Roselyn P Epps

I’m a pediatrician. I have been in private practice, I’ve taught and am still a professor of pediatrics and child health at Howard University. But I consider myself primarily a public health physician because most of my career has been in that field, ranging from work as a clinic physician in a child health clinic in the far Southeast Washington area to the Acting Commissioner of Health for the District of Columbia.

I only mention that because I am going to talk from the perspective of the consumer, the patient, the individuals, those who receive the services, as opposed to those who do research and/or deliver services. I’m going to speak briefly about channels for receiving information. What we really need is a marketing strategy for smoking cessation programmes. We have many piecemeal approaches to reaching individuals, but we don’t really have a strategy. And the tobacco companies fight back: they compete with each other; they do have a strategy. I would like you to think for a moment about what the consumer sees in the community, what the person at home sees.

First of all, they look at TV and see sporting events such as car races with blatant cigarette advertisements. On late night talk shows, guests frequently smoke. When they go to the movies, (and even the movies on TV), cigarettes are there. They’re on the tables. It may not be appropriate for the character, yet cigarettes are there. In the supermarkets, they’re up front at the checkout counter. At the gas station, cigarettes are there as well as billboards and posters.

And when you look at minority communities, cigarettes are even more apparent. The magazines are very colourful, but the back pages of practically every magazine feature cigarette ads. Also, minority newspapers promote cigarettes. I opened one of the minority newspapers recently and there was a two-page spread by a cigarette company, which was giving small grants ($2-5000) recognising people from community organisations. Most of the organisations are little known, but, nevertheless, cigarette companies are there, and they are part of the community. Cigarette companies give to charities. They don’t require research protocols and evaluation. All one has to do is have 501© IRS status, and they will give you money. That’s all that’s necessary.

And when you go into the low income areas, that’s where the billboards are. They advertise in restaurants and provide ashtrays and napkins. The convenience stores have posters in them. The communities are blanketed. Cigarette companies have a marketing plan. And then, once every six months, the smoker goes to the doctor and the doctor spends five or ten minutes and tells the patient that he or she should stop smoking. What is the impact? Think about it from the consumer’s stand-point.

As far as providers are concerned, the National Cancer Institute (NCI) has been conducting smoking cessation training for professionals. Recently, in the literature, I have seen much more about smoking cessation than I’ve ever seen before. I think people in all professions are becoming aware that they need to talk about smoking. I think we are reaching out, not only through the NCI’s nationwide programme, but also, through the NHLBI programme, to physicians, dentists and nurses. The ASSIST programme is going to be out in the community as well.

We’ve heard a lot about pharmacies, and I’d like to just take a moment to tell you about the real world of the pharmacy. As president of our state medical society, I was on a radio programme with a Washington, DC, pharmacist who said that the pharmacies are really being squeezed by a number of factors. First of all, pharmacies don’t exist in many of these communities we need to reach. Some are at supermarkets, but many people have no access to a pharmacist. It’s very important that pharmacists know about smoking cessation, but many of them don’t really come into contact with disadvantaged patients. Large numbers of people, middle income and low income, never talk to a pharmacist. Certainly, pharmacists need to be involved, but I don’t think you can count on their involvement as the sole answer.

I participated in a roundtable sponsored by the FDA, the Non-Prescription Drug Manufacturers Association, and the Consumer Product Safety Administration. They also talked about using the pharmacist. In many communities, people buy their OTC drugs from the local corner store, where probably the person who sells the product is behind a glass wall with cigarettes prominently displayed. I think we need to be realistic about what’s really going on in the world of the people we are trying to reach.

I’ve approached some of the Medical Board members about asking smoking cessation. questions. Professional schools should include smoking issues in the curriculum. We can put questions on Boards, but the subject needs to be in the textbooks. There needs to be a strategy beyond just putting it on the Boards.

At worksites, we talk about blue collar workers and the Unions. What about the pink collar workers, the people in the barber shops and the beauty parlors? That’s where a lot of the smoking goes on. As for the Unions, the tobacco industry is already there. Unions are the ones who lobby the DC City Council to get the smokers’ rights bills passed. They’re already there. It doesn’t mean that we don’t need to be there also, but be aware that they’re there before us.

I think adolescents are important. Dr Schydlower is here from the American Academy of Pediatrics, which cooperated with the NCI on the ‘Fifth A’, as we call it, which is to Anticipate. I think adolescence is late to attack smoking. The pediatrician, and those who see children, need to start with passive smoking exposure of newborns and young infants and the Fifth A. In our programme for Preventing the Onset of Tobacco Use, we start with a much younger age.

We talk about comprehensive: I listened to a number of speakers and I understand comprehensive, but I wonder what it means to a person who’s in the community? It doesn’t necessarily mean the same thing that it means to us.

