How to Synthesize Study Findings in a Review Article

To synthesize study findings in a review article, align studies around a common review question and comparable dimensions, then identify where findings converge, diverge, or depend on context or method. State what the studies collectively support rather than reporting each source one by one. Write the result as one qualified cross-study claim whose scope and certainty match the evidence, preserve credible exceptions and uncertainty, and verify that each part of the claim is traceable to the studies.

Sound synthesis depends on whether the studies are meaningfully comparable and on the context in which their findings were produced. Differences in methodology, populations, settings, measures, or other relevant conditions can affect whether findings support the same interpretation.

Evidence that converges under comparable conditions can support a coherent cross-study interpretation, while meaningful differences should remain visible rather than being forced into agreement. The strength and scope of the interpretation remain conditional on comparability, context, methodological differences, and the strength of the evidence.

Before comparing and connecting study findings in detail, establish a common synthesis frame that clarifies which evidence can be considered together and which differences need to be preserved during interpretation.

Table of Contents⌄

Establish a Common Synthesis Frame for the Studies

A synthesis frame defines what study findings will be compared and the shared basis on which that comparison will be interpreted.

Studies need this analytical frame before meaningful synthesis because findings can address related topics without representing the same comparison unit, common outcome, or conditions.

The frame should therefore be strict enough to support a valid comparison while preserving differences that matter to the review question.

Set the synthesis goal so the frame reflects the analytical question the review article needs to answer, then identify the comparison unit and the common concepts or outcomes that must align.

Comparability also depends on relevant conditions such as population, method, and context; differences may legitimately remain when they do not undermine the intended comparison.

This planning task is distinct from how you organize evidence before synthesis: organizing makes source material accessible, whereas the synthesis frame defines the common basis on which that material can be compared.

A visual synthesis frame can clarify which study findings share that basis and which differences must remain visible.

A short setup sequence keeps the synthesis goal, comparison unit, shared dimensions, and limits of comparability connected.

It establishes a defensible basis for focused comparison without treating legitimately different studies as equivalent.

1. Define the synthesis target: state the question or analytical claim that the combined study findings need to address.
2. Identify the comparison unit: specify the finding, outcome, concept, or relationship that will be compared across studies.
3. Select common dimensions: identify the concepts or outcomes that must have sufficiently similar meaning for the intended comparison.
4. Confirm meaningful comparability: check whether differences in population, method, context, or other relevant conditions permit the intended comparison while preserving differences that could change its interpretation.

Use the Review Question to Define the Synthesis Focus

The review question determines which findings and relationships belong in the synthesis by defining their analytical relevance to the synthesis focus.

Relevant findings directly inform the population, outcome, intervention or exposure, context, or concept specified by that question.

Evidence can still be informative without directly answering the review question, but it should not receive the same analytical emphasis as evidence that addresses the question itself.

Evidence relevance changes when a study examines a different population, measures a different outcome, evaluates another intervention or exposure, operates in a materially different context, or addresses a related but distinct conceptual focus.

These differences do not automatically exclude study findings; they determine whether those findings directly support the intended synthesis or provide only peripheral context.

The following checks keep peripheral findings from diluting the synthesis focus:

• Population: determine whether the studied population matches or meaningfully informs the population specified by the review question.
• Intervention or exposure: check whether the finding concerns the intervention, exposure, or relationship that the synthesis is intended to interpret.
• Outcome: confirm that the measured outcome directly represents, or is clearly relevant to, the outcome or concept under synthesis.
• Context: assess whether differences in setting or other contextual conditions change the finding's applicability to the review question.

A finding that directly addresses these elements belongs in the central synthesis when it bears on the analytical relationship defined by the review question.

A finding that is contextually interesting but addresses a different population, outcome, exposure, or conceptual relationship is peripheral and should remain secondary rather than shaping the main synthesis claim.

The visual relevance filter below clarifies that distinction without replacing the analytical judgment required by the review question.

Choose Consistent Dimensions for Comparing Findings

Compare study findings on the same decision-relevant comparison dimensions so equivalent attributes are examined across studies rather than whichever details each source foregrounds.

A consistent schema aligns comparable features while allowing their observed values or conditions to differ.

The dimensions that deserve the most emphasis depend on the review question and on whether a difference materially changes interpretation.

Core comparison dimensions commonly include population, setting, study design, measures, outcomes, direction of findings, and relevant limitations; magnitude can also be compared where studies report it in a meaningfully comparable form.

