November 9, 2026

Finding Research Gaps: Questions, Evidence, and Common Mistakes

A research gap is not a blank space on a topic map. It is a specific, defensible claim that the existing evidence does not adequately answer a question that matters for a defined population and outcome. Clinicians and me

Written byWisPaper TeamAI Research Workflow Team
Editorial illustration for Finding Research Gaps: Questions, Evidence, and Common Mistakes

A research gap is not a blank space on a topic map. It is a specific, defensible claim that the existing evidence does not adequately answer a question that matters for a defined population and outcome. Clinicians and medical researchers preparing an evidence brief, a research update, or a team discussion need to move from a broad topic to a narrow, answerable question, then test whether the literature actually leaves that question open. This guide walks through that workflow: diagnosing your starting state, building the artifacts that make a gap visible, and recognizing when the result is ready to present.

What does a research gap actually look like?

A usable gap statement has three parts: a population, an outcome, and a reason the current evidence falls short. "There is little research on hypertension" is not a gap. "For adults with resistant hypertension, it remains unclear whether home blood-pressure monitoring improves long-term cardiovascular outcomes compared with office monitoring" is closer, because it names who, what, and where the evidence stops.

The reason matters as much as the question. Evidence can fall short because studies are absent, because they exist but were conducted in different populations, because outcomes were measured differently, or because the available studies have limitations that weaken confidence. These are different gaps, and they call for different next steps. A gap caused by indirectness may be addressed by finding more applicable studies; a gap caused by inconsistent outcome definitions may call for a synthesis that harmonizes them.

Before searching, write your gap statement in one sentence and underline the population and outcome. If you cannot underline them, the question is still too broad.

Readers arrive at this task from different places. The table below maps common starting states to the first decision each one requires.

Starting stateWhat you likely haveFirst decision
Topic onlyA broad subject, no defined population or outcomeNarrow to one population and one outcome before searching
Question draftedA PICO-style question, no evidence mapDecide which study designs and sources count as relevant
Some papers collectedA folder of references, no synthesisScreen for eligibility and extract outcomes consistently
Brief in progressDraft claims, partial citationsVerify each claim against a source and check citation integrity
Team discussion prepA position, no shared evidence baseBuild a shared library and a documented search record

If you are at the topic-only stage, resist the urge to search immediately. A search run before eligibility criteria exist produces a pile of records you cannot screen consistently. Cochrane guidance is explicit that searches should be planned from eligibility criteria, combining free-text terms with subject headings and using multiple sources. That planning step is what turns a topic into a question.

How do you turn a topic into an answerable question?

Use a structured frame. For a clinical question, specify the population, the intervention or exposure, the comparator, and the outcome. Then add two practical constraints: the setting and the time frame. These constraints are not decoration; they determine which studies are applicable.

Take the resistant-hypertension example. The population is adults with resistant hypertension. The intervention is home blood-pressure monitoring. The comparator is office monitoring. The outcome is long-term cardiovascular events. The setting might be primary care, and the time frame might be at least twelve months of follow-up.

Now test the question against the literature you can find. If studies exist but follow patients for only three months, your outcome and time frame are where the evidence is thin. That is a more precise gap than "understudied." Write the question, the constraints, and the reason each constraint matters in a short protocol note. This note becomes the reference point for every screening decision.

Which sources and search tactics should you use?

Different sources serve different jobs. PubMed supports field tags, phrase searching, filters, Advanced Search, saved searches, Similar Articles, and citation lookup, and its search details show how terms were translated and combined. That transparency helps you document a reproducible strategy. Google Scholar supports searching by author and title, date sorting and limits, Cited by and Related articles, citation export, and alerts, which makes it useful for forward citation chasing and for locating alternative versions of a work.

A practical sequence is:

  1. Build a free-text and subject-heading strategy in a database that exposes its translation, such as PubMed.
  2. Run the same concepts in a second source to catch records the first missed.
  3. Use citation chasing to move forward from a key paper and backward through its references.
  4. Record the date, source, exact strategy, and number of records at each step.

Cochrane guidance supports using multiple sources, documenting the process, deduplicating records, and screening in stages. It also notes that multiple reports may describe one study, so link reports before counting studies. That linking step prevents the common error of treating three papers from one trial as three independent findings.

When you need to find academic papers through a scholarly search interface, WisPaper search provides a starting point for locating records; the public Scholar Search page explains the entry point for readers new to the tool. Neither replaces the documentation discipline above.

How do you screen records without losing the thread?

Screening in stages keeps the work auditable. First screen titles and abstracts against your eligibility criteria. Then screen full texts. At each stage, record how many records moved forward and why others were excluded.

PRISMA 2020 provides a reporting checklist, expanded explanations, an abstract checklist, and flow-diagram templates for original and updated systematic reviews, including transparent reporting of search, selection, and synthesis methods. Even a focused evidence brief benefits from the same logic: a flow diagram forces you to account for every record.

Two mistakes are common here. The first is changing eligibility criteria mid-screen without recording the change. The second is counting reports rather than studies. Both distort the evidence map and make the gap look larger or smaller than it is. Keep a screening log with the criterion applied and the reason for exclusion. If you are working with a team, agree on a sample of records and compare decisions before screening the rest.

