Specialised nursing tasks in cancer care and their effects

Aim and research questions

This umbrella review aimed to synthesise the results of reviews assessing the scope and effectiveness of roles and tasks of nurses specifically qualified in cancer care. The following research questions were addressed:

1. 1.
   Which specific tasks of oncology nurses are described and evaluated in the studies?
2. 2.
   Which formal qualification degrees and competencies are assigned to these specific nursing tasks?
3. 3.
   Which effects have specific tasks executed by oncology nurses on patient-related and economic care outcomes?

Eligibility criteria

We applied methodological and content-related inclusion criteria. Eligible studies were systematic reviews, meta-analyses, research syntheses and Health Technology Assessment (HTA) reports published in English or German language within the period 2016–2021 targeting an evaluative question. Syntheses of evidence had to meet the following criteria based on a modified version of the three-part systematic review definition (Pluye et al., 2016):

• •
  explicit and transparent research questions as well as inclusion and exclusion criteria,
• •
  a comprehensive and reported literature search strategy,
• •
  clearly stated methodology of data collection,
• •
  data extraction based on appropriate criteria,
• •
  detailed description and a critical appraisal of the included individual research studies.

Reviews had to meet additional content-related criteria: 1) focus on an evaluation of the effects of specialised nursing tasks and roles in cancer care, 2) outcomes related to patients, care procedures, resource use or economic impact, 3) and description of the nurses’ roles and tasks including assigned qualifications or competencies in a replicable way (Witzke et al., 2021).

Data sources and searches

We conducted systematic literature searches in the electronic databases MEDLINE (via Ovid), CINAHL, Epistemonikos (Epistemonikos is a large multilingual database of health evidence including over 100,000 systematic reviews), the International HTA Database, the JBI Database of Systematic Reviews and Implementation Reports and the Cochrane Database of Systematic Reviews. A comprehensive search strategy was applied to address the categories population/profession (e.g. nurs*), special roles/tasks (e.g. qualification) and cancer (care) (e.g. oncolog*). For MEDLINE, we used the “Systematic reviews or meta-analyses search filter” (University of Texas School of Public Health, 2021). We adapted the search strategy for each database (see Electronic Supplementary Material ESM1 for the MEDLINE strategy). We conducted forward (Google Scholar and PubMed) and backward citation tracking of included reviews to identify other potentially eligible reviews.

Study selection

The screening process was performed in Covidence (Veritas Health Innovation Ltd). First, titles and abstracts were independently assessed by two reviewers (ACR, BBH, KS, TW). Then, two reviewers (ACR, TW) independently assessed the full-texts for inclusion. Discrepancies were resolved by discussion or by the decision of a third reviewer.

Data collection

We extracted data using a standardised form based on the JBI data extraction tool, adapted to extract review features (e.g. data on cancer types and nursing roles). We pilot-tested the tool on three publications (Witzke et al., 2021). One reviewer extracted the data checked by a second reviewer. Disagreements were resolved by discussion or by a third reviewer.

Assessment of methodological quality

Reviews were assessed for methodological quality using an 11-item-checklist from the JBI umbrella review methodology working group to identify potential risk of bias in research syntheses comprising eleven questions. The authors added a 12th criterion to address whether two or more people screened the articles independently. Oriented towards the AMSTAR quality rating, the methodological quality was rated as high if 8–11 items, medium if 4–7 items and low if 1–3 items were met. Two reviewers conducted the critical appraisal (ACR, TW) independently. Disagreements were resolved by discussion or by a third reviewer.

Data synthesis

Given the clinical heterogeneity across the reviews, due to different study populations, an organised narrative synthesis approach was applied using text and evidence tables. Reviews were categorised by the Omaha System Intervention Scheme (Martin, 2005) respective the kind of interventions (i.e. nursing activities) studied and by the EONS Cancer Nursing Education Framework (European Oncology Nursing Society, 2018) regarding the category of cancer trajectory addressed. Both, the Omaha System and the EONS Cancer Nursing Education Framework are established frameworks in the field of nursing care and are applicable to the entire care process, including palliative care.

