Evaluating Primary Care Behavioral Counseling ...

OverviewRisky behaviors are a leading cause of preventable

morbidity and mortality, yet behavioral counselinginterventions to address them are underutilized inhealth care settings. Research on such interventionshas grown steadily, but the systematic review of thisresearch is complicated by wide variations in theorganization, content, and delivery of behavioralinterventions and the lack of a consistent languageand framework to describe these differences. TheCounseling and Behavioral Interventions WorkGroup of the United States Preventive Services TaskForce (USPSTF) was convened to address adaptingexisting USPSTF methods to issues and challengesraised by behavioral counseling intervention topicalreviews.

The systematic review of behavioral counselinginterventions seeks to establish whether suchinterventions addressing individual behaviorsimprove health outcomes. Few studies directlyaddress this question, so evidence addressing

whether changing individual behavior improveshealth outcomes and whether behavioral counselinginterventions in clinical settings help people changethose behaviors must be linked. To illustrate thisprocess, we present 2 separate analytic frameworksderived from screening topic tools that we developedto guide USPSTF behavioral topic reviews.

No simple empirically validated model capturesthe broad range of intervention components acrossrisk behaviors, but the 5 A’s construct—assess,advise, agree, assist, and arrange— adapted fromtobacco cessation interventions in clinical careprovides a workable framework to report behavioralcounseling intervention review findings. Weillustrate the use of this framework with generalfindings from recent behavioral counselingintervention studies. Readers are referred to theUSPSTF (www.preventiveservices.ahrq.gov or 1-800-358-9295) for systematic evidence reviewsand USPSTF recommendations based on thesereviews for specific behaviors.

M–47

Evaluating Primary Care BehavioralCounseling Interventions: An Evidence-based Approach

Evelyn P. Whitlock, MD, MPH; C. Tracy Orleans, PhD; Nola Pender, PhD, RN, FAAN;Janet Allan, RN, PhD, CS

From the Oregon Health & Science University Evidence-based Practice Center, Kaiser Permanente/CHR (Whitlock), Portland,Oregon; The Robert Wood Johnson Foundation (Orleans), Princeton, New Jersey; School of Nursing, University of Michigan(Pender), Ann Arbor, Michigan; School of Nursing, University of Texas Health Science Center (Allan), San Antonio, Texas.

The authors of this article are responsible for its contents, including any clinical or treatment recommendations. No statement inthis article should be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department ofHealth and Human Services.

Address correspondence to: Evelyn P. Whitlock, MD, MPH, Kaiser Permanente/CHR, 3800 North Interstate Avenue, Portland, OR97227-1110. E-mail: [email protected].

Reprints are available from the AHRQ Web site (www.preventiveservices.ahrq.gov), through the National Guideline Clearinghouse(www.guideline.gov), or in print through the AHRQ Publications Clearinghouse (call 1-800-358-9295 or [email protected]).

This chapter first appeared as an article in Am J Prev Med. 2002;22(4):267-284.

EpidemiologyIn 1998, the Agency for Healthcare Research and

Quality (AHRQ) reconvened the USPSTF to updateits recommendations for clinical preventive services.This Task Force represents primary care disciplines(nursing, pediatrics, family practice, internalmedicine, and obstetrics/gynecology), preventivemedicine, and behavioral medicine. Two evidence-based practice centers (EPCs)—Oregon Health &Science University and RTI–University of NorthCarolina—were contracted to prepare systematicevidence reviews that the USPSTF uses indeveloping its recommendations for preventive care.Although the USPSTF evidence-based methods arewidely applicable throughout medicine, to date theyhave been used primarily to assess services such aspreventive screening, rather than those requiringbehavioral counseling.1,2 The current USPSTFrecognized a 2-fold need: (1) to expand itsevidence-based approach to better assess behavioralcounseling interventions, and (2) to formulatepractical communication strategies for describingservices that are effective in changing behavior.

The Counseling and Behavioral InterventionsWork Group of the USPSTF adapted the USPSTFgeneric screening analytic framework, which guidessystematic reviews, to address behavioral topics morespecifically, and it has promoted a consistentorganizational construct for describing behavioralcounseling interventions. Clinicians are referred tocurrent products of the USPSTF(www.preventiveservices.ahrq.gov or 1-800-358-9295) for systematic evidence reviews of specificbehavioral counseling topics and related USPSTFevidence-based recommendations and clinicalconsiderations beyond the scope of this paper.

This paper has 3 purposes:

(1) To promote a broader appreciation of theimportance of behavioral counselinginterventions in clinical care and the context fortheir delivery.

(2) To describe the generic analytic frameworksdeveloped to guide the systematic review ofbehavioral counseling topics for the currentUSPSTF.

(3) To detail the practical organizational construct(the 5 A’s) adopted by the USPSTF to describeintervention research more consistently in orderto foster its application in clinical settings.

BackgroundHealthy People 20103 sets 2 major goals for the

United States: (1) to increase quality and years ofhealthy life, and (2) to eliminate health disparitiesamong different segments of the population. Thenext decade offers unprecedented opportunities forhealth care systems and providers to address thesegoals by promoting healthy lifestyles among thediverse populations they serve and by adoptingpolicies that will institutionalize preventive services.

Changing the health behaviors of Americans hasthe greatest potential of any current approach fordecreasing morbidity and mortality and forimproving the quality of life across diversepopulations.4 In their landmark paper, McGinnisand Foege5 linked 50% of the mortality in theUnited States from the 10 leading causes of death tolifestyle-related behaviors such as tobacco use, poordietary habits and inactivity, alcohol misuse, illicitdrug use, and risky sexual practices. These behaviorsremain problematic in today’s society despite havingbeen previously targeted for improvement.6 Thus,the U.S. Department of Health and Human Serviceshas designated 5 lifestyle factors as Healthy People2010 3 health indicators by which to track progressin improving the health of the nation over the nextdecade (Table 1). Improving health behaviors is animportant approach to health disparities, becausethose who are economically and/or sociallydisadvantaged, including those in low-incomeethnic/racial minority groups, disproportionatelybear the prevalence of risky health behaviors and theburden of preventable morbidity and mortality.7

The unabated impact of health-damagingbehaviors among Americans makes it imperative thathealth care providers and health care systemsseriously consider these behavioral issues and acceptthe challenge of routinely providing qualitybehavioral counseling interventions where proveneffective. The 1996 edition of the Guide to Clinical

M–48

Evaluating Behavioral Counseling Interventions


M–49

Health indicator 1997 baseline 2010 goals

Tobacco use (%)Cigarette smoking adults 24 12

American Indian/Alaskan Native 34 12Family income, poor level 34 12

Current tobacco use by youth (past 30 days) 43 21Smoking cessation attempts

Adults 43 75Pregnant women 12 30Adolescents (grades 9-12) 73 84

Overweight and obesity (%)Proportion of adults at healthy weight† 42 60

Mexican Americans 30 60Lower income (< 130% poverty threshold) 29 15

Obesity‡ in adults (≥ 20 years) 23 15Overweight/obesity in children and teens (6-19 years)§ 11 5

Physical activity (%)No leisure-time physical activity (≥18 years) 40 20

American Indians/Alaskan Native, African American, or Hispanic 46-54 20

Moderate physical activity||Adults (≥18 years) 15 30Adolescents (grades 9-12) 20 30

Substance abuseProportion of adults exceeding low-risk drinking guidelines (%)◊

Females 72 50Males 74 50

Alcohol-related auto deaths 6.1/100,000 4/100,000American Indian or Alaska Native 19.2/100,000 4/100,000People aged 15-24 years 11.7/100,000 4/100,000

High school seniors never using alcohol (%) 19 29Binge drinking (%)

Adolescents (12-17 years) 8.3 3High school seniors 32 11College students 39 20Adults 16 6

Youth (12-17) using marijuana in the last 30 days (%) 9.4 0.7High school seniors never using illicit drugs (%) 46 56

Responsible sexual behavior (%)Unmarried females (18-44 years) whose partners used condoms 23 50Teens abstain from sex or use condoms 85 95

Table 1. Healthy People 2010 leading health indicators*

* Other leading health indicators include mental health, injury and violence, environmental quality, immunizations, and access to health care.† 18.5 ≥ BMI ≤ 25.‡ BMI of ≥30 .§ ≥ 95th percentile of gender- and age-specific BMI from year 2000 U.S. growth charts.|| Moderate activity of 30 minutes a day 5 or more days a week.◊ Males > 14 drinks/week or > 4 drinks/occasion; females > 7 drinks/week or > 3 drinks/occasion. Note: BMI indicates body mass index. Source: From Healthy People 20103 Adapted from public domain document; also available online at http://www.health.gov/healthypeople.

Preventive Services by the USPSTF concluded:“Effective interventions that address personal healthpractices . . . [for] . . . primary prevention . . . holdgreater promise for improving overall health thanmany secondary preventive measures, such as routinescreening for early disease. Therefore, cliniciancounseling that leads to improved personal healthpractices may be more valuable than conventionalclinical activities, such as diagnostic testing.”1

Nevertheless, rates of behavioral counselingintervention by pediatricians, nurse practitioners,obstetrician-gynecologists, internists, and familyphysicians for the priority behaviors discussed abovestill fall far below national targets.3,8,9 In fact, gaps inthe delivery of clinical preventive services are greaterfor behavioral counseling than for screening orchemoprevention.10 This stems in part from therelative paucity of good research evidence to supportthe behavioral counseling interventionrecommendations in the 1996 Guide to ClinicalPreventive Services.1

The quality and quantity of good researchevidence for the effectiveness of behavioralcounseling interventions are increasing. Briefinterventions integrated into routine primary carecan effectively address the most common andimportant risk behaviors.11-22 The strongest evidencefor the efficacy of primary care behavior-changeinterventions comes from tobacco cessationresearch11,12,14,15,19 and, to a lesser extent, problemdrinking.11,16-19,21,22 Accumulating evidence also showsthe effectiveness of similar interventions for otherbehaviors.11,19,20 These interventions often providemore than brief clinician advice. Effectiveinterventions typically involve behavioral counselingtechniques and use of other resources to assistpatients in undertaking advised behavior changes.12,19

For example, intervention adjuncts to brief clinicianadvice may involve a broader set of health care teammembers (eg, nurses, other office staff, healtheducators, and pharmacists), a number ofcomplementary communication channels (eg,telephone counseling,22,23 video or computer-assistedinterventions,24–26 self-help guides,27 and tailoredmailings28), and multiple contacts with thepatient.12,14,19,29

Rationale for BehavioralCounseling Interventions inClinical Care

Health care providers and their staff play a uniqueand important role in motivating and assistingpatients’ healthy behavior changes. Patients reportthat primary care clinicians are expected sources ofpreventive health information and recommendationsfor patients.30 For instance, in a recent survey, thevast majority (92% to 98%) of adult members ofhealth maintenance organizations (HMO) indicatedthat they expected advice and help from the healthcare system in key behaviors, such as diet, exercise,and substance use.31 Similarly, health care providersgenerally accept32 and value their role in motivatinghealth promotion and disease prevention.33,34

Health care systems are natural settings forinterventions to improve health behaviors for manyindividuals because repeated contacts typically occurover a number of years. Interventions to helppatients change unhealthy behaviors, like treatmentsfor patients with chronic disease, often requirerepetition for modest effects over time. Continuityof care offers opportunities to sustain individualmotivation, assess progress, provide feedback, andadjust behavior change plans.35

In fact, most clinicians have multipleopportunities to intervene with patients on mattersrelated to health behavior change: patients youngerthan 15 years average 2.4 visits per person annuallyto office-based physicians, and those 15 years of ageand older average 1.6 to 6.3 visits per year, with visitfrequency increasing with age.36 Moreover, 93% ofchildren and youth and 84% of adults 18 years ofage and older have a specific source of ongoinghealth care.3 Not surprisingly, people with a usualsource of health care are more likely than thosewithout to receive a variety of clinical preventiveservices.3

The health care setting is not the only setting forapproaches to support healthy behaviors. The Guideto Community Preventive Services features evidence-based recommendations from the Task Force onCommunity Preventive Services for population-based interventions. Those recommendations

M–50


include policy or environmental changes orindividual and group interventions outside theclinical setting intended to change risky behaviors;reduce specific diseases, injuries and impairments;and address environmental and ecosystemchallenges.37 These preventive policies andapproaches complement the individually focusedinterventions that the USPSTF addresses.

