Open Access Research article

Mortality in Western Australian seniors with chronic respiratory diseases: a cohort study

Kristjana Einarsdóttir1*, David B Preen1, Frank M Sanfilippo1, Raylene Reeve2, Jon D Emery2 and C D'Arcy J Holman1

Author Affiliations

1 Centre for Health Services Research, School of Population Health, The University of Western Australia, 35 Stirling Highway, Crawley, 6009 Perth, Australia

2 School of Primary, Aboriginal and Rural Health Care, The University of Western Australia, 35 Stirling Highway, Crawley, 6009 Perth, Australia

For all author emails, please log on.

BMC Public Health 2010, 10:385  doi:10.1186/1471-2458-10-385

Published: 1 July 2010



Relatively few studies have examined survival by pharmacotherapy level and the effects of patient characteristics on mortality by pharmacotherapy level in older chronic respiratory disease (CRD) patients. This study aimed to investigate these issues in older (≥ 65) CRD patients in Western Australia.


We identified 108,312 patients ≥ 65 years with CRD during 1992-2006 using linked medical, pharmaceutical, hospital and mortality databases held by the Commonwealth and State governments. Pharmacotherapy classification levels were designed by a clinical consensus panel. Cox regression was used to investigate the study aim.


Patients using only short acting bronchodilators experienced similar, but slightly worse survival than patients in the highest pharmacotherapy level group using high dose inhaled corticosteroids (ICS) ± long acting bronchodilators (LABs) ± oral steroids. Patients using low to medium dose ICS ± LABs experienced relatively better survival. Also, male gender was associated with all-cause mortality in all patients (HR = 1.72, 95% CI 1.65-1.80) and especially in those in the highest pharmacotherapy level group (HR = 1.97, 95%CI = 1.84-2.10). The P-value of interaction between gender and pharmacotherapy level for the effect on all-cause death was significant (0.0003).


Older patients with CRD not using ICS experienced the worst survival in this study and may benefit from an escalation in therapeutic regime. Males had a higher risk of death than females, which was more pronounced in the highest pharmacotherapy level group. Hence, primary health care should more actively direct disease management to mild-to-moderate disease patients.