Chapter 04: Earning a Master’s in Public Health and a Landmark Study of Infections
Files
Loading...
Description
In this chapter, Dr. Elting explains the work she did under Dr. Gerald Bodey while also attending the UT School of Public Health to work on her Master’s and supplement the information she knew she lacked. She talks about the open environment that provided her with a good learning experience. She notes that she was interested in infections caused by multiple organisms. Dr. Elting explains that she did a large study of polymicrobial septicemia, which still may be the largest study ever conducted.
Next Dr. Elting explains the main lessons she learned during her Master’s program: that she could offer clinicians a view of entire populations (rather than a focus on individual patients); that she could strengthen studies by insisting that the basic research question be articulated. She also explains that she learned to communicate effectively with clinicians and non-biostatisticians by using graphic charts and pictures, rather than tables of numbers.
Identifier
EltingL_01_20150219_C04
Publication Date
2-19-2015
City
Houston, Texas
Interview Session
Linda S. Elting, DrPh, Oral History Interview, February 19, 2015
Topics Covered
The Interview Subject's Story - The Researcher; Discovery and Success; Overview; Definitions, Explanations, Translations; The Researcher; On Research and Researchers; Understanding Cancer, the History of Science, Cancer Research; The History of Health Care, Patient Care; Evolution of Career; Professional Practice; The Professional at Work
Transcript
Tacey A. Rosolowski, PhD:
Hmm. Now, tell me about the next kind of change, because then you went for your master's in public health.
Linda S. Elting, DrPh:
Yeah.
Tacey A. Rosolowski, PhD:
What about that decision?
Linda S. Elting, DrPh:
Well, I got to a place where I wasDr. Bodey's organization grew, and instead of doing just the clinical trials of antibiotics and working with infections, he was in charge of all the chemotherapy. And so I ended up with a bunch of research nurses working for me, coordinating those studies. So I was mostly just training new people, doing the analyses, and it became clear to me that while I could step in and do stuff, there was an awful lot I didn't know; particularly about statistics. So I just made the decision to get a master's degree. And I went to the School of Public Health, because it was a veryit was also a new organization, just sort of getting started here in the Medical Center and at UT [University of Texas]. And it was also very unstructured, just like where I was working. And you could walk into that school at that point and say, look, this is what I want to be doing when I walk out of here. Here are the skills I already have, these are the ones I know I need. That's where I want to focus. And somebody would say to you, "Okay, I'll be your advisor and you can do that, but I think you need to learn this too and have this exposure. But you're going to do it pass-fail, so it's not a huge risk to take this weird class that you don't feel prepared for." And so that was a really good experience for me. You know, I walked into classes and most of the students there had no clue what they wanted to be when they grew up, so they were just sort of floundering in that open environment, but it's great for me, you know, because I could tell every professor when I walked in the door, "I already know these things, but I don't know anything about this. You need to help me learn this. This is where I want to focus." And so it was a very good experience for me.
Tacey A. Rosolowski, PhD:
What did you do your thesis on? What was your research?
Linda S. Elting, DrPh:
I was still working at Dr. Bodey at that point, and I was very interested in infections caused by multiple organisms, because they were very deadly for patients in our hospital. When I first went to work at MD Anderson, most people with cancer died from infection. So we were developing antibiotics, and the worst of the worst infections were the ones caused by multiple organisms. And it had been decades since anyone actually did a study to describe the epidemiology of those problems; what are the most common ones, what antibiotics usually work? And if you think you have somebody who has this problem, what should you start with first? And so I did a large study on polymicrobial septicemias or bacteremias. I think it's still the largest one ever published
Tacey A. Rosolowski, PhD:
Really?
Linda S. Elting, DrPh:
even these many years after that. So that's what I did my thesis in.
Tacey A. Rosolowski, PhD:
And what were some of the bigI mean, obviously you learned new skillset areas. Were there some ways in which your own perspectives on data management-data analysis changed, and you started getting a philosophy about running epidemiological research?
Linda S. Elting, DrPh:
I guess a couple of things; the first thing that really was impressed on me as I studied epidemiology and the classic studies that have been done over time and things like that is that you have to take a population perspective. But everybody I worked with on a daily basis, being a clinician, took a perspective that was strongly influenced by the last individual they saw. And consequently, many of the sweeping statements that were made and some of the hypotheses they came up with occurred as a result of failing to take a population perspective. And so I felt that, particularly as I began doing some of my first research and preparing myself to do those things that what I could really add to a department full of clinicians who looked attake this individual perspective is the skills and the view of the whole population. And so that was probably one of the two most important things I learned. The other thing I learned is that, and it became more and more obvious to me, the things that we failed to do well at MD Anderson when we were doing studies, and that was that in doing our clinical trials and other projects, we failed to clearly specify the question. And consequently, we wasted some resources collecting data we didn't need to collect, and often we couldn't answer the most important question because we didn't do a good enough job at the beginning of specifying the question. And so, my mantra as a member of the department became, "What's the question,"every time somebody would sit down in my office, "I want to do this study." "What's the question? Let's get the question right, and then where's the data to support it?" You know, "Are we really going in a direction that is going to be relevant, or are we just going in a direction that is influenced by the last patient you saw, who is a train wreck, but those train wrecks are very rare." So I think that perspective was really, really helpful to me. The other thing that I learned there from studying the classic epidemiologic studies is the value of seeing the whole picture as a picture, and not numbers. And when I started working with using that idea with the clinicians I worked with, it's amazing the difference it made. You know, instead of showing them a table with numbers, I showed them a graph. That's so much more accessible than a bunch of numbers and P-values that we started getting through preliminary results and various things very fast, because I learned how to graph things properly and how to picture them instead of just write them in tables and numbers.
