Despite the clear health benefits that can be accrued through adopting a more active lifestyle (10,14,33,43), 60% of U.S. adults are not sufficiently active to achieve such benefits (43), and approximately 25% of adult Americans are essentially totally sedentary (43). There is a need for interventions to promote physical activity that can be widely implemented, especially among the inactive majority of the population. Evidence is accumulating that well-designed behaviorally based physical activity interventions can be effective (8,21,22).
Physicians and other health care providers can play an important role in reaching sedentary individuals and encouraging increases in their physical activity levels. Healthy People 2000 includes an objective on physician counseling for physical activity (42), and the U.S. Preventive Services Task Force recommends that physicians counsel healthy patients on physical activity (15). There is substantial interest in physical activity counseling among physicians. For example, in one study, 91% of physicians reported that they encourage their patients to engage in regular exercise (45), and in a study of family practice physicians, 39% reported counseling most of their inactive patients (34). Other data, however, indicate that these may be overestimates of physician counseling. Only 15% of sedentary adults who had visited a physician for a routine checkup in the past year reported being advised to exercise more (11). Similarly, a chart audit of physical examination visits to family physician residents indicated that physical activity counseling was documented in only 7% of visits, even after an educational intervention aimed at these physicians (25). The barriers to physical activity counseling in primary care are well known and include lack of time, reimbursement, and training in physical activity or behavior change counseling (36). Despite these barriers, it is important to develop physical activity interventions for primary health care settings because patients report a willingness to act on their doctor's advice (18,41,44,45). On average, U.S. adults see a physician or other health care provider two to four times a year (32), providing opportune moments for such counseling.
Physical activity counseling in primary care has been evaluated in five studies with adults. Three short-term studies (7,23,26) showed that physician advice or counseling was associated with increased patient physical activity levels. Two longer-term studies were multiple risk factor programs (13,24) that produced discrepant results. An intensive intervention (24) that featured physician visits specifically devoted to counseling regarding many facets of a health-promoting lifestyle and that included prompts and reimbursement led to 1-yr increases in physical activity. An intervention that relied on video and printed materials rather than direct interactions with a provider was not effective over either 4 or 12 months (13). It is possible that direct advice and counseling from a health care professional is more salient to patients because of perceived physician credibility and authority. These previous studies indicate that physical activity interventions in primary care can be effective, but they raise the possibility that more intensive interventions are needed for long-term effects. The relation of the intensity of an intervention to anticipated outcomes is important, because the cost of the program is likely to affect its potential for eventual widespread adoption. The present study was designed to investigate the amount of intervention in health care settings needed to produce long-term changes in physical activity while remaining feasible for delivery by a health care team. Use of health professionals in addition to the physician was deemed essential in light of the number of barriers impeding intensive physician counseling in this area (36). This trial will demonstrate whether such interventions can be successful in achieving long-term (i.e., 2-yr) increases in physical activity behavior and cardiorespiratory fitness and will examine the intensity of the intervention required to achieve change. The ultimate goal is to develop successful strategies that can be implemented in medical delivery systems throughout the U.S. The purpose of this article is to describe the theoretical background of the ACT intervention, the intervention methods, and intervention training and quality control procedures.
OVERVIEW OF THE ACTIVITY COUNSELING TRIAL (ACT)
The specific goal of ACT is to develop and evaluate two models of primary care practice-based physical activity counseling, differing in intensity, to determine their effectiveness in increasing and maintaining physical activity and cardiorespiratory fitness in sedentary participants. These two models are being compared against a standard care control condition. The study is a collaborative effort among clinical centers in California, Tennessee, and Texas; a data coordinating center in North Carolina; and the study sponsor, the National Heart, Lung, and Blood Institute. The study outcomes, eligibility criteria, recruitment strategies, and measurement schema are presented in the Blair et al. companion article (5) and will be summarized briefly here.
Self-report of physical activity level (caloric expenditure), as measured by the interviewer-administered 7-d Physical Activity Recall (PAR) (6,39), and cardiorespiratory fitness (maximal oxygen uptake) are the two primary outcomes for the trial. The need to maintain increases in physical activity to achieve health benefits has been documented, so the relatively long 24-month follow-up has been selected for the primary outcomes. In addition, the primary objectives will be addressed separately for the two genders because it is possible that intervention effects may differ in men and women (22). The study has been designed with sufficient power to evaluate gender-specific effects.
Secondary outcomes include short-term (i.e., 6-month) intervention effects on physical activity level and cardiorespiratory fitness; the effects on risk factors for cardiovascular disease risk (HDL-cholesterol, LDL-cholesterol, total cholesterol, triglycerides, blood pressure, plasma insulin, fibrinogen, body composition, dietary fat intake, smoking status, and heart rate variability) at 6 and 24 months; the effects on psychosocial factors (mood, self-efficacy, quality of life) at 6 and 24 months; and the relative cost-effectiveness of the two regimens after 24 months. In addition, all participants will be asked to report on the types of physical activities that they have been undertaking at 12 and 24 months.
