelity (e.g., Miller et al., 1993, Project MATCH Research Group, 1997, Stephens et al., 2000). Overall, MI has among the highest mean methodological quality in substance abuse treatment outcome research (Miller & Wilbourne, 2002). Although not presented in the current paper, several systematic reviews and meta-analyses examining MI in a variety of other health areas, such as substance abuse, have been published (e.g., Miller and Wilbourne, 2002, Noonan and Moyers, 1997).

MI was first developed as a method to help individuals decrease alcohol and drug abuse and has been empirically supported for the treatment of substance abuse. MI is a promising intervention to encourage health behavior change in general, in a variety of settings, alone or in conjunction with traditional interventions (Rollnick et al., 2002). Existing MI research on promoting health behaviors, however, has methodological limitations that must be addressed with future research. The purpose of the present paper is to critically review literature using MI to enhance the health care behaviors of patients in three areas (i.e., diet and exercise, diabetes, and oral health), since the first published article on MI (Miller, 1983) through 2008. Articles were acquired through a search of the MI website (www.motivationalinterviewing.org), through article and book reference lists, and through database searches (e.g., Psych Info and Medline); search keywords included the use of single terms (e.g., Motivational Interviewing, motivation) and the use of a combination of descriptive labels for each health area (e.g., Motivational Interviewing and diabetes; motivation and diet).

This review is unique in that a comprehensive, critical, and focused review of MI in the areas of diet and exercise, diabetes, and oral health has not been conducted. Although MI studies of diet, exercise, and diabetes have been examined in the literature, new studies have appeared since the prior reviews were published. No review of studies examining the effectiveness of MI in the area of oral health has yet been published. These three health areas were selected because they represent the growing empirical research examining the effectiveness of MI to encourage positive health behaviors. There were 37 empirical studies were reviewed: 24 in the areas of diet and exercise, 9 in the area of diabetes, and 4 in the oral health area. Over time, the numbers of MI studies across these three areas have increased, as evidenced by the number of published empirical studies annually: 1997 — 1 article; 1998 — 1 article; 1999 — 2 articles, 2001 — 2 articles, 2002 — 1 article, 2003 — 4 articles, 2004 — 3 articles, 2005 — 3 articles, 2006 — 4 articles, 2007 — 10 articles, and 2008 — 6 articles.

Health care professionals are in a unique position to promote dietary change and increased exercise in patients facing a variety of health risks and issues through imparting information, skills training, and fostering motivation. Although diet and exercise are discussed here as a separate topic, both play an important role in other health areas, including obesity, diabetes, and oral health. These two health areas of health behavior, while related in terms of impact on outcomes (e.g., body mass index [BMI]), may be distinct in terms of how MI (and other interventions) can positively impact change or maintenance. There have been 24 published empirical articles identified as utilizing MI as an intervention to modify diet and/or exercise behaviors. A summary of the empirical articles utilizing MI in the areas of diet and exercise is provided in Table 1. Overall, studies support the effectiveness of MI in the areas of diet and exercise (e.g., Berg-Smith et al., 1999, Bowen et al., 2002, Harland et al., 1999, Resnicow et al., 2000, Resnicow et al., 2001, Smith et al., 1997), both alone, and in combination with other interventions. Specifically, patients who received MI reported increased self-efficacy related to diet and exercise (e.g., Bennett et al., 2008, Resnicow et al., 2004), increased physical activity (e.g., Bennett et al., 2007, Carels et al., 2007, Harland et al., 1999; Hardcastle et al., 2008), reduced caloric intake (e.g., Befort et al., 2008), and increased fruit and vegetable consumption (e.g., Ahluwalia et al., 2007, Befort et al., 2008, Elliot et al., 2007, Hardcastle et al., 2008, Resnicow et al., 2000, Resnicow et al., 2001, Resnicow et al., 2004, Resnicow et al., 2005, Richards et al., 2006). Further, patients who received MI demonstrated decreased BMI (e.g., Hardcastle et al., 2008, Schwartz et al., 2007) after the intervention. Although MI was not always found to be more effective than other treatments, overall the findings support the clinical utility of MI in these areas.

The management of diabetes requires lifelong patient adherence to behaviors associated with diet restrictions, medication treatment, regular medical consultations, exercise regimens, restricted alcohol consumption, and smoking cessation (Clark and Hampson, 2001, Stott et al., 1996). Patients vary in their readiness and willingness to make recommended changes or to develop a lifestyle consistent with these stipulations, but most present with some degree of ambivalence about change (Rollnick, Kinnersley, & Stott, 1993).