Objectives and Scope ofBehavioral CounselingInterventions

Behavioral counseling interventions in clinicalcare are those activities delivered by primary careclinicians and related health care staff to assistpatients in adopting, changing, or maintainingbehaviors proven to affect health outcomes andhealth status. Common health-promoting behaviorsinclude smoking cessation, healthy diet, regularphysical activity, appropriate alcohol use, andresponsible use of contraceptives.

Behavioral counseling interventions occur all orin part during routine primary care and may involveboth visit-based and outside interventioncomponents. For instance, assessment of behavioralhealth risks may occur at the time of enrollment in ahealth plan or at the time of a clinical visit.Behavioral counseling may take place in routineprimary care visits and/or through telephonecontacts or personalized mailings of self-help guidesor materials. Referral to more intensive clinics in thecommunity also may be included. While theUSPSTF primarily evaluates interventions thatinvolve clinicians as part of routine primary care,USPSTF liaisons assigned to a particular behavioraltopic define the scope of clinical interventionapproaches reviewed for any given topic, such asproblem drinking or physical activity.

Behavioral counseling interventions differ fromscreening interventions in several important waysthat affect the ease and likelihood of their beingdelivered. Behavioral counseling interventionsaddress complex behaviors that are integral to dailyliving; they vary in intensity and scope from patientto patient; they require repeated action by both

patient and clinicians, modified over time, to achievehealth improvement; and they are stronglyinfluenced by multiple contexts (family, peers,worksite, school, and community). Further,“counseling” is a broadly used but imprecise termthat covers a wide array of preventive andtherapeutic activities, from mental health or maritaltherapy to the provision of health education andbehavior change support. Thus we have chosen touse the term “behavioral counseling interventions”to describe the range of personal counseling andrelated behavior-change interventions that areeffectively employed in primary care to help patientschange health-related behaviors. As with its use inother contexts, “counseling” here denotes acooperative mode of work demanding activeparticipation from both patient and clinician thataims to facilitate the patient’s independent initiativeand ability to cope.38 Engaging patients actively inthe self-management practices needed to change andmaintain healthy behaviors is a central componentof effective behavioral counseling interventions.

Theories and Models of BehaviorChange

Behavior change theories and models from thesocial and behavioral sciences explain the biological,cognitive, behavioral, andpsychosocial/environmental determinants of health-related behaviors. Thus they also defineinterventions to produce changes in knowledge,attitudes, motivations, self-confidence, skills, andsocial supports required for behavior change andmaintenance.39 The application of relevanttheoretical models to behavioral counselinginterventions is an important contribution tostrengthening health research in this area.40 Aliterature review of 1,174 articles evaluating healthbehavior, education, and promotion interventionspublished between 1992 and 1994 found that44.8% of these were explicitly theory based.41 Sixtheories and models addressing determinants ofhealth-behavior change at the intrapersonal,interpersonal, and environmental levels (Table 2)and 2 cross-theoretical key constructs/theories weremost commonly cited in this research. Promising, ifnot substantial, empirical evidence supports the


M–51

M–52


Level addressed Theory/model Focus Key concepts

Theories that address Health belief model Peoples’ perceptions Perceived susceptibilityhow individual factors of the threat of a health Perceived severitysuch as knowledge, problem and appraisal of Perceived benefits of actionattitudes, beliefs, prior behavior recommended Perceived barriers to actionexperience, and to prevent or manage Cues to actionpersonality influence problem Self-efficacybehavioral choices

Theory of reasoned People are rational Behavioral intentionaction/theory of beings whose intention Subjective normsplanned behavior to perform a behavior Attitudes

strongly relates to its Perceived behavioral controlactual performance through beliefs, attitudes, subjective norms, and perceived behavioral control

Stages of change/ Readiness to change or Precontemplationtranstheoretical model attempt to change a Contemplation

health behavior varies Preparationamong individuals and Actionwithin an individual over Maintenancetime. Relapse is a Relapsecommon occurrence and part of the normal process of change.

Theories that address Social cognitive Behavior is explained by Observational learning processes between the theory/social dynamic interaction among Reciprocal determinismindividual and primary learning theory personal factors, Outcome expectancygroups that provide environmental influences, Behavioral capacitysocial identity, support, and behavior Self-efficacyand role definition Reinforcement

Community Processes by which Participation and relevanceorganization/building community groups are Empowerment

helped to identify and Community competenceaddress common issue selectionproblems or goals

Social marketing The application of Consumer orientation commercial marketing Audience segmentation technologies to increase Communication channels the practice of healthy analysisbehaviors in order to Voluntary exchange of goodsimprove individual and and servicescollective well-being

Table 2. Six most commonly cited behavior change models, theories, and constructs—focus and key concepts

validity of all 8 theories in predicting or changinghealth behavior.41 In addition to those listed inTable 2, self-efficacy and social network/supportwere the other 2 most commonly cited constructs inthe current literature. Self-efficacy is an individual’slevel of confidence in his or her own skills andpersistence to accomplish a desired goal and predictsfuture behavior across a wide variety of lifestyle riskfactors.42 Social networks are a person-centered webof social relationships.43 These relationships providesocial support that can assist the individual through“stress-buffering” and other mechanisms.43

These theories focus on diverse, interacting levelsof influence on an individual’s behavior. On theintrapersonal level, multiple internal factorsinfluence an individual’s behavioral choices andactions, and there is considerable variability in thesefactors among individuals with the same objectivehealth behavior. For example, in the stages-of-change/transtheoretical model (Table 2), behavioralchange is thought of as an ongoing process withmultiple stages that often includes relapse andrecycling into renewed efforts to change.44 On theinterpersonal level, individual behavioral choicesoccur in a context that includes the influence ofsocial and environmental conditions in the familyand larger community.41,45

Behavioral influences operate within a broadlyconceptualized ecological paradigm emphasizing thata dynamic interaction between functional levels—intrapersonal, interpersonal, and the physicalenvironment—continues over an individual’slifetime, and that age, gender, race, ethnicity, andsocioeconomic status play a critical role in healthand health decisions.40,46 Similarly, the Institute ofMedicine47 recently concluded that “interventionsmust recognize that people live in social, political,and economic systems that shape behaviors andaccess to the resources they need to maintain goodhealth.”

According to another recent Institute of Medicinereport,40 there is an emerging consensus that socialand behavioral research and intervention effortsshould be based on this broader ecologic model thatincorporates and relates focused approaches acrosslevels. Thus, omission of any key dimension in

research or practice reduces the likelihood ofsuccessfully addressing problem behaviors, such assmoking.48 More than a brief overview of theoriesand models is beyond the scope of this paper andcan be found elsewhere.39-46

Although these theoretical constructs areunfamiliar to many clinicians, they can helppractitioners conceptualize the complex context inwhich individual behavioral choice occurs and thevariability among patients in their receptivity tobehavioral counseling interventions at any one time.These insights can clarify barriers, opportunities, andthe relative intensity of intervention needed tosuccessfully address behavior change for a givenindividual.

Generally speaking, less-intensive outside supportand intervention are needed for individuals withmore change-predisposing attributes than for thosewith fewer such attributes48,49 (Table 3). Scarceresources can be focused on strengthening anindividual’s factors favoring change and targeting themost intensive support to people with the fewestpre-disposing attributes. Theoretical perspectivesalso make clear the complementary role played bypolicies and practices in settings outside health carein promoting healthy behaviors across society.

The Clinician-PatientRelationship

As our understanding of behavioral counselinginterventions has become more sophisticated,interventions have evolved beyond the limits of one-on-one interactions between a clinician and apatient. However, the use of additional resourceswithin and outside the primary care setting tosupport the clinician by no means undermines theimportance of the clinician-patient relationship inpromoting behavior change. Effective cliniciancommunication is important for a variety of patientoutcomes.50,51 Clinician advice to change lifestylehabits is associated with increased efforts tochange52,53 and is effective in encouraging smokingcessation,11,12,14,15 reducing problem drinking,11,16 andmodifying some activity- and diet-associatedcardiovascular risk factors.11,20 Clinician advice is also


M–53

associated with increased satisfaction with medicalcare.30,54,55 Such advice has been suggested to “prime”patients, especially women, to attend to and act onsubsequent educational information.56 In a recentcross-sectional study among members of a managedcare organization,57 receipt of professional advice tochange was associated with a higher readiness tochange smoking, physical activity, and dietbehaviors. Preliminary data also suggest that advicefrom one’s health care provider based on personalhealth status is a very strong external cue to health-promoting action.58

The clinician employing an empathetic“partnership” approach avoids engendering resistanceto behavior change advice.59 Such an approachemphasizes the patient’s role in interpreting adviceand explores, rather than prescribes, how best toproceed. According to a Toronto consensusconference on doctor-patient communication,60

“effective communication between doctor andpatient is a central function that cannot bedelegated.”

The Potential Impact ofHealth Behavior-ChangePrograms in Clinical Care

Appreciating behavioral counseling interventionsrequires a true population-based medicineperspective (ie, intervening with individuals, butrecognizing that the health benefits may not be asclinically visible individually as they are clinicallymeaningful when considered for the whole).

Individually, brief behavioral counselinginterventions that are feasible in health care settingsoften have only modest behavior change impacts.For example, only 5% to 15% of those receiving anintervention make clinically significant changes, suchas quitting smoking12 or reducing heavy drinking.11

Even at a population level, overall risk factorstypically change only 1% to 20%.16,17,19,20,22,61

However, these “modest” impacts translate tosignificant benefits to the health of the population(and to multiple individuals) when systematicallyapplied to a large proportion of those in need.48,62-65

This opportunity for substantial public healthbenefit comes about only when behavior changeinterventions are applied broadly to entirepopulations of patients. Given this, population-based behavioral interventions generally offer a rangeof intervention options including motivationalstrategies designed for people not ready to change64

(see sidebar, “Impact of Health Behavior ChangePrograms”).

