Study design
This exploratory qualitative design study is part of a sequential multi-methods research project developing and testing the feasibility of a multicomponent tailored intervention targeting sedentary behaviour and physical inactivity among nursing home residents at risk of, or with sarcopenia. The qualitative results were reported following the COREQ checklist [19] to ensure comprehensive standards and transparency, while drawing on Braun and Clark’s guidelines for thematic analysis reporting [20].
Setting
Two public-private partnership nursing homes with 85 beds and 120 beds, respectively located in cities in Hunan Province, China. The localities comprise high-aged populations, with in 2023, 22.2% of people aged 60 and older, above the national average of 21.1% [21]. The nursing homes were selected based on facility type (i.e., size, nursing home provision) and geographical location within Hunan province to reduce logistical challenges related to transportation, communication, and institutional policies.
Sample size and selection of participants
Criterion-based purposive sampling was applied to residents (i.e., by age, risk of sarcopenia) to include participants central to the study’s focus. Maximum variation purposive sampling was used for staff (i.e., by professional roles, work experiences) to reflect diverse perspectives on the study’s topic. The planned sample size comprises approximately 20 residents and 10 staff, guided by the concept of information power [22] and related studies [23, 24], continuously evaluating participant group size based on study relevance and quality of data. Inclusion criteria for residents were those aged 60 years and over, with mental capacity to give informed consent, and at risk of sarcopenia. Residents’ risk of sarcopenia was identified using sarcopenia case-finding criteria recommended for research and clinical care using the Asian Working Group of Sarcopenia. The criteria consist of: (1) Presence of any of the following clinical conditions: functional decline or limitation; unintentional weight loss; depressive mood; cognitive impairment; repeated falls; malnutrition; and/or chronic conditions (e.g., heart failure, chronic obstructive pulmonary disease, diabetes mellitus, chronic kidney disease). And (2), if no clinical conditions above are present: low calf circumference (male < 34 cm, female < 33 cm), or score of Strength, Assistance in walking, Rising from a chair, Climbing stairs, and Falls (SARC-F) questionnaire ≥ 4, or score of SARC-F combined with Calf circumference (SARC-CalF) questionnaire ≥ 11. Inclusion criteria for staff were individuals working at the selected sites with three months or more experience in long-term care, including managers, registered nurses, and healthcare assistants.
Finally, 26 participants (14 residents and 12 staff members) included in this study. A total of 20 interviews were conducted, with 17 individual interviews (14 residents and 3 staff) and 3 small group interviews with 9 staff (Table 1). Although the final numbers slightly differed from the initial plan, the achieved sample was sufficient. No new themes emerged in the final interviews, and given the study’s narrow focus, sample specificity, and high-quality data, the sample provided adequate information power to address the research aims. Of the 30 participants approached, 2 residents were unable to participate due to severe hearing impairment and lack of cognitive capacity, and 2 staff were excluded due to limited working experience. Most participants were female (residents 57% and staff 83%). Residents were mean age 78.7 years. Most residents had a primary education (n = 6), followed by junior high (n = 4) and high school (n = 3), with one resident being without formal education. Their duration of residence in nursing home was evenly distributed across short-term (≤ 1 year, n = 5), mid-term (1–3 years, n = 4), and long-term (≥ 3 years, n = 5). Professional roles of staff included managers, registered nurses, and healthcare assistants, with most (n = 7) having over three years of experience in nursing homes.
Recruitment
Residents and staff were separately identified, approached, and recruited face-to-face. For residents, study flyers were posted on bulletin boards in the nursing home sites. Researcher (YM) introduced the study to potential participants using an information sheet and, with staff assistance, gauged residents’ willingness to participate. Residents mental capacity to give informed consent was assessed according to the Mental Capacity Act Code of Practice, focusing on understanding, retaining and weighing up relevant information, and communicating their decision [25]. Residents were asked to talk through detail in the information sheet to check understanding and recall, encouraged to discuss the pros and cons of participating with staff and/or family, and to express their choice clearly verbally or through other means [25]. Researcher (YM) made every effort to maximise residents’ autonomy, for example for those with impaired capacity, iterative communication methods (i.e., oral, and written information, body language) were used [26]. Any questions were addressed, and at least 24 h were provided for decision-making.
