Exploring subjective quality-of-life indicators in long-term care facilities: a mixed-methods research protocol

Abstract

Introduction

Improving quality of life has become a priority in the long-term care (LTC) sector internationally. With development and implementation guidance, standardised quality-of-life monitoring tools based on valid, self-report surveys could be used more effectively to benefit LTC residents, families and organisations. This research will explore the potential for subjective quality-of-life indicators in the interRAI Self-Reported Quality of Life Survey for Long-Term Care Facilities (QoL-LTCF).

Methods and analysis

Guided by the Medical Research Council Framework, this research will entail a (1) modified Delphi study, (2) feasibility study and (3) realist synthesis. In study 1, we will evaluate the importance of statements and scales in the QoL-LTCF by administering Delphi surveys and focus groups to purposively recruited resident and family advisors, researchers, and LTC clinicians, staff, and leadership from international quality improvement organisations. In study 2, we will critically examine the feasibility and implications of risk-adjusting subjective quality-of-life indicators. Specifically, we will collect expert stakeholder perspectives with interviews and apply a risk-adjustment methodology to QoL-LTCF data. In study 3, we will iteratively review and synthesise literature, and consult with expert stakeholders to explore the implementation of quality-of-life indicators.

Ethics and dissemination

This study has received approval through a University of Waterloo Research Ethics Board and the Social and Societal Ethics Committee of KU Leuven. We will disseminate our findings in conferences, journal article publications and presentations for a variety of stakeholders.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• For our first study, we will recruit a variety of long-term care (LTC) ‘insiders’, including residents and family members, interRAI researchers and leadership for a modified Delphi study.

• A limitation of the first study is the exclusion of residents living with moderate-to-severe cognitive impairment, who represent a large proportion of the LTC population.

• Our second study will use a comprehensive, linked dataset that includes the interRAI Self-Reported Quality of Life Survey for Long-Term Care Facilities and the interRAI Minimum Data Set.

• A major strength of study three is the use of a review methodology that is proven useful for unpacking complex interventions in a pragmatic manner. Researcher biases will be identified and mitigated throughout data collection and analysis with investigator triangulation, member checking and checklists.

Introduction

Background

Quality of life can be defined as how an individual perceives their life in relation to their culture, values, expectations and concerns.¹ Understanding quality of life is especially important in the long-term care (LTC) system. LTC is facing high rates of age-related chronic disease and pressure to maximise value.² As a highly controlled residential institution,³ LTC is also held responsible for maintaining resident quality of life by providing person-centred care and supporting physical and psychological health.² Quality of life overlaps conceptually with health-related quality of life, subjective well-being and social care-related quality of life.^{1 4 5} It is also a broad term, incorporating subjective, objective, personal and organisational elements.⁴ Therefore, it is unsurprising that quality of life is understood differently by different players in the LTC system⁶ and can be easily oversimplified or equated to happiness, quality of care or satisfaction.¹ Figure 1 shows a concept map describing the multiple dimensions of quality of life.¹

Figure 1. Quality of life concept map.

Our ability to describe and compare quality of life in LTC, identify challenges and evaluate solutions is strengthened by quantitative measurement.⁷ While listening to residents one-on-one plays an important role in care,⁸ collecting, analysing and discussing performance related to modifiable challenges, like staff responsiveness in supporting activities of daily living, allows care providers to make evidence-informed decisions and hold themselves more accountable for enhancing the quality of life.^{7 9} Additionally, monitoring residents’ quality of life may enable organisations to address challenges more cost-effectively. To illustrate, decision-makers can determine how to best invest in recreation programmes, food services or clinical programmes based on the expressed needs of residents rather than professional judgement alone.⁷ Measuring quality of life to identify resident needs may also help with addressing quality of life disparities.¹⁰ Since patient experience, healthcare spending and equity are each priorities for healthcare quality improvement within the Quintuple Aim Framework,¹¹ quality of life monitoring is an important component of an LTC quality improvement strategy.

