Interview with a Researcher – Martin Brodsky, PhD, ScM, CCC-SLP

Please give a brief background:  (Employment, interests of study, special honors/certifications etc.)

As a newly-minted clinician, straight out of my clinical fellowship that included settings of acute care, inpatient rehabilitation, outpatient rehabilitation, and skilled nursing facilities, I maintained a productive caseload as the only speech-language pathologist working at a major midwestern Level I trauma center. The combination of professional growth and the desire to advance excellence in patient care ultimately resulted in my shift to focusing on research after 4 years of full-time clinical work.

Although I started my PhD work studying aphasia and motor learning, I never strayed from my interest in swallowing and swallowing disorders. To that end, my doctoral dissertation combined the work and knowledge I gained in motor learning with swallowing in patients with Parkinson’s disease and healthy controls.

Some 13 years later, I’m an Associate Professor of Physical Medicine and Rehabilitation and Pulmonary and Critical Care Medicine at Johns Hopkins University. I have an active research program that is funded by the National Institutes of Health studying the effects of critical illness and critical care medicine on swallowing and the airway and their long-term outcomes while maintaining a clinical schedule and remaining active in leadership roles for ASHA and Dysphagia Research Society. Additionally, I’m an Associate Editor for Dysphagia, a Section Editor for Archives of Physical Medicine and Rehabilitation, and serve on the editorial board for the American Journal of Speech-Language Pathology.

 

If you could conduct any research study on any topic/issue (meaning money/funding, time, subjects, IRB etc. are NOT a problem!), what would it be? In other words, what’s your dream study?!

I’m not sure I would change anything from what I’m doing right now! My current research program in critical care as it pertains to swallowing, voice, and airway disorders is an area I’ve been thinking about for more than two decades. In essence, I’m doing my dream study! [👏Nice!!👏]

 

If you could change one thing about having to do research, what would it be?

This is an interesting question because the things that I would choose to change would be for the purpose of streamlining or making things easier. For example, the IRB at some institutions is not as streamlined as others. Although governance comes down from the Federal gov’t to guide the conduct of IRBs, the final rule ultimately resides on the local level. In the end, I strongly believe that all steps necessary to be completed in an IRB application are necessary for ethical research to be completed. And who would want to skimp on that?

Streamlining processes aside, it would be an absolute dream to have departmental money funding a research coordinator and/or research assistant instead of always trying to find grant money that is temporary at best. I personally believe that research staff should be better protected for the research mission than continuing to live with the “Where is my next paycheck coming from?” always in the back of their mind. At the very least, their dedication deserves our support.

 

What’s one thing you think clinicians should be excited about in our field regarding research?

Evidence. Whether clinicians are looking to justify staffing changes, purchases, or approaches to evaluation and treatment, the continuing flow of evidence that comes from research provides a strong basis for any discussion. The excitement that comes from these prospects is the feeling of confidence from standing on reasonably solid ground.

 

What’s one thing you think researchers should be excited about in our field regarding clinical practice?

The future. I’ve been in the field of speech-language pathology as a clinician for nearly 30 years and I can’t remember a time when I’ve seen so much participation by clinicians in research. It excites me to know that many of those who are “on the ground” and “in the trenches” are critically assessing their experiences and asking informed clinical questions that become well-executed studies completed by collaborative clinical research teams. Effectively, these clinicians have taken on the laudable additional roles as translators and implementors for the research mission. Because many more clinicians are actively participating in research, more questions can be asked and answered in shorter periods of time, the result of which is improved patient care and improved clinical and organizational efficiency. #yaasssss👏👏

 

I feel the Results section of studies can be the most misunderstood or difficult to navigate through, what tidbits or tips/tricks can you share to help clinicians process these sections?

In order to understand the Results section of the study, you need to first understand (or at least have the broad concept of) what was outlined in the Methods section. That is, how can you understand the answer if you don’t first understand how you got the answer or whether it was even possible to get that answer? Some Methods sections are going to be highly technical; others will be very straightforward. The ease of understanding the Methods section can be compared similarly to a recipe for making a soufflé vs. one that teaches you how to make a scrambled egg. Just because you’ve made a scrambled egg 1000 times does not mean you will be successful with making a soufflé. Let’s stick with the scrambled egg🍳…

Assuming you have never seen a scrambled egg, you would have no ability to recognize it as an egg at all. A scrambled egg doesn’t look like the thing that comes out of the shell as a raw egg🥚. Moreover, when compared with a soft-boiled egg, a hard-boiled egg, a poached egg, or a sunny side up egg, a scrambled egg resembles none of these…but they all resemble each other, right? 

Now, suppose I tell you that I cracked open a raw egg, whisked it well in a bowl, and then placed it in a pan over medium heat until all of the liquid became solid, then slid it on a plate. You would then understand how it was I got to the scrambled egg (i.e., the technique) and, therefore, what is sitting on a plate in that form (i.e., the answer). The “what” is the Results section.  

