The research question of this review is as follows: How are families involved in the care of hospitalized older adults? Qualitative research provides in-depth understanding of human experience and multifaceted social phenomena that cannot be conveyed as richly with quantitative data alone [24]. Family involvement in the care of hospitalized older adults is this kind of complex social phenomenon. We chose a QES, a systematic review of qualitative evidence, for this study given our goal to be comprehensive in characterizing the existing literature on the complex phenomenon of family involvement in the care of hospitalized older adults [25]. We have submitted our review protocol on PROSPERO (ID 465617) and adhere to PRISMA-P (Preferred Items for Reporting Systematic Reviews and Meta-Analyses Protocols) guidelines in this protocol (Multimedia Appendix 1) [26,27]. The review results will be reported in the final paper using the ENTREQ (Enhancing Transparency in Reporting Synthesis of Qualitative Research) guidelines [24].

Key elements of eligibility for inclusion in the review were specified using the SPIDER (Sample, Phenomenon of Interest, Design, Evaluation, Research type) tool, a standardized approach to systematic review eligibility criteria for qualitative systematic reviews as an alternative to the PICOT (Population, Intervention, Comparison, Outcome, Time) tool used for quantitative systematic reviews [28]. We define older adults as adults 65 years of age or older or as defined by an individual study’s authors as studying “older adults.” We use the term family broadly to refer to any family member, friend, or neighbor who is involved in a hospitalization of an older adult motivated by a personal relationship rather than financial remuneration [29]. For the purposes of eligibility in our QES, we do not distinguish between “family members” and “family caregivers” or require that this person be involved in the care of the older adult before or after a hospitalization. We define involvement broadly as inclusion in care processes and include search strategy cognate concepts such as engagement, participation, teamwork, collaboration, and coproduction [30,31]. Only original research studies using primary qualitative data collection and analysis methods and published in peer-reviewed journals are eligible for inclusion. Reference lists of reviews or protocols identified in our search will be manually checked to identify completed original studies that may be eligible in our review. Full inclusion and exclusion criteria, based on our SPIDER question, appear in Table 1.

Our review uses the following databases: MEDLINE (via Ovid), Embase (via Elsevier), PsycINFO (via Ovid), and CINAHL Complete (via EBSCO).

The search strategies used a mix of database-specific subject headings and keywords searched in the title and abstract for the following concepts: family, informal caregivers, older adults, acute care, and the hospital setting. The search was developed and executed by an expert medical librarian, with input on keywords from other authors. The search strategy was peer-reviewed by another expert librarian using a modified PRESS (Peer Review of Electronic Search Strategies) checklist [32]. The search was executed in all 4 databases on August 23, 2023 and yielded 8862 citations in total. The full reproducible search strategies for all databases are included here as a Multimedia Appendix 2.

Data elements to extract will include characteristics such as journal, journal discipline, publication date, study location, sample size, research design, data collection techniques, data analysis techniques, study setting, funding sources, and theoretical or conceptual framework. This descriptive data will be entered into the data extraction tool. The results section of each included study will also be extracted. All qualitative results related to family involvement in care will be extracted; we do not prespecify or prioritize particular outcomes of interest.

Should the number of eligible studies exceed our ability to perform a high-quality synthesis, we will use purposive sampling to select a sample of studies for synthesis and analysis [35]. We expect to use purposive sampling if the number of eligible studies after full-text review exceeds 40. We will prioritize studies that (1) take place in the United States (the review authors’ country), (2) include older adult patients with cognitive impairment (given the greater need for family involvement for patients with impaired decision-making capacity) [36], and (3) are data rich [37]. Data richness will be determined using a previously published data richness scale to score studies on a data richness scale ranging from 1 to 5, with a score of 4 or higher being considered data rich [37].

Two independent reviewers will independently assess the methodological quality of the included studies using the QualSyst tool for qualitative studies [38], with the resolution of final quality assessments reached through discussion. Quality assessments will be entered into Excel along with data extraction. We will report the findings of our quality assessment for each study in a table in the final manuscript.

Descriptive data about the studies (eg, journal name, journal discipline, and publication date) will be presented in a table and summarized narratively. We will analyze and synthesize the results from included studies using qualitative methods. Specifically, the results will be uploaded from into qualitative data analysis software such as Dedoose (SocioCultural Research Consultants, LLC), which we will use to facilitate our synthesis [39]. We will choose a synthesis method appropriate to the available pool of evidence, guided by published guidance on doing so, with the expectation that we will likely use thematic synthesis [40-42]. Thematic synthesis includes coding the text of each study’s results to develop descriptive themes, which are then used to generate analytical themes [41]. We anticipate using both inductive and deductive methods to identify themes. Deductively, we will consider domains identified in prior models of family involvement in outpatient and long-term care settings and in decision-making including caregiving context; facility context; family tasks, actions, and roles (eg, information exchange, emotional support, and relationship rapport); and patient and family outcomes [15,18,43]. We will meet as a team regularly to discuss review progress and iterate on emerging themes and will present a summary of our findings in a Summary of Qualitative Findings table.

