The selection of the information sources was based on the relation of scientific topics to the systematic review research questions and information retrieved from analysis regarding the optimal database combination for biomedical systematic reviews [12]. The review team split the review sources into three broad categories—scientific databases containing peer-reviewed publications, databases with gray literature, and scientific organization websites.

The inclusion criteria are set for the types of studies, types of participants, and extracted data items.

The exclusion criteria will be: studies not identifying the transmission of the fungal and parasitic diseases via SoHO donations; animal studies; in vitro studies; reviews, systematic reviews, and meta-analyses; letters to editors and opinion papers; theses and dissertations; and gray literature besides the published peer-reviewed conference proceedings relevant to the systematic reviews’ research questions.

Two members of the review team will select the eligible publications independently. A referee investigator will be asked to decide in case of a disagreement between the two reviewers. The Cohen kappa test will be used to measure the interrater agreement in selecting the eligible publications [15]. A study’s eligibility will be decided by screening the titles and abstracts using the EndNote software files, where the retrieved publications will be saved. The selection review team will ensure the identification and exclusion of retracted publications. The team will also ensure locating the inaccessible eligible publications’ full texts through emails to the lead authors and publication journals. The eligible publications without full texts will be listed in the systematic review with reasons. Finally, for transparency reasons, a complete list of the excluded publications will also be included in the systematic review.

An individual data extraction form (Cochrane Library model) [7] will be incorporated for each eligible publication. Two reviewers will extract the data independently from the eligible publications. A referee investigator will make the ultimate decision in case of a disagreement between the reviewers. A priori pilot data extraction will be used to include any missing data that were not initially considered or extracted. The data will be collected as two tables in Excel format, one each for the fungal and the parasitic disease. The data extraction–review team members will contact the corresponding authors via email if the outcome data in the full-text articles are unclear.

The priority of the outcomes of this systematic review are as follows: (1) intervention type (ie, number of donations, blood or blood components transfusion, type of transplantation); (2) the association of the presence of fungal and parasitic infections in SoHO donors, donations, and their transmission through SoHO donations with any physiological and demographic characteristics of the populations, including aspects such as environmental conditions; (3) the prevalence of the presence of fungal and parasitic infections in SoHO donors, donations, and transmission via SoHO donations; (4) odds ratios, risk ratios, risk differences, and hazard ratios of fungal and parasitic infections or contaminations in SoHO donors, donations, and their transmission via SoHO donations; (5) cause of the infection in donors or the contamination of donation; (6) the type of screening test of infection and the donor types; (7) hospitalizations and deaths; (8) region (country and continent); and (9) diagnostic method.

The review team will use the methodological design of each eligible publication to determine the risk of bias assessment. Three appropriate tools are going to be used: the Cochrane Library tool [16] for the risk of bias assessment in randomized controlled trials (RCT); the ROBINS-I tool (Risk of Bias in Nonrandomized Studies–of Interventions) [17] for the risk of bias assessment in non-RCT studies using an intervention, and the RTI item bank tool [18] for the risk of bias assessment in observational studies. Two reviewers will assess the risk of bias in the eligible publications independently, and a referee investigator will decide in case of a disagreement. The Cohen kappa test will be used to measure the interrater agreement in the evaluation results [15]. Finally, the risk of bias assessment results will be extracted in relevant tables and figures according to the Cochrane Library format [7].

The eligible studies with data not suitable for meta-analysis will be summarized into a qualitative description. In the case of the eligible studies with pertinent data for meta-analysis, a random or fixed-effect meta-analysis model will be used to account for heterogeneity due to differences in study populations, types of infections, interventions, study durations, and other factors. All meta-analyses will be conducted using the RevMan 5.3 software (Nordic Cochrane Centre, The Cochrane Collaboration) [19].

The prevalence for meta-analysis will be calculated using the following formula [7]

Standard errors for the meta-analysis will be calculated using this formula [7]

Standard errors will then be used for weighted proportions, and the RevMan 5.3 software [19] will be used to generate the forest and funnel plots. The funnel plots will only be generated for those meta-analyses that include more than 10 studies [7]. The odds ratios, risk ratios, and risk differences for meta-analyses will be calculated using a dichotomous, inverse variance method, referring to infection incidences of the individuals with an intervention (ie, transfusion, transplantation) against infection incidences of individuals without any interventions. The weighted proportions of such meta-analyses will be calculated based on each study’s sample size. The time-to-event outcomes (hazard ratio) for a meta-analysis will be calculated using the generic inverse variance, or an O and E variance fixed effect model [7].