A total of 874 sedentary adults without serious health problems, aged 35-75, have been recruited from primary care practices collaborating with the three clinical centers and randomly assigned to one of three hierarchical experimental conditions for 24 months: A: Standard Care Control; B: Staff-Assistance Intervention; or C: Staff-Counseling Intervention. In all three conditions, participants receive the same physician advice to increase their physical activity and are provided with current recommendations for physical activity (the CDC/ACSM recommendation of accumulating 30 min of moderate physical activity on most days of the week and the ACSM prescription for cardiorespiratory endurance; 2,35). In interventions B and C, behavior change strategies are provided to help participants increase their physical activity to meet these recommendations. The major difference between the B and C interventions is the type and frequency of contact. The three conditions represent a hierarchy, with all components in A included in B, and all those in B included in C. Thus, the conditions differ in amount of resources and personnel time needed. The interventions are delivered by centrally trained health educators (three at each clinical center) throughout the 24-month intervention period. Study participants and ACT-trained physicians are drawn from four medical practices in northern California, two medical practices in Dallas, and two medical practices in Memphis.
DESCRIPTION OF INTERVENTION AND CONTROL CONDITIONS
Participant Physical Activity Goals
Participants in all study conditions are given the same recommendations for physical activity by their regular primary care physician. Current national recommendations for physical activity form the basis of the ACT physical activity behavioral goals, which also are consistent with the long-term study goal of increasing total caloric expenditure by at least 2 kcal·kg−1·d−1.
There are different ways to meet the goal of increasing total daily energy expenditure by ≥2 kcal·kg−1·d−1, including a "moderate activity" approach and a "vigorous exercise" approach (or a combination of the two). The moderate activity approach is described by the recent CDC/ACSM recommendation on physical activity (35, p. 404): "Every American adult should accumulate 30 minutes of moderate intensity physical activity over the course of most, preferably all, days of the week." This approach also is consistent with recommendations in the NIH consensus report and the Surgeon General's Report on physical activity and health (33,43). Moderate intensity activity is perhaps best described as brisk walking (3-4 mph). This approach encourages fitting more physical activity into the daily routine by taking walks, climbing stairs, and engaging in active pursuits at home, on the job, and in leisure time. Multiple episodes of moderate intensity physical activity can be accumulated in sessions of at least 10 min to reach the total of 30 min a day.
The vigorous exercise approach typically involves more formal exercise modes and is described as various combinations of frequency, intensity, and duration of exercise sessions (2). An increase in total daily energy expenditure of at least 2 kcal·kg−1·d−1 can be accomplished by adding exercise 3 d·w−1 for 30-45 min per session at an intensity of 60-70% of maximal oxygen uptake (equivalent to approximately 70-80% of maximal heart rate). The vigorous exercise goal can be met by any endurance activity meeting the prescribed intensity, for example, jogging, swimming, or class aerobics.
The study goal of at least 2 kcal·kg−1·d−1 is consistent with the amount of activity that will produce significant increases in physical fitness and substantial health benefits. Because additional expenditure of energy via physical activity beyond the 2 kcal·kg−1·d−1 goal has been shown to produce added benefits to health and function (35,43), participants who are willing and able will be encouraged to increase their activity level beyond this basic goal. This further increase can be achieved most realistically by adding some vigorous exercise to a moderate activity approach, or by increasing the frequency, duration, or both for participants following the vigorous exercise approach. Although occupational activity is not an intervention target per se, incorporating methods for increasing routine forms of physical activity during the course of the work day (e.g., through stair climbing and walking) is included as part of the intervention message.
Theoretical Background for the ACT Interventions
The ACT interventions are drawn from several psychological models and theories of health behavior change that have been empirically tested. They have been simplified to ensure realistic interventions that can be utilized by a health care team. The primary theoretical basis for the ACT interventions and conceptualization of the intervention-related mediating factors is social cognitive theory (3). Social cognitive theory draws from the fields of operant learning, social psychology, and cognitive psychology. The theory describes a complex web of determinants in which behavior reciprocally influences, and is influenced by, factors within the person, variables associated with the behavior being targeted for change, and factors in the social and physical environments.
The intervention strategy is to alter key mediators of physical activity, because changing these constructs is expected to produce changes in physical activity. Important mediators have been selected on the basis of previous research on exercise adherence and determinants as well as social cognitive theory (9). Targeted personal factors include cognitions, emotions, and self-regulatory behavioral skills (e.g., self-monitoring, goal-setting). Self-efficacy, in this case perceptions of one's confidence to be physically active in multiple situations, has been shown to be an important predictor of physical activity (30,31). Therefore, the interventions are designed to enhance self-efficacy through promoting a series of successful experiences in meeting realistic physical activity goals. Other documented personal factors include enjoyment of physical activity, perceived benefits of and barriers to physical activity, and skills related to overcoming barriers to participation (9). The ACT interventions allow participants to select enjoyable and practical activities, assist them in developing solutions to barriers, and guide participants in identifying personal benefits of an active lifestyle. There is empirical support for the efficacy of the various intervention components used in ACT (8,9,22).
