Objectives: To evaluate the role of clinical scores assessing the risk of disease severity in patients with clinical suspicion of obstructive sleep apnea syndrome (OSA). The hypothesis was tested by applying artificial intelligence (AI) to demonstrate its effectiveness in distinguishing between mild–moderate OSA and severe OSA risk. Methods: A support vector machine model (SVM) was developed from the samples included in the analysis (N = 498), and they were split into 75% for training (N = 373) with the remaining for testing (N = 125). Two diagnostic thresholds were selected for OSA severity: mild to moderate (apnea–hypopnea index (AHI) ≥ 5 events/h and AHI < 30 events/h) and severe (AHI ≥ 30 events/h). The algorithms were trained and tested to predict OSA patient severity. Results: The sensitivity and specificity for the SVM model were 0.93 and 0.80 with an accuracy of 0.86; instead, the logistic regression full mode reported a value of 0.74 and 0.63, respectively, with an accuracy of 0.68. After backward stepwise elimination for features selection, the reduced logistic regression model demonstrated a sensitivity and specificity of 0.79 and 0.56, respectively, and an accuracy of 0.67. Conclusion: Artificial intelligence could be applied to patients with symptoms related to OSA to identify individuals with a severe OSA risk with clinical-based algorithms in the OSA framework.

Patients in intensive care units (ICUs) were at higher risk of worsen prognosis and mortality. Here, we aimed to evaluate the ability of the Simplified Acute Physiology Score (SAPS II) to predict the risk of 7-day mortality, and to test a machine learning algorithm which combines the SAPS II with additional patients’ characteristics at ICU admission. We used data from the “Italian Nosocomial Infections Surveillance in Intensive Care Units” network. Support Vector Machines (SVM) algorithm was used to classify 3782 patients according to sex, patient’s origin, type of ICU admission, non-surgical treatment for acute coronary disease, surgical intervention, SAPS II, presence of invasive devices, trauma, impaired immunity, antibiotic therapy and onset of HAI. The accuracy of SAPS II for predicting patients who died from those who did not was 69.3%, with an Area Under the Curve (AUC) of 0.678. Using the SVM algorithm, instead, we achieved an accuracy of 83.5% and AUC of 0.896. Notably, SAPS II was the variable that weighted more on the model and its removal resulted in an AUC of 0.653 and an accuracy of 68.4%. Overall, these findings suggest the present SVM model as a useful tool to early predict patients at higher risk of death at ICU admission.

The transition from adolescence to adulthood is a critical period for the development of healthy behaviors. Yet, it is often characterized by unhealthy food choices. Considering the current pandemic scenario, it is also essential to assess the effects of coronavirus disease-19 (COVID-19) on lifestyles and diet, especially among young people. However, the assessment of dietary habits and their determinants is a complex issue that requires innovative approaches and tools, such as those based on the ecological momentary assessment (EMA). Here, we describe the first phases of the “HEALTHY-UNICT” project, which aimed to develop and validate a web-app for the EMA of dietary data among students from the University of Catania, Italy. The pilot study included 138 students (mean age 24 years, SD = 4.2; 75.4% women), who used the web-app for a week before filling out a food frequency questionnaire with validation purposes. Dietary data obtained through the two tools showed moderate correlations, with the lowest value for butter and margarine and the highest for pizza (Spearman’s correlation coefficients of 0.202 and 0.699, respectively). According to the cross-classification analysis, the percentage of students classified into the same quartile ranged from 36.9% for vegetable oil to 58.1% for pizza. In line with these findings, the weighted-kappa values ranged from 0.15 for vegetable oil to 0.67 for pizza, and most food categories showed values above 0.4. This web-app showed good usability among students, assessed through a 19-item usability scale. Moreover, the web-app also had the potential to evaluate the effect of the COVID-19 pandemic on students’ behaviors and emotions, showing a moderate impact on sedentary activities, level of stress, and depression. These findings, although interesting, might be confirmed by the next phases of the HEALTHY-UNICT project, which aims to characterize lifestyles, dietary habits, and their relationship with anthropometric measures and emotions in a larger sample of students.

While personal characteristics have been evaluated as determinants of dietary choices over the years, only recently studies have looked at the impact of eating context. Examining eating context, however, can be challenging. Here, we propose the use of a web-app for the Ecological Momentary Assessment of dietary habits among 138 college students from Catania (Italy) and therefore for examining the impact of eating context on dietary choices. Eating away from home was associated with lower odds of consuming vegetables, fruits, and legumes and higher odds of consuming processed meat, salty snacks, and alcoholic drinks compared with eating at home. Eating in the company of other people was associated with higher odds of consuming vegetables, red meat, fish, legumes, milk, and sugar-sweetened beverages and lower odds of consuming nuts than eating alone. This study proposed a new way to capture and assess how eating environment might affect dietary habits. Based on our results, meal location and social context have significant effects on the dietary choices of college students, pointing to the need to incorporate these aspects into further epidemiological studies.

The efficacy and safety of varenicline for smoking cessation among individuals who smoke tobacco cigarettes and also use electronic cigarettes (known e-cigarettes or vapes) have not been studied. We aimed to address this knowledge gap and examine predictors for smoking abstinence. In this double-blind, placebo-controlled, single-centre randomised trial in Italy, we enrolled adults who had used an e-cigarette daily for at least 12 months and who also smoked at least one tobacco cigarette per day and had a willingness to quit smoking. 155 participants were randomly assigned to receive either varenicline (n = 78) or matched placebo (n = 77). Varenicline (1 mg, administered twice daily for 12 weeks) was given in combination with smoking cessation counseling in dual users with an intention to quit smoking. Participants in both treatment groups received the same smoking cessation counselling throughout the whole duration of the study. The trial consisted of a 12-week treatment phase followed by a 12-week follow-up. The primary efficacy endpoint was continuous abstinence rate (CAR) in weeks 4-12. Secondary efficacy endpoints were the CAR in weeks 4-24 and 7-day point prevalence of smoking abstinence at weeks 12 and 24. This study is registered in EUDRACT, 2016-000339-42. Between November 2018, and February 2020, 114 participants (61 in the varenicline group and 53 in the placebo group) completed the intervention phase at week 12 and 88 participants (52 in the varenicline group and 36 in the placebo group) completed the follow-up phase at week 24. CARs were significantly higher for the varenicline vs placebo at each time-point: 50.0% vs 16.9% (OR = 4.9; 95% CI, 2.3-10.4; P < 0.0001) between weeks 4 and 12; and 48.7% vs 14.3% (OR = 5.7; 95% CI, 2.6-12.3; P < 0.0001) between weeks 4 and 24. The 7-day point prevalence of smoking abstinence was also higher for the varenicline than placebo at each time point. Adverse events were rated as mild or moderate and rarely led to treatment discontinuation. Our findings indicate that inclusion of varenicline in a cessation programme for adults who smoke and use e-cigarettes with an intention to quit smoking could result in smoking abstinence without serious adverse events. In the absence of evidence from other smoking cessation methods, it could be useful to suggest the use of varenicline in cessation programmes specifically designed to help dual users stop smoking. Further research in larger and more generalisable populations is required to strengthen such a suggestion. Global Research Award for Nicotine Dependence, an independently reviewed competitive grants programmeme funded by Pfizer.