Introduction to the UTMB Resident Research Program
Welcome! As part of the scholarly activity portion of the residency program, you will be asked to complete a research project or other scholarly activity. This online module serves as an introduction to the research process.
With the completion of this module you should be able to:
Before you get started, here are a few hints about the module:
Which of the following interest you?
|
|
Yes |
No |
|
Academic Medicine |
|
|
|
Private Practice |
|
|
|
Fellowship Training |
|
|
|
General Pediatrics |
|
|
If you want to do academic or subspecialty, gaining a basic understanding of the scientific process and critical thinking in the realm of research is an integral process in both academia and fellowship training.
If you want to do general pediatrics or private practice, consider a quality improvement project or learning how to write a case report.
If you are completely unclear of your future plans, any of these projects are designed to allow you to become engaged in scholarly activity beyond the standard clinical practice training.
Things to take into consideration:
- your own interests
- what exists in the institution? what projects are currently being done?
- what you or your advisors or people in your chosen specialty doing? do they need help? can you do a spin off?
The development of a project can occur in different ways. Sometimes you will have career aspirations and a field of interest where you will need to find a mentor and a project. Other times you will have a project interest and need to find a mentor. Other times, you may simply have a mentor that you would like to learn more from and together you will identify a project. There are no rules on where to begin. Every project, person and situation is unique. You create the project that works best for you.
Mentors will help guide you through the trenches and provide invaluable experience and expertise to your project.
When looking for a research mentor, find a person who
When you have identified a potential mentor, ask for an appointment to discuss the possibility of working together. Be prepared to suggest an area of interest - you do not need to know your whole plan but do have an idea of what you would like to work on with that person. If the first interview is not productive, do not be discouraged. There are other opportunities available.
Now that you have discovered a topic of interest and chosen a mentor, these ideas need to gel into a project. We will cover the basics of defining your research project.
1.) Define your subject area of interest.
2.) Dissect the subject into manageable components.
3.) Determine what is known about the subject by reviewing previously published literature.
4.) Choose a component to study.
5.) Ask questions.
Start reading about your topic and take notes.
Keep your references.
Examples of the programs available:
While reading on your chosen topic, ask yourself the following questions:
Doing a project that has already been done well and is published is not a satisfying or worthwhile use of your time. You need to know what has been done and what questions have been brought up for future study.
You may also end up utilizing methodologies or statistics that you are reading about within your own study. You will do some of your tables, figures or graphs in a similar manner to what you are seeing in the literature on the topic.
Most importantly you will be learning how to word your question in a more scientific way simply by familiarizing yourself with the literature.
Choose one or more small, unanswered questions.
Why more than one? Because you must consider your resources and population.
When determining your study population you must consider the availability of the participants with the disease/condition you want to study.
Inclusion criteria: guidelines that identify particular characteristics that must be present for a subject to participate in the study
Exclusion criteria: guidelines that exclude participation in the study
Also you will need to consider the access you will have to patients or their charts. If you are interested in studying an extremely rare disease that has been seen once in the past 10 years at our institution, you should strongly reconsider the feasibility of the project.
What must you take into consideration?
- Access to patients (whether in person or their charts)
- Patients must satisfy your inclusion and exclusion criteria
- Prevalence of the disease/condition in the patient population
- Ability to intervene
Most research projects have an ultimate goal of publishing new findings within an area. So let's start at the end - the publication.
A research publication is broken down into the following sections:
We will use this model to guide you through the process of developing your project.
This is a synopsis of your entire project and its findings. Think of it as a 250-word summary of your entire project or the few sentences you would tell a friend when asking about your project.
The Abstract should include:
For the resident research committee initial review, you will need to construct an abstract focusing on:
The introduction is just that: an introduction to your project.
It should include the following:
1.) A brief description of what is already known about your particular project area.
2.) Significance of your project or the rationale of why your study was undertaken
3.) Specific hypothesis or study objectives
The Methods section of your paper may be the easiest to write, but it is one of the hardest to plan.
The following slides will guide you through the key things you will need to think about when planning your research design.
If you think the intervention being looked at will have a larger impact, you can work with a small sample size.
If the intervention will bring a smaller effect, then you will require a much larger sample size.
Things that must be taken into consideration when selecting your sample size:
Acceptable level of significance
Power of the study
Expected effect size
Underlying event rate in the population
Standard deviation in the population2
If it will not be possible to obtain an adequate sample, consider a pilot project.