I was director of the High Risk Young People’s Project for individuals 15 to 24 years of age in the District of Columbia. Everybody said, ‘You’ll never get young men to come’. We had more of them coming than any other group, but the reason was we had a comprehensive programme that meant something to them. We had sports physicals, we had GED training, we have employability training, things that meant something to them. So, when you’re going into the community, think about what’s comprehensive for the individual who’s receiving the services, not just for those who are delivering them.

It’s important that we, as educated professionals, get out of our offices and into the community. We cannot make decisions sitting in a vacuum-like office if we haven’t been out there. Believe me, people out there welcome us. There are not only the voluntary agencies to which we relate, like our own kind of voluntary agencies, but also almost 600 organisations that belong to the Independent Sector.

There’s another world out there, just as there is in the scientific world – people who are volunteering, who would welcome your input, who have access to smokers, and who can carry the smoking cessation message. I encourage you to go out there and take the message.

Jack Hollis

I want to offer a somewhat more sober message and maybe a bit of a wake-up call. We heard today that only about half of smokers report that they have ever received advice from a physician to stop smoking. Yesterday, Gary Giovino told us that, of patients who had seen a physician in the last year, only 36% reported that they received advice to stop at that visit. Even the most powerful intervention is going to be absolutely useless if it does not get used, and we need to become much more realistic about what we are asking to have happen in primary care settings.

Together with my colleagues, Vic Stevens, Thomas Vogt, and others at the Kaiser Permanente Center for Health Research, and Ed Lichtenstein at the Oregon Research Institute, we tried to come up with a more practical model. It had to be an approach that would be easier to implement and sustain in a busy, real life, primary care setting. We came up with a model that we thought would be both effective and practical and tested it in a randomised trial sponsored by the NCI. I would like to briefly share it with you.

We tested a clinic-based team approach for dealing with the tobacco issue in a way that does not put the bulk of the burden on the physician. As a first step, clinic assistants, medical assistants, or whoever puts the patient in the examination room assesses smoking status along with other vital signs. If the patient smokes, then the clinic assistant alerts the physician by flagging the chart. I am more optimistic than Ron is about tobacco assessment becoming a new vital sign. It is a key step and we have shown that it is possible to make this a routine part of practice.

In our team approach, the physician’s role is to deliver a cessation message that takes only about 30 seconds. Delivered at the end of the visit, it goes something like this: ‘Stopping smoking is the single most important thing you can do to protect your health, and as your physician, I want to strongly advise you to stop soon. I know that has got to be your choice, but I believe you can do it when you’re ready to try. So how about it? Are you thinking about quitting sometime in the next six months ? ’ If the answer is ‘ no, ’ the physician hands them a pamphlet, such as the one we call ‘Getting Ready to Get Ready to Quit Smoking’ and checks with them again at their next visit.

If the answer is, ‘Well, I am thinking about quitting,’ then the patient is referred to a trained nurse. The nurse first starts the patient on a 9-minute video. The video dramatically decreases the nurse’s task by modeling techniques other people have used to quit and dealing with many of the common concerns of smokers, such as fears about past failures and weight gain. The tone is upbeat, with a good deal of humour. At the end of the video, the nurse provides a self-help manual and tries to elicit a quit date. ‘So how about it? Are you seriously thinking about quitting in the next few weeks? Is there a specific date you think would work for you?’ Typically, the interaction with the nurse at the end of the video takes three to five minutes. If the patient sets a quit date, then we recommend that someone call them back the day after their quit date to see how they are doing. We have a staff member from Health Education make these brief calls. We also mailed smokers newsletters and other cessation reminders.

When we implemented this in two large Kaiser Permanente medical facilities, we found that all 60 of the medical care providers were happy to participate because it took so little of their time. Physician training required about one hour, a free lunch, and some in-clinic follow-up. Of all identified eligible smokers, 86% received advice over the one-year Intervention period, which tells us that physicians found this really worked well for them. Of the patients who received the advice, 87% went and saw the nurse counsellor as directed. So the patients also saw this as a reasonable clinic procedure.

We had four groups in this randomised trial. I will not describe all of them here, but our control group included a 30-second cessation message from the physician and the ‘Why Do You Smoke?’ pamphlet. Our three nurse- assisted treatment groups also included the brief physician message but added various nurse-delivered interventions. Compared to brief advice alone, the three nurse-assisted interventions significantly increased the percent of patients who set quit dates, attended a group programme, and made serious quit attempts.

We did not have a no-advice or usual-care condition, but from the literature I would expect about a 1-2% long-term sustained quit rate. In our case, we required the patient to be abstinent at both three- and twelve-month follow-ups to be considered a success. As described in the May 1993 issue of the Annals of Internal Medicine, the brief physician advice message yielded a 3.9% sustained quit rate, which is substantially better than usual care. But the nurse-assisted treatments yielded almost double that quit rate, 7.2%.

