Introduction

Background information

Service delivery in hospitals is labour-intensive and human resource requirements represent a disproportionate allocation of any availed health service budget [1, 2]. Whereas achieving optimal healthcare worker-to-patient ratios remains a global challenge [3,4,5], the majority of health systems are recognising the importance of a cost-effective and safe health workforce. However, inadequate finances for human resource recruitment and development [6, 7], limited capacity development opportunities, poor remuneration, difficult working conditions, and limited career advancement opportunities [7, 8] all contribute to the inability to attain optimal staffing ratios in healthcare facilities.

Broadly, the concept of human resources for health (HRH) not only includes primary care providers (i.e., physicians, nurses, or pharmacists) but also other assistive personnel like the administrators and care assistants who may not directly provide care to patients but provide operational support services and are, therefore, crucial to service delivery and the overall functioning of the healthcare system [8]. In 2020, the World Health Assembly [9] acknowledged that concerted efforts, including the implementation of the WHO Global Strategy on HRH [8] led to a reduction of shortages in healthcare workers by 3 million to 15 million [10]. Unlike other regions in the global north, countries east of the Mediterranean Sea and sub-Sahara Africa still experience dire staffing shortages and comparatively low healthcare worker-to-population ratios [7, 9].

The use of assistive personnel has been suggested as a task-shifting and skill-mix initiative that can help professional healthcare workers to optimise their shift time to focus on high-acuity or more technical tasks [11,12,13,14,15,16]. Conceptually, task-shifting is the rational re‐distribution or delegation of specific tasks among health workforce teams from the highly skilled to the less qualified/skilled staff [13]. On the other hand, skill-mix has been described as a multi-dimensional undertaking that incorporates performative elements (such as knowledge, skills, abilities, and competencies), intra-professional transversality (such as grade, level of expertise, education, and training), and inter-professional transversality of healthcare practice(i.e., a mix of posts, regulation, staff mix, and ratios). The relative proportions of highly skilled care providers and less skilled support staff represent an example of a staffing skill-mix. The current evidence suggests skill-mix tends to vary by fiscal year and country, and this is partly attributed to the adequacy of healthcare financing for HRH [7, 8, 17]. Consequently, to manage scarce human and financial resources efficiently while delivering needed care, healthcare systems adopt task-shifting and skill-mix strategies that target lower-level cadres whose emoluments might be less costly.

Some evidence suggests the nature and roles ascribed to unlicensed assistive staff vary across countries and healthcare systems. For example, whereas some hospitals in high-income countries assign their care assistants extended roles, such as phlebotomy and patient monitoring [18,19,20], in low-income settings, they seem to take up informally negotiated basic duties, such as general housekeeping, portering [18, 21,22,23], or supporting patients to perform activities of daily living [11]. Often, these assistants do not undergo any formal or professional pre-service training [8, 13, 24] based on a standard curriculum but might have different forms of informal or on-the-job training, mainly from registered nurses. However, a lot remains unclear or inconsistent on their scope of duties.

Some existing literature has examined the utilization of care assistants and highlighted knowledge gaps on how they are deployed in hospitals and how they affect patient care outcomes and experiences [25]. In addition, available reviews [21, 26, 27] have focused more on unlicensed assistive personnel who have undergone some form of structured formal training ranging from 6 months up to 2 years [11]. No review has focused on those with no formal pre-service training [8, 13, 24], which are a common cadre in many under-resourced healthcare settings, where clinical governance, licensure, and frameworks to regulate their activities may be limited [26]. Moreover, despite the wide utilisation of care assistants in hospitals, the global strategy on HRH hardly mentions the roles care assistants perform [11]. This leaves a lacuna in the clarity of their roles, while the scope of practice is mostly at the discretion of individual hospitals or supervisor. The use of this cadre of staffers is not invariable across most hospitals and begs for scrutiny. Thus, in this review, we coined and used the term hospital-based “ward/care assistants” (CAs) to describe this cadre of staff—lower-skilled assistive personnel who typically provide support to nurses in formal hospital settings. Consequently, this scoping review sought to answer the following questions:

Review questions

In formal hospital settings:

  1. 1.

