I've Got An Idea
Often, research ideas and hypotheses grow from the bedside in response to patient care. At other times, research needs to respond to community needs or epidemiological events. These clinical experiences can often be the driving force for clinical research. If you need to brush up on your understanding of evidence based medicine, you can review these modules.
When you encounter an idea or experience that inspires you, that is a good time to start thinking whether your idea can generate a research study. If you’d like to browse examples of your colleagues’ successes, you can find them celebrated here.
Great question – have you done a systematic review of literature yet? As part of evaluating your research question, it’s important to evaluate existing evidence, as it is possible your original question has been answered.
The CMH Medical Librarian can link you to great resources, including:
- A literature review search (request one here)
- Databases of publications (including making an article request on your behalf here)
- A template to keep track of literature found
- A template to visualize your results
- Pro Tip: You may want to search key components of your question using this PubMed search function; this resource may also help guide your process.
Systematic review of the literature is often time consuming but very informative. The time invested in examining the literature can be translated into writing a review article. In this way, one research project can yield more than one publication.
- Ex: During your literature review, you found nine studies on alcohol withdrawal and gabapentin use. One included a multicenter randomized controlled study and a meta-analysis addressing your specific research questions. You also find studies on other adjunctive therapies for alcohol withdrawal. What you did not identify was patient surveys on symptomatology or a specific cost analysis on this topic. You can tackle this gap in knowledge by designing a new study.
Don’t be dismayed if you find a number of studies on a topic. It is still possible that more information is needed, and you can investigate a gap in the literature by altering or tweaking your original research question. This process is considered a gap analysis.
- Ex: You decide to write a review article including more recent research since prior review articles were more than five years old. You could also change the question to evaluate cost analysis.
To better formulate your question, use the PICO (or PICOT) framework discussed in the next section (“How do I turn it into a research study?”). Often the question will dictate the research design so it’s important to evaluate the question.
- Ex: You are providing care for a patient with substance use disorder being hospitalized for alcohol withdrawal. During rounds the senior resident suggests adding adjunctive therapy of gabapentin to the benzodiazepine protocol. This generates an idea: Does adding gabapentin decrease risk of progressing to delirium tremens or reduce hospital stay or ICU admissions?
- Pro Tip: The PICO (or PICOT) framework is a commonly used tool for framing a question. This (2min) video breaks down the use of PICO (and how it can be used to jumpstart a solid Lit Review). You can also use helpful worksheets like this one (à la MadLibs). As helpful as PICO(T) can be, it isn’t bulletproof; check out this page citing PICO’s limitations. There are other frameworks for polishing your research question. These can be helpful if your research topic doesn’t nicely fit the PICO(T) format. This website breaks down a few other more common frameworks while this website offers a more exhaustive list, categorized by discipline or type of question.
For further guidance on how to begin shaping your idea into a study, grant proposal, conversation starter with an advisor, and/or manuscript, check out this guide to writing a research proposal.
The clinical question can guide the research design. The design may need adjustment to ensure feasibility based on resources.
Question Type ⇒ Ideal Type of Study
Therapy ⇒ RCT
Prevention ⇒ RCT > Cohort Study > Case Control
Diagnosis ⇒ Prospective, blind controlled trial comparison GS
Prognosis ⇒ Cohort Study > Case Control > Case Series/Case Report
Etiology/Harm ⇒ RCT > Cohort Study > Case Control
Cost analysis ⇒ Economic analysis
If your relationship with statistics is a little… rusty, you may want to peruse this document outlining a few common medical statistics concepts before you meet with GME Research Staff, or check out this online course to brush up on:
1) Simple ways to detect statistical errors in medical literature
2) How to avoid errors by using principles of effective statistics
3) How to use free, online statistics tools that don’t require a programming background
- How various study designs fall along the levels of evidence as well as practical considerations for how to best answer your research question. Check out these graphics for guidance.
Note that due to time and resource constraints, resident-led randomized control trials and meta-analyses are not currently supported.
