Objective and review questions

The primary objective of the proposed review is to synthesize evidence on the role of sociodemographic characteristics on the degree of perceived acceptability of DMHI. The research questions listed below have been informed by the primary objective of the proposed scoping review as well as the population, concept, and context (PCC) framework proposed by the Joanna Briggs Institute (JBI) manual of Evidence Synthesis [19]:

1. What are the different sociodemographic factors that have been considered in the evaluation of perceived acceptability of DMHI?
2. Which sociodemographic factors have been positively, negatively, or not associated with the perceived acceptability of DMHI?
3. Which DMHI have been positively, negatively, or not associated with perceived acceptability among specific sociodemographic groups? (Please see the Study Selection and Extraction section below for more details on how the various DMHI will be categorized)

Eligibility criteria

The PCC framework proposed by the Joanna Briggs Institute (JBI) manual of Evidence Synthesis [19] will inform the inclusion and exclusion criteria (Table 1) of the proposed scoping review.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Inclusion and exclusion criteria.

https://doi.org/10.1371/journal.pone.0301886.t001

Sources of evidence

The information sources that will be used to identify materials pertinent to the proposed review have been selected to ensure the review is as extensive and comprehensive as reasonably possible on the topic of inquiry. Four electronic databases will be examined using distinct search strategies (i.e., PsycInfo through Ovid, MEDLINE through Ovid, CINAHL through EBSCO, and Embase through Ovid). Moreover, the reference list of secondary sources identified as relevant to the topic of inquiry will be examined.

Search strategy

Initially, A1 developed the concept plan for the search strategy in consultation with A2. A search strategy was then developed by A1 and later refined in collaboration with A2 and an experienced librarian at the University of Montreal. As a first step in the development of the search strategy, a preliminary search of MEDLINE and CINAHL was conducted. These databases were selected due to their extensive assortment of literature on health research. This was then followed by the selection of keywords identified in the abstracts and titles of retrieved papers along with indexed terms used to categorize the publications. The keywords and index terms were considered to inform the development of a complete search strategy for MEDLINE (S1 File).

The search strategy incorporates three topics of interest, which are acceptability, digital technology, and mental health. For acceptability, one indexed keyword heading (i.e., Patient acceptance of healthcare) (line 1) was searched along with three sets of search terms (line 2, 3 and 4) relating to acceptability (e.g., client acceptance, perceived utility). The indexed keyword and the first three sets of listed terms were then linked with the Boolean operator “OR” (line 5).

For the topic of digital technology, 10 indexed keyword headings (e.g., Internet-based interventions, mobile applications) were searched (lines 6,7,8, 9,10,11, and 12) along with one set of search terms incorporating many concepts related to digital technology (e.g., digital, bots) (line 13). This set of search terms was complemented by a list of keywords proposed by Lal et al. [23] in a systematic review pertaining to the priority afforded to technology in government-based mental health strategy documents. The indexed keyword headings relating to digital technology along with the related set of search terms were then linked with the Boolean operator “OR” (line 14).

For the topic of mental health, four indexed keyword headings were searched (e.g., mental health, mental disorders) (lines 15,16,17, and 18) along with five sets of search terms relating to mental health (e.g., low mood, mania, bereavement) (Line 19,20,21,22, and 23). These initial sets of search terms were bonified by the mental health disorders listed in the ICD-11 [25] and DSM-V [26] along with their associated symptoms. The indexed keyword headings relating to mental health along with the sets of related search terms were then linked with the Boolean operator “OR” (line 24). Lines 5, 14, and 24 were then combined using the Boolean operator “AND” to identify relevant citations that will be considered for inclusion in the proposed scoping review. The search was then further limited by year of publication (i.e., 2013 to current) and language (i.e., French and English). With the guidance of an experienced librarian, A1 will adapt and implement the MEDLINE search strategy to three other electronic databases (i.e., PsycINFO, CINAHL, and Embase).

Study selection and extraction

The inclusion and exclusion criteria outlined above have been established to guide the selection of studies that are concordant with the primary objective of the proposed scoping review. The citations obtained from the literature search will be imported into Covidence, a software used to assist in evidence synthesis. Covidence has the ability to extract citations, remove duplicates, and allows for a multiphase review of citations by up to two reviewers [27].

First, as proposed in the JBI Manual of Evidence Synthesis [19], the study selection strategy for titles and abstracts will be piloted on a random subset of citations to ensure that reviewers have an adequate understanding of the eligibility criteria and selection process. The screening process will be guided by the eligibility criteria of the proposed review. To begin, a subset of 30 citations will be selected from the pool of obtained studies. These articles will then be screened independently by two reviewers. Once completed, the reviewers will discuss any discrepancies in the perceived eligibility of studies to determine if the eligibility criteria require any further clarifications. This process will be repeated until the eligibility agreement between reviewers reaches 90%. Once achieved, the complete screening of citations by A1 will begin. The citations that do not contain sufficient information in the titles and abstracts to assess eligibility or those that are deemed eligible will then undergo full-text screening.

