Accuracy of previous estimates for adult prolonged acute mechanical ventilation volume in 2020: update using 2000-2008 data.

OBJECTIVES: Patients requiring prolonged acute mechanical ventilation (mechanical ventilation ≥ 96 hrs) have hospital survival rates similar to those requiring <96 hrs of mechanical ventilation and consume approximately two-thirds of hospital resources devoted to mechanical ventilation care. Using 2000-2005 data, we previously estimated that their volume will go from approximately 250,000 cases in 2000 to 605,898 by year 2020. With 2006-2008 data becoming available, we explored the precision of our previous formulas and estimates.
DESIGN: We utilized National Inpatient Sample/Health Care Utilization Project of the Agency for Healthcare Research and Quality data from 2000 to 2008 to calculate historic annual age-adjusted prolonged acute mechanical ventilation incidence rates using estimated population statistics from the U.S. Census Bureau. To predict future growth by age group, we fit linear regression models to the historic incidence rate changes. Age-adjusted estimates were computed using population projections obtained from the U.S. Census Bureau.
SETTING: U.S. hospitals. PATIENTS: Nationally representative sample of U.S. hospital discharges with prolonged acute mechanical ventilation (International Classification of Diseases version 9 code 96.72).
INTERVENTIONS: None.
MEASUREMENTS AND MAIN RESULTS: Formulas based on the 2000-2005 data predicted the 2008 prolonged acute mechanical ventilation volume within a 1.9% margin. The historic annualized increase in adult prolonged acute mechanical ventilation increased from 5.0% to 5.2% after incorporating the 2006-2008 data. Factoring in the 2006-2008 data altered our 2020 prolonged acute mechanical ventilation estimates from 605,898 (95% confidence interval 456,695-779,806) to 625,298 (95% confidence interval 552,168-698,838), an upward revision of 3.2%.
CONCLUSION: Our original projections for growth of the adult prolonged acute mechanical ventilation population in the U.S. hospitals by the year 2020 are altered only slightly by adding the 2006-2008 data. Relatively precise prediction of the 2008 prolonged acute mechanical ventilation numbers by the original formulas lends internal validity to the methodology. By virtue of being a large, resource-intensive, and rapidly increasing population, this group of mechanical ventilation patients requires continued close monitoring over time to optimize our preparedness to meet their growing healthcare needs.