A dimension is core when it directly supports the review question, while an optional dimension matters when its variation changes the interpretation or comparability of a study finding.

Numerical comparison is not appropriate when studies represent conceptually different constructs or measures.

The following criteria keep equivalent attributes aligned while preserving study-specific conditions that carry interpretive meaning:

• Population and setting: compare who was studied and under which contextual conditions when those differences affect how findings can be interpreted together.
• Study design: distinguish relevant design characteristics when they qualify what can reasonably be inferred from the findings.
• Measures: compare what was measured and how the measure represents the concept or outcome under synthesis.
• Outcomes: align findings that address equivalent outcomes while retaining distinctions when outcome definitions differ materially.
• Direction of findings: record the direction or absence of the relationship or difference reported by each study, using the study's own substantive result rather than imposing a common numerical scale.
• Magnitude: compare the reported size of a finding only when the underlying measures and reporting basis permit a meaningful comparison.
• Limitations: retain study-specific constraints when they qualify the applicability, certainty, or interpretation of a finding.

A useful visual comparison frame shows how shared attributes can be applied consistently across studies while study-specific conditions that materially affect interpretation remain distinct.

Compare Findings to Identify Patterns, Themes, and Relationships Across Studies

Compare findings across studies by aligning comparable evidence and examining how the findings relate, rather than counting how often similar statements appear.

Cross-study patterns become meaningful when comparable findings show a recurring theme or relationship that remains interpretable across relevant study conditions.

Recurrence alone is insufficient because frequency does not establish the strength or importance of the supporting evidence.

The visual relationship map below shows how supporting, divergent, and qualifying findings can contribute to one cross-study pattern without reducing the synthesis to a count of studies.

Look for convergence where comparable findings support a similar interpretation and divergence where findings contradict, qualify, or depart from that interpretation.

Group findings into themes only when they express a meaningful shared concept or evidence relationship, not merely repeated wording.

Keep each proposed pattern traceable to its supporting studies and qualifying evidence so exceptions remain visible.

After the analytical relationship has been defined, visual representations can help you use tables and figures for synthesis by making cross-study relationships easier to inspect.

A pattern remains qualified when differences in context or study conditions limit how broadly the relationship can be interpreted.

The ordered comparison below converts extracted findings into defensible cross-study patterns by moving from comparable evidence to a qualified interpretation.

Each stage tests the relationship rather than treating similar statements as sufficient evidence of a pattern.

1. Align comparable findings: place findings addressing the same relevant outcome, concept, or relationship side by side so the comparison has a shared analytical basis.
2. Detect recurring or divergent relationships: identify where findings converge, contradict one another, or differ in direction or interpretation without treating frequency as evidence strength.
3. Group meaningful patterns: combine findings into a recurring pattern or theme when they express the same substantive relationship, while keeping the contributing studies traceable.
4. Test exceptions and conditions: examine whether the apparent pattern still holds across relevant populations, settings, methods, or other study conditions, and identify qualifying evidence where it does not.
5. Record the qualified pattern: state the cross-study relationship together with the supporting studies, contradictions, and conditions that limit or qualify its interpretation.

Group Findings by Shared Themes or Concepts

Group findings under a shared theme or concept when they express the same substantive idea, even when individual studies use different wording.

Conceptual similarity, rather than shared vocabulary alone, determines whether findings genuinely belong together.

The grouping should also remain relevant to the review question and represent a coherent relationship among findings with comparable meaning.

Use these criteria to test whether a proposed grouping reflects shared meaning:

• Conceptual similarity: the findings express the same underlying idea or phenomenon despite differences in terminology.
• Review-question relevance: the shared concept directly contributes to the relationship or issue the review question asks the synthesis to interpret.
• Comparable outcome meaning: the findings address outcomes or concepts with sufficiently similar substantive meaning for the intended grouping.
• Coherent relationship: the included findings relate to one another in a way that supports a collective interpretation rather than merely sharing a broad subject.

This distinction prevents themes from becoming storage labels rather than analytical categories.

Grouping findings for synthesis is also distinct from deciding how to choose an organizational structure for the review as a whole.

For example, one study might describe participants as having difficulty maintaining a behaviour, while another reports that the same behaviour declines when routine support is absent; if both findings express the underlying concept of difficulty sustaining that behaviour, they can form one conceptual grouping, and the grouping becomes an analytical theme only when the synthesis interprets what that shared relationship means across the studies.