M–54


1. Strongly wants and intends to change for clear, personal reasons.

2. Faces a minimum of obstacles (information processing, physical, logistical, or environmental barriers) to change.

3. Has the requisite skills and self-confidence to make a change.

4. Feels positively about the change and believes it will result in meaningful benefit(s).

5. Perceives the change as congruent with his/her self-image and social group(s) norms.

6. Receives reminders, encouragement, and support to change at appropriate times and places from valued persons and community sources, and is in a largely supportive community/environment for the change.

Table 3: Attributes from health behavior change theories and models that predispose an individual tosuccessful behavior change: 39,45,49

Impact of Health Behavior Change Programs

Highly efficacious, intensive group tobacco cessationapproaches12,48,64,66 have typically been perceived asproducing higher quit rates than primary carebehavioral counseling interventions. Groupapproaches produce quit rates of 30% to 40% butreach only a small proportion of highly motivatedsmokers volunteering for treatment (roughly 3% to5% of all smokers). Thus, their potential impact onthe prevalence of smoking (Impact = ParticipationRate x Efficacy) is substantially less thansystematically delivered primary care interventions,which can feasibly reach the 70% of smokers who

Continued on page M-55

Practical Approaches toOvercome Barriers toBehavioral and CounselingInterventions

Numerous barriers to preventive service deliverycontinue to exist in present-day health care settings,most of which are still organized mainly aroundsymptom-driven, acute illness care.67,68 These barriersinclude a focus on more medically urgent issues; lackof time; inadequate clinician training, self-confidence, or reimbursement; low patient demand;and lack of supportive resources.3,69 Further,feedback to clinicians about results of preventive careis largely non-existent or can even be negative.69 Forexample, clinicians or their staffs may never “hear”about the patients who followed through on areferral or made positive lifestyle changes, but mayencounter complaints about repeated advice to quitsmoking, even when voiced by only a few.

Unfortunately, most of these challenges areexacerbated for health behavior-changeinterventions. Thus, risk assessment and behavioralcounseling interventions are delivered even lessfrequently than screenings.8 Moreover, althoughclinicians increasingly agree that most health-promoting behaviors are important to patients’health,32 they report skepticism about patients’willingness to change these behaviors and abouttheir own ability to intervene successfully in theseareas.70,71 Clinicians often lack the knowledge, skills,and support systems to quickly and easily provide arange of different behavioral counselinginterventions, particularly in the limited time

available.69,72,73 These barriers provide an importantrationale for proposing a consistent overall approach(such as the 5 A’s, discussed below) for describingbehavioral counseling interventions across the rangeof topics in clinical care.

Evaluations of continuing medical educationefforts show that programs based on the principlesof adult learning that build clinician skills usinginteractive, sequential learning opportunities insettings such as workshops, small groups, andindividual training sessions appear to have thegreatest influence on clinician practices and patientoutcomes.74 Even relatively brief physician trainingalong these lines (2 to 3 hours) can improve thedelivery of clinical preventive services.75,76

However, clinician training may be efficaciousonly in the presence of an office-support programthat assists clinicians in carrying out behavioralcounseling interventions and incorporating theminto routine care.77,78 As Solberg et al79 has noted,“Without such systems, delivery of preventiveservices must depend on the memory, motivation,and time of individual clinicians.” Fortunately, wealso have a better understanding of the organizedoffice or health-plan processes that support thesystematic and consistent delivery of clinicalpreventive services. These systems typically consistof (1) preventive services guidelines; (2) basicsupport processes that identify and activate thosewho need a service, summarize needed services onthe patient chart, and remind the clinician during avisit; and (3) prevention resources to provide in-clinic and after-clinic counseling, support, andfollow-up.80 A recent randomized controlled trial81

reported that, compared with control practices,community family practices demonstratedsignificantly increased clinical preventive servicesdelivery 1 year after receiving practice-tailoredsystems support for preventive service delivery.Delivery of behavioral counseling interventions wasparticularly improved. The Put Prevention IntoPractice (PPIP) program, sponsored by AHRQ, hasa variety of materials to help make these services anintegral part of primary care. PPIP has developedtools to assist clinicians in determining whichclinical preventive services patients should receive,


M–55

Impact of Health Behavior Change Programs(continued)

visit their clinicians each year and result in 5% to 10%overall quit rates.

Applying a similar public health approach, modesteffective clinical interventions addressing problemdrinking21,22,62 and dietary change61 are projected tohave significant population impact when broadlydelivered.

and it produces guides and materials for servicedelivery in a variety of settings.82 PPIP also providesresources for patients to guide health maintenancedecisions and to keep track of their preventive care.

Ongoing innovations in the design and deliveryof behavioral counseling interventions can alsoaddress barriers, improve patient access, and increasetreatment effectiveness. Clinicians’ efforts areenhanced when the entire health care team takesappropriate and complementary roles in deliveringefficacious interventions.29,83,84 For example, healtheducators and nurse case managers who contact andsupport smokers between visits85 extend interventionopportunities beyond the initial primary care visit.Coordination with resources outside the clinicalsetting, such as programs and services throughvoluntary agencies and other community resources,can help patients conveniently access neededsupports after they leave the visit.67 This integrationmay increase health care system efficiency andimpact by creating congruence between clinicalinterventions and the broader community.86

Expanding communication technologies allow bothpassive and interactive use25 of telephones, videos,CD-ROMs, the Internet, and other computer-assisted venues to enhance and personalizebehavioral intervention content28,87 and to prolongcontact with the patient, while reducing the servicesthat must be directly provided by clinical staff.67

Such computer-based print, telephone, and videocommunications have boosted treatment outcomesover standard “one-size-fits-all” interventions inseveral behavioral areas (eg, smoking cessation anddiet modification), with greatest benefits sometimesseen in low-income populations.88-90 Although someof these technologies are relatively new and stillunder evaluation, advances in information andcommunication technologies hold great promise forenhancing intervention efficiency by automatingassessment, education, and patient contacts,especially for ongoing follow-up and support. Takentogether, these ongoing innovations offeropportunities to address key barriers to behavioralcounseling interventions in clinical settings.

Evidence-Based Methods forEvaluating BehavioralCounseling Interventions

We developed 2 interrelated generic analyticframeworks to guide the systematic review ofbehavioral topics (Figures 1 and 2). These analyticframeworks were derived from those developed forscreening topics.2 They separately frame the 2 mainquestions to consider when systematically reviewingrelevant clinical behavioral intervention research,namely: (1) Does changing individual healthbehavior improve health outcomes? (Figure 1) and(2) Can interventions in the clinical setting influencepeople to change their behavior? (Figure 2). Morein-depth key questions (KQs) for each mainquestion are detailed in the text notes on eachanalytic framework diagram, and the relevantsections of the diagram are numbered to correspondto these key questions.

Analytic Framework 1: DoesChanging Individual HealthBehavior Improve HealthOutcomes?

Clinical interventions are predicated on afoundation of epidemiological research thatadequately substantiates the link between particularbehaviors and health outcomes,2 as depicted inFigure 1 (Analytic Framework 1, KQs 1, 2, 5). Forinstance, there is strong consistent evidence thattobacco use, sedentary lifestyle, and improper dietlead to negative clinical and functional healthoutcomes,1 and, conversely, that smoking cessation,exercise improvement, and dietary improvement leadto positive clinical and functional health outcomes.However, few behavior change intervention studiesactually document long-term health outcomes (KQ8). Therefore, we usually must rely on linking upseparate bodies of evidence (represented here by the2 interrelated but separate analytic frameworks) todemonstrate whether clinical interventions improvehealth behaviors and lead to better health outcomes.

M–56


The USPSTF may elect to summarize, but notsystemically review, the evidence supporting the linkbetween health-behavior change and outcomes(shown here in Analytic Framework 1) when either:(1) the evidence has been reviewed in a previousUSPSTF report and addresses all issues of currentconcern, or (2) a good-quality systematic reviewconducted by another reputable body is availablethat meets USPSTF standards for grading evidenceand addresses the behaviors and outcomes that theUSPSTF is interested in. In such instances, AnalyticFramework 1 may be dispensed with altogether andattention focused on the literature addressinginterventions to effect the desired behavior change(discussed below under Analytic Framework 2).

However, even when an evidence review does not

formally undertake the key questions in AnalyticFramework 1, the epidemiologic evidence linkinghealth behavior change to health benefits illustratedin this diagram can help define appropriatebehavior-change outcome measures for thesystematic review of behavioral counselinginterventions represented by Analytic Framework 2(Figure 2). Ideally, behavior-change outcomemeasures of interest in a particular behavioral revieware defined as those related epidemiologically toreductions in morbidity and mortality directly (KQ6) or through intermediate outcomes (KQs 2 and 5linked together). For behaviors such as improperdiet and insufficient physical activity, intermediateoutcomes may include physiological risk factors,such as blood pressure, weight, and cholesterol level,


M–57

Sustainedhealth behavior

change

Behaviorcounseling

Intervention(s)in clinical care

7

Adverse effects

5

This linkaddressedin analytic

framework 2

Intermediate healthimprovement or risk

factor reductionReduction in morbidity

and/or mortality

Other postitiveoutcomes

1,2

8

6

4

3

Figure 1. Does changing individual health behavior improve health outcomes?Analytic framework 1

1. Do changes inpatients’ healthbehavior improvehealth or reduce riskfactors?

2. What is therelationship betweenduration of healthbehavior change andhealth improvement(ie, minimum duration,minimum level ofchange, change/responserelationship)?

3. What are the adverseeffects of healthbehavior change?

4. Does health behaviorchange produce otherpositive outcomes (eg,patient satisfaction,changes in otherhealth care behaviors,improved function,decreased use ofhealth careresources)?

5. Is risk factor reductionor measured healthimprovementassociated withreduced morbidityand/or mortality?

6. Is sustained healthbehavior changedirectly related toreduced morbidityand/or mortality?

7. Are behavioralcounselinginterventions in clinicalcare directly related toimproved health or riskfactor reduction?

8. Are behavioralcounselinginterventions in clinicalcare directly related toreduced morbidityand/or mortality?

Analytic framework 1–Key Questions

M–58


Interventioncondition*1

Clinicalpopulation

Adverse effects

Other postitiveoutcomes

1,2, 3, 4Interventioncondition*2

Measures ofbehavior change

Ongoing orsustained

behavior change

Adverse effects

Social-environmental

influences13

HealthCare systeminfluences

12 a-c

Assessment

5, 6, 7, 10, 11Intervention

...**

8, 9, 10, 11

Follow-upIntervention

Figure 2. Can interventions in the clinical setting influence people to change their behavior?Analytic framework 2

1. Are there distinctpatient groups forwhom differentassessment andbehavioral counselingintervention strategiesapply?

2. What patientcharacteristics (eg,sociodemographics—including age,race/ethnicity,gender—healthstatus, risk status,behavioral habits, andinterest in benefitsand barriers tochange) are critical toassess prior tobehavioral counselingintervention?

3. What are valid,reliable, feasible, andaccessible tools forbehavioralassessment ofpatients (and family,as appropriate)?

4 What are adverseeffects associatedwith behavioralassessment?

5. Do behavioralcounselinginterventions alterhealth behavior in thetargeted group?

6. What are the essentialelements ofefficaciousinterventions (ie, what,how, when, where, towhom, by whom, forhow often, and forhow long)?