For staff, the study was introduced during informal meetings. They were given at least 24 h to decide on participation, and informed that their decision would not affect their work or rights. Separate individual or small group interviews were scheduled for managers, nurses, and healthcare assistants, following informed consent.
Participation in the study was entirely voluntary, and no financial or material incentives were offered to either residents or staff. Participants were informed that they would not receive any personal or financial benefit from taking part, and that their decision to participate or not would have no impact on their care, treatment, or employment status. Participants were assured that they could withdraw from the study at any time and for any reason, without any adverse impact on their accommodation or care (for residents), or their employment or working conditions (for staff).
The researcher (YM) was highly attentive to the potential vulnerability of nursing home residents due to advanced age and multimorbidity. All participants were also informed about the potential consequences of participation, including the minimal risks and possible fatigue and emotional discomfort during interviews.
Data collection
Data collection involved semi-structured interviews conducted between January and March 2023, carried out individually with residents and either individually or in small groups with staff members. All data collection was completed by YM, a local of the study area, fluent in the local dialect and customs, and a registered nurse and PhD candidate in nursing trained in qualitative research and supported by an experienced qualitative researcher (CE). Interviews were conducted face-to-face in Chinese and with the researcher’s fluency in the local dialect enabling rapport with participants and understanding of cultural cues and dynamics during interviews. Individual interviews with residents were conducted in their rooms or public areas based on their choices. The median interview length was 49 min (range: 23–106 min), with adjustment into shorter interviews according to the situation of the interviewee. Individual or small group interviews with staff were undertaken in the nursing home, in an office or meeting area. In small group interview, semi-structured interviews are conducted with several people at the same time. It sought to prioritise individual input within the group, with for example directing questions to each person to ensure each could contribute [27].
To minimise fatigue and distress during interviews, participants were offered the option of multiple shorter interview sessions, with breaks as needed. For example, one resident completed the interview in three sessions averaging 15 min each. During interviews, participants were regularly reminded that they could pause or stop the interview at any time. The interviewer monitored participants for signs of discomfort and responded by offering breaks or stopping the interview if requested. No participant reported distress or chose to withdraw during or after the interview.
The interview topic guides included sections for collecting demographic data, exploring participants’ experiences with sitting/lying/reclining time reduction and physical exercise, and identifying reasons for engaging in or avoiding sedentary reduction and exercise. The Behaviour Change Wheel, centred around Capability, Opportunity, Motivation-Behaviour model [28], informed the design of the interview topic guides. Questions were designed to assess residents’ physical abilities, residents’ and staff’s understanding, knowledge, and skills (Capability) related to reducing sedentary behaviour and increasing physical activity for nursing home residents (Behaviour). External factors that might facilitate or hinder physical activity, such as social support from staff or family, available resources, and the physical environment (Opportunity) were covered. Participants’ reflective (e.g., beliefs and intentions) and automatic (e.g., emotions and habits) responses (Motivation) were also considered. Interview topic guides were adjusted separately for residents, senior staff (managers and nurses), and healthcare assistants (Supplementary material). Pilot interview conducted by YM with a researcher (YZ) and a nursing home resident refined and confirmed the topic guide’s relevance and accessibility. Interviews were digitally audio-recorded with consent. Field notes recording interview process, contextual factors, participant responses and reflexive journaling recording personal reflections were completed after interviews and used in the data analysis.
