Mortality predictors in disabling chronic obstructive pulmonary disease in old age.

OBJECTIVE: prospectively to evaluate predictors of mortality in elderly patients with disabling chronic obstructive pulmonary disease.
METHODS: 137 (69 men) outpatients, aged 60-89 (mean 73) years with symptomatic disabling chronic obstructive pulmonary disease. We collected baseline demographic and physiological data. Subjects completed the Manchester Respiratory Activities of Daily Living Questionnaire, the Brief Assessment Schedule Depression Cards a screening questionnaire for depression, the Breathing Problems Questionnaire measuring quality of life, and the Montgomery Asberg Depression Rating Scale measuring severity of depression. All subjects were followed prospectively and survival and mortality data were confirmed by contacting general practitioners and scrutinising hospital notes at 30 months.
RESULTS: the mean (standard deviation) of one second forced expiratory volume was 0.89 (0.3) litres. At 30 months, 44 patients (21 men, aged 61-89 [mean 75] years: 32% of the total) had died. Mean (standard deviation) baseline one second forced expiratory volume of those dying was 0.71 (0.2) litres. On logistic regression analysis, predictors of mortality were: Manchester Respiratory Activities Of Daily Living Questionnaire score (odds ratio=0.88, 95% confidence interval=0.80-0.97); pre-bronchodilator one second forced expiratory volume (odds ratio=0.04, confidence interval=0.005-0.32); body mass index (odds ratio=0.87, confidence interval=0.79-0.97); and long term oxygen therapy (odds ratio=3.17, confidence interval=1.04-8.36). Current smoking status, pack-years smoked, depression scores, quality of life scores, co-morbid diseases and social class did not predict mortality.
CONCLUSION: disability, use of long-term oxygen therapy, pre-bronchodilator lung function and body-mass index were independent predictors of mortality in elderly patients with severe chronic obstructive pulmonary disease.