Large-scale external validation and comparison of prognostic models: an application to chronic obstructive pulmonary disease.

BACKGROUND: External validations and comparisons of prognostic models or scores are a prerequisite for their use in routine clinical care but are lacking in most medical fields including chronic obstructive pulmonary disease (COPD). Our aim was to externally validate and concurrently compare prognostic scores for 3-year all-cause mortality in mostly multimorbid patients with COPD. METHODS: We relied on 24 cohort studies of the COPD Cohorts Collaborative International Assessment consortium, corresponding to primary, secondary, and tertiary care in Europe, the Americas, and Japan. These studies include globally 15,762 patients with COPD (1871 deaths and 42,203 person years of follow-up). We used network meta-analysis adapted to multiple score comparison (MSC), following a frequentist two-stage approach; thus, we were able to compare all scores in a single analytical framework accounting for correlations among scores within cohorts. We assessed transitivity, heterogeneity, and inconsistency and provided a performance ranking of the prognostic scores. RESULTS: Depending on data availability, between two and nine prognostic scores could be calculated for each cohort. The BODE score (body mass index, airflow obstruction, dyspnea, and exercise capacity) had a median area under the curve (AUC) of 0.679 [1st quartile-3rd quartile = 0.655-0.733] across cohorts. The ADO score (age, dyspnea, and airflow obstruction) showed the best performance for predicting mortality (difference AUCADO - AUCBODE = 0.015 [95% confidence interval (CI) = -0.002 to 0.032]; p = 0.08) followed by the updated BODE (AUCBODE updated - AUCBODE = 0.008 [95% CI = -0.005 to +0.022]; p = 0.23). The assumption of transitivity was not violated. Heterogeneity across direct comparisons was small, and we did not identify any local or global inconsistency. CONCLUSIONS: Our analyses showed best discriminatory performance for the ADO and updated BODE scores in patients with COPD. A limitation to be addressed in future studies is the extension of MSC network meta-analysis to measures of calibration. MSC network meta-analysis can be applied to prognostic scores in any medical field to identify the best scores, possibly paving the way for stratified medicine, public health, and research.

Influence of the COPD Assessment Test respiratory item score on the decision to hospitalize patients with disease exacerbation in a hospital emergency department. / Influencia de los ítems respiratorios del COPD Assessment Test (CAT) en la decisión de ingreso de las agudizaciones de enfermedad pulmonar obstructiva crónica (EPOC) atendidos en urgencias hospitalarias.

OBJECTIVES: The COPD Assessment Test (CAT) measures quality of life in patients with chronic obstructive pulmonary disease (COPD) as well as disease impact on activities of daily living. The questionnaire consists of 8 items related to breathing (cough, phlegm, chest tightness, and breathlessness) and other symptoms (low energy level, sleep disturbances, limitations on daily activities, and confidence when leaving the home). We investigated the relative impact of respiratory versus nonrespiratory scoring on the total CAT score at different moments in the course of COPD exacerbations: baseline (24 hours before an exacerbation), during the exacerbation, 15 days later, and 2 months later. To assess the influence of the respiratory item score on decisions to hospitalize patients treated for exacerbated COPD in our hospital emergency department (ED). MATERIAL AND METHODS: Prospective cohort study. We recruited patients who came to our ED for symptoms consistent with exacerbated COPD. Sociodemographic and clinical data were recorded. Clinical information, including treatments pleustarted in the ED and CAT scores, were also recorded. The event was defined as highly symptomatic if the patient's score was 3 points or higher on at least 3 of the 4 respiratory items at baseline. The outcome measures for the first objective were the total CAT score and item scores at the 4 time points before (baseline), during (ED), and after the exacerbation. The outcome for the second objective was hospital admission. RESULTS: A total of 587 patients were included. The mean (SD) total CAT score was 13.48 (7.29) at baseline, 24.86 (7.25) in the ED, 14.7 (7.47) at 15 days, and 13.45 (7.36) at 2 months. The respiratory item scores accounted for a mean 53.4% (20.76%) of the total score at baseline and 48.2% (11.47%) of the total score in the ED. Eighty-two patients (14.0%) were classified as being highly symptomatic. A total of 359 (61.2%) were admitted. Predictors of hospital admission were classification as highly symptomatic, odds ratio (OR, 3.045; 95% CI, 1.585-5.852, P .001), dyspnea at rest (OR, 2.906; 95% CI:1.943-4.346, P .001), and start of the following treatments in the ED: oxygen therapy (OR, 4.550; 95% CI, 3.056-6.773; P .0001), diuretic (OR, 1.754; 95% CI, 1.091-2.819; P = .02), and intravenous antibiotics (OR, 1.536; 95% CI, 1.034-2.281; P = .03). The model achieved an area under the receiver operating characteristic curve of 0.80 (95% CI, 0.763-0.836). CONCLUSION: Hospital admission from the ED is highly likely in patients with COPD exacerbation who have high baseline CAT scores, dyspnea at rest in the ED, and require oxygen therapy, diuretics, or intravenous antibiotics in the ED. The total CAT score and scores on respiratory items provide a tool for tailoring pharmacalogic and nonpharmacologic treaments and can facilitate follow-up evaluations.