Targeted social environmental mediators include modeling of physical activity by others and social support directly related to physical activity participation (9). Participants are initially shown an educational video that models targeted behavior and are assisted in making plans to increase social support for physical activity. Though it is consistent with social cognitive theory that access to appropriate activity facilities (e.g., health clubs), resources. (e.g., safe place for walking or jogging), and programs (e.g., affordable aerobics classes) promote regular physical activity (40), compelling data are lacking. Nevertheless, the interventions take into consideration the probable role of environmental influences on physical activity.
The self-regulation model of behavior change is derived from social cognitive theory and has been applied to many behaviors (17). Teaching the participants to apply self-monitoring, self-evaluation, and self-reinforcement through goal-setting, positive self-talk, and problem-solving is expected to enhance their abilities to integrate physical activity into their daily lives.
The transtheoretical model (37) specifies appropriate interventions for an individual moving through different stages of motivational readiness for change (e.g., precontemplation, contemplation, preparation, action, maintenance). The stage-of-change concept is used to guide assessment of motivational readiness to adopt a new behavior and to select the appropriate intervention techniques for each stage. For example, for contemplators, physical activity costs and benefits are evaluated. For those in action, the focus is shifted to behavioral strategies such as relapse prevention. This model has been successfully applied to physical activity assessment and intervention (28).
A hierarchical or additive relationship was utilized in the development of the three study conditions. By building on the strategies in one condition and adding on potentially effective strategies in each remaining condition, a determination of the relative amount of resources necessary for an effective intervention is possible. A Standard Care intervention (condition A) was used for the control condition to provide a minimal level of physician advice consistent with national recommendations for physician counseling. This method also provides a means whereby participating study physicians could be kept blinded to participant condition assignment, and it standardizes the physician advice. The Staff Assistance intervention (condition B) minimizes staff time and resources by employing empirically proven behavioral strategies delivered primarily through the mail following one interpersonal counseling session. In several previous studies, effective multicomponent interventions have been delivered directly by staff using either face-to-face or telephone-based individual counseling strategies (19,21), and this model serves as the basis for the additional components in the Staff Counseling intervention (condition C).
Implementing both B and C interventions in the context of an ongoing physician-patient relationship is expected to augment the potency of both approaches. The relation between the two interventions and the Standard Care Control condition is shown in Table 1.
As noted earlier, there has been a deliberate effort to use theories as a guide to the design of the interventions. The ACT interventions have components that are intended to alter the personal, social, and physical environment variables that are believed to mediate change in physical activity. Table 2 illustrates some of the linkages between theory and intervention components. Most of the mediator variables are targeted in several intervention components. Because the newsletters and telephone counseling are the central components of conditions B and C, respectively, they are designed to address most of the mediators. Not all mediators are addressed at each contact. Through periodic assessments, both the B and C interventions can provide assistance with changing the mediator that is most relevant at that time. The stage of change model is used to guide the tailoring of interventions.
Condition A: Standard Care. Participants in the Standard Care control condition receive physician advice consistent with national recommendations for physical activity counseling by physicians and other health care providers (42), but they do not receive behavioral counseling. For participants in this as well as in the other two conditions, physicians assess and recommended physical activity and refer the patient to the study health educator in a three-step process (27). The physician:
- assesses the participant's current level of physical activity with a simple self-report tool developed by ACT;
- compares the participant's current physical activity level with ACT physical activity goals and encourages the participant to increase activity level, recommends either the moderate activity or vigorous exercise approach, and provides examples of activities; and
- refers the participant to an on-site ACT health educator for further information and tells the participant that this information constitutes an important part of medical care.
The above process is repeated at each nonacute care visit to an ACT physician for the duration of the intervention period of the study. Physicians are blinded to each patient's intervention condition. ACT health educators work with each physician's office staff to ensure that ACT stickers are appended to the patient's chart, which cue the physician to deliver the ACT message.
There are many benefits of this simple protocol. It can be completed in a short time (estimated 2-4 min) and provides physical activity guidelines to all participants, including recommendations tailored to the participant's health needs and risks. The process assigns a role for physicians with which they are comfortable (conducting risk assessments, giving specific recommendations, and referring) and that the participants value. In addition, this approach is highly generalizable to most or all primary care settings. The physicians receive formal training in implementing the protocol, which they will likely perceive as improving their practices.
The physical activity recommendations are recorded in the medical chart so that the physician can review them at subsequent visits and the behavior change programs in the other two conditions can build on them. No additional physician visits are scheduled specifically for physical activity advice. The number of follow-up visits will thus be allowed to vary naturally across patients but, in light of the randomized design, is expected to be similar across all three study conditions. We expect that the average number of physician visits undertaken by ACT participants will be similar to the national average of two to four visits per year (32). The number of physician visits is monitored as part of the intervention. In subsequent clinic visits for routine care, the physician will ask about progress in physical activity. If the patient reports increasing activity, the physician will provide reinforcement and encouragement. If the patient reports no increase, the physician will restate the recommendations.
In the Standard. Care condition, the health educator provides the participant existing written materials on physical activity guidelines developed by national organizations (i.e., the American Heart Association; the National Heart, Lung, & Blood Institute). The health educator invites the participant to call if questions arise about the participant's physical activity program, at which time the health educator will answer questions about type and amount of physical activity and health benefits. Questions about behavior change methods are not answered; rather, the participant is referred to the written materials provided.