Table 2 shows a summary of nine published empirical studies examining the use of MI in either Type 1 or Type 2 diabetes management. Participants in five studies were diagnosed with Type 1 diabetes (i.e., Channon et al., 2003, Channon et al., 2007; Ismail et al., 2008, Knight et al., 2003, Viner et al., 2003), whereas four studies focused on individuals with Type 2 diabetes (i.e., Brug et al., 2007, Clark and Hampson, 2001, Smith et al., 1997, Smith-West et al., 2007). Lifestyle changes (e.g., diet and exercise) were the focus of MI, regardless of type of diabetes. Studies examining the use of MI (e.g., Channon et al., 2003) or MI in combination with other interventions (e.g., Smith et al., 1997) provided evidence for the effectiveness of MI in this health area. MI was found to be effective in assisting patients control glucose levels (e.g., Channon et al., 2003, Channon et al., 2007, Smith et al., 1997, Viner et al., 2003, Smith-West et al., 2007), increase physical activity (e.g., Smith-West et al., 2007), decrease weight (e.g., Smith et al., 1997, Smith-West et al., 2007), and engage in dietary changes (e.g., Brug et al., 2007, Clark and Hampson, 2001), both alone and in combination with other interventions. Further, MI appeared to contribute to additional successes, such as increased self-efficacy (e.g., Viner et al., 2003) and increased sense of control over diabetes (e.g., Knight et al., 2003).

In recent years, oral health problems have been associated with a variety of systemic health issues, such as diabetes (Jansson, Lindholm, Lindh, Groop, & Bratthall, 2006), cardiovascular disease (Beck and Offenbacher, 2005, Genco et al., 2002), low birth weight (Jeffcoat, Guers, Reddy, Goldenberg, & Hauth, 2001), premature birth (Jeffcoat et al., 2001, Offenbacher et al., 2006), and respiratory illness (Mojon, 2002). Engaging in appropriate oral health care not only can prevent many dental problems but also may impact other systemic health issues, such as diabetes, heart and lung disease, and stroke (Peterson, 2003). Although MI is relatively new to oral health, it seems to be a promising method to improve oral health status, based on four available articles on caries prevention or attendance at dental appointments. Table 3 provides a summary of these published empirical studies. There is other important work in the oral health arena that focuses on health behaviors not included in the present review, including tobacco and alcohol use. MI has shown promise in preserving the oral health of infants and other children, with the intervention targeted at their mothers (e.g., Harrison et al., 2007, Weinstein et al., 2004, Weinstein et al., 2006) (or perhaps their fathers, or other caregivers). MI also seems to have potential to address the self-care of adolescents and adults, with particular applications to special populations (e.g., pregnant women) whose oral health status may be related to other health outcomes (e.g., length of pregnancy). Delay or avoidance of timely dental care is an important area that MI ultimately may help to address, although the present data are quite preliminary (e.g., Skaret et al., 2003). With only four studies published in this area, the field invites further work.

There were 37 published empirical articles in the areas of diet and exercise, diabetes, and oral health that were reviewed. Each of the studies reviewed focused on health-related behaviors that require life-long adherence in order to achieve maximum health benefits. For example, consistent glycemic control is essential in the treatment of diabetes to ensure the best possible outcomes. As such, after behavior change is achieved, the focus shifts to maintenance of these behaviors. Individuals who received MI showed significant behavior changes with important implications for their health across the areas of diet and exercise, diabetes, and oral health, more often than not. In many cases, behavior changes were maintained over a long period of time (e.g., one or two years); however, follow-up data beyond two years in these areas were not available, but should be a focus of future research.

Section snippets

Assessment of major issues

Development and implementation of MI, a treatment vehicle through which information, motivation, and recommendations can be provided to patients, has immense clinical implications. MI been shown to be effective in a number of health areas, including diet and exercise, diabetes management, and oral health; however, several methodological concerns require additional focus in order to more confidently demonstrate the effectiveness of MI. Specifically, training and practice issues, treatment

Conclusions and future research directions

MI is a promising intervention to encourage positive health behavior change in medical and dental settings (Resnicow et al., 2002), as noted in the studies presented here. Specifically, MI may help individuals improve their oral health, and may enhance adherence to diet and exercise modification programs, and diabetes management. MI has been found to be more effective than no treatment, and in many cases, more effective than other active treatments (Burke et al., 2003). MI also may be more

References (86)

  • RichardsA. et al.

    Motivating 18- to 24-year-olds to increase their fruit and vegetable consumption

    Journal of the American Dietetic Association

    (2006)
  • StottN.C.H. et al.

    Professional responses to innovation in clinical method: Diabetes care and negotiating skills

    Patient Education and Counseling

    (1996)
  • WebberK.H. et al.

    Motivational interviewing in internet groups: A pilot study for weight loss

    Journal of the American Dietetic Association

    (2008)
  • WeinsteinP. et al.

    Motivating parents to prevent caries in their young children

    Journal of the American Dental Association

    (2004)
  • WeinsteinP. et al.

