(187b) Evaluating Hospital Performance Using Process Systems Engineering Tools

The health-care spending in the United States accounts for about 17% of US GDP, making the US the highest health-care expenditure per capita in the world. However, the United States has not seen an increase in life expectancy to match its huge outlay on health-care. Hospital care represents the single largest national health expenditure by the type of services, accounting for approximately 31% of total healthcare costs [1]. Therefore, there are areas for improvement in the health-care system and hospital care is one of them. In order to improve hospital care by improving patient outcomes while reducing cost, some metrics or indices are needed to quantify hospital performance.

The Healthcare Cost and Utilization Project (HCUP) is a family of health care databases and related software tools and products developed through a Federal-State-Industry partnership and sponsored by the Agency for Healthcare Research and Quality (AHRQ). HCUP databases bring together the data collection efforts of State data organizations, hospital associations, private data organizations, and the Federal government to create a national information resource of encounter-level health care data (HCUP Partners). HCUP includes the largest collection of longitudinal hospital care data in the United States, with all-payer, encounter-level information beginning in 1988. These databases enable research on a broad range of health policy issues, including cost and quality of health services, medical practice patterns, access to health care programs, and outcomes of treatments at the national, State, and local market levels [2].

Several quality indicators have been developed by AHRQ, including the Prevention Quality Indicators (PQIs), Inpatient Quality Indicators (IQIs), Patient Safety Indicators (PSIs), and Pediatric Quality Indicators (PDIs) [3]. However, most of these indicators are patient-outcomes related and none of them directly measures hospital performances. Driven by the need of a comprehensive hospital performance measure or indicator, in this work we propose to adapt some of the process systems engineering tools that have been developed for process and control performance monitoring to quantify hospital performance. For example, multivariate statistical methods such as principal component analysis (PCA) and partial least squares (PLS) have been developed to efficiently monitor the performance of large processes, and to rapidly detect and identify important process changes [4]. Because hospital operations and manufacturing processes share some similarities at systems level, we expect that the tools developed in process systems engineering are highly promising for extracting information such as hospital performance from the HCUP data.

However, there are many challenges in adopting these techniques to evaluating the performance of a hospital due to many unique characteristics of hospital operations. For example, for chemical process operations, data collected from various sensors in a DCS system are measurements of continuous variables such as temperature, pressure, flow rate and concentration, which make mathematical analysis of these data straightforward with various numerical computation tools. In contrast, HCUP data contains various data format, including both numbers and strings. Even for numbers, some of them are categorical such as hospital IDs and diagnosis categories; some of them are (binary) indicators such as smoke or not and drink or not; some of them are ordinal such as disease severity; and some of them are frequencies such as number of a certain procedure performed. Therefore, the multivariate methods mentioned above cannot be readily applied to this case. In this work, we will report our study of various variable transformations and investigation of statistical methods that work on mixed data types. In addition, we will present the proposed hospital performance indicator, its correlation and sensitivity to the factors in the HCUP database and the interrelations among these factors.

References:

[1] M. Hartman, A. Martin, P. McDonnell, A. Catlin, and N. H. E. A. Team, “National health spending in 2007: slower drug spending contributes to lowest rate of overall growth since 1998,” Health Aff., vol. 28, pp. 246–261, 2009.

[2] C. Steiner, A. Elixhauser, and J. Schnaier, “The healthcare cost and utilization project: an overview.,” Eff. Clin. Pract. ECP, vol. 5, pp. 143–151, 2001.

[3] M. Farquhar, “AHRQ quality indicators,” 2008.

[4] J. V. Kresta, J. F. MacGregor, and T. E. Marlin, “Multivariate statistical monitoring of process operating performance,” Can. J. Chem. Eng., vol. 69, pp. 35–47, 1991.