We have done preliminary cost-effectiveness analyses using national mortality rate data. Depending on which group we are talking about, the nurse-assisted treatments cost between $17 and $25 to deliver. The added costs needed to save one discounted year of life by adding nurse-assisted intervention to brief physician advice is roughly $1000. To put this in perspective, it takes roughly $11000 to save one discounted year of life with Propranolol treatment for mild to moderate hypertension and $102 000 to save a year of life with mammography screening in women aged 55 to 65 years. Clearly, nurse-assisted tobacco counselling is one of the most cost-effective medical procedures available.

We are now trying to implement this nurse- assisted system in different regions of Kaiser Permanente and other settings to test generalis- ability. We have modified the programme slightly to include stage-appropriate prompts at every visit that provide somewhat different messages to pre-contemplators, contempla- tors, and recent quitters. We have also built in a step-care protocol so that progressively more intensive interventions are provided at successive routine visits. For example, we encourage people to try to quit once or twice on their own, using self-help materials. Those who need more intensive help are referred to a group programme with nicotine replacement, if that seems to be appropriate.

As a final comment, we have learned that an absolutely critical part of any programme implementation effort is to have some mechanism for staff monitoring and feedback. During the early stages of implementation it is vital to track and provide feedback on whether the various components are being delivered or not. We have seen time and time again that what gets measured, gets done. Nurses are an important resource in the fight against tobacco and I encourage all of you to look for creative ways to involve them more actively in the process of helping smokers quit.

Marc Manley

Under the able assistance of Dr Tom Glynn, a series of trials were funded through the NCI looking at what health professionals could do in their practices to help people stop smoking. As those trials were wrapping up in the late 1980s, NCI, with the investigators, managed the production of a document that summarised the findings and gave some consensus recommendations. That book, called ‘How to Help Your Patients Stop Smoking,’ is one that I hope many of you have seen.

There was a study in the literature at that time that came from the UK. They had mailed books similar to ours to every General Practitioner (GP) in the UK and followed up with a survey of the impact of that mailing on GPs’ knowledge about smoking cessation. As you might guess, they had very dismal results, and it was very fortunate that we had that piece of information at the time or we might have wasted a lot of money on postage and books. We elected not to do mass mailings of instructional booklets for physicians and other clinicians and instead developed a set of training materials that would, in many ways, repeat the kinds of training programme that had gone on during the trials in the 1980s.

For the last two and a half years, we’ve been doing everything we can to get courses in smoking cessation taught to physicians, nurses, dentists, dental hygienists and others. And we’ve used what’s frequently called a train- the-trainer system, which was really the idea of Dr Roselyn Epps. With her leadership we’ve worked with a number of organisations who have the ability to bring clinicians to these programmes. I’d like to point out in particular the American Cancer Society and the American Medical Association as two groups that joined our programme early on and wanted to offer this kind of training service to their members and delivered it in many places around the country.

We originally planned to train 2000 people how to teach courses in smoking cessation techniques. We’re up to 1600 right now, so we’re getting near the end of that phase.

It’s been, in many ways, a volunteer programme. The NCI puts a lot of staff time into it, but it’s really been these organisations buying into it and individuals within these organisations being willing to come and attend our trainers’ workshops and then go out and do some teaching.

Clearly, this is not only the thing going on in smoking cessation training, nor should it be. This is a programme that’s meant to be a catalyst and to encourage clinicians around the country to learn more about it. As we’ve heard today, there’s reason to have more than just one programme in the country.

If you make the assumption that health care providers are human beings and follow the laws of human behaviour, then many of these issues that were brought up yesterday are applicable to changing health providers’ behaviour. When you sit in a room full of physicians, for instance, and teach them about smoking cessation, you often find that there are people in the room who are pre-contemplators and have no intention of ever changing the way they practice or the way they treat smokers. There are also people who are thinking about it a little bit, and there are people in action phases who are doing something about it but also need some help with maintenance. So really those same stages of change apply to clinicians as well.

To put it in another way, there’s also a big difference in the level of skills of different health care providers. Some people are quite comfortable counselling and some people want very little to do with counselling, so there is a big range of skills and attitudes among health care providers. It’s a tough thing to change their behaviour, especially in a group session, but the way medical and dental education works, that’s often the only option we have.

In the 4 ‘A’s approach, if they ask, advise, assist and arrange, it gives them an outline of what to do when they’re face to face with a patient. The nice thing about the 4 As is it makes it easy to convince these pre- contemplator clinicians: ‘You’ve only got to remember these four words and it will only take a couple of minutes. ’ That’s true to some degree, but everybody knows that there are some patients who require much more than two or three minutes, and it’s only an outline of an approach. So when you’re dealing with a group of clinicians, you’re selling some people on the idea all together, but some people are ready for more than just a very basic approach to the whole issue.

