Diagnostic approach used by U.S. general dentists following discovery of oral lesions suspicious for malignancy/premalignancy: Findings from the National Dental Practice-Based Research Network

Abstract

Oral cancer examinations seek to clinically identify early oral cancers and discover premalignancies prior to their progression to invasive cancer.

A cross-sectional study was conducted between April and August 2017 to explore and quantify the diagnostic approach used by United States (U.S.) general dentists (GDs) following discovery of an oral lesion suspicious for malignancy/premalignancy. U.S. licensed GDs who were clinically-active members of the U.S. National Dental Practice-Based Research Network were eligible to participate. Data for analysis were obtained via two questionnaires.

The majority of participants were males, practiced full-time, had a suburban primary practice, were born during the 1950s, and graduated dental school in the 1980s or 2000s.

After identifying an oral lesion suspicious for premalignancy/malignancy, the next action of most GD respondents (65.4%) was to refer the patient for consultation/biopsy. Most GDs (87.2%) referred to an oral and maxillofacial surgeon; 22% of GDs reported personally biopsying suspicious lesions.

There was a relatively weak association between completing an AEGD or GPR residency and subsequently personally performing biopsies on patients with oral lesions suspicious for malignancy/premalignancy (OR 1.33, p=0.06).

Most written referrals take place electronically and often include information, including lesion location (87.0%), lesion sign/symptoms (85.3%), lesion history (83.9%), radiographs (59.3%), medical history (50.6%), dental history (48.8%), and photographs (42.4%).

When a referral biopsy was performed, referring clinicians were most frequently informed of the findings via a written report (96.7%,when positive for malignancy/premalignancy and 95.4% when negative).

We present a snapshot of current actions taken by U.S. GDs following the discovery of oral abnormalities suspicious for premalignancy/malignancy.

Introduction

The purpose of oral examinations is to identify hard and soft tissue abnormalities including oral malignancies and oral potential malignant lesions (OPML) prior to progression to invasive cancer. A significant part of the long-term goal is to reduce morbidity and mortality from oral cancer and reduce the risk that a premalignant disorder will develop into a carcinoma.

The opportunistic visual-tactile soft tissue examination is an important component of a thorough dental patient oral evaluation and well within the skillset of the general dental practitioner. Most abnormalities identified during a visual-tactile oral mucosal examination are likely to be benign, with a smaller percentage being OPML or a malignancy. Unfortunately, it is not always possible to distinguish between malignant, premalignant, and benign lesions on the basis of a clinical visual-tactile examination alone, with the majority of lesions requiring a biopsy and histopathologic assessment for a diagnosis of an OPML or cancer.^1–5 There are exceptions like chronic Graft versus Host Disease (cGVHD) lesions where a biopsy is not needed except to assess a site for dysplastic malignant transformation.⁶

Previous articles have described a systematic approach for assessing and managing oral abnormalities suspicious for oral cancer and premalignancies.^3,7 The purpose of the current analysis was to explore and quantify the diagnostic approach that United States (U.S.) general dentists (GDs) use following the identification of potentially malignant oral lesions during a visual-tactile oral mucosal examination.

Methods

The collection of data used in the current analysis has been described previously.⁸ Briefly, a cross-sectional study was conducted between 4/12/2017 and 8/31/2017 under the auspices of the National Dental Practice-Based Research Network (Network), which is funded by the United States National Institute of Dental and Craniofacial Research. A total of two thousand licensed U.S. GDs who were practicing in the U.S. and clinically-active members of the Network were invited to join the study.

Invitations to participate in the Oral Cancer Examination (OCE) study were sent via e-mail, with the invitation including a link to an OCE questionnaire that had been designed specifically for the study. The self-administered questionnaire contained 25 question items, many with follow-up sub-questions. All items had been reviewed and evaluated by the Coordinating Center of the Network, Westat Instrument Design, Evaluation, and Analysis Services, key members of the Coordinating Center, and ten Network dentists and hygienists who completed the instrument and were individually interviewed for their understanding of each question.

Collected data were cleaned and imported into SAS, Stata, and SPSS for analysis.

Responses from two questions within the OCE questionnaire were used to form four groups of dentists that encompass the subsequent evaluation of their patients with oral lesions suspicious for malignancy or premalignancy.

