Understanding Factors Influencing Personal Care Workers Intentions to Leave: A Systematic Integrative Review

Abstract

Background and Objectives

The global need for older persons’ care, driven by demographic shifts such as an aging population and increased life expectancy, is outpacing the available care workforce. Nursing home operators face complex workforce challenges, including high turnover and low retention, particularly among personal care workers (PCWs). The objective of this review was to understand the factors influencing PCWs’ intention to leave.

Research Design and Methods

Following Whittemore and Knafl’s methodology and adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, an integrative review of peer-reviewed literature published between 1997 and 2024 was undertaken. A comprehensive search was conducted across several databases, including Business Source Complete, CINAHL Complete, Medline Complete, APA PsycINFO, and EMBASE. All studies underwent independent screening and quality assessment using the Mixed-Methods Appraisal Tool.

Results

Included were 21 articles reporting 20 studies using quantitative (n = 17), qualitative (n = 3), and mixed methods (n = 1), which identified various factors influencing PCWs’ intention to leave. These factors were categorized as individual, job and organizational, and socio-political-environmental. Demographic characteristics, employee satisfaction, health, career development, workplace conditions, leadership, empowerment, resident connections, and societal perceptions influenced personal care workers’ intentions to leave.

Discussion and Implications

The findings of this review reveal opportunities for nursing home operators and policy-makers to implement strategies to mitigate the high rates of turnover through PCW role and organizational reforms. Further, the findings underscore the need for qualitative research exploring the perceptions and experiences shaping the employment intentions of current and former PCWs.

Background

Globally, the population is living longer, with this trend expected to continue (Royal Commission into Aged Care Quality and Safety [RCACQS], 2021; United Nations, 2018). This increase in life expectancy has seen a change in patterns of disease and dependency, with new conditions emerging and complex comorbidities increasing, placing unprecedented demand for older persons’ care (Office of the Royal Commission, 2019). Across high-income countries, including Australia, the United Kingdom (UK), and the United States of America (USA), nursing homes (NH), also known as residential aged care, long-term care facility and care home are experiencing serious challenges in sustaining a skilled direct care workforce to support the growing population of older people (Chenoweth et al., 2014; RCACQS, 2021; Rosen et al., 2011). This is compounded by NH operators currently facing multidimensional workforce issues, including difficulty attracting staff, an increasing shortfall in the direct care workforce, high levels of staff turnover, and low levels of retention, in addition to a growing population of older people (Castle et al., 2007; Denny-Brown et al., 2020; RCACQS, 2021). These changes place increasing demands on services for older people, necessitating a significant increase in the direct care workforce, particularly personal care workers (PCWs), also known as certified nursing assistants, nursing assistants, personal care aides, and direct care workers.

Personal care workers comprise approximately 70% of the direct care NH workforce (Department of Health [DoH], 2021) in Australia, and over 60% in the UK (Health Education England, 2015) and the USA, respectively (Sengupta et al., 2022). They play a fundamental role in delivering care and offering essential physical and emotional support to residents in need of assistance with their activities of daily living (Cheng et al., 2019; Gibson et al., 2022). Yet, in Australia, the 2020 Aged Care Workforce Census revealed a 28% turnover among PCWs from November 2019 to November 2020 (DoH, 2021). For the same reporting period, 58% of NH operators reported vacancies for PCWs, with an average of eight vacancies per operator (DoH, 2021). Although PCWs perceive their role as professionally fulfilling (Dill et al., 2013), the high rate of turnover can have serious consequences, including a reduction in the quality of care and an increase in mortality rates for residents (Castle & Engbery, 2005; Spilsbury et al., 2011; Xerri et al., 2019). The Royal Commission into Aged Care Quality and Safety (RCACQS, 2021) placed Australia’s workforce supporting older people under intense scrutiny, revealing that the sector faces substantial challenges in recruiting and retaining direct care workers.

Guiding Framework

Authors of a comprehensive literature review of the research into factors that influence employment intentions identified three overarching domains: individual, organizational, and socioeconomic (Berguig & Abdelbaki, 2022). Individual domain holds significance to employees relating to their personal beliefs and life experiences, both within and outside the workplace. This includes sociodemographic and employment-related characteristics (Berguig & Abdelbaki, 2022). The organizational domain includes aspects that the organization can alter or influence, potentially affecting employees, such as organizational support, supervisor support, communication, and size of the organization (Berguig & Abdelbaki, 2022). Lastly, social-economic domain encompasses aspects that are outside the influence of employees such as economic market conditions exerting influence on rates of unemployment, labor patterns, and supply and demand dynamics (Berguig & Abdelbaki, 2022), thereby shaping work conditions and the policy and regulatory environment in which NH operates. For this review, these three domains were used to guide the analysis of the literature on factors influencing PCWs intentions to leave (ITL).

