This mixed-methods study consists of two phases: (1) quantitative survey gathering data regarding the experiences and perceptions of primary care among older adults, and (2) qualitative focus group discussion to gain an in-depth understanding of the factors influencing older adults’ perceptions of service quality, satisfaction, and trust in both public and private primary care services.
Phase 1 – Quantitative study by questionnaire survey
Eligibility criteria
The study population will include Hong Kong permanent residents aged 60 or above, who can read Chinese or speak Cantonese, and who have usual sources of care when he/she is sick or needs health advice (including Accident and Emergency). Those who are diagnosed with dementia or cognitive impairment, and severe vision impairments will be excluded from the study.
Sample size calculation
For sample size calculation, it should be 10 times the maximum number of measuring items pointing at the constructs in the partial least square path model [14]. There are 51 question items measuring 11 constructs in the questionnaire, the target sample size of this study will be 510 (51 × 10).
Recruitment
A cluster sampling focused on the geographical and socio-demographic factors will be adopted for the quantitative study. Cluster sampling is a widely utilised sampling technique for recruiting subjects from diverse geographical areas [15]. For the current study, the clusters are defined by 18 districts in Hong Kong and the median monthly household income according to Census and Statistics Department [16]. These clusters represent high- economic districts (e.g. Central and Western, and Tsuen Wan), middle-income districts (e.g. Sha Tin and Yau Tsim Mong), and low-income districts (e.g. Kwai Tsing and Sham Shui Po). To ensure representation across different socioeconomic groups, an equal number of participants will be recruited from each economic cluster. The questionnaire will take 15 to 20 min to complete. Participation is voluntary, and informed consent will be obtained.
The quantitative questionnaire survey will be conducted both online and through field surveys. The invitation message, along with questionnaire link will be sent to individuals and groups with connection to the investigators, colleagues and the research centre. As the study aims to obtain proportionate number of participants from each economic level, the research team will keep checking the distribution to balance the representation of participants. The research team will also visit the elderly community centres and private clinics in the field survey.
Measurement
The research study employs the Primary Care Assessment Tool (PCAT) as a guide for developing a more comprehensive primary care service to offer higher perceived service quality and experience to older adults. The questionnaire consists of two main sections (Table 1). The first main section is the validated Primary Care Assessment Tool Chinese Version (PCAT-C), contains seven domains with 37 items, covering first contact utilisation and first contact accessibility, ongoing care, coordination, comprehensiveness, family centeredness, community orientation [17]. First contact utilisation refers to the extent to which the source of care is used for various types of problems. First contact accessibility is used to measure access to the (usual) source of care. Ongoing care measures the continuing use of a regular source of primary care and the relationship between the source of care and the user. Coordination of care measures the recognition of prior and existing medical records and health needs. Comprehensiveness measures the services provided or received in primary care. Family-centeredness measures the involvement of family members in primary care. Community orientation measures the provision of community care by providers.
The second main section consists of 14 items, including four constructs about perceived service quality, satisfaction, trust, and revisit intention. Perceived service quality refers to an individual’s experience and impression of the healthcare services [18]. Satisfaction is about patients’ satisfaction with the healthcare services and providers. Patients’ trust refers to patients’ confidence in the healthcare providers as being capable and reliable in meeting their health needs [19]. Patients should feel comfortable and relaxed with the practitioners and staff of health facilities. Revisit intention indicates whether a customer will return to and consult the same doctor or not [18, 20]. For perceived service quality, 5 items are adopted from a study [21]. 3 items each are used to assess satisfaction and trust respectively [22]. The last 3 items measuring revisit intention are adopted [23, 24]. Five-point Likert-scale (1 = Definitely not, 2 = Probably not; 3 = Not sure; 4 = Probably; 5 = Definitely; or 1 = Strongly disagree, 2 = Disagree; 3 = Not sure; 4 = Agree; 5 = Strongly agree) will be used in the questionnaire.
Data analysis
The interaction effects among the constructs will be analysed using partial least squares-structural equation modelling (PLS-SEM). The SmartPLS 4.0 statistical software will test the reflective measurement model and structural model. Reliability (using the value of Cronbach’s alpha and combining reliability greater than 0.7 and 0.708 to check the internal consistency) and validity (including convergent and discriminant validity) of the measurement instruments will be assessed. The outer loadings should be greater than 0.6 to check the convergent validity. For checking discriminant validity, the Average Variance Extracted (AVE) of each construct should be greater than 0.5 [14], as well as the value of Fornell Larcker criterion and heterotrait-monotrait (HTMT) ratio will be assessed. Data analysis and collection will be conducted concurrently to determine the response rate, the number of surveys collected, and participant feedback. The responses and results from the questionnaires help the research team prepare for the focus group discussion questions in Phase 2.