As they appeared in the questionnaire, the questions used were as follows:

1. In the past (6) months, approximately how many biopsies did you personally perform on patients with oral lesions suspicious for premalignancy/malignancy? ___ (number)

2. In the past six (6) months, how many patients with oral lesions suspicious for premalignancy/malignancy did you refer for consultation/biopsy? ___ (number)

The four groups of GDs created were as follows:

• Personally performed biopsies and did not refer

• Referred only

• Both performed biopsies and referred

• Neither referred nor performed biopsies

We used a separate question in the OCE questionnaire to identify the type of practitioner to whom the GDs reported generally referring their patients with oral lesions suspicious for premalignancy or malignancy. The question provided eight generic responses (e.g., “oral medicine specialist,” “oral pathologist”) and the opportunity to select another, unspecified option [i.e., “other, specify”].

In addition, we used selected variables collected via the OCE and Network Enrollment Questionnaires to determine whether completion of an Advanced Education in General Dentistry (AEGD) or General Practice Residency (GPR) program was associated with personally performing oral biopsies.

Information on the referral process used by GD participants for their patients suspected as having potentially malignant lesions was obtained from the OCE study questionnaire and included the manner in which the referral was made and what information was generally provided to the surgeon or other specialist. We further ascertained the manner by which results of the consultation/biopsy were conveyed from the surgeon to the referring GD and patient.

In the analysis, standard statistical methods were used, including the inspection of frequency distributions, the computation of odds ratios (ORs) and their 95% confidence intervals (95% CI), as well as the construction of logistic regression models. Proportions were calculated using the number of GDs with a given trait as the numerator and the total number of respondent GDs for that variable as the denominator.

The study was approved by all Network-applicable Institutional Review Boards, and all study participants provided informed consent.

Results

A total of 1073 GDs responded to at least one question.

Table 1 presents the distribution of selected characteristics for the GDs who participated in the project. The majority of GDs were males, practiced full-time, had not attended a 1- or 2-year GPR or AEGD residency, had a suburban primary practice, and were born during the 1950s. Most participants had graduated from dental school during the 1980s or 2000s.

Table 1.

Characteristics of participating general dentists (n=1,073).

Sex	n (%)
Male	753 (70.5)
Female	315 (29.5)
Practice full- or part-time
Full-time (32+ hrs/wk)	891 (84.1)
Part-time (<32 hrs/wk)	168 (15.9)
Completed AEGD or GPR Residency
No	716 (66.9)
Yes	354 (33.1)
Primary practice location
Inner city urban	117 (11.0)
Urban (not inner city)	292 (27.4)
Suburban	498 (46.8)
Rural	157 (14.8)
Birth Cohort
<1950	101 ( 9.4)
1950 – 1959	378 (35.4)
1960 – 1969	205 (19.2)
1970 – 1979	251 (23.5)
1980	134 (12.5)
Dental graduation year
<1980s	192 (17.9)
1980s	349 (32.5)
1990s	204 (19.0)
2000+	328 (30.6)

N.B., in Psoter et al, 2019, graduation cohort-specific numbers for “Participants” as reported above were mistakenly reported in that article’s Table 1 as pertaining to “Non-Participants.” (i.e., in that table the reported values for “Participants” and “Non-Participants” were reversed).

The n for each variable does not total 1073 due to missing data.

As shown in Table 2, after identifying an oral lesion suspicious for premalignancy or malignancy during the preceding six months, nearly two-thirds of the dentists [698/1067, (65.4%)] reported that their next action was referring the affected patient for consultation / biopsy, while 194/1067 or 18.2%, reported referring some patients and personally performing biopsies on others. Only 42 of the 1067 GDs (3.9%) did not refer but personally performed biopsies; 12.5% (133/1067) reported no biopsies or referrals over the preceding six months.

Table 2.

Dentists by referral and biopsy status during the six months preceding administration of the OCE study questionnaire.

Personally Performed Biopsies
		No	Yes	Total
	No	133 (12.5%)	42 (3.9%)	175
Refer Patients
	Yes	698 (65.4%)	194 (18.2%)	892
	Total	831	236	1067 (100%)

As can also be ascertained from Table 2, during the six-month period prior to questionnaire completion, a total of 892 (698+194) or 83.6% (892/1067) of the general dentists whose patients had one or more suspicious lesions, reported referring their patients for consultation/biopsy while 236 (42+194) or 22.1% (236/1067) of the GDs reported personally biopsying one or more suspicious lesions. (Note: these categories are not mutually exclusive.)

Types of Practitioners to Whom Patients Were Referred

Participating dentists were asked to identify the practitioner type to whom they referred their patients with oral lesions suspicious for malignancy or premalignancy for consultation/biopsy. When all GDs who responded to the question (n=1069) were included, the majority (87.2%) reported referring to an oral and maxillofacial surgeon (OMFS), 5.4% of the GDs referred to oral pathologists, 2.5% to ENTs, and 2.4% to oral medicine specialists. The entire frequency distribution is presented in Table 3.