There are many ways to measure LTC residents’ quality of life, including family member reports, staff reports and direct observation. Proxy reports can provide information about externally observable aspects of quality of life and may be necessary to evaluate quality of life for residents living with more severe levels of physical or cognitive impairment.¹² However, proxy-reporting should not be used as a substitute when self-reporting is possible. Self-report is the gold standard for assessing quality of life among LTC residents, even for those living with mild-to-moderate cognitive impairment.^{13 14} Several self-reported quality of life surveys exist, including the interRAI Self-Reported Quality of Life Survey for Long-Term Care Facilities (QoL-LTCF)^{6 15} which will be the focus of this research.

InterRAI is an international research network that aims to improve the healthcare system by creating high-quality integrated assessments and screening instruments to support evidence-informed decision-making.¹⁶ The QoL-LTCF is a reliable and valid survey that was developed by interRAI in 2009, piloted and implemented in several countries for assessing subjective quality of life among LTC care recipients.^{13 14} It is administered by trained assessors who do not provide direct care to respondents. While the QoL-LTCF is not appropriate for assessing quality of life among residents living with severe levels of cognitive impairment, it has been used and researched among residents living with mild-to-moderate cognitive impairment.⁶ To complete the survey, LTC residents are asked to rate their experience on 50 statements and report how often each is true for them on a 5-point Likert scale from never to always. The quality-of-life domains in the QoL-LTCF are listed with examples in table 1. The QoL-LTCF also calculates five summary scales, covering social life, personal control, food, caring staff and staff responsiveness.¹⁴ Depending on which country the survey is being implemented, it also collects demographic information on couple status, gender, age, self-rated health and length of time living in the LTC home.¹⁵

Table 1. Quality-of-life domains in the interRAI Quality of Life Survey for Long-Term Care Facilities with example statements.

Quality-of-life domains	Example statements
Privacy	I can be alone when I wish.
Food and meals	I get my favourite foods here.
Safety and security	I feel my possessions are secure.
Comfort	This place feels like home to me.
Daily decisions (autonomy)	I decide when to go to bed.
Respect by staff	I am treated with respect by staff.
Staff responsiveness	Staff respond to my suggestions.
Staff-resident bonding	Some of the staff know the story of my life.
Activities	I participate in meaningful activities.
Personal relationships	Another resident here is my close friend.

The QoL-LTCF has several advantages. Compared with other surveys, like the Adult Social Care Outcomes Toolkit (ASCOT),⁵ the QoL-LTCF may be better suited for LTC homes that have already implemented other interRAI assessments.16,18 Integrated assessment systems can better support health system integration.¹⁹ For example, the interRAI Assessment for Long-Term Care Facilities (interRAI LTCF) and, its predecessor, the interRAI Minimum Data Set (interRAI MDS) are instruments for comprehensive geriatric assessment that have been mandated in several countries for examining resident health and functional status.^{20 21} The QoL-LTCF can be integrated with the interRAI LTCF and MDS. On the other hand, countries that have not implemented interRAI assessments may prefer to use different tools, like the minimum data set in England.²²

Another advantage of most interRAI assessments, including the interRAI LTCF, is that they can be used to generate risk-adjusted quality indicators. Quality indicators are quantitative measures based on standards of care that capture issues in quality²³ and quality-of-life indicators are a type of quality indicator that captures issues in quality of life. Quality indicators can support research, programme evaluation, decision-making and public reporting.²⁴ To be useful, quality indicators should be actionable,²⁵ acceptable to stakeholders, easy to interpret, sensitive to change, specific, reliable valid,²⁴ harmless to vulnerable groups and resistant to manipulation.²⁶ Additionally, to ensure fair quality comparisons, quality indicators are often risk adjusted.²⁷ Risk adjustment is a process that accounts for differences in resident populations.