Our 4 As is only one step in five steps of intervention, and the other steps have to do with setting up an office practice so that this kind of intervention is given routinely. I do think there’s good evidence that repeated routine delivery of these services and offering of services is essential. There’s also good reason to believe that if you don’t have some support within your practice environment, you’re not likely to do it and remember to do it on your own as a busy clinician.

So we need to continue to emphasize the importance of an office systems approach and of having staff members in an office setting who are a part of the programme and know their roles and responsibilities. We need to continue to encourage the use of reminder systems, for example, the use of making smoking status a vital sign. I, too, share what may be a hopelessly naively optimistic view that it can be a vital sign, but I think it’s a small step and an important one.

Other things that might motivate clinicians to change the way they treat smokers may go a bit beyond just what’s involved in their immediate practice environment. When we wrote the NCI manual, our model was a private practice where people sit down and decide this was something that they want to do. But many practice settings are different to that. There are large managed-care systems where standards of practice and rules of practice may be imposed a bit from the outside, and, in fact, that may be the wave of the future.

I think payment is also a motivating factor. It may not be people’s first motivating factor, but I think it’s an important issue, and so reimbursement clearly plays a critical role in all this.

The title of this conference is Who Quits? Who Pays ? but I think in between there might be the words ‘Who Treats?’ Who quits and who treats and who pays for the treatment- these are the important questions.

We’ve heard a lot about worksite cessation programmes and other kinds of group programmes, and I really think primary care providers and other clinicians, for that matter, have this service to lose, in a sense. They could grab a hold of this and say, ‘This is our domain, we are going to do this, and we are going to expect reimbursement; ’ or they may decide to say, ‘This really isn’t our ballgame and it’s better referred to a group programme. ’ I think that’s happening in certain situations, so it’s still the clinicians’ decision about how they’re going to proceed.

I know that certain Health Maintenance Organisations are now putting restrictions on how patients can receive, for instance, nicotine patches; patients have been told they have to attend a group programme in order to receive their prescription for patches. Many clinicians have complained to me that this takes them out of the loop, that they write the prescription for the patch, they refer, and that’s the end of their treatment. I think we need to look at those decisions very carefully because there may be better ways to include the clinicians in this.

The other thing that may encourage clinicians to do interventions more is their training, and not just their postgraduate training, but their medical school training and their residency training or other professional training. If we don’t find a way to build it in, we’ll continue to be moving a rock uphill, and I think we have to focus on schools now.

Also, and I think it’s demonstrated by the recent attention that nicotine patches got from the general public, patients dictate a great deal of what goes on in practice and what patients ask for is often what gets discussed in a clinical intervention. We have to think about patients asking for these kinds of services, and I certainly know that happened when patches were marketed so widely. No matter what you think about direct-to-patient advertising, this got more people into their physicians’ office asking about smoking cessation and probably had at least some positive benefit.

This leads me to the conclusion that this isn’t just a clinical issue; this is a much broader societal issue, as has been said. As we see more programmes such as mass media programmes or increased cigarette taxes, we’re giving patients more reason to want to stop smoking, and that’s going to, if we do our job well, drive them into the medical system asking for help.

So I think these broader political and economic issues play an important role in the clinical setting and that we need to continue to push what I call comprehensive tobacco control programmes, which include a whole range of interventions, including political, social, and economic interventions.

Deborah J Ossip-Klein

One point that has come out of this conference very clearly is that we have a range of interventions that work. We certainly need continuing research to upgrade and further enhance the effectiveness of our interventions. Simultaneously, we need delivery systems to get what works out now so that our interventions can have a public health impact. The two types of systems that I’ll be talking about will be self-help or self-directed delivery systems, and these will be telephone hotlines and also interactive systems to which John alluded a little while ago.

At one end of the continuum, we have the telephone hotlines, which have a number of advantages. They can be available to whole communities and thus provide assistance right when it’s needed. They can offer personal contact, but without the kind of face-to-face interaction that may deter many would-be quitters from seeking other sorts of formal services.

We know now from a number of years of experience that somewhere between 1% and 36% of the target population will call a hotline.¹ What that means is that, in order to be viable, hotlines must reach a very large population base, and, in fact, this is exactly what we want them to do from a public health perspective.

We know that call rates will improve with aggressive, ongoing promotion of the hot-line.^2,3 Hotlines only seem to be relevant to smokers when they’re at a point when they need help, so if they heard something about the hotline two or three weeks ago, that may not have had any impact on them. If they need some help right now and they see a message about the hotline, then they’re more likely to call.

