Physician-based smoking intervention: A rededication to a five-step strategy to smoking research

Addictive Behaviors. Vol. 22. No. 6. DD.835448.1997 Copyright Q i997 El&ier Science Ltd Printed in the USA. All rights reserved 0306403/97 $17.00 + .oO

PII s0306-4603(!v)00065-8

PHYSICIAN-BASED SMOKING INTERVENTION: A REDEDICATION TO A FIVE-STEP STRATEGY TO SMOKING RESEARCH JUDITH Division

of Preventive

K. OCKENE

and Behavioral

and JANE G. ZAPKA

Medicine,

University

of Massachusetts

Medical

School

Abstract - It is well established that physicians can have a significant effect on the smoking behavior of their patients. To do this. attention must be paid to putting in place multiple strategies or mechanisms in the organization where the physician practices, as well as in the macroenvironment (i.e., social and public policy). It has been questioned whether or not there is stagnation in the field of clinical smoking intervention requiring a rededication to basic research regarding smoking. With respect to physician-based smoking intervention, we alternatively suggest that recommitment to all phases of research is essential for moving forward physician-based smoking interventions in the rapidly changing health services and social environment. In this article, we first review the essential framework of the National Cancer Institute’s research science approach to cancer prevention and control. Evidence concerning physician-based interventions is then reviewed, followed by a schematic of a comprehensive framework for thinking about the process and intervention components needed for physicianbased smoking intervention to take place in the health-care setting, the impact they have, and the eventual outcome of such interventions. There is a discussion of the challenges for the delivery of smoking-cessation services presented by the rapidly changing healthy delivery system of the 1990s. Finally, we present recommendations concerning research priorities for physicianbased smoking intervention and the research funding process. 0 1997 Elsevier Science Ltd

Since the publication of the seminal research of Russell, Wilson, Taylor, and Baker (1979) regarding physician-delivered smoking intervention and the results of the five physician-delivered smoking intervention studies (Cohen, Stokey, Katz, Drook, & Smith, 1989; Cummings et al., 1989; Kottke, Brekke, Solberg, & Hughes, 1989; J. Ockene et al., 1991; Wilson et al., 1988), funded by the National Cancer Institute (NCI) in 1983, it has been well established that physicians can have a significant effect on the smoking behavior of their patients (Fiore et al., 1996; Manley, Epps, Husten, Glynn, & Shopland, 1991). More than 90% of the smokers who stop smoking in the United States do not use an organized program to help them stop, and most smokers who continue to smoke state that they would prefer to stop without the aid of a formal smoking-cessation program (U.S. Department of Health and Human Services, 1990). It is with this large group of smokers that physicians and other health-care providers can have a major impact. Almost 80% of the adult population has at least one contact per year with a physician (U.S. Department of Health and Human Services, 1994b), with the average yearly number of contacts (in 1994) being six per adult (National Center for Health Statistics, 1996). Thus, the physician as educator, facilitator, or counselor has the potential to be a powerful agent for smoking cessation (Glynn, Manley, & Pechacek, 1990; J. Ockene, 1987a). Individuals probably think more seriously about their health and smoking’s ef-

Requests for reprints should be sent to Judith K. Ockene, Medicine, Department of Medicine, University of Massachusetts Worcester, MA 01655; E-mail: [email protected]. 835

Division of Preventive and Behavioral Medical School, 55 Lake Avenue North,

J. K. OCKENE and J. G. ZAPKA

836

feet on it when in a physician’s office or in a hospital than at any other time. This opportunity for health promotion with smokers, ex-smokers, and would-be smokers should not be overlooked and may increase because of the recently proposed settlement of the attorneys general with the tobacco companies in which the latter agreed to pay for treatment of smoking. Unfortunately, although the opportunities are there, physician-based intervention often does not take place in “real-world” settings (J. Ockene, McBride, Sallis, Bonollo, & I. Ockene, in press), making this a major research and public health concern. Shiffman (1993) in his article “Smoking Cessation Treatment: Any Progress?” stated that there is stagnation in the field of intensive clinical smoking intervention requiring a rededication to basic research regarding smoking behavior and nicotine dependency to help move the field of smoking cessation forward. Contrary to Shiffman’s concern of stagnation, in the area of physician-based smoking intervention, there has been tremendous progress. However, using the model of the phases of cancer prevention and control research, developed by Greenwald (1982) and refined by Greenwald and Cullen (1985) and promulgated by the NC1 (see Fig. l), we believe late phase work regarding real-world applications (Phase V) has been underemphasized. Additionally, the rapidly changing social and medical environment raises some basic questions requiring some additional Phase I through Phase IV research. Therefore, we suggest that recommitment to research at all phases is essential and that a different process for judging and funding research may be needed. In this article, we first review the essential framework of the NC1 research science approach to cancer control. The evidence concerning the impact of physician-based smoking interventions and the intervention strategies necessary for its implementation in the health-care setting is then reviewed and summarized in a schematic of a comprehensive framework for thinking about the intervention process and components. We then discuss the challenges of the 1990s presented by a rapidly changing health delivery system that can profoundly affect smoking-cessation services. Last, we present our recommendations concerning research priorities and the research funding process.

