Population exacerbation incidence contains predictive information of acute exacerbations in patients with chronic obstructive pulmonary disease in telecare. (March 2018)

• Record Type: Journal Article
• Authors: Kronborg, Thomas; Mark, Lasse; Cichosz, Simon Lebech; Secher, Pernille Heyckendorff; Hejlesen, Ole
• Is Part Of: International journal of medical informatics. Volume 111(2018)
• Page Start: 72
• Page End: 76
• Publication Date: 2018-03
• Subjects: Chronic obstructive pulmonary disease -- COPD -- Exacerbation -- Hospitalization -- Telehealth; Medical informatics -- Periodicals; Information science -- Periodicals; Computers -- Periodicals; Medical technology -- Periodicals; Medical Informatics -- Periodicals; Technology, Medical -- Periodicals; Computers; Information science; Medical informatics; Medical technology; Electronic journals; Periodicals; Electronic journals; 610.285
• DOI: 10.1016/j.ijmedinf.2017.12.026
• Languages: English
• ISSNs: 1386-5056

Abstract

Highlights: We investigated whether the incidences of patient and population exacerbations contain predictive information for COPD exacerbations. Telehomecare data measured on a weekly basis was extracted from 57 patients experiencing 84 hospitalizations for COPD exacerbation. Population exacerbation incidence improves prediction in combination with physiological parameters and symptom questionnaires. Abstract: Objective: Chronic obstructive pulmonary disease (COPD) is a major global burden largely resulting from acute exacerbations. We investigated whether the incidences of patient and population exacerbations contain predictive information for continuous prediction of exacerbations in COPD patients. Methods: Data analysis was performed using home measurements from 1225 patients included in the large-scale telehomecare trial TeleCare North, where data supported 84 exacerbations occurring in 57 patients. Twenty-nine predictors were extracted and validated in two prediction models based on logistic regression. One model without and one model with inclusion of patient and population exacerbation incidences as potential predictors. The predictors were then evaluated by discriminative abilities between periods with and without exacerbation. Results: The optimal predictor combinations provided an average area under the receiver operation characteristics curve of 0.63 with exclusion; inclusion of the population exacerbation incidence provided a curve of 0.74 (p < 0.05). These results were based on a two-fold patient dependent cross-validation. Discussion: The present study has presented how the population exacerbation incidence contains predictive information in the continuous prediction of exacerbations in COPD patients. A system capable of predicting acute exacerbations could potentially prevent some cases of COPD-related complications and increase the health-related quality of life among COPD patients in telecare.