We also know that, when implemented correctly, hotlines can increase success rates, particularly for self-quitters, by somewhere between 19-47%. We know now that there are some things that won’t work with hotlines, and we also have an idea of the kinds of things that can improve their effectiveness. My group, and also, Tracy Orleans and her group, have demonstrated the effectiveness of both reactive and proactive telephone hotlines.^1,4

At the other end of the continuum we have the full interactive computerized systems for self-help and self-directed smoking intervention. Some of the components that might be included in these systems would be hotlines, personalised assessments and feedback, personalised intervention materials sent on a predetermined schedule, tracking of patients, generation of databases, and providing guidance to physicians and health care systems for intervention.

Essentially, what these kinds of systems can do is to take existing self-help interventions, and put them into a personalised interactive system that can get out to the public. A few variations of this approach are currently in existence, and there are several others in development. Notably, work by Prochaska and colleagues demonstrates the effectiveness of this type of system.^5,6 There are a number of advantages of these systems. First, they can also reach large populations and may provide viable self-help alternatives to the face-to-face clinics because they include many of the same components but do so in a self-help modality. Second, they may be able to help health care providers be more effective with their patients within the time constraints that they have for intervention. Third, these sorts of systems can continually be modified to incorporate Intervention components that are shown to be effective in controlled clinical trials.

Also, we can set standards for evaluating these kinds of systems in terms of their effectiveness. I’d like to suggest a couple of possible standards. First, have they been shown to be effective in controlled clinical trials, our real gold standards, and here’s where the funding agencies become increasingly important, and/or do they incorporate components that have been shown to be effective in controlled clinical trials, which may get them out a little more quickly.

And second, are they in conformance with national guidelines, and there are a couple of national guidelines that I think would be particularly relevant here: the NCI guidelines for physician intervention⁷ and the NCI guidelines for self-help intervention.⁸

Finally, there are some research needs in this area, and I’ll identify two. One is how can we best interface these kinds of systems with the existing or changing health care system to minimise the time requirements for health care providers yet maximise successful outcomes so that these systems become realistic and can have an impact. The other is how do we best mete out our contacts, telephone and/or mail contacts, so that we produce optimal outcomes. We may find that different dosing is appropriate for different populations, eg, high medical risk populations, populations that differ by age, populations at different stages of change, and so on.

In summary, what we see is an expansion of the concept of self-help, self-directed programmes with a couple of modalities listed here that can take the programmes that are shown to be effective in controlled clinical trials, and put them into delivery systems that get them out to the public in a meaningful way.

Nancy A Rigotti

I’d like to address the general problem of how those of us who work in the health care setting incorporate smoking cessation in our practices. It seems to me that we have three questions to answer. First, what should we be doing? Once we figure that out, how can we persuade and train practitioners to do it (and who are those people)? Finally, how can we make it happen in the organisations and institutions in which we work? I’ve been facing all these questions in the work that I’ve been trying to do at Massachusetts General Hospital in Boston, where I’ve been developing a comprehensive smoking treatment programme.

Let me reflect on the questions I’ve raised. First what should we be doing? I’m very impressed with what I’ve heard during the conference. We have good models that can be used at different levels in a stepped-care approach to treating smoking. There’s been a great deal of focus on the primary care provider, based on the rationale that 70% of smokers see a primary care doctor each year. I take issue with this 70% statistic because smokers are asked in surveys whether they have seen a doctor in the last year, not whether they have seen a provider whom they identify as their primary doctor. If a smoker has been to an emergency room to have a laceration treated, I’m not sure that we can expect that the doctor may have thought to ask about smoking while he or she stitched the patient’s wound. On the other hand, Neal Benowitz suggested that we should think about smoking like we think about hypertension. If you can’t see a doctor or go to an emergency room without having your blood pressure checked, regardless of why you’re there, perhaps we should treat smoking the same way. Perhaps we should adopt Mike Fiore’s idea and have asking about smoking be the fourth vital sign – something that happens every time a patient enters the health care setting. It’s a good idea, but it certainly isn’t how most of us in the medical business think about it at the moment.

It makes sense to focus on primary care providers. We’ve heard about the NCI models, which have been very helpful. I’ve taught them for years and see them as a very good scaffolding upon which individuals develop their own style of smoking counselling. My experience in teaching groups of physicians through the American College of Physicians and other organisations is encouraging. Over the years the proportion of grey hair on the heads of the people in the audience has increased, which I take to mean that the physicians interested in treating smoking are getting older. Originally, smoking was something that the younger people in medicine were interested in. Gradually it’s diffusing into older groups, and I think that that’s a positive sign. Ten years ago, the notion that doctors ought to talk about smoking was considered on the fringe of medical practice. Now it’s considered almost standard. Despite this progress, my impression is that too many of my colleagues consider that smoking is something someone should be addressing, but they’d prefer to avoid dealing with it themselves. Adopting the traditional model, they’d like to refer the problem to the appropriate specialist. As a result, I have reluctantly tempered my idealism that doctors can be taught a simple four-step model that will take three or four minutes. I realise that some physicians will never be interested in doing this, and some who are interested are not going to be very good at it. There are many practitioners who are or can become excellent smoking counsellors, but there are many who will not. This is why we will need flexible, stepped-care models of smoking treatment for the health care setting.