Prevention Program Centers &Community Oncology Program Cancer Control Science Program

-.c Epidrmiology

I * Hypothesis Dwelopmmt

___!

+_______I

II -b Methods Developnlcnt

(_______!

111 + COnvOllCd Intervention Trials

IV Defined Population Studies

+______!

+

V Dan-Iion and ImpI-on

+______:

Research and Community Settings

Fig. 1. Phases of cancer prevention control and research.

Physician-based

THE

NC1

smoking intervention

RESEARCH

837

FRAMEWORK

The stepwise phases of the research process recommended by the NC1 go from hypothesis development (Phase I) and methods development (Phase II) through demonstration and implementation (Phase V), with the intention of the eventual dissemination of a nationwide prevention and health services program. It was intended that research hypotheses and methods development for activities such as intervention and measurement would occur during Phases I and II, respectively. These more basic aspects of research would lead to Phase III studies or randomized controlled intervention trials (and sometimes case-control studies) needed to test the efficacy of an intervention (developed in Phase II studies) on a group of individuals who may be selected to optimize outcome and interpretation (i.e., they may be a relatively homogeneous group that facilitates research management but are not a representative sample). These Phase III studies would lead to Phase IV or defined population studies to test the export of an efficacious intervention when it is applied to a large, generalizable, distinct, well-characterized population. The target population could be institutions such as health-care settings. The information gained from Phase III and Phase IV research, or combinations of them, could be applied to a community at large in Phase V studies. Evaluation and quality control procedures must be in place to ensure proper implementation of knowledge gained from Phases III and IV and an ability to measure the public health effect. Although it is beyond the scope of this article to review all of the evidence related to physician-based smoking interventions in all five phases, this continuum should be kept in mind as we consider the information gained from physician-based smokingintervention research and as we make recommendations for future priorities. We will review provider-focused smoking-intervention studies, which are primarily Phase III and IV research. These efficacy and effectiveness studies were built on a relatively solid foundation of Phase I and Phase II studies.

EFFICACY

OF

PHYSICIAN-BASED

INTERVENTION

This review is organized according to several types of settings and study themes. Evidence from randomized

trials in the outpatient setting

Summarizing the smoking-intervention literature, the Agency for Health Care Policy and Research (AHCPR) smoking-cessation guideline (Fiore et al., 1996) recommends that simple advice by one’s physician to stop smoking is more effective than no advice at all and that as the dose of physician-delivered smoking intervention increases (ranging from 50 set to 15 min), the effect increases. Evidence for this statement is seen in Table 1, which includes the five NCI-supported studies noted earlier and a more recent study by Hollis, Lichtenstein, Vogt, Stevens, and Biglan (1993). Effect of multiple providers and intervention modalities on cessation The success of an intervention increases with the number of intervention

modalities employed and the number of health-care professionals involved (Fiore et al., 1996). The value of intervention by other providers and modalities is especially clear when physicians deliver very brief advice. Hollis et al. (1993) compared a 30-set physician intervention to the same intervention plus one of three nurse-assisted interventions.

Quit rates (in %) at 6 months: Intervention condition Returnees only 1. 1.3 2. 1.F 3. 7.0* 4. 6.3* All enrollees 1. 0.9 2. 5.0* 3. 4.0* 4. 3.8* Quit rates (in %) at 12 months (all enrollees): Intervention condition >l-week abstinencea 1. 7.1 2. 8.0 9-month abstinencea 1. 1.5 2. 2.6 Quit rates (in %) at 3 months: Intervention condition 1. 6.8 2.b 12.3** Quit rates at 3 and 12 months (in %) combined: Intervention condition Self report 1. 3.9 2. 7.2** Quit rates (in %) at 12 months (all enrollees): Intervention condition 1. 14.3 2. 12.0 3. 11.8

1. Control: received 1-hr lecture with four. step protocol 2. Gum: education plus charts identified with a prescription sticker for gum 3. Reminder: education plus charts identified with two stickers Did you talk about smoking? The patient agreed to the following quit date : _ 4. Gum & reminder: education plus #2 and #3 Quit for life 1. Controls: usual care 2. Training: 3-hr training for doctors plus office staff trained to implement reminder system

1. Brief doctor advice alone (30 set) 2. Three conditions of nurse-assisted quit (self-quit video, group, or clinic)

Doctors helping smokers 1. No assistance 2. Materials: received smoking-cessation manual for patients 3. Workshop: 6-hr skill building plus cessation manual

Randomized by doctor with panel of patients 1,420 patients from general medicine clinic of a city-county teaching hospital

Health maintenance organization (HMO) population Doctors grouped into 22 units Randomized by unit

HMO population Randomized by patient 2,707 patients

66 doctors randomized in intact groups of doctors’ practices

Cohen, Stokey, Katz, Drook, and Smith, 1989

Cummings et al., 1989

Hollis, Lichtenstein, Vogt, Stevens, and Biglan, 1993

Kottke, Brekke, Solberg, and Hughes, 1989

(Conhzued)

Results

Provider intervention conditions

Design

Investigators

Table 1. Physician-based smoking intervention randomized trials

Intact practices randomized 1933 patients

Wilson and colleagues (Wilson et al., 1988)

“Significant difference only for those who reported desire to quit. bNurse-assisted quit conditions were collapsed. RCT = Randomized Clinical Trial. *p < .05; **p < .ool.

