We conducted this study to identify the different stages at which delay can occur within the U.S. health care system with regard to diagnosis and treatment of oral cancer.

	SUBJECTS, MATERIALS AND METHODS

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
This study includes all patients seen in the Oral and Maxillofacial Surgery Clinic, University of California, San Francisco, between Jan. 1, 2003, and Dec. 31, 2007 (60 months), with a new diagnosis of squamous cell carcinoma of the oral cavity. Exclusion criteria included other malignancies, recurrences, long-standing dysplasia, proliferative verrucous leukoplakia and other premalignant conditions. We chose these criteria because many of these patients were being seen on a regular basis by a health care professional. We also excluded patients who received any part of their treatment—except for the first visit to a primary care clinician—elsewhere because we wished to study delays within one health care facility.

As part of an established protocol, all patients with head and neck cancer who are scheduled to receive any type of treatment at the University of California, San Francisco, are presented to a weekly multidisciplinary head and neck tumor board for discussion and treatment planning. Therefore, we defined the time intervals established as sequential and potential areas for delay to occur in the system as follows:

– T1: length of time, in days, from the time at which the patient first became aware of symptoms (subsequently shown to be oral cancer) to his or her first visit to a primary care clinician (physician or dentist). This time often was approximate because patients tended to estimate time in months, but sometimes they could identify the week or day when they first became aware of the cancer.

– T2: length of time, in days, from the time when the patient was seen by the primary care clinician to the time when he or she underwent a biopsy performed by the primary care clinician or the time when he or she was referred to a specialist.

– T3: length of time, in days, from the time when the biopsy was performed by the primary care clinician or the patient was referred to the time when the patient was seen by the specialist. In this study, the specialist always was an oral and maxillofacial surgeon who had an interest in head and neck pathology.

– T4: length of time, in days, from the time when the patient was first seen by the oral and maxillofacial surgeon at the University of California, San Francisco, to the completion of appropriate studies and investigations.

– T5: length of time, in days, from the completion of studies and investigations to the time when the patient’s case was presented to the head and neck tumor board for discussion and treatment planning.

– T6: length of time, in days, from case presentation to the head and neck tumor board to the time at which definitive treatment was begun. This was either the first day of radiation therapy or the day of surgery.

	RESULTS

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
During the five-year study period, we identified 50 new patients with biopsy-proven squamous cell carcinoma of the oral cavity who did not meet the exclusion criteria. Below are the time interval findings:

– T1: mean time from the patient’s first becoming aware of the problem to visiting a primary care clinician, 104.7 days (range, 0–730 days);

– T2: mean time from the patient’s visiting a primary care clinician to undergoing a biopsy or being referred, 35.9 days (range, 0–280 days);

– T3: mean time from the biopsy or referral to actually visiting a specialist, 17.7 days (range, 0–240 days);

– T4: mean time from visiting a specialist to undergoing appropriate tests, 10.1 days (range, 0–33 days);

– T5: mean time from completion of tests to case presentation to the head and neck tumor board, 20.7 days (range, 1–208 days);

– T6: mean time from case presentation to the head and neck tumor board to commencement of definitive treatment, 16.6 days (range, 1–76 days).

The mean total time from patients’ becoming aware of their problem to commencement of definitive treatment was 205.9 days (range, 52–786 days).

	DISCUSSION

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Oral cancer remains a public health problem, with approximately 8,500 people dying of the disease each year in the United States.¹ In our study, the longest single delay in receiving treatment was from the time the patient became aware of a possible problem to his or her first visit to a primary care clinician. Other studies have confirmed this delay.^18–22 The shortest time for a symptomatic patient to visit a clinician in this study was the same day that symptoms appeared, and the longest time was more than two years. Although the literature contains reports of oral cancer progressing slowly,²³ in general, a delay of more than two years, as occurred in one case in this study, worsens the patient’s overall long-term prognosis and complicates the treatment.