Previous researchers have predominantly focused on employment intentions and turnover within hospital settings, and principally among nurses (AI Zamel et al., 2020; de Vries et al., 2023; Flinkman et al., 2010; Hayes et al., 2006; Lee & Shin, 2020; Williamson et al., 2022). Systematic review authors have also explored factors influencing employment intentions among nurses (Lee, 2022; Lee & Shin, 2020; Yang et al., 2023). Additionally, researchers have mainly examined these factors across the entire direct care workforce, often encompassing nurses and/or enrolled nurses and Directors of Nursing as representatives of the personal care workforce (Cheng et al., 2019; Chenoweth et al., 2010; Jurij et al., 2021; Thwaites et al., 2023). Understanding PCWs’ perceptions and experiences of factors influencing their ITL is crucial to retaining PCWs. Identification of these factors will enable NH operators and executive staff to understand the workforce needs, facilitating the development of tailored retention strategies as well as reengagement initiatives for former employees, thereby reducing staff turnover and promoting recruitment among the personal care workforce. This article presents the findings of a systematic integrative review of existing literature to identify factors influencing PCWs’ ITL. To address this aim, the research question posed was: What factors influence the intentions of PCWs working in residential aged care to leave?

Design and Methods

After a broad systematic search of the literature exploring factors influencing intention to stay (ITS), ITL, and turnover among PCWs in NHs, an integrative review of the literature relating to ITL was conducted. The review protocol was registered in PROSPERO (registration number: CRD42022361689). The authors opted to present the review findings separately to provide a nuanced understanding of each phenomenon, enabling NH operators, NH workforce, policy-makers, and researchers to develop and implement targeted strategies to reduce turnover and improve employee retention. This review is one of three systematic integrative reviews that investigate factors that influence ITS, intention to leave, and turnover, among PCWs employed in NHs. Specifically, this paper reports the review findings regarding influences on ITL. The findings relating to ITS and actual turnover will be reported and published separately.

Design

An integrative review was chosen for its capacity to incorporate diverse methodologies, offering a comprehensive understanding of the phenomenon under exploration (Whittemore & Knafl, 2005). This approach enabled the synthesis of existing literature, providing insights into the current research landscape (Torraco, 2005), suggesting future research directions, and facilitating the translation of findings into practical application and policy initiatives (Whittemore & Knafl, 2005). The review was guided by Whittemore and Knafl’s (2005) methodological framework and is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRIMSA) guidelines (Page et al., 2021).

Data Sources

The search strategy, developed in consultation with a research librarian, was applied in a preliminary search conducted in May 2022 on the EBSCOhost platform, and refined iteratively where required. Relevant subject headings, keywords, synonyms, and truncations were used. Example key search terms for the broader systemic review included, “personal care worker,” “residential aged care,” “intention to leave,” “intention to stay” and “turnover.” Boolean Operators “AND” and “OR” were used to combine terms and concepts (Supplementary Table S1). The systematic search was conducted in July 2022 and updated in March 2024 in Business Source Complete, CINAHL Complete, Medline Complete, and APA PsycInfo using the EBSCOhost platform and extended to the EMBASE platform (excluding Medline). Additionally, reference lists of included articles were systematically searched for eligible articles using Scopus.

Article Selection

Articles were eligible for inclusion if they were: (a) published after July 1, 1997, aligning with the marketization of Australia’s older persons’ care sector when private NH operators were introduced; (b) published in an English language peer-reviewed journal reporting empirical studies; (c) articles reporting quantitative, qualitative, or mixed-methods studies, with no restriction on study design; (d) exploring the factors influencing ITL among PCWs and reported by the personal care workforce; (e) conducted in a NH. Additionally, articles reporting on participants from various job types and NH facilities were only considered if they presented separate findings for PCWs in NHs.

Eligible Studies

After deduplication, 5,178 articles underwent title and abstract screening. Subsequently, 4,734 articles were excluded based on title and abstract screening for the three phenomena of interest, “intention to stay,” “intention to leave,” and “turnover,” leaving 444 full-text articles for eligibility assessment. Full-text screening led to the inclusion of 20 articles reporting studies of ITL. An additional article was identified for inclusion through screening of reference lists of included articles, resulting in a total of 21 included articles reporting on 20 studies (see Figure 1). All articles were retained, regardless of the quality appraisal results, following guidelines from Hong et al. (2018).

