01 · The Question
How do you screen a study when the usual English summary is missing?
You find a citation that appears potentially relevant, but there is no English abstract. Perhaps the title is translated into English. Perhaps only the original-language title is available. You may have author names, a journal, publication year, keywords, indexing terms, or a link to a full text you cannot readily read.
At that point, excluding the study is certainly convenient. It is not necessarily methodologically justified.
A missing English abstract is a missing discovery aid, not evidence that the study is irrelevant. The task is to gather enough information from other sources to make the same substantive eligibility decision you would make for any other record.
03 · What You Need to Know
No English abstract does not mean no usable information
First establish what information actually exists
A record without an English abstract may still contain considerable bibliographic information. Depending on the database and publication, you may have an English translated title, original-language title, author names, affiliations, publication type, keywords, controlled vocabulary, journal title, year, DOI, other identifiers, or links to the publisher.
PubMed provides a useful illustration. Its records can include English titles, original or transliterated non-English titles, publication language, author keywords, subject indexing, and other citation fields. PubMed also states that English abstracts are taken from the published article and that the National Library of Medicine does not create one when an article lacks a published abstract.
So the absence of an English abstract should prompt an inventory of the remaining metadata rather than an immediate exclusion.
A translated title may be enough to establish potential relevance
A title can sometimes tell you that a study is clearly irrelevant or plausibly relevant. If the title explicitly concerns a population or phenomenon outside your review scope, that may support exclusion under the same rule applied to any other record.
If the title sounds relevant but provides too little information about design, participants, intervention, setting, or outcomes, the record remains uncertain. Do not force a decision from information that cannot support it.
No evidence of eligibility
The available metadata do not yet show whether the study meets the criteria.
Evidence of ineligibility
The available information positively shows that a predefined criterion is not met.
Only the second provides a substantive reason for exclusion.
Look for an abstract in the original language
No English abstract does not necessarily mean no abstract exists. The journal or another bibliographic source may provide an abstract in the article's publication language.
PubMed, for example, may include non-English abstracts supplied by publishers and provides a way to identify citations containing them.
If an original-language abstract exists, translating that abstract may be enough for initial screening. This can be substantially more efficient than translating the entire article before you know whether it is relevant.
Check the journal website rather than relying on one database record
Bibliographic records are not always the richest source of metadata. The publisher or journal website may provide additional keywords, an original abstract, supplementary files, an HTML version, bilingual metadata, or a downloadable full text.
Search using the article title, author names, journal, year, volume, issue, page range, DOI, or other identifiers. If the journal itself is difficult to locate because it is outside major indexes, use the broader methods for finding research published in journals not indexed by major databases.
Use indexing terms as clues, not substitutes for the study
Controlled vocabulary and database-assigned publication types can reveal useful information. A record indexed under a relevant population, intervention, disease, educational concept, or study type may warrant further investigation even when the abstract is unavailable.
These fields should not be overinterpreted. Indexing practices vary, not every record is fully indexed, and a subject heading does not necessarily establish that a study satisfies your exact inclusion criteria.
Think of indexing as another piece of evidence about whether the record deserves further assessment.
Search for the same study elsewhere
The absence of an abstract in one database does not prove that no English summary exists anywhere. Search the exact title, DOI, author names, and bibliographic details in other scholarly systems, repositories, publisher pages, and library resources.
The study may also have related outputs. A conference abstract, thesis, protocol, registry record, preprint, institutional record, or subsequent article could provide additional information, although you must verify that these records genuinely concern the same study rather than merely a similar project.
References and citations can provide context
If the study is cited by another paper, examine how the citing authors describe it. A later systematic review or related article may identify its population, design, intervention, or findings.
This information can help you decide whether obtaining a translation is worthwhile. It should not automatically replace verification from the original study when your review requires primary-study information.