OBJETIVO: El CAT (COPD Assessment Test) es un cuestionario de calidad de vida que mide el impacto que la enfermedad pulmonar obstructiva crónica (EPOC) está teniendo en el bienestar y vida diaria de los pacientes. Consta de 8 ítems divididos en 4 respiratorios y 4 no respiratorios. Conocer el impacto de las puntuaciones de los ítems respiratorios y no respiratorios en la puntuación CAT total, en diferentes momentos de la exacerbación de EPOC (24 horas antes de la exacerbación o basal, en la exacerbación, a los 15 días y a los 2 meses). Secundariamente, se valoró la influencia de los ítems respiratorios de la puntuación CAT total, en la decisión de ingreso de los pacientes atendidos por exacerbación de EPOC (EA-EPOC) en un servicio de urgencias hospitalario (SUH). METODO: Estudio de cohortes prospectivo. Se reclutaron pacientes que acudían al SUH con síntomas compatibles con EA-EPOC. La variable "Paciente respiratorio altamente sintomático"(PRAS) se definió como el paciente que tiene 3 puntos o más en al menos 3 de los 4 ítems respiratorios del CAT basal. Las variables de resultado fueron para el primer objetivo: la puntuación CAT total y desglosada por ítems, en los 4 momentos estudiados. Para el segundo objetivo fue el ingreso hospitalario. RESULTADOS: Se incluyeron 587 pacientes. La media de la puntuación CAT total basal fue 13,48 (7,29), en urgencias fue 24,86 (7,25), a los 15 días fue 14,7 (7,47) y a los 2 meses 13,45 fue (7,36). La proporción sobre la puntuación CAT basal total de los ítems respiratorios fue de 53,4% (20,76) y en el momento de llegar a urgencias del 48,2% (11,47). Los PRAS fueron 82 (14,0%). Ingresaron 359 pacientes (61,2%). Los predictores de ingreso hospitalario fueron: PRAS (OR 3,045, IC 95%: 1,585-5,852, p 0,001), disnea de reposo (OR 2,906, IC 95%: 1,943-4,346, p 0,001) y algunos tratamientos instaurados en el SUH (oxigenoterapia: OR 4,550, IC 95%: 3,056-6,773, p 0,001; diurético: OR 1,754, IC 95%: 1,091-2,819, p = 0,02; y antibiótico iv: OR 1,536, IC 95%: 1,034-2,281, p = 0,03). Este modelo logra un área bajo la curva COR de 0,80 (IC 95%: 0,763-0,836). CONCLUSIONES: En pacientes con EA-EPOC atendidos en urgencias, la alta puntuación de ítems respiratorios en el CAT basal, la disnea de reposo a su llegada al SUH y varios de los tratamientos instaurados en urgencias (oxigenoterapia, diuréticos y antibioterapia intravenosa) demostraron tener buena capacidad de predicción de ingreso hospitalario. La puntuación CAT total así como la puntuación en los ítems respiratorios del mismo son una herramienta que podría ayudar al clínico a individualizar el tratamiento o los controles posteriores.