Use of Mechanical Ventilation by Patients With and Without Dementia, 2001 Through 2011

Increasing demand for US critical care resources, including beds, intensivists, and invasive mechanical ventilation (IMV),^1,2 has placed substantial strain on the critical care system. Since 2000, elderly patients treated in the intensive care unit have received higher intensity care (and have experienced lower mortality rates) than historical cohorts.³ Yet certain populations of elderly patients exposed to intensive care experience substantial long-term adverse effects, including functional decline and excess mortality. Patients with dementia receiving IMV, for example, are at high risk for delirium, which confers a 3.2-fold increased risk of 6-month mortality.⁴ The increasing use of aggressive therapies suggests that demand for IMV in elderly populations will increase in the future, both among patients that are likely to benefit and among those with terminal illness. We examined temporal trends in IMV use by older patients with and without dementia and projected future use.

Data for this investigation came from the US Nationwide Inpatient Sample (NIS). Because the NIS is a public, deidentified data set, the Baystate Medical Center institutional review board has declared that research using the NIS does not meet regulatory criteria for human subjects research. Using data from 2001 through 2011, we included hospitalizations for patients 65 years or older who required IMV (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] procedure codes 96.70, 96.71, 96.72) with and without a diagnosis code for dementia.

Primary outcomes included changes in the number of hospitalizations involving IMV over time.⁵ Using population estimates obtained from the US Census, we calculated age-specific and sex-specific rates of IMV for 2001 through 2011. We used linear regression analysis to estimate a trend for each age-sex group and then used population estimates for 2015 and 2020 to estimate the number of patients receiving IMV per year nationally to 2020. Analyses accounted for sampling weights and were carried out using SAS, version 9.3.

We identified 4 854 860 (SD, 81 349) hospitalizations in patients 65 years or older who received IMV; of these, 452 338 (SD, 9354) had a code indicating a diagnosis of dementia. We observed a steady increase of approximately 30% in the number of hospitalizations of older patients who received IMV, from 386 364 (SD, 8079) in 2001 to 497 496 (SD, 12 929) in 2011 at an annual growth rate of 2.9% per year (Table). The prevalence of dementia diagnoses increased significantly (P < .001), expanding from 6.4% of patients receiving IMV in 2001 to 13.8% in 2011 at an annual growth rate of 11.4% per year (Figure). By 2020, we estimate that there will be 671 986 (SD, 19 922) hospitalizations of patients 65 years or older requiring IMV, of which 19.0% will have a diagnosis of dementia.

The use of IMV by populations 65 years or older is expected to double between 2001 and 2020, and growth in hospitalizations for patients receiving IMV with an ICD-9-CM diagnosis of dementia is outpacing, by a factor of 4, those for patients receiving IMV without this diagnosis. These projected IMV numbers are consistent with published data.^5,6

The use of ICD-9-CM codes to identify patients with dementia may be limited by poor sensitivity and lack of information about disease severity. Although better recognition of dementia among hospital providers may have contributed to some of the increase that we observed, the results of this study still have important implications for critical care resource planning. Given projected demand for IMV in the next 10 years, physicians and hospital administrators, already working in a strained system, face a potential crisis unless the critical care system is expanded or changes are made to temper current trends.

Efforts should therefore be made to promote earlier discussions about goals of care in elderly patients with end-stage terminal illnesses. This is most important for the subpopulations of patients (eg, frail elders older than 85 years, patients with end-stage dementia) who are least likely to benefit from IMV and at highest risk for worsening cognitive impairment and death.

Corresponding Author: Tara Lagu, MD, MPH, Center for Quality of Care Research, 280 Chestnut St, Third Floor, Springfield, MA 01199 (lagutc@gmail.com).

Published Online: April 28, 2014. doi:10.1001/jamainternmed.2014.1179.

Author Contributions: Drs Lagu and Lindenauer had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Lagu, Zilberberg, Tjia, Lindenauer.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Lagu, Tjia.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Pekow.

Obtained funding: Lagu.

Administrative, technical, or material support: Lagu, Lindenauer.

Study supervision: Lagu, Pekow.

Conflict of Interest Disclosures: None reported.

Funding/Support: The study was conducted with funding from the Tufts University School of Medicine Charlton Grant Research Program and from the Center for Quality of Care Research at Baystate Medical Center. Dr Lagu is supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award K01HL114745. Drs Lindenauer and Pekow are supported by the National Heart, Lung, and Blood Institute under award R18HL108810.

Role of the Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Meng-Shiou Shieh, PhD, Baystate Medical Center, analyzed the data for this study; Mihaela Stefan, MD, Baystate Medical Center and Tufts University School of Medicine, commented on an earlier draft; and Lauren Williams, BA, Baystate Medical Center, helped with creating the tables. They were not compensated for their contributions to the study.