Midwifery 29 (2013) 12111221

Contents lists available at ScienceDirect

Midwifery
journal homepage: www.elsevier.com/midw

A systematic review of qualitative evidence on barriers and facilitators

to the implementation of task-shifting in midwifery services
Christopher J. Colvin a,n, Jodie de Heer a, Laura Winterton a, Milagros Mellenkamp b,
Claire Glenton c, Jane Noyes d,e, Simon Lewin c,a, Arash Rashidian f
a

University of Cape Town, South Africa

University of Virginia, VA, USA
c
Norwegian Knowledge Centre for Health Services, Norway
d
Medical Research Council of South Africa, South Africa
e
Bangor University, UK
f
Knowledge Utilization Research Center and School of Public Health, Tehran University of Medical Sciences, Tehran, Iran
b

art ic l e i nf o

a b s t r a c t

Article history:
Received 7 January 2013
Received in revised form
3 May 2013
Accepted 7 May 2013

Objective: to synthesise qualitative research on task-shifting to and from midwives to identify barriers
and facilitators to successful implementation.
Design: systematic review of qualitative evidence using a 4-stage narrative synthesis approach. We
searched the CINAHL, Medline and the Social Science Citation Index databases. Study quality was
assessed and evidence was synthesised using a theory-informed comparative case-study approach.
Setting: midwifery services in any setting in low-, middle-, and high-income countries.
Participants: midwives, nurses, doctors, patients, community members, policymakers, programme
managers, community health workers, doulas, traditional birth attendants and other stakeholders.
Interventions: task shifting to and from midwives.
Findings: thirty-seven studies were included. Findings were organised under three broad themes: (1)
challenges in dening and defending the midwifery model of care during task shifting, (2) training,
supervision and support challenges in midwifery task shifting, and (3) teamwork and task shifting.
Key conclusions: this is the rst review to report implementation factors associated with midwifery task
shifting and optimisation. Though task shifting may serve as a powerful means to address the crisis in
human resources for maternal and newborn health, it is also a complex intervention that generally
requires careful planning, implementation and ongoing supervision and support to ensure optimal and
safe impact. The unique character and history of the midwifery model of care often makes these
challenges even greater.
Implications for practice: evidence from the review fed into the World Health Organisation's Recommendations for Optimizing Health Worker Roles to Improve Access to Key Maternal and Newborn Health
Interventions through Task Shifting guideline. It is appropriate to consider task shifting interventions to
ensure wider access to safe midwifery care globally. Legal protections and liabilities and the regulatory
framework for task shifting should be designed to accommodate new task shifted practices.
& 2013 Elsevier Ltd. All rights reserved.

Keywords:
Task shifting
Organisation of care
Systematic review
Qualitative research

Introduction
A key obstacle to the achievement of the maternal and child
health-related Millennium Development Goals (46) is the chronic
shortage and maldistribution of health workers in many countries

Corresponding author.
E-mail addresses: cj.colvin@uct.ac.za (C.J. Colvin), jddeheer@hotmail.com
(J. de Heer), winterton.laura@gmail.com (L. Winterton), mmm5cb@virginia.edu
(M. Mellenkamp), claire.glenton@kunnskapssenteret.no (C. Glenton),
jane.noyes@bangor.ac.uk (J. Noyes), simon.lewin@kunnskapssenteret.no (S. Lewin),
arashidian@tums.ac.ir (A. Rashidian).
0266-6138/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.midw.2013.05.001

(WHO, 2010). One important approach to addressing this human

resource problem is the redistribution of tasks between health workers, an idea sometimes referred to as task-shifting or task optimisation. Task shifting is one way of addressing the broader question of
the most effective and efcient skill mix in a health services context,
especially in settings with chronic shortages of health workers. By reorganising tasks and responsibilities more efciently and effectively
within the health workforce, policymakers hope to make better use of
existing human resources and expand and strengthen coverage of key
health interventions (WHO et al., 2007).
Midwives are a cadre of health worker that has long been
familiar with the concept of task shifting and its attending

1212

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

opportunities and challenges. The notion of the triple gap of

competencies, coverage and access to midwifery care recently
identied in the State of the World's Midwifery report speaks
succinctly to the global human resource crisis in maternal and
newborn health (UNFPA, 2011). In order to provide critical cover
for some of these gaps, midwives have long worked in complex
and often shifting and ambiguous relationship with other healthcare workers (Sandall, 2012). And the persisting crisis in human
resources for health will continue to put pressure on midwifery
services to move health-care tasks both to and from midwives in
an effort to maximise already thinly stretched human resources.
Important questions persist, however, around task shifting, in
both midwifery services and in other contexts. One set of questions involves the safety and effectiveness of task shifting. There is
growing evidence, from primary research and from quantitative
systematic reviews of effects that task shifting can be safe and
effective (Dovlo, 2004; Lewin et al., 2010; Bhutta et al., 2011;
Fulton et al., 2011; Pyone et al., 2012). This evidence is, however,
often mixed or ambiguous, with heterogenous effects and a wide
variety of methodological quality being the norm.
One reason for these weaknesses in the effectiveness evidence
is that the reorganisation of tasks among health workers is closer
to a complex health systems intervention than a narrow clinical
intervention. These kinds of complex interventions are more
difcult to assess empirically. It is also becoming clear that the
safety or effectiveness of task shifting depends as much on the
implementation and ongoing management of task shifting as it
does on the nature of the technical tasks being shifted (Callaghan
et al., 2010; Georgeu et al., 2012).
Addressing questions of implementation, however, requires a
different form of evidence, one focused on process, context and
mechanism. Process evaluations of task shifting interventions
(Glenton et al., 2011), and evidence syntheses of qualitative
evidence on task shifting, are required for understanding how
and why task shifting interventions might succeed in some
settings and not in others.
Given the uncertainty around the implementation, safety, and
effectiveness of some forms of task shifting, especially in critical
maternal and newborn health interventions, the WHO recently set
out to assess the relevant evidence in order to develop guidance
on task shifting in this context. This review is one of a series of
reviews that was used in the development of the WHO's recent
Recommendations for Optimizing Health Worker Roles to
Improve Access to key Maternal and Newborn Health Interventions through Task Shifting (OPTIMIZEMNH) (WHO, 2012) (http://
www.optimizemnh.org). While the more traditional quantitative
reviews used in the process assessed the evidence on safety,
effectiveness and efciency of task shifting initiatives in maternal
and newborn health, qualitative reviews like this one assessed
evidence regarding the barriers and facilitators to successful
implementation of task shifting.
This was the rst time that the WHO has included systematic
reviews of qualitative evidence in its ofcial guidelines. The aim of this
review on midwifery and task shifting was to synthesise qualitative
research on task shifting to and from midwives in order to identify
barriers and facilitators to their successful implementation.