7. Are behavioralcounselinginterventionsparticularly effectiveor ineffective inpatient subgroups?

8. How long are targetedbehavior changesmaintained afterbehavioral counselingintervention?

9. What type of ongoingassistance or supportis needed to achieveor maintain targetedbehavior changes?

10. Do behavioralcounselinginterventions produceother positiveoutcomes (eg,mediators of behaviorchange, changes inother healthbehaviors, andimprovements infunctioning)?

11. What are adverseeffects associated

with behavioralcounselingintervention?

12. Which of the followingsystems influencesfacilitate/impedebehavioralassessment and/orintervention?

a. Features of thehealth care team:attitude/motivation,professionaldiscipline(s),skills/training;

b. Features of thepractice setting:practice size andpatient makeup,workforce mix,incentives,resources, officesupport systems,materials;

c. Features of thehealth care system:type oforganization,location, populationcharacteristics,density,organizationalcharacteristics/poli-cies, administrativearrangement,decision supporttools, clinical

informationsystems,incentives, marketconditions,communityresources,political/legal/regulatory issues,accreditationissues.

13. What are the largersocial/environmentalinfluences thatdetermine whetherindividuals respond toappropriate behavioralcounselinginterventions andsuccessfully changetargeted healthbehaviors?

Analytic framework 2–Key Questions

*An interventioncondition is a distinctpatient sub-groupidentified through theassessment processthat receives aparticular interventionas part of their clinicalencounter.

**Evidence for eachintervention condition isreviewed in parallel.

through which reductions in morbidity and /ormortality are mediated. In reality, the preferredoutcome measures may not be widely available inthe literature, because behavioral outcomedefinitions often vary widely among studies.Sustained behavior changes potentially affect otheroutcomes of importance to the patient (changes inother behaviors or quality of life) or to the healthcare system (utilization or patient satisfaction) (KQ4), and may also induce adverse effects, such asincreased injury rates in those increasing physicalactivity (KQ 3).

As new epidemiologic evidence becomes available,the behavioral outcomes of interest to reviewers mayalso shift. For tobacco, illicit drug, and alcoholmisuse, abstinence has been the primary treatmentgoal and the most important behavioral outcome.Recently, increased attention has been paid to thehealth benefits from reducing smoking,91 increasingsafe needle use in intravenous drug users,92 andstressing moderation in alcohol use.93 Thus, futurereviews may include interventions addressing suchbehavioral outcomes.

Analytic Framework 2: CanInterventions in the ClinicalSetting Influence People toChange Their Behavior?

Once a behavior change has been clearly relatedepidemiologically to improved health outcomes, themost critical issue for clinicians is knowing whetherinterventions in the clinical setting help patientschange their behavior and, if so, how to deliver themeffectively and practically. Analytic Framework 2(Figure 2) contains the logic and critical questions tosystematically evaluate the evidence forrecommending specific strategies in clinical care topromote healthy behaviors.

Earlier USPSTF experience suggested the needfor studies that develop and validate risk-screeningand intervention-assessment tools and that examinethe efficacy or effectiveness of interventions based onthese assessments.94 Assessment (KQ 3) specifieshow best to identify patients in need of behavioralintervention and to measure quickly any keycharacteristics by which the intervention should be

individualized (KQ 1, 2). Assessment itself mayhave adverse effects, such as anxiety, misdiagnosis, ordistraction from appropriate care, which woulddetract from any overall benefit (KQ 4) (see sidebar,“Physical Activity Interventions”).

The next arrow or link in Analytic Framework 2examines whether clinical setting interventions areeffective in changing behavior (KQ 5) and specifyingfor whom (KQ 7). For behavioral counselinginterventions, no less than for other primary caretreatment regimens, it is critical to knowintervention details97 (KQ 6): What were the keyelements of the intervention, and to whom werethey delivered? How were they delivered—when,where, and by whom? What were the time andintensity of the intervention contact? How often andover what time period was the interventiondelivered? What was the total intervention “dosage”in terms of frequency, intensity and duration? Whatwere the extent and the duration of the treatmenteffect (KQ 8)?

Many successful interventions provide repeatedcontacts and supports that can be modified to fit theindividual path of change undertaken by the patient(KQ 9). The USPSTF also considers other benefits(KQ 10) or potential harms (KQ 11) associated withthe behavior change. Evaluation of intervention


M–59

Physical Activity Interventions

Using physical activity as an example, the majority ofadults may be sedentary, but not all who visit theclinician need an exercise intervention, and there is noway to determine the need for activity counselingwithout a specific assessment. In a recent study, fullyhalf of older adults in community-based medicalpractices who were willing to receive exercisecounseling were already active enough not to needfurther encouragement.95 Activity assessmentsinclude standard questions about the frequency,duration, and intensity of physical activity, as well asmedical factors that would dictate the exercise typeor regimen to prevent harms or complications.Exercise assessment is often individualized further toaddress motives, barriers, and supports for increasingactivity levels. The efficacy of exercise interventionsappears to be enhanced when varied according tofactors such as the patient’s readiness to change,exercise preferences, or past experiences.96

processes as well as content determines the extent,fidelity, and quality of interventionimplementation.47

Finally, the review can consider howcharacteristics of the health care setting influence thelikelihood that appropriate individuals will beidentified and will receive behavioral interventions(KQ 12a-c), and how larger socioculturalenvironmental forces influence individuals’ ability tochange their behavior (KQ 13).98-101 Sinceindividuals are embedded within social, political,and economic systems that shape their behaviors andconstrain their access to resources for change, it isimportant to incorporate these broader factors intoour evaluation of interventions.47

To gain the maximum benefit from interventionsin clinical settings, we need to extend ourperspective beyond efficacy (ie, it works in researchsettings) or even effectiveness (it works in real-worldclinical settings) to consider the degree to whichtested interventions are feasible for adoption intothose real-world clinical settings and sustainable overlong periods of time.31,63 These perspectives arecritical to realizing the public health benefits ofmodest clinical interventions.

Aligning Evidence WithUsefulness in Clinical Settings

Evidence-based analyses help define the mosteffective and efficient interventions for specific riskbehaviors. Unfortunately, the state of the evidencefor behavioral counseling interventions precludes asimple, consistent approach to conducting andreporting the results of these evidence reviews,particularly across a variety of behaviors. Lack ofdetail and inconsistency of terms describingbehavioral interventions in published reportsseriously hamper rigorous reviews and limit thepotential for research replication. Similarly,methodologic approaches to these topics are evolvingas we consider whether and how specialmethodologic considerations apply regardingadequacy of research design or unique threats tointernal and external validity when evaluatingbehavioral counseling interventions. These issues areimportant to understand, particularly given the gapbetween available behavioral research and current

standards of high-level evidence developed for otherfields of medicine.47 However, under the best ofcircumstances, it remains to be seen how far we cango in specifying standardized approaches forclinicians to the variety of patients for a variety ofbehaviors. There may be a limit as to how well wewill ultimately be able to define any standardizedapproach, given the multiplicity of factors (patient,family, community, clinician, and health caresetting) influencing behavioral change, and the rangeof states within each factor. This is an importantarea for ongoing research.

Thus, the current literature, while muchimproved over the past, may still be insufficient tounequivocally define for the clinician what does anddoes not work across all primary care behavioralcounseling interventions. However, given theprevalence and health impact of unhealthybehaviors, clinicians may still use the time andresources readily available to them to reinforce theimportance of healthy behaviors with their patients.For detailed evidence-based consideration ofbehavioral counseling interventions for specificbehaviors, readers are referred to the USPSTFrecommendations (and associated systematicreviews).102

Given the inconsistencies in terms andintervention descriptions in the current behavioralcounseling intervention literature, the USPSTFdecided to use a unifying construct to describe theseinterventions more consistently across a range ofapproaches and behaviors. The USPSTF alsorecognized the need to contribute to thedevelopment of a new conceptual and linguisticsynthesis for health behavioral counselinginterventions in clinical care. Given that no singleempirically validated model captures the broad rangeof interventions across risk behaviors, the USPSTFchose to adopt the 5 A’s construct because it wasjudged to have the highest degree of empiricalsupport for each of its elements and because of itsuse in the existing literature. We describe and thenillustrate the use of this construct in the next sectionof the paper, which also updates the 1996 USPSTFsummary of the range of research-supportedstrategies for clinicians interested in deliveringbehavioral counseling intervention in clinical care.1

M–60


The 5 A’s OrganizationalConstruct for ClinicalCounseling

BackgroundThe 4 A’s construct (ask, advise, assist, arrange)

was originally developed by the National CancerInstitute to guide physician intervention in smokingcessation.103 Recently, the Canadian Task Force onPreventive Health Care proposed that clinicians usea 5 A’s construct (adding an agree step) to organizetheir general approach to assisting patients withbehavioral counseling issues (W.Elford, CanadianTask Force on Preventive Health Care, personalcommunication, December 2000). The U.S. PublicHealth Service12 used the A’s construct to report onhigh-quality, controlled clinical trials in tobaccocessation, many conducted in primary care settingsto test brief, feasible population-level interventions.The A’s construct has also been applied to briefprimary care interventions for a variety of otherbehaviors.70,75,95

To be congruent with the U.S. Public HealthService and Canadian Task Force concepts of the Asconstruct, we adopted the following terminology todescribe minimal contact interventions that areprovided by a variety of clinical staff in primary caresettings:

Assess: Ask about/assess behavioral health risk(s)and factors affecting choice of behaviorchange goals/methods.

Advise: Give clear, specific, and personalizedbehavior change advice, includinginformation about personal healthharms/benefits.

Agree: Collaboratively select appropriatetreatment goals and methods based on thepatient’s interest in and willingness tochange the behavior.

Assist: Using behavior change techniques (self-help and/or counseling), aid the patient inachieving agreed-upon goals by acquiringthe skills, confidence, andsocial/environmental supports for behavior

change, supplemented with adjunctivemedical treatments when appropriate (eg,pharmacotherapy for tobacco dependence,contraceptive drugs/devices).

Arrange: Schedule follow-up contacts (in person orby telephone) to provide ongoingassistance/support and to adjust thetreatment plan as needed, includingreferral to more intensive or specializedtreatment.

Rationale and Strategies forImplementing the 5 A’s

The content of each step in the 5 A’s constructnecessarily varies from behavior to behavior, butclinical intervention targeting any behavior changecan be described with reference to these 5intervention components. While we promote theidea of a unifying construct to describe behavioralcounseling interventions across behaviors, weacknowledge that the type and intensity of behaviorchange strategies needed may vary by the complexityof the change, whether the behavior is being addedor deleted, and by factors individual to the patient,as described in the “Theories and Models ofBehavior Change” section above. Our briefdescription of each “A” of this unifying constructuses selected examples from recent research to detailcurrent options and challenges in providingbehavioral counseling interventions in clinical care.