Educating patients Health care providers must place greater emphasis on educating patients about the importance of visiting a clinician as soon as oral symptoms develop. They can visit a dentist or a physician, and the clinician should schedule an early appointment for these patients.²⁴ Health care providers’ offices should not encourage patients to visit an emergency department, because this is not its primary function.

Reasons for delay There are a number of reasons why people do not visit clinicians soon after noticing symptoms. One is the financial barrier that exists for many patients who are at risk of developing oral cancer. Interestingly, Yu and colleagues²⁵ recently studied 102 patients with new diagnoses of oral cancer at the University of Toronto. They reported a mean initial delay of 21.7 weeks in visiting a health care professional, although the median delay was only 4.5 weeks. This median delay of 4.5 weeks was remarkably shorter than the median delay of 18.4 weeks found at our institution. This disparity may be explained by the fact that Canada’s health care system is publicly funded.

In the United States, studies have shown that health insurance status is an important predictor of the stage at which patients with oropharyngeal cancer first visit a clinician.²⁶ Compared with patients with private health insurance, those with no insurance and those with Medicaid had a greater likelihood of being diagnosed with overall advanced-stage cancer.²⁶ Although it is a daunting task, the dental profession needs to find a way to eliminate financial restraints that prevent patients from seeking care when symptoms first appear. Increased access to health care, including dental care, for uninsured patients likely would allow for more frequent detection of early-stage cancers.

Government agencies, universities and dental clinics have made attempts to diagnose cancer early, via population screening and the use of various visual and chemical detection methods. Routine screening of asymptomatic people in an attempt to diagnose oral cancer early is not practical and does not meet any epidemiologic guidelines for a successful screening program; attempts to implement such programs generally have not been effective.^27,28 In general, for those considering a regional or national screening program, the following are needed²⁹:

– an identifiable and diagnosable disease that is an important health problem;

– availability of an accepted treatment for patients with recognized disease;

– a recognizable latent or early symptomatic stage;

– an adequate understanding of the natural history of the disease;

– a simple, cost-effective and reliable test for the disease that has acceptable rates of false-positive and false-negative findings;

– a screening process that is ongoing and not a one-time project;

– detection of the disease via screening that improves the outcome.

Routine screening of the general population for oral cancer cannot meet the above guidelines.³⁰ Nevertheless, screening of certain high-risk groups may result in earlier detection and earlier treatment.³¹ Such high-risk groups might include those with alcoholism, those who smoke^32,33 and those who had developed a carcinoma previously, because it is known that patients who have had a malignancy in one part of the body are more liable to develop one in another site.³⁴ In particular, Kramer and colleagues³⁵ reported that those who have had an oral carcinoma have a greater than 9 percent overall chance of developing a second oral carcinoma throughout their lifetime, while Day and Blot³⁶ found second primary malignancies at a rate of 3.7 percent per year.

Oral cancer testing Developing and validating an acceptable test for detecting oral cancer with minimal false-positive and false-negative results has been problematic. Screening via clinical examination alone—as can be done by a general dentist during the patient’s routine dental examination—or via oral health fairs generally has resulted in poor detection rates, mainly because the high-risk population does not visit a dentist regularly or attend screenings at health fairs. In particular, this applies to patients with a history of alcohol abuse.³⁷ With routine screening of asymptomatic patients, examination alone may identify areas such as leukoplakia and atrophy; however, a biopsy is necessary to make a diagnosis, and this is a time-consuming, uncomfortable and relatively expensive technique that cannot be applied to a general population.³⁸ Therefore, determining the true sensitivity and specificity of a clinical examination is not possible without comparing it with the gold standard of scalpel biopsy.

Developing and validating an acceptable test for detecting oral cancer with minimal false-positive and false-negative results has been problematic.