The Omaha System provides a taxonomy for classifying nursing interventions against four categories of nursing tasks and related nursing activities, i.e. tasks. The categories are: “Teaching, Guidance, and Counselling”, “Treatments and Procedures”, “Case Management”, and “Surveillance”. Each category contains a list of nursing tasks (subcategories) building a scope of tasks (Martin, 2005). Based on the information provided by the reviews, we assigned reported interventions into these four categories and related subcategories. Additional tasks were added under “Other” to the most fitting category (decided by discussion). All review authors, except JR, were involved in allocating tasks.

The qualification levels (e.g. Advanced Nurse Practitioner [ANP] or advanced training in oncology nursing) reported for single interventions were extracted and cross-mapped against the identified tasks to explore the range of qualification levels required to fulfil the scope of tasks. Any qualification building upon first cycle nursing degree education not explicitly reported as reflecting a formal ANP degree or case management training was classified as “Advanced Training”, irrespective of the course scope and length.

The EONS Cancer Nursing Education Framework comprises eight modules. We applied the first five modules which reflect the cancer trajectory: “Risk Reduction, Early Detection and Health Promotion in Cancer Care”, “Cancer Pathophysiology and the Principles of Treatment Decision Making”, “Cancer Treatment, Patient and Occupational Safety”, “Supporting People Living With, Through and Beyond Cancer” and “Supporting People with Advanced Disease and at End of Life”. We added a further category for reviews that could not clearly be assigned to one of these categories or represent several steps of the cancer trajectory named “Diagnosis to Survival/End of Life Care”. Reviews that only included one study focusing on a certain category of the cancer trajectory were assigned to another category representing the majority of the studies (European Oncology Nursing Society, 2018).

The outcomes health-related quality of life (HRQoL), symptom burden, health service/resource use and costs as well as satisfaction and perceived benefits were considered to assess the effects of the nursing tasks. Four categories were created to provide a summary of the effectiveness: beneficial intervention (effective) (++/+), no difference exists in the investigated comparison, or the results are unclear or mixed (?), the results suggest the intervention is detrimental or less effective than the comparator (–) (Aromataris et al., 2020). The clear beneficial effect (++) was limited to effect estimates based on meta-analyses. To be assigned to the category (+) or (–), the majority of results of individual studies had to point out a beneficial or detrimental effect, respectively. If the majority of results showed no clear or mixed effects, the category (?) was assigned. Because of the way the results were reported in the reviews, no thresholds were applied. The assessment of the effectiveness was carried out by two people independently and resolved through extensive discussion in case of differing assessments. Therefore, the classification of an effect was based on an interpretation of the authors (ACR, TW) except for meta-analyses. Results from reviews based on a single study were automatically classified as unclear.

We applied the corrected covered area method to account for the overlap of individual studies in each of the included reviews (Pieper et al., 2014).

Study selection

From an initial pool of 2,657 references, eleven reviews were included within the narrative synthesis of the final review consisting of 148 original studies. The main exclusion reasons comprise that the intervention was not delivered by a nurse, an unsuitable study design, and a lack of details on the search strategy or quality assessment of individual research studies (see review flow for details in ESM2).

Characteristics of included reviews

The eleven reviews were published between 2016 and 2021 and include 148 individual studies (published from 1980 to 2020) conducted in various countries in Asia, America, Europe and Oceania (see ESM3). Two meta-analyses (Li et al., 2020; Suh & Lee, 2017), eight systematic reviews (Browall et al., 2017; Brown et al., 2021; Chan et al., 2020; Hussain Rawther et al., 2020; Kobleder et al., 2017; Rodríguez-Matesanz et al., 2020; Schneider et al., 2021; Zweers et al., 2016) and one scoping review (Molassiotis et al., 2021) were included. The majority of reviews (n = 6) included mixed cancer types, n = 4 focused solely on breast and n = 1 on gynaecological cancers. Most interventions covered were classified into the trajectory categories “Cancer Treatment, Patient and Occupational Safety” and “Supporting People Living With, Through and Beyond Cancer”. Only one review focused on the early detection of cancer. The settings differed including hospital, outpatient, domestic, community and palliative care.