Article Screening

Three authors (B. O’Keefe, E. Yuen, and B. Dow) independently screened references in a two-step process using Covidence systematic review software. The first author screened the titles and abstracts of all references, whereas the second and third authors each screened half of the articles against the inclusion criteria. In the second step, the same authors independently screened the full texts of the articles using the same inclusion criteria. Discrepancies were resolved through verbal discussions to achieve consensus.

Data Extraction

Data on study characteristics, including author, year, location, category of interest (ITS, ITL, and turnover), sampling strategy, response rate, study design, outcome measures, data analysis, findings, and limitations were extracted first from the quantitative studies (Supplementary Table S2), followed by the qualitative and mixed-methods studies (Supplementary Table S3). The data were extracted by one author (B. O’Keefe) and verified for accuracy and confirmability by a second author (S. Perlen) (Miles & Huberman, 1994).

In relation to the quantitative studies, variables that were not significant were not described in the narrative synthesis below but are included in Supplementary Table S4. Data were extracted for the most advanced analysis (e.g., statistical models, if associations were tested and then modeling was undertaken). All variables were scrutinized, and only data for statistically significant variables were extracted. Furthermore, variables that were identified as statistically significant in the preliminary analysis but were not tested in the most advanced analysis stage were also extracted (Table 1 and Supplementary Table S2). The findings were compared and used to structure the narrative synthesis.

Quality Appraisal

The methodological quality of the included articles was assessed using the Mixed-Methods Appraisal Tool (MMAT) version 2018 (Hong et al., 2018) to support analyses of study bias, completeness, and transparency. Two authors (B. O’Keefe, E. Yuen) independently conducted a quality assessment for all studies. The methodological quality of each study was assessed using a scale ranging from 1 to 5, where 1 denoted low quality and 5 denoted excellent quality, signifying all criteria were met (Hong et al., 2018). Disagreements were resolved through discussion to achieve a consensus. Applying the MMAT facilitated the identification of both methodological strengths and weaknesses across the studies included in this review. Table 1 presents a summary of the MMAT scores, with further details available in Supplementary Table S5.

Data Analysis

Adhering to Whittemore’s methodological framework (2005), data were analyzed using a comprehensive classification system, involving five stages: data reduction, data display, data comparison, conclusion drawing, and verification process (Miles & Huberman, 1994; Whittmore & Knafl, 2005) to manage data collected from diverse methodologies. Once variables with statistical significance were extracted from the quantitative data, the qualitative data were thematically analyzed (Braun & Clarke, 2006). Through an iterative process, the quantitative and qualitative data were compared and integrated by identifying recurring patterns (Cooper, 1998) and then organized into codes, subthemes, and overarching themes. The final level of analysis involved categorizing each theme into the applicable domain of individual, organization, and socioeconomic using a data display table. The data displayed in the table were verified for accuracy and confirmability by a second author (S. Perlen) (Miles & Huberman, 1994), after which a narrative summary was produced. A detailed synthesis of the findings is outlined in Supplementary Table S6, illustrating the themes and subthemes that were generated from integrating the quantitative and qualitative data. Heterogeneity across study designs, sample compositions, differing measurement instruments, and variable specifications prevented meta-analyses and cross-study comparisons.

Results

Study Characteristics

The main characteristics of the studies included in the review are presented in Table 1. Among the 21 articles reporting on 20 studies investigating ITL, authors of only two studies provided a definition of ITL, defined as turnover intention in both studies (Supplementary Table S7). Articles authored by Yeatts and Cready (2007) and Yeatts et al. (2010) reported on the same study, which was a mixed-methods study conducted in the United States. Of the 21 included studies, the first article was published in 2003. Subsequent articles were not published until 2007 with up to three articles published per year until 2011. After a hiatus, further articles were published in most years and in small numbers from 2014 to 2023. The majority of studies were conducted in the United States (n = 11). Other countries included Canada (n = 2), Hong Kong (n = 2), Australia (n = 1), France (n = 1), Sweden (n = 1), Taiwan (n = 1), Korea (n = 1), and China (n = 1). Participant sample sizes ranged from 9 to 5,513, and the number of NH facilities included ranged from 1 to 582. Across included studies, participants were predominantly female (73% to 98%). Study designs comprised quantitative cross-sectional (n = 14), quantitative longitudinal (n = 3), qualitative (n = 3), and mixed methods (n = 1). Additional information about the study characteristics is provided in Supplementary Tables S2 and S3.