Condition B: Staff Assistance intervention. The Staff Assistance intervention builds on the physician advice and materials of the Standard Care condition. A guiding principle in developing this intervention is for it to be feasible for delivery in a wide variety of primary care settings without requiring extensive staff support. The intervention relies primarily on interactive mail to extend clinic staff efforts yet provides tailoring to individual needs and continued contact to promote maintenance.
At the initial session after referral by the physician during the same clinic visit, the health educator shows a 17-min video, developed by ACT. The video provides background on the importance of physical activity, presents physical activity goals, and provides role modeling for setting realistic goals. The health educator then reviews the physical activity goals as presented by the physician, reiterates the physician's recommendations, and individualizes the importance of health and quality of life benefits related to physical activity. In an interactive, individualized manner, different options for meeting physical activity goals based on the participant's abilities, needs, and preferences are discussed, and a short-term activity plan is developed. A community resource guide (developed at each ACT site) is consulted to inform the participant about the options for classes and other programs available in the community, and to facilitate matching community programs to participants.
An interactive mail-based component, consisting of newsletters and mail-back cards, is implemented that is designed for low-literate patients and uses simple concepts. The 24 newsletters are written to increase cognitive and behavioral skills of individuals in various stages of motivational readiness. The newsletters include site-specific inserts to highlight upcoming local physical activity events or physical activity-related organizations. Each newsletter is accompanied by a postage-paid mail-back card. The card provides a place to report the minutes and type(s) of activity engaged in over the past week, an indication of whether the participant is ready to increase activity in the near future, and a checklist of possible barriers to participation. Feed-back sheets that provide specific information and advice are sent by the health educator based on the participant's responses to questions concerning barriers to adherence.
To encourage self-monitoring, participants are given an electronic step-counter (Yamax, Digi-Walker, Model SW-200) and a large magnetic, erasable monthly calendar, which they are instructed to attach to their refrigerator. The Yamax step-counter is a small, lightweight device that is clipped onto the belt or waistband and can be worn during waking hours. The Yamax Digi-Walker has been found to measure steps at a variety of speeds reliably and accurately (4). Participants are instructed to write down their minutes of activity and the number of steps from the step-counter on the monthly calendar, which is then used to complete the physical activity self-monitoring logs (i.e., mail-back cards).
A relatively low level of telephone contact is allowed in the Staff-Assistance intervention. One staff-initiated call is made 1 wk after the initial visit to assess success in meeting the first goal, set subsequent goals, provide social support, and problem-solve. The call is designed to last 5-10 min. Participants are encouraged to call the health educator to ask questions or to request assistance with problem-solving.
Participants who do not make regular mail contact with the project (i.e., return less than 70% of their mail-back cards during the previous 6 months) are called by the health educator once every 6 months. The primary purpose of these telephone calls is to stimulate involvement in the mail intervention program. Reasons for lack of involvement are assessed, participant problem-solving is encouraged, and involvement in at least some intervention components is planned.
A simple and low-cost incentive system is used to promote continued involvement with the mail intervention. The incentives include inexpensive items (e.g., key holders, sports bottles) based on the accumulation of points for returning the mail-back cards. The incentive system is not linked to physical activity behavior itself.
Brief, structured counseling sessions with the health educator are conducted in conjunction with subsequent nonacute naturally occurring physician visits, which last approximately 10-15 min. Specific visits for physical activity counseling do not occur.
Condition C: Staff-Counseling intervention. The Staff-Counseling intervention represents a more staff-intensive intervention approach for promoting physical activity over a 2-yr period in the primary care setting. Participants assigned to this condition receive the physician assessment and recommendation, as described for the control condition, as well as all of the components of the Staff Assistance intervention, plus additional components that provide substantially more interpersonal contact and counseling. The additional components of this intervention include on-going telephone counseling, additional in-person counseling sessions, and behavior change classes. The participant is apprised of these additional components at the initial session.
The telephone counseling component is designed to provide frequent, individualized counseling. Approximately 15 health educator-initiated telephone counseling sessions occur during the first year. Information provided by the participant on the mail-back card is used in the counseling. The counseling is used to evaluate success in meeting physical activity goals, update physical activity goals, problem-solve around barriers to adherence, discuss future barriers and plans to effectively cope with them (i.e., relapse prevention), and provide reinforcement and social support. Feed-back sheets and other materials are mailed out after the telephone counseling session, as appropriate. The basic schedule of telephone contacts is flexible to meet the needs of individual participants (e.g., somewhat more frequent if adherence is low or marginal). The schedule for telephone contacts is once weekly for the initial two wk, biweekly for the following 6-wk period, and once a month for the remainder of the first year.
At the beginning of the second year of counseling, the health educator and participant jointly decide how frequent subsequent contacts will be. The recommended frequency of contacts is monthly for participants with adequate adherence, potentially bimonthly for outstanding adherers who prefer somewhat less contact and more frequent than once a month for those participants for whom it is deemed necessary.