Adjunctive techniques Alternative diagnostic techniques such as chemiluminescence³⁹ and toluidine blue staining⁴⁰ indicate which areas of the oral cavity may be more suspicious than others and may direct the choice of biopsy site, but they are useful only as an adjunct in diagnosing existing lesions. Thus, chemiluminescence and toluidine blue staining are not true screening tests.⁴¹

Brush biopsy A noninvasive biopsy such as the brush biopsy technique can identify atypical cells,⁴² but any such detection must be followed by an incisional biopsy, which again is impractical for the general population. The brush biopsy is intended to assess clinical lesions that may not otherwise have undergone a biopsy, because the level of suspicion for cancer in these lesions is low.^43,44 The brush biopsy technique has been well-accepted by dentists and has helped detect malignant and premalignant lesions in early stages that otherwise would have been watched or ignored.⁴² There also are anecdotal reports of false-positive⁴⁵ and false-negative⁴⁶ results with brush biopsy techniques. To our knowledge, no studies have compared cytology with scalpel biopsy in a large number of seemingly innocuous lesions identified on clinical examination that typically would not undergo a biopsy.⁴¹ Further studies are needed to clarify the sensitivity and specificity of the brush biopsy technique for lesions of low suspicion.

A recently developed detection system, the VELscope (LED Dental, White Rock, British Columbia, Canada), is thought to enhance dentists’ ability to detect oral cancer after the tissue in question is exposed to fluorescent light. The premise is that abnormal cells will absorb fluorescent light and appear dark, while normal mucosa will reflect light and glow. To date, the effectiveness of this device has been described only in observational and case studies.^47,48 Although it can detect abnormal tissue, it remains to be seen if it can distinguish cancer from other abnormal oral tissues, such as aphthous ulcers. In addition, biopsy ultimately still is required.⁴⁹

With routine screening of an asymptomatic population being unrealistic and ineffective for the detection of oral cancer,⁵⁰ most cases are diagnosed when a patient with symptoms visits a clinician. In practice, some patients may make a special visit to a dentist when symptoms develop, or some may wait until their next six-month checkup before mentioning a sore or an ulcer.⁵¹ Other patients may visit a primary care physician, and this applies in particular to patients who have medical insurance but may not have dental insurance. Studies have shown that, in general, physicians are capable of diagnosing oral carcinoma and referring patients appropriately,⁵² but health care providers in oral specialties are known to be better trained in this area.⁵³ In our study, some patients had been seen by a physician or a dentist on a number of occasions with the patient complaining about a persistent oral problem before he or she finally underwent a biopsy or was referred.

Self-examination One intriguing concept in increasing awareness of oral cancer is to institute a self-examination campaign⁵⁴ similar to the monthly breast self-examination campaign. A visual examination of the oral cavity in a mirror on a monthly basis would be fairly easy to perform and could result in the detection of some lesions. Patients would be instructed to lift up their tongues to view the floor of the mouth, move the tongue to the right and left of the mouth to evaluate the lateral surfaces and pull both cheeks laterally to examine the vestibules, gingivae and buccal mucosa. Self-examination has the potential to enable patients to detect asymptomatic cancers at early stages. Not only does self-examination increase the frequency with which the oral cavity is screened, but it is done at no burden to the health care system beyond patient education (for example, brochures).

Patients without symptoms might visit a general dentist every six months and a physician every year. Any self-examination conducted between these intervals might result in the detection of lesions and expedite visits to health care professionals before symptoms develop. Again, the main difficulty lies in communicating this message to patients who are at greatest risk. National oral health organizations have recognized this and have increased their efforts to educate the public.

In 2002, the American Dental Association received a five-year grant from the National Cancer Institute that focused on oral cancer prevention education for practicing dentists in the United States.⁵⁵ More than 60 continuing education programs have been held, focusing on the key components of oral cancer prevention, including risk assessment and risk reduction for tobacco and alcohol use, chemoprevention, early detection and early diagnosis. The American Association of Oral and Maxillofacial Surgeons also has initiated a task force on early detection of oral cancer.⁵⁶ Despite these efforts, the majority of consumers do not know that dentists screen for oral cancer.^57,58 Through continued public educational efforts, however, patients will be better informed regarding oral cancer screenings provided by dentists.