There are arguments for and against risk-adjusting subjective quality-of-life indicators. Proponents argue that some individuals systematically rate their care as better or worse than others.²⁸ For example, compared with people with intact cognition, persons with moderate cognitive impairment may rate the quality of food and meals better, and their experience with autonomy, feeling respected and staff-resident bonding may be reported as worse.²⁹ Non-modifiable individual characteristics, such as cognitive impairment, could explain about 20% of the variation in LTC resident quality-of-life reporting.²⁹ Risk-adjusting quality-of-life indicators could improve the fairness of quality comparisons for LTC decision-making and public reporting.^{29 30} However, the chosen risk adjustment variables should be correlated with the quality indicator of interest, should not be indicators of good or poor quality of care themselves, should vary between organisations and should be unaffected by related care processes.^{28 31}

In contrast, concerns have been raised about risk-adjusting subjective quality-of-life indicators. First, if risk adjustment is done poorly, important differences in quality could be negated or exaggerated.³² Second, because of the risk-adjustment process, the voices of some residents could be misrepresented as ‘more important’ than others, leading to unequal treatment.²⁸ Third, some clinical factors such as depression or pain could affect subjective quality-of-life ratings but would not be appropriate risk adjusters because they are themselves important indicators of subjective quality of life. Finally, due to the complex and subjective nature of quality of life, incorrect assumptions could be made regarding the relationships between resident characteristics and quality-of-life outcomes. There is no gold standard to establish the validity of a quality-of-life monitoring system, so some researchers recommend opting for simpler quality measurement approaches.³²

Rationale

While there is a large amount of research on quality of life in LTC,³³ there are gaps when it comes to LTC quality-of-life indicators. Most research on quality indicators thus far has focused on technical, medical and observable elements of care.²⁵ This may be due, in part, to the lack of focus on subjective quality of life in earlier versions of the interRAI-LTCF or MDS assessment.²⁵ However, technical and medical indicators only represent a small part of what makes an LTC home high quality.²⁵ While the interRAI LTCF offers quality-of life-indicators related to objective or observable issues, such as pain, mood or behaviour symptoms, it has a limited focus on subjective quality of life. Some research has investigated quality-of-life indicators in LTC.^{5 34} For example, the ASCOT was developed to measure social care-related quality of life, a concept that overlaps with quality of life for social care services more broadly.⁵ Although it has been proven useful in several countries, it is not integrated in the interRAI suite of assessments. The QoL-LTCF may be a good candidate for subjective quality-of-life indicators, but it is still unclear which statements and scales on the survey are the most important to stakeholders, whether risk adjusting the subjective quality-of-life indicators is feasible or appropriate,³⁵ and if and how subjective quality-of-life indicators could be best implemented to improve the lives of LTC residents.²⁴

Aims and objectives

The purpose of this research is to explore the development and implementation of LTC quality-of-life indicators for the QoL-LTCF. This study will address the following research objectives and questions:

Research objectives

1. Engage with stakeholders to evaluate the importance of statements and scales related to LTC resident quality of life in the QoL-LTCF.

2. Critically examine and test the feasibility of risk-adjusting quality-of-life indicators for the QoL-LTCF.

3. Explore the potential for quality-of-life indicators in LTC homes and how they could be implemented to encourage quality improvement.

Research questions

1. What statements and scales in the QoL-LTCF do resident and family advisors, interRAI researchers and leadership from LTC organisations consider important indicators of quality of life in LTC?

2. Is it appropriate and feasible to risk-adjust quality-of-life indicators for the QoL-LTCF? If so, what implications should researchers consider?

3. How, for whom, in what circumstances and why should quality-of-life indicators be implemented or not in LTC homes?

We hypothesise that participants will identify several statements as important indicators of quality of life in LTC, including staff respect and responsiveness.³⁴ In this list, we expect that there will be at least three statements which can be further evaluated in study 2 for feasibility and appropriateness and that complex implications and additional questions will be discussed that challenge assumptions around quality of life in LTC. Finally, we hypothesise that we will be able to generate implementation recommendations for quality-of-life indicators in LTC.

Methods and analysis

This mixed-methods research will contribute to a PhD dissertation and will be conducted for about 10 months beginning in January 2024. It includes three related studies: a modified Delphi study to address objective 1 and research question A, a feasibility study to address objective 2 and research question B, and a realist synthesis to address objective 3 and research question C. A workflow diagram describing the research is shown in figure 2.

