Summary of findings for the main comparison. Summary of qualitative findings.
| Summary of review finding | Studies contributing to the review finding | CERQual assessment of confidence in the evidence | Explanation of CERQual assessment |
| Type of task | |||
| Recipients of care had mixed views about the expansion of tasks undertaken by nurses. They preferred doctors when the tasks were more 'medical' in nature and they accepted nurses for preventive care and follow‐ups. | Bennett 2013; Boyle 2016; Branson 2008; Cheek 2002; Clendon 2001; Clendon 2003; Coker 2009; Courtenay 2010; Flowers 2008; Leipert 2011; Perry 2005; Rosemann 2006 | Moderate confidence | Due to minor concerns about methodological limitations and moderate concerns about relevance. |
| Doctors in most studies also preferred that nurses performed only non‐medical tasks. | Abbott 2013; Bailey 2006; Branson 2008; Coulter 2000; Georgeu 2012; Ivers 2011; Kraus 2017; Lindblad 2010; Lorch 2015; Marsden 2004; Rosemann 2006; Ross 2015; Stenner 2010; Stephen 2018; Twinn 1999; Voogdt‐Pruis 2011 | Moderate confidence | Due to minor concerns about methodological limitations and relevance; and moderate concerns about coherence |
| Nurses were comfortable with, and believed they were competent to deliver, a wide range of tasks, but particularly emphasised tasks that were more health promotive/preventive in nature. | Abbott 2013; Albers‐Heitner 2011; Bailey 2006; Carryer 2017; Dennis 2016; Francis 2013; Georgeu 2012; Hamel 2017; Hart 2012; Kraus 2017; Lindblad 2010; Peterson 2007; Stephen 2018 | Moderate confidence | Due to minor concerns about methodological limitations, adequacy and relevance |
| Accessibility and quality of care | |||
| Recipients in most studies believed that nurses were more easily accessible than doctors. | Basaleem 2009; Cheek 2002; Coker 2009; Fortin 2010; Georgeu 2012; Leipert 2011; Marsden 2004; Perry 2005; Ross 2015; Stenner 2011 | High confidence | — |
| Both doctors and nurses saw doctor‐nurse substitution and collaborative practice as a way of increasing quick access to care for certain tasks such as maternity care and prescriptions. | Kaasalainen 2013; Ljungbeck 2017; Lovink 2018; Perry 2005; Peterson 2007; Poghosyan 2017 | Moderate confidence | Due to minor concerns about methodological limitations and relevance; and moderate concerns about adequacy |
| Recipients of care in most studies were satisfied with nurses' social skills. Recipients' perceptions of nurses' technical skills were mixed. | Bennett 2013; Boyle 2016; Branson 2008; Coker 2009; Corneli 2008; Dennis 2016; Duane 2015; Fortin 2010; Friman 2011; Hart 2012; Leech 2007; Leipert 2011; Parfitt 2007; Peterson 2007; Ross 2015; Stenner 2011; Stephen 2018 | Very low confidence | Due to minor concerns about methodological limitations; and serious concerns about coherence |
| Health professionals, including doctors, nurses, policymakers and other healthcare providers, believed that doctor‐nurse substitution led to improvements in the quality of care. | Abbott 2013; Boyle 2016; Carryer 2017; Coulter 2000; Dierick‐van Daele 2010a; Kaasalainen 2013; Leipert 2011; Ljungbeck 2017; Lorch 2015; Marsden 2004; Nkhata 2016; Perry 2005; Rustagi 2015a; Stenner 2010 | Moderate confidence | Due to minor concerns about methodological limitations and coherence |
| Doctor‐nurse communication | |||
| A close doctor‐nurse relationship characterised by trust and mutual respect helped nurses to expand and develop their roles. | Burns 2009b; Francis 2013; Georgeu 2012; Hamel 2017; Lovink 2018; Mills 2008a; Peterson 2007; Poghosyan 2017; Schadewaldt 2016; Vetter‐Smith 2012; Voogdt‐Pruis 2011 | Moderate confidence | Due to moderate concerns about methodological limitations and minor concerns about relevance |