Once the screening of titles and abstracts is completed, a three-step process will be used to determine which studies will be included in the proposed scoping review. First, the full-text screening strategy will be piloted on 15 citations selected from the pool of obtained studies. These articles will then be screened independently by two reviewers. Once completed, the reviewers will discuss any discrepancies in the perceived eligibility of studies to determine if the eligibility criteria require any further clarifications. This process will be repeated until the eligibility agreement between reviewers reaches 90%. Once achieved, the complete screening of full-text citations by A1 will begin. A second reviewer will then validate all the studies deemed eligible by A1. If discrepancies arise in the process of eligibility assessment, A1 will meet with the second reviewer to reach a final consensus. If a consensus cannot be reached, a third reviewer will be consulted.

Once eligible articles have been selected, the process of data extraction and charting will begin. To facilitate the process of extracting data and ensure its validity, a legend will be developed. This legend will clearly identify and describe the information that should be included in each category of the charting tool. The categories and their corresponding data will be organized in an Excel sheet. As a first step, we will consider data categories listed in the JBI Manual of Evidence Synthesis [19]. These include basic descriptive information found within citations (i.e., title, Digital Object Identifier (DOI), author(s), and year of publication), as well as the country of origin, sample size, study design, study objective(s), and key outcomes. Key outcomes will be identified based on their degree of responsiveness to the research questions put forth by the proposed scoping review.

Furthermore, these categories will be further complemented by data relating to the PCC framework and research questions of the proposed review. Considering the population component of the PCC framework, we will extract data relating to the study sample, including their health status (e.g., diagnosed with a major depressive disorder) and reported sociodemographic characteristics (e.g., age, gender, race, sexual orientation). In relation to the concept component of the PCC framework, the conceptualization of acceptability (e.g., perceived ease of use) put forth in the articles along with the method(s) used to evaluate acceptability (e.g., self-report questionnaire) will be extracted. Lastly, considering the context component of the PCC framework, we will extract information on the type of DMHI evaluated (e.g., ICBT), the purpose of the innovation (e.g., assessment), its mental health target (e.g., anxiety), and the technology used (e.g., smartphone).

The data charting tool that will be used is subject to an Iterative process. As data are extracted, additional categories may emerge that warrant inclusion in the charting tool. As recommended in the JBI Manual of Evidence Synthesis [19], the data charting tool will be piloted by extracting data from 15 studies. This process will help ensure that the data charting tool is comprehensive and easy to use. Two independent reviewers will be involved in the pilot data extraction process. Any discrepancies that may arise during the data extraction or categorization process will be discussed by two reviewers to reach a final consensus. If consensus cannot be reached, a third reviewer will be consulted to examine and resolve any identified discrepancies. Once completed and sufficient concordance among extracted data has been achieved, A1 will begin the data extraction process for the complete list of eligible citations. The extracted data will then be validated by a second reviewer.

Data synthesis and presentation

First, a table will be used to represent the various DMHI that have had their acceptability evaluated through a consideration of sociodemographic characteristics. This table will include data on the various types of DMHI (e.g., ICBT), their mental health targets (e.g., anxiety), their purpose (e.g., assessment), and the technology used (e.g., smartphone). Secondly, a table will be presented to illustrate the different sociodemographic characteristics that have been considered in the evaluation of acceptability of DMHI. The extracted data will be summarized and presented using descriptive statistics (e.g., frequency and percentages).

Finally, key findings on the role of sociodemographic characteristics on the perceived acceptability of DMHI will be presented in tables, including information on whether the association is positive (i.e., high perceived acceptability), neutral (i.e., no association) or negative (i.e., low perceived acceptability). These findings will be categorized based on the technology examined (e.g., smartphone app), sociodemographic factor(s) of interest (e.g., age), and whether they are relative (e.g., men demonstrated a higher level of perceived acceptability than women for this technology) or absolute (e.g., men demonstrate a high level of perceived acceptability for this technology).

For absolute acceptability, the cut-off scores reported in the literature for the various measures used to measure acceptability across studies will be used to indicate whether the reported scores represent high or low acceptability. If such scores are not available, the acceptability scores that have been evaluated using the same measures across studies will be contrasted. For studies that utilized ad-hoc measures to examine the perceived acceptability of users, the author’s interpretation of these scores will be reported in the proposed review. The approach used to interpret the various scores will be finalized depending on how acceptability is being measured across the included studies.

Furthermore, as the data extraction strategy is piloted, additional findings may be reported based on their relevance to the inquiries put forth in the proposed scoping review. Moreover, throughout the screening and extraction processes, notable patterns may emerge across studies, which may not have been specified in this protocol. These may include methodological limitations (e.g., many studies did not report on the socio-demographics of study dropouts), the overrepresentation of specific digital mental health innovations (e.g., virtual reality), or their mental health targets (e.g., depressive symptoms).

Moreover, depending on the volume of findings uncovered (i.e., over 100 studies included), it may be necessary to use additional forms of visual representation, such as graphs and Venn diagrams. These may be used to provide readers with a complementary depiction of findings to facilitate greater comprehension.