Table 3.

Practitioners to whom patients with oral lesions suspicious for premalignancy or malignancy were generally referred (n=1069).

	n (%)
Oral / Maxillofacial Surgeon	932 (87.2)
Oral Pathologist	58 (5.4)
Ear Nose Throat Specialist	27 (2.5)
Oral Medicine Specialist	26 (2.4)
Periodontist (Other)	9 (0.8)
Head & Neck Surgeon	3 (0.3)
Depends on lesion (OMF, ENT or Oral Pathologist)	3 (0.3)
Dermatologist	2 (0.2)
Primary Care Physician	2 (0.2)
Cancer hospital, Oral Oncologist	2 (0.2)
Five GDs reported that they did not refer patients with oral lesions:
Do not refer patients with oral lesions	2 (0.2)
Suspicious for Malignancy / Premalignancy
Do not refer patients with oral lesions;	3 (0.3)
patients are referred to me
Total	1,069 (100)

Note: Two GDs who responded to the type of practitioner question did not respond to questions regarding personally performing biopsies and referrals.

When the analysis of practitioner type was limited to the 892 GDs who reported having referred at least one patient with an OPML or cancer during the previous six months, the distribution was similar to that shown in Table 3, with 88.6% of the GDs reporting that they referred to an OMFS (distribution not shown).

AEGD or GPR Residency Training

Among the 1070 GD respondents, 33.1% (354/1070) reported having completed either an AEGD or GPR postgraduate residency. While 25.7% (91/354) of the AEGD/GPR graduates reported having personally biopsied a lesion suspicious for oral malignancy/premalignancy during the preceding six months, 20.7% (148/716) of the GDs who had not completed such a residency also personally performed biopsies during the same time period. The corresponding crude odds ratio (OR) and 95% confidence interval (95% CI) for personally biopsying suspicious lesions by AEGD/GPR status was 1.33 (95% CI 0.99–1.79, p=0.06). Controlling simultaneously for practicing full- or part-time, gender, birth cohort, and dental graduation cohort yielded an adjusted OR of 1.35 (95% CI: 0.99–1.84), p=0.06.

When AEGD and GPR graduates were analyzed separately as regards personally performing biopsies, the crude OR for AEGD completers was 0.97 (95% CI: 0.59–1.61; p=0.91 and for GPR graduates, 1.27 (0.91–1.76); p=0.15. Controlling for practicing full- or part-time, gender, birth cohort, and dental graduation cohort resulted in an adjusted OR of 1.02 (95% CI: 0.60–1.72), p=0.94 for AEGD graduates and 1.30, 0.93–1.81, p=0.13) for GPR completers.

Referrals

When referring a patient with an OPML or malignancy, the purpose of the requested referral appointment was most often communicated to the referral practitioner using some form of written correspondence sent by the referring clinician (63.9%). Most written referrals were reported to take place electronically, sometimes involving back-and-forth e-mails between offices, and depending on the situation, could include photographic and radiographic images, faxes, and medical/dental records. In addition, written communication and ancillary information was reportedly hand-carried by the patient to the referral practitioner 42.3% of the time. When telephone calls were used to arrange a referral appointment, the staff reportedly initiated the calls 30.0% of the time while the practitioner called 28.8% of the time.

Information Provided with Referral

Table 4 presents the frequency distribution for items that GDs provide when patients are referred for a consult or biopsy. The most frequently reported information was lesion location (87.0%), lesion signs/symptoms (85.3%), and lesion history (83.9%). Notably lower percentages of general dentists reported routinely providing radiographs (59.3%), medical history (50.6%), dental history (48.8%), and photographs (42.2%).

Table 4.

Information routinely provided when referring patients with lesions suspicious for malignancy or premalignancy.

	n *	(%)
Lesion location	933	87.0
Signs/symptoms	915	85.3
Lesion history	900	83.9
Radiographs	636	59.3
Medical history	543	50.6
Dental history	524	48.8
Photographs	455	42.4
Diagnostic adjunct results, e.g., VELscope	129	12.0
Diagnostic salivary test results	19	1.8
Total	1,073	100

^*the number of GDs indicating that they provide the identified item when making a referral.

