According to Mueller (2000), the skill mix within the team is one component to consider in a staffing model. Skill mix includes staff with assorted skills, qualifications, experience, proficiencies, and scope of practice (Backhaus et al. 2014). For example, a diverse skill mix may include nurses with degrees, newly qualified nurses, and caregivers who are not nurses. The skill mix is essential in LTCFs because the team needs higher levels of skill as residents become more frail (Boscart et al. 2018; Koopmans, Damen & Wagner 2018).

In South Africa, the skill mix includes registered nurses (RNs), comprising professional and general nurses (Republic of South Africa 2022), enrolled nurses (ENs), enrolled nurse assistants or auxiliary nurses (ENAs) (Republic of South Africa 2005), and caregivers (Republic of South Africa 2010a). Professional nurses hold 4-year bachelor’s degrees, provide comprehensive care by applying scientific principles across diverse service delivery contexts, and are allowed to manage the nursing department of an LTCF. General nurses have 3-year diplomas and provide general nursing care. General nurses, for example, promote public health, prevent disease, and address health problems by delivering nursing care to the overall population. When a healthcare unit operates within a larger facility, general nurses may manage a subunit but are not allowed to oversee the total nursing department of an LTCF (Republic of South Africa 2022). The ENs have 2 years of training and provide basic nursing care under RN supervision; however, this legacy qualification was phased out. The ENs with legacy qualifications remain part of the South African workforce and are therefore included in this review. The ENs are responsible, for example, for monitoring residents’ vital signs and reactions to medication, assisting with procedures, wound care, and hygiene (South African Nursing Council 1984). The ENAs have 1 year of training. A professional or general nurse delegates basic nursing care to ENAs. Basic nursing care includes assisting healthcare users with activities of daily living in accordance with prescribed standards of care, to promote and maintain their health status (Republic of South Africa 2022). A caregiver in South Africa includes anyone performing caregiving tasks in the formal sector but is not a qualified nurse (unlicensed staff) (Republic of South Africa 2006). Accredited training programmes are available for South African caregivers, such as the unit standard ‘Provide care to a frail person’. This programme comprises 12 credits, lasts 120 h, and is a National Qualification Framework (NQF) level 1 qualification. Alternatively, a longer community care programme is available at NQF levels 1, 2, or 3 (South African Qualifications Authority 2022). Caregivers assist residents with activities of daily living, including meal support, mobility, hygiene, and toileting (Republic of South Africa 2015).

There is a global shortage of health workers, and more than 50% of this shortfall comprises RNs and midwives (WHO 2022). This shortfall often leads to the replacement of RNs with nursing assistants to save costs (Shin & Hyun 2015). Substituting RNs with less-qualified staff may increase the number of less-qualified nurses in the skill mix, potentially leading to poorer resident outcomes (Harrington et al. 2020), while more RNs in the skill mix result in, for example, fewer urinary tract infections (Cho et al. 2020).

Staffing models should also include staffing levels, i.e. the quantity of caregivers and nurses (including all categories) available in the LTCF for resident care (Butler et al. 2019). Staffing levels are frequently expressed as staff-to-resident ratios, or as hours of care per resident day (HPRD). Authors found that residents’ activities of daily living improved with higher nurse staffing levels, resulting in lower hospitalisation rates (Harrington et al. 2020), and fewer adverse events and fewer deaths (Griffiths et al. 2018). Conversely, inadequate nurse staffing levels may compromise resident safety, lead to higher nurse workloads, burnout, and job dissatisfaction (Al-Jumaili & Doucette 2018; Griffiths et al. 2018).

According to Mueller (2000), staff allocation is another component to consider in a staffing model. Therefore, along with having sufficient nurses and caregivers and the right skill mix (different categories of nurses and caregivers), the LTCFs must assign staff in accordance with the residents’ acuity (Beckett et al. 2021; Butler et al. 2019), as the residents’ acuity determines the type of nursing care needed (Brennan & Daly 2009; Juvé-Udina et al. 2019). Long-term care facilities must also consider the legal parameters within which each nurse category may practice and which tasks they may perform (Republic of South Africa 2022), as determined by the country’s regulatory authority or statutory professional council, to ensure adherence to legislation. Caregivers’ job descriptions must comply with applicable legislation, and LTCFs must ensure caregivers are competent to perform assigned tasks.

