A web-based prediction score for head and neck cancer referrals

Appropriateness of two-week wait head and neck cancer referrals to a district general hospital

Objectives The NHS advise urgent referral of patients with suspected head and neck cancers to secondary care to be seen via a two-week wait pathway. The objective of this review was to analyse the two-week wait head and neck cancer referrals to a district general hospital and to identify the prevalence of oral cancer.Materials and methods Patients referred via an urgent two-week wait cancer pathway during the period of 12 October 2020 to 19 January 2022 were identified. Data were extracted and analysed for referral source, patient sex, whether or not a biopsy was undertaken, and the number of patients with a final positive cancer diagnosis.Results Overall, 883 two-week wait referrals were received. Most referrals came from general medical practitioners (50%) followed by general dental practitioners (37%). A total of 379 patients (46%) underwent a biopsy, special investigations, or internal referral to another speciality. The overall prevalence of cancer was 6.2%. Most referrals received were for commonly occurring benign conditions.Conclusion Despite many two-week wait suspected cancer referrals, only a small percentage of patients go on to be diagnosed with head and neck cancer. These results highlight the number of avoidable referrals, which ultimately impact patient waiting lists and clinician time.

Risk prediction models for head and neck cancer: A rapid review

Background

Cancer risk assessment models are used to support prevention and early detection. However, few models have been developed for head and neck cancer (HNC).

Methods

A rapid review of Embase and MEDLINE identified n = 3045 articles. Following dual screening, n = 14 studies were included. Quality appraisal using the PROBAST (risk of bias) instrument was conducted, and a narrative synthesis was performed to identify the best performing models in terms of risk factors and designs.

Results

Six of the 14 models were assessed as "high" quality. Of these, three had high predictive performance achieving area under curve values over 0.8 (0.87-0.89). The common features of these models were their inclusion of predictors carefully tailored to the target population/anatomical subsite and development with external validation.

Conclusions

Some existing models do possess the potential to identify and stratify those at risk of HNC but there is scope for improvement.

Interventions to improve early cancer diagnosis of symptomatic individuals: a scoping review

OBJECTIVES

To summarise the current evidence regarding interventions for accurate and timely cancer diagnosis among symptomatic individuals.

DESIGN

A scoping review following the Joanna Briggs Institute's methodological framework for the conduct of scoping reviews and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews checklist.

DATA SOURCES

MEDLINE (Ovid), CINAHL (EBSCOhost) and PsycINFO (Ovid) bibliographic databases, and websites of relevant organisations. Published and unpublished literature (grey literature) of any study type in the English language were searched for from January 2017 to January 2021.

ELIGIBILITY AND CRITERIA

Study participants were individuals of any age presenting at clinics with symptoms indicative of cancer. Interventions included practice guidelines, care pathways or other initiatives focused on achieving predefined benchmarks or targets for wait times, streamlined or rapid cancer diagnostic services, multidisciplinary teams and patient navigation strategies. Outcomes included accuracy and timeliness of cancer diagnosis.

DATA EXTRACTION AND SYNTHESIS

We summarised findings graphically and descriptively.

RESULTS

From 21 298 retrieved citations, 88 unique published articles and 16 unique unpublished documents (on 18 study reports), met the eligibility for inclusion. About half of the published literature and 83% of the unpublished literature were from the UK. Most of the studies were on interventions in patients with lung cancer. Rapid referral pathways and technology for supporting and streamlining the cancer diagnosis process were the most studied interventions. Interventions were mostly complex and organisation-specific. Common themes among the studies that concluded intervention was effective were multidisciplinary collaboration and the use of a nurse navigator.

CONCLUSIONS

Multidisciplinary cooperation and involvement of a nurse navigator may be unique features to consider when designing, delivering and evaluating interventions focused on improving accurate and timely cancer diagnosis among symptomatic individuals. Future research should examine the effectiveness of the interventions identified through this review.

Application of Artificial Intelligence in Community-Based Primary Health Care: Systematic Scoping Review and Critical Appraisal

BACKGROUND

Research on the integration of artificial intelligence (AI) into community-based primary health care (CBPHC) has highlighted several advantages and disadvantages in practice regarding, for example, facilitating diagnosis and disease management, as well as doubts concerning the unintended harmful effects of this integration. However, there is a lack of evidence about a comprehensive knowledge synthesis that could shed light on AI systems tested or implemented in CBPHC.

OBJECTIVE

We intended to identify and evaluate published studies that have tested or implemented AI in CBPHC settings.

