Defining Scope, Delimitations and Limitations

Overview

Every study has boundaries and imperfections. Transparent research distinguishes the scope of what the study addresses, the delimitations deliberately chosen by the investigator and the limitations that constrain the strength or reach of the conclusions.

The scope defines the study’s substantive and operational coverage: population, setting, time, variables or phenomena, design and intended inference. Delimitations are intentional choices that narrow this scope—for example, selecting one clinical setting, age group or primary outcome. Limitations are features that may weaken validity, precision, transferability or completeness, whether they arise from the design, available data, implementation or unavoidable constraints.

These terms are sometimes used inconsistently across disciplines. The most important practice is to describe each boundary clearly, explain why it exists and state how it affects interpretation [1–3].

Why this matters

A study can be criticised unfairly for not answering a question outside its intended scope. Conversely, investigators may hide preventable weaknesses by calling them delimitations. Clear boundaries help readers judge what the study was designed to establish and what remains uncertain.

Limitations are not a ritual paragraph stating that time and money were limited. They should identify mechanisms that could distort results, the likely direction or nature of the effect, steps taken to reduce it and the conclusions that remain supportable. Transparent limitation reporting is necessary for appropriate use of evidence [3,4].

Defining study scope

Scope answers: What exactly does this study cover? It should be consistent with the question and objectives.

Population

Who or what is included? Specify age, condition, eligibility, unit of analysis and relevant exclusions. Distinguish the target population from the accessible source population.

Setting

Identify the healthcare level, institutions, communities, laboratories, databases or jurisdictions involved. Context affects implementation and applicability.

Time

State recruitment period, exposure period, follow-up duration, historical data window and outcome timing.

Variables, phenomena and outcomes

Define which exposures, interventions, outcomes, experiences or processes are central. State what the study does not attempt to measure when ambiguity is likely.

Design and inference

Clarify whether the study estimates prevalence, examines association, evaluates an intervention, develops prediction, explores experience or maps evidence. The scope of inference should not exceed the design.

Analytical or conceptual perspective

Some studies focus on patient, provider, organisational or policy perspectives. Economic, equity or implementation perspectives may require additional data that are not automatically included.

Scope statement

illustrative

This study will include women diagnosed with gestational diabetes who deliver at the study hospital from January to December 2027. It will estimate documented postpartum glucose testing within 12 weeks and examine pre-specified patient and service factors associated with completion. Tests recorded in the hospital and participating public primary-care laboratories will be included. The study will not evaluate long-term diabetes incidence or the effectiveness of a new follow-up intervention.

A scope statement reduces ambiguity and prevents readers from assuming that broader outcomes were intended.

Delimitations

Delimitations are reasoned choices that define manageable and scientifically coherent boundaries. Examples include:

• selecting a clinically meaningful age group;
• focusing on one disease stage;
• restricting to settings with a shared care pathway;
• choosing a 12-week outcome window;
• prioritising one primary outcome;
• excluding people for whom the intervention is contraindicated; or
• limiting a qualitative study to participants with direct experience of the phenomenon.

A delimitation should have a methodological, clinical, ethical or feasibility rationale. “Because it is easier” is incomplete. The investigator should explain what the choice improves and what applicability it may reduce.

Limitations

Limitations are conditions that restrict what can be concluded. They may be anticipated during protocol development or discovered during conduct.

Design limitations

Cross-sectional designs may have ambiguous temporality. Uncontrolled before-and-after evaluations may be affected by secular trends. Small qualitative samples may not capture all relevant perspectives, although statistical representativeness is not their aim.

Selection limitations

Convenience recruitment, low participation or differential loss to follow-up may create systematic differences between included and target populations.

Measurement limitations

Self-report, imperfect instruments, inconsistent records or missing linkage can misclassify exposure or outcome. The consequence depends on how error relates to study groups and outcomes.

Confounding and causal limitations

Observational comparisons may retain residual or unmeasured confounding. Statistical adjustment does not guarantee causal identification.

Precision limitations

A small sample or rare outcome may produce wide confidence intervals. This limits certainty but should not be described merely as “lack of power” without presenting the actual precision.

Context and transferability limitations

A single setting may have distinctive pathways, resources or populations. Applicability to other contexts requires reasoned judgement rather than a binary claim of generalisability.

Implementation limitations

Protocol deviations, intervention contamination, incomplete adherence, staff turnover or changing policy may alter the intervention or exposure actually studied.

Assumptions are different

An assumption is a condition accepted for the purpose of design or analysis, such as assuming that a linked record corresponds to the correct participant or that a measurement scale functions similarly across language groups. Assumptions should be justified and assessed where possible.

An assumption can become a limitation when it is uncertain and materially affects conclusions. Listing assumptions does not remove the need to investigate their plausibility.

Writing limitations well

A useful limitation statement has four parts:

1. Name the mechanism: What feature creates the limitation?
2. Explain the consequence: Which estimate, interpretation or population may be affected?
3. Describe mitigation: What was done to prevent, measure or analyse the problem?
4. Set the interpretive boundary: What conclusion remains defensible and what should not be claimed?