The LTCFs face staffing deficiencies worldwide (Sato et al. 2017). These shortages lead to lower staffing levels and inadequate staff mixes that do not meet the recommended standards for LTCFs, especially regarding RNs (Brühl, Planer & Hagel 2018). With inadequate staffing and a diluted skill mix, LTCFs struggle to assign competent staff, especially when residents require higher levels of care (Estabrooks et al. 2020). Likewise, South African LTCFs also face nurse shortages and an inappropriate skill mix (Republic of South Africa 2010b).

The objective of the scoping review was to explore the implementation of nurse and caregiver staffing models in resource-rich and resource-constrained LTCF settings. The aim was to map the available literature on the implementation of LTCF nurse and caregiver staffing, focusing on skill mix, staffing levels, and staff allocation. The research question was: ‘What are the characteristics of staffing models implemented in resource-rich and resource-constrained contexts?’ The scoping review was guided by three sub-questions: ‘(1) How do LTCFs implement staffing models regarding staffing levels, skill mix, and allocation of tasks? (2) Is allocating tasks aligned with the nurses’ scope of practice and the caregivers’ job descriptions as described in the relevant country’s legislation? and (3) Is the allocation of staff aligned with the acuity of the residents?’

This scoping review was conducted using the JBI methodological framework (Joanna Briggs Institute 2015) and the Peer Review of Electronic Search Strategies (PRESS) (eds. Aromataris et al. 2024). A priori protocol was developed to guide the scoping review (unpublished).

The eligibility criteria for selecting sources were based on the Participants, Concept, and Context (PCC) framework (Peters et al. 2020). Evidence was included if caregivers (or their counterparts in other countries) or any category of nurses, regardless of ethnicity, age, or gender, provided direct care to residents in LTCFs. Studies were considered if they contained at least two of the three central concepts identified by Mueller (2000), i.e. staffing levels, skill mix, and staff allocation according to residents’ acuity. Studies were included if they were conducted in formal long-term care settings for older people, thus, over 60 years old (WHO 2022), in either resource-rich or resource-constrained contexts. Studies within community settings (e.g. home-based care), studies conducted in languages other than English, studies for which the full articles were unavailable, and studies in which most data were collected before 2010 were excluded. The researchers decided to include data collected after 2010 because South Africa published its staffing model only in that year (Republic of South Africa 2010a) and to ensure the inclusion of more updated literature.

Sources included primary research studies employing quantitative and qualitative methods, as well as grey literature (national and provincial legislation, acts, guidelines), policies, conference articles, dissertations, and theses. Scoping and systematic reviews were excluded because they contained synthesised evidence; however, Google Scholar was used to search their reference lists for additional articles.

A qualified librarian assisted with developing a search strategy to identify relevant sources using Medical Subject Headings (MeSH) terms, Boolean operators, and the keywords in the abstract. Pilot screening was performed to ensure that the keywords produced appropriate results. After that, a three-step search process was followed (Peters et al. 2020), where the first-level limited search involved searching for text words in titles, abstracts, and indexed keyword lists in PubMed and MEDLINE. Table 1 provides an example of search strings for PubMed and MEDLINE.

After the initial search, the strategy was adapted. Two reviewers independently used the same search strings to conduct a comprehensive search in PubMed and MEDLINE, CINAHL, Cochrane Library–Wiley, and Sabinet African Journals. The full search strings for all the databases are available as Online Appendix 1. Lastly, the researchers manually searched the reference lists of the selected studies for additional studies, which were then searched on Google Scholar. The original search took place in December 2022 and a final search was conducted in December 2024.

JBI’s three-step process was used (Joanna Briggs Institute 2015). Two reviewers (ECB and JCB) independently screened the titles and abstracts manually against the inclusion criteria (Figure 1). After the initial title and abstract search, the two reviewers retrieved the full articles and conducted independent assessments. Discussions between the two reviewers resolved any disagreements about whether studies should be included or excluded from the review. A third reviewer (MMvdH) conducted a randomised assessment to verify the accuracy of the selection process and inclusion criteria.

