Machine learning algorithms could potentially be used to classify patients referred on the two-week wait pathway for suspected head and neck cancer. Patients could be classified into ‘predicted cancer’ or ‘predicted non-cancer’ groups.

A variety of machine learning algorithms were assessed using the clinical data of 5082 patients. These patients had previously been referred via the two-week wait pathway for suspected head and neck cancer to two separate tertiary referral centres in the UK. Outcomes from machine learning classification were analysed in comparison to known clinical diagnoses.

Variational logistic regression was the most clinically useful technique of those chosen to perform the analysis and patient classification; the proportion of patients correctly classified as having ‘non-cancer’ was 25.8 per cent, with a false negative rate of 1 out of 1000.

Machine learning algorithms can accurately and effectively classify patients referred with suspected head and neck cancer symptoms.

To assess the role of video endoscopy in evaluating velopharyngeal incompetence and investigate a possible relationship between velopharyngeal incompetence type and speech defect in cleft palate patients.

A prospective study of 28 pre- or post-operative cleft palate patients with speech defects who attended Plastic Surgery–Cleft Palate and ENT out-patient clinics was performed. The velar defect type was determined using a flexible endoscope and findings were video recorded. Speech pathology was assessed using the cleft palate audit protocol for speech.

A significant, clinically relevant relationship was noted between the perceived characteristics of hypernasality and velopharyngeal insufficiency type. Hypernasal speech was a definite clinical indicator of velopharyngeal incompetence, and the type 1 velopharyngeal defect was most common. Type 1 velopharyngeal coronal-type dysfunction was strongly associated with hypernasality (p < 0.05). When speech substitution was noted, type 2 velopharyngeal (or sagittal) incompetence could be predicted (p < 0.05).

In the management of cleft palate patients, it is important that surgical correction of the defect and achieving velopharyngeal competency for speech are performed simultaneously. Pre-operative velopharyngeal endoscopy with speech assessment will define the anatomical and functional bases for velopharyngeal correction and assist in planning and tailoring the pharyngeal flap.

The Camberwell Assessment of Need – Forensic Version (CANFOR) is a standardised assessment tool specifically designed to assess needs for care in forensic psychiatric populations. The original English version of the instrument has shown good psychometric properties. The aim of this study was to validate the Italian version of the CANFOR-staff tool.

After translation and back-translation, the Italian CANFOR tool was administered to a sample of 50 forensic psychiatric patients. Convergent validity was tested using the Brief Psychiatric Rating Scale (BPRS) and the Global Assessment of Functioning (GAF) by applying Kendall's tau-b. Inter-rater and test–retest reliabilities were measured by ICCs for need dimensions (total and unmet) and Cohen's kappa coefficients for individual need items.

Regarding convergent validity, a higher number of needs (total and unmet) were associated with more severe psychiatric symptoms (BPRS). Higher numbers of unmet needs were also associated with lower levels of global functioning (GAF). ICCs for total and unmet needs scores indicated a good level of agreement for inter-rater reliability and a very good level for test–retest, respectively. Regarding the specific items, inter-rater Cohen's kappa was high (moderate to very good agreement) for 18 items in relation to the presence of a need and for 15 items in the rating of an unmet need, whereas Cohen's kappa for test–retest reliability was very high for all the items in the presence of a need and high for 18 of the unmet need domains.

The Italian version of CANFOR has adequate psychometric properties. It can be considered a promising instrument for the assessment of needs of forensic psychiatric patients.

To investigate and compare the performance of head mirrors and headlights during otolaryngological examination.

The illuminance and illumination field of each device were measured and compared. Visual identification and visual acuity were also measured, in 13 medical students and 10 otolaryngology specialists.

The illuminance (mean ± standard deviation) of the LumiView, Kimscope 1 W and Kimscope 3 W headlights and a standard head mirror were 352.3 ± 9, 92.3 ± 4.5, 438 ± 15.7 and 68.3 ± 1.2 lux, respectively. The illumination field of the head mirror (mean ± standard deviation) was 348 ± 29.8 grids, significantly greater than that of the Kimscope 3 W headlight (183 ± 9.2 grids) (p = 0.0017). The student group showed no statistically significant difference between visual identification with the best headlight and the head mirror (score means ± standard deviations: 56.2 ± 9 and 53.3 ± 14.1, respectively; p = 0.3). The expert group scored significantly higher for visual identification with head mirrors versus headlights (59.7 ± 3.3 vs 55.2 ± 5.8, respectively; p = 0.0035), but showed no difference for visual acuity.