Translate in stages rather than translating everything immediately
Translation resources can be allocated according to the decision being made.
| Stage |
Possible language support |
Purpose |
| Initial identification |
Title and metadata translation |
Determine whether the record could plausibly be relevant. |
| Abstract screening |
Translation of an original-language abstract |
Assess obvious eligibility and decide whether full text is required. |
| Full-text eligibility |
Targeted or complete full-text assistance as needed |
Determine whether all required inclusion criteria are satisfied. |
| Critical appraisal |
Reliable access to relevant methodological sections |
Assess study quality or risk of bias using the chosen framework. |
| Data extraction |
Sufficient translation of methods, results, tables, figures, and relevant supplementary material |
Extract and interpret the required data accurately. |
JBI guidance explicitly notes that English abstracts of non-English papers can help reviewers decide whether to retrieve a full paper and seek translation. When no English abstract exists, the same logic can be applied one step earlier by obtaining enough translation of the available original-language information to make that preliminary decision.
Machine translation can help with triage, but verification should match the stakes
Machine translation can be useful for determining whether a title or abstract appears broadly relevant. Its reliability is not uniform across languages, disciplines, technical terminology, or document quality.
A low-stakes triage decision and extraction of a primary outcome are not equivalent tasks. As the consequence of misunderstanding increases, so should the level of language verification. For consequential methodological and numerical information, assistance from someone proficient in the language and familiar with the subject may be warranted.
Search design still matters before screening begins
Records without English abstracts are harder to screen, but they can also be harder to retrieve in the first place. If your search relies heavily on English title and abstract words, some of these records may never enter the candidate set.
This is why you may need to translate selected search terms into other languages, use controlled vocabulary, search appropriate regional databases, or follow citation networks when multilingual evidence is important.
Language restrictions can affect more than the number of studies
Cochrane recommends avoiding language restrictions in comprehensive systematic-review searches where possible because eligible studies may otherwise be missed and language or indexing bias can be introduced. It also recommends seeking translation assistance when needed.
The practical significance varies across evidence bases. A language restriction may have little effect in one review and matter substantially in another. That uncertainty is precisely why the possible effect of language restrictions on a review's conclusions should be considered rather than assumed.
Watch Out
Do not code “no English abstract” as a reason for exclusion unless your protocol explicitly and defensibly makes English-language accessibility an eligibility criterion. Otherwise, you are replacing substantive eligibility criteria with a convenience criterion after seeing the records.
04 · A Practical Example
How to handle a promising citation with no English abstract
Hypothetical Example
A potentially relevant local-language study
A researcher reviewing digital learning in secondary education finds a citation from a regional database. The article is published in another language. Its title has an English translation, but no English abstract is available.
Inspect the metadata The title clearly concerns digital learning and secondary-school students. The record also provides authors, journal, year, keywords, and a DOI.
Locate the original record The researcher follows the DOI to the journal website and finds an abstract in the publication language.
Translate for screening The abstract is translated sufficiently to determine the population, intervention, study design, and outcomes.
Retain the study The translated abstract suggests that the study meets the initial criteria, so the full article proceeds to eligibility assessment.
Obtain stronger language support Because the study remains eligible, the relevant full-text methods and results are assessed with suitable language assistance before appraisal and extraction.
The researchers did not need a polished translation of the entire paper merely to decide whether it deserved further attention. They increased the level of translation as the study moved closer to contributing evidence.
07 · A Quick Checklist
When a potentially relevant study has no English abstract
Before excluding the record, check:
Whether an English translated title or useful bibliographic metadata are available.
Whether the database provides controlled vocabulary, publication type, keywords, or other indexing clues.
Whether an abstract exists in the original publication language.
Whether the journal or publisher website contains richer metadata than the database record.
Whether the article can be found in another database, repository, or scholarly discovery system.
Whether references, citations, or related publications provide useful context about the study.
Whether a translation of the title or abstract is sufficient for the current screening decision before translating more of the article.
Whether stronger language expertise is needed before final eligibility, appraisal, or data extraction.
Whether any unresolved language limitations have been recorded and will be reported transparently.