Metaphors aside, it seems that most people have difficulty reading the results section because of the numbers associated with statistics and perhaps the interpretation of statistics tables(🙋‍♀️🙋‍♀️🙋‍♀️). Doing a little extra reading about statistics will behoove any reader. One of the best sources that I’ve seen in the literature is a series of articles written by Phillip Sedgwick in the British Medical Journal (BMJ). The articles are written for the non-statistician, basically, people like you and me who are trying to understand peer-reviewed research. He generally starts out the articles with a case or an example, followed by a multiple-choice question related to the topic of the article (and yes, he gives the answer!). 

After this, he goes into some explanation of the topic, ending the article with a one-paragraph summary. These articles are very approachable in that they are generally 2 to 3 pages and so very well illustrate the points! The “library” he created between 2010 and 2015 spans roughly 80 articles.  If you were to type the phrase (without quotes) “Sedgwick P[au] and BMJ” into the PubMed search bar, all will become clear! The only question that will remain is, “Which one will you choose first?”😉 

[**edit note: not all Sedgwick’s reviews are open access😩, BUT Dr. Brodsky was so kind to also suggest you can download the pre-published versions on ResearchGate, and can also search Nikolaos Pandis in PubMed for similar explanations too!🤓]

 

Could you pick one technical jargon (e.g. “linear regression” “ANOVA” “two-tailed test” etc.)  to explain in a relatable and easily understandable way (real-life, simple examples get bonus points😉)?

Regression is all around us😮. Simply, regression analyses are used to understand how individual variables (called predictor variables) contribute to the outcome variable. In a sense, regression weighs the contribution each of the variables has on the outcome variable, effectively determining their association. There are many different kinds of regression (e.g., linear, logistic, polynomial, Poisson), each depending on the type of data (e.g., categorical, continuous) you are using.

Although you may not have recognized it at the time, a growth chart is perhaps one of the easiest and most common examples of data using regression. Most often, you’ll see combinations of age with length and weight.  Charts are separated by sex. On any chart, you will see curvilinear lines that represent percentiles based on these numbers. These lines represent thousands, if not millions of data points that have been gathered across many years of data collection in order to understand normal growth. The curvilinear lines that you see on this chart are, in fact, regression lines, or the plot of association between age and either length or weight.

What’s the one thing you think is important for practicing clinicians to know/understand when reading research?

P-values are probabilities, not certainties. Just because one study was successful in its treatment, showing a low p-value, doesn’t mean the study that comes after that one will be able to replicate the first one’s results. Even in studies with large numbers of patients that are well-controlled, we are still only dealing in the world of probabilities and levels of confidence.🤨

A real-life example can be seen on the nightly news when you listen to the weather forecast. In keeping with the theme of the previous question’s answer, meteorologists use regression as prediction models to deliver the forecast. 

Example 1: If you were to track a 10-day forecast for the same 10 days, you will see ranges of temperatures and ranges of precipitation narrow as you move closer and closer to the day of interest.🤯🤯

Example 2: Lately, we’ve all heard about how many cases (or deaths) of COVID-19 we can expect during the next month, in 3 months, etc. Large amounts of data are placed into statistical models while making certain assumptions (e.g., temperature, population density). The statistical modeling uses regression, and the range of values cited in the reports we hear are the 95% confidence intervals that are derived from the regression model.

 

What’s one thing you think is important for researchers to know/understand about clinical practice?

We need to listen. Clinicians are the ones “on the ground” using the instruments, the tools, and the therapies that researchers create. But what good is an improperly used instrument, an unused tool, or ignored therapy because it wasn’t practical to implement? Listen to the clinicians around you… they have a lot to say!👏😁👏

 

What is something you believe researchers could do better to #bridgethegap?

Reach out to clinicians for their input and join them in understanding and experiencing the complexities of clinical activities. Clinicians with all levels of experience can provide meaningful contributions, whether it’s an idea during a conversation or insights into the condition/disease/patient right in front of them. Create opportunities for researcher-clinician interactions.

 

What is something you believe clinicians could do better to #bridgethegap?

Don’t be afraid to reach out to researchers. Most peer-reviewed published articles contain the primary author’s contact information—use it! Become involved in research programs, journal clubs, lab discussions. Research is a team sport.  Join the team. 

 

 

Can you provide your contact email if clinicians want to reach out? (Honors system for everyone to be respectful of your time)

Email: mbbrodsky@jhmi.edu

Twitter: @MBBrodskyPhD

LinkedIn: http://bit.ly/Brodsky_LI

ResearchGate: http://bit.ly/Brodsky_RG

 

---

 

A HUUGE thank you for not only all the amazing work that Dr. Brodsky has dedicated to our field over the decades but also for his passion and time giving these awesome responses!!!👏👏😍👏👏😍👏👏