To explain the reluctance of some, and to put smoking counselling in the perspective of a primary care practitioner, let me give you a list of things that the US Preventive Services Task Force advises a physician seeing a new patient to do in the 20,30 or 40 minutes allotted to that appointment. He or she is not only supposed to talk about tobacco, which is very important, but about many other issues. For example, cholesterol: has the patient ever had his or her cholesterol measured? What kind of diet does the patient follow? What is his or her weight? If it’s a young woman, might she have an eating disorder? What about the level of physical activity? Does the patient use seat belts or have a gun at home? What about alcohol use or other substance use? We need to ask about sexual practices and sexual behaviour. Have they ever been tested for HIV? Are there any risk factors for AIDS? Does the patient know what the risk factors for AIDS are? If it’s a woman who is premenopausal, what does she do about contraception? If a woman is menopausal, has she thought about oestrogen replacement therapy? That is not a quick discussion. What about immunisations ? Then we get to the cancer screening issues: pap smears, mammograms and breast self-examination. Do you know how to do it ? Often we need to teach them how. We now have newer biomarkers for cancer, like prostate- specific antigen (PSA) for men, or CA125, a potential screening test for ovarian cancer in women. The American Cancer Society has now recommended that men over the age of 50 be screened with PSA, so the appropriateness of this test needs to be covered. This is not a single simple discussion because it requires talking to patients about the concept of positive predictive value, ie, whether a positive test is more likely to be a true positive or a false positive.

Even though I care about doing smoking counselling more than many of my colleagues and probably do more of it, there are many other issues that I have to deal with in practice. I’m very persuaded about the value of the work that Jack Hollis described today. This system would limit what I would be expected to do and connect me into a coherent system that would take care of it for me. I think that’s the way we’re going in primary care practice.

As a result of the work of Jack Hollis and others, we have good models for handling the smoker in primary care. Even with the best model, there are going to be failures, and then you need an approach for smokers who need more intensive intervention. First, who will do this work? Administratively, will the practitioners be based in a Department of Medicine, a Department of Preventive Medicine (if there is one), or in a Department of Psychiatry or Psychology? Should it be the domain of substance abuse treatment? It’s not so clear. What is clear is that the content is multidisciplinary. There are now some very nice models being developed for this more intensive level of treatment. I’d like to acknowledge Richard Hurt and his colleagues at the Mayo Clinic, who have developed an excellent model of how to use the whole health care setting. Other institutions around the country are also developing these models: Duke, the University of Wisconsin, and Kaiser-Permanente Hospitals in Portland, Oregon.

Another approach to using the health care setting to promote smoking cessation would be to ask what are the teachable moments ? There has been some work on that, which we haven’t talked much about here. For example, we know that one of the best ways to quit smoking in this country is to have your first myocardial infarction (MI). About one-third of the people who have their first MI don’t smoke again without any specific anti-smoking intervention. However, (1) they’re frightened to realise that smoking is not only bad for smokers, it’s bad for them personally, and (2) they are not allowed to smoke for the duration of their hospital stay – about a week nowadays. These effectively serve as an anti-smoking intervention.

We recently investigated smoking after coronary bypass surgery and found that 50% of people who undergo these operations don’t smoke for five years. That’s good news, in a way. On the other hand, these people have been through a $20-30 000 procedure whose outcome is improved if they quit smoking, and 50% still go back to smoking. Suddenly, the outcome looks less positive and indicates a need for change. Unfortunately, we found out that early in-hospital intervention with these people does not increase their quit rates. New approaches need to be developed and tested to maximise medical outcomes and make these procedures more cost-effective.

There are other teachable moments that might be considered: pregnancy or just being admitted to hospital, since hospitals are now almost uniformly smoke-free. We don’t really know how much nicotine withdrawal smokers have when they’re admitted. In practice, we don’t seem to see a lot of it, but it’s possible that withdrawal is occurring and is not recognised as such, since withdrawal symptoms are nonspecific. Regardless, I think that we need to look at ways to develop smoking interventions in hospitals.