| Nurses might find it difficult to communicate effectively with colleagues in stand‐alone practices or vertical programmes of care. | Basaleem 2011; Broyles 2012; Flowers 2008; Rustagi 2015a; Walker 2015 | Moderate confidence | Due to moderate concerns about methodological limitations; and minor concerns about relevance and adequacy |
| Doctors' trust in and acceptance of nurses was a critical factor that shaped the extent of nursing practice. | Abbott 2013; Bailey 2006; Burns 2009b; Coulter 2000; Dennis 2016; Duane 2015; Francis 2013; Friman 2011; Georgeu 2012; Hamel 2017; James 2003; Kraus 2017; Leech 2007; Lindblad 2010; Mabelane 2016; Mills 2008a; Ross 2015; Stenner 2010 | Moderate confidence | Due to minor concerns about methodological limitations; and minor concerns about relevance |
| Financial issues might damage the relationship between doctors and nurses. | Coulter 2000; Lovink 2018; Mills 2008a; Peterson 2007; Poghosyan 2017; Ross 2015; Schadewaldt 2016 | Moderate confidence | Due to minor concerns about methodological limitations, coherence and adequacy; and moderate concern about relevance |
| Educational and training system | |||
| Nurses felt they had gained additional skills through task‐shifting. However, they believed that further training and education could increase their skills, job satisfaction and motivation; allow them to work more independently; and increase others' acceptance of their professional roles. | Albers‐Heitner 2011; Burns 2009b; Courtenay 2010; Dennis 2016; Duane 2015; Francis 2013; Friman 2011; Furin 2011; Hart 2012; Ivers 2011; Kassean 2005; Lindblad 2010; Maddox 2016; Mills 2008a; Mills 2008b; Mkhabela 2008; Rustagi 2015a; Stenner 2010; Stenner 2011 | Moderate confidence | Due to minor concerns about methodological limitations and relevance |
| Nurses had concerns about their training in terms of adequacy, equity and quality. | Broyles 2012; Drew 2002; Drew 2003; Francis 2013; Georgeu 2012; Hart 2012; Mabelane 2016; Maddox 2016; McKenna 2015; Nkhata 2016 | Moderate confidence | Due to minor concerns about methodological limitations and relevance |
| Awareness and understanding of the strategy | |||
| Recipients of care in many studies had limited knowledge about nurses' roles in primary care, nurse models of care, and any differences between nurse‐led and doctor‐led care. | Basaleem 2009; Branson 2008; Cheek 2002; Clendon 2001; Halcomb 2013; Leipert 2011; Lovink 2018 | Moderate confidence | Due to moderate concerns about relevance and methodological limitations |
| Continuity of care | |||
| Doctors in some studies felt that doctor‐nurse substitution improved the continuity of care and believed that recipients of care would prefer to see the same nurse rather than different doctors. | Marsden 2004; Ross 2015 | Moderate confidence | Due to moderate concerns about adequacy and relevance |
| Recipients of care in some studies were concerned over the continuity of care provided by nurses and felt insecure if they lost contact with their doctors. | Branson 2008; Fortin 2010; Georgeu 2012; Stephen 2018 | Low confidence | Due to minor concerns about methodological limitations, and moderate concerns about relevance and adequacy |
| Motivation and incentives | |||
| Internal motivators most frequently cited by nurses regarding task‐shifting were psychological (including personal development and being respected) and professional (improving the quality of care). | Albers‐Heitner 2011; Burns 2009b; Coulter 2000; Drew 2002; Drew 2003; Friman 2011; Furin 2011; Georgeu 2012; Hamel 2017; James 2003; Ljungbeck 2017; Petrova 2015; Ross 2015 | High confidence | — |