Randomized by patients 1286 patients of primary care clinics

Ockene and colleagues (Ockene et al., 1991)

1. Usual Care 2. Cued only with gum 3. Cued plus MD training with gum (trained)

1. Advice only (AO) (6.7 min) l MD advice l Chart stickers 2. Brief patient centered counseling (CI) (12.3 min) l advice plus patient received patientcentered counseling l self-help booklet l resource list l followup visit/telephone call within 2 weeks l letter of encouragement signed by MD l chart stickers 3. Same as condition 2 plus free NCG (CI +NCG) pius (13.9 min) l Visit (3-5 min) with assistant for instructions on gum use 50% of patients were randomized to maximal followup consisting of 3 telephone calls and 3 follow-up letters

Table 1. (continued)

*p < .05

Smoking quit rates at 1 year: MD treatment condition 3-month sustained quit (validated) (in%) 1.4.4 2.6.1 3. 8.8 Attempts to quit (self-reported) (in%) 1.36.4 2.60.7 3.71.9

6-month cessation (self-report) for combined minimal and maximal conditions: MD treatment condition Total (in %) 1.9.1 2.11.9 3. 17.4 Overall effect p < ,025 l-year maintained cessation was significant between groups

840

J. K. OCKENE and J. G. ZAPKA

There was little difference between the latter three conditions; however they had a significantly greater effect on cessation than did the very brief physician advice alone. Of note is that only 9.5% of smokers referred to the group intervention attended at least one session, underlining the importance of interventions delivered by providers and modalities in the practice setting; it is part of the smoker’s “natural environment” (J. Ockene, 1987b) and does not require them to go elsewhere. Other adjuncts such as video, computer-driven expert-systems, or telephone counseling are ways to reduce physician burden. Effect of physician time on cessation

Several studies have demonstrated that the greater the dose of physician intervention, the greater the cessation outcome. For example, Wilson et al. (1988) demonstrated the greatest effect in the condition in which physicians spent the most time delivering an intervention (see Table 1). Similarly, in the J. Ockene et al. (1991) study, the more time the physician spent (6.7 min to 13.9 min), the greater the cessation effect. By contrast, in the Hollis et al. (1993) study, only 30 set of advice were delivered by the physician with the additional intervention time spent by a nurse. Although far less physician time was required in the Hollis et al. (1993) study, the cessation rate was substantially lower than that found in the former two studies. Important considerations that remain to be addressed regarding the best combination of time and effort for physicians and other providers include at least two: (a) whether support staff can offer some skills and attributes to enhance what the physician does (e.g., interviewing skills that can help facilitate motivation, rapport, better follow-up, offering another concerned professional’s opinion and (b) whether shifting some tasks to support staff will increase the probability that interventions will be delivered to a large percentage of patients. The product of efficacy and penetration is key. Examples illustrate the need for studies along the continuum from Phase I to Phase IV. Effect of multiple components on cessation

Randomized trials have demonstrated the need to use multiple intervention components for physician-delivered intervention to have significant effect. For example, of the five NCI-funded trials, all used some level of physician education, ranging in time from 1 to 3 hr. However, the three that demonstrated efficacy of physician-delivered intervention also used a reminder system intervention (Cohen et al., 1989; J. Ockene et al., 1991; Wilson et al., 1988) and one (J. Ockene et al., 1991) used additional intervention from other providers (telephone counselors). A descriptive study of physician practice behavior found that if physicians record smoking status in a location where they will be reminded (e.g., on a problem list or a flow sheet) patients are more than twice as likely to report that their physicians told them to stop smoking (McBride, Plane, Underbakke, Brown, & Solberg 1997). Prevention trials also have focused on the systematic use of medical record and computer reminders (Davis, Thomson, Oxman, & Haynes, 1995; Grimshaw & Russell, 1993; McPhee, Bird, Fordham, Rodnick, & Osborn, 1991) to improve the implementation of preventive services. Another proposed method for ensuring smoking-cessation intervention is to expand the vital signs to include an assessment of smoking status (Fiore et al., 1995). This simple institutional change increases markedly the proportion of patients who report that their health-care providers ask and counsel them about smoking cessation (Robinson, Laurent, & Little, 1995).