A sizable proportion of the GDs (43.1%, 461/1069) did not know whether the medical/dental insurance plans of their patients required the initial referral to a primary care physician (gatekeeper). On the other hand, 51.0% (545/1069) of the respondents indicated that a gatekeeper was not required while 5.9% (63/1069) reported that a gatekeeper was mandatory.

When a referral appointment for a patient with a suspected oral premalignancy or malignancy was scheduled, a reported 51.2% of the GD offices verified with either the patient or host office that the appointment had been kept.

After a biopsy was performed, referring clinicians were most frequently informed of the findings via a written report (96.7% when positive, 93.4% when negative for malignancy or premalignancy) although findings were additionally conveyed via telephone calls.

Most GDs (59.4%) assumed the practitioner who performed the surgery would contact the patient regarding the findings. To support that assumption, nearly half of the GDs (48.7%) indicated that they generally receive a letter from the specialist who performed the surgery saying that the biopsy findings had been discussed with the patient. Some GDs indicated that they (33.3%) or their staff (9.8%) would follow-up with the patient, with a small percentage (2%) of GDs indicating that they would forward a copy of the relevant histopathologic report to their patient.

Discussion

In a previous analysis based upon Network data, we estimated the annualized per dentist number of oral lesions suspicious for premalignancy/malignancy discovered by United States general dentists and the annualized per dentist number of histologically-confirmed cancers subsequently diagnosed.⁸ More recently, an article based upon vignette data from the same parent study explored cues that dentists utilize in the early detection of potentially malignant lesions.⁹ In the current manuscript, we report on actions taken by Network general dentists beginning with their visual-tactile identification of an oral lesion suspicious for malignancy/premalignancy and continuing through the histopathologic diagnosis.

In the present study of United States GDs, we found the most frequently reported course of action following initial lesion identification was to refer the patient for consultation/biopsy, generally to a specialist and most frequently to an oral and maxillofacial surgeon. A similar referral strategy has also been reported for the dental profession in other developed countries. For example, in a study of British primary care dentists, it was concluded that when most general practitioners had any doubt regarding how a patient’s presenting signs, symptoms, and risk profile should be utilized in coming to a diagnostic decision, the dentist’s default action was to refer.¹⁰ In a consistent vein, the latest United Kingdom (UK) NICE Pathways (2021) recommends that GDs consider patients with a lump in the oral cavity or on the lip, as well as persons with an oral erythroplakia or erythroleukoplakia, be suspected as being on a potential oral cancer trajectory and scheduled for a referral appointment to occur within two weeks of lesion discovery.¹¹

Studies conducted in Australia have also reported that general dentists often refer their patients with oral lesions suspicious for premalignancy or oral cancer to specialists.^12,13 One Australian study found that the majority of dentists referred patients with oral lesions suspected of containing a malignant or premalignant component to oral and maxillofacial surgeons, oral medicine specialists, or oral pathologists.¹² In another Australian study, Mariño et al (2017) reported that nearly 86% of Australian oral health professionals (defined as dentists, dental hygienists, dental therapists, oral health therapists) indicated that their first action after identifying a suspicious, non-readily diagnosable lesion was referral to an “oral medicine specialist.”¹³

While oral soft tissue examinations, systematically performed, can reveal potentially malignant disorders, such evaluations can also be instrumental in discovering conditions with probable benign diagnoses (e.g., pyogenic granulomas). The detection of such abnormalities can extend the overall comprehensiveness and utility of the examination by identifying lesions that otherwise might have gone undetected. Because some lesions initially diagnosed by the GD as innocuous might have a malignant component, their early identification can serve as a reference to be compared against in the future, keeping in mind that the standard for the diagnosis of many intraoral lesions has long been a biopsy and histopathologic assessment.^2–5

It should be underscored that in the current study, when a patient was reported as being referred by a general dentist, the referral could result in a biopsy being performed by the surgeon who received the referral. Therefore, it should not be assumed that referred patients did not receive a timely biopsy.

We found that during the six months prior to questionnaire administration, 22% of participating GDs reported having personally biopsied at least one lesion suspicious for oral cancer or premalignancy. Based upon our analysis, there was some evidence that GDs who had completed an AEGD or GPR post-doctoral residency were slightly more likely to personally perform biopsies on oral potentially malignant lesions than their GD counterparts who had not completed an AEGD/GPR residency. The observed crude OR of 1.33 can be interpreted as the odds of personally performing a biopsy among AEGD/GPR residency completers were 1.33 times, or 33%, higher than the odds of personally performing a biopsy among GDs who had not completed an AEGD/GPR residency. The OR suggests a relatively weak association between completing an AEGD or GPR residency and subsequently performing a biopsy on potentially malignant oral lesions.