Despite the advantages of headlight illumination, head mirrors provided better, shadow-free illumination. Despite no differences amongst students, head mirrors performed better than headlights in experienced hands.

The Functioning Assessment Short Test (FAST) is a useful instrument for the assessment of overall functioning of people with bipolar disorder, showing good psychometric properties. The aim of this study is to validate the Italian version of FAST.

Translation and back-translation of the original FAST Spanish version were performed. Participants with bipolar disorder (n = 132) and healthy controls (n = 132) completed the FAST as a part of an assessment package including the Montgomery–Asberg Depression Rating Scale and the Young Mania Rating Scale. Internal consistency, inter-rater reliability, construct and discriminant validity were assessed.

The FAST Italian version showed good internal consistency, inter-rater reliability and discriminant validity. The cut-off discriminating patients from controls was 15, with a sensitivity of 0.79 and a specificity of 0.80. Principal component analysis with oblique rotation showed factor loadings consistent with the a priori structure of the instrument.

This study confirmed the psychometric properties of FAST and extended its generalization and validity to the Italian population.

Knowing the pulse rate of a patient in a medical emergency can help to determine patient acuity and the level of medical care required. Little evidence exists regarding the ability of a 911 layperson-caller to accurately determine a conscious patient's pulse rate.

The hypothesis of this study was that, when instructed by a trained emergency medical dispatcher (EMD) using the scripted Medical Priority Dispatch System (MPDS) protocol Pulse Check Diagnostic Tool (PCDxT), a layperson-caller can detect a carotid pulse and accurately determine the pulse rate in a conscious person.

This non-randomized and non-controlled prospective study was conducted at three different public locations in the state of Utah (USA). A healthy, mock patient's pulse rate was obtained using an electrocardiogram (ECG) monitor. Layperson-callers, in turn, initiated a simulated 911 phone call to an EMD call-taker who provided instructions for determining the pulse rate of the patient. Layperson accuracy was assessed using correlations between the layperson-caller's finding and the ECG reading.

Two hundred sixty-eight layperson-callers participated; 248 (92.5%) found the pulse of the mock patient. There was a high correlation between pulse rates obtained using the ECG monitor and those found by the layperson-callers, overall (94.6%, P < .001), and by site, gender, and age.

Layperson-callers, when provided with expert, scripted instructions by a trained 911 dispatcher over the phone, can accurately determine the pulse rate of a conscious and healthy person. Improvements to the 911 instructions may further increase layperson accuracy.

Scott G, Clawson J, Rector M, Massengale D, Thompson M, Patterson B, Olola CHO. The accuracy of emergency medical dispatcher-assisted layperson-caller pulse check using the Medical Priority Dispatch System protocol. Prehosp Disaster Med. 2012;27(3):1-8.

The trauma symptom inventory (TSI; Briere, 1995) is a useful instrument for the assessment of post-traumatic and common trauma-related mental health symptoms. The purpose of the study was to validate the Italian version of the original TSI.

Participants from non-clinical (n = 285), clinical (n = 110) and post-traumatic (n = 30) samples completed the TSI as part of a battery that included self-report measures of trauma exposure [MMPI-2 PK scale and Impact of Event Scale-Revised (IES-R)] and of psychological symptoms [brief symptom inventory (BSI) and symptom questionnaire (SQ)]. TSI validity scales were compared with MMPI-2 validity scales in order to assess convergent validity.

The TSI Italian version showed adequate internal consistency reliability and a good convergent validity. Discriminant function analysis indicates a classification accuracy of TSI scales of 90% for true-positive and 91.4% for true-negative post-traumatic stress disorder (PTSD) cases. A revised three-factor structural model, which demonstrated an adequate and the best fit for the data, was proposed.

The study extended the generalization and validity of TSI and provided some suggestions for eventually revisiting factorial structure of the questionnaire.