Figure 2. Project workflow diagram. QoL-LTCF, Quality of Life Survey for Long-Term Care Facilities.

Study 1: insider perspectives on subjective quality-of-life indicators for the InterRAI QoL-LTCF – a modified Delphi study

Study design

To begin, we will conduct a modified Delphi study to evaluate the importance of different outcome measures related to quality of life. This study is considered ‘modified’ because the first round of surveys will not entail open brainstorming, but instead evaluate a fixed set of statements and scales from the QoL-LTCF surveys. We will also deviate from the traditional Delphi approach by collecting data with focus groups.³⁶ Data collection and analysis will occur from February to May 2024.

Sampling and recruitment

We aim to recruit about 100 individuals from LTC-related organisations in Canada, the USA and Flanders, Belgium using purposive sampling. We will categorise participants into two panels:

1. Professional experience panel: This panel will include about 40 individuals who are engaged in LTC quality improvement networks. We will email members of the interRAI research network to participate as many are familiar with the QoL-LTCF and motivated to participate in interRAI-related research. We will also recruit individuals from organisations like the Senior’s Quality Leap Initiative, a community of practice of 14 LTC organisations aiming to improve the quality of life and quality of care for residents.³⁷

2. Lived experience panel: This panel will be composed of about 60 LTC residents and care partners engaged in LTC resident and family advisory councils or committees across the Senior’s Quality Leap Initiative.³⁸ We will include individuals who report having the time and ability to participate and sufficient cognitive and communication skills to provide informed consent.

Individuals under the age of 18 or not fluent in English, French or Flemish will be excluded. Recruitment will end once we meet or exceed the target sample size.

Data collection and analysis

Before collecting data, we will obtain approval from all research ethics boards necessary and informed consent from panellists. The data will be collected in one of two ways:

1. Asynchronous: For the professional experience panel, data will be collected electronically with 20 min, quasi-anonymous Qualtrics surveys. There will be three survey rounds, each open for 3 weeks. To maintain panellists’ motivation, we will send reminder emails three business days before the survey closing.³⁹ Detail on our asynchronous data collection and analysis plan is shown in table 2.

2. Synchronous: For the lived experience panel, data will be collected through focus groups. The focus groups will employ discussions for consensus-building around the importance of statements and scales on the QoL-LTCF.³⁶ In each LTC home, the survey will be administered using flexible methods by webinar-trained staff not providing direct care to participants. One or more focus group sessions may be needed depending on the preferences of the group. The focus groups may also be audio recorded if participants agree to it beforehand. To begin, the focus group facilitator will present the first section of statements from the focus group survey. Then, the resident or family advisor group will discuss and rate the importance of each statement on a scale from 0 (extremely unimportant) to 10 (extremely important). Once all of the statements and scales have been discussed and rated as a group, the facilitator will ask the group what they believe are the five most and least important quality-of-life statements in the survey. Relevant comments will be noted by the facilitator, then they will return the surveys via email to the PhD researcher for analysis.

Table 2. Data collection and analysis for the asynchronous Delphi surveys.

	Survey 1	Survey 2	Survey 3
Goals	Narrow down a list of important candidate quality indicators for measuring quality of life in LTC and analyse the demographic data.	Narrow down a list of important candidate quality indicators for measuring quality of life in LTC.	Evaluate a set of candidate quality indicators for acceptability, content coverage, proportional representation and possible contamination.64
Presurvey preparation	Distribute the Qualtrics link via email and employ informed consent procedures.	Provide plain-language group feedback on the synthesised results of survey 1 and distribute the Qualtrics link via email.	Provide plain-language group feedback on the synthesised results of survey 2 and distribute the Qualtrics link via email.
Response format	Quantitative: Demographic questions (ie, role and country) and feedback collected on a list of statements and scales according to an 11-point Likert scale from least to most important.Qualitative: Opportunity to suggest changes, express ideas, provide explanations and ask questions in free text boxes.	Quantitative: Feedback collected on a list of statements and scales that have not yet achieved consensus according to an 11-point Likert scale from least to most important.Qualitative: Opportunity to suggest changes, express ideas, provide explanations and ask questions in free text boxes.	Quantitative: Feedback collected on a list of statements and scales according to an 11-point Likert scale from complete disagreement to complete agreement on whether the indicator set (1) covers the right items, (2) is an acceptable size and (3) does not include items considered irrelevant or misleading.Qualitative: Opportunity to suggest changes, express ideas, provide explanations and ask questions in free text boxes.
Postsurvey follow-up	Share a summary of individual responses and appreciation email; develop survey 2 and group feedback document.	Share a summary of individual responses and appreciation email; develop survey 3 and group feedback document.	Share a summary of individual responses, group feedback document and appreciation emails.