methodology for summarising data from multiple qualitative

studies (Noyes, 2009). We used a four-stage narrative synthesis
design (Popay et al., 2006) using thematic analysis informed by the
SURE conceptual framework (described further below) with comparative case analysis across low, middle and high income
contexts.
Study inclusion criteria
Types of study methodology
Studies including any type of qualitative method of data
collection and analysis. Mixed method studies were eligible
provided it was possible to extract the ndings derived from
qualitative research.
Types of studies and settings
Studies from low-, middle-, and high-income countries (LMICs
and HICs) were included. Studies conducted in hospitals, clinics,
and communities were included as long as midwives were a
central part of the reorganisation of tasks under review. Study
participants could include midwives, nurses, doctors, patients,
community members, policymakers, programme managers, community health workers, doulas, traditional birth attendants (TBAs)
and other stakeholders.
Types of interventions
Studies reporting on interventions addressing specic taskshifting initiatives between midwives and either other health
workers or other birth attendants or community-based health
volunteers. Working from the International Confederation of
Midwives' denition of a midwife (2011), we understood a midwife to mean a skilled health-care worker with one or more years
of legally recognised and regulated training, usually at the level of
a registered nurse, who delivers antenatal care, delivery and
postnatal care to women. As discussed below, some midwives
may not frame their work as primarily biomedical in focus but we
were interested for the purposes of this review on task-shifting
initiative in midwives who were recognised in some way as a
formal part of the biomedical health system. We did not include
traditional, community, lay, or village midwives or other nonbiomedically trained TBAs in this denition of midwife. We did
include studies, however, where tasks were shifted between
biomedically-trained midwives and these other forms of midwives
and TBAs.
Deciding what constituted a concrete task-shifting intervention
was sometimes difcult given that there are no commonly
accepted criteria for identifying which new tasks signicantly
expand or reduce the scope of work for midwives. We read
abstracts and full texts of articles to determine (a) if there were
specic tasks being shifted from one cadre to another (rather than
the creation of new tasks not previously done by anyone) and (b) if
the report authors or the study participants described the new
tasks as representing a signicant shift in their previous roles and
responsibilities. We also limited this review to studies that were
specically relevant to the maternal and newborn health focus of
the OptimizeMNH guidelines.

Methods
Review design
We undertook a qualitative systematic review. As with systematic reviews of effectiveness, reviews of qualitative data should
be carried out in a systematic and transparent way and the last
few years have seen signicant development in systematic review

Exclusions
We excluded (a) studies assessing general attitudes of stakeholders to midwives or to task shifting in the absence of a specic
task-shifting initiative, (b) studies proposing training programmes
or detailing training needs and curriculum requirements in the
absence of empirical research on task-shifting interventions themselves, and (c) studies assessing the internal reorganisation of

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

midwifery practices without any reference to new tasks shifted

between midwives and other cadres.

Search strategy and selection process

Search strategy
To identify eligible studies, we searched CINAHL, Medline and
the Social Science Citation Index. For the database searches, we
used variations of the term midwife as well parteras and sagefemme. We included French, Spanish and English studies in the
search, though no French or Spanish studies from these databases
were found to meet the inclusion criteria.
We also included three termsndings, interviews, and qualitativeas a way of identifying studies with qualitative data (see
http://www.york.ac.uk/inst/crd/intertasc/qualitat6.htm
for
a
description of this search strategy). We did not include terms
related to task shifting or skill mix since initial testing revealed
that these terms were very rarely used in studies on midwives.
There were no restrictions on date of publication. The search
extracted 800 records from CINAHL (excluding Medline matches),
2097 records from Medline, and 3002 records from SSCI (see
Supplementary le A for the search strategies).

Study selection
Assessment of the eligibility of studies can vary between
review authors. We addressed this challenge in two stages. First,
three review authors independently assessed the rst 200
abstracts for inclusion in the review. Each reviewer's list of
included articles and accompanying rationale was compared with
the list from the other reviewers and discrepancies were discussed
and resolved. Inclusion and exclusion criteria were also rened
and claried during this process. The remaining abstracts were
then independently reviewed by two reviewers and the lists then
compared again and resolved. Finally, a third reviewer consulted
on discrepant cases and also examined the notes kept by the other
two reviewers to ensure accuracy and consistency.

Quality assessment and data extraction

Assessment of study quality
Two researchers assessed the quality of included studies using
an adaptation of the Critical Appraisal Skills Programme (CASP)
quality-assessment tool for qualitative studies (see http://www.
casp-uk.net for more information). However, no studies were
excluded on the basis of quality assessment. Rather the quality
assessment process was used to identify weaknesses in study
methodologies and better interpret and assess their ndings. We
describe below (in the Data synthesis section) how we integrated
these study quality assessments into the interpretation of the
review ndings.

Data extraction and management

Once the study selection process was concluded and quality
assessed, studies were divided among three review authors who
then extracted the data from the studies using standard templates.
Initial data extraction captured the study characteristics, including
setting, participants, type of task shifting reviewed, training and
levels of education required, etc. Subsequent data extraction
collated the qualitative ndings of the studies using a template
derived from the SURE framework for assessing factors affecting
the implementation of health systems interventions (see below,
SURE Collaboration, 2011).

1213

Data synthesis
The reviewers synthesised the qualitative evidence using a
broadly comparative case-study approach informed by tools and
techniques outlined in the narrative synthesis frameworksee
Table 1 (Popay et al., 2006).
The rst step in this approach is to develop a preliminary
synthesis of the ndings using a starting conceptual framework.
For this review, initial data extraction was performed using a
template informed by the SURE framework. Published lists of
barriers for implementing changes in health care often show a
high degree of overlap (Oxman and Flottorp, 2001; Hanson et al.,
2003; Grol et al., 2007; Fretheim et al., 2009). As part of another
project, these lists were reviewed and a theory-informed framework for identifying barriers to implementing a policy option was
developed (SURE Collaboration, 2011).
This SURE framework includes the following factors:
(a) knowledge and skills; attitudes regarding programme acceptability, appropriateness and credibility; and motivation to change
or adopt new behaviours among recipients of care, providers of
care, and other stakeholders; (b) health system constraints
(including accessibility of care, nancial resources, human
resources, educational system, clinical supervision, internal communication, external communication, allocation of authority,
accountability, management or leadership (or both), information
systems, facilities, patient ow processes, procurement and distribution systems, incentives, bureaucracy, and relationship with
norms and standards); and (c) social and political constraints
(including ideology, short-term thinking, contracts, legislation or
regulations, donor policies, inuential people, corruption, and
political stability).
Three reviewers used this template to extract key ndings and
identify emerging themes across the studies. The denitions and
boundaries of each of these emerging themes were discussed
among the authors throughout the data extraction process. Finally,
one author synthesised these themes and categories into an
overall framework for facilitators and barriers that was then
reviewed by the other review authors for accuracy and
comprehensiveness.
The second step was to explore whether certain factors related
to the intervention or the context of the intervention were
associated with differences in the facilitators and barriers that
we had identied. Specically, we explored whether there were
differences between HICs and LMICs in the barriers and facilitators
identied, and whether there were differences between taskshifting initiatives involving simple tasks and ones delivering
more complex tasks.
The authors used the SURE framework to guide the analysis for
two reasons. First, it provides a broad but relatively comprehensive list of possible factors that could inuence implementation.
Second, the other reviews of qualitative research that were part of
the OptimizeMNH guidelines used the framework as a way to
facilitate cross-cutting analysis of task shifting barriers and facilitators across the reviews. While we used the framework as a
guide, our application of it was not rigid and the nal synthesis of
ndings is only loosely based on its categories.

Assessment of certainty of ndings

There are few methods for assessing the certainty of the
ndings in systematic reviews of qualitative evidence. To assess
how much certainty can be placed in the qualitative evidence for
each review nding, we have therefore chosen to apply an
innovative approach, which we refer to as the CerQual (certainty
of the qualitative evidence) approach. This approach was developed by a number of authors during the evidence syntheses for

1214

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

Table 1
Four-stage review design (Popay et al., 2006).
Main elements of 4 stage narrative synthesis

1. Developing a theoretical model of how the

interventions work, why and for whom

2. Developing a preliminary synthesis

Purpose

Tools and
frameworks
used

 To inform decisions about the review question and what types of studies to review
 To contribute to the interpretation of the review's ndings
 To assess how widely applicable those ndings may be

The SURE
framework
MDGs 4,5,6

 To organise ndings from included studies in order to:

Thematic
analysis

Identify and list the facilitators and barriers to implementation reported

Explore the relationship between reported facilitators and barriers

3. Exploring relationships in the data

 To consider the factors that might explain any differences in the facilitators and/or barriers Comparative
case analysis

to successful implementation across included studies

 To understand how and why interventions have an effect

4. Assessing the robustness of the synthesis
product

 To provide an assessment of the strength of the evidence for drawing conclusions about the CerQual tool
facilitators and/or barriers to implementation identied in the synthesis. Generalising the
product of the synthesis to different population groups and/or contexts

the Optimize4MNH guideline development process and is

described in more detail elsewhere (Glenton et al., 2013).
The CerQual approach uses the term certainty to indicate how
likely it is that the review nding happened in the contexts of the
included studies and could happen elsewhere. Certainty is thus
composed of two factors: the methodological quality of the
individual studies contributing to a review nding and the
plausibility of each review nding.
We used three levels to indicate the certainty of the qualitative
evidence. Findings with a high certainty were drawn from
generally well-conducted studies showing high levels of plausibility. Moderate certainty ndings signalled concerns regarding
either the methodological quality of the studies or the plausibility
of the review nding. Where there were concerns regarding both
the methodological quality and the plausibility of the review
nding, the nding was classied as of low certainty. Please see
Supplementary le B for a table detailing the full review ndings
along with assessments of their certainty.