An examination of the methodologies across the reviewed studies revealed notable influences on the results shaped by qualitative and quantitative approaches. Qualitative studies emphasized emotional, relational, and contextual factors, offering a nuanced understanding of PCWs’ experiences, whereas quantitative studies prioritized measurable sociodemographic, employment and organizational characteristics. Over time, both methodologies have evolved, particularly within quantitative research broadening in scope to include a wider array of independent and dependent variables. This shift has enabled a more comprehensive examination of the factors influencing PCWs’ employment intentions. Although core sociodemographic factors such as age, marital status, and education attainment along with job and pay satisfaction have remained constant focal points, recent studies have included a broader range of workplace conditions, organizational resources and even examined perceptions of older persons’ care services. No significant differences were found in the factors influencing PCWs’ employment intentions across the countries where the studies were conducted.

Narrative Synthesis of Included Studies

Table 2 presents the results of the synthesis, illustrating the themes that were generated and categorized into the domains of individual, organizational, and socioeconomic. The themes are described in the following sections.

Individual

Personal and employment characteristics.—

Employee age was associated with ITL, with younger employees exhibiting a higher ITL (Anderson, 2008) compared with older employees (Backman et al., 2023; Pélissier et al., 2018; Rosen et al., 2011; Sloane et al., 2010; Stearns & D’Arcy, 2008). Specifically, Stearns and D’Arcy (2008) found that employees aged between 46 and 60 years exhibited a lower ITL than those in other age brackets (18–30 years and 31–35 years). Pélissier et al. (2018) found that employees aged 40–50 years were less likely to leave the profession compared to those aged under 30. Marital status was also a factor (Hsu & Yang, 2022), with Stearns and D’Arcy (2008) finding widowed employees demonstrated a higher ITL than those who were separated, divorced and/or never married. Higher levels of education attainment were associated with a greater ITL (Decker et al., 2009; Stearns & D’Arcy, 2008), with Sloane et al. (2010) finding that education levels below high school were associated with lower ITL. Noncitizen status correlated with an increased ITL within 1 year (Sloane et al., 2010), and race also played a significant role, with White employees less likely to express ITL compared to non-White employees (Rosen et al., 2011). Additionally, speaking a native language other than the national language was associated with lower ITL (Backman et al., 2023).

Part-time employees exhibited a lower ITL in one study (Castle et al., 2007), whereas full-time employees were associated with reduced ITL in another (Stearns & D’Arcy, 2008). Tenure and experience yielded mixed results. In one study, longer tenure was associated with a lower ITL compared to shorter tenure (Rosen et al., 2011). Similarly, Yan et al. (2024) found that increased time in the industry was associated with a lower ITL. In another study, those with more than 3 years of experience had lower ITL than those with less than 2 years of experience (Decker et al., 2009). Sun et al. (2023) found that employees with 5–10 years of experience showed higher ITL compared to those with less than 5 or more than 10 years of experience. In another study, less than 5 years of experience was associated with a higher ITL compared to 6 or more years (Sloane et al., 2010). Additionally, having a greater number of previous jobs was associated with higher ITL (Castle et al., 2007).

Employee health and well-being.—

Deteriorating physical health (Chon & Kim, 2020), physical and emotional fatigue (Chon & Kim, 2020; Gao et al., 2015; Sabi Boun et al., 2023), acute and chronic fatigue (Yan et al., 2024), spinal and lower-limb complaints (Pélissier et al., 2018), and adverse physical health (Chon & Kim, 2020) were all associated with higher ITL. Furthermore, employee mental health, including burnout (Yan et al., 2024) and emotional distress (Rosen et al., 2011), were associated with higher ITL. Greater acceptance of video gaming technology moderated the association between exhaustion and ITL, reducing ITL (Chen et al., 2021).

Employee satisfaction and engagement.—

Employee satisfaction and engagement were inversely associated with ITL. Low job satisfaction was associated with increased ITL (Anderson, 2008), whereas higher levels of job satisfaction (Hsu & Yang, 2022; Rosen et al., 2011) and higher overall job satisfaction were associated with lower ITL (Decker et al., 2009). Higher intrinsic job satisfaction levels (Decker et al., 2009) and high scores in job satisfaction subscales of work schedules, training, and rewards (Castle et al., 2007) along with work engagement (Hsu & Yang, 2022) were associated with lower ITL (Hsu & Yang, 2022). Sloane et al. (2010) identified an association between not recommending one’s job to a family member or friend and higher employee ITL.

