Fast track program reduced time-to-treatment initiation by 40%

Denmark implemented a national fast track system in 2008 that gave patients with head and neck cancers the highest priority in the Danish health care system.

Proponents hoped the program would allow patients to undergo surgery or radiation therapy sooner, reducing the risk of tumor progression and death.

The fast track program reduced by 4 weeks the length of time patients had to wait between their initial meeting with health care providers and their first treatment, according to findings of the Danish Head and Neck Cancer Group presented at the Multidisciplinary Head and Neck Cancer Symposium in January.

Barbara Burtness, MD, a medical oncologist at Fox Chase Cancer Center who specializes in head and neck cancers, led a round table discussion during the HemOnc Today section editors’ retreat to further explore the findings and their potential implications.

The round table discussion follows.

Dr. Burtness: Some years ago, there had been exposés in the Danish press about how patients with head and neck cancers were undergoing their workup, and it took a very long time for the patient to make it to definitive treatment in the single-payer system they had in Denmark.

This became a political issue in Denmark, and the Danish health care system instituted what they called the fast track program.

Under the fast track program, patients with head and neck cancers are meant to be
guaranteed a reasonable time course in their workup, diagnosis and time-to-treatment initiation, and they have reserved extra slots in head and neck clinics and made several other changes to the way they process these patients to allow that to occur.

Cai Grau, MD, DMSc, a professor of radiation oncology at Aarhus University Hospital in Aarhus, Denmark, and lead author of the study, presented data from 474 head and neck cancer patients who were treated either in 2002, before the implementation of the fast track system, or in 2010, 2 years after the system was implemented. The data showed the program reduced the time to initial treatment from 69 days to 41 days.

The investigators do not have any data that show whether this is changing outcomes for patients, although there are some data in the literature that suggest that long time-to-treatment initiation in head and neck cancers can be associated with upstaging and worse outcome.

There are a number of ways to look at this. First, this is a report from a single-payer system. Although their problems may have arisen because of having a single-payer system, they also were able to streamline their response to it. In the literature in Europe and in the United States, many reports about delay have to do with public hospitals. Some of the reports focus on the proportion of time and the delay that comes from the patient and the proportion of treatment delay that comes from the medical care setting.

In the United States, there are a couple new factors that introduce delay, even for patients who are not in public hospitals. The first is that, with patients who have HPV-associated cancer who present with big bulky cystic neck nodes, radiologists can mistake these for branchial cleft cysts. A young healthy man goes in with a big neck mass, gets a CT scan and hands a report to his primary care physician that says branchial cleft cyst. If the PCP doesn’t process that a man in his 40s or 50s shouldn’t be developing a new branchial cleft cyst, the result can be a multi-month delay before the diagnosis is made.

Also, we are doing very fancy reconstructions now for people with oral cavity cancers, but these are very long procedures. You have to have the plastic surgeons and the head and neck surgeons scheduled in the operating room at the same time, and that can
introduce delay.

So I thought it was impressive that they were able to get their time-to-treatment initiation down from 69 days to 41 days, and I wonder if we always achieve the 41 days in our system.

Increased resources

Joseph Aisner, MD: We have two kinds of single-payer systems in this country. One is the Veterans Affairs, and they are heavily burdened with head and neck cancers. The other is a place like the California systems, where they have large population bases. Do we have any idea what the timetable is in those?

Dr. Burtness: No.  I was unable to find anything about head and neck cancers in the VA system.  I do know that the VA system had been troubled with a longer than 90-day period to treatment initiation for lung tumors, and they have been trying to get their centers to focus on bringing that down, with some success.

Ralph Green, MD: It wasn’t clear to me how they achieved this, other than by focusing on it. There must have been some increase in resource allocation. The question is: To what extent did they have to increase those resources, and is there any message in terms of application to other tumors?

Dr. Burtness: They did increase resources. They developed a dedicated telephone hotline. They had reserved slots in the ENT and radiology departments. Dr. Grau actually commented on how they were staying late on days when extra head and neck cancer patients showed up. They had faster pathology reporting, so these cases were supposed to move to the head of the line in pathology departments. They conducted their tumor boards twice a week, while most of us are still doing it weekly. There was no kroner (currency of Denmark) figure associated with that, though, in their presentation.

Dr. Green: Were there any perceptible diminutions in other areas? When you shift allocation without any incremental input, you would suspect that perhaps patients received a lesser quality of care if they had a diagnosis other than head and neck cancer.

Dr. Burtness: My impression was that this had become kind of a political hot point, and extra resources were added into the system to deal with this.

Streamlined processes

Dr. Aisner: I wonder if an approach to this in terms of looking at its science would be to use what Dilts did for study design. That is, apply an engineering flow process to how the patients come into the system and go out of the system. He has done a number of spectacular reviews of what happens in the life of a protocol. He picked up a Dell computer and he said, “You bought this. You have a custom-made Dell. You got it in 2 days because everything is modular in the way you design it. You cut every screw.” He said that it’s a generic problem in medicine. We challenged him to get the NCI to do his process review. Where he has worked, and where people have paid attention to that, the timetable went from 680 days for the gestation protocol to about 200 days. So that kind of process — particularly for the single-payer systems, where the variables can be plotted very cleanly — should be strongly advocated.