In our study, the mean time from the initial appointment with a primary care physician or dentist to a biopsy or referral to a specialist was 35.9 days. This is longer than one would wish, particularly when we consider the fact that in some cases, there was a delay of almost 10 months from the time when the patient first visited a clinician to the time when the patient underwent a biopsy.

The British National Health Service, in its guidelines for general medical practitioners,⁵⁹ recommends that clinicians refer patients to specialists in the following circumstances:

– any oral signs or symptoms for which a definitive diagnosis cannot be made and that do not resolve within six weeks;

– unexplained oral red or white patches that are painful, swollen or bleeding;

– unexplained oral ulcer present for longer than three weeks;

– unexplained tooth mobility (unless the teeth are loose, clinicians may refer patients to a general dentist).

Delays in referral by primary care clinicians have led to lawsuits and claims of malpractice,^60,61 and some evidence shows that the number of such malpractice claims may be increasing.⁶² An early biopsy or referral of a patient with an unknown lesion is indicated,⁵¹ and a biopsy of an unknown lesion that does not resolve within three weeks is essential. Any patient who is referred to a specialist because of a suspicious lesion should be given an appointment as soon as possible.

In our study, the mean time from referral to a specialist to actually being seen by a specialist was 17.7 days, which is appropriate. The longest delay was 240 days, and this was because of a patient-generated delay. This specific delay illustrates the need for patients to realize the potential seriousness of the problem and the wisdom of arranging and keeping early appointments.

In most cases, there was an inevitable delay between the patient’s seeing the specialist and undergoing necessary tests and other investigations. These included plain radiographs, computed tomography, magnetic resonance imaging, positron emission tomography, direct and indirect laryngoscopy, triple endoscopy and appropriate laboratory tests. The longest time for these tests to be performed in our study was 33 days. In some cases, a delay was caused by the need to obtain insurance authorization for some studies. In addition, in some cases, there were delays of a few days in the scheduling of imaging studies. The mean delay of 10.1 days in this study appears appropriate.

Head and neck tumor board The mean delay from completion of imaging and other investigations to presentation to the head and neck tumor board was 20.7 days (range, one to 208 days). In two cases, the delays were more than 200 days, thus elevating the mean number. Again, these delays were patient-generated and resulted from patients who were difficult to contact and difficult to persuade of the importance of presentation to the head and neck tumor board. If we exclude these outliers, the mean time is reduced to 13.2 days, which is appropriate because some patients could not attend the tumor board presentation during the week after their diagnosis.

Definitive treatment The mean time from presentation to the head and neck tumor board to commencement of definitive treatment was 18.6 days, which appears appropriate because there often was a delay in obtaining insurance authorization and in scheduling an operating room (frequently two to three weeks for each). Similarly, there often was an unavoidable delay in commencing radiation therapy, because of the need to obtain insurance authorization and for any planning (such as calculating radiation dosage) that was necessary.

Overall delay The overall delay from the patient’s first becoming aware of symptoms to commencement of definitive treatment was a mean of 203.8 days, which is just less than seven months (range, more than two months to 2years). This overall delay appears unacceptable; we would hope that the interval would be eight weeks or less. No patients in our study underwent definitive treatment in less than eight weeks.

	CONCLUSION

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Delays in the diagnosis and treatment of oral cancer occur in a number of stages in the referral and treatment process. The longest single delay in our study occurred between the patient’s first becoming aware of symptoms and visiting a primary care clinician. Health care professionals need to place more emphasis on early self-referrals, as well as accelerate other portions of the referral process, particularly in referrals from a primary care clinician to a specialist. Screening and self-examinations of high-risk groups might improve outcomes for patients who self-refer late in the course of their oral cancer. The practical issues of how high-risk people are identified, how they are screened and how often they are screened require further study.