A custom tool was used for data abstraction, and two reviewers then randomly tested it on three articles to revise it as needed (available as Online Appendix 2). The general information was recorded in a tabulated format (Joanna Briggs Institute 2015) and included the charting of authorship, publication year, country of origin, income level (high, medium, or low-income country), data sources used, study design, sample description and size, study aims, settings, and outcomes. The researchers did not critically appraise the included studies because the purpose of scoping reviews is to map available evidence irrespective of quality (Peters et al. 2020). Themes were used to organise the results.

Trustworthiness was ensured through a detailed methodology description to promote transparency, enhance the credibility and reliability of the findings, and enable readers to evaluate the transferability of the scoping review to their own settings (Batten & Brackett 2021). A research question and sub-questions were developed, eligibility criteria were set, keywords were identified, and the search strategy was outlined. The inclusion of various electronic databases facilitated method triangulation (Batten & Brackett 2021). Gathering data from varied sources facilitated data triangulation (Lincoln & Guba 1985). The researchers practised reflexivity by considering their preconceived ideas and assumptions (Dwyer 2009), and carefully reflected on whether the data related to the research question and the study’s purpose, as well as with the concepts of staffing levels, skill mix, and staff allocation aligned with residents’ acuity, nurses’ scope of practice, and caregivers’ job descriptions.

An application for full ethical approval was made to the Stellenbosch University’s Health Research Ethics Committee and ethics consent was received on 18 November 2022. The ethics approval number is S22/10/216.

Figure 1 contains a PRISMA flow chart to display the study selection and screening process, and reasons for excluding studies. Through four database searches, 1115 studies were identified. A manual search of the reference lists identified an additional study. The final sample included 20 studies.

Table 2 presents the characteristics of the studies. The studies were published between 2012 and 2022. All the data were collected in LTCFs in high-income countries. Data from low- and middle-income countries were unavailable. Nevertheless, resource constraints may also occur in LTCFs in higher-income countries, regardless of income level, making the results from these studies applicable. The studies included the United States (n = 8, 40%) (Chappell, Kirkham & Seitz 2022; Gorges & Konetzka 2020; Lerner 2013; Schnelle et al. 2016; Shin 2013; Shin, Park & Huh 2014; Yang et al. 2021; Yoon et al. 2022), South Korea (n = 6, 30%) (Kim & Han 2018; Lee et al. 2022; Shin 2018, 2019; Shin & Hyun 2015; Shin & Shin 2019), Canada (n = 1, 5%), (McCloskey et al. 2015) Germany (n = 2, 10%) (Zimmermann & Pfaff 2018; Zirves, Demirer & Pfaff 2021), and the Netherlands (n = 1, 5%) (Tuinman et al. 2016). One study collected data from the United States, Canada, England, Germany, Norway, and Sweden (n = 1, 5%) (Harrington et al. 2012), while another collected data from the United States and South Korea (n = 1, 5%) (Lee, Shin & Harrington 2015).

Eleven studies (n = 11, 55%) used cross-sectional designs (Chappell et al. 2022; Gorges & Konetzka 2020; Harrington et al. 2012; Lerner 2013; McCloskey et al. 2015; Shin 2018; Shin et al. 2014; Shin & Shin 2019; Tuinman et al. 2016; Yang et al. 2021; Yoon et al. 2022), four studies (n = 4, 20%) used correlational designs (Schnelle et al. 2016; Shin 2013; Zimmermann & Pfaff 2018; Zirves et al. 2021), three studies (n = 3, 15%) were longitudinal (Kim & Han 2018; Shin 2018, 2019), and two studies (n = 2, 10%) were descriptive (Lee et al. 2015, 2022). Populations included RNs, licensed practical or vocational nurses (LPNs or LVNs), nurse assistants (NAs), certified nursing assistants (CNAs), care workers, and residents. Most studies (n = 15, 75%) collected secondary data from government staffing websites, payroll-based journal data, medical insurance websites, government documents, research articles, reports, and quarterly data from nursing homes (Chappell et al. 2022; Gorges & Konetzka 2020; Harrington et al. 2012; Kim & Han 2018; Lee et al. 2015, 2022; Lerner 2013; Schnelle et al. 2016; Shin 2013, 2019; Shin & Shin 2019; Yang et al. 2021; Yoon et al. 2022; Zimmermann & Pfaff 2018; Zirves et al. 2021). Five studies (n = 5, 25%) collected primary data (McCloskey et al. 2015; Shin 2018; Shin & Hyun 2015; Shin et al. 2014; Tuinman et al. 2016).