| Nurses believed that external motivators such as improved working conditions and financial incentives could act as an incentive to take on more responsibilities. | Flowers 2008; Francis 2013; Furin 2011; Hamel 2017; Hart 2012; Ljungbeck 2017; McKenna 2015; Mills 2008a; Nkhata 2016 | Moderate confidence | Due to moderate concerns about methodological limitations and minor concerns about relevance |
| Doctors valued the contribution of nurses in collaborative practices when this reduced their own workload. | Coulter 2000; Dierick‐van Daele 2010a; Drew 2002; Drew 2003; Georgeu 2012; Hamel 2017; Kaasalainen 2013; Ljungbeck 2017; Lorch 2015; Lovink 2018; Marsden 2004; Peterson 2007; Stenner 2010 | Moderate confidence | Due to minor concerns about methodological limitations and relevance; and moderate concerns about relevance and coherence |
| In settings where a proportion of doctors' revenues came from fee‐for‐service payments, doctors expressed negative reactions towards doctor‐nurse substitution. | Coulter 2000; Lorch 2015; Peterson 2007 | Low confidence | Due to minor concerns about methodological limitations; moderate concerns about adequacy and serious concerns about relevance |
| Resources (financial, infrastructures, facilities, and drugs and equipment) | |||
| A shortage of resources, including human resources, equipment and supplies, and lack of equity in how organisational resources were allocated, sometimes negatively impacted on the effective implementation of doctor‐nurse substitution strategies. | Abbott 2013; Basaleem 2009; Basaleem 2011; Coker 2009; Flowers 2008; Friman 2011; Leech 2007; Mabelane 2016; Mills 2008a; Mkhabela 2008; Nkhata 2016; Poghosyan 2017; Schadewaldt 2016; Vetter‐Smith 2012; Voogdt‐Pruis 2011; Walker 2004 | High confidence | — |
| Recipient of care flow processes and referrals | |||
| An appropriate referral system for recipients of care was important for the effective implementation of doctor‐nurse substitution strategies. | Basaleem 2011; Bennett 2013; Duane 2015; Lovink 2018 | Moderate confidence | Due to minor concerns about methodological limitations, relevance and adequacy |
| Management and leadership vision | |||
| Experienced leadership was a facilitator of smooth implementation of doctor‐nurse substitution strategies. | Burns 2009a; Leech 2007; Ljungbeck 2017; Mills 2008b; Petrova 2015; Poghosyan 2017 | High confidence | — |
| Nurses and recipients reported dissatisfaction with the huge number of documents and reports that needed to be completed in connection with doctor‐nurse substitution strategies. | Basaleem 2011; Flowers 2008; Georgeu 2012 | Moderate confidence | Due to minor concerns about methodological limitations and moderate concerns about adequacy |
| Doctor‐nurse professional boundaries and role clarity | |||
| Clear role definitions were critical in the successful implementation of doctor‐nurse substitution strategies. | Coulter 2000; Drew 2002; Drew 2003; Flowers 2008; Hamel 2017; Kraus 2017; Lindblad 2010; Lovink 2018; McKenna 2015; Mills 2008a; Peterson 2007; Poghosyan 2017; Schadewaldt 2016; Stephen 2018 | Moderate confidence | Due to minor concerns about methodological limitations and moderate concerns about relevance |
| Supervision | |||
| Where nurses were supervised by doctors, the quality of this supervision was central to the building of confidence in both partners. | Coulter 2000; Courtenay 2010; Drew 2002; Drew 2003; Kassean 2005; Kraus 2017; Lindblad 2010; Ljungbeck 2017; Mkhabela 2008 | Moderate confidence | Due to minor concerns about methodological limitations and adequacy |
| Nurses in LMIC settings appeared to lack effective supervision. | Basaleem 2011; Leech 2007 | Very low confidence | Due to serious concerns about adequacy |
aAdopted from the SURE Collaboration 2011; World Health Organization.
CERQual: Confidence in the Evidence from Reviews of Qualitative research; LMIC: low‐ to middle‐income country.