Physician-based smoking intervention

841

These studies reinforce the principle that educating physicians is not enough, and it is essential to affect the whole system in which they work if long-term change in their practice is to occur. There also is growing recognition that, if one is to ensure effectiveness in real-world settings, it is important to work with clinic staff to define procedures that maximize the degree of organizational fit of the desired intervention with daily clinic operations (Roberts-Gray & Scheirer, 1988; Shortell& Kaluzny, 1986). In addition, there has been an evolving awareness that a policy intervention component also is needed that includes the use of covered prevention benefits, incentives, feedback, and other reinforcements if providers are to be motivated to intervene in real-world settings (J. Ockene, McBride, et al., in press; Solberg et al., 1996). The role of public and organizational policy alternatives need testing. Examples of questions regarding systems interventions and policy interventions include: (a) What are the best incentives or combination of incentives for physicians and/or patients? (b) What are the most effective and efficient strategies that can be used to cue providers to intervene? and (c) How can each of these best be implemented in different types of settings and systems? Effect of patient illness on cessation

A physician’s impact on a patient’s smoking also varies greatly with the visit context. The patient who is symptomatic or already ill is more likely to respond to physician advice than the patient who feels well and has few if any symptoms or abnormal findings (I. Ockene &J. Ockene, 1992; J. Ockene et al., 1992). Randomized trials have shown high biochemically verified sustained quit rates (50%63%) at 6 months for patients who are advised to quit smoking after having a myocardial infarction or after having been told they have coronary heart disease and significantly higher rates (65%-75%) for patients who receive additional counseling by health educators or nurses by telephone or in person (DeBusk et al., 1994; J. Ockene et al., 1992). These cessation rates are directly related to the severity of disease (Frid et al., 1991; J. Ockene et al., 1992). Cessation rates are lower (30%-40%) when intervention is directed at smokers who are not ill but are at high risk for infarction (Rose & Hamilton, 1978). Unfortunately, similar to “healthy patients,” relapse occurs after intervention stops and continues for several years (Rosa1 et al., 1997). Therefore, more research on relapse prevention with individuals who already are ill is necessary.

A

COMPREHENSIVE PHYSICIAN-BASED

FRAMEWORK INTERVENTION

FOR

Thus, there are considerable data that demonstrate physician-based interventions have a significant effect on cessation rates. The more time the physician spends, the greater the effect, and the more intervention modalities used and providers intervening, the greater the effect. The health of the smoker also affects the outcome. Education of providers that builds knowledge, confidence, and skills must occur in the context of multiple theory-driven strategies or mechanisms (i.e., reminder and tracking systems, feedback, and policies) in the organization or microenvironment where the physician practices (Kottke, Brekke, & Solberg, 1993; Ockene, Lindsay, Hymowitz, et al., in press; Pommerenke & Dietrich, 1992). There also has been a growing conviction that enabling mechanisms must be in place in the macroenvironment (i.e., social and public policy) to encourage the implementation of organizational procedures and individual provider behavior (Schauffler & Parkinson, 1993).

J. K. OCKENE and J. G. ZAPKA

842

Figure 2 presents a framework that reflects these findings and thinking about the critical intervention components needed, given the complexities not only of patient behavior but also of provider and organizational behavior. The major interventions or components affecting provider behavior noted in Figure 2 are grouped broadly into provider education, clinical systems and procedures, and organizational and health policy (J. Ockene, McBride, et al., in press). Critical to the framework is the understanding that individual factors and peer community norms also affect providers and patients (noted as effect modifiers in Fig. 2). A model useful for understanding the importance of the combination of strategies to increase providers’ use of efficacious interventions is Green and Kreuter’s (1991) precede/proceed model. It includes strategies similar to those noted in Figure 2: predisposing strategies (e.g., guidelines, education, and insurance coverage), practice-enabling strategies (e.g., office reminders and a system to follow patients), and reinforcing strategies to reinforce use of preventive practices (e.g., feedback and performance measures). CHALLENGES

AND PRACTICE

COMPLEXITIES IN THE

OF

CLINICAL

’90s

Progress in the effective delivery of comprehensive smoking-cessation services cannot be reviewed without highlighting the massive and ongoing structural and process changes within the health-care system itself. The past decades have witnessed environmental trends including restructuring of physician payment systems (fee for service to capitation); vertical and horizontal integration of services delivery; consolidation of practices into network affiliations; increasing for-profit status of major providers; increasing role of payors in determining benefit packages; increasing presence of managed-care approaches to address problems in access, quality, and cost; and increasing

Intervention Components PlWidW Education

Health Care Provider-Assisted Smoking Intervention: ramework of Intervention Components, Impact and Outcomes

Fig. 2. Health-care provider-assisted components, impact, and outcomes.

smoking

intervention:

