Letter to the editor in reference to : "A web-based prediction score for head and neck cancer referrals"

Tikka, Theofano and Paleri, Vinidh (2020) Letter to the editor in reference to : "A web-based prediction score for head and neck cancer referrals". Clinical Otolaryngology, 45 (1). p. 156. ISSN 1749-4478

[img] Text (Tikka-Paleri-CO-2020-Letter-to-the-editor-in-reference-to-a-web-based-prediction)
Tikka_Paleri_CO_2020_Letter_to_the_editor_in_reference_to_a_web_based_prediction.pdf
Accepted Author Manuscript
Restricted to Repository staff only until 17 October 2020.

Download (393kB) | Request a copy from the Strathclyde author

    Abstract

    Sir, We read with interest the article by Lau et al (2018) on the design of a web‐based prediction score for head and neck cancer (HNC) referrals. The authors state that there are no similar scoring systems available in the literature that are web‐based and applicable to the two‐week‐wait referrals. Over the last years, risk calculators for common cancer have been extensively explored and are available online, aiming to improve cancer referral pathways and detection.1, 2 A HNC risk calculator (HNC‐RC) based on symptoms and demographics has also been developed and is available online (Sensitivity: 74.8%; specificity: 65.9%; and overall predictive power (AUC): 0.77).3 It has also been externally validated maintaining its discriminatory ability (sensitivity: 79.3%; specificity: 68.6%; and AUC: 0.81). In their abstract, Lau et al have stated that logistic regression and artificial network machine approached have been used. Despite this, only the former was employed, as mentioned in their methodology section, due to time‐related restrictions and effort required to ensure an error‐free algorithm.