Investigation into opportunities for early detection of oral cancer

Purkayastha, Mitana (2018) Investigation into opportunities for early detection of oral cancer. PhD thesis, University of Glasgow.

Abstract

Background: Early detection strategies for oral cancer aim to decrease the mortality rates and improve outcomes of the disease through early diagnosis and treatment. Guidance and regulatory bodies have an expectation that general dental practitioners will be able to promptly detect and refer patients with suspected oral cancerous lesions. However, the opportunities for early detection of oral cancer in primary dental care settings (particularly considering the low overall volume of the disease, the potentially increasing incidence rates, and the possibility of certain communities exhibiting particularly high rates) have not yet been investigated. This thesis examines the feasibility of early detection of oral cancer in primary dental care services, and undertakes risk-stratification to identify “high-risk” communities that can be utilised to target future early detection efforts. It further explores potential or missed opportunities for early detection in dental and other healthcare settings (both primary and secondary care), and assesses the feasibility of exploring routes to diagnosis.

Aim: The aim of this thesis was to investigate opportunities for the early detection of oral cancer in Scotland by measuring the current burden of the disease, examining the feasibility of early detection in a dental setting, and exploring the potential role of alternative health care settings in early detection efforts.

Methods: Descriptive epidemiological and data linkage cohort studies utilising national routine administrative health datasets were undertaken. The descriptive epidemiological analysis included all cases of head and neck cancer diagnosed between 1975 and 2012 and registered on the Scottish cancer Registry and annual midterm population estimates. These data were used to examine the incidence trends between 1975 and 2012 and the projected burden up to 2025 by individual subsites (oral cavity cancer, oropharyngeal cancer, and laryngeal cancer), age, sex, health board region, and socioeconomic status.
The cohort study included all patients diagnosed with oral cancer between 2010 and 2012 and registered on the Scottish Cancer Registry. The individual patient data were linked to NHS dental service activity in the two years prior to diagnosis, and this linked cohort dataset and published NHS Scotland dental workforce and registration and participation statistics were used to examine dental attendance rates and the feasibility of early detection of oral cancer in the primary dental care setting.
The individual patient data from the cohort were also linked to the hospital outpatient, hospital inpatient/day case, primary dental care, and general practitioner prescription databases. These four healthcare services were selected based on data availability. The linked data were used to examine all healthcare service contacts made by the cohort in the two years prior to referral. Additionally, a preliminary exploration of the referral period (defined as the one-month period prior to diagnosis) was also undertaken.

Results and conclusions: The findings of this thesis showed that the incidence rates of head and neck cancer had increased in Scotland between 1975 and 2012, and this appeared to be largely driven by a dramatic rise in the rates of oropharyngeal cancer in recent decades. This burden was predicted to continue to rise up to 2025, with the rates of oropharyngeal cancer bypassing the rates of oral cavity cancer, which were expected to exhibit only a modest increase. Males, individuals above 60 years of age, and those from the most deprived areas of Scotland consistently exhibited the highest rates of cancer, irrespective of subsite. Moreover, an almost dose-like effect was seen to exist, with the rates of cancer increasing with the level of deprivation. Therefore, contrary to previous reports that oropharyngeal cancer exhibited an inverse socioeconomic profile, Scotland country-level data showed that those from the most deprived areas consistently bore the greatest incidence burden of head and neck cancer.
Despite these increasing trends, the overall burden of oral cancer in Scotland was relatively low, and just over half of the cohort examined in this thesis had not contacted a general dental practitioner in the two years prior to diagnosis, thus automatically limiting opportunities for early detection. Dentists were estimated to potentially encounter one patient with oral cancer every 10 years, one patient with oral cavity cancer every 17 years, and one patient with oropharyngeal cancer every 25 years. Therefore, strategies for early detection must consider the rarity of oral cancer incidence and the poor dental attendance patterns of patients, and the expectations of dentists in these efforts must be tempered. These results also highlight the importance of improving access and uptake of dental services among those at the highest risk of developing oral cancer (i.e. those from the most deprived communities).
When examining the linked cohort data and undertaking a look-back analysis of their healthcare service contact history, just under half (45%) of the patients diagnosed with oral cancer were seen to have actually visited a primary care dental service clinic in the two years prior to the start of the referral period. However, the majority of the patients with oral cancer had contacted one of the four healthcare services examined (hospital outpatient, hospital inpatient/day-case, primary dental care, and general practitioner prescription) at least once over the same period, suggesting that there were potential or missed opportunities for the early detection of oral cancer in primary dental care and alternative healthcare settings. The proportions of patients contacting the four services increased closer to the start of the referral period, as did the mean number of contacts made with each service. Although not all of these instances would have necessarily been associated with missed opportunities for early detection, it was highly likely that there were potential or missed opportunities amongst at least some of the patients with oral cancer.
The two most common services contacted most recently before the start of the referral period were general practitioner prescription and hospital outpatient, and there was a possibility that these services were the sources of referral. The hospital specialties contacted most frequently during the one-month referral period were ENT, oral surgery, oral and maxillofacial surgery, and general surgery, suggesting that these contacts were likely to have been associated with the signs and symptoms of oral cancer. While no significant opportunities for the early detection of oral cancer in hospital or secondary care settings were identified, these findings demonstrated considerable potential in other primary care settings, particularly general medical practices and community pharmacies.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: head and neck cancer, screening, epidemiology, early detection.
• Subjects: R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine; R Medicine > RK Dentistry
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing > Dental School
• Supervisor's Name: Conway, Professor David
• Date of Award: 2018
• Depositing User: Mitana Purkayastha
• Unique ID: glathesis:2018-9080
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 09 May 2018 14:42
• Last Modified: 21 Jun 2018 07:26
• URI: https://theses.gla.ac.uk/id/eprint/9080

Related URLs

• Article DOI