This is a crucial decision for your systematic review, and one that many researchers grapple with. The ideal of a systematic review is to be as comprehensive and unbiased as possible, and restricting by language introduces a potential source of bias (language bias). However, practical constraints are very real.

Here's a structured approach to help you decide whether to restrict to English or include other languages (specifically Chinese):

The Ideal vs. Reality

• Ideal: To minimize bias and achieve true comprehensiveness, you would include all relevant studies regardless of language.


• Reality: Including multiple languages, especially complex ones like Chinese, significantly increases the time, resources, and expertise required.

Key Considerations to Guide Your Decision

You need to weigh the potential benefits of including Chinese literature against the significant practical challenges.

1. The Nature of Your Research Question (PICO)

• Is the phenomenon/intervention/population predominantly studied in China or has a strong Chinese cultural context?


• Example: If you're reviewing traditional Chinese medicine (TCM) interventions, studies on specific Chinese diseases, or public health interventions uniquely implemented in China, then a significant body of relevant literature will be in Chinese. Restricting to English would likely miss the most important evidence.


• Example: If you're reviewing a universal surgical technique or a globally developed drug, then while some studies might be in Chinese, the primary evidence base might still be global and well-represented in English literature.


• Is there a known history of research in your field being published primarily in non-English journals, particularly Chinese ones?


• For many fields, especially clinical trials and certain public health areas, there's a strong publishing culture in China within Chinese-language journals.

2. Preliminary Evidence of Language Bias (The "Pilot Search")

This is perhaps the most important step. Don't guess; investigate.

• Conduct a pilot search in English databases: Use your preliminary search terms in PubMed, Embase, Web of Science, etc. Get a sense of the volume and relevance.


• Conduct a parallel pilot search in key Chinese databases:


• CNKI (China National Knowledge Infrastructure): Most comprehensive for academic journals, theses, conference proceedings.


• Wanfang Data: Another major database, strong in medicine and technology.


• SinoMed (China BioMedical Literature Database): Specifically for biomedical literature.


• CBM (Chinese Biomedical Literature Database): Similar to SinoMed.


• Compare the results:


• Volume: How many hits do you get in Chinese databases compared to English?


• Relevance: How many seem highly relevant based on titles/abstracts (even if machine-translated for a first glance)?


• Overlap: Is there significant overlap (i.e., are the same studies published in both English and Chinese journals)? This is less common than you might think. Many Chinese studies are only published in Chinese.


• Quality/Impact (initial assessment): Are there highly cited or seminal works in Chinese that you would miss?


• Look for existing systematic reviews on your topic: Did they include Chinese studies? What did they find? Did they report a significant impact of including them?

3. Available Resources (Time, Funding, Team Expertise)

This is the practical bottleneck.

• Language Proficiency: Do you have team members who are native or highly proficient in Chinese and English? This is critical for:


• Developing accurate Chinese search strategies.


• Screening titles and abstracts.


• Retrieving full texts.


• Extracting data accurately.


• Critically appraising the quality of studies.


• Resolving discrepancies.


• Note: Simply having someone who can translate is not enough. They need to understand the nuances of research methodology, terminology in your field, and systematic review processes.


• Time: Including Chinese literature will add significant time to every stage of your review. Translation alone is time-consuming.


• Funding: If you don't have in-house expertise, hiring professional translators or research assistants fluent in Chinese can be expensive.


• Software/Tools: Some review software has limited support for non-Latin characters, though most modern ones are better.

4. Potential Impact of Language Restriction

• Bias: Restricting to English can introduce language bias, potentially leading to an overestimation or underestimation of intervention effects if studies published in Chinese journals have systematically different findings (e.g., "positive" results more likely to be published in English, "negative" results in local journals, or vice-versa).


• Completeness: You risk missing important studies, potentially leading to a less robust or even misleading conclusion.


• Generalizability: If your question is globally relevant, but you exclude a major region's research output, the generalizability of your findings might be compromised.

Making the Decision:

Based on the above considerations, you can categorize your situation:

Option A: Strongly Consider Including Chinese Literature

• If: Your research question has a strong Chinese context, preliminary searches indicate a large volume of relevant and unique studies in Chinese, AND you have sufficient resources (especially a proficient Chinese-speaking team member for all review stages).


• Justification: This choice significantly enhances the comprehensiveness and reduces language bias for your specific topic.

• If: Your preliminary searches suggest that while some Chinese literature exists, it's either not overwhelming, not central to the primary evidence, AND/OR you lack the necessary linguistic resources (time, funding, expertise) to adequately incorporate it.


• Justification: You must clearly state your language restriction in your protocol and final paper. You need to justify why you made this decision (e.g., "due to resource constraints," "pilot searches did not indicate a substantial, unique body of evidence that would significantly alter findings," or "the scope of the review was limited to globally accessible literature").


• Crucial: You must also acknowledge this limitation in your discussion section and discuss its potential impact on your findings and generalizability. For example, "Findings of this review are limited to English language literature and may not fully capture the global evidence base, particularly from China, where relevant research may be published in Chinese."

• If: You want to be thorough but are severely resource-constrained.


• Approach:


• Perform comprehensive searches in both English and Chinese databases.


• Screen titles and abstracts in both languages (using automated translation tools for an initial pass if no native speaker is available for this stage, but be very cautious).


• If a non-English abstract seems highly relevant, try to find an English version or collaborate with a Chinese speaker for full-text assessment.


• Ultimately, if full text cannot be adequately translated and extracted, you might have to exclude it at the full-text review stage.


• Justification & Limitation: You would report your full search strategy, note the number of non-English studies identified, and explicitly state that only studies for which full-text translation and data extraction were feasible were included, acknowledging the potential for bias.

Recommendation

Given your statement that "most relevant literature may be in Chinese," my strong recommendation is to prioritize the pilot search (consideration #2) and assessment of your resources (consideration #3).

• Conduct the pilot searches in both English and Chinese databases first. This will give you concrete data on the volume and likely relevance of Chinese literature.


• Honestly assess your team's linguistic capabilities and your overall time/funding budget.


• Then, make an informed and well-justified decision.