DOI: 10.1148/radiol.2391051114
(Radiology 2006;239:300-301.)
© RSNA, 2006
A Fisher Exact Test Will Be More Proper
Yuan-Ya Liao, MD*,
Tu-Sheng Lee, MD, PhD
and
and Yu-Min Lin, MD, MMS
Departments of Surgery* and Orthopaedics, Taichung Veterans General Hospital, 160, Sec 3, Chung-Kang Road, Taichung, Taiwan 407, Republic of China
e-mail: ymlin{at}vghtc.gov.tw
Editor:
We read with interest the well-written article in the June 2005 issue of Radiology, entitled "Abductor Tendons and Muscles Assessed at MR Imaging after Total Hip Arthroplasty in Asymptomatic and Symptomatic Patients," by Dr Pfirrmann and colleagues (1). The numbers of asymptomatic and symptomatic patients were 25 and 39, respectively, and differences in the frequencies of findings between the two groups were tested by using 
2 analysis (1). We have some concern about the small sample sizes by using 2 analysis. When sample sizes are small, the 2 test yields poor results, and the Fisher exact test is preferred. A general rule of thumb is to use it when either the sample size is less than 30 or the expected number of observations in any one cell of a 2 x 2 contingency table is less than five (2). We repeated the 2 analysis and performed the Fisher exact test and found fortunately that the results were similar.
As orthopedic surgeons, what also intrigues us most is how Drs Dora and Notzli repaired the abductor during primary total hip arthroplasty and revision for the torn abductors. During closure for the direct lateral approach, 5-mm Mersilene sutures are used to repair the abductors (3). The method for repairing the abductors varies with the institution. Will the authors describe some surgical details for us?
 |
References
|
|---|
- Pfirrmann CW, Notzli HP, Dora C, Hodler J, Zanetti M. Abductor tendons and muscles assessed at MR imaging after total hip arthroplasty in asymptomatic and symptomatic patients. Radiology 2005;235:969–976.[Abstract/Free Full Text]
- Applegate KE, Tello R, Ying J. Hypothesis testing III: counts and medians. Radiology 2003;228:603–608.[Abstract/Free Full Text]
- Thompson RC Jr, Hardinge K. Direct lateral approach. In: Sledge CB, ed. The hip. Philadelphia, Pa: Lippincott-Raven, 1997; 15–32.
Response
Christian W. A. Pfirrmann, MD
Department of Radiology, Orthopedic University Hospital, Balgrist, Forchstrasse 340, Zurich CH-8008, Switzerland
e-mail: christian@pfirrmann.ch
Our data and statistics were reviewed by a biostatistician. A study addressing the surgical technique and the clinical results is currently being performed in our orthopedic department. We are hoping to be able to present our analysis and results soon.
The statistical consultant responds:
The authors do not state that the 2 method is wrong, they simply point out that if sample sizes are small, sometimes 2 is not the best test and exact statistics would work better. Some people use the stated criteria of the authors of the letter; others use weaker criteria. Thus, there is no "problem" per se with the test used.
TOP
References
