Objective sputum colour assessment and clinical outcomes in bronchiectasis: data from the European Bronchiectasis Registry (EMBARC).

BACKGROUND: A validated 4-point sputum colour chart can be used to objectively evaluate the levels of airway inflammation in bronchiectasis patients. In the European Bronchiectasis Registry (EMBARC), we tested whether sputum colour would be associated with disease severity and clinical outcomes. METHODS: We used a prospective, observational registry of adults with bronchiectasis conducted in 31 countries. Patients who did not produce spontaneous sputum were excluded from the analysis. The Murray sputum colour chart was used at baseline and at follow-up visits. Key outcomes were frequency of exacerbations, hospitalisations for severe exacerbations and mortality during up to 5-year follow-up. RESULTS: 13 484 patients were included in the analysis. More purulent sputum was associated with lower forced expiratory volume in 1 s (FEV1), worse quality of life, greater bacterial infection and a higher bronchiectasis severity index. Sputum colour was strongly associated with the risk of future exacerbations during follow-up. Compared to patients with mucoid sputum (reference group), patients with mucopurulent sputum experienced significantly more exacerbations (incident rate ratio (IRR) 1.29, 95% CI 1.22-1.38; p<0.0001), while the rates were even higher for patients with purulent (IRR 1.55, 95% CI 1.44-1.67; p<0.0001) and severely purulent sputum (IRR 1.91, 95% CI 1.52-2.39; p<0.0001). Hospitalisations for severe exacerbations were also associated with increasing sputum colour with rate ratios, compared to patients with mucoid sputum, of 1.41 (95% CI 1.29-1.56; p<0.0001), 1.98 (95% CI 1.77-2.21; p<0.0001) and 3.05 (95% CI 2.25-4.14; p<0.0001) for mucopurulent, purulent and severely purulent sputum, respectively. Mortality was significantly increased with increasing sputum purulence, hazard ratio 1.12 (95% CI 1.01-1.24; p=0.027), for each increment in sputum purulence. CONCLUSION: Sputum colour is a simple marker of disease severity and future risk of exacerbations, severe exacerbations and mortality in patients with bronchiectasis.

Cost-sensitive ordinal classification methods to predict SARS-CoV-2 pneumonia severity.

OBJECTIVE: To study the suitability of costsensitive ordinal artificial intelligence-machine learning (AIML) strategies in the prognosis of SARS-CoV-2 pneumonia severity. MATERIALS & METHODS: Observational, retrospective, longitudinal, cohort study in 4 hospitals in Spain. Information regarding demographic and clinical status was supplemented by socioeconomic data and air pollution exposures. We proposed AI-ML algorithms for ordinal classification via ordinal decomposition and for cost-sensitive learning via resampling techniques. For performance-based model selection, we defined a custom score including per-class sensitivities and asymmetric misprognosis costs. 260 distinct AI-ML models were evaluated via 10 repetitions of 5×5 nested cross-validation with hyperparameter tuning. Model selection was followed by the calibration of predicted probabilities. Final overall performance was compared against five well-established clinical severity scores and against a 'standard' (non-cost sensitive, non-ordinal) AI-ML baseline. In our best model, we also evaluated its explainability with respect to each of the input variables. RESULTS: The study enrolled n = 1548 patients: 712 experienced low, 238 medium, and 598 high clinical severity. d = 131 variables were collected, becoming d ' = 148 features after categorical encoding. Model selection resulted in our best-performing AI-ML pipeline having: a) no imputation of missing data, b) no feature selection (i.e. using the full set of d ' features), c) 'Ordered Partitions' ordinal decomposition, d) cost-based reimbalance, and e) a Histogram-based Gradient Boosting classifier. This best model (calibrated) obtained a median accuracy of 68.1% [67.3%, 68.8%] (95% confidence interval), a balanced accuracy of 57.0% [55.6%, 57.9%], and an overall area under the curve (AUC) 0.802 [0.795, 0.808]. In our dataset, it outperformed all five clinical severity scores and the 'standard' AI-ML baseline. DISCUSSION & CONCLUSION: We conducted an exhaustive exploration of AI-ML methods designed for both ordinal and cost-sensitive classification, motivated by a real-world application domain (clinical severity prognosis) in which these topics arise naturally. Our model with the best classification performance exploited successfully the ordering information of ground truth classes, coping with imbalance and asymmetric costs. However, these ordinal and cost-sensitive aspects are seldom explored in the literature.

Cost Effectiveness of the 13-Valent Pneumococcal Conjugate Vaccination Program in Chronic Obstructive Pulmonary Disease Patients Aged 50+ Years in Spain.

BACKGROUND: Patients with chronic obstructive pulmonary disease (COPD) are at elevated risk of pneumococcal infection. A 13-valent pneumococcal conjugate vaccine (PCV13) was approved for protection against invasive disease and pneumonia caused by Streptococcus pneumoniae in adults. This study estimated the incremental cost-effectiveness ratio (ICER) of vaccinating COPD patients ≥50 years old with PCV13 compared with current vaccination policy (CVP) with 23-valent pneumococcal polysaccharide vaccine. METHODS: A Markov model accounting for the risks and costs for all-cause non-bacteremic pneumonia (NBP) and invasive pneumococcal disease (IPD) was developed. All parameters, such as disease incidence and costs (€; 2015 values), were based on published data. The perspective of the analysis was that of the Spanish National Healthcare System, and the horizon of evaluation was lifetime in the base case. Vaccine effectiveness considered waning effect over time. Outcomes and costs were both discounted by 3% annually. RESULTS: Over a lifetime horizon and for a 629,747 COPD total population, PCV13 would prevent 2224 cases of inpatient NBP, 3134 cases of outpatient NBP, and 210 IPD extra cases in comparison with CVP. Additionally, 398 related deaths would be averted. The ICER was €1518 per quality-adjusted life-year (QALY) gained for PCV13 versus CVP. PCV13 was found to be cost effective versus CVP from a 5-year modelling horizon (1302 inpatient NBP and 1835 outpatient NBP cases together with 182 deaths would be prevented [ICER €25,573/QALY]). Univariate and probabilistic sensitivity analyses confirmed the robustness of the model. CONCLUSIONS: At the commonly accepted willingness-to-pay threshold of €30,000/QALY gained, PCV13 vaccination in COPD patients aged ≥50 years was a cost-effective strategy compared with CVP from 5 years to lifetime horizon in Spain.

[Multidisciplinary guidelines for the management of community-acquired pneumonia]. / Guía multidisciplinar para la valoración pronóstica, diagnóstico y tratamiento de la neumonía adquirida en la comunidad.

Community-acquired pneumonia (CAP) is an infectious respiratory disease with an incidence that ranges from 3 to 8 cases per 1,000 inhabitants per year. This incidence increases with age and comorbidities. Forty per cent of CAP patients require hospitalization and around 10% of these patients are admitted in an Intensive Care Unit (ICU). Several studies have suggested that the implementation of clinical guidelines has a positive impact in the outcome of patients including mortality and length of stay. The more recent and used guidelines are those from Infectious Diseases Society of America/American Thoracic Society, published in 2007, the 2009 from the British Thoracic Society, and that from the European Respiratory Society/European Society of Clinical Microbiology and Infectious Diseases, published in 2010. In Spain, the most recently released guideline is the Sociedad Española de Neumología y Cirugía Torácica-2011 guideline. The present guidelines GNAC are designed to be used by the majority of health-care professionals that can participate in the care of CAP patients including diagnosis, decision of hospital and ICU admission, treatment and prevention. The Centro Cochrane Iberoamericano (CCIB) has participated in summarizing the previous guidelines and in the bibliography search. For each one of the following sections the panel of experts has developed a table with recommendations classified according to its evidence, strength and practical applicability using the Grading of Recommendations of Assessment Development and Evaluations (GRADE) system: 1. Epidemiology, microbiological etiology and antibiotic resistances.2. Clinical and microbiological diagnosis.3. Prognostic scales and decision of hospital admission.4. ICU admission criteria. 5. Empirical and definitive antibiotic treatment.6. Treatment failure. 7. Prevention.

Impact of guidelines on outcome: the evidence.

Pneumonia continues to be the main cause of death due to infection in the world, and it produces a high consumption of healthcare resources. The guidelines established by the scientific societies improve the care of patients with pneumonia. One way of evaluating the effect of the guidelines is to analyze their impact on the prognosis of the infection. To evaluate this effect, cohort studies have been performed using before-after, observational, cost-effectiveness, and, to a lesser degree, randomized designs. The most recent studies show that the implementation of the guidelines is accompanied by an increase in the process of care percentage and a lower inpatient hospital mortality rate- including the first 48 hours and after 30 days. These findings are consistent across various studies, and they have been confirmed in patients admitted to the intensive care unit. Clinical stability is also reached earlier in patients hospitalized for community-acquired pneumonia (CAP) when the antibiotic treatment is begun early and complies with the recommendations. Finally, the choice of antibiotics that adhere to the guidelines is cost-effective in CAP requiring hospitalization, which is responsible for 80% of the total cost of this disease.