In this study, operational definitions of terms [9] were assigned to the following:

In order to comprehensively and systematically clarify the attributes of family spirituality and the influencing factors that cause deterioration of family spirituality, the study conducted data triangulation from literature reviews and family interviews [11] [12] , and directed content analysis [13] based on the theoretical framework of the six systems of CSFET.

Using the literature database PubMed, a search was performed with “family spiritual*” as the key word. In order to identify the most recent papers, a search for literature between 2010 and 2023 (conducted in December 2023) yielded a total of 4,826 hits. The inclusion criteria were that the paper must be an original article; that it must be a study targeting a family or multiple family members; and that it must describe family spirituality. The exclusion criteria were that the paper is about personal spirituality and is written in a language other than English. After reading the titles and abstracts of the papers, 165 potentially relevant papers were obtained, and the papers were carefully read and confirmed. The 20 papers that met the inclusion and exclusion criteria were subjected to analysis [14] - [33] . The literature search was conducted in March 2020, October 2021, and October 2022, and was used for analysis at each stage. The literature searched in December 2023 was finally included in the analysis.

The target families for the semi-structured interviews were families with an adult member aged 65 years or older. This is because we assumed that families with older adults have experienced a variety of events over their years of family life, with numerous opportunities to consider the meaning of their family’s existence, which represents a key aspect of family spirituality. We explained this study to the manager of a housing complex and the manager of a visiting nurse station in an urban area of Japan, and recruited the target families. We contacted the families who agreed to participate in the study and set the date, time, place, and method of the interview. In principle, the family interview was conducted with multiple family members, but individual interviews with only one key informant were also possible upon their request. In both cases, the interviews were concerned with family spirituality, not individual spirituality, as based on the operational definitions of family spirituality in this study. The interviews were conducted from May 2020 to January 2022, and each family was interviewed for about 120 minutes to collect sufficient family data.

Semi-structured interviews were conducted using an interview guide based on the Family Environment Assessment Index (FEAI) [34] . The FEAI was created as a family assessment tool based on the CSFET, and is a collection of sample questions in family interviews/meetings, to enable accurate collection of family information necessary for family assessment. The FEAI includes family spirituality. In this study, a pilot test of the interviews was conducted with four families as a convenience sample in advance. After that, based on the opinions of the families, the question wording was clarified and the interview guide was revised to suit the purpose of this study. An example of the completed interview guide is shown in Table 1 .

The interviews were conducted by two researchers, one of whom was mainly responsible for the interviews and the other for writing the field notes. With the consent of the subjects, all interviews were recorded using an IC recorder, and the recorded contents were transcribed verbatim after each interview while referring to the field notes, and on each occasion analysis was performed.

In order to ensure the complementarity of the analysis, data sources consisted of the descriptions in the papers and the verbatim transcripts of the interviews. The data sources were analyzed using the following procedure while creating an audit trial of data analysis processes based on the data triangulation protocol [11] [12] .

Three researchers independently read the data sources repeatedly to fully understand the contents related to the meaning of family existence. After that, sentences, paragraphs, or sections related to the attributes of family spirituality and the influencing factors of the decline of family spirituality were identified and written in the code column of the codebook. Codes are small, meaningful units. The influencing factors of the decline of family spirituality were classified into risk/causal/promoting factors, preventive/inhibitory/suppression factors, and context-sensitive factors. The classification was based on the similarities and differences of the contents of the codes, and the codes with common contents were classified into subcategories and categories to increase the level of abstraction [35] . Then, using directed content analysis [13] , each category was classified into the six CSFET systems and written into the code, subcategory, category, and system columns of the codebook.

A primary coding framework and categories were constructed to compare the codes, subcategories, categories, and systems from the data sources and eliminate discrepancies. The three researchers confirmed the analysis by re-checking the codes, subcategories, and categories to confirm that they were converging on the range of attributes of family spirituality and influencing factors of declining family spirituality.

The codebook was constantly revisited after initial coding, until categories were saturated and no new categories emerged. Saturation was determined to have been reached when interviews had been conducted with 12 families. After that, an interview was conducted with one more family, and saturation was verified, so the codes, subcategories, categories, and systems were finalized through the consensus of all three researchers.