Once we have formulated a list of themes, we will use them to develop a conceptual model of family involvement in the care of hospitalized older adults to inform intervention development, measurement approaches, and policy needs. We will build on previously published conceptual models of family involvement in health care processes to refine our conceptual model for family involvement in care during hospitalizations of older adults [15,43]. As lead author, JBV will draft the initial model from themes identified in our synthesis of included papers. The study team will then meet iteratively to refine the model. Afterwards, we will obtain community input by eliciting feedback on our themes and conceptual model from family caregiver informants as a form of member checking [44,45]. We plan to perform this step through the Duke Clinical and Translational Science Institute Community Engaged Research Initiative and the Durham VA Veteran Research Engagement Panel to gather feedback and further refine themes and the conceptual model. We expect this step to take place in the spring or summer of 2024.

Our multidisciplinary review team includes members with expertise in hospital medicine, general internal medicine, geriatric medicine, nursing, information science, evidence synthesis, social work, and health services research. Throughout all stages of the review process and subsequent model development, we will each be cognizant of how our own backgrounds, views, and beliefs could influence our decisions regarding study eligibility decisions and the interpretation of the data. We will include discussions of potential bias in our regular team meetings. As a review team, we have a variety of backgrounds, which may help to overcome systematic bias in decision-making and analysis. Seven have clinical backgrounds (4 in medicine: JBV, BPG, KMG, SNH; 2 in nursing: MHG and TOO; 1 in clinical social work: MRS). Five are family caregiving researchers (JBV, BPG, TOO, CVH, and MHG) and 4 of us have formal training in qualitative methods (JBV, BPG, MHG, and TOO). One is a senior medical librarian (SC) and 1 is a health economist (CVH). Four have formal training in evidence synthesis (JBV, SC, KMG, and TOO). Six have been involved in the care of hospitalized older adults as family members (JBV, MHG, MRS, SNH, KMG, and CVH). We all work in academic settings. As lead author, JBV will keep an account of the review process and decisions made to document and reflect as the project progresses.

The importance of family involvement in health care delivery has been increasingly recognized [9,10,15,18,29], but family roles, tasks, and interactions with the professional clinical team in the hospital have received less attention than in the outpatient and community settings. Systematic review of family involvement in discharge planning for hospitalized patients supported policy implementation of the Care Act in the majority of US states requiring systematic family involvement in discharge planning [11]. More comprehensive evidence synthesis of how families are involved with other aspects of care for hospitalized older adults has the potential to similarly impact intervention development and policy change that can positively affect public health. The proposed work aims to complete this synthesis via a systematic review of existing qualitative literature regarding family involvement in the care of hospitalized older adults, followed by the development of a conceptual model of family involvement in the care of hospitalized older adults.

The strengths of the proposed study include its comprehensive approach and the focus on an incompletely understood phenomenon of interest. Systematic reviews of qualitative research are growing in use, but our QES method can still be considered an innovative strength [46]. The multidisciplinary team, including physicians, nurses, a librarian, a social worker, a health economist, and health services researchers with expertise in qualitative methods, evidence synthesis, and clinical care, is another strength of the proposed study. Limitations will likely stem from ambiguity in the phenomenon of interest, which may lead us to miss relevant studies that exist in the literature. This ambiguity surrounding the similar concepts of involvement, engagement, participation, teamwork, and other terms has been previously described [30,32]. We attempt to ameliorate this potential limitation by including as many cognate concepts in our search strategy as keywords based on prior work. Another limitation is intentional and that is the lack of inclusion of quantitative research. It is possible that prior experimental or observational studies that collected quantitative data only would give important information about the experiences of family involvement in the care of older adults that we will not be able to assess in a review of qualitative data alone. However, we maintain that a qualitative approach will identify richer data than that available with existing quantitative data. We anticipate that our included studies will identify roles that family members play. This will help identify narrower topics of interest within the larger phenomenon of family involvement, which will be better suited to a systematic review of quantitative studies in the future (eg, a systematic review of interventions designed to involve families for the purposes of delirium prevention for older adults).

This paper describes the protocol for a QES of family involvement in the care of hospitalized older adults. Themes identified through this review will be used to develop a conceptual model of the phenomenon of interest. We anticipate that the review and conceptual model will be used in the future to improve the health and well-being of hospitalized older adults and their families through optimizing family involvement.