    Motivating mothers to prevent caries: Confirming the beneficial effect of counseling

    Journal of the American Dental Association

    (2006)
  • AhluwaliaJ.S. et al.

    Pathways to health: Cluster-randomized trial to increase fruit and vegetable consumption among smoker in public housing

    Health Psychology

    (2007)
  • AmrheinP.C. et al.

    Client commitment language during motivational interviewing predicts drug use outcomes

    Journal of Consulting and Clinical Psychology

    (2003)
  • BeckJ.D. et al.

    Systematic effects of periodontitis: Epidemiology of periodontal disease and cardiovascular disease

    Journal of Periodontology

    (2005)
  • BefortC.A. et al.

    Journal of Behavioral Medicine

    (2008)
  • BennettJ.A. et al.

    Motivational interviewing to increase physical activity in long-term cancer survivors

    Nursing Research

    (2007)
  • BennettJ.A. et al.

    A telephone-only motivational intervention to increase physical activity in rural adults: A randomized controlled trial

    Nursing Research

    (2008)
  • Berg-SmithS.M. et al.

    A brief motivational intervention to improve dietary adherence in adolescents

    Health Education Research

    (1999)
  • BrodieD.A. et al.

    Motivational interviewing to promote physical activity for people with chronic heart failure

    Journal of Advanced Nursing

    (2005)
  • BurkeB.L. et al.

    The efficacy of motivational interviewing and its adaptations

  • BurkeB.L. et al.

    The efficacy of motivational interviewing: A meta-analysis of controlled clinical trials

    Journal of Consulting and Clinical Psychology

    (2003)
  • BurkeB.L. et al.

    The emerging evidence base for motivational interviewing: A meta-analytic and qualitative inquiry

    Journal of Cognitive Psychotherapy

    (2004)
  • ButterworthS. et al.

    Effect of motivational interviewing-based health coaching on employees' physical and mental health status

    Journal of Occupational Health Psychology

    (2006)
  • CarelsR.A. et al.

    Using motivational interviewing as a supplement to obesity treatment: A stepped-care approach

    Health Psychology

    (2007)
  • ChannonS.J. et al.

    A multicenter randomized controlled trial of motivational interviewing in teenager with diabetes

    Diabetes Care

    (2007)
  • ChannonS. et al.

    A pilot study of motivational interviewing in adolescents with diabetes

    Archives of Disease in Childhood

    (2003)
  • ColbyS.M. et al.

    Brief motivational interviewing in a hospital setting for adolescent smoking: A preliminary study

    Journal of Consulting & Clinical Psychology

    (1998)
  • DunnC. et al.

    The use of brief interventions adapted from MI across behavioral domains: A systematic review

    Addiction

    (2001)
  • ElliotD.L. et al.

    The PHLAME firefighters' study: Feasibility and findings

    American Journal of Health Behavior

    (2004)
  • ElliotD.L. et al.

    The PHLAME (Promoting Healthy Lifestyles: Alternative Models' Effects) firefighter study: Outcomes of two models of behavior change

    Journal of Occupational and Environmental Medicine

    (2007)
  • FuemmelerB.F. et al.

    Psychosocial mediation of fruit and vegetable consumption in the Body and Soul effectiveness trial

    Health Psychology

    (2006)
  • GreavesC.J. et al.

    Motivational interviewing for modifying diabetes risk: A randomized controlled trial

    British Journal of General Practice

    (2008)
  • HarlandJ. et al.

    The Newcastle exercise project: A randomised controlled trial of methods to promote physical activity in primary care

    British Medical Journal

    (1999)
  • HarrisonR. et al.

    Effects of motivational interviewing on rates of early childhood caries: A randomized trial

    Pediatric Dentistry

    (2007)
  • HertiageJ. et al.

    Dilemmas of advice: Aspects of delivery and reception in interactions between health visitors and first-time mothers

  • HettemaJ. et al.

    Motivational interviewing

    Annual Review of Clinical Psychology

    (2005)
  • IsmailK. et al.

    Motivational enhancement therapy with and without cognitive behavior therapy to treat type 1 diabetes

    Annals of Internal Medicine

    (2008)
  • JanssonH. et al.

    Type 2 diabetes and risk for periodontal disease: a role for dental health awareness

    Journal of Clinical Periodontology

    (2006)
  • JeffcoatM.K. et al.

    Periodontal infection and preterm birth

    Journal of the American Dental Association

    (2001)
  • Cited by (419)

    View all citing articles on Scopus

    This paper is based on a preliminary doctoral examination in the West Virginia University Department of Psychology, submitted by the first author and supervised by the second author.

    ☆☆

    We would like to thank William Fremouw, Ph.D. for his involvement in the preliminary examination process. Appreciation also is extended to colleagues in the Anxiety, Psychophysiology, and Pain Research Laboratory in the Department of Psychology at West Virginia University for their support of this project.

    View full text