A visual grouping can clarify how differently worded findings connect through one substantive concept without treating shared terminology as the basis for the theme.

Combine Findings That Converge on the Same Pattern

Convergent findings can support one shared claim when they align on the same substantive pattern, but the claim should retain differences in context, method, and strength of support that affect its interpretation.

The degree of convergence depends on the consistency of direction across relevant supporting studies and on whether important conditions qualify that agreement.

The number of supporting studies can describe the extent of support, but it does not by itself determine claim strength or certainty.

Distinguish the convergence state before stating the shared pattern:

• Full convergence: relevant findings consistently support the same direction and shared pattern, with no identified differences among the compared studies that materially change the interpretation; the synthesized claim can state the common finding while remaining bounded by the contexts represented.
• Partial convergence: findings support the same general pattern, but meaningful differences in direction, method, context, relevance, or evidence strength qualify the consistency of support; the shared claim should state both the agreement and the qualification.
• Convergence under specific conditions: findings converge only within particular populations, settings, methods, or other relevant conditions; the synthesized claim should attach those conditions directly to the shared pattern rather than generalizing beyond them.

For example, if several relevant studies are consistent with an association under one context but the relationship is weaker or absent under another condition, a synthesis can state: “The findings converge on the association under the first context, while differences under the second condition limit how broadly that pattern can be generalized.” This wording combines the common finding with its boundary conditions instead of treating convergence as certainty.

Separate Complementary Findings From Genuine Contradictions

Complementary findings describe different but compatible parts of the same phenomenon, whereas a genuine contradiction occurs when findings point in opposing directions on a comparable question under sufficiently similar conditions.

An observed difference should therefore be classified only after establishing a shared comparison basis.

Findings that address different outcomes, populations, contexts, measurements, or study designs can remain compatible even when their reported results appear different.

The comparison below separates additive evidence from evidence that points in opposing directions on a comparable basis.

Comparison basis	Complementary findings	Genuine contradiction
Comparable question	Address different aspects of the same broader question.	Address the same substantive question but support opposing interpretations.
Population and context	Differ in populations, settings, or conditions in ways that can make both findings compatible.	Concern sufficiently comparable populations and contexts, so contextual differences do not account for the opposing direction of findings.
Measurement	Measure different outcomes or different aspects of the phenomenon, allowing the results to add distinct information.	Measure sufficiently comparable outcomes or constructs but report findings that point in opposing directions.
Study design	Use designs that answer different aspects of the question, so the findings can contribute different forms of evidence.	Remain opposed after relevant study design differences are considered rather than differing because the designs address different relationships.
Direction of findings	Findings differ without making incompatible claims about the same relationship.	Findings support opposing interpretations of the same relationship on the shared comparison basis.

Before labeling conflicting findings as a contradiction, compare the question addressed, population, context, measurement, study design, and direction of findings.

A difference on any of these dimensions can change the classification from genuine conflict to complementary evidence or an apparent conflict created by different conditions.

For example, one study could report a positive association in one population while another reports no association in a different population; these findings can remain compatible when the population difference defines distinct conditions for the observed relationship.

They constitute a genuine contradiction only when the relevant comparison basis is sufficiently aligned and the findings still support opposing interpretations.

Explain Why Study Findings Agree or Conflict

Agreement or conflict across studies requires a plausible explanation grounded in documented study attributes rather than an assumption about why the findings differ.

An observed difference becomes interpretable when a corresponding difference in population, context, study design, measurement, or another relevant condition could reasonably affect the findings.

The explanation should remain traceable to the studies being synthesized and should preserve uncertainty when more than one account is consistent with the observed pattern.

Candidate explanations can be separated into substantive or contextual differences and methodological differences.

This distinction helps connect a documented study condition to a possible interpretive effect without treating cross-study variation as proof of causation:

• Substantive or contextual explanations: differences in population, setting, context, or relevant conditions may help explain why an association or outcome appears in one study but differs in another, particularly when those differences change the circumstances to which the finding applies.
• Methodological explanations: differences in study design, sampling, measurement, or analytical method may help explain variation when studies capture the same phenomenon in different ways or under different methodological constraints.

For example, if two studies examine a similar relationship but use different populations and different measurement approaches, either difference may be a plausible explanation for their conflicting findings.