Assess

Because behavioral risks are largely invisible andare rarely the main reason for seeking clinical care,explicit systems for behavioral risk-factor assessmentin clinical populations serve 2 purposes. First, theyidentify all those in need of some intervention for agiven behavior (eg, sedentary or underactiveindividuals vs already active).96 Second, they gatherdata needed to target (group) those needing differentinterventions and, if warranted, to individualize(tailor) brief interventions for maximumeffectiveness or health benefit.104

Depending on the behavior, groups are targetedfor intervention by factors such as current practices(eg, current tobacco users vs recent quitters),12


M–61

intention (eg, intending to breast-feed vs not),105

readiness to change the behavior (eg, soon vs not),106

and presence of medical/physiological factorsdefining treatment options (eg, pregnant vs not).Within target groups, moderating factors such asage,107 gender,108 ethnicity,109 comorbidity, or healthliteracy110 can help clinicians individualize (tailor)intervention emphasis104 once such tailoring hasbeen proven beneficial. Such assessment forintervention individualization may be delayed to alater point in the A’s process12 (see “Agree” sectionbelow). Assessment can also identifycontraindications to intervention, such as generalpromotion of physical activity in the presence ofrecent morbidity96 or the safety and appropriatenessof nicotine replacement therapy as a behavioraltreatment adjunct.12

Systematic, routine assessment is the foundationfor proactive behavioral counseling interventions,particularly to realize their public health potential.For instance, having a system in place to identifyand document tobacco-use status triples the odds ofclinician intervention.12 Adequate assessment canhelp the clinician consider patient priorities andmedical risks, particularly among those withmultiple behavioral risks.111 Little research currentlyexists in effective methods for prioritizing amongcompeting behavioral risks, but ongoing work by theBehavior Change Consortium, sponsored by theNational Institutes of Health, may help address theseissues112 (see sidebar, “Assesment Strategies”).

Advise

As discussed above, clinician advice establishesbehavioral issues as an important part of health careand enhances the patient’s motivation to change.Such advice is most powerful when personalized byspecifically linking the behavior change to thepatient’s health concerns, past experiences, family, orsocial situation,119 and tempering it with theindividual’s level of health literacy.120 Clinicianadvice primarily gives the cue to action, while other

M–62


Assessment Strategies

Ideal assessment strategies for clinical practicesettings are feasible, brief, and able to be interpretedor scored easily and accurately, and they enhanceintervention appropriateness and effectiveness.113,114

Assessment ranges from a few focused questionsadded before the clinician visit (“Have you usedtobacco products at all in the last 7 days? If yes, areyou seriously thinking about quitting in the next 6months? If no, have you used them in the last 6months?”115) to more comprehensive tools, such ashealth risk appraisal (HRA). An HRA is a multi-pagequestionnaire that collects patient information toidentify risk factors and is typically used to producean individuated feedback report to promote health,

Continued

Assessment Strategies (continued)

sustain function, and prevent disease. HRA feedback,alone or in combination with single-sessioncounseling by a clinician, is generally ineffective inproducing behavior change,111 but the HRA can be alow-cost, easy method to gather data systematicallyabout a variety of modifiable health behaviors andrelated factors.

Challenges for behavioral assessments include thetension between accuracy and feasibility.116 To bepractical, many tools are abbreviated to require aslittle patient and clinician time as possible; thus, goodevaluations consider both accuracy and applicabilityfor any assessment approach. Most behaviorsbesides tobacco use—such as poor diet, physicalinactivity, or risky sex—are complex to assessbecause clinicians need some details of usualpractices, such as the frequency, intensity, andduration of various physical activities96 or “usual”intake of specific food items, both to identify individualcandidates for intervention and to measure theirprogress.116,117 One approach to the demands of amore lengthy assessment is to obtain briefassessment by telephone in advance of the clinicvisit.95,117 This has been shown to produce reasonablyaccurate results, at least for physical activity.118

Assessments rely on self-report and recall ofcustomary behavior, and these can suffer from lapsesin individual memory, errors in estimation, and theimprecise mapping of self-reported activities tomeaningful, physiologically related measures.116

Overall, when reliable biological or biomechanicalmarkers are available for comparison, self-reportedhealth behaviors and risk factors tend tounderestimate the proportion of general-populationindividuals considered “at risk.”113 Accuracy and self-disclosure are enhanced by selecting assessmenttools designed to maximize the accuracy of self-reportinformation.113

health professionals and media provide thedetails.29,56 In this scenario, the clinician is auniquely influential catalyst for patient behaviorchange69 and is best supported by a coordinatedsystem to accomplish and maintain that change.

Feedback from current or previous assessmentscan help personalize health risks and health benefitsas well as enhance motivation for change.59 Well-delivered advice supports the patient’s self-determination.121 Using minor qualifications suchas, “As your physician, I feel I should tell you,” foran advice message, rather than “You should,” is asubtle but powerful way to convey respect for, andavoid undermining, patient autonomy (see sidebar,“Advice Strategies”).

Agree

Here the patient and clinician “come to commonground”51 on area(s) where behavior change is to beconsidered or undertaken. When both agree thatchange is warranted, they then collaborate to definebehavior-change goals or methods. The importanceof collaborative care and patient agreement in acourse of action was not explicit in the original 4 A’smodel, but medical thinking has shifted over recentdecades to greater patient participation in manyaspects of medical care.124 Increasingly, treatmentdecisions are based on clinician-patient agreementafter considering treatment options, consequences,and patient preferences.125 Shared decision-making isspecifically recommended by the USPSTF forpreventive services that involve conflicting or highlyindividualized risk-benefit trade-offs.126 Similarly, acollaborative approach that emphasizes patientchoice and autonomy is critical in behavioralcounseling intervention, where the patient retainsultimate control.

Patient involvement in decision-making aboutbehavior change offers important benefits, evenwhen decisions involving competing risks andbenefits are not the overriding concern. Patientswho are actively involved in health care decisionshave a greater sense of personal control,127 animportant factor for successful behavior change.Also, patient involvement in decisions promoteschoices based on realistic expectations and patientvalues,128 which are important determinants ofpatient adherence or compliance.129 Patient-centeredapproaches in which the patient and clinicianmutually agree on specific changes may require lessvisit time than provider-centered ones130 (see sidebar,“Agreement Strategies”).


M–63

Advice Strategies

Effective clinician advice has several importantelements. Personalized feedback can be biological(laboratory or physiological test results), normative(compared with results for others of the same age,race, and gender), or ipsative (compared with one’sprevious scores). How the clinician’s advice isdelivered matters—a warm, empathetic, and non-judgmental style elicits greater cooperation and lessresistance, particularly for patients not currentlyinterested in change.59,119 A respectful, individualizedapproach first considers patient interest in changebefore warning about health risks or trying to convincethe patient to take action.122 Helpful clinician advicealso emphasizes the clinician’s confidence in thepatient’s ability to change the behavior (building self-efficacy), and reassures the patient that there aremultiple ways to approach successful change andsources to support the behavior change once it isundertaken.119 Acknowledging a patient’s previoussuccess in making changes can also boost thepatient’s confidence. Even considering all theseelements, advice messages can be compactlyconstructed and short (30 to 60 seconds), particularlywhen coupled with additional assistance. Someclinicians are reluctant to advise patients becausepeople seeking clinical care are not consciouslyseeking medical advice about their behavior.However, well-delivered advice is actually associatedwith improved satisfaction among smokers54 and otherpatients with behavioral risk factors.30 Expertsrecommend providing anticipatory advice forpreventing risky sexual activity or tobacco, alcohol,and illicit drug use to all members of specialpopulations, such as adolescents, even before riskybehaviors are evident.123

Agreement Strategies

Additional questions will help frame the rest of theintervention. For example, current tobaccointervention guidelines recommend assessing whetherthe patient is willing to make a quit attempt within thenext 30 days.12 If not, subsequent behavior-changeassistance will consist of a motivational intervention tobolster confidence and readiness and addressenvironmental and other barriers to change. If the


Assist

In providing assistance, the primary care clinicianor other health care staff offers additional treatmentto address barriers to changes, increase the patient’smotivation and self-help skills, and/or help thepatient secure the needed supports for successfulbehavior change. Effective primary careinterventions seek to teach self-management andengage problem-solving/coping skills, therebyenabling the patient to undertake the nextimmediate step(s) in the targeted behavior change.70

Those not ready to commit to making a specificbehavior change in the near future often benefitfrom assistance strategies that explore ambivalenceand enhance motivation.59 As emphasized earlier,additional assistance through effective behavior-

change techniques need not be provided directly bythe primary clinician solely within the context of aprimary care visit. Clinicians may provide assistancethrough referral to other health care staff within theclinic or outside in the larger health care system orcommunity. Importantly, such approaches typicallyinvolve multiple communication channels andintervention methods, which also improveintervention outcomes.12,19

Additional assistance within or outside thepatient visit is likely to produce better outcomesthan minimal-contact, advice-only treatment. Forexample, even though 1-3 minutes of advice andcounseling have been found to double smokers’ 6-month quit rates, time-intensive interventions andmore numerous contacts produce even bettereffects.12 Increasing the total contact time in anintervention (time per intervention X number ofcontacts) from the minimal 1 to 3 minutes to morethan 30 minutes doubles the long-term quit ratesyet again. Similarly, a recent analysis at the U.S.population level estimated the expected ex-smokeryields of increasing the proportion of physicians whoprovide systematic advice (1-3 minutes) to theirsmoking patients from 60% to 90%. That estimatewas compared with also providing additionalcounseling assistance (10 minutes) by the clinicianor other staff for the 50% of advised smokersinterested in quitting.66 The results showed thatincreasing rates of physician advice alone wouldyield an additional 63,000 quitters per year.Coupling the higher advice level with briefcounseling assistance would increase annual quittersby a factor of 10 (630,000) (see sidebar, “AssistanceStrategies”).

M–64


Assistance Strategies

Assistance techniques vary according to the behaviorand the individual patient’s needs but includepractical counseling (problem-solving skills training) toreplace the problem behavior with new behaviors andto tackle environmental and physiological barriers tochange. Assistance also can include direct supportfrom the health care provider/team, guidance inobtaining social support from friends and family, theprovision of self-help materials to support self-change

Agreement Strategies (continued)

patient is ready to take action, then further behavioralcounseling is provided, along with adjunctivemedication or medical devices, if appropriate. Formany behaviors, a few brief questions such as “Howimportant is it for you to…“ or “How confident are youthat you can…” easily assess a person’s motivationand confidence to change a particular behavior, andquickly identify the most promising avenues for furtherassistance.121 This type of open-ended exchange canengage even the minimally interested patient in anonthreatening way that may also increaseknowledge, self-confidence, and motivation.

Actively engaging a patient’s agreement beforeproceeding with further behavioral counseling canalso prevent resistance.121 Agreement considers themultiple treatment or intervention options available tohelp the patient achieve selected behavior changegoals. For instance, patients can select home-basedor fitness center options to increase their activitylevels, nicotine fading or “cold turkey” approaches tosmoking cessation, the use of varied contraceptivemethods and/or abstinence to prevent pregnancy, andthe choice of a wide variety of approaches toimproving diet. Moreover, for each of these changes,patients can often choose between reliance on self-help and more intensive clinic methods, based onpreference and perceived need for the more intensiveskill training and higher levels of social support thatclinic-based and face-to-face counseling provide. Forpeople with multiple behavioral risks, agreement isneeded about which behavior change(s) to tackle first.


