What is a research gap?
A research gap is the specific, unanswered problem inside a topic — what is still unknown, uncertain, conflicting, under-studied, or not yet applied to your population or setting. A clear gap is your defensible contribution and your main shield against a “not novel” rejection.
The 5 steps to find one
- Capture the observation — write what you noticed in one sentence (e.g. “Diabetic inpatients on my service seem to fall more often”).
- Broaden into angles — list the possible drivers (glucose variability? medications? mobility?).
- Rapid literature scan — read TL;DRs in PubMed and Semantic Scholar; ask Consensus whether the evidence agrees. Don’t read 40 papers.
- Sort into three buckets — Established (don’t repeat), Conflicting (settle the debate), Missing (nobody answered). Your gap lives in the last two.
- Refine to feasible — narrow to a specific population, a measurable outcome, and a realistic timeframe.
Free AI tools that speed this up
AI accelerates discovery; you stay the judge of novelty, feasibility and ethics. A free stack covers most of the work:
| Tool | Best for |
|---|---|
| Perplexity | Quick, sourced topic scan |
| Semantic Scholar | Discovery + paper TL;DRs |
| Consensus | Does the evidence agree? |
| Elicit | Extract findings into a table |
| Research Rabbit | Map the citation network |
Golden rule: never cite an AI-generated reference without opening the real paper.
Frequently asked questions
What is a research gap in clinical research?
It is the specific unanswered question within a topic — something unknown, uncertain, conflicting, under-studied, or not yet applied to your population or setting. It is what makes a study novel and publishable.
How do I find a research gap quickly?
Capture a clinical observation, broaden it into angles, run a rapid literature scan (read TL;DRs, not full papers), sort findings into Established/Conflicting/Missing, then refine the gap to a feasible study.
Which free AI tools help find a research gap?
Perplexity and Semantic Scholar for discovery, Consensus to check whether evidence agrees, Elicit to extract findings into a table, and Research Rabbit to map citations — all have free tiers.
Can AI write my literature review or find citations for me?
AI can accelerate discovery and drafting, but it can invent citations and overstate novelty. Always open and verify the real paper before citing it; you remain the judge of quality and novelty.
What makes a research gap strong enough to publish?
A strong gap is specific (defined population, measurable outcome, clear setting) and sits in the Conflicting or Missing bucket — it resolves a debate or answers something no one has answered for your context.