The synthesised results answered the first research question: ‘How do LTCFs implement staffing models regarding skill mix, staffing levels, and task allocation according to residents’ acuity?’

Countries’ national staffing standards varied regarding their skill mix (Harrington et al. 2012; Lee et al. 2015; McCloskey et al. 2015; Shin 2019; Shin & Shin 2019; Zimmermann & Pfaff 2018; Zirves et al. 2021). According to studies conducted in 2018 and 2021, Germany required that half of the care staff must be RNs, spread over 24 h but had no additional specific requirements for the rest of the staff (Zimmermann & Pfaff 2018; Zirves et al. 2021). Norway, Sweden, and England did not specify staffing standards (Harrington et al. 2012). In 2019, South Korea required one RN or CNA per 25 residents for nursing homes exceeding 30 beds, and one RN or CNA for nursing homes with fewer than 10 beds and up to 30 beds. The remaining staff comprised care workers (Shin 2019; Shin & Shin 2019). Studies conducted between 2012 and 2015 reported that federal United States standards required one RN to be on duty for 8 h every weekday, with an RN and an LPN covering the remaining two 8-h shifts. A director of nursing must be full-time employed, and there must be sufficient staff to ensure residents’ well-being. The different states within the country may set standards that exceed the federal government’s staffing standards (Harrington et al. 2012; Lee et al. 2015).

Authors reported in 2012 and 2015 that Canada’s different provincial governments determined their staffing standards (Harrington et al. 2012; McCloskey et al. 2015).

In the studies, the proportion of RNs among the total nurse and caregiver staff ranged from 13.2% (Yang et al. 2021) to 56.6% (Zimmermann & Pfaff 2018). In South Korea, an RN or CNA may be used interchangeably, with the proportion of RNs or CNAs indicated in the study as 15.87% (Lee et al. 2022). Yang et al. (2021) indicated in 2021 the proportion of LPNs in their United States study as 23.4%, while McCloskey et al. (2015) in Canada in 2015 indicated 40%. The proportion of caregivers ranged from 40% in Canada (McCloskey et al. 2015) to 84.13% in South Korea (Lee et al. 2022).

Registered nurses undergo between 3 and 4 years of training (Zirves et al. 2021). Compared to other nurse categories and caregivers, RNs use higher levels of clinical judgement, decision-making, and critical thinking when providing care (Chappell et al. 2022). In the United States, the category of LPN completes a 1-year full-time nursing course (Yang et al. 2021). South Korean CNA category’s job description seems comparable to LPNs in the United States (Lee et al. 2022). Certified nursing assistants in South Korea do not train at nursing colleges but attend private nursing institutes or occupational high schools (Shin 2019). Care workers in South Korea receive 240 h of home care training (Shin 2018, 2019) and have job scopes comparable to those of the United States CNAs (Lee et al. 2022). Unlicensed staff, such as care workers without nursing qualifications, are often used (Shin 2018, 2019).

Staffing levels, i.e. the number of staff required to provide care HPRD, also varied between provinces and countries. In 2011, the HPRD was 1.9 in Alberta, Canada, and 2.0 in Saskatchewan, Canada. The HPRD in Florida was 3.9 and 2.56 in South Carolina in the United States (Harrington et al. 2012). In Germany, residents highly dependent on care require ≥5 HPRD for assistance or ≥4 HPRD for basic care (Zimmermann & Pfaff 2018). In South Korea, national standards stipulate that for every 25 residents, there must be one CNA or RN, and that the care worker-to-resident ratio be maintained at 1:2.5 (Lee et al. 2022; Shin 2019). England’s 2011 regulations did not specify LTCF staffing levels; however, the registration authority set those levels based on residents’ needs (Harrington et al. 2012).

