Home     |    Contact ASA     |     Join ASA!    |     Members Only     |    Retail Store   |    Advertising Information
 
ASA NEWSLETTER
 
 
February 2004
Volume 68
Number 2

Letters to the Editor


The Relative Cost of Being a Resident

Having just returned from Kilimanjaro Christian Medical College (KCMC) in Tanzania, Africa, I read with interest the Jill E. Beland, M.D., (October 2003) account of the trials of being an American anesthesiology resident.

Her article reminded me of the two anesthesiology residents I had the pleasure to work with at KCMC. They don’t have resident or fellowship salary issues because they don’t receive a salary during training. They have to seek out funding to pay for their postgraduate education. Once they complete their training, they can look forward to a yearly income of about $2,000. There are no fellowships available so this is not a major career issue. The stress of having to balance studying and family, when you get home from the hospital, is not a big problem because the one with a family is living in a dormitory six hours by bus away from his pregnant wife and 2-year-old daughter. There is no stress about deciding what books to invest in because there is no bookstore selling anesthesiology books. They can relax about all the hassle of dealing with invasive monitoring and PACU patient management because there are no facilities for invasive monitoring and there is no PACU. They just have to try to find a functioning ECG monitor, a blood pressure cuff that fits and the one functional oximeter in the department for their patient.

Interpreting lab results is not problematic because they often are not available. Due to staff shortages, they often get additional learning opportunities by working after being “on call.” They also get the experience of delivering anesthesia care to a population with a high prevalence of HIV.

Perhaps when Dr. Beland finishes her training and accepts her first six-figure yearly salary, she might enjoy participating in the ASA Overseas Training Program and commiserating with the anesthesiology residents she meets there about “the cost of being a resident.”
 
Wendy J. Watson, M.D.
Brookfield, Wisconsin


Statistics Reveal That Statistical Analysis Is Flawed

Several articles in the November 2003 ASA NEWSLETTER (“Performance and Outcomes Measurement”) appear to contain the same assumption that the accumulation of ever-increasing amounts of data in a relational database will permit the discovery of new knowledge and relationships among the data. Unfortunately this may not be true.

Since we can perform an arbitrarily large number of queries on any given database, the reliability (“P-value”) of the conclusions should be adjusted to reflect the number of past, present and future conclusions. As we increase the number of comparisons (hypothetical relationships within the data), the likelihood that any individual relationship is only due to chance increases. One method to compensate for this is to decrease the threshold P-value. For example if we make five comparisons, we should consider using P<0.01 (0.05/5). This is a special case of the Bonferroni inequality.1 If we do not do this, our “new knowledge” may be entirely factitious.

This problem was illustrated by an article in the Wall Street Journal several years ago on large customer databases developed by grocery corporations for marketing. After the expenditure of millions of dollars, one of these companies was able to “discover” that customers who bought diapers were likely to buy baby food but not much else.

Another difficulty with large relational databases is the need to classify data. Waste of data is inherent in any classification system that does not store all the original data. A photograph compressed into a JPEG file contains much of, but not all, the original data and can never be re-expanded into a perfect copy of the original.

Large databases may be useful for suggesting what to investigate with future randomized, controlled trials but do not deliver any “scientific proof” by themselves.

Peter H. Norman, M.D.
M. Denise Daley, M.D.
Houston, Texas


References:

1. Glantz SA. Primer of Biostatistics. 4th ed. New York, NY: McGraw-Hill; 1997:90.d



Thanks to Mark J. Lema, M.D., Ph.D.

Dr. Lema, I am not at all certain how the members of our ASA can thank you for your editorial wisdom and sense of humor. You have guided the NEWSLETTER with unusual insight and, when needed, tough, but appropriate, commentary. The “Aphorisms” have been especially welcome!

Many of us are sorry that you are retiring from your post as Editor, but understand that there is more to life than the Editorship of the NEWSLETTER. Thanks for all the efforts for the benefit of the anesthesia community!

Steven R. Young, M.D.
Indianapolis, Indiana



 

FEATURES

Reclaiming Critical Care


ARTICLES

DEPARTMENTS


The views expressed herein are those of the authors and do not necessarily represent or reflect the views, policies or actions of the American Society of Anesthesiologists.

NL Archives

Information for Authors