Even though we don’t have fully developed models of how best to promote smoking cessation in the health care setting, we have developed a number of good models. Ultimately, how do we get what we have learned adopted widely in our hospitals and health care facilities? How can it get into the consciousness of the reimbursers as an activity that should be covered? Reimbursement policies are determining how smoking cessation is being defined and delivered. Consider nicotine replacement. Its method and duration of use is being defined not by the FDA, not by package inserts, not by physicians, not by the behavioural scientists who probably know better than anybody what a comprehensive behavioural programme is, but by third-party reimbursers. As an example, consider the situation in Massachusetts, where my colleagues and I are trying to set up a comprehensive smoking cessation service at Massachusetts General Hospital, modelled on what Richard Hurt and his colleagues have done in Minnesota. In doing so, we investigated insurance coverage for the nicotine patch. What we found was that nicotine patches are covered, not beforehand but after the fact. There is reimbursement after the purchase of the patch for one course of treatment in a smoker’s lifetime, but only after they show evidence of having attended a smoking cessation programme, which is usually defined as three sessions of four hours, and includes certain types of behavioural modification methods. Insurers specifically exclude counselling in a physician’s office as qualifying for nicotine patch reimbursement. Furthermore, they don’t pay for the counselling programme they require. The smoker has the choice of paying for the patches or for the programme. I submit that this is the reimbursement policy I would develop if I wanted to minimise use of a medication.

So, in summary, I think we need to solve some of the organisational issues first. What might be helpful to people like me would be a way of collecting the information about model programmes and strategies that have been developed around the country. Whether in a book or a conference, that information would be useful for people who’d like to start programmes and, I hope, organisations who might consider paying for them.

Questions and answers

ELLEN GRITZ: Much of the central work in developing the theory, content, and delivery modalities for smoking cessation services has been undertaken primarily by behavioural scientists and psychologists. I wonder what anyone on this panel, perhaps Jack, might comment on how to broaden the mandate, and the ability to treat and deliver patients using tobacco-dependence diagnoses, given Nancy’s comments. I would also like to ask what the rest of us have heard about the difficulty of even establishing the treatment in a medical care setting, much less allowing psychologists and other mental health professionals to treat legitimately patients with smoking-dependence disorders.

JACK HOLLIS: Working in a Health Maintenance Organisation (HMO) or a large medical care system, I think psychologists and others have a really important role to play in organising those systems. I don’t think that they, at least at the primary care level, are the appropriate ones to be there in the clinic providing the services, although they probably have an important role to play as referral agents, running group programmes and more intensive kinds of programmes like those that have been mentioned several times.

ROBERT ROBINSON: Ron, I would be interested in knowing in terms of the mass media campaign that you developed in Michigan how you handled competing interests, since it was defined as health promotion. I’m thinking of AIDS and infant mortality – serious competitive issues in terms of those dollars – and whether or not it was an issue ?

I was struck by your comments about the minimal role of the public health system. There’s an infrastructure there and they access high-risk populations, but what I heard that wasn’t said was that the infrastructure is minimal, so that one questions exactly how one can utilise it. This brings me to the need for capacity building. When you talked about the needs of communities, you listed at least three research initiatives. I would like to invert that. I think what we need first in communities, is capacity building and infrastructure development. Once we lay a foundation there, it would be efficacious to go in and develop some research protocols to evaluate the efficacy of what we have. But at the moment we have nothing, and if all we offer is research that implicitly has no service component or no capacity building component to it, I think we’re backing in with our behinds rather than with our heads. What I would propose is that we really look at the need for capacity development and infrastructure development in communities, and then follow up with some research.

RONALD M DAVIS: I agree with that. I think there’s sometimes a tendency to over-evaluate, although, certainly, the greater problem has been underevaluation. But I agree with you about inverting the priorities for delivering these things in the community.

Even though the involvement of local public health departments or the public health system has been small, I think their potential role is great. When you consider the fact that half of the vaccinations delivered in this country are delivered by local health departments, which primarily reach the indigent population, you can see the potential for local health department involvement in public health. They’re just not doing it in terms of health behaviours and chronic disease prevention.

In terms of how we’ve allocated mass media dollars to tobacco, out of the nine million dollars raised from the computer software tax, two million was allocated for the media, one million for AIDS media messages and up to a million for tobacco, although in the last year or so we’ve gone down from a million to about 600000 for tobacco because we’ve taken the rest for breast and cervical cancer screening and an anti-violence campaign. The competition has actually been more acutely felt in the last year or so.

To answer Ellen’s question-my understanding of the reimbursement system is that, historically, the greatest priority has been to reimburse physicians for the services they deliver, or at least they get the first reimbursement before other health care providers get reimbursed. I would think in smoking cessation that one way in which psychologists and others could be reimbursed would be to do it, ‘under the supervision of a physician’.

