Pulmonary disease among inpatient decedents: Impact of schizophrenia

Abstract

Objectives

Determine the risk associated with schizophrenia for common pulmonary illness (pneumonia and chronic obstructive pulmonary disorder (COPD)) during the last year of life.

Methods

Inpatient decedents in Veterans (VA) hospitals in 2002 (N = 27,798) were identified. Logistic regression modeled diagnosis of pulmonary illness in either the final year or final admission as a function of schizophrenia, smoking history and other covariates.

Results

Among decedents, 943 (3%) had schizophrenia, 3% were women, most were white (76%) or African–American (18%), and average age at death was 72.4 years (SD 11.5). Three-fifths received VA outpatient care in the year prior to death. Among those with schizophrenia, only two-fifths had outpatient care. Pneumonia was more common among schizophrenia patients (38% vs 31%) as was COPD (46% vs 38%). In models controlling for history of smoking and other covariates, schizophrenia was a risk factor for pulmonary disease in the last year of life (OR = 1.9, 95% CI 1.6–2.2) but less so for final-stay pulmonary disease (OR = 1.5, 95% CI 1.3–1.7).

Conclusions

VA inpatient decedents with schizophrenia were at increased risk for pneumonia and COPD, independent of smoking indicators. Clinicians treating schizophrenia patients need to be especially alert to potential comorbid medical conditions and ensure vulnerable patients receive appropriate care.

Introduction

Patients with schizophrenia are known to be at increased risk for numerous comorbid medical conditions including diabetes, coronary artery disease, and digestive and liver disorders as compared to non-schizophrenia patients (Brown et al., 2000, Curkendall et al., 2004, Daumit et al., 2002, Newman and Bland, 1991). It has been suggested that they are also at increased risk of some pulmonary disorders such as community-acquired pneumonia and chronic obstructive pulmonary disorder (COPD) (Chafetz et al., 2005, Sokal et al., 2004). Risk may be increased by numerous factors including high rates of lifetime cigarette use (70–80% of schizophrenia outpatients aged < 67 smoke) (Brown et al., 2000, de Leon et al., 1995, Goff et al., 2005), poor self-management skills, frequent homelessness, and poor diet (Brown et al., 2000, Casey, 2005, Dalack et al., 1999, Goff et al., 2005, Kermode et al., 1998). In addition, psychotropic medications may increase respiratory irregularities (Nishikawa et al., 1992, Wilcox et al., 1994, Youssef and Waddington, 1989) or risk of respiratory disorders arising from neuroleptic malignant syndrome (Aruna and Murungi, 2005, Liam and Ong, 1990). However, previous studies of pulmonary disease and mortality in patients with schizophrenia have been limited in scope, broadly focused, and set primarily outside the United States. It is unknown whether these factors lead to increased pulmonary disease-related mortality among veterans with schizophrenia.

COPD and community-acquired pneumonia are the 4th and 7th leading causes of death in the United States, and are responsible for over 200,000 deaths each year (Anderson and Smith, 2005). Although substantial attention is paid to the psychiatric and behavioral management of patients with schizophrenia, many barriers impede the detection and treatment of their medical conditions, a circumstance that has been implicated in excess unforeseen deaths (Copeland et al., 2006). Patients with schizophrenia often lack insight into their own psychiatric illness (Pedrelli et al., 2004). This limitation may extend to other signs and symptoms of disease that would motivate the non-schizophrenia patient to seek timely and appropriate medical care (Goldman, 1999). Thus, patients with schizophrenia may experience delays in the diagnosis of disease, leading to more acute comorbidity at the time of diagnosis and premature mortality. In addition, at-risk Veterans Health Administration (VA) patients, including those with schizophrenia, have lower adherence to treatment recommendations (Dolder et al., 2003, Valenstein et al., 2002) and may receive fewer preventive measures such as smoking cessation (de Leon et al., 2005) or influenza vaccinations (Weaver et al., 2004, Zimmerman et al., 2003) leading to increased rates of pulmonary-related deaths. Smoking cessation improves survival for persons with chronic lung disease (Pauwels et al., 2001), and community-acquired pneumonia rates have been linked to lower rates of influenza vaccination, especially during influenza season (Nichol et al., 1999). Current smoking may be sharply reduced among older, sicker patients, as suggested by VA surveys reporting overall rates of 35% among younger enrollees vs 11% among older (65+) enrollees (Goldman et al., 2004, Miller et al., 2001). Yet poor pulmonary health from a lifetime of smoking and other factors, combined with impaired self-care, may allow a treatable acute illness to become life-terminating. Therefore, we hypothesized that a diagnosis of schizophrenia would predict increased likelihood of pulmonary disease among VA inpatient decedents, especially during the final inpatient stay in which death occurred. The purpose of this study was to examine in a retrospective observational study whether schizophrenia was associated with increased likelihood of COPD or pneumonia diagnosis in the year prior to death.