How do you judge evidence level, applicability, and indirectness?

These are separate judgments, and conflating them is a frequent source of overclaiming.

Evidence level describes the design and conduct of the studies. A randomized trial and an observational cohort answer questions differently, and the level appropriate to your question depends on what you are asking.

Applicability asks whether the study population, setting, and care context resemble yours. A trial conducted in a different health system may still be informative, but its applicability to your setting needs to be stated.

Indirectness is a narrower judgment: whether the intervention, comparator, or outcome in the study matches the one in your question. A study that measured a surrogate outcome is indirect for a question about clinical events.

Limitations are the weaknesses within a study, such as small samples or short follow-up. They affect confidence in that study's result, not whether the study is relevant to your question.

A decision table helps keep these apart:

JudgmentQuestion to askEffect on your gap claim
Evidence levelWhat design and conduct does this study have?Determines how much weight the result can carry
ApplicabilityDoes the population and setting match mine?Narrows or widens the gap for my context
IndirectnessDo intervention, comparator, and outcome match my question?Identifies where evidence is adjacent but not direct
LimitationsWhat weakens this study internally?Tempers confidence without changing relevance

If most available studies are direct but limited, your gap is about certainty. If they are well conducted but indirect, your gap is about applicability. State which one you mean.

Where do common mistakes creep in?

Most weak gap claims fail at one of five points.

Searching before defining eligibility. Without criteria, you cannot tell whether a missing paper is absent from the literature or absent from your search. Cochrane's emphasis on planning searches from eligibility criteria addresses this directly.

Relying on one source. A single database or search engine will miss records. Multiple sources and citation chasing reduce this risk.

Treating reports as studies. Link multiple reports of one study before counting. Otherwise a single trial can appear to be a body of evidence.

Confusing absence of evidence with evidence of absence. A gap in your search results is not proof that no studies exist. It may reflect indexing, language, or terminology. Rerunning searches before finalizing, as Cochrane recommends, catches some of this.

Letting citations drift from claims. A reference that does not support the sentence it is attached to undermines the whole brief. Verify each citation against the source, and check that bibliographic details are real. For BibTeX citation screening, TrueCite checks BibTeX against real academic databases to flag hallucinated references.

How do you organize the evidence so the gap becomes visible?

An evidence table is the artifact that turns a pile of papers into a gap statement. Columns should include study, population, intervention or exposure, comparator, outcomes, follow-up, design, and the judgments from the previous section. Rows should be studies, not reports.

Once the table is complete, read down each column. If the population column is uniform but the outcome column is sparse for your outcome of interest, the gap is visible. If follow-up is short across every study, that is your gap. The table also shows where studies cluster, which helps you avoid claiming a gap that the literature already fills.

Save the papers you rely on in one place so the table and the citations stay connected. My Library lets you save or upload papers and organize them in the WisPaper library. When you need to ask questions based on papers uploaded to My Library or added from search results, Library QA supports that step. Use it to interrogate your own collected evidence, then confirm each answer against the source text.

How can WisPaper support each step of the workflow?

The workflow above has distinct stages, and different WisPaper areas fit different stages. Use them where they match the task rather than as a single pass.

  • Finding records: WisPaper search for scholarly search, with the English homepage as a general entry point for new users.
  • Organizing evidence: My Library for saving and uploading papers, and Library QA for asking questions grounded in those papers.
  • Longer research tasks: Scholar Agent fits literature review, inspiration, or reproduction use cases when the task is larger than a single search.
  • Project-level work: Research Projects is the area for creating and managing a new project when the brief spans multiple sessions or collaborators.
  • Citation integrity: TrueCite for BibTeX screening against real academic databases.
  • Idea generation: Idea Discovery takes research topics, background, references, and experimental constraints to generate research ideas when you are moving from a gap to a proposed study.

A general introduction to WisPaper as an AI research assistant for finding, screening, organizing, and working with academic papers is available on the WisPaper homepage.

When is the gap statement ready to present?

A gap statement is ready when a colleague can trace it back to your evidence table and search record without asking you where a claim came from. Use this checklist:

  • The population and outcome are named in one sentence.
  • Eligibility criteria are written and were applied consistently.
  • The search strategy, sources, dates, and record counts are documented.
  • Reports are linked to studies before counting.
  • Each study is judged separately for evidence level, applicability, indirectness, and limitations.
  • The evidence table shows where the literature is sparse or indirect for your specific question.
  • Every citation in the draft has been checked against its source.
  • Searches were rerun before finalizing.

If any item is missing, the gap may still be real, but the brief is not yet defensible. The checklist is also a useful agenda for a team discussion: assign each item an owner and a date.

FAQs

Test it against multiple sources and citation chasing before concluding. If a concept appears under different terminology, try free-text variants and subject headings. PubMed's search details show how terms were translated and combined, which helps you see whether a concept was missed. Rerun searches before finalizing, as Cochrane recommends. If the gap persists across sources and strategies, you can describe it with more confidence, while still stating the limits of your search.