The qualification levels varied from first cycle nursing education to nurses with advanced training courses or studies/special training for case management to advanced nursing practice qualifications. The qualification level was not reported in two reviews (Rodríguez-Matesanz et al., 2020; Suh & Lee, 2017) (Table 1). ESM3 shows the key characteristics of the eleven included reviews.

Table 1 Roles and Tasks by the Omaha System Intervention Scheme (Martin, 2005)

Categories and activities based on the Omaha System Intervention Scheme (Martin, 2005)	Browall et al. (2017)	Brown et al. (2021)	Chan et al. (2020)	Hussain Rawther et al. (2020)	Kobleder et al. (2017)	Li et al. (2020)	Molassiotis et al. (2021)	Rodríguez-Matesanz et al. (2020)	Schneider et al. (2021)	Suh and Lee (2017)	Zweers et al. (2016)
Notes: Assignment based on the number of studies to a certain qualification level. Qualification levels: ○ = first cycle nursing degree, □ = advanced training, Δ = Advanced Nursing Practice degree, x = qualification level not reported.
Teaching, Guidance & Counselling
Provide information and materials (health education)		□	□	□	□/Δ	□	○/□	x	Δ	x	○/□/Δ
Encourage action and responsibility for self-care and/or coping		□	□	□	□/Δ			x		x	○/□/Δ
Assist in decision-making		□	□		□/Δ		○/□
Assist in problem-solving			□	□				x
Counselling and support	□	□	□	□	□/Δ	□	○/□	x	Δ	x	○/□/Δ
Other										x
Treatments & Procedures Activities to prevent, decrease or alleviate signs and symptoms.
Provide medical services/symptom management (i.e. wound care, specimen collection, resistive exercises, medication prescriptions)	□			□			○/□	x	Δ	x	○/□/Δ
Case Management Activities to facilitate service delivery, improve communication among service providers, promote assertiveness & guide toward use of appropriate resources.
Coordination	□	□	□	□	□/Δ	□			Δ
Advocacy
Referral		□	□		□/Δ	□			Δ	x
Surveillance Activities to identify status in relation to a given condition or phenomenon.
Detection			□				○/□
Measurement							○/□
Critical analysis			□
Monitoring		□	□		□/Δ		○/□		Δ		○/□/Δ
Other										x

Table 1 Roles and Tasks by the Omaha System Intervention Scheme (Martin, 2005)

View as image HTML

Methodological quality

The methodological quality was classified as high in ten reviews and medium in one review (see ESM4). Three reviews provided sparse information on the search strategy, and the likelihood of publication bias was only assessed in the two meta-analyses. Half (n = 6) of the reviews indicated that the screening of potential eligible publications was done by two people independently (additionally added criterion).

Analysis of overlap

The overlap analysis showed a slight degree of overlap with a corrected covered area value of 3,9% (Pieper et al., 2014). The detailed calculations can be traced in the supplements (see ESM5 and ESM6).

Findings of the synthesis

Effectiveness of specialised nursing interventions

In total, eight reviews reported on HRQoL and symptom burden, six on health service/resource use and costs and nine reviews on satisfaction and perceived benefits. Table 2 summarises the effects on these outcomes, stratified for the categories of the cancer trajectory (cancer trajectory category [CTC]) and the scope of tasks according to the Omaha classification. One review (Schneider et al., 2021) was not included in this synthesis due to assumed selective reporting.