Weak statement:

This study was limited by time and sample size.

Improved statement:

Recruitment was restricted to one hospital and 214 eligible women were included. The resulting estimate describes this care pathway and may not transfer to settings with integrated laboratory systems. The 95% confidence interval will be reported to show the precision of completion and association estimates rather than treating the sample size alone as a limitation.

Prospective planning for limitations

Limitations should be anticipated during protocol development. Early recognition allows mitigation through better sampling, additional data sources, validation substudies, standardised procedures, blinded assessment, follow-up strategies or sensitivity analyses.

PROTOCOL CHECKLIST

Plan study boundaries prospectively

• Define population, setting, study period and follow-up explicitly.
• List deliberate exclusions and justify each one.
• Identify major design, selection, measurement and confounding limitations.
• State which limitations can be prevented, measured or analysed.
• Pre-specify sensitivity analyses or validation checks where feasible.
• Explain how the limitations constrain causal, predictive or generalisable claims.
• Record protocol changes that alter the original scope.

Worked example

WORKED EXAMPLE

Classifying boundaries in the postpartum study

Scope: Women with gestational diabetes delivering at one tertiary hospital; documented testing within 12 weeks; patient and service factors; linked public-sector records.

Delimitations: Exclude women who move out of the catchment area before discharge because follow-up records cannot be linked; focus on the first 12 weeks because this is the specified service target; prioritise completion rather than long-term diabetes incidence.

Limitations: Tests performed in unlinked private facilities may be missed; some service variables are recorded inconsistently; the single-hospital pathway may differ from other institutions; observational associations may be confounded.

Mitigation: Ask participants about external testing, verify a sample against available documents, define variables in a data dictionary, describe non-participation, adjust for pre-specified confounders and conduct a sensitivity analysis assuming plausible numbers of missed tests.

Interpretation: The study can estimate documented completion within the linked pathway and identify associations that support service planning. It cannot prove that a particular barrier causes non-completion or that the estimate applies unchanged to all hospitals.

Appraising boundaries and limitations

APPRAISAL CHECKLIST

Check the scope and limitations

• Is the scope explicit and aligned with the question and objectives?
• Are population, setting, time, variables and intended inference defined?
• Are delimitations clearly identified as investigator choices and justified?
• Could any stated “delimitation” actually be a preventable source of bias?
• Are limitations described by mechanism rather than generic labels?
• Is the likely effect on validity, precision or transferability explained?
• Are mitigation and sensitivity analyses reported where feasible?
• Do the conclusions remain within the stated boundaries?
• Are limitations balanced rather than used to dismiss the entire study or minimise serious weaknesses?

Suggested protocol and reporting wording

Common mistakes

Knowledge check

What is the difference between a delimitation and a limitation?

Answer and reasoning

A delimitation is an intentional boundary chosen to define a coherent and feasible study. A limitation is a feature that constrains validity, precision, completeness or applicability. A choice can create a limitation, so the rationale and consequence must both be described.

Is recruitment from one hospital always a limitation?

Answer and reasoning

It is a scope boundary and deliberate delimitation. It may limit transferability when the hospital, pathway or population differs from other settings. The consequence should be explained rather than assumed.

Why is ‘small sample size’ an incomplete limitation statement?

Answer and reasoning

The relevant issue is the resulting uncertainty. Report confidence intervals, sparse outcomes or instability and explain which conclusions are imprecise. A sample can be small yet adequate for a focused qualitative aim or precise enough for a common outcome.

How should investigators discuss possible missed outcome records?

Answer and reasoning

Describe how records may be missed, whether misclassification differs across participant groups, the likely effect on estimates, efforts to capture or validate outcomes and any sensitivity analyses. Do not merely say that records were incomplete.

Key take-home messages

SUMMARY

Key take-home messages

• Scope defines what the study covers; delimitations are deliberate boundaries; limitations constrain conclusions.
• Boundaries should be explicit across population, setting, period, variables, design and inference.
• Delimitations require justification and may trade broader applicability for coherence or validity.
• Limitations should be explained by mechanism, consequence, mitigation and interpretive boundary.
• Precision is communicated through uncertainty intervals, not by labelling a sample simply “small”.
• Conclusions must remain within the study’s designed and observed scope.

References

1. World Health Organization Regional Office for the Eastern Mediterranean. A practical guide for health researchers. Cairo: World Health Organization; 2004. https://iris.who.int/handle/10665/119703
2. Browner WS, Newman TB, Cummings SR, Grady DG, Huang AJ, Kanaya AM, et al. Designing clinical research. 5th ed. Philadelphia: Wolters Kluwer; 2022.
3. Price JH, Murnan J. Research limitations and the necessity of reporting them. Am J Health Educ. 2004;35(2):66–67. https://doi.org/10.1080/19325037.2004.10603611
4. Ross PT, Bibler Zaidi NL. Limited by our limitations. Perspect Med Educ. 2019;8(4):261–264. https://doi.org/10.1007/s40037-019-00530-x