The planned systematic review will be conducted according to the current Cochrane guidelines for qualitative evidence synthesis [21] and will be reported in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [22]. The relevant PRISMA 2020 checklist [22] can be found in Multimedia Appendix 1. The review team includes members trained in systematic review methodology. The protocol has been registered in the Open Science Foundation registry. In case any amendments to this protocol are made during the review process, changes and related reasons will be reported in the final article.

We employed the feasible, interesting, novel, ethical, and relevant (FINER) approach [23] to test the applicability of the research question expressed via the PerSPecTIF statement. The outcome of the FINER approach is presented below:

Studies investigating opening-up strategies of innovation ecosystems for increasing the participation of more diverse innovation stakeholders will be included. We will consider observational studies that provide original data, independently of the design and methodology adopted.

We will exclude narrative reviews, systematic reviews, perspectives, opinion articles, and other publications that do not include original data.

The selected scientific databases will be searched from January 1, 2000, to present. All searches will be updated prior to submission of the final manuscript, in case the date of the initial search is more than 12 months older than the date of submission [24,25]. New records will be screened and evaluated based on the inclusion and exclusion criteria set above.

We anticipate a large body of the literature on the topic of this systematic review to be in English. No articles will be excluded based on language, but our search will be conducted using English terms. Articles in languages other than the ones spoken by the coauthors of this study (English, Greek, and German) will be translated into English using Google Translate following previous methodologies [26]. If Google Translate does not generate a good translation or if we are in doubt about the translation generated for a non-English manuscript, the paper will be assessed by a native speaker.

We will perform a keyword algorithmic search in scientific databases that focus on research topics relevant to our research question. These databases are as follows: Web of Science [27], Cochrane Library [28], Scopus [29], and Social Science Research Network [30].

The search algorithm will be built according to the PerSPecTIF statement [20] described in Table 1 and will be adjusted to the environment of each database. An example of translating the PerSPecTIF statement to appropriate indexing terms based on existing guidelines [21] is shown in Table S1 (Multimedia Appendix 2). A preliminary search on the Web of Science database using these search terms revealed a total of 1301 records. The algorithm is also provided in Multimedia Appendix 2.

In the algorithmic searching we will use Booleans (ie, OR, AND, NEXT, etc) and possibly truncations as well as wildcard characters (*, ?). Finally, we will also check the reference lists of the included studies for potential eligible publications that did not appear in the initial search approach. To comply with the Multi Engine Command Instrument Rating (MECIR) requirement and Conduct of Systematic Reviews in Toxicology and Environmental Health Research (COSTER) recommendations regarding the updating of the searches within 12 months prior to publication of a systematic search [24,25], we will rerun all searches shortly before the final analyses, and any further relevant studies identified will be retrieved for inclusion.

The above-mentioned pilot algorithm was judged by the review team as applicable and appropriate to be used in the official searching procedure. As such, 2 members of the review team, GN and EM, independently conducted the official searching in the selected scientific databases (from the date of their inception to October 2021). The search algorithm was “translated” from one database to another so that it is compatible with the corresponding website’s search engine. ADF confirmed no disagreement between the team members GN and EM in applying the algorithms in all 3 databases.

The retrieved studies from the searching procedure will be managed using the Rayyan online platform [31]. Two reviewers (GN and ADF) with previous experience in conducting systematic reviews will independently screen the retrieved publications for eligibility. A third reviewer (EM) will act as a referee and resolve any potential conflicts. Then, all reviewers will share their notes to confirm and finalize the selection of studies. All 3 reviewers will be provided with written instructions on the selection process and will undergo pilot testing of the systematic review selection procedures on a small subset of the records retrieved. The methodology of the selection process is as follows:

A PRISMA flow diagram will be created to describe in detail the procedures of searching and selection of publications in this systematic review [22,32]. A full list of the excluded articles will be provided in the final systematic review paper.

For each eligible publication, an individual data extraction form will be incorporated. Two review team members will independently extract data from the eligible publications and a referee- investigator will make an ultimate decision in case of a disagreement between review team members. A priori pilot data extraction will be used so that the 2 review team members agree on including missing data that were not initially considered or data that do not need to be extracted. Each team member will extract their data in an Excel spreadsheet (Microsoft Corp). In cases where the presented data are unclear in the articles’ full text, the data extraction review team members will contact the corresponding authors via email to retrieve them.

According to the existing guidelines [21], we will assess methodological strengths and limitations as a marker of study rigour using the Critical Appraisals Skills Programme (CASP) tool [33]. CASP is a previously published, commonly used, and validated tool to assess methodological strengths and limitations of qualitative studies [21]. Also, it has shown an interrater agreement of >92% [34]. According to the existing guidelines [21], the aim of this assessment will not be to calculate total quality scores but to provide evidence for a discussion on the studies, their “risk to rigour,” as well as whether their methodological limitations may have affected the systematic review findings.

We will use the thematic synthesis method to produce syntheses that can subsequently be integrated with an intervention review or analysis.

Based on the existing guidelines [21], we will use the GRADE-CERQual tool to assess confidence in the qualitative synthesized findings. This tool evaluates 4 components (relevance, methodological limitations, adequacy, and coherence) to provide an overall assessment of confidence in the synthesized qualitative findings.