Findings

Records identified through

database searching
(n = 6120)

CASP quality
appraisal tool

Additional records identified

through other sources
(n = 38)

Records after duplicates and those

without abstracts removed (n = 5899)

Records screened
(n = 5899)

Records excluded
(n = 5854)

Full-text articles
assessed for eligibility
(n = 45)

Full-text articles
excluded, with reasons
(n = 8)

Studies included in
qualitative synthesis
(n = 37)

Fig. 1. Flow diagram of search and inclusion process.

Overview of study contexts and interventions

A total of 5899 titles and abstracts were identied for screening. We included 37 papers in this review. See Fig. 1 for a ow
diagram of the search and inclusion process and see Table 2 for
characteristics of these included studies. Given the large number
of references to these studies in the narrative below, studies
included in the review will be referenced below in square brackets
by their study ID (found in the rst column of Table 2) rather than
their full in-text citation.
Most studies assessed the perspectives of midwives themselves. Other common participant groups were health workers
and supervisors. A few studies also included recipients of midwifery services.
Of the 37 studies, 26 were based in ve high-income countries:
Australia, Canada, USA, Sweden and the UK. Eleven studies were
based in eight low- and middle-income countries: Angola, Dominican Republic, Guatemala, Jordan, Kenya Indonesia, Mexico, and
South Africa. Most of the studies in high-income countries took
place in urban, hospital settings. The studies from low- and middleincome countries took place in both urban and rural settings and
had a more even mix of community, clinic and hospital contexts.
Midwifery services were run by either NGOs or local and national

governments, sometimes in collaboration, and ranged from small

pilot studies to large-scale national programmes.
Though the term midwife has a commonly accepted, generic
meaning in much of the academic literature, the studies included
reveal a wide range of scopes of practice for midwives and widely
varying practice contexts. The reorganisation of tasks in these
studies also took varying forms. Tasks could be shifted from
midwives (horizontally, to other nurses, or vertically, more junior
health-care workers (HCWs) or to lay health workers) or to
midwives (horizontally, from other nurses, or vertically, from
doctors or other more senior HCWs). The majority of the studies
reviewed shifting of tasks to midwives rather than from midwives
(see Table 3).
The studies described a wide range of factors driving taskshifting initiatives. Many of the European studies situated task
shifting in the context of public service reforms, drives for greater
effectiveness and efciency, and moves towards more decentralised and community-based forms of health services [3, 7, 21].
A few studies, from Sweden [21], the UK [7] and Guatemala [22],
also cited a growing emphasis within biomedical care on neoliberal principles such as throughput, scale, efciency, routinisation

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

1215

Table 2
Characteristics of included studies.
Study details

Task shifting intervention

Intervention context
Country

Study
ID

Author (Year)

Data collection method

Participants

Tasks

Direction
of shift

Akhavan and
Lundgren (2010)
Battersby (2010)

Interviews

Midwives

Interviews, survey

Midwives

From
Sweden
Midwife
To midwife England

Questionnaire

Midwives & GPs

Psychosocial
support
Prevention/
promotion
Antenatal care

Questionnaire

Midwives

Interviews, FGDs, observations,

document reviews
Interviews, document reviews
Interviews, FGDs, observation

Co-ordinators and directors of

birthing centres, midwives
Health care-workers, family and
community members
Community midwives

Dickson-Tetteh and
Billings (2002)
Dietsch (2010)

Interviews, observation

Midwives

Interviews, observation
Interviews, FGDs

11

Doring and Nolte

(1999)
Edwards (2008)

Skilled birth attendants and

traditional midwives
Midwives

Interviews

12

Foster et al. (2005)

Interviews, FGDs, observation

13

Foster et al. (2006)

Interviews, observation

14

Foster et al. (2004)

15

Fulton (2002)

Interviews, observation,
document reviews
Document reviews

16

Grifn et al. (2009)

Questionnaire

17

Interviews, observation

18
19
20
21

Kennedy and Lyndon

(2008)
Lindberg et al. (2005)
Lipp (2008)
Lumsden (2005)
Lundgren (2010)

22

Maupin (2008)

2
3
4
5
6
7
8
9
10

23
24

Battersby and
Thomson (1997)
Bench and Fitzpatrick
(2007)
Collin et al. (2000)
DAmbruoso et al.
(2009)
Deery (2005)

FGDs
Interviews
Interviews
Interviews

Interviews, observation,
document review
McKenna et al. (2003) Interviews

McNeill and Alderdice

(2009)
Mitchell (2003a
Mitchell (2003b)
Persson et al. (2011)

Interviews

Interviews

31
32

Pettersson et al.
(2001)
Prowse and Prowse
(2008)
Reiger and Lane
(2009)
Rogers et al. (2003)
Shaban et al. (2012)

Interviews
FGDs

33

Smid et al. (2010)

Interviews

34

Stevens et al. (2011)

Interviews

35

Tsouroui (2011)

Interview, observation, survey

36

Walker et al. (2004)

FGDs

37

Widmark et al. (1998) Interviews

25
26
27
28
29
30

Interviews
Interviews
Interviews

Interviews, observation, survey

Interviews, FGDs

Advanced
clinical skills
Advanced
clinical skills
Advanced
clinical skills
Advanced
clinical skills
Abortion
Neonatal care

Advanced
clinical skills
Midwives
Advanced
clinical skills
Auxiliary nurses
Advanced
clinical skills
Auxiliary nurses
Advanced
clinical skills
Nurses, traditional midwives
Advanced
clinical skills
Midwives
Advanced
clinical skills
Maternity support worker
Psychosocial
support
Nurse and midwife
Psychosocial
support
Midwives
Neonatal care
Midwives and nurses
Abortion
Midwives
Neonatal care
Primparous and multiparous women Psychosocial
support
Midwives and community members Advanced
clinical skills
Midwives and midwife students
Psychosocial
support
Midwives
Prevention/
promotion
Midwives
Neonatal care
Midwives
Neonatal care
Midwives
Prevention/
Promotion
Midwives
Psychosocial
support
Midwives
Psychosocial
support
Midwives and doctors
Psychosocial
support
Midwives
Neonatal care
Midwives
Advanced
clinical skills
Doctors, other health-care workers
Psychosocial
support
Midwives and doulas
Psychosocial
support
Pregnant women, midwives,
Prevention/
healthcare assistants
Promotion
Midwives
Advanced
clinical skills
Midwives
Prevention/
promotion