The 95% CI and heterogeneity between studies will be evaluated using the I² statistic. The results for heterogeneity will be considered statistically significant at P<.10, while the I² index interpretations will be made based on earlier guidelines [7]. Small study effects, potentially caused by publication bias, will be assessed using the funnel plots [7]. The risk of bias assessments will be incorporated in data synthesis and used for the forest plots [7]. Finally, the standardized mean difference (SMD) (ie, the difference in mean outcomes between groups/standard deviation of outcomes among participants) will be used for the meta-analysis studies that assess the same outcome using different measurement scales [7].

The reporting of the eligible publications will be checked independently by two review team members, while a referee will make the final decision in case of a discrepancy between them. The interrater agreement in the evaluation results will be tested using the Cohen kappa test [15]. The 25-item checklist of the CONSORT (Consolidated Standards of Reporting Trials) [20] will be adopted for the eligible RCTs. The 22-item STROBE checklist (Strengthening the Reporting of Observational studies in Epidemiology) [21] will be employed for the eligible observational studies and published conference proceedings. A score will be calculated for each eligible study following a previous methodology [22]. The reporting scores of the eligible studies will not be used for assessing the methodological quality of the eligible studies; however, they will be used for measuring the reporting quality as a risk factor for the critical appraisal of eligible papers due to missing information [23].

Two reviewers will independently appraise the implications and applicability of the findings of the systematic reviews using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) analysis [24]. The GRADE analysis rates the quality of the best available evidence, which can be used for developing healthcare recommendations and guidelines. The four-level GRADE ratings classify the quality of evidence as high, moderate, low, and very low. The low-quality evidence characteristics include an observational study, risk of bias, inconsistent results, indirectness, imprecision, and publication bias. The high-quality evidence characteristics include an RCT study, a large magnitude of effect, demonstration of a dose–response relationship, and if residual confounding factors that are plausibly expected to reduce or increase the demonstrated effect do not actually reduce or increase it. A critical appraisal of the current systematic review process will be performed using the AMSTAR (A Measurement Tool to Assess Systematic Reviews) checklist [25], a critical appraisal tool for assessing the quality of systematic reviews.

The two systematic reviews aim to identify the scientific evidence on transmission risks of fungal and parasitic diseases via SoHO globally. The retrieved evidence would assist an evidence-based risk assessment of the fungal and parasitic transmission through SoHO donations and evaluating the available prevention strategies within the European Union. The study results will be used to create an evidence pool containing the geographical data for transmission risks, genetic, physiological, and demographic characteristics of the infected populations, the infection cause and the type of screening tests for donors, and diagnostic methods of the infections. It is also possible that a meta-analysis of these data will be incorporated, which will further strengthen the evidence-based risk assessment approach.

The systematic review process has many strengths. The protocol followed the PRISMA-P guidelines. The searching procedure used robust algorithms with standardized indexing terms to retrieve records that had different words to describe the same concept and information beyond the words in the title and abstract [7]. We will use well-established tools [16,18] to evaluate the included studies. Additionally, to reduce bias, two investigators will work independently on screening the data for eligibility, risk of bias assessment, data extraction, and the CONSORT and STROBE scores. The search procedure of the systematic reviews had no restrictions regarding the date of publication, study design, and language. The GRADE analysis will allow an excellent evaluation of the quality of the outcomes.

The systematic review process has limitations too. We excluded gray literature, incorporating a publication bias. Nevertheless, the inclusion of gray literature may itself introduce bias, and one reason to include the gray literature is the absence of peer-review sources [7]. Another limitation that we foresee is that we have accepted the peer-reviewed in vitro studies as eligible, even though it is scarce to identify such studies. We have included these studies to increase the transparency and validity of our systematic review process.

These systematic reviews will form the basis for developing a risk assessment of fungal and parasitic disease transmission via SoHO.