Career development and advancement.—

Gao et al. ’s (2015) study of nursing assistants working in NH and undertaking a nursing course found limited opportunities for career progression, a preference to work in acute care upon course completion, limited acquisition of on-the-job knowledge and experience, and reduced employee capability, all of which were linked to increased ITL. Additionally, active engagement in classroom discussions sponsored by an employer was associated with a decreased ITL (Stearns & D’Arcy, 2008). Elevated levels of employee capability (Gao et al., 2015) and career adaptability (Sun et al., 2023) were associated with reduced ITL. Sun et al. (2023) also demonstrated that psychological capital mediated the relationship between career adaptability and ITL, underscoring the significance of specific dimensions of psychological capital, including concern, control, curiosity, and confidence, in reducing ITL.

Organizational

Workplace conditions and working environment.—

Employees grappling with working conditions, workload expectations, the nature of their work (Chon & Kim, 2020), limited organizational resources (Gao et al., 2015), high levels of extrinsic effort pertaining to pressures placed on employees in their work environment (Pélissier et al., 2018), and disparities in workload management (Sabi Boun et al., 2023) were associated with higher ITL. These challenges were compounded by persistent workforce shortages (Chon & Kim, 2020; Gao et al., 2015). However, higher staffing levels of registered nurses and nurse aides (Castle et al., 2007), along with higher levels of coworker support (Zang et al., 2014) and access to in-person specialist staff (Gao et al., 2015) were associated with lower ITL.

Several workplace characteristics were associated with reduced ITL, including access to lifting devices (Stearns & D’Arcy, 2008), employment in facilities with over 100 beds (Sloane et al., 2010), availability of employee benefits such as paid vacation/personal days (Rosen et al., 2011; Stearns & D’Arcy, 2008) and provision of paid sick leave and health insurance (Decker et al., 2009) were associated with reduced ITL. Additionally, availability of training when needed (Yeatts et al., 2010) was associated with lower ITL, whereas employment in an organization affiliated with a chain of organizations (Decker et al., 2009) was associated with higher ITL.

Effective leadership and supervision.—

Superior supervisor qualities and respect from supervisors (Stearns & D’Arcy, 2008), high levels of supervisor/management support (Chon & Kim, 2020; Zhang et al., 2014), higher perceived leadership quality (Backman et al., 2023), and effective supervision (Parsons et al., 2003) were associated with reduced ITL. Bethell et al. (2018) identified a relationship between supervisory support and ITL that was mediated by job satisfaction. This indicated that higher levels of supervisory support and job satisfaction were associated with reduced ITL. Additionally, Bethell et al. (2018) reported the relationship between supervisory support, job satisfaction, and intent to leave was contingent upon the happiness levels of employees.

Conversely, negative professional interactions with managers (Chon & Kim, 2020), managers’ failure to keep employees informed (Parsons et al., 2003), supervisors who did not treat all employees equally (Sloane et al., 2010), challenging relationships with nursing peers (Sabi Boun et al., 2023), and employees perceiving their relationships with management as “moderately satisfactory” or “unsatisfactory” (Pélissier et al., 2018) were associated with higher ITL.

Employee engagement and empowerment.—

Dissatisfaction with work challenges (Sloane et al., 2010), increased absenteeism (Yeatts & Cready, 2007), insufficient workplace respect (Zhang et al., 2014), and job insecurity (Pélissier et al., 2018) were associated with elevated ITL. Restricted decision-making (Sabi Boun et al., 2023), diminished decision-making authority (Zhang et al., 2014), and limited personal growth opportunities (Parsons et al., 2003) were also associated with increased ITL.

Residential connections and challenges.—

Sufficient time to assist residents with activities of daily living (Decker et al., 2009) and weak attachment to residents among employees (Pélissier et al., 2018) were associated with higher ITL. Additionally, increased ITL was associated with resident-staff abuse (Sabi Boun et al., 2023), resident-to-resident aggression, resident-to-staff aggression (Yan et al., 2024), elevated aggressive incidents from residents (Snyder et al., 2007), and “moderate” and “severe” intellectual deterioration of residents, related to those nearing the end of life (Pélissier et al., 2018).

Compensation.—

Perceived inadequate income (Chon & Kim, 2020) and dissatisfaction with pay/wages, and benefits provided by the employer (Sloane et al., 2010) were associated with increased ITL. Conversely, higher levels of pay satisfaction, particularly among those who were somewhat satisfied (Decker et al., 2009), and higher hourly wages (Stearns & D’Arcy, 2008) were associated with decreased ITL.