Setting

HIC/
LMIC

Hospital

HIC

Hospital

HIC

To midwife England

Other

HIC

To midwife England

Hospital

HIC

To midwife Canada

Clinic

HIC

To midwife Indonesia

Clinic

LMIC

To midwife England

Hospital

HIC

To midwife South Africa

Hospital

LMIC

From
Kenya
midwife
From
South Africa
Midwife
To midwife England

Hospital

LMIC

Hospital

LMIC

Hospital

HIC

Hospital

LMIC

From
midwife
From
midwife
From
midwife
To midwife

Dominican
Republic
Dominican
Republic
Guatemala

Hospital

LMIC

Hospital

LMIC

England

Hospital

HIC

From
England
midwife
To midwife USA

Hospital

HIC

Hospital

HIC

Sweden
Wales
England
Sweden

Other
Hospital
Hospital
Home

HIC
HIC
HIC
HIC

Guatemala

Home

LMIC

Hospital

HIC

Hospital

HIC

Clinic
Hospital
Clinic

HIC
HIC
HIC
LMIC

To midwife England

Peripheral
sites
Hospital

HIC

To midwife Australia

Hospital

HIC

To Midwife England
To midwife Jordan

Hospital
Hospital

HIC
LMIC

To midwife Mexico

Hospital

LMIC

From
Australia
midwife
To midwife England

Hospital

HIC

Hospital

HIC

To Midwife Australia

Hospital

HIC

To midwife Sweden

Clinic

HIC

To Midwife
To Midwife
To midwife
From
midwife
To midwife

From
England
midwife
To midwife Northern
Ireland
To midwife England
To midwife England
To midwife Sweden
To midwife Angola

1216

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

Table 3
Tasks shifted and direction.
Tasks shifted to midwives

Tasks shifted from midwives

High dependency care/managing chronic or critical illness in pregnant women

Midwife-led care where the midwife is responsible for overall care
Neonatal care to maintain a continuity of care with midwife, mother and child
Genetic screening and counseling
Abortion services

Emotional support during labour

Clerical/administrative work
Some clinical management duties (like monitoring postpartum bleeding)

and what one study from the UK cited as the industrial model of
childbirth (Kirkham, 2003, cited in 7).
Studies from LMICs tended to focus less on efciency gains and
acute hospital-based care and more on simply covering major gaps
in obstetric care in their health system as a motivation for task
shifting [1, 12, 20, 32, 33, 34, 36]. In the case of abortion, pressures
to shift abortion-related tasks from doctors to midwives came
from political forces, client demand, lack of access in rural areas,
and a shift to medical abortion that made it easier for midwives to
deliver abortion services in the community [8, 19, 24].
Finally, task shifting emerged in several of the studies as
something that often happened without an ofcial plan or formal
rationale in place. In these studies, lack of staff, poor access to care,
poor outcomes, or unclear divisions of roles and responsibilities
typically resulted in what one study described as ad hoc forms of
task shifting among health staff [8, 19, 24].
Dening and defending the model: complex reactions to task shifting
in midwifery
Many of the challenges of task shifting described in the studies
centred the idea of a midwifery model of care and the tensions
that task shifting brought to bear on the ideals and relationships
embedded in this model.
Contrasting models of birth
A strong nding across many of the studies, and in particular
the ones from HICs, involved the construction of the midwifery
model of care and its sometimes profound conicts with what
was generally characterised as the biomedical model of care. One
study, for example, argued that this midwifery model of care rests
on the conviction that pregnancy is a natural and non-pathological
process; midwives favour a comprehensive approach to expectant
women, emphasizing preventive and qualitative dimensions of
care and encouraging the participation of women and their
spouses [5, p. 16].
By contrast, the medical model, often alluded to but poorly
described in these studies, is said to be organised around concepts
of pathology and abnormality, concerned with managing risk and
liability, and devoted to protecting the status of scientic medical
knowledge and technology. Features of these two ideal-typical
models are compared in Table 4 (Table 4).
Most of the studies in HIC settings described how tensions
between these models impacted the reorganisation of tasks to and
from midwives. Task-shifting initiatives in midwifery services
were thus often experienced as either an afrmation or a threat
to the perceived midwifery model.
It was an emphasis on the continuity, holism and womancenteredness of care between mother and midwife that had the
most impact on whether shifting tasks to or from midwives was
perceived as acceptable and feasible to those involved. Task
shifting that increased the ability of midwives to provide more
holistic or continuous care was readily accepted in principle by
midwives and mothers alike. Example of this kind of task shifting
included the performance by midwives of neonatal examinations

[20, 25, 26, 31] and the move to midwife-led models of care found
in, for example, birthing centres run by midwives [3, 5].

Practical complications of task shifting for midwives

In practice, however, even forms of task shifting that were
consistent with the midwifery model often presented other
complications. Most of the studies, in fact, reported that task
shifting put pressure on the midwifery model of care regardless of
whether midwives supported the practical or ideological justications for the changes.
For example, team midwifery and maternity unit team
approaches, intended to improve integration of services, communication, and continuity of care, were criticised by midwives in
some studies as disrupting the relationship between midwife and
mother [3, 7, 23, 34]. In another study, baby-friendly initiatives to
promote breast feeding and best practices infant care were
accepted by midwives in principle but also prevented them from
providing mothers sufcient information on feeding alternatives
and hampered their ability to solicit, respect and support the
feeding choices made by mothers [2].
There were a large number of studies in which new clinical
tasks were shifted to midwives that were either more complex,
more focused on illness, or less centrally related to pregnancy and
delivery than their standard practice. These include critical illness
management/high dependency care, management of gestational
diabetes mellitus, genetic screening, cervical cancer screening,
abortion services, and other sexual and reproductive health
services [4, 8, 19, 24, 27, 30, 35, 37].
Although midwives often supported the intention behind many
of these up-skilling initiatives (with abortion generating the most
ambivalence) and derived job satisfaction from them, they also
changed their caring practices and relationships to mothers in
sometime signicant ways. Even the generally well-accepted
shifting of neonatal care and examination before hospital discharge to midwivesa practice that was seen to deepen the
holistic relationship of total care of mother and child [20, 25,
31]also brought concerns over increased workload, uncertain
liability, and fear of missing rare abnormalities [3, 4, 20, 26].
Finally, a number of studies described often quite difcult and
ambivalent relationships between midwives and either doulas,
traditional birth attendants (TBAs) or other birth supporters such
as maternity support workers [21, 29, 33, 34]. The presence of
these additional birth supporters, while consistent with a general
emphasis on a less medicalised, more community- and familybased approach to childbirth [1, 12, 18, 33], nonetheless often
complicated the relationship between midwives and mothers.
In most cases, the use of doulas or TBAs seemed to push the
midwife into a more medical role and the TBA into a more
supportive and advocacy-oriented role [1, 21]. Major conict could
emerge between doulas or TBAs and midwives when midwives
felt as though they were not being respected and heard and TBAs
felt ignored in the labour room [34]. One study even reported
claims of verbal abuse and physical violence between skilled birth
attendants (including midwives) at a hospital and TBAs [9].

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

1217

Table 4
Midwifery and medical models of care.
Midwifery model

Medical model

Pregnancy and birth as normal

Continuous care and support for the mother by one person over a long
period of time
Woman-centred and relationship-based approach to care that prioritises
the experience of the mother
Anticipation and avoidance of unnecessary medical support and
intervention
Value of experiential, embodied, intuitive knowledge
Focus on mother's birth experience
Importance of personal trust between midwife and mother
Provider training focused on the normal
Perception of pain during labour as natural in most cases

Pregnancy and birth as abnormal/dangerous

Care distributed among several health-care providers, generally short-term and
discontinuous
Efciency- and effectiveness-based approach to care that prioritises care as part of a broader
system of health-care provision
Preparation/preference for medical support and intervention
Value of biomedical knowledge
Focus on anticipating and managing risk and liability
Importance of clinical condence of mother in health-care provider
Provider training focused on the abnormal
Perception of pain during labour as a medical symptom to be alleviated.

Although not all interactions between midwives and other

birth supporters produced tension, these relationships generally
proved complicated for the ways they disrupted the (often
idealised) midwifery model of singular care.