    What roles and duties are performed by ward/care assistants?

  2. 2.

    What impact do ward/care assistants have on patient care?

  3. 3.

    What are the perceptions and experiences of nurses towards ward/care assistants?

  4. 4.

    What clinical or organisational governance frameworks exist to regulate the activities of ward/care assistants?

Methods

Design

We conducted a scoping review following the Joanna Briggs Institute and PRISMA–ScR evidence synthesis and reporting guidelines [28,29,30]. These guidelines describe the best practices for evidence synthesis from protocol development, search strategies, data extraction, interpretation, and reporting of scoping review results.

Protocol registration

The protocol for this review was registered on the Open Science Framework registries [31].

Search strategy and data sources

We conducted our literature search in PUBMED, CINAHL, PsychINFO, EMBASE, Web of Science, and Scopus with additional targeted searches from Google Scholar and citation chasing, particularly for grey literature. The last search was conducted on 20th June 2022. We combined the search terms using Boolean operators and adapted them for each of the electronic databases. The comprehensive list of the keywords used is shared in Additional File 1 and a sample full search strategy in Additional File 2. There was no time limitation on the search period. However, language was restricted to English-published papers only. If a full text was completely irretrievable, the reviewers attempted to contact the corresponding author for the full text, otherwise, the paper was excluded from charting and synthesis.

Eligibility criteria and data items

We used the Participants–Concept–Context framework [32, 33] to describe the inclusion and exclusion criteria. Under “Participants”, we only included papers that reported on CAs. We defined CAs as hospital-based staff that support healthcare professionals to provide non-clinical, low-skilled basic tasks to a patient or within a ward or clinic—but mainly supporting nurses’ work. They would ordinarily not have any formal professional, technical training before working, licensing, or regulatory requirements save for a high-school level education with some level of on-the-job (in-service) training, particularly from nurses. We, therefore, excluded literature that reported on CAs who had undergone some formal or professional pre-service training [8, 13, 24] or who require licensure before working. Pre-service training was considered formal or professional if there is a defined curriculum and the training duration exceeded 6 months.

Under “Concepts”, we sought to document the characterisation of CAs as unlicensed assistive personnel in formal hospital settings, their ascribed roles and duties, and their impact on patient care and nurses’ experience. We also sought to map regulatory frameworks and clinical governance mechanisms at the workplace. We only included papers that reported on these review objectives.

Finally, for “Context” we considered studies that report CAs working or providing support to nurses in both in-patient and outpatient care settings, including nursing care homes. CAs providing services in the community or individuals’ homes of their patients—also known as community home-based care—were excluded. In addition, literature from across the globe irrespective of the World Bank’s income group [34] was included. We considered both primary and secondary research papers as well as relevant grey literature. Thesis and dissertations, conference abstracts, seminar reports, case reports/series, and animal studies were excluded from this review. Additional File 3 provides a detailed description of the eligibility criteria.

Data management and synthesis

Data management and analysis steps involved study selection, data extraction, data synthesis, and reporting.

Study screening and selection

Two reviewers (VK and OO) independently conducted literature searches in June 2022. The search results were then consolidated and screened for relevance against the eligibility criteria, initially by title and abstract then later by full text. The screening stage was done independently by two reviewers (VK and OO) with the aid of The EndNote reference manager [35]. Reasons for ineligibility were documented at the full-text screening stage and are reported in the results section (Fig. 1). Inconsistencies or disagreements on the eligibility of a paper were jointly discussed between the two reviewers at both the abstract and full-text review stages. In case inconsistencies or disagreements could not be resolved by the two reviewers, an additional third reviewer (AI) was invited to act as a tiebreaker in a joint discussion for consensus building.