Social-economic

Community and societal perceptions.—

Few socioeconomic factors were identified. Castle et al. (2007) found that a greater concentration of NH in five counties was associated with a decreased likelihood of ITL. Conversely, challenging perceptions of the CNA workforce (Sabi Boun et al., 2023) and negative societal perceptions (Gao et al., 2015), particularly when employees perceived their role as undervalued, were associated with higher ITL.

Discussion

Given the multidimensional workforce issues encountered by NH operators, understanding the factors that influence PCWs ITL represents a crucial step towards efforts to mitigate these workforce issues and build a strong and stable workforce. In this review, 21 articles were included, comprising qualitative, quantitative, and mixed-methods studies that identified factors influencing PCWs’ ITL. A cross-sectional design, investigating ITL through predetermined variables, was used in the majority of studies included in this review. Heterogeneity in the included studies precluded the conduct of meta-analyses, primarily due to variations in objectives, methodologies, and measurement tools.

Studies from nine countries met the eligibility criteria. However, differences exist in the approaches to providing services and supports to older persons across these countries. These differences include legal and regulatory frameworks, funding mechanisms, financing models, operational requirements, quality standards, staffing levels, societal preferences regarding the delivery of services and supports to older people, along with disparities in each country’s political systems, economic conditions, social institutes, demographics and cultural and religious norms (Dyer, van den Berg, et al., 2019, Dyer, Valeri, et al., 2019; Eager et al., 2019; McMaughan et al., 2020; Zimmer et al., 2016).

Spanning more than 26 years, the small number of studies included in this review highlights the scarcity of empirical investigation on this subject. Categorizing the factors associated with ITL into the individual, organizational, and/or socioeconomic domains, not only enabled the examination of similarities and differences across these domains but also contributed to the existing body of knowledge. In this review, most of the factors extracted were categorized into the organizational (n = 17 studies) domain, indicating the pronounced influence of the organization on an employee’s ITL.

In this review, the factors that were categorized into the individual domain encompassed a myriad of sociodemographic and employment-related characteristics that intersect to shape employees’ ITL. The findings of this review highlight the influence of sociodemographic characteristics such as age, marital status, education, and race, along with employment-related characteristics such as work experience, tenure, and job status as pivotal factors in predicting ITL among less experienced PCWs. However, disparities in the influence of these characteristics existed across different countries, indicating the need to consider the role of these factors as predictors of turnover intentions in different contexts.

Mirroring a growing global population of older people (Yan et al., 2024), is an aging workforce; however, the findings of this review generally indicate that younger employees exhibit a greater ITL than their older counterparts (Anderson, 2008; Backman et al., 2023; Pélissier et al., 2018; Rosen et al., 2011; Sloane et al., 2010; Stearns & D’Arcy, 2008). Higher ITL among younger and less experienced PCWs points to a critical turnover risk for NH operators targeting strategies to retain newer employees and preventing early-career burnout. ITL was also found to be associated with deterioration in PCWs’ physical and mental health (Chon & Kim, 2020; Gao et al., 2015; Yan et al., 2024; Pélissier et al., 2018; Sabi Boun et al., 2023). Research has revealed that, among aging workers, an increase in work-related injuries was associated with an increased frequency of work requirements, excessive physical effort, lifting heavy loads, and stooping/kneeling/crouching (Baidwan et al., 2018) mirroring the requirements of a PCW’s role. PCWs experience physical and psychological aggression from residents (Pélissier et al., 2018; Sabi Boun et al., 2023; Xiao et al., 2021; Yan et al., 2024). Researchers have found that compared to other health care settings, PCWs working in NHs are 23 times more likely to encounter aggressive behavior from residents (Xiao et al., 2021). Therefore, strategies to modify the work environment such as workload expectations, leadership support, and staffing levels along with promoting employee diversity are important. To reduce ITL, NH operators should prioritize preventive health and safety measures, implement workplace violence prevention strategies, and modify the work environment.

An increasing number of culturally and linguistically diverse PCWs are working in NHs (DoH, 2021), with positions often filled by migrant workers to address workforce shortages (Kumar et al., 2022), highlighting the importance of the review finding that noncitizen status and race were associated with increased ITL (Backman et al., 2023; Rosen et al., 2011; Sloane et al., 2010). Immigrant PCWs are more prone to expressing ITL and experiencing disrespect from residents and their families (Sloane et al., 2010). Previous studies have supported these findings, showing migrants experienced racial discrimination, prejudices (Berdes & Eckert, 2001; Cuesta et al., 2016; Goel & Penman, 2015), and lack of respect (Berdes & Eckert, 2001). Therefore, fostering a supportive work environment that values diversity and inclusion and promotes cultural competence and safety among all staff members, residents, and family members, recognizing and respecting the unique backgrounds and experiences of migrant PCWs, is essential.