Table 2 Results for stages of cancer trajectory

	Outcomes**	HRQoL (n = 8)					Symptom burden (n = 8)					Health service/resource use and costs (n = 6)					Satisfaction and perceived benefits (n = 9)
Cancer Trajectory Stages based on the EONS Cancer Nursing Education Framework*	Intervention type	n(++)	n(n+)	n(?)	n(–)		n(++)	n(+)	n(?)	n(–)		n(++)	n(+)	n(?)	n(–)		n(++)	n(+)	n(?)	n(–)
Notes: Applicable Reviews***: Browall et al. (2017)1, Brown et al. (2021)2, Chan et al. (2020)3, Hussain Rawther et al. (2020)4, Kobleder et al. (2017)5, Li et al. (2020)6, Molassiotis et al. (2021)7, Rodrigues-Matesanz et al. (2020)8, Suh and Lee (2017)9, Zweers et al. (2016)10. Results from reviews that are only based on a single study (in bold type) were classified as unclear. n(++)/n(+) = Study results indicate the intervention is beneficial (effective), n(?) = number of/Reviews indicating no difference/unclear/mixed, n(–) = results suggest the intervention is detrimental or less effective. HRQoL = Health-related quality of life.* Cancer Trajectory Stages based on the EONS Cancer Nursing Education Framework (European Oncology Nursing Society, 2018). ** Outcomes guided by a systematic review examining the effects of nurse-led services in chronic disease management (Chan et al., 2018). *** Schneider et al. (2021) not included due to assumed selective reporting.
Risk Reduction, Early Detection and Health Promotion in Cancer Care	Teaching, Guidance & Counselling											16						16
	Treatments & Procedures
	Case Management											16						16
	Surveillance
Cancer Pathophysiology and the Principles of Treatment Decision-Making
Cancer Treatment, Patient and Occupational Safety	Teaching, Guidance & Counselling		12	33,4,7				41–3,7	14				23,7	21,4				23,7	22,4
	Treatments & Procedures			24,7				21,7	14				17	21,4				17	14
	Case Management		12	23,4				31-3	14				13	21,4				13	22,4
	Surveillance		12	23,7				32,3,7					23,7					23,7	12
Supporting People Living With, Through and Beyond Cancer	Teaching, Guidance & Counselling			61–3,5,7,8				11	42,3,5,7					31,3,7				22,7	51,3–5,8
	Treatments & Procedures			31,7,8				11	17					21,7				17	31,4,8
	Case Management			41–3,5				11	32,3,5					21,3				12	41,3–5
	Surveillance			42,3,5,7					42,3,5,7					23,7				22,7	32,3,5
Supporting People with Advanced Disease and at End of Life
Diagnosis to Survival/End of Life Care	Teaching, Guidance & Counselling			42,3,5,9					43,5,9,10					23,5					25,9
	Treatments & Procedures			19					29,10										19
	Case Management			42,3,5,9					33,5,9					23,5					25,9
	Surveillance			33,5,9					43,5,9,10					23,5					25,9

Table 2 Results for stages of cancer trajectory

View as image HTML

Key findings

This umbrella review includes eleven reviews on the effectiveness of specialist nurses’ tasks on specific health outcomes and cost of care among PABC without focusing on specific cancer types, age groups of PABC or settings. Overall, the review demonstrates that across the cancer trajectory, especially nurses with advanced training or an ANP degree were responsible for fulfilling the scope of tasks identified, aiming to have an impact on outcomes like HRQoL, symptom burden, health service/resource use and costs and patient satisfaction. This was particularly the case with case management tasks (Table 1).

In this umbrella review, various tasks in the fields “Teaching, Guidance, and Counselling”, “Treatments and Procedures”, “Case Management”, and “Surveillance” were identified, performed mainly by nurses with qualifications beyond initial training. Only tasks in direct nursing care were the subject of this umbrella review. The results point to an international trend of nursing specialisation, especially in the task fields of health education and counselling, case management and surveillance in the care of PABC. However, based on the aggregated information from the included reviews used here, no conclusions can be drawn about the exact task profile and the roles of the specialised nurses within the care teams in the implementation of these tasks. Based on international definitions of roles and tasks, it can be assumed that specialised nurses below the Master’s degree level are mainly responsible for planning, direct implementation and evaluation of the corresponding care processes. In contrast, specialised nurses at the Master’s degree level have additional roles in the needs-based coordination and organisation of care processes, in the application of evidence-based knowledge and quality development (16-2International Council of Nurses, 2020). Overall, available data on task profiles and formal qualifications indicate that specialised nursing tasks in cancer care are entrusted to nurses qualified at various advanced levels beyond first cycle degrees at level 6 of the European Qualifications Framework (Council of the European Union, 2017).

