Longitudinal Analysis of Efficiency in Multiple Output Dialysis Markets

Abstract

Provider efficiency in the dialysis industry in the U.S.A. has been of great interest for a variety of parties mainly because of the continuing growth in the number of such patients and providers and in the industry's costs. This study examined technical efficiency longitudinally among the multiple-output producers of freestanding facilities, as the dominant group of providers, using the DEA-based Malmquist index. Nationally representative data were obtained from Independent Renal Facility Cost Report Data Files for the years 1994 through 2000. The resulting sample comprised 140 facilities that had operated throughout the seven study years and jointly produced all dialysis outputs with nonzero inputs. The results show that over the period 1994-2000, on average multi-output, freestanding dialysis facilities did not achieve improvement in productivity. Decomposition of the Malmquist productivity indices showed improvement in technical efficiency but at the same time regress in technologies with potential to improve their quality of care. Negative change in technology was the major source of negative movement in productivity. The study concludes that under the fixed-price payment policy, multiple dialysis outputs are produced efficiently, but it may be that the quality of care is being sacrificed.

The economics of cancer care: longitudinal changes in provider efficiency

“…These include: (1) adjusted operational expenses less personnel costs, adjusted with both case mix and geographical wage indices; (2) hospital size, measured in terms of number of staffed beds annually; and (3) total full time and part time staff. All three of these inputs were consistent with previous studies using similar metrics, most notably Ozgen and Ozcan [21], which applied Malmquist techniques to longitudinal analysis of dialysis markets with multiple outputs. While there are other choices of inputs (e.g., nursing hours or value of medical supplies) and outputs (e.g., research volumes), the variables selected were all: (1) publicly available; and (2) reliably measured using consistent data sources and measurement methodologies.…”

Measurement of efficiency of the dialysis sector in Turkey using data envelopment analysis

“…Instead, different types of dialysis modalities are considered to be the outputs of a dialysis facility [11]. The dialysis literature defines the three output categories as outpatient dialysis, dialysis training, and home dialysis treatments, thereby taking into account variation in resource use [12][13][14]. Home dialysis is not a common modality of dialysis in Turkey; although it has gained attention through the "Home Dialysis Circular" which was effective in 2007, so far this service has been provided by private facilities in only one province.…”

“…Three types of inputs are plausible for a typical dialysis production process: labor measured as numbers of fulltime equivalent (FTE) physicians, registered nurses, and other medical staff, capital measured as the total number of dialysis machines, and operating costs [12][13][14][15][16]. Unfortunately, due to data constraints this study was only able to use two of these inputs, labor and capital.…”

“…Although the majority of facilities are publicly financed, private enterprises have also increased by 16%, up from 34% in 1999 to 50% in 2008. Dialysis machines have increased from 3723 in 1999 to 13,879 in 2008, a nearly 373% increase. In 2008 private providers owned 63% of the total number of machines in the country, an increase of nearly 22% since 1999.…”

The role of safe practices in hospitals’ total factor productivity

“…However, such efforts to control health care costs have been shown to come at the expense of quality in some settings. 8 The interest in VBP has led to numerous studies focusing on hospital inefficiency and the environmental and cultural conditions that make change difficult. such as Data Envelopment Analysis (DEA) 12,13 and Stochastic Frontier Analysis 14 to measure hospitals' X-inefficiency levels.…”