framework

of intervention

Physician-based smoking intervention

843

demands on the provider to be cost effective and efficient. These trends are complex and synergistic and beyond the scope of this review. The turmoil concomitant with a complex and evolving system change, however, has contributed to the barriers and facilitators that affect the delivery of preventive services, including smoking-cessation services. We here comment briefly on several trends that particularly affect physicianbased interventions and the need for continued commitment to research in this area. The continuing changes in the financing and delivery of care and the growth of managed care has led to the increased demand for evidence-based recommendations and practical applications to smoking and other health-related problems (Center for the Advancement of Health, 1997). The recent revision of the U.S. Preventive Services Task Force (1996) Guide to Clinical Preventive Services highlights the need for more emphasis on preventive counseling services and the strong evidence supporting the importance of smoking intervention. In addition, Putting Prevention Into Practice, the U.S. Public Health Services program (U.S. Department of Health and Human Services, 1994a), and The Canadian Guide to Clinical Preventive Healthcare (The Canadian Task Force on the Periodic Health Examination, 1994) are important guides for organized delivery systems to implement preventive services. In addition, implementation by organizations of the AHCPR clinical practice guideline for smoking cessation (Fiore et al.? 1996) will be a major boost for considering the elements of quality smoking-cessation programs. It promulgates a systems approach component directed at health-care administrators, insurers, managed-care organizations, and purchasers (U.S. Department of Health and Human Services, 1997, p. l), in which it is noted, “The most successful smoking cessation programs are supported by institutional policies. They incorporate reimbursement practices, clinical and systems procedures, incentives for providers, and clinician education.” These new policy recommendations, which focus on explicit coverage of preventive services, if responded to by organizations, can provide an important impetus to provide such services. They establish clinical tobacco interventions as part of standards of practice, therefore motivating physicians and organizations to provide high-quality help to their patients. Consumers also are requesting preventive services, which also can motivate physicians and organizations to provide them. New standards developed by accreditation bodies such as the Joint Commission on Accreditation of Healthcare Organizations can facilitate a voluntary systematic approach to prevention services (Larose, 1995). Another important development for encouraging smoking intervention practices by primary care physicians is the draft of the “Health Plan Employer Data Information Set” (HEDIS) 3.0, which includes two measures on smoking (HEDIS 3.0,1997). One measure is for the identification of smokers within each managed-care plan, and one is intended to track cessation rates among members (Zapka, Merriam, & J. Ockene, 1977). The complex and ever-changing environment described earlier has put new and renewed pressures on the day-to-day practice of medicine. These include pressures to increase efficiency, including seeing more patients, and to consider multiple needs of patients, not only the presenting acute symptoms. Although there is a focus on implementation of preventive services, especially in the managed-care setting, issues of prevention often may not compel action in the clinical setting, especially an overburdened one (Kottke et al., 1993). Time demands on providers need to be considered not only from the perspective of patient care but also for provider education. Traditional provider education and training have usually included strategies such as grand rounds, workshops, seminars, and

844

J. K. OCKENE

and J. G. ZAPKA

minicourses. Given the limited time available for physicians and nurses to participate in continuing education opportunities, we must be more creative about where and how they can take place. Educational efforts need to involve a variety of professionals, because multiple providers, in collaboration with physicians-for example, nurses (Hollis et al., 1993) and pharmacists (Baluch, 1995)-have a significant effect on smoking. Additionally, administrative sanction is vital. An example of an enabling approach to preventive services has been implemented by the Group Health Cooperative of Puget Sound (Thompson, Taplin, McAfee, Mandelson, & Smith, 1995). There is an emphasis on each of the intervention components noted in Figure 2, with a commitment to education and systems development by the chief executive officer/medical director. When training events have an institutional commitment, they will happen, and providers will attend them. In the physician-assisted lipid-lowering study, the Worcester Area Trial for Counseling in Hyperlipidemia conducted, in a closed-panel health maintenance organization (J. Ockene et al., 19954, a 3-hr training program with 100% attendance of 100 physicians. There were four time choices for the sessions, and the health maintenance organization administration supported it and allowed time for the providers to attend training.

RECOMMENDATIONS SMOKING

FOR INTERVENTION

PHYSICIAN-BASED RESEARCH

This review suggests several important areas for research to help move the field of physician-based smoking intervention forward. The following three recommendations are not exhaustive. They include topics that would benefit from studies representative of all five phases of the basic to community spectrum illustrated in the NC1 research process model. It is important to note that there is evidence to indicate that the framework suggested in Figure 2 is appropriate. However, we have no Phase V studies that clearly demonstrate its usefulness. This gap will be discussed shortly. Stepped-care model and patient-treatment matching model

Abrams et al. (1996) proposed a stepped-care and patient-treatment matching model for delivery of smoking-cessation treatment into mainstream health care in order to be able to treat the general population of smokers at a reasonable cost. The model includes brief counseling by physicians for the “easier” smokers and moves to more extensive interventions for the more difficult-to-treat smokers. The proposed model has not been tested and lends itself to the need for reinforcement of hypotheses (Phase I) and development of assessment and intervention methods (Phase II) in order to eventually be able to determine the efficacy (Phase III) and effectiveness (Phase IV) of such a model and its application in real-world settings (Phase V). Questions needing answers include: Does such a model work for all types of smokers? Can a feasible, useful needs assessment tool be designed? Who is best served? What kind of reminder systems work best? What are the incentives? Are there optimal combinations of provider-delivered interventions? Who is responsible for the management of each smoker? Multiple risk factor patients

Providers generally see patients with multiple risk factors and multiple problems, and increasingly they are told, “Don’t just intervene in smoking, but do smoking, do