The interpretation should therefore state which documented study attributes are consistent with the observed divergence and avoid selecting one explanation as definitive unless the study evidence can distinguish among the alternatives.

When several explanations remain plausible, the synthesis should retain that uncertainty rather than converting an evidence-traceable possibility into a causal claim.

Relate Differences to Population, Context, and Conditions

Contextual differences can change how an apparently similar finding should be interpreted across studies.

A finding observed in one population, setting, time period, or exposure condition may have different applicability when those study conditions change.

The relevant comparison is therefore not only whether the findings look similar, but whether the studies are sufficiently comparable on the contextual attributes that define where and to whom the finding applies.

The table maps contextual differences to changes in observed findings and to cautious implications for interpretation and transferability.

Contextual attribute	Study-specific condition	Observed finding	Interpretive implication
Population	Different populations or subgroups are represented across studies.	The direction or apparent strength of a finding differs between the studied groups.	The finding applies most directly to the population in which it was observed and should not be transferred unchanged to another subgroup without supporting evidence.
Setting	Studies are conducted in different institutional, social, geographic, or practical settings.	A relationship is observed in one setting but differs or is absent in another.	Setting differences may limit comparability and may indicate that applicability is bounded by the contexts represented.
Timing	Studies observe the phenomenon at different periods, durations, or stages.	The observed pattern differs according to when the outcome or relationship is measured.	Interpretation should remain tied to the timing represented by each study rather than assuming the finding is stable across periods.
Exposure or other conditions	Participants or study units experience different levels, forms, or circumstances of the relevant exposure or condition.	The finding is present under one condition but differs or is absent under another.	The apparent difference may be compatible with condition-specific applicability, but it should not be presented as a causal effect unless the evidence supports that inference.

Each contextual difference should be connected to the documented study state, the observed finding, and a qualified interpretation.

Differences in population, setting, timing, or conditions can modify comparability and may limit generalization, but they do not by themselves establish why the findings differ.

For example, if a relationship is observed in one subgroup under a particular setting but is not observed in another subgroup studied under different conditions, the first finding should not be generalized automatically to the second group.

The defensible interpretation is that applicability remains bounded by the contexts represented until evidence supports transfer beyond them.

Relate Differences to Study Design and Methods

Methodological differences can plausibly account for variation in findings even when studies address a similar question because study design and methodological choices affect what is observed and how it is interpreted.

Differences in sampling, operational definition, measurement, follow-up, or analytical approach can therefore make apparently comparable findings represent different observations.

These differences warrant consideration when they could materially affect the finding, but they should not be treated as definitive explanations without supporting evidence.

The table connects methodological choices to the kinds of differences they can create in observed results and the resulting implications for interpretation.

Methodological attribute	Study choice	Potential effect on observation	Interpretive implication
Study design	Studies use designs that observe or estimate the relationship in different ways.	The findings may represent different forms of evidence about the same question.	A design difference can limit direct comparability and should qualify conclusions that combine the findings.
Sampling	Studies use different sampling approaches or selection criteria.	The resulting samples may represent different portions of the population relevant to the question.	Variation in findings may be consistent with sampling differences, but the difference does not establish that sampling produced the variation.
Operational definition and measurement	Studies define or measure the same underlying concept differently.	The measurement instruments may capture different aspects or representations of the concept.	Apparently conflicting findings may not be directly equivalent when the measured construct differs in substantive meaning.
Follow-up	Studies observe outcomes over different follow-up periods.	The recorded finding may represent the outcome at different stages or observation periods.	Differences in follow-up may limit comparison when the timing of observation changes the meaning of the reported finding.
Analytical approach	Studies analyse the evidence using different models, variables, or analytical procedures.	The reported estimates or relationships may differ according to what the analysis represents or adjusts for.	The analytical difference may help explain variation, but its relevance depends on whether it materially changes interpretation of the finding.
Methodological limitation	A study has a documented methodological constraint relevant to the finding.	The constraint limits what can be observed or inferred from that finding.	The limitation should qualify the synthesis only to the extent that it affects the comparison or interpretation at issue.

Methodological differences are most relevant when a specific study choice can plausibly change the observation being compared; differences unlikely to alter the meaning of the finding need not receive the same interpretive weight.

For example, if two studies address the same concept but one uses a narrowly defined measurement instrument while another uses a broader operational definition, their different findings may partly reflect what each instrument captures rather than a substantive conflict.