METHODS

We conducted a systematic scoping review informed by an earlier study and the Joanna Briggs Institute (JBI) scoping review framework and reported the findings according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analysis-Scoping Reviews) reporting guidelines. An information specialist performed a comprehensive search from the date of inception until February 2020, in seven bibliographic databases: Cochrane Library, MEDLINE, EMBASE, Web of Science, Cumulative Index to Nursing and Allied Health Literature (CINAHL), ScienceDirect, and IEEE Xplore. The selected studies considered all populations who provide and receive care in CBPHC settings, AI interventions that had been implemented, tested, or both, and assessed outcomes related to patients, health care providers, or CBPHC systems. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Two authors independently screened the titles and abstracts of the identified records, read the selected full texts, and extracted data from the included studies using a validated extraction form. Disagreements were resolved by consensus, and if this was not possible, the opinion of a third reviewer was sought. A third reviewer also validated all the extracted data.

RESULTS

We retrieved 22,113 documents. After the removal of duplicates, 16,870 documents were screened, and 90 peer-reviewed publications met our inclusion criteria. Machine learning (ML) (41/90, 45%), natural language processing (NLP) (24/90, 27%), and expert systems (17/90, 19%) were the most commonly studied AI interventions. These were primarily implemented for diagnosis, detection, or surveillance purposes. Neural networks (ie, convolutional neural networks and abductive networks) demonstrated the highest accuracy, considering the given database for the given clinical task. The risk of bias in diagnosis or prognosis studies was the lowest in the participant category (4/49, 4%) and the highest in the outcome category (22/49, 45%).

CONCLUSIONS

We observed variabilities in reporting the participants, types of AI methods, analyses, and outcomes, and highlighted the large gap in the effective development and implementation of AI in CBPHC. Further studies are needed to efficiently guide the development and implementation of AI interventions in CBPHC settings.

Patterns of urgent hoarseness referrals to ENT-When should we be suspicious of cancer?

BACKGROUND

Current UK referral criteria stipulate that hoarseness should be persistent to merit 2 week wait (2WW) or urgent suspicion of cancer (USOC) referral. This study delineates patterns of hoarseness presentation with a view to assisting referral pathways, and whereby reassurance could be provided.

METHODS

A pre-existing database of patients referred with hoarseness under the urgent suspicion of cancer (USOC) category was analysed. Univariate and multivariate analyses were performed on a variety of demographic and comorbid features to produce odds ratios (OR) of features either related or not related to laryngeal cancer.

RESULTS

Of 698 consecutive hoarseness referrals were studied. In these referrals there were 506(73%) with persistent hoarseness and 192(27%) with intermittent hoarseness. The most significant patient variables related to laryngeal cancer were persistent hoarseness (OR 4.97), recreational drug use (OR 4.94), male gender (OR 4.01) and weight loss (OR 3.75). Significant patient variables present not related to laryngeal cancer diagnosis were intermittent hoarseness (OR 0.2), the presence of cough (OR 0.2), globus sensation (OR 0.25) and recent viral infection (OR 0.29).

CONCLUSION

The strongest association with cancer is seen in patients that are persistently hoarse. Patients with fluctuating hoarseness do not need an "urgent suspicion of cancer" referral. Additional demographic referral information could help to streamline the referral of these patients, and reassure others.

Recent Advances in the Application of Artificial Intelligence in Otorhinolaryngology-Head and Neck Surgery

This study presents an up-to-date survey of the use of artificial intelligence (AI) in the field of otorhinolaryngology, considering opportunities, research challenges, and research directions. We searched PubMed, the Cochrane Central Register of Controlled Trials, Embase, and the Web of Science. We initially retrieved 458 articles. The exclusion of non-English publications and duplicates yielded a total of 90 remaining studies. These 90 studies were divided into those analyzing medical images, voice, medical devices, and clinical diagnoses and treatments. Most studies (42.2%, 38/90) used AI for image-based analysis, followed by clinical diagnoses and treatments (24 studies). Each of the remaining two subcategories included 14 studies. Machine learning and deep learning have been extensively applied in the field of otorhinolaryngology. However, the performance of AI models varies and research challenges remain.

An audit to analyse the two-week wait pathway at an oral cancer specialist district general hospital

Background The incidence of head and neck cancers is increasing, alongside a decrease in associated mortality. Currently, medical and dental practitioners can refer patients to be seen urgently within two weeks. The appropriateness of these referrals has been disputed. In 2020, the Department of Health aims for patients to be given cancer diagnoses within 28 days from referral.

Methods A retrospective audit was conducted for all patients referred under the two-week wait pathway in a six-month period. In the first cycle of this audit, one month's worth of urgent referrals were analysed; given the small sample size, very few recommendations could be made. The audit cycle was repeated and it analysed six months' worth of data, which gave a much more representative study. All patients were analysed to see if the 14-day period had been breached. Positive cancer patients were further assessed to see if their diagnosis had been given within 28 days and treatments within 62 days.

Results Of the 569 patients seen, there was a positive malignancy diagnostic yield of 7.38%. Nineteen patients breached the 14-day wait. Of the positive patients, 45.2% received their diagnosis more than 28 days from referral, and 22.2% of these patients received treatment after 62 days.