In this review, age, marital status, education, and race were identified as the most significant sociodemographic characteristics related to ITL, whereas work experience, tenure, and job status were the key employment-related characteristics. However, the findings of this review highlight variability in the direction of influence of these factors on ITL, which may be attributed to the country in which the study was conducted. Therefore, sociodemographic and employment-related characteristics that are important in one country relating to ITL may be different in another country.

Based on the findings of this review, job satisfaction had a significant influence on PCWs’ ITL (Anderson, 2008; Castle et al., 2007; Decker et al., 2009; Hsu & Young, 2022; Rosen et al., 2011), congruent with another study (Choi & Johantgen, 2012), that showed an inverse relationship between job satisfaction and ITL. Castle et al. (2007) reported the pivotal role of job satisfaction, particularly in terms of satisfaction with work schedules and training in reducing PCWs’ ITL. Addressing these facets of job satisfaction may be a promising avenue for NH operators to diminish turnover intentions.

A substantial number of factors relating to employee ITL were within the organization domain, underscoring the important role that organizational factors play in ITL. Although most factors identified in this review are modifiable, such as work environment, workload management, staffing levels, leadership support and supervision, training and development, and compensation, some factors are not modifiable, including the nature of the work, workforce shortages, and facility characteristics, such as being affiliated with a chain and facilities over 100 beds. Nursing home operators should review and, where required, implement necessary changes in modifiable factors to enhance employee satisfaction and retention. Although nonmodifiable factors cannot be changed such as the sociodemographic characteristics of PCWs, NH operators can mitigate their impact through supportive strategies and training, fostering a more supportive and satisfying work environment that reduces turnover intentions.

The findings of this review shed light on the impact of supervisor qualities and leadership practices on employees’ ITL. Effective leadership and supervision play a crucial role in mitigating ITL. Positive supervisor qualities, supervisor support, effective supervision, and leadership practices are linked to lower ITL (Backman et al., 2023; Chon & Kim, 2020; Parsons et al., 2003; Stearns & D’Arcy, 2008; Zhang et al., 2014). Conversely, negative professional interactions with managers, unsupportive peer relationships, and dissatisfaction with management increase ITL (Chon & Kim, 2020; Parsons et al., 2003; Sloane et al., 2010; Pélissier et al., 2018). Leadership that engages and empowers PCWs, values, and acknowledges PCWs’ contribution, and promotes decision-making authority was found to be associated with higher ITL. This finding is congruent with other research findings where empowerment practices were positively associated with building a strong and stable workforce, reducing PCWs’ employment intentions of leaving (Berridge et al., 2018).

This review revealed a paucity of studies investigating factors related to the socioeconomic domain. Societal perceptions of services and supports delivered to older people (Gao et al., 2015; Sabi Boun et al., 2023) were associated with increased ITL, aligning with existing research that showed the perception of unappreciation and undervaluation among PCWs plays a substantial role in turnover, originating from policies and practices that engender sentiments of personal and professional disregard (Bowers et al., 2003). Addressing societal misconceptions and enhancing community support may reduce ITL and foster a more supportive environment for PCWs.

Like other developed countries (excluding the USA), Australia’s services and supports delivered to older people predominantly rely on government funding. The federal government assumes a crucial and multifaceted role in policy development, regulation, and funding to ensure high-quality care for older Australians. Demonstrating its extensive authority, the Australian government has the capacity to enforce federal legislation and implement policies to tackle the enduring problem of high turnover rates. An illustration of this authority was evident in the 2023 Australian federal budget, which allocated an additional $11.3 billion to support the Fair Work Commission’s resolution to increase by 15% the wages of NH workers covered by the Aged Care Award and the Nurses Award (Department of Health and Aged Care, 2023). In Australia, there is a much larger role for the government to play in shaping the policy and regulatory environment for PCWs to shift their employment intentions from an ITL to employee retention.

The absence of a universally accepted definition of “ITL” presents a significant challenge to benchmarking and undertaking comparability testing in research. Although all the studies included in this review investigated ITL, only two provided an explicit definition (Chon & Kim, 2020; Hsu & Yang, 2022), highlighting the absence of a consistent definition of the term “intention to leave.” Establishing a clear, accepted, and consistently used definition of ITL is crucial for advancing understanding, facilitating measurement and comparisons, enabling cross-study synthesis, and informing evidence-based practices.