Phase 2 – Qualitative study: focus group discussion and in-depth interviews
Sampling and recruitment
A purposive sampling approach will be used for the Phase 2 focus group discussions. To ensure a balance between diversity and meaningful interaction, each focus group will consist of 4 to 12 participants, a size that facilitates the sharing of individual perspectives while maintaining group diversity [25, 26]. Published study suggest that 90% of thematic content can be captured with three to six focus groups, and that four groups are generally sufficient to achieve data saturation [27]. Therefore, five focus group discussions will be conducted, each comprising 10 participants. These sessions will involve at least 25 older adults along with their family members or caregivers. In total, 50 participants will be recruited for this phase of the study. Twenty-five older adults who participated in Phase 1 will be invited to join focus groups based on their willingness to participate further, their ability to share their experiences. To ensure the sample includes a range of socio-demographic backgrounds (e.g., age, gender, district, and income level), these participants will be identified during the Phase 1 survey, with an opt-in question to assess interest in a follow-up qualitative session. In addition, their family members or caregivers will be invited if they are involved in the older adult’s primary care experiences and express willingness to participate. Older adults will be asked to nominate a caregiver or family member who actively supports them in their healthcare decisions or accompanies them to medical appointments. Caregivers will be contacted through the older adult. We will then provide them with study information and invite them to join the same focus group discussion. There will be five focus groups, each consisting of approximately 10 participants (including older adults and caregivers). Each session will last 1.5 to 2 h.
Measurement
The findings obtained from Phase 1 will help design the guiding questions used in the focus group discussion. Semi-structured interviews will be employed. Questions regarding the expectations of primary care service will also be asked, such as “What are your expectations of GOPCs, DHCs/EHCs?”. In addition, some guiding questions will be taken for reference from the qualitative study on community-based primary health care for older adults [28], like “What are some of the words you would use to describe ’quality of care’?” “What are the barriers or challenges that prevent you from accessing or using primary care services? How can they be overcome or improved?” and “What are your expectations or preferences for primary care services in Hong Kong? What kind of services or supports do you need or want as an older adult?”.
Data analysis
The team members will host the focus group discussions and take notes of the conversation with the participants. The research assistant will audio-record the discussions and accomplish verbatim transcription into Cantonese for data analysis. Audio recordings will be anonymised, stored securely, and accessible only to the research team. At least two investigators will validate the transcription before data entry. By using the Qualitative Analysis Guide of Leuven [29], data analysis will be processed to inform potential adjustments during data collection. The first process of data analysis will involve preparing the coding stage. A narrative report will be written, and the essence of the informants’ experiences and opinions will be articulated. The investigators will develop the conceptual schemes from the narrative report. Thereafter, they will discuss and validate the conceptual schemes by re-reading the informants’ experiences and opinions to determine whether the conceptual schemes can connect the data (transcripts). The research team will address and resolve discrepancies during regular meetings throughout the project. The second process of data analysis will be conducted using Qualitative Data Analysis (QDA) Miner 5 of Provalis Prosuite (Version 6.0.16) to identify key themes and insights from the focus group discussions. This qualitative software will facilitate coding, generation of categories and development of themes. Two investigators will conduct the initial coding of the transcriptions independently and consult a third investigator in the event of any disagreements regarding the identified themes.
Themes, categories, and codes will be analysed and interpreted with all definitions documented in a comprehensive code book. Investigators will discuss these to ensure consistency and reliability. Code saturation will be assessed by reviewing each group’s code development process and the transcripts [30]. Data analysis will continue until saturation is reached, which indicates that no new themes or codes emerge from the analysis. The results will provide an in-depth examination of the informants’ experiences, perceptions, and opinions regarding primary care, mainly focusing on perceived service quality, satisfaction, trust, and revisit intention.
Triangulation of quantitative and qualitative findings
Triangulation refers to the use of multiple methods to study a phenomenon [31]. The project’s triangulation of quantitative and qualitative findings will confirm results, yield more comprehensive data, enhance validity, and deepen the understanding of perceived service quality, satisfaction, trust, and revisit intentions related to primary care for older adults.
Any potentially conflicting findings will be discussed and resolved by team members.
link