Perspectives of other health-care workers and mothers on task

shifting
The perspectives of other health-care workers and of mothers
on the acceptability of task shifting to and from midwives
appeared in these studies but not with great regularity.
Doctors who were not part of a specic task-shifting initiative
involving midwives generally had a more sceptical attitude
towards greater midwife involvement in maternal care, as compared with doctors in other studies who were more directly
involved with and, in turn, supportive of task shifting [3, 25, 30].
In studies of initiatives in which doctors were directly involved,
the picture is slightly more mixed. One study of the introduction of
midwives to a hospital in Quebec reported turf battles, professional conict, distrust and poor integration during the project [5].
In another study [3], however, community midwives and general
practitioners in the UK reported good working relationships and
trust and doctors reported feeling comfortable to refer women to
midwife-led care if requested. Importantly, however, GPs in this
study were the rst point of contact for women entering the
health service and rarely referred women to midwives on their
own initiative.
Mothers expressed a range of opinions about birth care
providers. Some preferred doulas or other TBAs as their primary
birth supporter [1, 6, 9, 33], some preferred midwives [20, 26, 32,
36] and some preferred doctors [31]. Studies from LMICs and with
vulnerable populations in HICs found that women often preferred
the cultural familiarity of doulas or other TBAs over midwives.
Hospital-based midwives or community-based ones that had been
hired and trained through the public health system enjoyed less
approval [6]. Midwife-led care in hospitals enjoyed high acceptability across a range of HICs and LMICs, especially in those that
did not historically offer midwifery or other forms of womancentred care in the health system [3, 34, 36].

Training, supervision and support challenges in midwifery task

shifting
In addition to task shifting's challenges, both ideological and
practical, to the midwifery model of care, there were also important training, supervision and support challenges presented by
task shifting.

Beyond knowledge and skills: initiating and sustaining changes in

practice
Task shifting initiatives generally required some form of training and follow-up support and supervision to be effective. The
studies, however, described signicant challenges in initiating and
effectively sustaining changes in practice among midwives.
Studies that assessed training programmes for new clinical
knowledge and skills found that midwives generally had no
problem absorbing new information and practicing new techniques [4, 8, 15, 19, 20, 23, 24]. Three reported that on-the-job
learning and participatory approaches were the most effective [5,
12, 13]. Midwives expressed the greatest anxiety around
tasks where they were expected to undertake complex new
responsibilities with little substantive training [4, 11]. Midwives
who had taken the direct route in midwifery training, bypassing
the standard nurse-training process, were found in some
studies to be signicantly less condent in handling sick women
than their counterparts who had completed nurse training [4, 11,
22].
Whatever the initial education, experience or training, most of
the studies addressing training argued that ongoing support and
clinical supervision were critical [4, 7, 13, 28]. Most, however, also
described task-shifting initiatives where this kind of support and
supervision were sorely lacking. Even in the high-dependency
units in one study [11], there was no real training, assessment or
education provided and practice on these units appeared uncoordinated and quality of care low. Other programmes, like the
abortion care programme described in another study [8], enjoyed
high-quality initial training and strong buy-in but still suffered
from anaemic follow-up support and supervision.
Several studies also noted the difculty of translating new
knowledge and skills into effective clinical judgment and decisionmaking. In one study [6], for example, a life-saving training
programme in Indonesia improved village midwives' ability to
diagnose life-threatening conditions correctly but did not improve
their clinical management or skills. Midwives in rural Angola were
able to correctly use a partograph but the training had not
improved their critical decision-making during delivery [28]. The
use of protocols as a support to clinical decision-making also
received mixed reviews [4,14].
Lack of co-ordination in the broader work environment could
also make sustaining task shifting difcult. If new skills, like the
ability to recognise and manage postpartum bleeding among
auxiliary nurses, were not accommodated in the scopes of work
of other staff, little often changed in practice [14]. Indeed, there
was a general lack of knowledge among HCWs about the respective training and roles of other HCWs. Doctors, in particular, often
seemed unaware of the knowledge and skill sets of the midwives
they worked with [3, 4, 5, 16, 17, 29, 32, 33].

1218

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

Finally, changes in technical practice also involve changes in

the personal meanings, habits and identities of midwives and
other staff involved in the task shifting. Even in settings where the
ideological aspects of midwifery were not perceived to be under
threat, many studies described difculties in changing entrenched
habits, settled and hierarchical relationships, and deeply personal
meanings attached to the work health-care workers do [7, 18, 22,
29, 31].

Teamwork and task shifting: navigating the interprofessional terrain

A nal set of factors emerged that both enabled and constrained task shifting: teamwork and the interprofessional context.
We identied four factors that affected the function of these
teams: change and uncertainty in the work environment; social
status and power; communication, co-ordination and continuity of
care; and trust, responsibility and accountability.
Change and uncertainty in the workplace

The ambivalence of up-skilling

Many of the reviewed studies addressed task shifting that
could be described as up-skilling of midwives, introducing them
to either more complex/illness-related tasks or to a wider range of
tasks than their initial midwifery training had anticipated. These
included critical illness management/high dependency care, management of gestational diabetes mellitus, genetic screening, cervical cancer screening, abortion services, and other sexual and
reproductive health services [4, 8, 15, 19, 24, 27, 30, 35, 37].
These forms of task shifting entailed a number of advantages
for some of the midwives in these studies, including increased
status and promotion opportunities [3], a sense of achievement
and clinical condence [28, 29], heightened job satisfaction from
being able to help sicker or a greater number of people [27, 31],
improved overall practice and skills [31], and in some cases,
improvements in quality [16] and continuity of care for mothers
[18, 31]. Neonatal examinations were credited in one study with
having nearly all of these advantages [20].
But this up-skilling came with a number of challenges as well.
Some have already been highlighted above. Poor clinical support
and supervision (7), inadequate training [1, 11, 22, 24], haphazard
implementation of new programmes and working relationships
[4], and insufcient educational preparation [1, 4, 11, 35, 37] could
undermine the ability of midwives to practice new skills with
condence. Up-skilling could also be done without any involvement of the midwife in clinical decision-making [11, 37]. Fear of
liability and an unclear regulatory environment were also particularly threatening when the task shifting involved more complex
tasks or engagement with sicker patients [5, 10]. For some, upskilling also meant a threat to continuity of care and/or increases
in workload [3, 26, 31]. Midwives trained, for example, in
advanced resuscitation were likely to face extra work requirements and interruption of their engagement with individual
women birthing at any particular time [15, 29].
Finally, for some, up-skilling represented a threat to concepts,
practices, and relationships embedded in the midwifery model of
care. Genetic screening and counselling for Down's Syndrome, for
example, proved not only technically complex but it also anticipated abnormality in the pregnancy in a way that was foreign to
the midwifery model of care [24, 37]. Midwives here struggled to
balance a belief in the normalcy of pregnancy with the idea that
pregnancy was simultaneously a risky time [3, 5, 17, 35].
Management of gestational diabetes mellitus brought with it
similar concerns about complexity and abnormality. It also
entailed signicant shifts in the relationship between midwife
and mother by asking midwives to ensure that women behaved a
certain way and met a set of medical targets to manage their
condition. Midwives felt they were policing women rather than
supporting them, acting uncomfortably as medical guardians and
moral keepers [17, 27]. Abortion care task shifting also challenged
the conception of what a midwife's work properly was. Even
among strong advocates for abortion rights, abortion worked
against the idea of the normalcy and celebration of pregnancy
central to the midwifery model [19].