Fig. 1
figure 1

Summary of search results and records screening

Full size image

Charting the data

Two reviewers (VK and OO) used a pre-developed template on Microsoft Excel to aid in the charting of key pre-specified study information crucial in answering the review questions, including bibliographic details, study context and characteristics, concepts, roles and duties, training, regulatory/governance mechanisms, patient care outcomes, nurses’ experiences related to the utilisation of the CAs. Charting was done jointly by two reviewers (VK and OO). The synthesis, interpretation, and reporting of the findings of this review were then guided by the PRISMA–ScR guidelines [29].

Critical appraisal of the individual papers

Quality and risk of bias assessment were not undertaken, since our eligible papers yielded multiple types of papers and methods that were heterogenous in methodology, region, care setting, participant selection, and reporting of their findings (illustrated in Additional File 4). This would have meant multiple critical appraisal tools. Moreover, quality assessment is not a mandatory requirement for scoping reviews, since they do not aim to synthesise the ‘strength’ of evidence from literature but rather to provide an overview of the available evidence[36]. Thus, the reviewers agreed to include all the 73 eligible papers in data charting and synthesis.

Synthesis

The Joanna Briggs Institute’s evidence synthesis manual [28, 30] guided our synthesis approach. With the aid of NVIVO 12 Plus [37], we conducted a thematic content analysis. Data from the charting template was uploaded onto the NVIVO 12 Plus program followed by open coding onto nodes that helped to answer our review questions. The nodes were then refined and grouped into sub-themes and themes. The process generated themes on how CAs are characterised, their ascribed duties, patient care outcomes, and sentiments (i.e., views, feelings, or opinions) of nurses’ regarding their experiences working with CAs. The emerging themes were then assigned sentiment labels on whether they had a positive, neutral, or negative effect on care or experiences of care. Frequencies of mention of titles that characterise CAs were generated from each source and presented in a word cloud. Similarly, each specific task charted from individual sources was coded into a theme on tasks. These specific tasks were then grouped into broad categories for easier interpretation and reporting. The number of records mentioning a task and the task frequency across the records were then generated for graphical presentation. Finally, clinical, and organisational governance mechanisms were extracted, curated, and tabulated as reported in the individual sources.

Results

Reporting is guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for Scoping Reviews (PRISMA–ScR) guidelines [29].

Search and screening results

Figure 1 shows a flow diagram adapted from the PRISMA 2020 statement [38]. It provides a summary of the sources searched and the records assessed for eligibility. We obtained a total of 85,251 records from electronic databases and grey literature searches. After screening for eligibility and excluding records based on the eligibility criteria, a total of 73 records were included in the final full-text review, data charting, and synthesis.

Summary statistics

Table 1 provides a summary of key characteristics of the included records.

Table 1 Characteristics of included records
Full size table

Characteristics of sources

Majority of the papers were observational studies followed by grey literature. Records from high-income settings accounted for a majority of the eligible full texts. That is, United Kingdom (UK) had the highest number of sources (n = 19 [25%]) followed by the United States of America (n = 13 [17%]) and Australia (n = 7 [9%]), respectively. Japan, Israel, Benin, Hong Kong, Malawi, Brazil, and Uganda had 1 report each.

The hospital setting

Only 47 (n = 64%) of the papers described the type of care setting. Our synthesis established that these relevant sources yielded a mix of rural and urban (n = 22), inpatient-only (n = 18) and outpatient-only (n = 2), nursing care homes (n = 12), both inpatient and outpatient (n = 9) with a mix of children and adult care settings. Moreover, of the sources that mentioned a type of setting, public and private hospitals were reported in 30 (40%) and 13 (17%) papers, respectively. Twelve papers (16%) had a mix of both private and public. Additional File 4 provides more information on the setting as reported in the included papers.

Characterising CAs

Conceptually, we note that many titles are currently used for the identification or description of CAs in different countries and regions (Additional File 5). Only a few select similarities exist, for instance, “nursing assistant” is used in Australia, Hong Kong, China, New Zealand, the United States of America (USA), and the UK. UK has the highest number of terms/name variations that describe CAs (> 21), followed by Australia and Canada (> 11 each) then the USA and Hong Kong China (> 7 each). Consequently, the term “Assistant” is the most common stem word followed by “support staff/worker” in what are typically compound terms for CAs. Figure 2 shows a word cloud illustration of these variations. The other variations are summarised in Additional File 6 and in order of the most common to the least common.