To ensure the trustworthiness of the analysis, the following was conducted: One, as an investigator triangulation [36] , in order to develop a comprehensive understanding of phenomena, two researchers skilled in family interviews/meetings were used to conduct the interviews. In addition, three family nursing researchers were constantly involved in performing integrative analysis of the data to minimize the risk of researcher bias. Two, the analyses were performed by triangulating data sources from the literature and family interviews [37] . Three, the analyses were first performed by three researchers independently. Following multiple meetings with all, any disagreements were resolved through discussions, and, when necessary, another family nursing researcher reviewed the disagreements until a consensus was reached. Four, we recorded an audit trail [38] of data analysis processes and analytic decisions, and were able to assess at any time whether the findings were grounded in the data. Five, the research findings and interpretations were shared and discussed in regular meetings attended by eight family nursing researchers, and the analysis was carried out repeatedly based on peer debriefing from different perspectives [39] . And six, the results of the analyses were returned to the interviewed families for verification through family member checking [40] and these results were reflected in the analysis.

This study was conducted following review of and approval by the institutional review board of the university to which the researcher belongs (approval number: No. 938). The interview participants were given written and verbal explanations of the study’s purpose, methods, benefits and risks; confidentiality obligations; freedom to withdraw midway; and that no inconveniences would be incurred for withdrawing midway or for non-participation, and written consent was obtained. In addition, care was taken to minimize the burden on the participants by informing them that they were not obliged to respond to questions they preferred not to answer, and by allowing them to take breaks as necessary.

Full adherence was given to research ethics when handling papers. Cited papers and quoted sections were clearly identified to avoid copyright infringement. When using papers, strict adherence was given to copyright law and sources clearly indicated. We also strived for accurate interpretation of the research content, with efforts to avoid deviating from the authors’ intentions.

Table 2 lists 20 papers that were utilized for the literature review [14] - [33] . The family characteristics were: one family of individuals with no disease; five families of children with chronic diseases/cancer; eight families of individuals with a terminal illness; three families with older adults; and three families of bereaved individuals. The countries studied were: nine in United States; two each in Australia and Iran; one each in China, Thailand, South Korea, Ireland, Lebanon and Canada; and one multinational study. Eight of the studies were quantitative and 12 qualitative.

Interviews were conducted with 23 people from 12 families, with the average interview time 125.3 minutes (SD = 41.4). The demographic characteristics of the participants and their families are indicated in Table 3 . Eleven families (84.6%) were nuclear families and two families (15.4%) were extended families. With regard to religious beliefs, 20 people (85.0%) were Buddhists, and three people (15.0%) were non-believers.

a. As fractions are rounded off, the percentage may not total 100%.

A total of 21 categories and 73 subcategories were extracted for the attributes of family spirituality ( Table 4 ). Based on the CSFET, the 21 categories were classified into 6 systems, with six categories and 26 subcategories in family system unit; five categories and 13 subcategories in the family internal environment; two categories and five subcategories in the micro system; one category and one subcategory in the macro system; three categories and 13 subcategories in the supra system; and four categories and 15 subcategories in the family chrono environment system.

Note: CSFET = Concentric Sphere Family Environment Theory; Fsu = family system unit; Int = family internal environment; Mic = micro system; Mac = macro system; Sup = supra system; Chr = family chrono environment system.

Altogether, 20 categories and 29 subcategories were extracted as influencing factors for deterioration of family spirituality ( Table 5 ). These consisted of six categories and eight subcategories of risk/causal/promoting factors; 11 categories and 17 subcategories of preventive/inhibitory/suppression factors; and three categories and four subcategories of context-sensitive factors were identified. For example,

Note: CSFET = Concentric Sphere Family Environment Theory; Fsu = family system unit; Int = family internal environment; Mic = micro system; Mac = macro system; Sup = supra system; Chr = family chrono environment system.

the context-sensitive factor “upon the death of a family member” can become a risk/causal/promoting factor when the number of family members decreases and the family loses the meaning of its existence upon the death of a family member. Conversely, it can also become a preventive/inhibitory/suppression factor when the death of a family member increases contacts between other family members and bolsters family ties.