Arrange

Arranging follow-up challenges us toreconceptualize behavioral risk factors as chronicproblems that change over time.67 No matter howintensive the initial assistance, some form of routinefollow-up assessment and support through repeatvisits, telephone calls, or other contact is generallydeemed necessary in behavior change interventions.For one thing, follow-up contacts provide theopportunity to evaluate and adjust the behavior-change plan. Usually, this is accomplished by brieflyrepeating the first 4 A,s (assess, advise, agree, assist)to update the behavior-change plan, taking into

account the patient’s intervening efforts, experience,and current perspective. Follow-up allows forsupport of behavior-change maintenance134 andrelapse prevention for those who have already madesome significant behavior change.93,135 In general,follow-up is best scheduled within a relatively shorttime period (eg, 1 month), although the timing canbe geared to provide support for a specific event(such as calling a few days after a set quit-smokingdate). After initial intervention follow-up, futurecontacts are often spaced at successively longerintervals to provide needed support and continuityin a gradually reduced manner (see sidebar,“Arranging Implementation and Follow-up”).


M–65

Arranging Implementation and Follow-up

Behavioral interventions can involve “stepped-care”approaches, similar to those used for hypertensionmanagement, with the need for referral to moreintensive treatment or outside resources determinedafter evaluating response to briefer, less-intensiveinterventions during follow-up.48 Simply notifyingpatients that follow-up will occur seems to be apowerful motivating factor,136 communicating that thebehavior change is important and that follow-upassistance will be available if needed. Clinical staffcan systematically arrange follow-up assessment andsupport through repeat clinical visits, telephone calls,or other methods of contact between the patient andthe health care system. Completion rates for follow-up and outside referral are important implementationprocess measures.

Recent advances in health communications can assistboth clinicians and patients as they engage inappropriate adjustment of the behavior-change plan.For example, interactive computer programs coupledwith the capacity for individually tailored output cantrack individual progress and adjust health promotionstrategies to respond to the individual’s preferences,rate of progress, and changing environments.137 Thediversity of populations that clinicians serve increasesthe importance of adjusting behavior change plans tothe culture, social circumstances, and economicstatus of clients; such adjustment of health behaviorchange plans over time and across changingcircumstances is an area where many healthprofessionals need increased preparation andexpertise.138

Assistance Strategies (continued)

efforts, and the provision or prescription ofappropriate medication or medical devices (eg,pharmacotherapy for tobacco dependence,contraceptives for prevention of unplanned pregnancy,and dietary supplements for certain weight lossregimens). Other effective behavior-changetechniques include modeling and behavioral rehearsal,contingency contracting, stimulus control, stress-management training, and the use of self-monitoringand self-reward.131

Involving a variety of staff and using diverse,complementary intervention methods improve thefeasibility and the effectiveness of providing furtherbehavior change assistance. Interactive videos candeliver standardized portions of behavioral counselinginterventions.24 Telephone counseling and well-developed self-help materials provide additionalchannels for efficiently delivering effectiveinterventions.23,27 If proven effective, computer-driveninterventions will someday offer direct, interactivepersonalized contact through computer kiosks or theInternet that bypasses use of office staff andresources.26 Within certain health care environments,such as managed care and health maintenanceorganizations, staff outside the clinical settingundertake written and telephone counseling that canresult in feedback to the provider or medical chart.68

For settings with few of these options, the delivery ofappropriate behavior change assistance is morefeasible if intervention activities are spread acrossclinical staff (eg, clinician, nurse, medical assistant,and receptionist).29,132,133

ConclusionsBehavioral counseling interventions in clinical

settings are an important means of addressingprevalent health-related behaviors, such as lack ofphysical activity, poor diet, substance (tobacco,alcohol, and illicit drug) use and dependence, andrisky sexual behavior that underlie a substantialproportion of preventable morbidity and mortalityin the United States. Important advances in theways primary care interventions have been packagedhave resulted from the past 2 decades of research.Most importantly, brief interventions designed to fitinto everyday practice have been found to produceclinically meaningful changes in the population for agrowing number of behavioral risk factors.

Future progress will depend on further refinementof the science supporting behavioral counselinginterventions in clinical care through ongoingbehavioral research and further development ofstandards and methods for the reporting andsystematic review of behavioral counselinginterventions. These advances will facilitatesubsequent recommendation development forbehavioral counseling topics. They will alsofacilitate the identification of common, as well asunique, key elements of behavioral counselinginterventions across behaviors and populations and,thus, enhance their practical implementation by realclinicians and real patients in everyday clinicalsettings.

AcknowledgementsThe authors gratefully acknowledge early

manuscript review and suggestions by outsideexperts: Sue Curry, PhD; Russ Glasgow, PhD;Michael Goldstein, MD; and Pat Mullen, DrPH; aswell as the support and critical review by othermembers of the work group (Karen Eden, PhD;Mark Helfand, MD, MPH; Peter Briss, MD; JudithHarris, BSN; Russell Harris, MD, MPH; Al Berg,MD, MPH; Mary Burdick, PhD); the UNC andOregon EPCs (Kathleen Lohr, PhD and GaryMiranda, MA); the USPSTF (Steve Woolf, MD,MPH and Paul Frame, MD); and AHRQ (DavidAtkins, MD, MPH).

References1. U.S. Preventive Services Task Force. Guide to Clinical

Preventive Services. 2nd ed. Washington DC: Officeof Disease Prevention and Health Promotion, 1996.

2. Harris RP, Helfand M, Woolf SH, et al. Currentmethods of the U.S. Preventive Services Task Force.A review of the process. Am J Prev Med.2001;20(suppl 3):21-35.

3. U.S. Department of Health and Human Services.Healthy People 2010 (Conference Edition).Washington, DC: U.S. Government Printing Office;2000.

4. Koop EC. Health promotion and disease preventionin clinical practice. In: Lawrence RS, Woolf SH,Jonas S, eds. Health Promotion and Disease Preventionin Clinical Practice. Baltimore, MD: Williams andWilkins; 1996:vii-ix.

5. McGinnis JM, Foege WH. Actual causes of death inthe United States. JAMA. 1993;270(18):2207-2212.

6. National Health Center for Health Statistics. HealthyPeople 2000 Review. Hyattsville, MD: Public HealthService; 1999.

7. Orleans CT, Ulmer C, Gruman J. The role ofbehavioral factors in achieving national healthoutcomes. In: Boll TJ, Bennet-Johnson S, eds.Handbook of Health Psychology. Washington, DC:American Psychological Association (in press).

8. Ewing GB, Selassie AW, Lopez CH, McCutcheon EP.Self-report of delivery of clinical preventive servicesby U.S. physicians. Comparing specialty, gender, age,setting of practice, and area of practice. Am J PrevMed. 1999;17(1):62-72.

9. Lemley KB, O’Grady ET, Rauckhorst L, Russell DD,Small N. Baseline data on the delivery of clinicalpreventive services provided by nurse practitioners.Nurse Pract. 1994;19(5):57-63.

10. Partnership for Prevention. Why Invest in DiseasePrevention? Washington, DC: Partnership forPrevention; 1999.

11. Ashenden R, Silagy C, Weller D. A systematic reviewof the effectiveness of promoting lifestyle change ingeneral practice. Fam Pract. 1997;14(2):160-176.

12. Fiore MC, Bailey WC, Cohen SJ, et al. TreatingTobacco Use and Dependence. A Clinical PracticeGuideline. Rockville, MD: U.S. Dept. of Health andHuman Services; 2000. AHRQ publication No. 00-0032.

M–66


13. Israel Y, Hollander O, Sanchez-Craig M, et al.Screening for problem drinking and counseling bythe primary care physician-nurse team. Alcohol ClinExp Res. 1996;20(8):1443-1450.

14. Kottke TE, Battista RN, DeFriese GH, Brekke ML.Attributes of successful smoking cessationinterventions in medical practice. A meta-analysis of39 controlled trials. JAMA. 1988;259(19):2883-2889.

15. Fiore MC, Bailey WC, Cohen SJ, et al. SmokingCessation. Clinical Practice Guideline No. 18.Rockville, MD: U.S. Department of Health andHuman Services, Public Health Service, Agency forHealthcare Research and Quality; 1996. AHRQPublication No. 96-0692.

16. WHO Brief Intervention Study Group. A cross-national trial of brief interventions with heavydrinkers. Am J Public Health. 1996;86(7):948-955.

17. Wilk AI, Jensen NM, Havighurst TC. Meta-analysisof randomized control trials addressing briefinterventions in heavy alcohol drinkers. J Gen InternMed. 1997;12(5):274-283.

18. Bien TH, Miller WR, Tonigan JS. Brief interventionsfor alcohol problems: a review. Addiction.1993;88(3):315-335.

19. Mullen PD, Simons-Morton DG, Ramirez G,Frankowski RF, Green LW, Mains DA. A meta-analysis of trials evaluating patient education andcounseling for three groups of preventive healthbehaviors. Patient Educ Couns. 1997;32(3):157-173.

20. Brunner E, White I, Thorogood M, Bristow A, CurleD, Marmot M. Can dietary interventions changediet and cardiovascular risk factors? A meta-analysisof randomized controlled trials. Am J Public Health.1997;87(9):1415-1422.

21. Kahan M, Wilson L, Becker L. Effectiveness ofphysician-based interventions with problem drinkers:a review. CMAJ. 1995;152(6):851-859.

22. Fleming MF, Barry KL, Manwell LB, Johnson K,London R. Brief physician advice for problemalcohol drinkers. A randomized controlled trial incommunity-based primary care practices. JAMA.1997;277(13):1039-1045.

23. Zhu SH. Telephone quitlines for smoking cessation.In: National Cancer Institute. Population Basedsmoking Cessation: Proceedings of a Conference onWhat Works to Influence Cessation in the GeneralPopulation. Smoking and Tobacco ControlMonograph No. 12. Bethesda, MD: U.S.

Department of Health and Human Services,National Institutes of Health, National CancerInstitute; 2000. NIH Pub. No. 00-4892.

24. Stevens VJ, Glasgow RE, Toobert DJ, Karanja N,Smith KS. Randomized trial of a brief dietaryintervention to decrease consumption of fat andincrease consumption of fruit and vegetables. Am JHealth Promotion. 2002;16(3):129-134.

25. Marcus BH, Nigg CR, Riebe D, Forsyth LH.Interactive communication strategies: implicationsfor population-based physical-activity promotion.Am J Prev Med. 2000;19(2):121-126.

26. Fotheringham MJ, Owies D, Leslie E, Owen N.Interactive health communication in preventivemedicine: internet-based strategies in teaching andresearch. Am J Prev Med. 2000;19(2):113-120.

27. Curry SJ. Self-help materials. In: National CancerInstitute. Population Based Smoking Cessation:Proceedings of a Conference on What Works to InfluenceCessation in the General Population. Smoking andTobacco Control Monograph No. 12. Bethesda,MD: U.S. Department of Health and HumanServices, National Institutes of Health, NationalCancer Institute; 2000. NIH Pub. No. 00-4892.

28. Kreuter MW, Strecher VJ, Glassman B. One size doesnot fit all: the case for tailoring print materials. AnnBehav Med.1999; 21(4):276-283.

29. Kottke TE, Edwards BS, Hagen PT. Counseling:implementing our knowledge in a hurried andcomplex world. Am J Prev Med. 1999;17(4):295-298.