LTClong-term care

Once the Delphi and focus group surveys are submitted, we will listen to all of the focus group audio recordings, take notes on key points of conversation and synthesise the data into a single Excel file. Then, we will ensure the data are stratified by role (ie, resident, researcher, etc) and perform frequency analysis of the importance scores. We will select our candidate quality indicators for study 2 from the list of statements where 70% of the professional and lived experience panels agreed that the item was important or extremely important. Statements that do not reach 70% consensus will still be reported on but will not be considered suitable for quality indicator development, as in previous research.⁴⁰ Comments and notes from the surveys and audio recordings will be examined using content analysis, and then grouped into themes while staying as true to the original wording as possible. The themes should contextualise the importance scores, justify quality indicator selection and provide additional depth for the discussion. A summary of the responses will be shared with participants in June 2024 for member checking.

Study 2: exploring quality-of-life indicator risk adjustment – a feasibility study

Study design

In our second study, we will explore the feasibility of risk-adjusting of quality-of-life indicators. As part of this effort, we will test a risk adjustment procedure using observational data from QoL-LTCF datasets, including a Canadian dataset that links QoL-LTCF and interRAI MDS assessment results and consult with expert stakeholders. Data preparation and analysis will occur from May to July 2024 and expert consultation will occur from July to September 2024.

Data sources

We currently have access to data that were collected from January 2015 to December 2016 (n=331) and June 2011 to August 2013 (n=662) across 10 LTC homes in Ontario, Canada, but additional data may be obtained. Trained volunteers, peer support staff or staff not providing direct care to respondents administered the QoL-LTCF on paper or electronically, and trained assessors administered the interRAI MDS. Once the data were submitted to interRAI Canada, they were linked by an interRAI data manager. More details on the target populations, recruitment and data collection can be found in the relevant user manuals.^{15 21}

Variables

Three variables of interest will be selected as candidate quality indicators for this study based on the quantitative and qualitative findings of the modified Delphi study, the degree of variation between homes, and whether they represent distinct concepts. While more than three variables could serve as candidate quality indicators in the future, the Medical Research Council framework states that exploring feasibility on a smaller scale can support decision-making about future larger-scale evaluations and implementation initiatives.⁴¹ We will also identify useful risk-adjustment variables from the demographic variables in the QoL-LTCF and from the wider range of demographic variables, summary scales and health-related variables (eg, impairment in activities of daily living, cognitive impairment and disease diagnoses) in the interRAI MDS. As mentioned previously, risk-adjustment variables should be correlated with the quality indicator of interest, should not be indicators of good or poor quality of care themselves, should vary between organisations and should be unaffected by related care processes.^{28 31}

Data analysis

We will begin analysis by ensuring the data are clean, excluding missing values and performing exploratory analyses. We will exclude new admissions with a length of stay of under 3 months since their experiences may overly reflect the stress and uncertainty associated with relocation, termed ‘relocation stress syndrome’.^{42 43} Observations with missing length of stay data will also be excluded. Preliminary analysis will entail creating frequency tables and conducting χ² tests for each candidate indicator. To ensure that the indicators are relevant, candidate indicators with a prevalence of less than 3% will not be considered for risk adjustment (eg, if less than 3% of residents in the entire sample agree that ‘I am bothered by the noise in my home’, it will no longer be considered as a candidate quality indicator for the present study).⁴⁰

Then, we will apply risk-adjustment methods adapted from Jones et al.⁴⁴ The first step will involve stratifying the data for each quality-of-life indicator into high-risk, medium-risk and low-risk groups, divided at the 20th and 80th percentile of a relevant risk adjustment variable, such as the Resource Utilisation Groups (RUG-III) casemix algorithm,⁴⁴ cognitive performance or age group. In each stratum, we will generate observed indicator scores for the individual LTC homes and the organisation-wide reference population. Risk adjustment will then proceed for each indicator using at least one of two methods described in table 3.