Additional in-person counseling sessions are offered as necessary throughout the intervention period, particularly during the first few months, to meet individual participant needs. The sessions can be recommended by the health educator or requested by the participant. The content of such sessions is similar to that included in the telephone counseling sessions, i.e., problem-solving around barriers to adherence, discussion of future barriers and plans to effectively cope with them, and provision of reinforcement and social support. Occasional home visits by the health educator are optional and are considered on an individual basis depending on the participant's needs, barriers, and preferences, and availability of staff and resources.
Behavior change classes occur throughout the intervention period at least once each week at each clinical center. The classes have primarily a behavioral skill building focus, address adoption and maintenance issues, and facilitate group support. Although the classes are optional, participants are strongly encouraged to attend. The classes are designed to meet the needs of participants at different stages of motivational readiness for change and focus on behavior change strategies such as goal setting, social support, reinforcement, and problem-solving. Group leaders are trained to be sensitive to providing extra encouragement and assistance to those thinking about change or beginning to make small changes and to shape their activity goals in small increments. The topics presented in each class vary, so a participant can attend over a long period and not feel the classes are repetitive. The application of behavior change methods and group problem-solving facilitate relapse prevention for longer-term participants. Supervised physical activity is not offered because the focus is on a behavioral intervention model. To encourage class attendance, regular reminders are given via telephone or mail, and attendees receive tokens toward inexpensive prizes (e.g., discount coupons to local restaurants or retail stores).
Staff Adherence to Intervention Protocols
Adherence to the study protocol by ACT intervention staff is crucial to the integrity of the ACT study. Numerous methods for monitoring and promoting adherence and providing feedback are employed. Methods to enhance protocol adherence include the use of standardized intervention materials that have been centrally developed; structured guides for patient contacts that include checklists of items to be covered; standardized training of intervention staff; and detailed training manuals given to health educators that describe procedures and highlight common problems and their solutions.
In addition, a computerized tracking system via lap-top computers is used to enable measurement of various aspects of protocol adherence. The primary purpose of the tracking system, which was developed by personnel at the ACT Coordinating Center with input from the ACT Intervention Subcommittee, is to ensure that the intervention protocol is being followed. The tracking system was designed to document the actions of key ACT professionals (i.e., the health educators and ACT physicians); prompt specific actions by the health educators (e.g., telephone contacts for condition C participants; monthly newsletters for condition B and C participants); and generate reports approximately quarterly to decision-makers involved with the ACT intervention to facilitate quality control. At each participant's initial randomization/counseling visit with the health educator, the health educator records in the tracking system the tasks and activities that are covered as well as the time spent with the participant. All subsequent contacts and interactions with the participant are tracked in a similar manner for all participants.
Participant Adherence to Intervention
Participant adherence to each component of the intervention is expected to be related to success in changing behavior, so steps are taken to enhance level of patient participation. Methods to enhance participation include presentation of the intervention under the auspices of the participant's primary care provider; frequent contact; reinforcement for completion of mail-back cards and other intervention-specific materials through an incentive system; and newsletters, which contain "hidden clues" that earn the participant additional points toward incentives if they are reported on the participant's mail-back card.
ACT provides intervention training for physicians, primary care clinic staff, and health educators. For physicians and clinic staff, the "training the trainers" approach is used, in which central training has been provided to "master trainers" from each clinical center who then return to their sites to train others. All of the health educators have been trained centrally. Training emphasizes standardization of delivery of the various phases of the intervention and consistent collection of process measures. Training manuals are used and retained for future reference by ACT staff members.
Training methods are based on steps shown to be effective in a wide variety of applications (1,27). Counseling methods are demonstrated by the trainers in a role-play context. Role-plays are conducted in small groups, so each learner has ample opportunity to practice under supervision and receive feedback.
Physician training. Interested primary care physicians who work in a practice targeted by an ACT Clinical Center were trained during a 1-h session at their practice setting conducted by an experienced ACT physician trainer. The ACT physician trainer serves as a liaison between the physicians and his/her ACT clinical center throughout the study. The aim of the training was to promote standardization of the protocol and physician counseling behaviors among participating physicians. Viewing of videotapes showing appropriate initial and follow-up ACT physician visits was used as part of the standardization training. Ongoing evaluation of physician performance is accomplished through tracking system feedback reports that monitor the extent to which participants receive the initial physician counseling (indicated via a physician-signed physical activity assessment and counseling form and participant reports of physician counseling in response to health educator queries) as well as follow-up visit advice (indicated via participant reports of physician counseling in response to health educator queries). Based on these tracking system reports, the ACT physician trainer at each clinical center initiates physician retraining as indicated. In addition, physicians answer questions concerning their counseling on surveys administered following the end of the recruitment period and at the end of the study.
Clinic staff training. Primary care clinic staff (i.e., receptionists, nurses, other clinic staff) are trained to implement procedures for participant flow that will assure appropriate notification of the health educator of appointments and arrival of ACT participants. They also provide ACT health educators with patient charts so that the health educator can flag charts to prompt the physician to deliver advice. The office staff have received training through scheduled meetings with the ACT physician trainer and health educators and through informal discussions with the health educators.