- Feasibility – this page provides an overview
Ah, the million dollar question. To set yourself up for success, chat with GME Research Staff about how realistic study completion is given the amount of time you have to dedicate to research and the number of years remaining in your training.
Consider the IRB approval timeline; all patient-centered research requires IRB approval prior to initiation.
- The IRB applications have three review categories: Exempt, Expedited, and Full Board Review.
- Research that does not require IRB approval and is feasible to complete in a shorter time frame includes review articles, opinion papers, and educational and some non-patient based research.
- Based on what type of IRB review your study requires, IRB approval may take a few weeks to many months, assuming no amendments are needed.
- Case reports, case series, retrospective analysis are usually considered expedited and have the fastest turnaround time.
- The IRB meets the first Tuesday of every month except August and January. Completed applications need to be submitted according to the IRB schedule. Partial applications will not be considered. You should not contact the IRB for research specific questions; instead speak to the GME Research Staff.
- Consult this sample timeline or this estimation of how long various steps take to plan your study submission. Completed applications need to be submitted according to the IRB schedule requirements. IRB meeting due dates can be found on the CMH Institutional Review Board website on the intranet. To avoid a delay due to preventable resubmission, please review the document Common Challenges with IRB Apps.
Consider time involved for patient enrollment and monitoring. Depending on the research design and background, you may need to complete a power analysis with the assistance of the GME Research Staff to get an estimate of the number of people needed in your study to determine a statistical difference.
If you choose to use the EHR, an IT analyst can usually provide a data report within about a month. Please note that some data points are more easily pulled (e.g. weight or gestational age, which are discrete numbers) vs others (written notes). Depending on the data points needed for your study, you may need to go into a patient’s chart to manually abstract the data, which can impact your timeline.
Consider if your recruitment plan is feasible. Think through how many subjects would need to be enrolled per month/week/day to gather enough data to complete my study within a timely manner.
For further information, visit Feasibility Considerations for guidance on timeline, logistics, funding, CMH capacity and PI responsibilities, data, recruitment, and consenting.
Residents with a GME-approved budget and GME-approved and IRB-approved proposal are eligible for GME Resident Research Project Funding.
You must receive approval from both the IRB and your Program Director before your project can begin. The Program Director and ADGME will make the funding decision, and you may purchase supplies in advance with approval from the ADGME.
You MUST have receipts for reimbursement.
- Common covered expenses include: supplies/equipment, participant compensation for time or costs incurred due to participation. Expenses that often exceed allowable expenses: clinical coordinator, cost of drugs, blood draws, any and all imaging not covered by insurance for standard of care.
Monies directed toward a specific vendor have a minimum turnaround time of 30 days.
GME will provide financial support to attend a regional or national conference if your abstract or research is accepted for poster or podium presentation.
Let’s get you a dream team. Whose expertise (content or logistical) do you need at the table?
You will need a PI with expertise in the field and an expert in statistics to complete the analysis. All team members will need certification in human subjects research (CITI training) within the last three years and, if appropriate, training in obtaining informed consent for research participation. For most types of GME research, the CITI GCP – Social and Behavioral Research Best Practices for Clinical Research course will satisfy the IRB requirement.
- Go to the Collaborative Institutional Training Initiative (CITI) training page on the CMH Institutional Review Board website on the intranet to complete the course.
- Save an electronic copy of the certificate and send a copy to GME Research Department (email@example.com; firstname.lastname@example.org).
- The certificate is valid for three (3) years.
Your teammates will need time for pre-study training and to assist in the project. When approaching colleagues about joining your study team, remember that people are giving of their limited time. In line with their level of involvement, research team members should be listed as investigators, names on presentation, etc.
Playing the long game: it is never too early to think through any publication ambitions. Utilize the CMH Authorship Policy as a guide for who to include on a publication team, how to determine authorship order, and authorship responsibilities. All research members accessing PHI must be listed on the IRB application.