Data analysis
Interviews were transcribed verbatim by YM and a bilingual (Chinese and English) postgraduate student with transcription experience and checked for accuracy by YM. Data underwent codebook thematic analysis (framework method, primarily inductive with supplemental deductive) [29, 30]. This was achieved by combining inductive data analysis with deductive theoretical interpretation to enhance the relevance and applicability of results. MAXQDA 2020 software was used for analysis. The analysis process started with transcription and data familiarisation. Inductive coding was then conducted line by line by two Chinese researchers (YM and LC) in Chinese, for half of the interviews (n = 10). Four bilingual (Chinese and English) researchers (YM, HC, YZ and LC) reviewed the coding and translated the codes, sub-themes/themes, and key quotes in English. An English working analytical framework was developed after initial coding, reviewed, discussed, and agreed upon by the wider team (CE, MM and AB) and bilingual researchers (HC, YZ and LC), with iterations continuing until no new codes were identified. This framework was then applied to subsequent transcripts, with some codes combined and no new codes developed. Data were summarised into a framework matrix using a spreadsheet, included references to illustrative quotations.
The Ecological Social Theory [31] was used post-coding to contextualise and organise emerging themes and codes within a multi-system perspective, covering microsystem (individual), mesosystem (organizational), exosystem (family), and macrosystem (societal factors), that influence sedentary behaviour and physical inactivity in nursing homes. This theory recognises the ecology of human growth and development, emphasising the complex interplay of individual, socio-cultural, and environmental factors [31]. We interpreted and discussed the results with consideration of field notes and reflexive journaling. Our findings were reported in a descriptive way.
In addition, a Theory of Change logic model, previously developed through a systematic review [32], provided a theoretical model for interventions addressing sedentary behaviour and physical inactivity in nursing homes. Theory of Change is a pragmatic framework used to plan, describe and evaluate the processes through which a desired change or outcome is expected to occur [33]. Theory of Change presented in a logic model visually illustrate the relationships between intervention activities and desired outcomes, outlining underlying assumptions and contextual factors. The findings of this study were applied afterward to update and contextualise the model, particularly for Chinese nursing homes.
Establishing rigour and trustworthiness
Credibility and auditability are crucial criteria for assessing the rigour and trustworthiness of qualitative data [34]. Three strategies were implemented to strengthen credibility and auditability: (a) audio-recording and verbatim transcription of the interviews, (b) investigator triangulation was applied during data analysis through independent coding followed by group meetings for data analysis and interpretation, (c) maintaining a detailed audit trail, which included comprehensive records of data collection methods, coding decisions, analysis procedures, and their rationales [34], and (d) data triangulation was achieved by including participants from diverse stakeholder groups, residents and staff in various roles (nurses, healthcare assistants, and managers), to ensure a broad range of perspectives on the phenomenon under study. Chinese transcripts were initially coded in their original language, then reviewed and translated by four bilingual team members, resulting in an English working analytical framework that was reviewed, discussed, and agreed upon by co-authors until no new codes emerged. In line with Braun and Clarke’s guideline on thematic analysis reporting [20], YM, as the lead researcher, used reflexive journaling to engage in critical reflections on how the position as a PhD student and registered nurse with limited clinical experience shaped interactions with participants and influenced the research process [35].
Ethical considerations
Ethical approval for this study was obtained from the King’s College London (KCL) Research Ethics Committee (Ref: HR/DP-22/23-33808). The researcher (YM) conducted fieldwork in China with the support of the China Scholarship Council. Although not affiliated with a Chinese institution, formal permission to conduct the study was granted by the management of the two participating nursing homes in Hunan province, who acknowledged the KCL ethical approval as sufficient.
To protect participants’ privacy, all transcripts were anonymised by removing any personally identifiable information and assigning unique ID number to each participant. The audio recordings were securely stored on a password-protected device accessible only to the research team. After transcription, the audio files were deleted to further ensure confidentiality. The anonymised transcripts are stored securely and will be destroyed within the specified time in accordance with the ethically approved data management plan.
All procedures adhered to the principles of the Declaration of Helsinki [36], including obtaining written informed consent from all participants, ensuring autonomy, confidentiality, and the protection of vulnerable groups.