Numerous studies were excluded because they did not include PCWs in their sample. These studies relied on organizational representatives’ perceptions of PCWs’ intentions rather than including the workers themselves. An example is the Australian Government’s Aged Care Workforce Census, which offers a periodic snapshot of the workforce. Notably, the latest Census (DoH, 2021) is based on data collected from management without direct employee involvement. This review focused on PCWs in NHs, examining factors that influence PCWs’ ITL. Direct insights from these workers regarding their employment intentions are vital to inform strategies to retain the largest segment of the Australian direct care workforce.

An employees’ ITL is prompted by either a personal or work-related event or by a gradual withdrawal process firstly commencing with job resignation and potentially culminating in leaving the profession entirely (Morrell, 2005). By proactively investigating and monitoring PCWs’ employment intentions through various mechanisms such as annual reviews or employee pulse surveys, NH operators can identify and address factors contributing to ITL, thereby mitigating the withdrawal process. This review delineates factors influencing PCWs’ ITL across individual, organization, and socioeconomic domains, facilitating NH operators’ comprehension of the influences on the employment intentions of the largest segment of their workforce, PCWs. Such knowledge could be used to inform strategies to facilitate a shift from an employee’s ITL to a commitment to stay within both the organization and the profession.

Strengths and Limitations

The strength of this study lies in its systematic, comprehensive review of literature, including diverse methodologies without restrictions on study design. Searching several databases, coupled with the use of a broad range of search terms, significantly broadened the scope and comprehensiveness of the articles included in this review. However, despite the implementation of a comprehensive search strategy, there is the possibility that certain studies may have been overlooked, thereby potentially limiting the identification of statistically significant variables associated with ITL. The conclusions drawn from this review predominantly rely on the findings of studies using cross-sectional research designs. Additionally, due to the variety of outcome measurement tools used by authors of the included studies, comparing the results posed a significant challenge, and therefore, the results of this review should be interpreted with caution.

Given that the study designs described in most of the included articles were cross-sectional, it is possible that some PCWs expressed their ITL without subsequently leaving their employment. In addition, as surveys were predominantly used in the included studies, understanding of PCWs’ perspectives and experiences that influence their employment intentions, as well as turnover, is limited. Finally, relying exclusively on findings from articles published in English may have introduced language bias, due to the omission of valuable research reported in other languages.

Implications and Conclusion

In this review, the findings of 20 studies, published in 21 articles over a 26+ year period, were synthesized to identify factors influencing PCWs’ ITL. These were categorized into individual, organizational, and socioeconomic domains. Most factors were categorized into the organization domain, indicating the significant influence the organization has on influencing employees’ ITL. For NH operators and policy-makers, the development of a systematic approach to address organizational factors offers a promising avenue for mitigating the persistent issue of high PCW turnover to build a strong and stable workforce. However, the findings of this review reveal a need for greater focus in future research on the impact of social, economic, and political forces on working conditions and ensuring the safety, health, and welfare of employees in both research and policy initiatives (Sorensen et al., 2021).

To advance research in this field and offer evidence-based insights into the factors shaping the ITL of PCWs in NHs and given the limited number of qualitative studies in the field, we recommend that future exploratory research delve into the perceptions and experiences of PCWs that underly the factors that influence ITL. This exploration should involve both current and former PCWs in NH, including those who have left the sector, rather than relying solely on participants reporting ITL as a proxy for turnover as observed in previous research.

Funding

This work was supported by the Postgraduate Research Scholarship of Deakin University. The research did not receive any specific grants from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of Interest

None.

Data Availability

Data may be available from the corresponding author upon request.

Acknowledgment

The authors would like to thank Louisa Sher, Scholarly Services Librarian at Deakin University, for providing her expertise in the development of the search strategy and guidance on the databases that were ultimately selected to run the search strategy.

Author Contributions

The protocol was developed by B. O’Keefe and reviewed by E. Yuen and A. M. Hutchinson. B. O’Keefe led the document search, article selection, data extraction, quality review, data synthesis, and manuscript preparation. A. M. Hutchinson and E. Yuen contributed to the overall study design. All authors independently screened articles for selection, and A. M. Hutchinson and E. Yuen independently conducted quality review of the included articles. A. M. Hutchinson and S. Perlen provided critical feedback on data extraction, results, and interpretation. The manuscript was drafted by B. O’Keefe and substantial critical contributions were made by all authors. All the authors reviewed, revised, and edited the final version of the manuscript.

Ethical Approval

Not applicable.

Preregistration Statement

This study was not preregistered.

References