Health educator training. A minimum of two full-time health educators for each clinical center are employed to deliver the three interventions. All health educators have a minimum of a bachelor's level degree in health education or psychology. A 2.5-d central health educator training seminar was delivered approximately 1 month prior to the randomization of the first patient. All health educators attended a second central training session approximately 1 year later.
All ACT health educators were trained to deliver all three interventions (A, B, and C). The training included behavioral counseling approaches and skills, using the specific ACT intervention materials, and using the computerized tracking system to document delivery of intervention components and time spent in intervention. In addition, the training included methods to maintain a clear distinction between intervention components delivered to participants in the three conditions, to avoid behavioral counseling for condition A, and to deal with difficult cases.
Documentation of intervention delivery and the extent to which the interventions are received by patients serves many purposes, especially when the interventions are complex, as in ACT. Process data are used to document the quantity of intervention components delivered, the extent to which the intervention protocol is followed, barriers to full implementation, the quality of the interventions, participant evaluations of each intervention component, and physician and health educator satisfaction with the intervention components. Three types of process measures are being collected: intervention quantity, intervention quality, and participant and physician evaluation.
The purpose of intervention quantity measurement is to document the extent to which the participants are exposed to different intervention components. This information can also be used to evaluate the relation between quantity of intervention received and participant outcomes. The computerized tracking system is being used to collect measures related to intervention quantity, including delivery of advice by physicians, duration and number of contacts with the health educator, attendance at classes, and duration and number of telephone counseling sessions.
The computerized tracking system is also being used to assess whether the intervention procedures are implemented in the manner in which they were planned, i.e., intervention quality. This includes tracking the timeliness with which different components of the intervention are delivered (e.g., newsletter mailings) and received (e.g., return of participant mail-back cards). In addition, information is collected on quality by audio tape-based ratings of the content and quality of health educator counseling sessions judged by three senior ACT behavioral scientists not directly involved in intervention delivery, as well as by periodic on-site observation and evaluation of health educator intervention delivery by a behavioral science expert from the ACT Coordinating Center.
The purpose of participant evaluation is to obtain the participants' subjective evaluations of each intervention component. This evaluation is particularly important in multicomponent interventions such as ACT and will be collected at the end of the study. Data obtained from participant ratings can be used to predict outcomes and to inform modifications to the program for later dissemination. In addition, suggestions and feedback written by participants on condition B and C mail-back cards as well as noted during the telephone contacts received by condition C participants are discussed during the regular ACT Intervention Subcommittee conference calls (see below).
The purpose of ACT physician evaluation is to obtain subjective evaluations of the physician-based components of the intervention from participating ACT physicians. This type of evaluation is being collected at the end of the recruitment period as well as at the end of the study.
Intervention Oversight Committee
To ensure the timely development and implementation of the ACT interventions, the ACT Intervention Subcommittee was formed at the initiation of the trial. Consisting of at least two representatives from the three clinical centers, the co-ordinating center, and the NHLBI project office, the intervention subcommittee meets regularly to ensure the quality and adequate delivery of the interventions at all clinical sites. The subcommittee has conducted the training sessions described earlier and has initiated the other quality control measures. The subcommittee reviews a quarterly summary of intervention tracking system data to assure intervention consistency with protocol and identify areas requiring feed-back. Telephone conference calls among senior members of the subcommittee as well as the health educators themselves are conducted regularly (i.e., at least monthly) to discuss issues and solve problems. These conference calls serve as a means for maintaining high levels of quality and consistency across all centers.
The Activity Counseling Trial was designed to address important questions related to physical activity interventions in primary care. The ultimate goal of the project is to contribute to improvements in public health, so the design of the study was guided by competing concerns about the efficacy and practicality of the interventions. It is assumed that both efficacy and generalizability of the interventions are affected, in opposite ways, by the intensity of the intervention. The hierarchical design of the experimental conditions allows for the intensity of the interventions, reflected by the number of components and amount of staff time required, to be compared with physical activity and fitness outcomes.
The Standard Care condition is similar to national recommendations for preventive interventions in primary care (15), but it represents a substantial improvement over current practice. The type and amount of physician involvement in this condition is considered an appropriate level of involvement for physicians, although some physicians may want to be more personally involved in counseling. The ACT study tests whether additional intervention components in primary care that go well beyond the involvement of the physician are effective in improving outcomes.
The Staff Assistance condition is designed as a feasible, relatively low-cost interactive mail intervention that can be implemented in a wide range of clinical settings. The advantages of this intervention are the relatively low cost, ease of implementation, ability to individualize mailed materials to some extent, and capacity for long-term contact. Disadvantages include lack of relevance for illiterate participants, the possibility that participants will not read or use materials, potential dissatisfaction with lack of personal contact, general low intensity of the program, and lack of direct evidence that this type of intervention can be effective in increasing long-term physical activity. Some changes in clinic practices would be needed to implement the Staff Assistance program, but they would not be major. A nurse or allied health professional could be trained to deliver the initial counseling session. The interactive mail system that is the core of the intervention could be managed by either an assigned member of the clinic staff or through an external organization. The cost of implementing the Staff Assistance program over 2 yr is expected to be relatively modest, and the generalizability of the program is expected to be high.