Physician-based

smoking intervention

845

diet, do alcohol, do partner abuse, do everything; plus do this in a very brief amount of time.” Therefore, we need to develop hypotheses (Phase I), understand the process of multiple risk factor change and develop integrated interventions (Phase II) in order to help providers do the right thing (J. Ockene, McBride, et al., in press). Only a few studies to date have investigated interventions for multiple risk factors in health-care settings (Family Heart Study Group, 1994; Imperial Cancer Research Fund OXCHECK Study Group, 1995; Logsdon, Lazaro, & Meier, 1989) and none has addressed the actual integration of interventions. They each have addressed the risk factors individually. Examples of questions regarding the process of patient behavior change and provider behavior change needing answers include the following: Should patients sequence their approach to multiple risk factor change, or can they be changed simultaneously? How should priorities for change be set? Do different types of people do better with different physician-based approaches? What factors determine which approach is best? Scale development is also needed in multiple risk factor studies in order to measure needs and change adequately. Once hypotheses and methods are developed for studying provider-based interventions for patients with multiple risk factors, efficacy and effectiveness studies will need to be done to test the interventions developed. Education of providers

Given that there are many skills that providers could be taught regarding smoking intervention, that there is a need for brief interventions, and that we do not know which intervention steps are the most important (effective), a determination of the latter could contribute to the field of provider-based smoking intervention (J. Ockene et al., 1994). In addition, the ever-decreasing time available for continuing education requires that attention in Phase III and Phase IV studies be shifted to the development and assessment of creative ways for relatively brief education and training. On-site academic detailing (Daly et al., 1993; Soumerai & Avorn, 1990) could respond to the limited amount of time providers have available. Phase V: Demonstration

and implementation emphasis

Solberg et al. (1996) raised a critical research question, “How do we implement guidelines?” There is little evidence that anyone has learned exactly how to do this well. We do know, however, that there are things we can do to ensure better implementation of guidelines (Grimshaw & Russell, 1994; Pommerenke & Weed, 1991). Guidelines and standards for prevention can potentially overcome barriers that impede clinicians from implementing optimal intervention on tobacco use by motivating organizational change that may not otherwise occur. However, organizations must be willing to implement the guidelines and set up the systems needed to facilitate their utilization. A new accountability could increase the need for clinician education about brief smoking intervention and for staff training and ultimately improve our effectiveness in helping smokers quit. Studies conducted under well-controlled experimental conditions, as well as demonstration projects conducted under less well-controlled conditions in the clinical setting, provide the foundation for the next generation of effectiveness studies that address the role of various factors in the health-care environment in the delivery of services. However, such factors (e.g., reimbursement policies, covered benefits) do not lend themselves well to tightly controlled randomized trials. Use of quasi-experimental designs and application of qualitative strategies are needed. This represents a different

846

J. K. OCKENE and J. G. ZAPKA

challenge to the research community. As noted by Dr. Tom Glynn (personal communication June 199’7) Director of NC13 Division of Cancer Prevention and Control, “reviewers have a hard time approving and allocating millions of dollars to studies which have few controls in place.” The scientific community will need to consider paradigms other than randomized trials for Phase V studies in health-care settings. The rapidly changing health-care environment also is a challenge to the process of funding of research. The process must provide opportunities for researchers to “turn on a dime” in a health-care environment that is changing so rapidly. In such an environment, funds must be allocated relatively quickly. The current extended lag time between proposal and award results in lost opportunities for many natural experiments of innovation. Reviewers also must be willing to fund ideas that are creative and innovative, keeping in mind that innovation is what will keep research in the health-care environment moving forward. REFERENCES Abrams, D., Orleans, C., Niaura, R., Goldstein, M. G., Prochaska, J. O., & Velicer, W. (19%). Integrating individual and public health perspectives for treatment of tobacco dependence under managed health care: A combined stepped care and matching model.Annals of Behavioral Medicine, 18(4), 290-304. Baluch, W. (1995). Pharmacists’ role in a smoking-cessation program at a health maintenance organization. American Journal of Health-System Pharmacy, 52,287-293. The Canadian Task Force on the Periodic Health Examination. (1994). The Cunadinn guide to clinical preventive health care. Ottawa, Ontario: Health Canada. Center for the Advancement of Health. (1997). Changing health-related behavior in managed care settings.

Washington, DC: Center for the Advancement of Health. Cohen, S., Stokey, G., Katz, B., Drook, C., & Smith, A. (1989). Encouraging primary care physicians to help smokers quit. Annals of Internal Medicine, 110(S), 648-652. Cummings, S., Coates,T. J., Richard, R. J., et al. (1989)Training physicians in counseling about smoking cessation: A randomized trial of the “Quit for Life” program. Annals of Internal Medicine, 110,64&647. Daly, M., Balshem, M., Sands, C., James, J., Workman, S., & Engstrom, P. (1993). Academic detailing: a model for in-office CME. Journal of Cancer Education, 8(4), 273-280. Davis, D., Thomson, M., Oxman, A., & Haynes, R. (1995). Changing physician performance. A systematic review of the effect of continuing medical education strategies Journal of the American MedicalAssociation, 274,7OlX705.