Table 3. Risk adjustment methods⁴⁴.

Method	Steps	Relevant formulas
Indirect standardisation: Uses multivariable logistic regression to adjust for clinically relevant covariates (eg, age, RUG-III) and produces three scores adjusted at the resident level.	Calculate expected indicator scores within each risk group (high, medium and low risk) using data from the population of interest and formula A. Generate three performance ratios by dividing observed indicator scores by expected indicator scores. Multiply the performance ratios by the observed indicator scores for the reference population. Transform using the logit function.	Formula A:${\displaystyle y_{RL}=\frac{e^{{\beta }_0+{\beta }_1{c}_1+{\beta }_2{c}_2\dots +{\beta }_n{c}_n}}{1+e^{{\beta }_0+{\beta }_1{c}_1+{\beta }_2{c}_2\dots +{\beta }_n{c}_n}}}$Where,$y_{RL}$ is the expected indicator score for the risk level of interest.${\beta }_0$ is the fixed, unknown intercept.${\displaystyle {\beta }_1\dots {\beta }_n}$ are the fixed, unknown regression coefficients corresponding to each covariate.${\displaystyle c_1\dots c_n}$ are the mean covariate values.
Directstandardisation: Applies standard population weights to adjusted indicator scores, yielding a single composite score adjusted at the facility level.	Multiply the resident-adjusted scores for each risk group by the reference population weights (ie, proportion in strata). Sum into a single composite score, as shown in formula B.	Formula B:${\displaystyle S_{adj}=\mathrm{\Sigma }\left(S_{RL}\times P_{RL}\right)}$Where,$S_{adj}$ is the composite indicator score.$S_{RL}$ is the adjusted indicator score for each risk level.$P_{RL}$ is the proportion in strata for each risk group, calculated by dividing the size of the risk group by the size of the reference population.

RUGResource Utilisation Group

Quality appraisal and feasibility assessment

Finally, we will appraise the risk-adjusted quality-of-life indicators using the Appraisal of Indicators through Research and Evaluation (AIRE) instrument and informally consult with up to 10 expert stakeholders for feedback on the findings. The stakeholder group will be composed of English-speaking representatives from the Senior’s Quality Leap Initiative, Flemish Institute for Care Quality and/or interRAI who were recruited for the Delphi study and have experience using or developing quality indicators. To consult the stakeholders, we will ask the question ‘do you have any thoughts you would like to share about risk-adjusting quality-of-life indicators?’ and additional prompts as needed. Once stakeholder comments have been collected and reviewed by the research team, we will meet to further discuss the feasibility and appropriateness of risk-adjusting quality-of-life indicators. Specifically, we will weigh the ethical, methodological and practical implications of developing a set of quality-of-life indicators, including concerns about utility, data availability, cultural differences and privacy.

Study 3: exploring the implementation of quality-of-life indicators in LTC homes – a realist synthesis

Study design

Our final study will be a realist synthesis, scheduled from May to October 2024. Realist syntheses are less rigid than systematic reviews⁴⁵ but still useful for synthesising several sources of information. Our aim will not be to evaluate the effectiveness or efficiency of quality-of-life indicators. Instead, realist syntheses unpack the ‘black box’ of complex interventions by producing two types of theory⁴⁶: (1) middle-range theories provide structured explanations of social phenomena using context+mechanism=outcome (CMO) configurations and (2) programme theories explain how and why specific programmes yield different outcomes in different contexts.⁴⁷ By combining middle-range and programme theories, we aim to explain if and how quality-of-life indicators could be implemented to promote quality improvement in LTC homes, for whom, in what circumstances and why. We will perform the realist synthesis in six iterative steps.