The three categories of the supra system are “Family believes in religious teachings”; “Family believes that religion is a blessing”; and “Family believes that family spirituality and religion can sometimes overlap”; all of which are related to religion. For families who have faith in religion, recognizing the meaning of the family’s existence can be said to be related to religion. It has been shown that spirituality is related to the ability of families to deal with risks and recover from crises [41] . Family spirituality is also related to the formation of resilient families [42] . Thus, when facing a family crisis, families are likely to overcome the crisis by channeling religious blessings. Regardless of whether they have religious faith or not, it is possible that they believe that family spirituality and religion are sometimes intertwined, since religion and religious teachings provide comfort to families when a family member is sick or when the family faces difficulties.

In the family chrono environment system, the category of “Family believes that family members should share time” that recognizes the meaning of the family’s existence is understood. It was revealed that “family events” are a way for family members to interact with each other and build good family relationships and promote family well-being [43] . Family togetherness is a characteristic of healthy families, and includes the quality of time family members spend together as well as the amount of time they spend together [44] . Nursing professionals will need to create private spaces and opportunities for communication among family members to improve family well-being.

Risk/causal/promoting factors for deterioration of family spirituality include “disharmony in family relationships” in the family system unit, which has been shown to result in the breakdown of family trust and problems in communication. Families in a situation of well-being are characterized by active communication, spiritual well-being, gratitude, and love [45] . When disharmony in family relationships occurs, family trust is destroyed, making it difficult for family members to support each other, while reducing family well-being, which in turn impacts negatively on beliefs that recognize the meaning of the family’s existence. Good family spirituality is reflected by good harmony and communication between family members [46] , and furthermore, family relationships and family well-being [23] are improved, enabling the family to find meaning in its existence. Disharmony in family relationships, therefore, is thought to act negatively on family spirituality.

As for preventive/inhibitory/suppression factors for deterioration of family spirituality, “Medical professionals being close to the family” in the micro system indicates that it is important for medical professionals to suppress the decline in family spirituality. This is related to the micro system attribute of family spirituality, “Family believes that mutual understanding with medical professionals is necessary,” and it is assumed that by medical professionals being close to families, mutual understanding between the family and medical professionals can be deepened. Hohashi [47] advocates the Family Care/Caring Theory, which focuses on families, and states that nursing professionals can be close to families by understanding families and by understanding themselves as nursing professionals. A decline in family spirituality reflects on a family’s problems, issues, difficulties, or suffering, and by understanding this, family nursing based on the Family Care/Caring Theory will enable a decline in family spirituality to be prevented [47] [48] .

In the supra system, “religious consciousness” is one of the context-sensitive factors for deterioration of family spirituality. It has been found that differences in family religious beliefs can be risk/causal/promoting factors for a decline in family spirituality, or conversely, preventive/inhibitory/suppression factors. For sick family members, belief in religion can provide emotional support and prevent a decline in family spirituality. On the other hand, it is possible that an incurable illness can lead to spiritual suffering, which can lead in turn to a decline in family spirituality, or that a family member who is a non-believer won’t obtain a sense of security from religion, even in the event of a family crisis. In other words, differences in the manner religion are dealt with are thought to have different effects on decline in family spirituality [3] , and nursing professionals will need to be aware of a family’s religious background when providing support for a decline in family spirituality.

In this study, using CSFET as a framework, we were able to obtain a comprehensive grasp of the attributes of family spirituality and the influencing factors of decline in family spirituality. Nursing professionals can refer to the attributes of family spirituality when providing family spirituality support. By controlling the influencing factors of a decline in family spirituality, moreover, nursing interventions can be provided to prevent a decline in family spirituality. The attributes of family spirituality and the influencing factors of decline in family spirituality belong to all systems of CSFET, and it became clear that intervention is required not only within the family, but also in the external environment. In the future, we hope to develop a measurement scale for assessing family spirituality, using CSFET as a framework, that will holistically improve family well-being, and apply it to clinical practice.

In the literature review, few studies on family spirituality were found to exist, and the particulars of family spirituality and personal spirituality may be ambiguous. In addition, owing to their sources, many of the literature reviews may reflect some degree of bias toward Western culture. All interview subjects consisted of families with older adults, which may limit the generalizability of the results. The research period coincided with the COVID-19 pandemic, which may have provided participants with greater opportunities to consider the meaning of family, and this may have influenced the results. Moreover, 85% of the interview participants were Buddhists, so a study in countries or regions where other religions are practiced may produce different results.