Dr. Burtness: I agree. That would be very useful. It would be important to figure out whether the magnitude of the delay they’re getting rid of is meaningful. I imagine that, for many patients, it’s not meaningful, but the 69 days is partly driven by some outliers who take a lot longer, and getting rid of the long delays is quite likely to be meaningful. For HPV negative cancers, upstaging can occur with a 30-day delay. Finding out the proportion of patients in which there is a change in stage, or figuring out if there’s any difference in outcome, would be important.

Although these data come from a single-payer system, I brought them up because we have issues with getting patients through their workup and treatment initiation in head and neck cancers in our own system. As an example, the patient may be diagnosed in the primary care setting, and then has to wait for an appointment for the biopsy, then has to wait for an appointment with the specialist, and then he or she has to book the patient with a reconstructive surgeon. In some cases, the patient may be unhappy with the treatment plan and go to get a second opinion. Even in our system, patients are probably pushing 2 months in most cases.

Joseph R. Bertino, MD: I remember the days when you could hospitalize a patient, work them up in 3 days and get treatment started the next week.

Munir Ghesani, MD: We have patients who are coming from two systems. One is a PCP referred to ENT, and then ENT referring to us for imaging, followed by radiation/oncology. They all have learned the team approach. Right from the beginning, there will be a set distribution list. It’s already made. They just send it out by email. Within days, everything is worked up, and that’s on one side of the range.

The other side is a clinic patient who does not have a designated PCP. He or she goes to the clinic for one visit and, for the next referral, has to wait maybe 6 days later. Then the third clinic generates the referral for imaging.

In that case, sometimes we see a prescription written on a certain date and even get to the patient for scheduling 5 weeks later. So there’s a big paradox, and it’s so interesting that if we see where the patient came from, we already know how much delay there is. Of course, it’s not scientific, but we see the difference in the staging of the disease.

David H. Ilson, MD, PhD: At the more comprehensive cancer centers, you walk in the door and you see everyone the first day. You don’t have this fragmented approach that you see Dr. A., and then it takes another 2 weeks to see Dr. B. Increasingly, we’re seeing this multidisciplinary approach, and you sort of hit the ground running. The patient is seen and their treatment plan is set up from day 1. If they are not seen the same day, they get referrals with very fast turnaround time.

Another aspect, particularly for head and neck cancers and even esophagus cancer, is the use of induction therapy. If you’re going to do chemotherapy and radiation, if you start chemotherapy early, you can be planning radiation while the patient is already getting active treatment. 

Cause vs. diagnosis

Harry S. Jacob, MD, FRCPath(Hon): The concept of earlier diagnosis and treatment of cancers, which has been this golden idea forever, is under great attack. The argument being that it’s the biology of many tumors rather than the timing of the tumor that’s important. Underlying this study is the concept that the earlier the diagnosis and treatment, the better. It may not be. But it seems to me — especially in something like head and neck cancers — you have two major causes that I know about. One is tobacco and, possibly, alcohol-induced, and the other is probably sexual, a papillomavirus epidemic that probably is occurring. In the latter situation, it’s a viral kind of a thing, so you already have a natural difference in pathophysiology of what this tumor is going to do relative to the timing of diagnosis compared with, let’s say, the tobacco-induced cancer. Is there a difference in prognosis or outlook for those two kinds of tumors, or a suggestion that timing would be important for one but might not be for the other?

Dr. Burtness: They are very different cancers, and they have a very different prognosis with treatment. We know less about the difference in their natural history because we don’t have a lot of untreated HPV-associated cases. The cases that arise from tobacco and alcohol exposure have a high mutational burden. They’re very locally invasive, and they tend to grow fairly quickly. The cases that sparked action in Denmark involved those types of cancer, and they were seen to have changed stage during the waiting period. In that kind of head and neck cancer, stage is very closely associated with the likelihood of success with treatment.

The HPV-associated cancers arise in patients who are younger and healthier. They’re much more treatment-responsive, even at a higher stage. My guess would be that the treatment delay would be more serious for a patient with the traditional tobacco-associated type of cancer than for the HPV-associated cancers, but again, it would be a subject worthy of study because I don’t think that we really know that.

 Dr. Aisner: Another reason why it’s important to look at this in single-payer systems is that, depending on what community we’re in, about 10% to 20% of the population is no longer carrying much insurance. If we look at the head and neck population, it’s probably two times what we see in the base population. We’re seeing 12% or so uninsured in our place. One-quarter or more of the head and neck patients come in and we can’t even start on them until they get charity care cleared by the hospital. We can’t even do the first test to them, let alone a biopsy. So this often causes a month or 2 months delay before we can even start. Unfortunately, it’s a growing issue in this country. We have to look carefully at the kind of system that can look at this, straightened out for this factor. The VA is a marvelous place to do this kind of an analysis, as are any of the closed systems.