Step 1: scoping the literature

First, we will informally scope the literature on quality-of-life indicators in LTC within two relevant bodies of literature, gathering key references from PubMed, Google Scholar and Connected Papers. To facilitate knowledge synthesis, we will compile all references in a Zotero reference management software library and note relevant concepts from abstracts and figures in an Excel spreadsheet.

Step 2: stakeholder interviews

Next, we will purposively recruit and conduct informal, virtual interviews with up to 10 stakeholders who were previously engaged in the Delphi study. Engaging subject matter experts can improve the efficiency of reviews.⁴⁸ Therefore, we aim to include researchers and healthcare leaders with experience using quality indicators. We will exclude individuals who do not speak English or do not have access to a computer. Stakeholders will be asked the following interview questions (adapted as needed):

1. In your opinion, are quality-of-life indicators useful (or could they be useful) for quality improvement in LTC homes? Why or why not?

   • Prompt 1: Who would benefit or not from quality-of-life indicators if they were implemented in LTC?

   • Prompt 2: How could quality-of-life indicators be implemented? Are there circumstances that would make them easier or harder to implement in LTC?

2. What theories, papers, practice guidelines or other resources do you know of that could help with implementing quality-of-life indicators in LTC?

3. Are you interested in being sent the results of this study or being personally acknowledged in the final publication?

The interviews should take about 30 min to complete and will be audio recorded to facilitate note-taking. Any resources identified by the stakeholders will be searched for and added to the reference library.

Step 3: searching process

We will formally search white and grey literature for additional references. In collaboration with a university librarian, we will compile a list of search terms related to “quality of life”, “quality indicators”, “implementation” and “long-term care”. To identify relevant white literature, we will search PubMed via the National Library of Medicine (NLM), Scopus and CINAHL via EBSCOhost, with the search terms while restricting the search to papers in English. For the grey literature search, we will use the Trip Database and Google Search, screening titles and summaries in the first five pages of the searches for government reports and documents from credible organisations. We will prioritise case studies, qualitative research and reports from reputable sources and exclude unpublished manuscripts and conference abstracts.

Step 4: document selection and appraisal

Two researchers will then independently screen titles and abstracts from the collected literature using Covidence review management software.⁴⁹ After removing duplicates, we will screen the full texts and appraise them using questions adapted from Pawson’s ‘fitness-for-purpose’ criteria⁵⁰: (1) ‘what were the findings? Is the study design suited to make those claims?’ and (2) ‘do the findings build on or challenge the emerging pattern? Is a shift in scope needed?’. Disagreements between reviewers will be resolved through discussions with the research team.

Step 5: data extraction

Once the papers have been reviewed, we will extract and analyse relevant information using a realist lens. Specifically, we will label the text with the theory-driven codes CMO to separate relevant aspects of each article. Then, we will apply data-driven subcodes. Using modified thematic analysis, the subcodes will be listed and organised into themes based on their frequency and contributions to the emerging theory.⁵¹

Step 6: analysis and synthesis

Finally, we will synthesise the codes, structure them into themes, incorporate insights from the stakeholder interviews, and develop middle-range and programme theories. We will primarily employ mind-mapping to generate theory by arranging the ‘pieces of the puzzle’ into charts and figures. The interview participants will then be invited to approve or reject a one-page summary of the research findings via email. The steps of the realist synthesis will be iteratively repeated as needed, and the study will conclude when our themes meet the conceptual depth criteria of range, complexity, subtlety, resonance and validity.⁵²

Patient and public involvement

Patients and the public were not involved in designing the research protocol or determining the aims and objectives. However, in study 1, we will involve LTC residents and care partners in focus groups to gather their perspectives on quality of life in LTC, which inform parts of studies 2 and 3. We will also disseminate our results to all interested participants.

Discussion

Summary of expected outcomes

This research will contribute to the growing body of literature on quality-of-life indicators. We expect to evaluate the importance of several quality-of-life statements and scales, examine the feasibility of risk-adjusting quality-of-life indicators for the QoL-LTCF and describe if and how LTC quality-of-life indicators could be best implemented.