Fig. 2
figure 2

Common descriptors for CAs

Full size image

Objective 1: Duties performed to CAs in hospital settings

Duties ascribed to CAs

A majority (78%, n = 58) of the records mentioned specific tasks performed by CAs in their care setting or facility. Our review was able to chart and curate 58 different tasks and these were grouped into 7 broad categories (illustrated in Table 2). That is, direct patient care (n = 53 records), housekeeping (n = 26 records), clerical and portering (n = 19 records each), patient flow management and ordering laboratory tests (n = 4 records each). Emergency response and first aid was the least reported task category.

Table 2 Duties undertaken by CAs
Full size table

Table 2 unpacks the specific tasks performed under broad and sub-categories. It was noted that vital signs monitoring, patient hygiene, and feeding are top of the list of direct care duties, whereas environment and surface cleaning, stock taking, equipment care and device functionality checks are the most common housekeeping tasks. However, a few outstanding papers had CAs performing more extended roles that require an extra level of knowledge and skills, i.e., flu vaccination, drug injections [24, 39, 40], catheterisation [24, 40,41,42,43,44], phlebotomy [44,45,46,47,48,49], electrocardiogram (ECG) monitoring [39, 44, 46, 49,50,51,52], wound/colostomy care [24, 27, 39,40,41,42, 44, 49, 51, 53], resuscitation [54], and requesting laboratory tests [52, 55,56,57]. These extended roles are observed mainly in high-income countries (Australia, Canada, and UK, and the USA) and not in low- and middle-income countries (LMICs).

Objective 2a: Impact of CAs on patient care and their experiences

Only 20 (27%) papers reported some form of patient care outcomes. With the aid of NVIVO 12 Plus [37], a thematic analysis approach was used to examine and curate the impact of CAs on patient care. The emerging themes were then assigned sentiment labels on whether they had a positive, neutral, or negative effects on care. As shown in Table 3, the papers report a mix of both positive and negative effects.

Table 3 Impact of CAs on patient care
Full size table

Objective 2b: Nurses’ perception and experiences towards CAs

Only 23 (32%) papers reported on nurses’ experience of working with CAs. With the aid of NVIVO, a sentiment analysis approach was used to code and curate themes related to nurses’ experiences of working with CAs at the individual level. The emergent themes were grouped into either positive, neutral, or negative sentiments as illustrated in Table 4. Additional file 7 is a summary illustration of what is reported as per the synthesised records.

Table 4 Nurses’ sentiments on utilisation of CAs
Full size table

Objective 3: clinical and organisational governance frameworks that regulate activities of CAs

Objective 3a: Regulatory and clinical governance mechanisms

Only 33 (45%) of the records mentioned some form of an organisational regulatory or clinical governance mechanism for the CAs—8 in the UK [26, 27, 39, 41, 42, 48, 54, 73, 81,82,83], 6 in the USA [26, 51, 65, 70, 84,85,86,87], 3 were from Taiwan China [75, 88], 2 from Canada [24, 40], and 2 from Australia[26, 51], and 1 each from Japan [72], Brazil [68], Sweden [76], Kenya [89], Malawi [66], and Uganda [90]. Moreover, Brazil, Kenya, Malawi, and Uganda are the only LMICs reporting some form of clinical governance mechanism. However, for Kenya and Uganda, it is largely a proposed framework and not an already operationalised one.

We note that these mechanisms vary substantially within and across the 15 countries reviewed (Additional File 7 and Additional File 8). However, notable similarities in some countries include a requirement for completion of a competency-based training curriculum in a work setting and that there is delegation and supervision by a qualified (registered) nurse. The majority of countries lack a legislative framework that standardises or regulates the training of CAs. Moreover, nearly, all the countries (93%) do not have a task-shifting/sharing policy that guide the delegation and supervision of tasks.