30. Kottke TE, Solberg LI, Brekke ML, Cabrera A,Marquez M. Will patient satisfaction set thepreventive services implementation agenda? Am JPrev Med. 1997;13(4):309-316.

31. Vogt TM, Hollis JF, Lichtenstein E, Stevens VJ,Glasgow R, Whitlock E. The medical care systemand prevention: the need for a new paradigm. HMOPract. 1998;12(1):5-13.

32. Wechsler H, Levine S, Idelson RK, Schor EL,Coakley E. The physician’s role in health promotionrevisited—a survey of primary care practitioners. NEngl J Med. 1996;334(15):996-998.

33. Valente CM, Sobal J, Muncie HL, Jr, Levine DM,Antlitz AM. Health promotion: physicians’ beliefs,attitudes, and practices. Am J Prev Med.1986;2(2):82-88.

34. Levine DM. The physician’s role in health-promotionand disease prevention. Bull N Y Acad Med.1987;63(10):950-956.


M–67

35. Podl TR, Goodwin MA, Kikano GE, Stange KC.Direct observation of exercise counseling incommunity family practice. Am J Prev Med.1999;17(3):207-210.

36. U.S. Bureau of the Census. Statistical Abstract of theUnited States: 1998. 118th ed. Washington, DC;1998.

37. Zaza S, Lawrence RS, Mahan CS, et al. Scope andorganization of the Guide to Community PreventiveServices. The Task Force on Community PreventiveServices. Am J Prev Med. 2000; 18(suppl 1):27-34.

38. Nupponen R. What is counseling all about—basicsin the counseling of health-related physical activity.Patient Educ Couns. 1998;33(suppl 1):S61-S67.

39. Elder JP, Ayala GX, Harris S. Theories andintervention approaches to health-behavior change inprimary care. Am J Prev Med. 1999;17(4):275-284.

40. Smedley BD, Syme SL, eds. Promoting health:Intervention strategies from social and behavioralresearch. Institute of Medicine. Washington DC:National Academy Press; 2000.

41. Glanz K, Lewis F, Rimer B. Linking theory, research,and practice. In: Glanz K, Lewis F, Rimer B, eds.Health Behavior and Health Education: Theory,Research and Practice. 2nd ed. San Francisco,CA:Jossey-Bass; 1999.

42. Abrams DB, Emmons KM, Linnan LA. Healthbehavior and health education: The past, present,and future. In: Glanz K, Lewis F, Rimer B, eds.Health Behavior and Health Education: Theory,Research and Practice. 2nd ed. San Francisco,CA:Jossey-Bass; 1999:453-478.

43. Heaney CA, Israil BA. Social networks and socialsupport. In: Glanz K, Lewis F, Rimer B, eds. HealthBehavior and Health Education: Theory, Research andPractice. 2nd ed. San Francisco, CA:Jossey-Bass;1999:179-205.

44. Prochaska JO, DiClemente CC. Stages and processesof self-change of smoking: toward an integrativemodel of change. J Consult Clin Psychol.1983;51(3):390-395.

45. National Cancer Institute. Theory at a Glance: AGuide for Health Promotion Practice. Bethesda, MD:National Institutes of Health, National CancerInstitute; 1995. NIH Publication No. 95-3896.

46. Stokols D, Allen J, Bellingham RL. The socialecology of health promotion: implications forresearch and practice. Am J Health Promot.1996;10(4):247-251.

47. Institute of Medicine. Health and Behavior: TheInterplay of Biological, Behavioral, and SocietalInfluences. Washington, DC: National AcademyPress; 2001.

48. Abrams DB, Orleans CT, Niavra RS, Goldstein MG,Prochaska JO, Velicer WF. Integrating individual andpublic health perspectives for treatment of tobaccodependency in managed care: A combined stepped-care and matching model. Ann Behav Med.1996;10(4):20-304.

49. Orleans CT. Treating nicotine dependence in medicalsettings. In: Orleans CT, Slade J, eds. NicotineAddiction: Principles and Management. New York:Oxford University Press; 1993:148-150.

50. Safran DG, Taira DA, Rogers WH, Kosinski M,Ware JE, Tarlov AR. Linking primary careperformance to outcomes of care. J Fam Pract.1998;47(3):213-220.

51. Stewart MA. Effective physician-patientcommunication and health outcomes: A review.CMAJ. 1995;152(9):1423-1433.

52. Galuska DA, Will JC, Serdula MK, Ford ES. Arehealth care professionals advising obese patients tolose weight? JAMA. 1999;282(16):1576-1578.

53. Hunt JR, Kristal AR, White E, Lynch JC, Fries E.Physician recommendations for dietary change: theirprevalence and impact in a population-based sample.Am J Public Health. 1995;85(5):722-726.

54. Hollis JF, Bills R, Whitlock E, Stevens VJ, MulloolyJ, Lichtenstein E. Implementing tobaccointerventions in the real world of managed care. TobControl. 2000;9(suppl 1):I18-I24.

55. Schauffler HH, Rodriguez T, Milstein A. Healtheducation and patient satisfaction. J Fam Pract.1996;42(1):62-68.

56. Kreuter MW, Chheda SG, Bull FC. How doesphysician advice influence patient behavior? Evidencefor a priming effect. Arch Fam Med. 2000;9(5):426-433.

57. O’Connor PJ, Rush WA, Prochaska JO, Pronk NP,Boyle RG. Professional advice and readiness tochange behavioral risk factors among members of amanaged care organization. Am J Manag Care.2001;7(2):125-130.

58. Jones T, Fowler MC, Hubbard D. Refining a tool tomeasure cues to action in encouraging health-promoting behavior—the CHAQ. Am J HealthPromot. 2000;14(3):170-173, iii.

M–68


59. Emmons KM, Rollnick S. Motivational interviewingin health care settings. Opportunities and limitations.Am J Prev Med. 2001;20(1):68-74.

60. Simpson M, Buckman R, Stewart M, et al. Doctor-patient communication: The Toronto consensusstatement. BMJ. 1991;303(6814):1385-1387.

61. Kristal AR, Curry SJ, Shattuck AL, Feng Z, Li S. Arandomized trial of a tailored, self-help dietaryintervention: the Puget Sound Eating Patterns study.Prev Med. 2000;31(4):380-389.

62. Parish DC. Another indication for screening andearly intervention: problem drinking. JAMA.1997;277(13):1079-1080.

63. Glasgow RE, Vogt TM, Boles SM. Evaluating thepublic health impact of health promotioninterventions: the RE-AIM framework. Am J PublicHealth. 1999;89(9):1322-1327.

64. Prochaska JO. A stage paradigm for integratingclinical and public health approaches to smokingcessation. Addict Behav. 1996;21(6):721-732.

65. Rose G. Sick individuals and sick populations. Int JEpidemiol. 1985;14(1):32-38.

66. Hollis JF. Population impact of clinician efforts toreduce tobacco. In: National Cancer Institute.Population Based Smoking Cessation: Proceedings of aConference on What Works to Influence Cessation in theGeneral Population. Smoking and Tobacco ControlMonograph No. 12. Bethesda, MD: U.S.Department of Health and Human Services,National Institutes of Health, National CancerInstitute; 2000:129-154. NIH Pub. No. 00-4892.

67. Glasgow R, Orleans CT, Wagner EH, Curry SJ,Solberg LI. Does the chronic care model serve also asa template for improving prevention? MilbankQuarterly. 2001;79(4):579-612.

68. Thompson RS, Taplin SH, McAfee TA, MandelsonMT, Smith AE. Primary and secondary preventionservices in clinical practice: Twenty years’ experiencein development, implementation, and evaluation.JAMA. 1995;273(14):1130-1135.

69. Thompson RS. What have HMOs learned aboutclinical prevention services? An examination of theexperience at Group Health Cooperative of PugetSound. Milbank Q. 1996;74(4):469-509.

70. Goldstein MG, DePue J, Kazuira A. Models forprovider-patient interaction: Applications to healthbehavior change. In Shumaker SA, Schon EB,Ockene JK, McBeem W.L., eds. 2nd ed. TheHandbook of Health Behavior Change. New York:

Springer Pub. Co.; 1998:85-113.

71. Orleans CT, George LK, Houpt JL, Brodie KH.Health promotion in primary care: A survey of U.S.family practitioners. Prev Med. 1985;14(5):636-647.

72. American College of Preventive Medicine. 1998National Prevention in Primary Care Study.Washington, DC: American College of PreventiveMedicine; 1998.

73. Timmerman GM, Reifsnider E, Allan JD. Weightmanagement practices among primary care providers.J Am Acad Nurse Pract. 2000;12(4),113-116.

74. Davis D, O’Brien MA, Freemantle N, Wolf FM,Mazmanian P, Taylor-Vaisey A. Impact of formalcontinuing medical education: Do conferences,workshops, rounds, and other traditional continuingeducation activities change physician behavior orhealth care outcomes? JAMA. 1999;282(9):867-874.

75. Ockene JK, Ockene IS, Quirk ME, et al. Physiciantraining for patient-centered nutrition counseling ina lipid intervention trial. Prev Med. 1995;24(6):563-570.

76. Marcus BH, Goldstein MG, Jette A, et al. Trainingphysicians to conduct physical activity counseling.Prev Med. 1997;26(3):382-388.

77. Leininger LS, Finn L, Dickey L, et al. An officesystem for organizing preventive services: A report bythe American Cancer Society Advisory Group onPreventive Health Care Reminder Systems. Arch FamMed. 1996;5(2):108-115.

78. Ockene IS, Hebert JR, Ockene JK, et al. Effect ofphysician-delivered nutrition counseling training andan office-support program on saturated fat intake,weight, and serum lipid measurements in ahyperlipidemic population: Worcester Area Trial forCounseling in Hyperlipidemia (WATCH). ArchIntern Med. 1999;159(7):725-731.

79. Solberg LI, Kottke TE, Brekke ML, Conn SA,Magnan S, Amundson G. The case of the missingclinical preventive services systems. Eff Clin Pract.1998;1(1):33-38.

80. Solberg LI, Kottke TE, Brekke ML. Will primarycare clinics organize themselves to improve thedelivery of preventive services? A randomizedcontrolled trial. Prev Med. 1998;27(4):623-631.

81. Goodwin MA, Zyzanski SJ, Zronek S, et al. Aclinical trial of tailored office systems for preventiveservice delivery: The Study to Enhance Prevention byUnderstanding Practice (STEP-UP). Am J Prev Med.2001; 21(1):20-28.


M–69

82. Agency for Healthcare Research and Quality. PutPrevention into Practice (PPIP). Available at:www.ahrq.gov/clinic/ppipix.htm. Accessed onDecember 19, 2001.

83. Hollis JF, Lichtenstein E, Vogt TM, Stevens VJ,Biglan A. Nurse-assisted counseling for smokers inprimary care. Ann Intern Med. 1993;118(7):521-525.

84. Burns DM. Smoking cessation: Recent indicators ofwhat’s working at a population level. In: NationalCancer Institute. Population Based Smoking Cessation:Proceedings of a Conference on What Works to InfluenceCessation in the General Population. Smoking andTobacco Control Monograph No. 12. Bethesda,MD: U.S. Department of Health and HumanServices, National Institutes of Health, NationalCancer Institute;2000:1-24. NIH Pub. No. 00-4892.