Three forms of change and uncertainty in the workplace

emerged as important. First, many health systems have experienced a great deal of change in the role of midwives, especially in
the last 20 years [7, 18, 22, 29, 33, 37]. These changes were
reported to be often poorly co-ordinated and communicated, and
at times chaotic. They have strained midwives emotionally and
physically and increased pressure on midwifery's relationshipbased, woman-centred care by bringing midwives increasingly
into team-based, bureaucratic and technocratic environments.
Second, there appeared to be quite a number of grey areas of
practice for midwives, situations where they undertook tasks
because they had been asked by doctors or other nurses, because
of critical service gaps, or because of the urgency of the situation
[11, 16, 25]. Even in more stable situations, like the shared care
model of antenatal care, what precisely was shared in shared care
was often not clear [3, 5].
Thirdly, and probably most importantly, lack of clarity around
job descriptions, performance assessment, and the policy and legal
contexts for midwifery care were described in several studies as a
barrier to task shifting [5]. One study [16], for example, described a
well-entrenched national maternity support worker programme
in the UK for which there was no nationally approved job
description and no qualications framework. Another [10]
described midwife frustration at poorly worded and communicated policies, lack of specicity in nursing policies with respect to
midwives, pressure from doctors to either do more or less than the
law allows, and ambiguous criteria that allow midwives to perform certain procedures if they feel capable to do them.
Social status and power: hierarchy and interpersonal engagement
Differences in social status and the impact of hierarchical
relationships emerged as another important barrier [3, 24, 33].
Although some studies described these hierarchies as fairly well
entrenched and accepted, if grudgingly [3], others [5] described
more signicant turf battles between midwives and obstetricians,
obstetric nurses, and family doctors.
Several studies [9, 12, 33] described conicts between midwives and other skilled birth attendants and TBAs, with midwives
asserting their authority through their connection to the medical
system, and at times even scapegoating TBAs for poor maternal
outcomes.
Relationships between midwives and nurses could also be
difcult. One study [17] reported that in the teams of nurses and
midwives they described, nurses often took on the role of policy
enforcers. Although this gave them a certain measure of power, it
also put them in the position of the bad guy vis--vis midwives
and the mothers in their care. Nurses reported feeling marginalised, as neither doctor nor midwife, and uncertain in their role.
Communication and co-ordination
The routine challenges of communication and co-ordination
also proved difcult in several studies where midwives worked in
teams. Even in successful initiatives, like one to train midwives in

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

advanced resuscitation of newborns [15], communication was a

persistent challenge. Similarly, nurse managers in another study of
maternity support workers [16] identied communication and
teamwork as two of their top training needs.
Team midwifery units appeared, ironically, to also suffer from
problems of communication and co-ordination [4, 7, 36]. Midwives
in one study [7] who were not used to working as a team were
slow to build trust, especially when individual versus group
liability was unclear, leading to pseudo-cohesion that undermined group communication and co-ordination. Similar challenges were reported in studies on the management of critical
illness/high-dependency care [4, 11]. These units brought team
members with widely differing skills and skill levels together
from junior midwives to specialist cliniciansand despite the high
stakes, communication remained weak.
In general, regardless of the context, communication and coordination seemed to be the easiest between midwives (though by
no means uncomplicated), slightly less free and effective between
midwives and nurses, and relatively weak between midwives and
doctors.
Trust, responsibility and accountability
Issues of trust, responsibility and accountability were another
factor shaping the success of task shifting. Doctors expressed
anxiety around responsibility and accountability when midwives
took over or shared roles with them [3, 5, 25]. Their anxiety often
stemmed from their lack of knowledge of midwifery training and
practice, their perception of variability in midwife skill and
experience, and unclear regulations governing liability in these
contexts [4, 5].
Midwives had similar concerns [22]. Team midwifery models of
care proved complicated with midwives in one study trying to
maintain individual responsibility for their practice in a group
context [36]. Midwives in another study also felt vulnerable when
working with doulas given the inuence they can have over
birthing plans [34].
Midwives, however, were generally described as having a
different conception of responsibility from other health professionals. One study of midwife integration in Canada [5], for
example, argued that, These midwives had a history of marginal
practice and a culture favouring alternative and community care
[that led them] to develop particular conceptions of risk, professional responsibility and the clientprovider relationship that were
obstacles to co-operation with other health-care providers and to
the integration of midwives [p. 18]. Midwives, unlike doctors,
tended to not speak of their practice in terms of liability but rather
trust and condence [30].

Discussion
One of the challenges of this review was the difculty of
dening task shifting in the context of midwifery services. Task
shifting interventions were often not labelled as such (in contrast
to more frequent use of the term among lay health worker
programmes) and we could not identify other search terms that
would reliably serve the same purpose. The broadness of the
review question also limited the degree to which multiple studies
could be identied to contribute ndings on specic issues that
emerged in the data analysis. In several cases, there are only one or
two studies that addressed a specic barrier or facilitator.
Many of the challenges of task shifting in the context of
midwifery services involved the unique role, status and identity
of midwives in the medical setting and the impact of this on
efforts to reorganise services. At stake were both ideological

1219

differences about the most appropriate form of care during

pregnancy and childbirth as well as different professional roles,
responsibilities and organisation of practices. These tensions
emerged across most of the studies despite wide variation in the
actual practice of midwives in the studies.
Overall, midwives reported that a range of task shifting
scenarios challenged the midwifery model of care. Sometimes
task shifting disrupted one-on-one relationships and continuous
care. At other times, task shifting challenged midwives' natural
and non-interventionist approaches to childbirth. Even taskshifting initiatives that were consistent with midwifery's model
of holistic and continuous care could prove difcult in practice
since these new skills also meant greater workloads and potentially fragmented care.
Where new tasks and skills were shifted to midwives from
doctors or nurses, the studies reported few problems in acquiring
new knowledge and skills. Of greater concern was frequent poor
planning, poor follow-up support and training, and poor integration of new tasks into the broader organisation of care. Task
shifting in these studies was often not well planned or consistently
implemented and indeed ran the full spectrum from improvisational, ad hoc forms of task shifting to highly formalised
programmes.
Differences in social status and power among lay and medical
staff could also prove to be a barrier to task shifting as could
uncertainty around these new roles, responsibilities and liabilities.
Communication and co-ordination among staff members was
another common challenge reported. Some of these challenges
reected general weaknesses in the health system but others
resulted from tensions between midwifery and biomedical models
of care.
Many of the studies addressed task-shifting initiatives to upskill midwives. These forms of task shifting brought midwives a
number of advantages but sometimes came at a price. Poor clinical
support and supervision, inadequate training, and haphazard
implementation could undermine their condence in these new
skills. Fear of liability and an unclear regulatory environment also
threatened task shifting. Finally, midwives sometimes struggled to
balance a belief in the normalcy of pregnancy and the importance
of minimal intervention with the idea that pregnancy was simultaneously a risky time that required increased vigilance and
concern.
These ndings contribute both to a better understanding of
task shifting and skill mix in midwifery services as well as to the
broader literature on task shifting in primary health care (WHO
et al., 2007; Callaghan et al., 2010; Fulton et al., 2011). As a
systematic review of global qualitative evidence, it integrates
experience from around the world on task shifting and identies
important commonalities and differences in the implementation
of task shifting for midwifery in different settings.
Some of the review ndings will not be surprising to those
practising and conducting research in the eld of midwifery. The
tensions between the biomedical and midwifery models of care,
the contrasting notions of risk and responsibility among health
staff, and the complicated relationships between midwives,
nurses, and doctors, for example, are common themes in the
broader literature. As syntheses of the existing literature, rigorous
systematic reviews indeed often produce ndings that reect but
also bring condence to some of the current consensus in a eld.
We also made what we believe are more novel ndings, such as
the nding that shifting new tasks to midwives that were
consonant with the ideological model of midwifery care (e.g.
neonatal examinations) could nonetheless put pressure on this
model in practice as these added tasks increased workloads and
disrupted the one-to-one relationship between mother and
midwife.