The Staff Counseling intervention is designed to be a near-optimal, yet feasible, program designed to maximize treatment efficacy. The advantages of adding behavior change classes and frequent individual telephone counseling include the increased intensity of intervention, active out-reach to all participants, multiple options for participating in the intervention, increased ability to tailor the intervention to individual needs, and previous demonstration of efficacy of similar approaches in other settings (20). The primary disadvantage is that the cost and complexity of the program may make the Staff Counseling intervention difficult to generalize to some primary care settings. Other disadvantages include the training and supervision needed for health educators, and the fact that the intensity of the program may be burdensome to some participants. Major changes in the staffing of primary care practices, or, alternatively, the establishment of a system of formal relationships or referral mechanisms between such practices and appropriate community agencies who could be contracted with to provide the intervention would be needed before the Staff Counseling program could be widely adopted. However, national initiatives support the need to devote more resources to issues such as physical activity counseling (15,35,43). It is possible that future practice patterns will be able to accommodate interventions similar to the Staff Counseling program. If the Staff Counseling program is substantially more effective than the other conditions, health care planners will be challenged to develop methods for altering practices or referral patterns so that the Staff Counseling intervention can be disseminated.
Other factors that influence the efficacy of behavior change interventions have been considered in addition to intensity. Quality of implementation is an essential ingredient of effective programs. The best program cannot be effective if it is poorly implemented. High levels of implementation in ACT are promoted through extensive training and supervision of intervention staff. Implementation of intervention components and adherence to protocols are monitored throughout the study, and periodic feedback based on computer records is provided to each clinical center and individual health educator.
Both the Staff Assistance and Staff Counseling interventions have been developed based on a great deal of accumulated behavioral science literature. The behavior change theories that serve as the framework for the interventions are also the basis for many other effective interventions (12,16). The interventions target the psychological and social mediators that have been most strongly correlated with physical activity in many studies (9). The specific intervention strategies and techniques are based on results from previous evaluations of interventions to change physical activity (8,29,38) and other health behaviors. They are expected to be efficacious for improving physical activity and fitness in sedentary adults, practical for implementation in primary care settings, and reasonably cost-effective. The results of the ACT study will reveal to what extent the interventions fulfill these expectations.
1. Albright, C. L., J. W. Farquhar, S. P. Fortmann, et al. Impact of a clinical preventive medicine curriculum for primary care
faculty: results of a dissemination model. Prev. Med.
2. American College of Sports Medicine. Guidelines for Exercise Testing and Prescription, 5th ed.,
L. Kenney (Ed.). Baltimore: Williams & Wilkins, 1995.
3. Bandura, A. Social Foundations of Thought and Action.
Englewood Cliffs, NJ: Prentice-Hall, 1986.
4. Bassett, Jr. D. R., B. E. Ainsworth, S. R. Leggett, et al. Accuracy of five electronic pedometers for measuring distance walked. Med. Sci. Sports Exerc.
5. Blair, S. N., W. B. Applegate, A. L. Dunn, et al. Activity Counseling Trial (ACT): Rationale, design, and methods. Med. Sci. Sports Exerc.
6. Blair, S. N., W. L. Haskell, P. Ho, et al. Assessment of habitual physical activity by a seven-day recall in a community survey and controlled experiments. Am. J. Epidemiol.
7. Calfas, K. J., B. J. Long, J. F. Sallis, et al. A controlled trial of physician counseling to promote the adoption of physical activity. Prev. Med.
8. Dishman, R. K. and J. Buckworth. Increasing physical activity: a quantitative synthesis. Med. Sci. Sports Exerc.
9. Dishman, R. K. and J. F. Sallis. Determinants and interventions for physical activity and exercise
. In: Physical Activity, Fitness, and Health: International Proceedings and Consensus Statement,
C. Bouchard, R. J. Shephard, and T. Stephens (Eds.). Champaign, IL: Human Kinetics, 1994, pp. 214-233.
10. Fletcher, G. F., S. N. Blair, J. Blumenthal, et al. Statement on Exercise
: Benefits and Recommendations for Physical Activity Programs for All Americans. A Statement for Health Professionals by the Committee on Exercise
and Cardiac Rehabilitation of the Council on Clinical Cardiology, American Heart Association. Circulation
11. Friedman, C., R. C. Brownson, D. E. Peterson, and J. C. Wilkerson. Physician advice to reduce chronic disease risk factors. Am. J. Prev. Med.
12. Glanz, K., F. M. Lewis, and B. Rimer (Eds.). Health Behavior and Health Education: Theory, Research and Practice.
San Francisco: Jossey-Bass, 1990.
13. Graham-Clarke, P. and B. Oldenburg. The effectiveness of a general-practice-based physical activity intervention on patient physical activity status. Behav. Change
14. Hahn, R. A., S. M. Teutsch, R. B. Rothenberg, and J. S. Marks. Excess deaths from nine chronic diseases in the United States, 1986. J.A.M.A.