For example, in family planning clinics, a lot of nurse-practitioners get reimbursed and can prescribe oral contraceptives if it’s done ‘under the direction of a physician’, which can be done quite flexibly.

But the other way in which I think it could be facilitated would be through certification, a point I made earlier. If we certify psychologists or any other professionals to do cessation therapy, I think insurance would be much more likely to cover it rather than opening it up to a long list of providers.

CAROLE TRACY ORLEANS: I’m hearing a theme emerging from today’s presentations and yesterday’s, which is basically that we need to develop smart systems, not only for the patients, but for the providers and the delivery systems. The NCI panel, Tom Glynn, Jessie Gruman, and Gail Boyd wrote that we have to make smoking cessation cues persistent and inescapable. I think we have to make smoking cessation or nicotine addiction treatment persistent and inescapable, and we’ve got an enormous amount to learn quickly about how to use the managed care environment creatively: to build chart audits; to put nico- tine-dependence treatment into the routine practice of primary care or to provide forms and materials that make those kinds of chart audits easier, and to start reimbursing physicians by whether or not they do that; to build simple health risk assessments (HRAs) to relieve the overburdened primary care physician who’s trying to take preventive medicine seriously by using the increasing capabilities we have in computer-based expert information systems to give HRA feedback in a cost- effective way in the primary care setting.

The other comment I wanted to make is that the pharmaceutical sales representative’s detail visit provides a wonderful model of how to train physicians to do just what we’re talking about: brief interventions at the office that will get the attention of and maybe get the patient started on a more extended intervention. We ought to be in a partnership with the pharmaceutical companies marketing nicotine replacement therapies to start to infuse the ‘Four A’s’ in what they’re doing. The opportunities for follow-up through the pharmaceutical detail strategy are impressive.

RICHARD HURT: I want to answer Nancy’s question about reimbursement, but before I do that, I want to give you just a brief summary of what we’ve done. Nancy’s experiencing exactly what we experienced in 1985, which was a lot of resistance and a lot of inertia, in trying to bring a nicotine-dependence treatment programme to a major medical centre. We were able to do that after a three-year planning effort, and in 1988 began our consultation services, which have now broadened to provide the full range of treatment services, including the stepped-care model we talked about yesterday. We have provided services for over 7400 patients to date, and have provided training for people from various medical centres in the country, including Indiana University and Henry Ford, among others, that have come to see what we’re doing.

To answer the reimbursement question, we surveyed 91 patients who had been seen for a consulation in 1991. The business office at our place does routine surveys of patients to find out what reimbursement was available. We had 19 different insurance companies that were represented, 10 of which covered services and four covered 80 to 100% of the services, so that’s the answer to your question.

CYNTHIA RAND: We’ve had a lot of discussion about physician advice in the office with patients, and we’re aware of the statistics that show the number of patients that report that they haven’t received that advice. Those of us who work in the area know that sometimes patients get it but don’t hear it, and there is a question of what it is they do hear. I want to share one of our observations when we took people into the Lung Health Study, all these middle-aged folks with pulmonary impairment. One of the questions we asked them when we interviewed them was ‘Has a doctor ever told you you had to quit smoking ? ’ A very common response from all these folks with shortness of breath and pulmonary impairment was no, but if you probed a little bit further, they’d say, well, he didn’t say I had to quit smoking.

I think that we need to understand that what people hear may be very different from what we think we’re saying. I have worked with many residents at Johns Hopkins and have listened to the type of advice they give. They believe they’re giving smoking cessation advice, but what they’re giving are generic messages, like on the cigarette packet. ‘You realise smoking is very bad for you.’ ‘You realise that smoking causes cancer.’ ‘You realise...’ People are innundated with these generic messages.

I don’t think there are smokers out there that don’t know in some amorphous way that smoking is bad for them. They do not need to hear their physician reiterate the generic message. The role of the physician is to make it personal and direct. When a physician takes the time to connect with a patient and say ‘You are at risk, I am concerned about you, ’ patients don’t walk out of the office saying my doctor didn’t mention smoking; they do that when the message is broad and based on smoking being bad. ‘Of course it’s bad, everyone knows that, but I’m not at risk. I’m not in danger. ’ So I would suggest if you’re working with physicians, training them how to deliver the message, be very clear on what it is they should say because they may not be saying anything very effective.

ROSELYN P EPPS: That is a part of the training programme, to personalise the message and not just say people shouldn’t smoke but give the reasons why the individual who’s visiting the physician shouldn’t smoke.

JOHN SLADE: I’d just like to point out the progress that’s being made, especially in Canada, with making warning labels very specific and detailed, with a brief message covering 25% of the pack, some smoke content information on the side, and then the opportunity to get a pamphlet that provides large amounts of information and potential opportunities for public health messages.