JMIR Publications

ABSTRACT

Background:

Conversational agents, defined for the purpose of this work as computer programs designed to simulate two-way human conversation using language, potentially supplemented with non-language modalities, offer promising avenues for health interventions in different populations across the life course. There is a lack of open access and user-friendly resources to illuminate research trends and gaps and to pinpoint expertise across international centers.

Objective:

Our aim is to provide an overview of all relevant evidence on conversational agents in health and well-being across the life course. Specifically, our objectives are to identify, categorize and synthesize, through visual formats and a searchable database, primary studies and reviews in this research field.

Methods:

An evidence map was selected as the type of literature review that optimally responded to our aim. We systematically searched eight databases (MEDLINE, CINAHL, Web of science, Scopus, Cochrane, ACM, IEEE, JBI and Google Scholar), and will perform backward citation searching of all studies included. First-stage of a double-stage screening, based on abstract and title only, was conducted using predetermined eligibility criteria for primary studies and reviews; an operational screening procedure was developed to streamline consistent procedures across the team. Double data extraction will be performed onto previously piloted data collection forms. We will appraise systematic reviews using AMSTAR 2. Primary studies and reviews will be treated separately in the analysis. Data will be synthesized through descriptive statistics, bivariate statistics and subgroup analysis if appropriate, and through high-level maps, such as scatter and bubble charts. Development of the searchable database will be informed by the research questions and data extraction fields.

Results:

As per April 2021 the literature search in eight databases was concluded, yielding a total of 16351 records. First-stage screening, based on abstract and title only, led to the selection of 1282 records for primary studies and 151 records for reviews. These will be subjected to second-stage screening. A glossary with operational definitions to support the study selection and the data extraction stages was drafted. The anticipated completion date is October 2021.

Conclusions:

Our wider definition of conversational agent and the broad scope of this evidence map will explicate trends and gaps in this field of research. Additionally, it will contribute to avoid fragmented research efforts and wasteful redundancies, particularly through the development of a searchable database of studies. Finally, as part of the eCCo project (https://ecco.esel.pt), our work will also inform the development of an international taxonomy on conversational agents in health and well-being, contributing to standardize terminology and categorization.