Only a selection of relevant outcomes could be included, some of them were reported differently (e.g. symptom burden is often included in cancer-specific quality of life). Consistent with the current state of research (Charalambous et al., 2018; Tuominen et al., 2019; Wu et al., 2021), this umbrella review provides initial, albeit inconsistent, indications of possible effects of specialised nursing roles and tasks in cancer care. Another umbrella review, including nine reviews published from 2007 to 2016, explored the effectiveness of nursing interventions on the lives of PABC (Tuominen et al., 2019). Our umbrella review builds upon this existing evidence and expands it by incorporating all care targets, including screening and diagnostic, and a particular focus on nurses’ qualification levels. However, its findings confirm the inconclusive evidence of specialised nursing roles and tasks on the outcomes considered. While more and more studies and reviews are emerging, there is still no clear picture of the effects on important outcomes like HRQoL. Given the dominating unclear/mixed results on effectiveness, well-reported studies with underlying theoretical frameworks containing core outcomes are needed to allow meta-analyses.

As stated, the themes of communication, leadership and management, as well as research utilisation, were not individually mapped within the scope of this review, assuming that they are part of every task of a specialised nurse to some extent. A Cochrane review evaluating the effectiveness of communication skills training for healthcare professionals working with PABC illustrates the importance of good communication for high-quality health care (Moore et al., 2018).

Four reviews (Browall et al., 2017; Chan et al., 2020; Suh & Lee, 2017; Zweers et al., 2016) discussed that the lack of a theoretical framework made it difficult to interpret reasons and drivers for effectiveness and non-effectiveness highlighting the use of a framework as a quality indicator. In most cases, it is unclear whether a framework to develop and evaluate the interventions such as the Medical Research Council (MRC) framework on complex interventions was used in the studies (Skivington et al., 2021). It remains unknown which processes and mechanisms are underlying the interventions’ effect or lack thereof. Thus, it is not possible to draw conclusions about which parts of the interventions or which circumstances influenced the outcomes.

Only a few studies focused on nurse-led follow-up, therefore, the same studies were cited in the reviews, forming the basis for the conclusions on effectiveness. This overlap must be considered when interpreting the results. No review focused on supporting people with advanced disease and at end of life demonstrating the need for RCTs in this field and a systematic review on the growing evidence (e.g. Schenker et al., 2021).

Limitations

The scope of the search was extensive to address the umbrella review questions without restrictions to a specific cancer type, specialist nurse intervention, outcome, or instrument used, leading to different results. Thus, the heterogeneity of reviews limited their comparability. As the searches were conducted in just a few databases and only English and German articles were considered, this may pose a risk of not all relevant reviews being identified and included in the review. On the other hand, since the most important databases were consulted and additional grey literature was included, this risk is considered low.

It is noted that gaps or lack of currency in evidence undermine an overview of reviews; hence umbrella reviews are assumed to be only as good as the systematic reviews and the included individual original/primary studies they are based on (Thomson et al., 2010). This was also evident in this umbrella review. Interventions and the nurses’ qualifications were often poorly described and lacked detail, hindering their systematic analysis. Because of the aggregated data and the complex nature of interventions, the classification of tasks according to the Omaha System Intervention Scheme as well as the assignment to the categories of care across the cancer trajectory was challenging, resulting in overlapping categories and leaving room for interpretation. For example, most of the reported case management tasks also included educational aspects. An error-proneness due to poor reporting and missing information cannot be ruled out. The use of the Omaha System Intervention Scheme, although an established and widely used tool to categorise nursing interventions, may have simplified the scope of some interventions. In addition, the summary presentation of intervention types and outcomes, given in Table 2, does not allow any statements on causality, except for the meta-analyses.

Future research

In this umbrella review, a wide variety of specialised tasks could be identified leading to the requirement of ANP and specialisation through advanced training courses. Although several reviews on nurse-led case management were located (Joo & Liu, 2019; Wu et al., 2021), they did not meet our inclusion criteria because they included observational studies next to RCTs with no subanalysis. Future research should address this aspect by conducting more methodological high-quality studies, especially RCTs, on the effectiveness of nurse-led case management in cancer care.