1220

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

However, this systematic review also makes a number of other

contributions to understanding task shifting in midwifery. Firstly,
it provides an assessment of the certainty of each key review
nding and in doing so, points to important issues that may make
intuitive sense but for which there may be less abundant or lower
quality evidence. The ndings, for example, around the role of
clinical experience in taking on more complex technical tasks and
the important differences between narrow technical skill and
deeper clinical judgement come from only a few studies in limited
contexts (see Supplementary le B, Findings 1719), suggesting the
need for further research.
By reviewing evidence from a range of contexts and settings
globally, this review also identies important contextual variations
in task shifting. We reported on, for example, some important
differences between high-income and low-/middle-income settings when it comes to the drivers and meanings of task shifting
and the role of doulas, TBAs and other birth supporters.
By using the framework of task shifting, this review also
brings the existing literature on the organisation of tasks and
services in midwifery under a new conceptual umbrella, allowing
connections to be made across different areas of midwifery
research and between research on midwifery and other cadres of
primary health care. Our initial search for studies, for example,
revealed a near total absence of the term task shifting in the
midwifery literature. The term skills mix produced more relevant
studies but this concept is broader than that of task shifting. By
bringing the task shifting lens to research in midwifery, this
review provides a platform for connecting the long-standing but
under-explored experiences of task shifting in midwifery, to
research and policy conversations happening in relation to other
cadres. The differences and commonalities between experiences of
task shifting in midwifery and in other cadres can, for example, be
readily seen by comparing the key ndings reported in the
Optimize4MNH guidelines.
Another strength and contribution of this review is its use of a
theory-informed conceptual framework for the implementation of
health system interventions. The SURE framework was synthesised
from existing conceptual frameworks and empirical evidence on
the key barriers and facilitators to implementing health systems
interventions. It provides an approach for organising and assessing
evidence about health systems interventions in a way that ensures a
more rigorous, thorough, and consistent review of the available
evidence. The subsequent analysis is both richer and better able to
be compared and contrasted with other evidence on and theories of
implementation. This review is the rst review of task shifting in
midwifery services that we know of that uses this kind of
theoretically-informed health systems approach.
Use of the SURE framework in the WHO guideline development
process also allowed for the ndings from this review to be linked
meaningfully to the broader debate around task shifting and
maternal and newborn health among all cadres of health workers
(Sandall, 2012). This not only strengthened the quality of the
review itself but also helped to ensure its ndings had a wider
relevance and impact. The nal recommendations, along with a
summary video and annexes describing the quantitative and
qualitative evidence reviewed, can be found at http://www.opti
mizemnh.org.

Conclusion
A number of conclusions follow from the ndings summarised
above. Health-care workers should be adequately informed of both
specic changes in practice entailed by task shifting as well as the
general scope of practice and training for midwives. Although it is
clear that doulas, TBAs and other birth supporters can be valuable

sources of emotional support and cross-cultural brokering, the

relationships between midwives and these supporters can be
contentious and clearer denitions of roles, responsibilities and
liabilities is important.
Task-shifting initiatives involving midwives should also take
into account the complex relationships and social hierarchies
between midwives, obstetric nurses, doctors, and others when
redesigning roles and responsibilities, anticipate conicts and
ambiguities, and develop ways of managing these dynamics
inter-professionally. Similarly, effective task shifting generally
requires ongoing supervision and support as well as careful
integration into clinical protocols and the broader delivery of care.
This is a medium-term process that can require signicant
management planning and support.
Finally, vague legal and regulatory environments can be important barriers to successful task shifting. As far as possible, the legal
protections and liabilities and the regulatory framework for task
shifting should be designed to accommodate these new practices
and should be communicated to all those involved.
Many of the conclusions described above that follow from the
review ndings are not specic to midwifery and its particular
history and characteristics. Rather they have to do with need to
strengthen health systems and improve human resource development and management capacities overall. That said, if policymakers and practitioners can nd ways to support the effective
use of task shifting to and from midwives in the provision of
critical maternal and newborn health interventions, these
improvements should be felt concretely, in the improved health
and survival of mothers and children around the world.

Conicts of interest
None declared.

Acknowledgements
We would like to express our appreciation to A. Metin Glmezoglu,
Dr. Joao Paulo Dias de Souza and the other members of the
OptimiseMNH guideline technical committee for their support of this
project. Funding for this project was received from the UNDP/UNFPA/
UNICEF/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction, Department of
Reproductive Health and Research, World Health Organization, and
the Alliance for Health Policy and Systems Research.

Appendix A. Supporting information

Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.midw.2013.05.001.

References
Akhavan, S., Lundgren, I., 2010. Midwives' experiences of doula support for
immigrant women in Swedena qualitative study. Midwifery 28, 8085.
Battersby, S., 2010. An evaluation of midwives' knowledge of formula feeding and
their role in supporting mothers who formula feed their infants. Journal of
Family Health Care 20, 192197.
Battersby, S., Thomson, A.M., 1997. Community midwives' and general practitioners' perspectives, of antenatal care in the community. Midwifery 13, 9299.
Bench, S., Fitzpatrick, J., 2007. Recognition and management of critical illness by
midwives: implications for service provision. Journal of Nursing Management
15, 348356.
Bhutta, Z.A., Lassi, Z.S., Mansoor, N., 2011. Systematic Review on Human Resources
for Health Interventions to Improve Maternal Health Outcomes: Evidence from
Developing Countries. Aga Khan University, Karachi.

