Comparison of methods to identify outliers observed in health services small area variation studies

To, Teresa; Williams, J. Ivan; Wu, Keyi; Theriault, Marc-Erick; Goel, Vivek
December 2003
Statistical Methods in Medical Research;Dec2003, Vol. 12 Issue 6, p531
Academic Journal
journal article
Small area variation analysis (SAV) is an established methodology in health services and epidemiological research. The goal is to demonstrate that rates differ across areas, and to explain these differences by differences in physician practice styles or patient characteristics. While the SAV statistics provide an overall variation estimate, they do not provide a statistical means to identify significant outliers. We compared the chi-square (chi2) test with three approaches in determining significant outliers in SAV. We used data from the Canadian Institute for Health Information (CIHI) for Ontario residents discharged between 1989 and 1991. Coronary artery bypass surgery, hysterectomy and hip replacement data were used to compare four statistics in determining outliers: the chi2 test, Swift's approximate bootstrap confidence interval (ABC), Carriere's T2 (T2) with simultaneous confidence intervals (SCI), and Gentleman's normalized scores (GNS). Both the ABC and SCI correct the skewness of the distribution of the adjusted rates. With large data, confidence intervals calculated by the normal or the ABC methods are indistinguishable. The T2 can be applied to also nonbinary events. For binary events, it is asymptotically the same as the chi2. The GNS ranks the rates, but the distribution of these ranks does not differ significantly from that of the adjusted rates. We concluded that when using large data with binary events, there is little advantage in using the ABC, SCI or GNS over the commonly known chi2. The chi2 remains a useful tool in small area variation analysis to 'screen' or flag potential differences beyond chance alone.


Related Articles

  • Hysterectomy: Variations in Rates Across Small Areas and Across Physicians' Practices. Roos, Noralou P. // American Journal of Public Health;Apr1984, Vol. 74 Issue 4, p327 

    This analysis focuses on the practice of hysterectomy across 33 hospital catchment areas of one Canadian province, using claims data from the Manitoba health insurance system. Hysterectomy rates varied five-fold across hospital areas. The availability of hospitals and physicians was unrelated to...

  • Geographic Variations in Breast Cancer Survival Among Older Women: Implications for Quality of Breast Cancer Care. Goodwin, James S.; Freeman, Jean L.; Mahnken, Jonathan D.; Freeman, Daniel H.; Nattinger, Ann B. // Journals of Gerontology Series A: Biological Sciences & Medical ;Jun2002, Vol. 57 Issue 6, pM401 

    Presents a study that examined breast cancer incidence, survival and mortality in a number of health service areas covered by the Surveillance, Epidemiology and End Results (SEER) program for aged women. Factors of the variation in incidence of breast cancer; Uses of the mortality data from the...

  • Creating a Geographical Information Systems-based Spatial Profile for Exploring Health Services Supply and Demand. Murad, Abdulkader A. // American Journal of Applied Sciences;2011, Vol. 8 Issue 6, p644 

    Problem statement: GIS has several techniques and functions that can be used for health services planning. Each one of these functions can be applied for different health related issues. Approach: The first part of this study reviews the relevant GIS functions that are used by health care...

  • Effects of residential relocation on mortality and morbidity among elderly people. DANERMARK, BERTH D.; EKSTRÖM, MATS E.; BODIN, LENNART L. // European Journal of Public Health;1995, Vol. 6 Issue 3, p212 

    This study analyses the effects of residential relocation among elderly people, in terms of mortality and health service consumption. The analyses are based on a register covering 22,579 persons. Analyses at different levels are carried out a logistic regression analysis, a longitudinal analysis...

  • Tolker leserne forskningsresultater i trÃ¥d med forfatternes intensjoner? Ringerike, Tove; Glenton, Claire; Vist, Gunn Elisabeth; Jamtvedt, Gro; Nylenna, Magne // Norsk Epidemiologi;2013, Vol. 23 Issue 2, p231 

    Background: Research results can be communicated in various ways, for instance through numbers, words, and combinations of these. Within clinical research, in addition to presenting the estimate of effect, it is also desirable to say something about our confidence in this result, for example by...

  • Multilevel Analysis in Health Services Research. Coughlin, Steven S. // Open Health Services & Policy Journal;Jan2009, Vol. 2, p42 

    The article reflects on the multilevel analysis in health services research in the U.S. He notes that the multilevel analysis techniques are used in health care systems' observational and randomized studies. He points out that multilevel analytic techniques have possible advantages for health...

  • „HUMANITIES GREY HAIRS“ - ASPECTS OF THE AGEING OF SOCIETY. Tomaszewska-Hołub, Beata; Domaszewicz, Wacław // Proceedings of the 7th Annual International Scientific Conferenc;2011, p316 

    Modern demographic transformations that take place in European societies cause the process of ageing to become irreversible and persistent. This is happening as a result of the simultaneous occurrence of such factors as average life expectancy growth, fertility decrease and a negative birth...

  • CANADA.  // Health in the Americas (9789275115879);2002, Vol. 1, p125 

    The article offers information on the economic condition, health care system and vital statistics of Canada. It has an extensive social security network, including old age pension, family allowance, employment insurance, and social assistance. Basic health care, such as physician and hospital...

  • Inequality in the geographical distribution of general practitioners in England and Wales 1974-1995. Gravelle, Hugh; Sutton, Matthew // Journal of Health Services Research & Policy;Jan2001, Vol. 6 Issue 1, p6 

    Objectives: To compare geographical inequality in the distribution of general practitioners (GPs), other resources and mortality around 1995 in England and Wales; to measure trends between 1974 and 1995 in inequality of GP distribution; to examine the implications of different need adjustments...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics