How to Design a Simple Research Study

As a surgical trainee, your time is relentlessly squeezed between theatre lists, clinic, on-call shifts and exam preparation. Designing a research study often feels like an impossible luxury reserved for those out-of-programme academics. However, building a robust, achievable project is entirely within your reach if you focus on surgical pragmatism over scientific perfection.

Start with a Highly Specific Clinical Question

Every monumental piece of surgical research began as a simple, irritating question muttered in a scrub room. The biggest mistake trainees make is starting with a broad, sweeping topic. You do not have the funding, the time, or the statistical manpower to answer a question like, "What are the outcomes of total knee replacements?" You do, however, have the capacity to ask, "In patients under sixty undergoing a primary total knee replacement for osteoarthritis, is there a difference in two-year revision rates between ceramic-on-polyethylene and metal-on-polyethylene bearings?"

To ensure your idea is realistic, you must ruthlessly distil it using the PICO framework. This is not merely a tedious academic hurdle; it is the architectural blueprint for your entire project. PICO forces you to define your patient population, the intervention, the comparator, and the outcomes.

The PICO Breakdown

• P (Population): Who exactly are you looking at? Define the age, condition, and setting. "Patients with distal radius fractures" is too vague. "Adults over 65 presenting to a single trauma unit with isolated, closed distal radius fractures" is a measurable cohort.
• I (Intervention): What is the specific exposure, procedure, or treatment? "Arthroplasty" is unhelpful. "Reverse total shoulder arthroplasty" is precise.
• C (Comparator): What are you measuring the intervention against? This could be a different surgical technique, a conservative management pathway, or a historical cohort.
• O (Outcome): What is the exact endpoint? Crucially, choose an outcome that is realistic to capture. Patient-reported outcome measures (PROMs) are excellent but notoriously difficult to collect with high follow-up rates in busy trauma centres. Objective outcomes like re-operation rates, readmission within thirty days, or implant survivorship pulled from a registry are far more robust and achievable for a time-poor trainee.

Assess Feasibility Before You Commit

The absolute most common reason surgical trainees abandon a research project is poor initial feasibility assessment. You might have a brilliant PICO, but if the data does not exist, or you cannot access it, the study will bleed your time and yield nothing. Before you write a single line of a protocol, conduct a brutally honest audit of your resources.

Ask yourself how many patients actually meet your inclusion criteria in your hospital over a given period. A quick search of your trust’s theatre coding system will tell you immediately if you are looking at fifty patients a year or five. Then, determine where the data lives. Is it locked in a fragmented paper notes system, or is it neatly housed in a digital electronic health record?

Ethics and governance are often viewed as terrifying bureaucratic monsters, but understanding the landscape early will dictate your study design.

Governance Pathways

• Service Evaluation: Uses routine data already collected, with no intention of changing clinical practice. This usually only requires local hospital registration, making it highly attractive for trainees.
• Clinical Audit: Measures current practice against an established, evidence-based gold standard. Again, this requires only local audit department registration.
• Research: Generates new knowledge or tests a novel hypothesis. This almost always requires formal ethical approval from a national research body and site-specific governance sign-off. This process is lengthy and should only be pursued if your question absolutely demands it.

Choose the Right Study Design for the Real World

When time is your most scarce commodity, the elegance of your study design lies in its simplicity. Prospective, randomised controlled trials are the gold standard of evidence-based medicine, but they are logistical nightmares for a standalone trainee juggling full-time clinical work. Instead, lean heavily into designs that leverage existing infrastructure.

A retrospective cohort study is the workhorse of surgical trainee research. By using historical data from your hospital’s electronic records, theatre management systems, or local registries, you can conduct a high-yield study without waiting years for prospective data to mature. The key is to acknowledge the inherent biases—such as missing data or selection bias—and design your analysis to mitigate them.

Case series can be valuable, particularly for rare presentations or novel surgical techniques, but they rank low on the hierarchy of evidence. If you are going to present a case series, ensure the cases are genuinely fascinating, impeccably documented, and accompanied by a thorough review of the existing literature.

When to use a systematic review

If you are entirely bereft of local data, or if your clinical commitments are unusually punishing for the next few months, a systematic review and meta-analysis is an excellent alternative. These require rigorous methodology and an understanding of PRISMA guidelines, but they allow you to dictate your own working hours from the comfort of your laptop. Collaborating with a local university library specialist can save you weeks of agonising over search strategies and database parameters.

Crafting a Protocol That Actually Works

A protocol is not just a document to appease an ethics committee; it is your operational manual. The moment you begin data collection without a clearly written protocol, you are engineering disaster. Protocols prevent data dredging—the unethical practice of endlessly torturing a dataset until a statistically significant, yet clinically meaningless, result accidentally emerges.

Your protocol must explicitly detail your inclusion and exclusion criteria. Be cautious not to stack your exclusions so high that you create an artificial, impossible patient cohort that no longer reflects the reality of the patients walking through your trauma clinic doors.

Equally important is your statistical analysis plan. You do not need to be a biostatistician, but you must know what tests you will run before you collect the data. If you do not know how the data will be analysed, you do not know which variables you need to extract. If you are comparing a continuous variable (like the Visual Analogue Scale for pain) between two groups, you need a t-test or a Mann-Whitney U test. If you are comparing categorical data (like the proportion of patients requiring re-operation), you need a Chi-square test. State these in the protocol.

Finally, include a sample size calculation early. If your calculation tells you that you need a thousand patients to prove your hypothesis, but your hospital only sees fifty such cases a decade, the study is dead in the water. You must pivot your question or collaborate with other hospitals to pool data.

Assembling Your Team and Leveraging Collaboratives

Surgical research is an undeniably team-based endeavour. Operating in isolation is the fastest route to burnout and substandard results. As a trainee, your primary goal is to deliver the project, but your secondary goal is to network and learn the research process from those further along the pathway.

You need a supervisor or a consultant champion. Look for a surgeon who values academic inquiry and, crucially, has a track record of supporting trainees to publication. Their role is to help you refine the clinical relevance of the question and to provide institutional weight when you are asking busy ward clerks for access to patient notes.

Data collection is the most laborious phase, so divide and conquer. Recruit fellow trainees or eager medical students to help with the electronic data extraction. In return, offer them co-authorship. This teaches them the mechanics of research and preserves your evenings.

Furthermore, tap into the immense power of surgical collaborative networks. Over the last decade, trainee-led collaboratives have revolutionised how we conduct multicentre surgical research. By joining an established collaborative, you can contribute to massive, high-impact trials or cohort studies using standardised protocols. This allows you to generate robust data from your specific hospital, act as the local principal investigator, and earn authorship on papers published in premier surgical journals without the burden of designing the entire study yourself.

The Mechanics of Data Collection and Management

The way you collect and store your data will determine the quality of your study. The cardinal rule of data management is to maintain a clean, standardised dataset.

Create a data collection proforma—a simple spreadsheet where every column represents a specific variable. Ensure that data is entered uniformly. If one person records "Right" and another records "R" or "1", your dataset will become an unmanageable mess. Implement a data dictionary, a separate document that explicitly states how every variable should be coded.

Clinical data is never perfect. Patients will inevitably have missing radiographic measurements or undocumented time intervals. Decide in your protocol how you will handle missing data. You cannot simply leave cells blank or delete the patient entirely, as this introduces significant bias. You must clearly document your methodology.

Patient data security is an absolute, non-negotiable priority. Never, under any circumstances, store identifiable patient information on personal laptops, local hard drives, or unencrypted memory sticks. Create a pseudonymised or fully anonymised dataset from the very beginning. Assign every patient a unique study number. Keep the "key" that links the study number to the hospital number locked securely on a hospital trust computer, completely separate from your research dataset.

Data Analysis, Interpretation and Writing Up

Once you have your data, the analysis begins. This is where many clinicians panic. However, the most critical aspect of data analysis is not the mathematical computation, but the clinical interpretation. Statistics are a tool to help you understand the clinical reality, not a magic box that automatically generates conclusions.

If you have followed your statistical analysis plan, you simply execute the tests you outlined months ago. Focus on what the data is actually telling you, rather than what you hoped it would say. A negative result—that a new surgical implant shows no significant superiority over the traditional one—is just as valuable to the surgical community as a positive one. Do not fall into the trap of twisting your data to find a positive outcome just to satisfy a perceived publication bias.

When writing your manuscript, adhere strictly to standard reporting guidelines depending on your study design. The STROBE checklist is the gold standard for observational studies, detailing exactly what must be included in your methodology and results. Following these checklists will dramatically improve the quality of your writing and smooth the peer-review process.

Target your journal wisely. Submitting a retrospective single-centre case series to a high-impact international journal may result in months of frustrating rejections. Consider regional orthopaedic journals or highly respected specialty-specific publications that value the clinical reality of district general hospital practice.

Surgical research does not require you to abandon your clinical training or sacrifice your exam preparation. By grounding your study in an answerable question, leveraging the data right in front of you, and executing a rigorous, simple design, you can produce meaningful work that advances orthopaedic knowledge and accelerates your surgical career.