Our analysis, which stratified on type of residency program (i.e., AEGD or GPR), suggests that GPR graduates were somewhat more likely to personally perform biopsies than completers of AEGD programs. Our observation is in keeping with some (13,14), but not all (15), previous research linking GPR and AEGD training with the provision of specific clinical services.^14–16 It is noteworthy that the observed residency-specific ORs in our analysis were neither clinically nor statistically significant. If either AEGD or GPR residencies are associated with personally performing biopsies on oral potentially malignant lesions, the relationships appear weak.

A number of advantages and disadvantages have been reported previously as regards general dentists personally performing biopsies on oral lesions suspicious for malignancy or premalignancy. Advantages include the patient’s prior familiarity with the GD/staff/office, which could result in reducing patient fears, thereby limiting the occurrence of diagnostic and treatment delays.¹⁷ Disadvantages, as often identified by the GDs themselves, include medico-legal concerns, a perceived lack of practical biopsy skills/experience among GDs, and a dearth of confidence in interpreting biopsy results.^17–19 Limited biopsy skills, due to educational background or experience on the part of general dentists, could manifest, for example, as a disproportionate excess of crush artifacts in oral biopsies obtained by GDs relative to OMFSs as observed in a Spanish study.²¹

In the current analysis a large proportion (>40%) of the GD respondents did not know whether a gatekeeper was required by their patients’ insurance plans for the referral/management of suspicious oral lesions; however, a majority of GDs (>50%) indicated that a gatekeeper was not required and fewer than 10% reported that a gatekeeper was mandatory.

As described by the general dentist respondents, the process by which patients are referred for consultation / biopsy, is generally initiated using a written format via either e-mail or an electronic referral form. Items frequently submitted with the referral – either electronically or by other means – include lesion location, signs/symptoms, and lesion history. The lower percentages of dentists who routinely providing radiographs, medical history, dental history, and photographs is notable. However, radiographs might be considered less important to provide for soft tissue lesions. Biopsy results, whether positive or negative for malignancy, were reportedly sent by the consulting specialist to the referring dentist and patients.

Just over half of the GD offices reported that they confirmed that their patients with lesions suspicious for oral cancer/premalignancy had kept their referral appointments. This study did not assess the reasons for not confirming that the appointments was kept. We can speculate that this may include such things as busy office staff in which no one is assigned the task, not knowing patients well enough to recognize who will reliably attend their appointment or not, or other reasons. Any strategy for ensuring the appointment is kept will decrease the delay in diagnosis and should include stressing with the patient the importance of keeping the referral appointment.

Potential strengths and limitations of the current project were addressed as part of our initial report of study findings.⁸ Primary study strengths include access to a large database of general dentists through the National Dental Practice-Based Research Network and the use of data from a questionnaire designed specifically for the project as a whole. The primary study limitation was that all GD data were self-reported and could result in some level of bias being introduced into our results. In addition, dentists who agree to participate in a research network are not necessarily representative of all practicing GDs, and the study sample was weighted towards males, suburban practice, and being born during the 50’s. However previous network questionnaire studies have found that network dentists report diagnostic and treatment practices that are similar to that of non-network dentists.^{16, 22–27} This similarity remains consistent when network dentists are compared to non-network dentists using the American Dental Association (ADA) survey of dental practice.²⁸

Summary

After discovery of an oral lesion suggestive of premalignancy or malignancy, the majority of U.S. general dentists refer their patient to a specialist, most frequently to an oral and maxillofacial surgeon, for a consult / biopsy. One-fifth of the GDs reported personally biopsying some or all oral lesions suspicious for premalignancy or malignancy.

Compared with National Dental PBRN GDs who had not completed a dental residency, our findings suggest that GDs who completed an AEGD or GPR residency program had only a slightly higher odds of personally performing biopsies on their patients with oral potentially malignant lesions. The odds were especially low among AEGD graduates compared to GDs who had completed a GPR. None of these were statistically significant.

Items frequently submitted with the referral include lesion location, signs/symptoms, and lesion history. A lower percentage of dentists routinely provide radiographs, medical history, dental history, and photographs. Biopsy results, whether positive or negative for malignancy, are routinely sent by the consulting specialist to the referring dentist and patients.

Our analysis of Network data provides a snapshot of current actions taken by U.S. GDs following the discovery of an oral abnormality suspicious for premalignancy/malignancy. The findings have further utility as a useful mile marker when evaluating trends in the approach used by United States GDs into the future.