Comparison of Recording Positions of Physical Activity in Patients with Severe COPD Undergoing LTOT

Physical activity as a predictor of thirty-day hospital readmission after a discharge for a clinical exacerbation of chronic obstructive pulmonary disease

RATIONALE

Because physical inactivity in chronic obstructive pulmonary disease (COPD) predicts health care use and mortality, we prospectively evaluated the relationship of this variable to the frequency of 30-day readmissions after a hospitalization for an exacerbation.

METHODS

Consented patients discharged after an exacerbation of COPD were asked to wear a GT3X+ accelerometer (ActiGraph, Pensacola, FL) continuously on the wrist for 30 days after hospital discharge. Vector magnitude units (VMU), the sum of movements in three planes over each minute of use, were recorded. Higher physical activity for each minute was defined by a VMU threshold of at least 3,000 counts. Those patients with fewer than 60 minutes/day over the first week were considered inactive.

MEASUREMENTS AND MAIN RESULTS

Fifty-four study patients were discharged from the hospital and 38 underwent activity testing. In the latter subgroup, all-cause hospital readmissions within 30 days occurred in 12 patients (32%). Minutes of higher physical activity per day over the first week after discharge were considerably lower in those eventually readmitted than in those who remained as outpatients: 42 ± 14 (SE) versus 114 ± 19 minutes, respectively (P = 0.02). In addition, physical activity decreased over time in those who were eventually readmitted, but increased in those who were not readmitted. Those with lower physical activity over Week 1 after discharge were more likely to have 30-day all-cause readmissions than those with higher activity: odds ratio, 6.7, P = 0.02. In multivariate testing, both physical inactivity and a history of two or more hospitalizations for exacerbations in the preceding year predicted 30-day readmission.

CONCLUSIONS

These findings underscore the importance of physical activity as a predictor of this type of health care use outcome.

Classification of exacerbation episodes in chronic obstructive pulmonary disease patients

BACKGROUND

Chronic obstructive pulmonary disease (COPD) is a progressive disease affecting the airways, which constitutes a major cause of chronic morbidity and a significant economic and social burden throughout the world. Despite the fact that in COPD patients exacerbations are common acute events causing significant and often fatal worsening of symptoms, an accurate prognostication continues to be difficult.

OBJECTIVES

To build computational models capable of distinguishing between normal life days from exacerbation days in COPD patients, based on physical activity measured by accelerometers.

METHODS

We recruited 58 patients suffering from COPD and measured their physical activity with accelerometers for 10 days or more, from August 2009 to March 2010. During this period we recorded six exacerbation episodes in the patients, accounting for 37 days. We were able to analyse data for 52 patients (369 patient days), and extracted three distinct sets of features from the data, one set of basic features such as average, one set based on the frequency domain and the last exploring the cross-information among sensors pairs. These were used by three machine-learning techniques (logarithmic regression, neural networks, support vector machines) to distinguish days with exacerbation events from normal days.

RESULTS

The support vector machine classifier achieved an AUC of 90% ± 9, when supplied with a set of features resulting from sequential feature selection method. Neu- ral networks achieved an AUC of 83% ± 16 and the logarithmic regression an AUC of 67% ± 15.

CONCLUSIONS

None of the individual feature sets provided robust for reasonable classification of PA recording days. Our results indicate that this approach has the potential to extract useful information for, but are not robust enough for medical application of the system.