Limitations

This research may be affected by some limitations. First, no financial compensation is available for participants due to resource constraints. Instead, we will express our gratitude with appreciation letters. Second, participant biases may limit the generalisability of this research. For example, the Delphi study may be susceptible to selection bias since we will only recruit resident and family advisors connected to large organisations committed to quality improvement. While purposeful sampling offers advantages, like the efficient recruitment of information-rich cases,⁵³ individuals who are not resident or family advisors or living in LTC homes without major quality improvement programmes may have different perspectives on quality of life. We have also chosen to exclude residents with moderate to severe dementia from the focus groups as they cannot provide their own informed consent. These residents represent a large proportion of the LTC population and may have different experiences around quality of life that are better captured with proxy-reported instruments.⁵⁴ Finally, the focus groups may be susceptible to social desirability bias and groupthink.⁵⁵ Running separate focus groups at multiple LTC homes may reduce the effect of this bias on the results. We will employ investigator triangulation, member checking, and checklists (eg, AIRE) to minimise bias throughout our research.⁵⁶

Strengths and contributions

This research also has several strengths, including our mixed-methods approach, stakeholder engagement and use of the QoL-LTCF. First, we will use quantitative and qualitative research methods, including surveys, focus groups, interviews, secondary data analysis and literature searches to obtain and describe multiple perspectives related to quality of life in LTC. Some researchers have suggested that applying mixed methods and combining subjective and objective data supports the development of more trustworthy and useful quality indicators.⁵⁷ Specifically, the Delphi technique is widely seen as valuable for selecting candidate quality indicators.⁵⁸ Second, this study will engage a diverse array of stakeholders, including LTC residents, care partners, researchers and organisational leaders. Participant triangulation can improve the trustworthiness of research⁵⁹ since different stakeholders have unique perspectives on quality of life in LTC.⁶ Third, using the QoL-LTCF offers several advantages. The QoL-LTCF is a robust, reliable and valid survey for evaluating subjective quality of life among LTC residents.^{13 14} Since it is part of an international third-generation assessment system,¹⁶ it can be linked to other robust, reliable and valid measurement instruments such as the interRAI LTCF. Therefore, the QoL-LTCF is well positioned to streamline the piloting and implementation of quality indicators, a task rarely performed by other researchers.⁶⁰

Finally, the present research may offer several research and practice contributions. Depending on our findings, quality indicators developed from this study could facilitate benchmarking and comparisons between quality of life in LTC and other outcomes, such as decline in mood, worsening pain, decline in activities of daily living and more. Researchers using interRAI data could also be enabled to highlight gaps in care, advocate for more proactive quality evaluation and support approaches that enhance resident quality of life like person-centred or relationship-centred care. Our findings could also open avenues for exploring relatively new research methodologies like multiattribute utility analysis, which optimises decision-making with composite measures based on preference-weighted sources of information.⁶¹ Regarding practice contributions, we expect that evaluating the statements and scales in the QoL-LTCF will improve its utility for supporting LTC residents without dementia or mild-to-moderate dementia. Moreover, the knowledge we generate on risk-adjusting and implementing quality-of-life indicators, could be valuable for organisations hoping to measure, benchmark and compare their performance with others.

Ethics and dissemination

This research and all relevant materials were reviewed and approved by a University of Waterloo Research Ethics Board (#45825, #45826, #45956) and Social and Societal Ethics Committee of KU Leuven (G-2023-6931 R3). There are no known or anticipated risks to the participants of this study, who will be required to provide informed consent. We will inform participants that their responses are confidential, they can withdraw from the study at any time up until the results are submitted for publication, and their data will be erased if they choose to withdraw. All primary data will be securely stored and retained by the LTC organisations and/or researchers for 5 years following the conclusion of the study.

Beyond academic publications and conference presentations, we plan to present our research findings to several stakeholder groups, such as the Senior’s Quality Leap Initiative. Our dissemination strategies will be informed by research best practices.^{62 63}