DeBusk, R., Houston Miller, N., Superko, H., et al. (1994). A case-management system for coronary risk factor modification after acute myocardial infarction.Annals of Internal Medicine, 120,721-729. Family Heart Study Group. (1994). Randomised control trial evaluating cardiovascular screening and intervention in general practice: Principal results of the British Family Heart Study. British Medical Journal, 308,31>320.

Fiore, M., Bailey, W., Cohen, S., et al. (1996). Smoking cessation: Clinical Practice Guideline No. 18 (AHCPR Publication No. 96-0692). Rockville. MD: U.S. Deoartment of Public Health and Human Services. Publit Health Service,Agency for Health Care Policyand Research. Fiore, M., Jorenby, D., Schensky, A., Smith, S., Bauer, R., & Baker,T. (1995). Smoking status as the new vital sign: Effect on assessment and intervention in patients who smoke. Mayo Clinic Proceedings, 70,209213.

Frid, D., Ockene, I., Ockene, J., Merriam, I’., Goldberg, R., Kristeller, J., & Barrett, S. (1991). Severity of angiographically proven coronary artery disease predicts smoking cessation. American Journal of Preventive Medicine, 7(3), 131-135.

Glynn, T., Manley, M., & Pechacek, T. (1990). Physician-initiated smoking cessation program: The National Cancer Institute trials In P. Engstrom, B. Rimer, & L. Mortenson (Eds.), Advances in ctmcer control: Screening and prevention research (pp. 11-25). New York: Wiley-Liss Green, L., & Kreuter, M. (1991). Health promotion planning:An educational and environmental approach (2nd ed.). Mountain View, CA: Mayfield Publishing Company. Greenwald, P (1982). New directions in cancer control. Johns Hopkins Medical Journnl, 151,209-213. Greenwald. P.. & Cullen. J. (1985). The new emohasis in cancer control. Journal of the National Cancer Institute, 74(5), 543-551.’ ~ ’ Grimshaw, J., & Russell, 1. (1993). Effect of clinical guidelines on medical practice: A systematic review of rigorous evaluations. Lancer, 342,1317-1322. Grimshaw, J., & Russell, 1. (1994). Achieving health gain through clinical guidelines 11:Ensuring guidelines change medical practice. Quality in Health Care, 3,45-51.

Physician-based

smoking

intervention

847

HEDIS 3.0 (1997). Washington, DC: The National Committee for Quality Assurance. Hollis, J., Lichtenstein, E., Vogt, T., Stevens, V., & B&Ian, A. (1993). Nurse-assisted counseling for smokers in primary care.Annals of Internal Medicine, 118,521-525. Imperial Cancer Research Fund OXCHECK Study Group. (1995). Effectiveness of health checks conducted by nurses in primary care: Final results of the OXCHECK study. British Medical Journal, 310, 1099-1104. Kottke, T., Brekke, M., & Solberg, L. (1993). Making “time” for prevention services. Mayo Clinic Proceedings, 68,785-791. Kottke, T., Brekke, M., Solberg, L., & Hughes, J. (1989). A randomized trial to increase smoking intervention by physicians: Doctors helping smokers, Round 1. Journal of the American Medical Association, 261(14), 2101-2106. Larose, S. (1995). JCAHO initiates development of performance measures for preventive services. Inside Preventive Cure, 2,7-8. Logsdon, D., Lazaro, C., & Meier, R. (1989). The feasibility of behavioral risk reduction in primary medical care.American Journal of Preventive Medicine, 5,249-256. Manley, M., Epps, R., Husten, C., Glynn, T., & Shopland, D. (1991). Clinical interventions in tobacco control: A National Cancer Institute training program for physicians. Journal of the American Medical Association, 268(22), 3172-3173. McBride, P E., Plane, M. B., Underbakke, G., Brown. R., & Solberg, L. I. (1997). Smoking screening and management in primary care practices.Archives of Family Medicine, 6,165-172. McPhee, S., Bird, J., Fordham, D., Rodnick, J., & Osborn, E. (1991). Promoting cancer prevention activities by primary care physicians: Results of a randomized, controlled trial. Journal of the American Medical Association, 266,538-544. National Center for Health Statistics. (1996). Health, United States, 1995. Hyattsville, MD: Public Health Service. Ockene, I., & Ockene, J. (Eds.). (1992). P revention of coronary heart disease. Boston, MA: Little, Brown and Company. Ockene, J. (1987a). Physician-delivered interventions for smoking cessation: Strategies for increasing effectiveness. Preventive Medicine, 16,723-737. Ockene, J. (1987b). Smoking intervention: The expanding role of the physician. American Journal of Public Health, 77(7), 782-783. Ockene, J., Adams, A., Pbert, L., Luippold, R., Hebert, J. R., Quirk, M., & Kalan, K. (1994). The physiciandelivered smoking intervention: Factors that determine how much the physician intervenes with smokers. Journal of General Internal Medicine, 9,379-384. Ockene, J., Kristeller, J., Goldberg, R., Hosmer, D., Amick, T., & Kalan, K. (1991). Increasing the efficacy of physician-delivered smoking intervention: A randomized clinical trial. Journal of General Internal Medicine, 6.1-8. Ockene, J., Kristeller, J., Goldberg, R., Ockene, I. S., Barrett, S., Pekoe, I?, & Merriam, P. (1992). Smoking cessation and severity of disease:The Coronary Artery Smoking Intervention Study. Health Psychology, 11(2), 119-126. Ockene, J., Lindsay, E., Hymowitz, N., Giffen, C., Purcell,T., Pomrehn, P., & Pechacek,T. (in press). Tobacco control activities of primary care physicians in the Community Intervention Trial for smoking cessation (COMMIT). Tobacco Control. Ockene, J., McBride, I?, Sallis, J., Bonollo, D., & Ockene, I. (in press). Synthesis of lessons learned from cardiopulmonary preventive interventions in healthcare practice settings.Annals of Epidemiology. Ockene, J.. Ockene, I., Quirk, M., Hebert, J. R., Saperia, G. M., Luippold, R. S., Merriam, P A., & Ellis, S. (1995). Physician training for patient-centered nutrition counseling in a lipid intervention trial. Prevenrive Medicine, 24,563-570. Pommerenke, E, & Dietrich. A. (1992). Improving and maintaining preventive services. Part 1: Applying the patient model. Journal of Family Practice, 34(l), 88-91. Pommerenke, E, & Weed, D. (1991). Physician compliance: Improving skills in preventive medicine practices. American Family Physician, 43(2), 56CL568. Roberts-Gray, C., & Scheirer, M. (1988). Checking the congruence between a program and its organizarional environment. In K. J. Conrad and C. Roberts-Gray (Eds.), Evaluating program design: New directions for program evaluation (Vol. 40). San Francisco, CA: Jossey-Bass. Robinson. M., Laurent, S., & Little, J. (1995). Including smoking status as a new vital sign: It works! Journal of Family Practice, 40(6), 556-563. Rosal, M., Ockene, J., Ma, Y., Hebert, J. R., Ockene, I. S., Merriam. I!, & Hurley, T. G. (1997). The Coronary Artery Smoking Intervention Study (CASIS): Five-year followup. Manuscript submitted for publication. Rose, G., & Hamilton, P. (1978). A randomized controlled trial of the effect on middle-aged men of advice to stop smoking. Journal of Epidemiology and Community Health, 32,275-281. Russell, M., Wilson, C., Taylor, C., & Baker, C. (1979). Effects of general practitioners’ advice against smoking. British Medical Journal, 2,231-235. Schauffler. H., & Parkinson, M. (1993). Health insurance coverage for smoking cessation services. Healfh Education Quarterly, 20(2), 185-206.