Tacey A. Rosolowski, PhD:
I was going to ask you a kind of related question, because you were talking about that difference in perspective between the public health epidemiological mentality and the clinical mentality; that just seemed like a gap that was made for miscommunication.
Linda S. Elting, DrPh:
Mm-hmm. (laughter)
Tacey A. Rosolowski, PhD:
And so I was going to ask you, well, how did that all happen? Are there some other areas where you've seen emerge where there's kind of a communication gap between a biostatistician and a clinician, I mean a gap that needs to be overcome with effective communication strategies?
Linda S. Elting, DrPh:
I guess because I have worked with some of the best statisticians in the world, and they're very knowledgeable about medicine, the gaps are not so obvious, or so influential. I have never felt that it was effective, even to show really expert people only numbers and only significance levels, because it's just a different way to think. And I have always thought that the way to bridge that is with pictures.
Tacey A. Rosolowski, PhD:
Mm-hmm.
Linda S. Elting, DrPh:
That is in graphs, and that's how most statisticians work these days, I think.
Tacey A. Rosolowski, PhD:
Is with more visualization.
Linda S. Elting, DrPh:
Mm-hmm. Mm-hmm.
Tacey A. Rosolowski, PhD:
Yeah. I'm blanking on the guy who does the beautiful datathe Beautiful Data books, but I'll think of it.
Linda S. Elting, DrPh:
Ed [Edward] Tufte.
Tacey A. Rosolowski, PhD:
That'sEd Tufte. That's right. (laughter)
Linda S. Elting, DrPh:
I have his books right up there.
Tacey A. Rosolowski, PhD:
There you go! (laughter) Yeah, I've admired his communications styles
Linda S. Elting, DrPh:
Yeah.
Tacey A. Rosolowski, PhD:
very much, too, sure. Yeah, how to inspire with information, not basically put people to sleep.
Linda S. Elting, DrPh:
Mm-hmm, yeah. Yeah. Or, have people miss it. You know, I was involved in some studies with a colleague from a different university for a while, and he would show survival curves to patients and families. And what he discovered was that people never looked at the end of the curve. They looked at the difference between the lines right on the left side of the graph. So if you had a treatment that was really good initially, but actually worse at the end, patients and their families didn't pick up on that. And if you took it to the extreme, no treatment usually is best at the very beginning because there's no toxicity and no deaths. And then you might have something that actually cures people, but it didn't. The patients missed it.
Tacey A. Rosolowski, PhD:
They didn't see that either?
Linda S. Elting, DrPh:
All the stuff off at the right hand side of the graph was missed. So that really impressed on me how innumerate our society is.
Tacey A. Rosolowski, PhD:
Yeah.
Linda S. Elting, DrPh:
How bad we are at synthesizing data in our heads.
Tacey A. Rosolowski, PhD:
Mm-hmm.
Linda S. Elting, DrPh:
Because when I look at those graphs, the first thing I look at is the far right hand corner at the overall survival of the two groups. And no matter how close they are here, down here at ten years, if one's better, that's what I want.
Tacey A. Rosolowski, PhD:
Mm-hmm.
Linda S. Elting, DrPh:
But people missed that entirely.
Tacey A. Rosolowski, PhD:
Yeah. But that's almost like you already know that that's where the punchline of a data story is.
Linda S. Elting, DrPh:
Sure. Yeah. That's where it is for me.
Tacey A. Rosolowski, PhD:
Yeah.
Linda S. Elting, DrPh:
And they looked at that the next year.
Tacey A. Rosolowski, PhD:
Yeah.
Linda S. Elting, DrPh:
And no treatment was better than a curative treatment, in some cases, so
Tacey A. Rosolowski, PhD:
Hmm.
Linda S. Elting, DrPh:
Visualizing that is an important way to do it.
Recommended Citation
Elting, Linda E. DPh and Rosolowski, Tacey A. PhD, "Chapter 04: Earning a Master’s in Public Health and a Landmark Study of Infections" (2015). Interview Chapters. 708.
https://openworks.mdanderson.org/mchv_interviewchapters/708
Conditions Governing Access
Open