A similar qualification applies when sampling choices produce samples with different coverage of the population relevant to the question.

These methodological choices provide plausible explanations for variation, but uncertainty should remain when the study evidence cannot establish that a particular choice actually produced the observed difference.

Turn Cross-Study Patterns Into a Critical Synthesis

Critical synthesis transforms a cross-study pattern into an evaluated interpretation of what the evidence collectively supports, rather than merely restating individual studies.

Move from the identified pattern to the supporting evidence, then evaluate the consistency, relevance, methodological limitations, contradictions, and contextual boundaries that affect interpretation.

The same cross-study pattern can support different conclusions depending on whether the evidence is broadly consistent, limited to particular settings, weakened by methodological constraints, or challenged by credible contradictory findings.

Interpretation should therefore explain not only what the pattern is, but why the combined evidence warrants a particular level of claim strength.

The resulting synthesized claim should state the shared pattern together with the limitations or conditions that define how far the evidence can reasonably support it.

The sequence below makes the analytical reasoning visible without reducing critical synthesis to a fixed formula.

It moves from description to evaluation, interpretation, and a qualified synthesized claim.

1. State the pattern: identify the cross-study pattern that appears across the relevant findings without repeating each source separately.
2. Evaluate the support: weigh how consistent and relevant the evidence is, including methodological limitations, contradictions, and contextual boundaries that may qualify the pattern.
3. Interpret the relationship: explain what the evaluated pattern collectively means and why the supporting and conflicting evidence changes its significance.
4. Formulate the qualified claim: write a synthesized claim that reflects the evidence strength and explicitly retains the limitations or conditions that restrict its scope.

Evaluate the Relevance and Limitations of Evidence Behind Each Pattern

A cross-study pattern supports a synthesis claim only to the extent that the evidence behind the pattern is relevant to the review question, sufficiently comparable, reasonably consistent, and appropriately qualified.

Evidence quality should therefore be evaluated in relation to the specific claim rather than through a generic score or checklist.

Some evidence conditions strengthen support, while others narrow the permissible scope or require more cautious wording.

The criteria below connect each condition to its implication for claim strength.

Claim-strength criteria

• Directness: evidence that directly addresses the review question, relevant population, exposure or intervention, and outcome can support a more direct synthesized claim; indirect evidence warrants narrower wording.
• Comparability: studies examining sufficiently comparable concepts, outcomes, and conditions can contribute more coherently to the same pattern; material differences may restrict how confidently their findings can be combined.
• Methodological limitations: limitations that materially affect observation or interpretation can reduce confidence in the pattern, while limitations unrelated to the intended claim may have little effect on claim strength.
• Consistency: a coherent direction across relevant findings strengthens support, whereas unresolved contradiction or substantial variation requires qualification rather than automatic rejection of the evidence.
• Applicability: evidence drawn from populations, settings, and conditions that match the target context supports greater applicability within that context; mismatches narrow the claim to the circumstances actually represented.
• Uncertainty: unresolved methodological, contextual, or evidential uncertainty warrants more cautious wording and can limit how much weight the pattern carries.

The importance of each criterion depends on how directly it affects the intended claim, so not all limitations should receive equal weight.

A methodological limitation that changes measurement of the central outcome may materially reduce claim strength, whereas a minor limitation unrelated to that outcome may not alter interpretation.

Likewise, consistent evidence can still justify only a narrow claim when applicability is limited to a particular population or setting.

When unresolved evidence extends beyond the current synthesis, it may help identify gaps from the evidence without turning this evaluation into research-gap analysis.

For example, a well-conducted study may provide strong evidence for an outcome in one narrowly defined population but have limited directness to a synthesis focused on a different population or outcome.

The finding can remain credible within its original scope while carrying less weight for the broader claim, so the appropriate response is to narrow the claim rather than treat the evidence as generally weak or unusable.

Write Synthesized Claims From Multiple Studies

A synthesized claim should state the cross-study interpretation first, then ground that interpretation in the evidence pattern across multiple studies and qualify it where context, inconsistency, or limitations matter.

The analytical claim should remain the grammatical and semantic center, with individual studies functioning as support rather than becoming a source-by-source sentence chain.

A strong synthesized claim therefore reflects what the evidence collectively supports, how consistent that support is, and the conditions that limit its scope.