J. K. OCKENE and J. G. ZAPKA

848

Shiffman, S. (1993). Smoking cessation treatment: Any progress? Journal of Consulting and Clinical Psychology, 61.718-722.

Shortell, S., & Kaluzny, A. (Eds). (1986). Health care management:A text in organization theory and behavior. New York, NY: Wiley. Solberg, L., Kottke, T., Brekke, M., Calomeni, C. A., Conn, S. A., & Davidson, G. (19%). Using continuous quality improvement to increase preventive services in clinical practice-Going beyond guidelines Preventive Medicine, Z&259-267.

Soumerai, S., & Avom, J. (1990). Principals of educational outreach (“academic detailing”) to improve clinical decision making. Journal of American Medical Association, 263(4), 549-556. Thompson, R., Taplin, S., McAfee, T., Mandelson, M., & Smith, A. (1995). Primary and secondary prevention services in clinical practice: Twenty years’ experience in development, implementation, and evaluation. Journal of the American Medical Association, 273(4), 1130-1135. U.S. Department of Health and Human Services. (1990). The healrh benefits of smoking cessation: A report of the Surgeon General (DHHS Publication No. CDC 90-8416). Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services. (1994a). Clinician’s handbook of preventive services: Purring prevention into practice. Washington, DC: Office of Disease Prevention. U.S. Department of Health and Human Services (1994b). Currenr esrimares from rhe National Health Interview Survey, 1993 (Series 10: Dara from rhe National Healrh Survey No. 190) (DHHS Publication No. PHS 95-1518). Hyattsville, MD: Public Health Service, Centers for Disease Control, National Center for Health Statistics U.S. Department of Health and Human Services. (1997). Smoking cessation:A systems approach (AHCPR Publication No. 97-0698). Rockville, MD: Public Health Service, Agency for Health Care Policy and Research. U.S. Preventive Services Task Force. (1996). Guide ro clinical preventive services: Reporr of rhe U.S. Preventive Services Task Force (2nd ed.). Baltimore, MD: Williams &Wilkins. Wilson, D., Taylor, W., Gilbert, R., Best, J. A., Lindsay, E. A., Willms, D. G., & Singer, J. (1988). A randomized trial of a family physician intervention for smoking cessation. Journal of rhe American Medical Associarion,260(11), 1570-1574. Zapka, J., Merriam, P., & Ockene, J. (1977). Smoking cessation benefits in HMOs. HMO Practice, 11(l),

27-33.