Treatment tolerance

Dr. Burtness: I wanted to go back to [Dr. Ilson’s] comment about the multidisciplinary team. I’m also at a comprehensive cancer center. That’s also how we function. But if you look at the patients’ itinerary before they get to us, a lot of them had a biopsy 30 days ago. We can hit the ground running and we can have all three of us see them on the same day, and speech has them by the end of the week, and so forth. But a fragmented system also introduces delays, and I just would be curious if the experience at Memorial Sloan-Kettering is similar to ours; that even among those patients who have a diagnosis, it’s a fair bit of time before they get into the system.

Dr. Ilson: A lot of the delay occurs before they come in — a delay in diagnosis and, of course, there are psychosocial issues with substance-abusing patients. But, of course, a key issue in treatment delay is that there’s just an impact on these patients’ abilities to tolerate treatment. There is nutritional compromise, there are swallowing problems. The longer you wait to treat them, the less likely they’re going to be able to tolerate intensive therapy. That’s a whole other aspect of aerodigestive cancers.

Dr. Aisner: That’s why looking at this in an engineering diagram would make a tremendous amount of sense to try to figure out where the delays are and what efficiencies are picked up by understanding where everything is going.

Dr. Burtness: As patients get upstaged, the treatment gets more and more morbid in head and neck cancer — 9 weeks of induction, 7 weeks of radiation, the high-dose platinum. Earlier-stage patients may get away with two modalities instead of three. I remember an uninsured patient I’m taking care of who was booked for an operation at another hospital and then charity care didn’t happen there. By the time she came to us, she had lost the opportunity to have a resection.

Psychological considerations

Dr. Jacob: What about the psychological aspects of this for somebody who thinks they might have — and possibly does — have cancer and is now being fiddled around with in a fragmented system and waiting 90 days or more to get some kind of definitive answer? I would think this would be a major reason to fast track this kind of diagnosis. Am I wrong about that? It’s a terrible time that people put in before they are definitively taken care of. Even with the possibility of better OS, I would think a quality-of-life analysis would be interesting.

Dr. Burtness: They did not discuss that at all in the Danish study. Some of the US literature makes the point that the delay can originate with the patient. Coming to grips with the diagnosis, and with the morbidity of the treatment plan, can certainly be difficult for some patients.

Dr. Aisner:  That’s a good point, but there’s a second side to this sword. It’s exemplified in some of what’s happening in breast cancer today, where there’s a demand to have the answer to the biopsy the same day. The problem is, when you push the system that fast, it makes mistakes. When you start to push it too hard, that’s when it frays.

A. Koneti Rao, MD: No matter who comes in, some of these problems are still relevant. No matter what the diagnosis is, if there are multiple subspecialties involved, there’s inordinate delay. It’s fragmented and it’s probably true in most places that you can’t render care that requires multiple groups to be involved in an efficient manner.

Dr. Green: The definitive diagnosis ultimately comes from the pathology. There is a necessity in most systems, certainly in ours, with a first-time diagnosis of cancer to get a concurrent opinion by another pathologist. That part is not that difficult. In teaching institutions, there is the component that the trainees, the house staff and, in our case, the pathology residents review the cases. There’s constant attention paid to accelerating the process but without skipping the steps. Depending, of course, on the tumor, there’s always the question of obtaining concurrence from, say, fine-needle aspiration to actual excision, from frozen section to definitive H&E, and these steps — depending on the type of tumor — are important.

Then the patient component comes in — the denial process, and their sense that maybe they should get a second opinion. When you add up all these components, just as far as the pathology part of the process is concerned, that’s something that is considerable, and there are numerous little mini-steps in the process, and I’m sure the same applies all the way through the system. Trying to reduce the time is probably a gargantuan effort. It takes a lot of “engineering.”

Dr. Bertino: So what are we going to do about it?

Dr. Burtness: For head and neck cancers, we need to figure out if these delays are harming patients. If they are, the multidisciplinary teams need to have approaches that involve fast tracking, scheduling the scans and getting the time required for treatment simulation down, and making sure there is operating room time in big enough blocks for head and neck cancers that reconstructions can be done. The other part of the problem, the delay in diagnosis, is not necessarily a hem/onc problem, but may arise out in the community. A better appreciation of the nature of this HPV epidemic so that people are more alert to what they were seeing on CT scans would also be helpful.

Dr. Aisner: Another dimension of this is where surgery and other aspects of treatment are going to happen. My sense is that surgical approaches fare better in well-organized head and neck programs that have the full spectrum of care and they do a lot of cases as opposed to the ENT guy who thinks he can do it but only does six cases a year.

Dr. Burtness: In esophagus cancer, we’ve already seen insurers favoring high-volume centers because there are clear data that if you have your resection at a hospital that does a fair number of these operations, and where the ICU manages a fair number of post-esophagectomy patients, you will have a better survival. If we were to develop those data in head and neck cancers, payers might similarly favor having head and neck surgery done at a center of excellence because management is complex. It involves multiple teams, and the risk of dropping the ball as you pass from one team member to the other is high. This can be minimized with multidisciplinary, high-volume, high-expertise teams.