Claim wording should match the direction, consistency, context, and evidence strength of the underlying findings.

When multiple studies show a similar direction under comparable conditions, the claim can be stated more directly; when findings are mixed, context-dependent, or methodologically limited, the wording should include an explicit qualification.

The implication should follow from the evaluated evidence pattern rather than from citation quantity alone.

This keeps the interpretation proportional to the support behind it and avoids overstating certainty.

Connect Studies Around One Analytical Point

A synthesis paragraph should be organized around one analytical point, with multiple studies functioning as evidence for that point rather than being reported one source at a time.

Begin with a clear topic claim, then connect the studies according to the relationship among their findings.

The writer's analytical voice should remain responsible for the paragraph-level interpretation, while the studies support, extend, qualify, contrast with, or help explain that interpretation.

Make each relationship explicit instead of relying on transition words alone.

Agreement shows that studies support the same analytical point; extension adds a further dimension; qualification limits the scope of the point; contrast identifies a meaningful difference; and explanation links that difference to a plausible interpretive condition.

The paragraph should then return to the analytical point by stating what these evidence relationships collectively imply.

This gives the paragraph a clear semantic endpoint rather than leaving the reader to infer why the studies were placed together.

Preserve Exceptions, Uncertainty, and Conflicting Evidence

Exceptions, mixed results, and unresolved conflicting evidence should remain visible whenever they change the claim scope or confidence of a synthesis.

A credible outlier can identify a boundary condition that narrows an otherwise consistent pattern, while uncertainty can require qualification even when the dominant direction of evidence remains clear.

Conflicting evidence should support competing interpretations or cautious wording when the available studies do not justify a single resolution.

The criteria below connect each evidence state to the claim adjustment it requires.

• Exception: a credible finding departs from the dominant pattern under a relevant condition; treat the exception as a potential boundary condition and narrow the claim scope when that condition materially affects applicability.
• Uncertainty: the evidence leaves an important aspect of the pattern unresolved; reduce confidence in the corresponding interpretation and use only the qualification needed to reflect what remains uncertain.
• Mixed results: relevant findings support different directions or levels of consistency without establishing one interpretation across the represented conditions; state the mixed evidence explicitly and avoid wording that implies uniform agreement.
• Unresolved conflict: sufficiently comparable findings support competing interpretations that cannot be reconciled by the studies being synthesized; retain both interpretations and withhold a firm conclusion where choosing between them would exceed the evidence.

A single outlier does not automatically overturn a broader pattern, but it can materially change the wording when it is credible and identifies a condition under which the pattern does not hold.

For example, if the evidence is otherwise consistent across the represented studies but one credible finding diverges in a distinct population, the synthesis can describe the pattern as consistent within the populations that support it while identifying the divergent population as an exception.

This preserves the dominant interpretation without extending it beyond the evidence distribution.

Verify That Synthesized Claims Are Supported by the Evidence

Every synthesized claim should be traceable to the relevant studies and proportionate to the supporting evidence.

Verification should confirm that the claim represents the complete relevant evidence pattern rather than only the findings that support its main direction.

A claim passes only when its wording, scope, and certainty align with the evidence that supports, limits, or conflicts with it.

Check the claim against supporting evidence, conflicting evidence, contextual limits, methodological limits, and uncertainty without reopening the synthesis or introducing new interpretations.

Traceability should make it possible to identify which studies support the claim and which evidence requires qualification.

The verification also tests whether the claim answers the synthesis focus and whether its strength remains proportionate to what the complete evidence pattern can support.

Verification checks

• Traceability: Can the synthesized claim be traced to the relevant supporting studies and the findings that provide its evidence support?
• Evidence completeness: Does the claim represent relevant supporting evidence, conflicting evidence, credible exceptions, and mixed results rather than only the dominant pattern?
• Contextual qualification: Does the claim preserve contextual limits that narrow the populations, settings, conditions, or circumstances to which the interpretation applies?
• Methodological qualification: Are methodological limits retained when they materially restrict the comparison, interpretation, or claim strength?
• Uncertainty: Does the wording express unresolved uncertainty without implying greater confidence than the evidence supports?
• Synthesis alignment: Does the claim answer the synthesis focus while keeping its scope and certainty proportionate to the complete relevant evidence pattern?

If any check fails, revise the synthesized claim or revisit the comparison logic rather than attempting to compensate by simply adding more citations.