Chappell et al. (2022) conducted a cross-sectional study and found that across 10 436 United States LTCFs, the mean staffing level for RNs, LPNs, and nursing assistants was 3.69 HPRD. Harrington et al. (2012) compared staffing standards across countries in 2022, with the HPRD values as follows: United States: 3.9 HPRD, Canada: between 2.1 and 3.3 HPRD, England: 4.26 HPRD, and Sweden, with the highest staffing levels at 5.19 HPRD. A comparison of nurse staffing between South Korea and the United States showed that the United States RNs provided an HPRD of 0,63, versus 0,47 for RNs or CNAs in South Korea (Lee et al. 2015). In a study conducted across seven Canadian LTCFs, the staffing level averaged 3.1 HPRD (McCloskey et al. 2015).

Higher total staffing levels seemed to be linked to lower prescription rates of antipsychotic medications, better ability of staff to address residents’ behavioural and psychological symptoms (Chappell et al. 2022), and to allow RNs more time to explore possible causes of disruptive behaviours (Yoon et al. 2022). The probability that an LTCF had only one COVID-19 case was not affected by the number of staff, although more staff were associated with better COVID-19 case control (Gorges & Konetzka 2020). Lee et al. (2022) further found that a higher care workers-to-residents ratio was associated with more infections and higher mortality rates; thus, when more care workers were employed, but fewer RNs, infections and mortality rates were higher. Low staffing levels were associated with high workloads (Yoon et al. 2022) and the inability of the staff to assist residents with their daily living activities (Schnelle et al. 2016).

Four studies described the effect of increased staffing levels on resident outcomes. A 1-h increase in overall staffing HPRD led to a 0.75% reduction in the improper prescribing of antipsychotic medication. When the RNs’ HPRD was increased by 1 h, it led to a decrease of 2.25% in the improper prescribing of antipsychotic medications (Chappell et al. 2022), a 5.72% decrease in residents with bed rest (Shin 2018), and 6.8% fewer resident falls (Shin & Hyun 2015). A 12% increase in RN HPRD resulted in a 3% improvement in quality of care, while a 20% increase in RN HPRD led to a 5% – 8% improvement in quality of care (Shin 2019). A 1-h increase in LPN HPRD decreased inappropriate antipsychotic medicine prescribing by 1.83% (Chappell et al. 2022). However, when increasing the HPRD of CNAs and care workers by 1 h, aggressive behaviours in residents increased by 4.238%, residents on bed rest increased by 5.047% (Shin 2018), and there was a 5.3% increase in feeding tube use (Shin & Hyun 2015).

Only three studies reported a link between skill mix, staffing levels, and organisational outcomes. In Lerner’s (2013) study in 2013, 225 United States LTCFs with a mean total staffing of 1.71 HPRD received 9.76 deficiency citations for violating federal regulatory care standards. While some LTCFs received no citations, others received up to 51 deficiency citations (Lerner 2013). Similarly, Yoon et al. (2022) found that the likelihood of receiving deficiency citations was lower when staffing levels were 3.61 HPRD rather than 3.51 HPRD. In addition, low staff turnover leads to better quality of care, whereas high RN turnover worsens residents’ outcomes.

Staff allocation and the use of nurses and caregivers showed similarities across the countries. The roles and responsibilities of the RNs include taking responsibility for the overall nursing process (Zimmermann & Pfaff 2018), being accountable for evaluating overall nursing outcomes (Yoon et al. 2022), providing health education, conducting residents’ physical assessments (Lee et al. 2022), and overseeing nursing assistants and additional care staff (Zimmermann & Pfaff 2018). Registered nurses also tended to care for residents requiring higher levels of care (Tuinman et al. 2016). Between 3.5% and 28.9% of RNs’ time is spent on non-value-added resident-related activities, such as refilling stock items, searching for equipment and stock, issuing linen, and locating other staff. Furthermore, RNs spend about 16.9% of their duty time on walking through the unit and 41.7% on indirect resident care, including documentation and communication (McCloskey et al. 2015).