Independent variable: factor that the researcher has control over and can be manipulated (i.e. what you think will affect the dependent variable)
Dependent variable: factor that changes with manipulation of the independent variable
Extraneous variables: factors outside from what you are studying that are sources of experimental error. Although they are not manipulated in the study design, they alter the magnitude of the relationship between what you are studying and the outcome.
Confounding variables: type of extraneous variable that is statistically related to or correlated with the independent variable. As the independent variable changes the confounding variable changes with it in some way so that the dependent variable is affected -- leading the researcher to erroneously believe there is a causal relationship between the independent and dependent variables. This type of variable is not always known or acknowledged, which results in poor internal validity and the inability to determine a causal relationship.
Good research does not occur by accident. It requires a carefully detailed plan and execution.
The research design:
1.) Details your procedure
2.) Ensures that the procedure is adequate to obtain answers to the research question
3.) Allows you to foresee and avoid pitfalls and obstacles
Not all change in the dependent variable will be due to the independent variable. There are other influences that you must take into consideration and account for within your design.
The change in the dependent variable is due to a combination of factors, including change related tothe independent variable, change related to extraneous variables and change due to chance.
How are you going to determine the change in the dependent variable attributed solely to the independent variable? An adequate study design allows you to quantify the effect of other factors, minimize them or eliminate them entirely.
Control group: group exposed to the standard treatment with all conditions of the experiment remaining the same except for the experimental variable
Randomization: research participants are assigned by chance to either the experimental group or the control group with the goal of reducing as bias as much as possible; through randomization all participants have an equal chance of being influenced by the same bias.
Before you begin collecting any data for your project, you will need to work out the kinks on exactly how you will collect it and what you will be collecting.
What information will you need to collect?
By whom? Will you need help carrying this out?
When will you have time?
If you are collecting a large amount of data, you will need to develop an organized way to make sense of it all. This will likely be accomplished by a spreadsheet. It should be made prior to collecting any data and will include all of the variables within your study.
What will you be comparing, and how will you make that comparison?
There are numerous statistical tests that you can perform with your data, and you will need to consider which one you will use prior to data collection.
Some of the most commonly used tests in resident research are as follows:
Student's t-test: used to compare the means of two samples
ANOVA: used to compare the means of more than two samples, used for continuous variables
Correlation coefficient: indicates the strength and directional relationship of two random variables, can range from -1 to +1.
You will not have to become a master of statistics during your research project, but you will need to know how to best test your data.
You obviously have not generated results from your project yet, but it is tremendously helpful and time saving to think about the data you want to collect and how it will land in this section. This will save you from collecting too much data, and more importantly it will save you from having to go back to retrieve data that you did not collect the first time around. Planning this section will also help you get your information into the correct format so that it is easier to work with once you get to your results section.
Things to think about:
Will you have a demographics table? What will it look like?
What other tables or figures would relay your question and results?
What points do you hope to be able to make with all of this work?
The Discussion section is the interpretation of the results of your study and the rationale or support for all of the conclusions you have made regarding the data collected. It is not merely a summary of your main points, although you often must summarize them here. Use fresh new language to summarize those key points and include the rationale for your interpretations.
An important point: Is the question you are asking worthy of publishing if your hypothesis is incorrect?
This goes into your introduction, but we saved it for last to demonstrate how what one starts with as a question evolves in the process of research planning into the ultimate hypothesis for the project.
"The frustration of the transmutation is what turns your confabulation into a publication!"
- Dr. Cara Geary with the aid of Dr. Karen Shattuck
The hypothesis is your research question. In it's very simplest form, the hypothesis statement is a testable proposition that includes:
1.) what you are manipulating (independent variable)
2.) the outcome you wish to explain (dependent variable)
3.) the proposed relationship between them - type or direction of the effect
The Conclusion is not just a summary or restatement of your hypothesis and results.
Include:
1.) a brief summary sentence of your results
2.) reflection of the significance of these results - what did you accomplish with your study?
3.) limitations of your study
4.) possible future research questions that were brought up due to your findings
The end of module survey can be found here.
1. Kumar, R. Research Methodology. 3rd Ed. Thousand Oaks, CA: Sage. 2011.
2. Kirby A, Gebski V, Keech AC. Determining the sample size in a clinical trial. Med J Aust. 2002;177:256–7.