15. Harris, S. S., C. J. Caspersen, G. H. Defriese, and E. H. Estes. Physical activity counseling for healthy adults as a primary preventive intervention in the clinical setting. J.A.M.A.
16. Kanfer, F. H. and A. P. Goldstein. Helping People Change.
New York: Pergamon Press, 1986.
17. Karoly, F. and F. H. Kanfer (Eds.). Self-Management and Behavior Change: From Theory to Practice.
New York: Pergamon, 1988.
18. Kelly, M. P. Health promotion in primary care
: taking account of the patient's point of view. J. Adv. Nursing
19. King, A. C. Community intervention for promotion of physical activity and fitness. Exerc. Sport Sci. Rev.
20. King, A. C., W. L. Haskell, C. B. Taylor, et al. Group vs. home-based exercise
training in healthy older men and women. J.A.M.A.
21. King, A. C., W. L. Haskell, D. R. Young, et al. Long-term effects of varying intensities and formats of physical activity on participation rates, fitness, and lipoproteins in men and women aged 50-65 years. Circulation
22. King, A. C., S. N. Blair, D. E. Bild, et al. Determinants of physical activity and intervention in adults. Med. Sci. Sports Exerc.
23. Lewis, B. S. and W. D. Lynch. The effect of physician advice on exercise
behavior. Prev. Med.
24. Logsdon, D. N., C. M. Lazaro, and R. V. Meier. The feasibility of behavioral risk reduction in primary medical care. Am. J. Prev. Med.
25. Madlon-Kay, D. J., P. G. Harper, and C. J. Reif. Health promotion counseling in residency training. J. Gen. Intern. Med.
26. Marcus, B. H., M. G. Goldstein, A. Jette, et al. Training physicians to conduct physical activity counseling. Prev. Med.,
27. Marcus, B. H., B. M. Pinto, M. M. Clark, et al. Physician-delivered physical activity and nutrition interventions. Med. Exerc. Nutr. Health
28. Marcus, B. H. and L. R. Simkin. The transtheoretical model: applications to exercise
behavior. Med. Sci. Sports Exerc.
29. Martin, J. E. and P. M. Dubbert. Behavioral management strategies for improving health and fitness. J. Cardiac Rehabil.
30. McAuley, E. The role of efficacy cognitions in the prediction of exercise
behavior in middle-aged adults. J. Behav. Med.
31. McAuley, E., K. S. Courneya, D. L. Rudolph, and C. L. Lox. Enhancing exercise
adherence in middle-aged males and females. Prev. Med.
32. National Center for Health Statistics, U.S. Dept. of Health and Human Services. Health United States, 1995. Hyattsville, MD: Public Health Service, 1996; DHHS Publication No. (PHS) 96-1232.
33. NIH Consensus Development Panel on Physical Activity and Cardiovascular Health. Physical activity and cardiovascular health. J.A.M.A.
34. Orleans, C. T., L. K. George, J. L. Houpt, and K. H. Brodie. Health promotion in primary care
: A survey of U.S. family practitioners. Prev. Med.
35. Pate, R. R., M. Pratt, S. N. Blair, et al. A recommendation from the Centers for Disease Control and prevention and the American College of Sports Medicine. J.A.M.A.
36. Pender, N. J., J. F. Sallis, B. J. Long, and K. J. Calfas. Health care provider counseling to promote physical activity. In: R. K. Dishman (Ed.), Exercise Adherence (2nd ed.).
Champaign, IL: Human Kinetics, 1994, pp. 213-235.
37. Prochaska, J. O. and C. C. DiClemente. States and processes of self-change in smoking: toward an integrative model of change. J. Consult. Clin. Psychol.
38. Rejeski, W. J. Motivation for exercise
behavior: a critique of theoretical directions. In: Motivation in Sport and Exercise,
G. C. Roberts (Ed.). Champaign, IL: Human Kinetics, 1992, pp. 129-157.
39. Sallis, J. F., W. L. Haskell, P. D. Wood, et al. Physical activity assessment methodology in the Five City Project. Am. J. Epidemiol.
40. Sallis, J. F. and N. Owen. Ecological models. In: Health Behavior and Health Education: Theory, Research, and Practice (2nd ed.),
K. Glanz, F. M. Lewis, and B. K. Rimer (Eds.). San Francisco: Jossey-Bass, 1996, pp. 403-424.
41. Taylor, R. B., J. R. Urea, R. H. Butler, et al. Aerobic exercise
: physician beliefs and practices. Fam. Pract. Res. J.
42. U.S. Department of Health and Human Services. Healthy People 2000; National Health Promotion and Disease Prevention Objectives.
U.S. Government Printing Office, DHHS Pub. No. (PHS) 1991;91-50212.
43. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Physical Activity and Health: A Report of the Surgeon General.
Atlanta, GA: 1996.
44. Valente, C. M., J. Sobal, H. L. Muncie, Jr., et al. Health promotion: Physician's beliefs; attitudes, and practices. Am. J. Prev. Med.
45. Williford, H. N., B. R. Barfield, R. B. Lazenby, and M. C. Olson. A survey of physicians' attitudes and practices related to exercise
promotion. Prev. Med.