C.J. Colvin et al. / Midwifery 29 (2013) 12111221

Callaghan, M., Ford, N., Schneider, H., 2010. A systematic review of task-shifting for
HIV treatment and care in Africa. Human Resources for Health 8.
Collin, J., Blais, R., White, D., Demers, A., Desbiens, F., 2000. Integration of midwives
into the Quebec health care system. LEquipe dEvaluation des Projets-Pilotes
Sages-Femmes. Canadian Journal of Public Health 91, I16I20.
DAmbruoso, L., Achadi, E., Adisasmita, A., et al., 2009. Assessing quality of care
provided by Indonesian village midwives with a condential enquiry. Midwifery 25, 528539.
Deery, R., 2005. An action-research study exploring midwives support needs and
the affect of group clinical supervision. Midwifery 21, 161176.
Dickson-Tetteh, K., Billings, D.L., 2002. Abortion care services provided by registered midwives in South Africa. International Family Planning Perspectives 28,
144150.
Dietsch, E., 2010. The experience of being a traditional midwife: relationships with
skilled birth attendants. Rural Remote Health 10, 1481.
Doring, C., Nolte, A., 1999. Problems midwives experience with legislation and
their needs in this regard. Health SA Gesondheid 4, 7177.
Dovlo, D., 2004. Using mid-level cadres as substitutes for internationally mobile
health professionals in Africa. A desk review. Human Resources for Health 2.
Edwards, L., 2008. How prepared are midwives to provide high-dependency care?
A study exploring midwives experiences when caring for highly dependent
women on delivery suite. Midwifery Digest 18, 224230.
Foster, J., Anderson, A., Houston, J., Doe-Simkins, M., 2004. A report of a midwifery
model for training traditional midwives in Guatemala. Midwifery 20, 217225.
Foster, J., Regueira, Y., Burgos, R.I., Sanchez, A.H., 2005. Midwifery curriculum for
auxiliary maternity nurses: a case study in the Dominican Republic. Journal of
Midwifery & Womens Health 50, e45e49.
Foster, J., Regueira, Y., Heath, A., 2006. Decision making by auxiliary nurses to assess
postpartum bleeding in a Dominican Republic maternity ward. Journal of
Obstetric, Gynecologic, & Neonatal Nursing 35, 728734.
Fretheim, A., Munabi-Babigumira, S., Oxman, A., Lavis, J., Lewin, S., 2009. SUPPORT
Tools for evidence-informed health policymaking (STP). 6. Using research
evidence to address how an option will be implemented. Health Research
Policy and Systems 7, S6.
Fulton, B.D., Schefer, R.M., Sparkes, S.P., et al., 2011. Health workforce skill mix and
task shifting in low income countries: a review of recent evidence. Human
Resources for Health 9.
Fulton, T., 2002. Midwife advanced resuscitators of the newborn: a successful
practice change. Midwifery Digest. 12, 537539.
Georgeu, D., Colvin, C., Lewin, S., et al., 2012. Implementing Nurse Initiated and
Managed Antiretroviral Treatment (NIM-ART) in South Africa: a qualitative
process evaluation of the STRETCH Trial. Implementation Science 7.
Glenton, C., Lewin, S., Scheel, I., 2011. Still too little qualitative research to shed light
on results from reviews of effectiveness trials: a case study of a Cochrane
review on the use of lay health workers. Implementation Science 27, 53.
Glenton, C., Colvin, C.J., Carlsen, B., et al., 2013. Barriers and facilitators to the
implementation of lay health worker programmes to improve access to
maternal and child health: qualitative evidence synthesis (Protocol). Cochrane
Database of Systematic Reviews.
Grifn, R., Dunkley-Bent, J., Malhotra, G., 2009. Building capacity to care: learning
for maternity support workers. British Journal of Midwifery 17, 711.
Grol, R., Bosch, M., Hulscher, M., Eccles, M., Wensing, M., 2007. Planning and
studying improvement in patient care: the use of theoretical perspectives.
Milbank Quarterly 85, 93138.
Hanson, K., Ranson, M., Oliveira-Cruz, V., Mills, A., 2003. Expanding access to
priority health interventions: a framework for understanding the constraints to
scaling-up. Journal of International Development 15, 114.
International Confederation of Midwives, 2011. ICM International Denition of the
Midwife. International Confederation of Midwives, The Hague.
Kennedy, H.P., Lyndon, A., 2008. Tensions and teamwork in nursing and midwifery
relationships. Journal of Obstetric, Gynecologic, & Neonatal Nursing 37,
426435.
Kirkham, M., 2003. Birth Centres. A Social Model for Maternity Care. Book for
Midwives. Elsevier Science Limited, London.
Lewin, S., Munabi-Babigumira, S., Glenton, C., et al., 2010. Lay health workers in
primary and community health care for maternal and child health and the
management of infectious diseases. Cochrane Database of Systematic Reviews
CD004015.
Lindberg, I., Christensson, K., Ohrling, K., 2005. Midwives experience of organisational and professional change. Midwifery 21, 355364.
Lipp, A., 2008. A woman centred service in termination of pregnancy: a grounded
theory study. Contemporary Nurse 31, 919.
Lumsden, H., 2005. Midwives experience of examination of the newborn as an
additional aspect of their role: a qualitative study. Midwifery Digest 15,
450457.

1221

Lundgren, I., 2010. Swedish womens experiences of doula support during childbirth. Midwifery 26, 173180.
Maupin, J.N., 2008. Remaking the Guatemalan midwife: health care reform and
midwifery training programs in Highland Guatemala. Medical Anthropology
27, 353382.
McKenna, H., Hasson, F., Smith, M., 2003. Training needs of midwifery assistants.
Journal of Advanced Nursing 44, 308317.
McNeill, J., Alderdice, F., 2009. Exploring the perspective of midwives involved in
offering serum screening for Downs syndrome in Northern Ireland. Journal of
Clinical Nursing 18, 28882896.
Mitchell, M., 2003a. Midwives conducting the neonatal examination: part I. British
Journal of Midwifery 11, 1621.
Mitchell, M., 2003b. Midwives conducting the neonatal examination: part II. British
Journal of Midwifery 11, 8084.
Noyes, J., 2009. Qualitative research and Cochrane reviews. In: Higgins, J., Green, S.
(Eds.), Cochrane Handbook for Systematic Reviews of Interventions Version
5.0.2 [updated September 2009]. The Cochrane Collaboration.
Oxman, A., Flottorp, S., 2001. An overview of strategies to promote implementation
of evidence based health care. In: Silagy, C., Haines, A. (Eds.), Evidence Based
Practice, 2nd edn. BMJ Publishers, London.
Persson, M., Hornsten, A., Winkvist, A., Mogren, I., 2011. Mission impossible?
Midwives experiences counseling pregnant women with gestational diabetes
mellitus. Patient Education and Counseling 84, 7883.
Pettersson, K.O., Svensson, M.L., Christensson, K., 2001. The lived experiences of
autonomous Angolan midwives working in midwifery-led, maternity units.
Midwifery 17, 102114.
Popay, J., Roberts, H., Sowden, A., 2006. Guidance on the Conduct of Narrative
Synthesis in Systematic Reviews. A Product from the ESRC Methods Programme. Institute of Health Research, Lancaster.
Prowse, J., Prowse, P., 2008. Role redesign in the National Health Service: the effects
on midwives work and professional boundaries. Work, Employment & Society
22, 695712.
Pyone, T., Sorensen, B.L., Tellier, S., 2012. Childbirth attendance strategies and their
impact on maternal mortality and morbidity in low-income settings: a
systematic review. ACTA Obstetricia et. Gynecologica 91, 10291037.
Reiger, K.M., Lane, K.L., 2009. Working together: collaboration between midwives
and doctors in public hospitals. Australian Health Review 33, 315324.
Rogers, C., Bloomeld, L., Townsend, J., 2003. A qualitative study exploring midwives perceptions and views of extending their role to the examination of the
newborn baby. Midwifery 19, 5562.
SURE Collaboration, 2011. Identifying and addressing barriers to implementing
policy options. In: Collaboration, S. (Ed.), SURE Guides for Preparing and Using
Evidence-Based Policy Briefs. World Health Organization, Geneva, Version 2.1.
Sandall, J., 2012. Every woman needs a midwife, and some women need a doctor
too. Birth 39, 323326.
Shaban, I., Barclay, L., Lock, L., Homer, C., 2012. Barriers to developing midwifery as
a primary health-care strategy: a Jordanian study. Midwifery 28, 106111.
Smid, M., Campero, L., Cragin, L., Gonzalez Hernandez, D., Walker, D., 2010. Bringing
two worlds together: exploring the integration of traditional midwives as
doulas in Mexican public hospitals. Health Care Women International 31,
475498.
Stevens, J., Dahlen, H., Peters, K., Jackson, D., 2011. Midwives and doulas
perspectives of the role of the doula in Australia: a qualitative study. Midwifery
27, 509516.
Tsouroui, M., 2011. Routinisation and constraints on informed choice in a one-stop
clinic offering rst trimester chromosomal antenatal screening for Downs
syndrome. Midwifery 27, 431436.
UNFPA, 2011. The State of the Worlds Midwifery: Delivering Health, Saving Lives.
UNFPA, Geneva.
WHO, 2010. Increasing Access to Health Workers in Remote and Rural Areas
Through Improved Retention. Global Policy Recommendations. WHO, Geneva.
WHO, 2012. Optimizing Health Worker Roles to Improve Access to Key Maternal
and Newborn Health Interventions Through Task Shifting. World Health
Organization, Geneva http://www.optimizemnh.org. (last accessed 24 January
2013)..
WHO, PEPFAR, UNAIDS, 2007. Task Shifting: Rational redistribution of Tasks among
Health Workforce Teams. Global Recommendations and Guidelines. WHO,
Geneva.
Walker, S.B., Moore, H.D., Eaton, A., 2004. North Queensland midwives experience
with a team model of midwifery care. Australian Journal of Midwifery 17,
1722.
Widmark, C., Tishelman, C., Lundgren, E.L., et al., 1998. Opportunities and burdens
for nurse-midwives working in primary health care. An example from
population-based cervical cancer screening in urban Sweden. Journal of Nurse
Midwifery 43, 530540.