Authors found a lack of clarity between the roles of nurse categories and caregivers (Lee et al. 2022; McCloskey et al. 2015; Shin 2019; Shin & Hyun 2015; Shin & Shin 2019; Yang et al. 2021; Yoon et al. 2022). Licensed practical nurses appeared to work beyond their scope of practice when RNs were unavailable, but below their scope of practice when CNA shortages occurred (Yoon et al. 2022). In addition, poor role clarity led to RNs and LPNs performing various tasks that could have been delegated to unlicensed care workers (McCloskey et al. 2015). Care workers provide most of the direct resident care under the supervision of RNs and LPNs (Schnelle et al. 2016; Shin 2019; Yang et al. 2021; Zirves et al. 2021).

Acuity-based staffing is not commonly used in most countries, including the United States and Canada, which provided an answer to this scoping review’s third question: ‘Is the allocation of staff aligned to the acuity of the individual residents?’ Harrington et al. (2012) stated that not implementing acuity-based staffing standards may be the reason for the inability to address residents’ higher-level care needs. Germany may be an exception, as residents’ dependency levels informed staffing decisions (Zimmermann & Pfaff 2018). Care omissions were about 22% when residents’ acuity levels were high (Schnelle et al. 2016). However, the authors found that residents received similar care despite differing needs (Tuinman et al. 2016).

The skill mix standards varied across and within countries, from no specific set standards to prescribing explicit staffing standards. South Africa also has explicit mandatory staffing standards, irrespective of province, whether urban or rural, and whether LTCFs were private or state-subsidised (Republic of South Africa 2010a). The RN numbers in the overall nurse and caregiver staffing in the studies appear low due to some countries’ standards, such as South Korea, where CNAs may be used instead of RNs (Cho et al. 2020; Shin et al. 2021). In contrast to the proportion of nurses in the skill mix in the included studies, the proportion of unlicensed caregivers in the total staffing was as high as 84.13% (Lee et al. 2022), despite LTCF residents requiring higher levels of (Bae & Kim 2020; Kim & Han 2018; Tuinman et al. 2016). South African staffing standards for frail residents require that 33% of the total staff be RNs, but 50% of the total RNs may be replaced by ENs, resulting in 16.5% RNs and 16.5% ENs. The remaining 66% of the staff may be ENAs, but 50% of the total ENAs may be replaced by caregivers, thus 33% ENAs and 33% caregivers (Republic of South Africa 2010a).

Authors found that more highly educated healthcare professionals are needed (Kim & Han 2018) because higher-qualified staff can better address residents’ higher care needs, thereby improving resident outcomes (Chappell et al. 2022). Overall, the quality of resident care is improved, with lower infection and mortality rates when more RNs are employed rather than care workers (Lee et al. 2022), leading to better quality of life for residents, less decline in residents’ functional abilities (Shin et al. 2021), and fewer pressure ulcers (Clemens et al. 2021; Jutkowitz et al. 2023; Shin et al. 2021). Thus, before determining the skill mix needed, LTCFs should identify which educational levels and skills are required to meet residents’ dependency levels and care needs (Harrington et al. 2012). In South African LTCFs, it may be more viable for financially struggling facilities to not only appoint professional nurses but also make greater use of general nurses (diploma-qualified). However, LTCFs should analyse gaps in general nurses’ skills and address them through in-service training. Although RNs are typically employed at a higher cost than LPNs or CNAs (or ENs and ENAs in South Africa), using RNs offers potential cost-effectiveness in contrast to applying a low-cost nurse staffing model. Investing in more RNs may improve the quality of care, result in fewer deficiency citations (Harrington et al. 2020; Perruchoud et al. 2021), and reduce costs associated with adverse events (Cho et al. 2020; Mukamel et al. 2023).