85. Orleans CT, Schoenbach VJ, Wagner EH, et al. Self-help quit smoking interventions: effects of self-helpmaterials, social support instructions, and telephonecounseling. J Consult Clin Psychol. 1991;59(3):439-448.

86. Jeffery RW. Risk behaviors and health: Contrastingindividual and population perspectives. Am Psychol.1989;44(9):1194-1202.

87. Noell J, Glasgow RE. Interactive technologyapplications for behavioral counseling: Issues andopportunities for health care settings. Am J Prev Med.1999; 17(4):269-274.

88. Campbell MK, DeVellis BM, Strecher VJ,Ammerman AS, DeVellis RF, Sandler RS. Improvingdietary behavior: the effectiveness of tailoredmessages in primary care settings. Am J PublicHealth. 1994;84(5):783-787.

89. Skinner CS, Campbell MK, Rimer BK, Curry S,Prochaska JO. How effective is tailored printcommunication? Ann Behav Med. 1999;21(4):290-298.

90. Strecher VJ, Kreuter M, Den Boer DJ, Kobrin S,Hospers HJ, Skinner CS. The effects of computer-tailored smoking cessation messages in familypractice settings. J Fam Pract. 1994;39(3):262-270.

91. Hughes JR. The future of smoking cessation therapyin the United States. Addiction. 1996;91(12):1797-1802.

92. Normand J, Vlahov D, Moses LE, eds. PreventingHIV Transmission: The Role of Sterile Needles andBleach. Washington, DC: National Academy Press;1995.

93. Weingart KR, Manlatt AG. Sustaining change:Helping those who are still using. In: Miller WR,Heather N, eds. Treating Addictive Behaviors. 2nd ed.New York: Plenum Press; 1998:337-351.

94. Atkins D, DiGuiseppi CG. Broadening the evidencebase for evidence-based guidelines: A research agendabased on the work of the U.S. Preventive ServicesTask Force. Am J Prev Med. 1998;14(4):335-344.

95. Pinto BM, Lynn H, Marcus BH, DePue J, GoldsteinMG. Physician-based activity counseling:Intervention effects on mediators of motivationalreadiness for physical activity. Ann Behav Med.2001;23(1):2-10.

96. Eden KB, Orleans CT, Mulrow CD, Pender NJ,Teutsch SM. Counseling to Promote Physical Activity.Systematic Evidence Review. Rockville, MD:Agency for Healthcare Research and Quality; 2001(in process).

97. Altman DG, Schulz KF, Moher D, Egger M,Davidoff F, Elbourne D et al. The revisedCONSORT statement for reporting randomizedtrials: Explanation and elaboration. Ann Intern Med2001;134(8):663-694.

98. House JS, Williams DR. Understanding and reducingsocioeconomic and racial/ethnic disparities in health.In: Institute of Medicine. Promoting Health:Intervention Strategies From Social and BehavioralResearch. Smedley BD, Syme SL, eds. Washington,DC: National Academy Press; 2000:81-124.

99. Fuligni AD, Brooks-Gunn J. The health developmentof young children: SES disparities in health. In:Institute of Medicine. Promoting Health: InterventionStrategies From Social and Behavioral Research.Smedley BD, Syme SL, eds. Washington, DC:National Academy Press; 2000:170-216.

100. Perry CL. Preadolescent and adolescent influenceson health. In: Institute of Medicine. PromotingHealth: Intervention Strategies From Social andBehavioral Research. Smedley BD, Syme SL, es.Washington, DC: National Academy Press;2000:217-253.

101. Emmons KM. Behavioral and social sciencecontributions to the health of adults in the UnitedStates. In: Institute of Medicine. Promoting Health:Intervention Strategies From Social and BehavioralResearch. Smedley BD, Syme SL, eds. Washington,DC: National Academy Press; 2000:254-321.

102. Agency for Healthcare Research and Quality. U.S.Preventive Services Task Force (USPSTF). Availableat: http://www.ahcpr.gov/clinic/uspstfix.htm.Accessed on: 12-19-01.

M–70


103. Glynn TJ, Manley MW. How to Help Your PatientsStop Smoking. A Manual for Physicians. Bethesda,MD: National Cancer Institute; 1989. NIH Pub:89-3064.

104. Rakowski W. The potential variances of tailoring inhealth behavior interventions. Ann Behav Med.1999;21(4):284-289.

105. Sikorski J, Renfrew MJ. Support for breastfeedingmothers (Cochrane Review). In: The CochraneLibrary. Oxford: Update Software; 2001.

106. Calfas KJ, Long BJ, Sallis JF, Wooten WJ, Pratt M,Patrick K. A controlled trial of physician counselingto promote the adoption of physical activity. PrevMed. 1996;25(3):225-233.

107. Rimer BK, Orleans CT, Fleisher L, et al. Doestailoring matter? The impact of a tailored guide onratings and short-term smoking-related outcomes forolder smokers. Health Educ Res. 1994;9(1):69-84.

108. Royce JM, Corbett K, Sorensen G, Ockene J.Gender, social pressure, and smoking cessations: TheCommunity Intervention Trial for SmokingCessation (COMMIT) at baseline. Soc Sci Med.1997;44(3):359-370.

109. King TK, Borrelli B, Black C, Pinto BM, MarcusBH. Minority women and tobacco: implications forsmoking cessation interventions. Ann Behav Med.1997;19(3):301-313.

110. Gazmararian J, Baker D, Williams M, et al. Healthliteracy among Medicare enrollees in a managed careorganization. JAMA. 1999;281(6),545-551.

111. Strecher VJ, Kreuter MW. Health risk appraisal froma behavioral perspective: present and future. In: HyerGC, Peterson KW, Travis JW, Dewey JE, Foerster JJ,Framer EM, eds. Society of Prospective MedicineHandbook of Health Assessment Tools. Pittsburgh, PA:the Society; 1999:75-82.

112. National Institutes of Health. Behavior ChangeConsortium. Available at:www1.od.nih.gov/behaviorchange/index.htm.Accessed on December 19, 2001.

113. Newell SA, Girgis A, Sanson-Fisher RW, SavolainenNJ. The accuracy of self-reported health behaviorsand risk factors relating to cancer and cardiovasculardisease in the general population: A critical review.Am J Prev Med. 1999;17(3):211-229.

114. Ofstead C, Gobran D. Practical guidelines formaximizing the impact of assessment. In: Hyer GC,Peterson KW, Travis JW, Dewey JE, Foerster JJ,

Framer EM, es. Society of Prospective MedicineHandbook of Health Assessment Tools. Pittsburgh, PA:the Society; 1999:155-160.

115. Whitlock EP, Hollis JF. From research to practice:Helping medical systems integrate nurse-assistedtobacco intervention into routine clinical practice.Seattle, WA: Presented to International Conferenceof the American Thoracic Society; 1995.

116. Bond D, Hyner GC. Lifestyle-specific outcomemeasures. In: Hyer GC, Peterson KW, Travis JW,Dewey JE, Foerster JJ, Framer EM, eds. Society ofProspective Medicine Handbook of Health AssessmentTools. Pittsburgh, PA: the Society; 1999:25-32.

117. Beresford SA, Curry SJ, Kristal AR, Lazovich D,Feng Z, Wagner EH. A dietary intervention inprimary care practice: The Eating Patterns Study.Am J Public Health. 1997;87(4):610-616.

118. Ainsworth BE, Bassett DR, Jr, Strath SJ, et al.Comparison of three methods for measuring thetime spent in physical activity. Med Sci Sports Exerc.2000;32(9 suppl):S457-S464.

119. Miller WR, Rollnick S. Motivational Interviewing:Preparing People to Change Addictive Behavior. NewYork: Guilford Press; 1991.

120. Ad Hoc Committee on Health Literacy for theCouncil on Scientific Affairs. Health literacy: Reportof the Council on Scientific Affairs. JAMA.1999;281(6),552-557.

121. Rollnick S, Mason P, Butler C. Health BehaviorChange: A Guide for Practitioners. Edinburgh/NewYork: Churchill Livingstone; 1999.

122. Solberg LI, Kottke TE. Patient perceptions: Animportant contributor to how physicians approachtobacco cessation. Tob Control. 1998;7(4):421-422.

123. American Medical Association. Guidelines forAdolescent Preventive Services (GAPS):Recommendations Monograph. Chicago: AmericanMedical Association; 1997.

124. Little P, Everitt H, Williamson I, et al. Preferences ofpatients for patient centered approach toconsultation in primary care: Observational study.BMJ. 2001;322(7284):468-472.

125. Frosch DL, Kaplan RM. Shared decision making inclinical medicine: Past research and future directions.Am J Prev Med. 1999;17(4):285-294.

126. Woolf SH, Atkins D. The evolving role ofprevention in health care: Contributions of the U.S.


M–71

Preventive Services Task Force. Am J Prev Med.2001;20(suppl 3):13-20.

127. Lerman CE, Brody DS, Caputo GC, Smith DG,Lazaro CG, Wolfson HG. Patients’ PerceivedInvolvement in Care Scale: Relationship to attitudesabout illness and medical care. J Gen Intern Med.1990;5(1):29-33.

128. Miller WR. Enhancing motivation for change. In:Miller WR, Heather N, eds. Treating AddictiveBehaviors. 2nd ed. New York: Plenum Press;1998:121-132.

129. Donovan JL, Blake DR. Patient non-compliance:Deviance or reasoned decision-making? Soc Sci Med.1992;34(5):507-513.

130. Adams A, Ockene JK, Wheller EV, Hurley TG.Alcohol counseling: Physicians will do it. J GenIntern Med. 1998;13(10):692-698.

131. Bandura A. Social Foundations of Thought andAction: A Social Cognitive Theory. Englewood Cliffs,NJ: Prentice Hall; 1986.

132. Dickey LL, Gemson DH, Carney P. Office systeminterventions supporting primary care-based healthbehavior change counseling. Am J Prev Med.1999;17(4):299-308.

133. Glasgow RE, Whitlock EP, Eakin EG, LichtensteinE. A brief smoking cessation intervention for womenin low-income planned parenthood clinics. Am JPublic Health. 2000;90(5):786-789.

134. Orleans CT. Promoting the maintenance of healthbehavior change: Recommendations for the nextgeneration of research and practice. Health Psychol.2000;19(suppl 1):76-83.

135. Donovan DL. Continuing care: Promoting themaintenance of change. In: Miller WR, Heather N,eds. Treating Addictive Behaviors. New York: PlenumPress; 1998:317-336.

136. Lichtenstein E, Glasgow RE. Smoking cessation:What have we learned over the past decade? JConsult Clin Psychol. 1992;60(4):518-527.

137. Krueter J, Farrell D, Olevitch L, Brennan L.Tailoring Health Messages: CustomizingCommunication With Computer Technology.Mahwah, NJ: Erlbaum; 2000.

138. Pender N. Health Promotion in Nursing Practice.Stamford, CT: Appleton and Lange; 1996.

M–72


AHRQ Pub. No. 03-518May 2002

Evaluating Primary Care Behavioral Counseling ...

Documents

Transcript of Evaluating Primary Care Behavioral Counseling ...