This review will include randomized controlled trials, cluster randomized controlled trials, and nonrandomized studies (NRS). NRS have been defined as quantitative studies estimating the effectiveness of an intervention (harm or benefit) that does not use randomization to allocate units to comparison groups [47]. We will include prospective cohort studies, case-control studies, controlled before-and-after studies, interrupted-time-series studies, and controlled trials with inappropriate randomization (quasiexperimental studies) [48-50]. We will consider publications in English, French, German, and Portuguese.

This review will consider studies involving registered HCPs, including those with bachelor’s, master’s, or doctoral degrees in physiotherapy (PTs) and nursing (registered nurses [RNs], NPs) and who are delivering primary health care, including nursing and physiotherapy students. Physical therapists and PTs will be considered synonymous.

We will include all types of primary health care settings such as private practices, community and health maintenance organization practices, community and private primary health care settings, hospital outpatient departments, practices in hospital settings, and hybrid primary health care practices including community and private practices, health maintenance organizations, and outpatient departments.

We will examine all types of educational interventions aimed at improving the EBP delivered by RNs, NPs, and PTs to adults living at home as part of active primary health care.

Based on the Cochrane Effective Practice and Organization of Care taxonomy of interventions [51], we will consider educational interventions targeting health care organizations and health care professionals (Textbox 1). We will exclude interventions targeting the regulatory, economic, or financial aspects of EBP.

The review’s primary outcome measures will be increased or decreased beliefs, knowledge, implementation, and integration of EBP among RNs, NPs, and PTs active in primary health care settings (measured using methods [52,53] such as questionnaires, interviews, chart analysis, and self-reporting by RNs, NPs and PTs [53]), with a focus on dichotomous (yes/no), ordinal or continuous beliefs, and implementation or integration rates or scores.

The review’s secondary outcome measures will be the production of systematic reviews; numbers of journal clubs organized; numbers of grand rounds organized; development of EBP guidelines or practice guidelines for care or case management; and the implementation of EBP programs, mentor coaching, or tutorial programs.

In collaboration with the medical librarians (MS and PM) and using predefined search terms, we will conduct a systematic literature search for published articles in the following electronic databases, from inception until October 31, 2020: Medline Ovid SP (from 1946), PubMed (NOT Medline[sb]; from 1996), Embase.com (from 1947), CINAHL Ebesco (from 1937), the Cochrane Central Register of Controlled Trials Wiley (from 1992), PsycINFO Ovid SP (from 1806), Web of Science Core collection (from 1900), PEDro (from 1999), the JBI Database of Systematic Reviews and Implementation Reports (from 1998), and the Trip Database (from 1997). We will also conduct a hand search of the bibliographies of all the relevant articles and a search for unpublished studies using Google Scholar, ProQuest Dissertations and Theses dissemination, Mednar, and WorldCat. The search will be completed by exploring the grey literature in OpenGrey and the Grey Literature Report from inception until October 31, 2020.

The search syntax of the included databases will serve as the basis for all search strategies, using descriptors (EMTREE and Medical Subject Headings [MeSH]) and text terms with Boolean operators “AND” and “OR.” The syntax consists of 4 search themes intersected by the Boolean terms “AND” and “OR.” The descriptor terms included in the health occupations of RNs, NPs, and the allied health occupations of PTs are described in Textbox 2, and descriptor terms and keywords included in the search strategy for educational interventions on EBP are described in Textbox 3.

In addition to searching electronic databases, we will conduct a hand search of the bibliographies of all relevant articles and search for unpublished studies. We will consider publications in English, French, German, and Portuguese. Multimedia Appendix 1 presents the syntax used in all selected databases.

Two pairs of reviewers (HV and PM, RH and MS) will independently screen the titles and abstracts identified in searches in order to assess which studies meet the inclusion criteria. Disagreements will be resolved through discussion, or, if needed, a consensus will be reached after discussion with the co-authors (AGM and FP).

Two pairs of reviewers (HV and PM, RH and MS) will independently assess the full-text articles to ensure that they meet the inclusion criteria. Disagreements will be discussed and resolved with the co-authors (AGM and FP). A flowchart of the trial selection process has been drawn in accordance with the PRISMA-P statement [44] (Multimedia Appendix 2).

Data extraction will be conducted independently by 2 pairs of authors (HV, RH, FP) using a specially designed, standardized data extraction form (Multimedia Appendix 3). Discrepancies will be resolved through discussion and consultation with the co-authors (FP, RH, FP).

The following information will be extracted from each included study: (1) study authors, year of publication, and country where the study was conducted; (2) study characteristics (including setting and design, duration of follow-up, and sample size); (3) participants’ characteristics (eg, profession, employment [% vs hours/week], employer, sex, age); (4) characteristics of interventions (eg, description and frequency of educational interventions, health care professionals involved); (5) characteristics of usual care group; and (6) types of outcome measures (Multimedia Appendix 3).

Two reviewers (HV and RH) will independently assess the risks of bias in all the randomized and nonrandomized studies of interventions (NRSIs) included. Disagreements will be resolved through discussion and consultation with the co-authors (HV, RH, FP).

We will use the validated Cochrane Risk of Bias Tool, version 2.0 [54], to assess the risk of bias in randomized trials and nonrandomized studies. This is based on 5 domains: (1) bias arising from the randomization process, (2) bias due to deviations from intended interventions, (3) bias due to missing outcome data, (4) bias in the measurement of the outcome, and (5) bias in the selection of the reported result. Each of these 5 domains will be rated as one of the following: (1) low risk of bias, (2) some concerns, or (3) high risk of bias. Declaring that a study has a particular level of risk of bias in any individual domain will mean that the study as a whole has a risk of bias.

We will use the validated Robins-I tool for assessing the risk of bias in NRSIs [55]. This tool covers 2 dimensions and 7 domains through which bias might be introduced into an NRSI: (1) pre-intervention and at intervention (bias due to confounding, bias in the selection of study participants, and bias in the classification of the intervention) and (2) post-intervention (bias due to deviations from intended interventions, bias due to missing data, bias in the measurement of outcomes, and bias in selection of the reported result) [55]. Any disagreements in quality assessments will be resolved through discussion.