Objective 3b: Training/capacity development

Only 41 (55%) of eligible papers reported some level of training requirement (pre-service or in-service) and this was reported in several countries, including Australia, Benin, Canada, Israel, Kenya, Malawi, the Republic of Ireland, Taiwan, China, UK, and the USA. Sources from Brazil, Hong Kong, China, Sweden, Uganda, and Japan did not report any form of training requirement before or after the recruitment of CAs.

Our review establishes that nearly all the CAs are required to undertake onboarding training and continue with in-service competency skills training at their own pace. The skills development period varies substantially across all the countries reviewed. For instance, the minimum in-service (on-the-job) training duration ranged between 1 and 48 h [61, 66, 70], whereas the maximum in-service training period was undertaken between 126 and 672 days [26, 51, 91].

Training topics/content

The majority of the theory and practical learnings covered topics and skills related to basic nursing care (e.g., taking vital signs, simple wound care, taking weight and height measurements, specimen collection, or patient hygiene), workplace health and safety (including cleanliness and basic first aid), communication skills, infection prevention and control (including equipment processing), anatomy and physiology, confidentiality, privacy, and dignity as the most common training topics. The least common topics mentioned include family support/centred care, health promotion, human growth and development, food, and nutrition, and counselling. A detailed list is found in Additional File 9.

Discussion

This review aimed to map evidence for the characterisation of lower-skilled support and CAs in formal hospital settings. We note that there is a lot of inconsistency and substantial variation in the terms or titles for CAs, their training, the scope of practice, and regulatory mechanisms within and across the reports included from 15 countries. Moreover, we note that the two commonly used terms “assistant” and “support worker” do not align with the ILO’s ISCO 5132, 5133, and 3231 descriptions [92] that use “Institution-based personal care workers” or “nursing aid” to refer to the titles and roles ascribed to CAs. Thus, our synthesis points out the need to review the ILO’s ISCO nomenclature for this occupational group.

Our evidence mapping indicates a substantial amount of empirical literature on task-shifting/sharing between nurses and CAs in high-income countries but largely understudied in LMICs. However, overall, some evidence suggests CAs may contribute to improved quality of patient care by availing nurses time to concentrate on high-acuity and critical care activities [11, 12]. There is, however, contrary evidence suggesting the involvement of CAs in patient care may pose a risk to patient safety and quality of care [26, 48, 93, 94]. In addition, the evidence for CAs’ ascribed roles is generally mixed and their role boundary with professional nurses is even more blurry from the patients’ perspective.

Importantly, this review reveals that CAs take up a range of roles in clinical and care settings some of which are informally negotiated based on competency levels, years of experience, confidence, or supervision effort required. These insights are similar to observations by Just et al. [63] while examining the role of CAs in end-of-life care and McKenna et al. in their review on how CAs’ roles affect patient safety and care quality. The majority of the CAs’ assigned tasks are direct patient care activities, including vital signs monitoring, assisting with patient hygiene and elimination needs, and support with patient medication—all of which help to meet essential patient care needs. This suggests that with good mentorship, competency-based training, and appropriate supervision, such tasks could be progressively assigned to lower-cadre assistive personnel. Interestingly, we also note that there are higher-level extended roles undertaken by the CAs, namely, basic life support, giving injections, wound care, ECG monitoring, catheterisation, and sample(specimen) collection. However, in support of other studies elsewhere [63, 95], these tasks are mostly informally negotiated, setting-specific, and would require a higher level of skills and training. Notably, the extended roles were common in USA, Canada, UK, and Australia, which could be attributed to deliberate staff upskilling incentives and/or the availability of good on-the-job competency development programmes in these settings. More importantly, clarity of role boundaries and accountability mechanisms that avoid role conflict with the qualified nursing workforce appears to be missing. Overall, the nature of pre-service, and on-the-job training requirements determine the roles and depth to which tasks are assigned and executed by the CAs [24, 40, 45, 60, 64, 81, 96, 97].