Studies have shown that staffing levels in LTCFs vary across countries, with 2.3 HPRD in the United States (Schnelle et al. 2016) and 5.19 HPRD in Sweden (Harrington et al. 2012). United States experts have suggested that LTCFs provide 4.55 HPRD, although many countries provide lower HPRD (Harrington et al. 2012). A minimum of 2.57 HPRD is prescribed for frail residents in South Africa (Republic of South Africa 2010a). Staff burnout may follow when staffing levels are low, as higher workloads increase stress, impede staff’s ability to assist residents with daily activities, and may lead to inadequate care (Perruchoud et al. 2021; Schnelle et al. 2016). Adverse events, such as increased use of psychotropic medications in residents, may occur when RN staffing levels are too low. Conversely, higher RN staffing can contribute to positive resident outcomes. Higher RN hours may help address residents’ behavioural issues when RNs have enough time (Yoon et al. 2022), thereby reducing the need for antipsychotic medications (Chappell et al. 2022). Fewer resident deaths were associated with higher staffing levels (Cho et al. 2020), emergency department visits and rehospitalisations were less (Yang et al. 2021), fewer residents were on bed rest, and there was less need to restrain residents (Shin 2018). Long-term care facilities can incur higher costs when they are required to use agency staff to fill staffing gaps, rather than increase baseline staffing levels above the minimum mandated standards (Griffiths et al. 2021).

Despite evidence that residents’ needs are increasing, most LTCFs did not use acuity-based staffing methods at the time of completing this review (Chappell et al. 2022; Zimmermann & Pfaff 2018). Authors found that residents receive similar care despite differing needs (Tuinman et al. 2016). Moreover, care omissions were higher when residents’ acuity was higher (Schnelle et al. 2016). Thus, before staff planning, LTCFs should assess residents’ needs, consider their acuity levels, and align staff’s educational levels, skills, and experience with those needs to ensure residents receive appropriate care (Harrington et al. 2012; Tuinman et al. 2016).

Despite the clear scope of practices for nurses, there is an increased lack of clarity between the roles of different nurse categories and caregivers (Lee et al. 2022; McCloskey et al. 2015). To reduce costs, LTCFs appear to rely on caregivers and LPNs rather than RNs. Consequently, less-qualified staff may assume responsibilities beyond their qualifications. Lee et al. (2022) found that 22% of South Korean LTCFs did not employ RNs but instead used CNAs, even though RNs and CNAs have different scopes of practice and qualifications (Lee et al. 2022; Perruchoud et al. 2021). Likewise, RN shortages may lead LPNs to work beyond their scope of practice, risking overextension (Yang et al. 2021). Despite the RNs’ and LPNs’ training and skill set, the results indicated that they perform activities that could have been delegated to lower-qualified staff (McCloskey et al. 2015). Registered nurses spend significant time on indirect resident care activities, such as walking through the units, reviewing documents, and recordkeeping (McCloskey et al. 2015; Tuinman et al. 2016).

However, these non-direct resident activities should support residents’ care. Registered nurses should review workflow processes and feel comfortable assigning tasks to care workers when those tasks do not benefit residents.

Consequently, RNs’ efficiency could be enhanced, and they could concentrate on completing tasks within the full scope of their practice (McCloskey et al. 2015). Furthermore, despite being unlicensed, caregivers provide most of the direct resident care and carry significant responsibilities (Lee et al. 2022; Yang et al. 2021). Thus, role clarifications are needed to ensure that tasks are assigned to the appropriate staff, enabling resource optimisation and facilitating quality care (McCloskey et al. 2015). In South Africa, policies that comply with the legal boundaries of caregivers’ job scope may prevent undesirable task-shifting from higher- to lower-qualified staff.

This scoping review confirmed that adequate staffing is viewed differently between countries and that staffing involves more than only having enough staff with the right qualifications. The review also highlights the importance of aligning the allocation of tasks with the nurses’ scope of practice, the caregivers’ job descriptions, and the acuity of the residents.

During this scoping review, insights were gained into nurse and caregiver staffing in LTCFs. Developing a protocol to delineate the scoping review plan strengthened the review. Transparency was enhanced by using two independent reviewers to conduct a search strategy and document all the phases of the review process. A limitation of the scoping review was a lack of evidence from low- to middle-income countries. This may provide an inaccurate or biased perspective of the current state of skill mix, staffing levels, and staff allocation practices in LTCFs. Limiting the search to English-language studies from 2010 may have narrowed the scope of the review, potentially excluding relevant articles and introducing publication bias.

Further research is needed on nurse and caregiver staffing, especially in LTCFs in low- and middle-income countries. Given that caregivers provided most of the care, it is necessary to examine whether staffing deficiencies were associated with adverse events. In addition, research is needed to determine whether residents and family members are comfortable receiving most of their care from caregivers rather than from higher-qualified nurses.