Conclusion The department performed well despite the high number of referrals. This audit has touched on some key issues which have been discussed in detail in this article. Furthermore, this audit recommends a concerted effort to improve oral cancer detections skills among GDPs and GMPs. While all referrals may be appropriate from a primary care point of view, this audit makes it apparent that better differentiation is needed between malignant and routinely manageable lesions. All secondary care units alongside general practitioners can learn from the findings of this audit.

Head and neck cancer risk calculator (HaNC-RC)-V.2. Adjustments and addition of symptoms and social history factors

OBJECTIVES

Head and neck cancer (HNC) diagnosis through the 2-week wait, urgent suspicion of cancer (USOC) pathway has failed to increase early cancer detection rates in the UK. A head and neck cancer risk calculator (HaNC-RC) has previously been designed to aid referral of high-risk patients to USOC clinics (predictive power: 77%). Our aim was to refine the HaNC-RC to increase its prediction potential.

DESIGN

Following sample size calculation, prospective data collection and statistical analysis of referral criteria and outcomes.

SETTING

Large tertiary care cancer centre in Scotland.

PARTICIPANTS

3531 new patients seen in routine, urgent and USOC head and neck (HaN) clinics.

MAIN OUTCOME MEASURES

Data collected were as follows: demographics, social history, presenting symptoms and signs and HNC diagnosis. Univariate and multivariate regression analysis were performed to identify significant predictors of HNC. Internal validation was performed using 1000 sample bootstrapping to estimate model diagnostics included the area under the receiver operator curve (AUC), sensitivity and specificity.

RESULTS

The updated version of the risk calculator (HaNC-RC v.2) includes age, gender, unintentional weight loss, smoking, alcohol, positive and negative symptoms and signs of HNC. It has achieved an AUC of 88.6% with two recommended triage referral cut-offs to USOC (cut-off: 7.1%; sensitivity: 85%, specificity: 78.3%) or urgent clinics (cut-off: 2.2%; sensitivity: 97.1%; specificity of 52.9%). This could redistribute cancer detection through USOC clinics from the current 60.9%-85.2%, without affecting total numbers seen in each clinical setting.

CONCLUSIONS

The use of the HaNC-RC v.2 has a significant potential in both identifying patients at high risk of HNC early thought USOC clinics but also improving health service delivery practices by reducing the number of inappropriately urgent referrals.

Two week wait referral criteria - heading in the right direction?

OBJECTIVES

The National Institute for Health and Care Excellence referral guidelines prompting urgent two-week referrals were updated in 2015. Additional symptoms with a lower threshold of 3 per cent positive predictive values were integrated. This study aimed to examine whether current pan-London urgent referral guidelines for suspected head and neck cancer lead to efficient and accurate referrals by assessing frequency of presenting symptoms and risk factors, and examining their correlation with positive cancer diagnoses.

METHODS

The risk factors and symptoms of 984 consecutive patients (over a six-month period in 2016) were collected retrospectively from urgent referral letters to University College London Hospital for suspected head and neck cancer.

RESULTS

Only 37 referrals (3.76 per cent) resulted in a head and neck cancer diagnosis. Four of the 23 recommended symptoms demonstrated statistically significant results. Nine of the 23 symptoms had a positive predictive value of over 3 per cent.

CONCLUSION

The findings indicate that the current referral guidelines are not effective at detecting patients with cancer. Detection rates have decreased from 10-15 per cent to 3.76 per cent. A review of the current head and neck cancer referral guidelines is recommended, along with further data collection for comparison.

Referrals for head and neck cancer in England and The Netherlands: an international qualitative study of the views of secondary-care surgical specialists

One-year survival after head and neck cancer in England has been reported to be worse than that in Europe, despite five-year conditional survival being similar, which implies that patients present later in England. One country with better rates is The Netherlands. There are many possible causes, one of which may be the system of referral from primary to secondary care. We have compared the views of secondary care specialists in the two countries about their systems for referral, and identified aspects that might have an impact on outcomes. We organised semistructured qualitative interviews of surgical specialists in head and neck cancer in England and The Netherlands (n=12 in each). The most common theme was communication between primary care and specialists. Surgeons in England identified this as the aspect most lacking under the English "two-week" rule, while Dutch specialists felt that the good communication in their system was one of its best points. Other themes included the educational needs of primary care practitioners, criticism of "tick box" referrals in England, and too many patients referred who do not have cancer. Overall, specialists in both countries identified good aspects of their respective referral systems, but those in England felt strongly that the "two-week" rule/NICE guidance system could be improved with better direct communication between primary and secondary care, which might improve the speed and quality of referrals, reduce unnecessary ones, and assist in educating primary care physicians. It is not clear whether such improvements would improve survival, but further research and piloting of such a system should be considered in England.