How CAs are recruited and deployed in formal care settings has the potential to affect patient care outcomes and shape nurses’ experiences of care delivery. We infer that hospitals choose between two strategies of integration of CAs in the hospital setting: a substitutive model [50, 73] where the CAs are employed to cover the shortage of nursing staff and a supplemental model where the CAs are added as a layer to an existing proportionately ‘optimal’ staff. However, these strategies have only been explicitly reported in the UK and Australia [50, 73, 98]. Still, the choice of either approach is very much inconsistent or unclear across and within most care settings [50, 73]. This could be explored further in future research. Supplementation of nurses with CAs creates an incremental effect on skill-mix and is generally linked with positive patient outcomes and clinical staff experiences, including promoting continuity of quality and effective care through fewer work interruptions, reduced waiting time, and overall, reduction in the risk for missed care. On the other hand, the substitutive strategy is associated with a reduced nurse–patient contact time, leading to concerns about patient safety [14, 69, 71], role deprivation, loss of professional identity, reduced job satisfaction, and unclear accountability lines for actions [53, 63, 95]. Noteworthy, as illustrated in Table 3 and Table 4, a few papers had impacts and experiences that conflict with others. For instance, patient safety is not only viewed in a positive sense (i.e., reduced patient injury [48, 54] but equally in a negative sense (i.e., heightened risk to patient safety [42, 50, 68]. Similarly, job satisfaction among nurses has both negative [54] and positive[59, 79] sentiments from different settings. Understandably, as reported elsewhere [99, 100] the effects related to patient safety and satisfaction present potential for medical–legal issues, although this implication was not observed in the current review.

With the ever-increasing strain on healthcare globally, many healthcare systems have been pushed to adopt mechanisms to optimize care delivery amidst limited workforce capacities [43]. Coupled with the slim evidence on specific safe staffing mix ratios, CAs have become a norm in healthcare systems albeit, largely informally in most countries. Although a few studies reported some form of regulatory or clinical governance mechanism, the general picture is a lack of standardised regulatory frameworks for training, employment, degree of task-shifting/task-sharing, the delegation of duties, and supervision and accountability of the CAs. In LMICs, this cadre of staff remains marginalized and or unofficially recognised. Effective management and utilisation of this workforce in healthcare remains disjointed and could lead to either exploitation or underutilisation [43, 73, 85]. Moreover, the lack of clarity on their contribution to patient safety and quality of care begs further investigation. In essence, the adoption and use of CAs in healthcare require a careful approach to sustain professional accountability and avoid over-dilution of skill-mix [42, 94, 97, 101].

The synthesis of the finding of this review is cognisant of some limitations. First, since this is a review, its methodology is unable to make causal inferences on how CAs affect patient care outcomes. However, our findings help to highlight gaps that could be addressed by robust empirical study designs. Second, our eligibility was limited to records in English-language only. However, despite adopting an open search period and across multiple databases and registries, we only retrieved and synthesised literature from 15 countries. This could imply this topic remains under-studied in the countries with no available literature.

Conclusion

In summary, the nomenclature for CAs is largely inconsistent and variably used. Second, the application of task-shifting strategies in hospitals is fragmented and this is observed both within and across countries. Third, the effect of CAs on patient care outcomes or nurses’ experiences remains unclear as both positive and negative sentiments have been reported in equal measure. Consequently, the synthesis of these findings has several implications. First, we provide evidence that the 2008 version of the International Labour Organization’s International Standard Classification of Occupations [ILO’s ISCO] [92] and related national occupation classifications should be updated for a more clear and more meaningful nomenclature for CAs and other related assistive personnel. Second, collating and mapping empirical evidence on regulatory and clinical governance mechanisms for CAs and their impact on patient care as well as nurses’ experiences present an opportunity to advance conversations on future research on HRH for assistive personnel. For instance, the amount of net savings or costs incurred relative to gains on service delivery and efficiency from assistive personnel in healthcare facilities. Third, these findings bring to the fore, a need for the operationalisation of context-specific policy guidelines and strategies for task-shifting/sharing, including the training of CAs, within a larger framework for norms and standards for HRH management. Such guidance will need to accommodate a diverse and changing landscape of CA roles in hospital settings.