patient safety in maternity care: towards better outcomes · 2017. 4. 4. · patient safety in...

Patient Safety in Maternity Care:

Towards Better Outcomes

Suzanne Veronica Sinni RN, RM, BN, Grad Cert Neonatal Intensive Care, MBL

A thesis submitted for the degree of Doctor of Philosophy

at Monash University in 2016

Department of Obstetrics and Gynaecology

School of Clinical Sciences

Table of Contents

Abstract ............................................................................................................................................ ii Peer-reviewed publications arising from research ........................................................................... v Oral presentations at learned scientific meetings during candidature ............................................. vi Poster presentations at learned scientific meetings during candidature ......................................... vii Institutional oral presentations during candidature ......................................................................... ix General Declaration .......................................................................................................................... x Acknowledgements ........................................................................................................................ xv CHAPTER ONE ........................................................................................................................... 17 General overview .......................................................................................................................... 17 1.1 Background .......................................................................................................................... 18 1.2 Literature review and Casey Hospital .................................................................................. 19

1.2.1 Evaluation methodologies ......................................................................................... 19 1.2.2 Application to the research design ............................................................................. 22 1.2.3 Patient safety .............................................................................................................. 22 1.2.4 Casey Hospital maternity service .............................................................................. 25 1.2.5 References .................................................................................................................. 32

1.3 Hypotheses ........................................................................................................................... 36 1.4 Aims ..................................................................................................................................... 36 1.5 Thesis structure .................................................................................................................... 37

CHAPTER TWO .......................................................................................................................... 40 Patient safety: a literature review to inform an evaluation of a maternity service ...................... 40 2.1 Preamble ............................................................................................................................... 41 2.2 Patient safety: a literature review to inform an evaluation of a maternity service ............... 42

CHAPTER THREE ..................................................................................................................... 47 Designing a clinical audit tool to measure processes of pregnancy care .................................... 47 3.1 Preamble ............................................................................................................................... 48 3.2 Designing a clinical audit tool to measure processes of pregnancy care ............................. 49

CHAPTER FOUR ........................................................................................................................ 60 Measuring pregnancy care: towards better maternal and child health ....................................... 60 4.1 Preamble ............................................................................................................................... 61 4.2 Measuring pregnancy care: towards better maternal and child health ................................. 62

CHAPTER FIVE .......................................................................................................................... 68 Perinatal staff perceptions of safety and quality in their service ................................................. 68 5.1 Preamble ............................................................................................................................... 69 5.2 Perinatal staff perceptions of safety and quality in their service ......................................... 70

CHAPTER SIX ............................................................................................................................. 79 Discussion and closing marks ...................................................................................................... 79 6.1 Summary of the findings ...................................................................................................... 80 6.2 Limitations ........................................................................................................................... 81 6.3 Directions for future research .............................................................................................. 82 6.4 Conclusions .......................................................................................................................... 83

Page | ii

Abstract

“A healthy start to life” is a health research priority of the National Health and Medical

Research Council (NHMRC). Such a start to life is, in no small way, dependant upon both

a healthy pregnant woman and high quality care during her pregnancy, labour and

postpartum. In Australia, 99 of every 100 babies are born in hospital. While their mothers

may receive care during their pregnancy in an array of settings, acute public health services

are the major provider of care.

When Monash Health (formerly Southern Health), Victoria’s largest provider of maternity

care, opened a new maternity service, encompassing outpatient antenatal, inpatient

intrapartum and immediate postpartum, and domiciliary postpartum pregnancy care, in a

new green-field site hospital a formal evaluation of the service was planned to provide both

detailed information about the quality of the new service and insights into opportunities for

care enhancement at its other two existing sites. The work detailed in this thesis is the

result of that evaluation and of a broader view at the provision of safe maternity care. The

underlying premise of the evaluation was that the highest priority for the service was

safety. As such, I endeavoured to measure the safety of the service. The approach that I

took to measure safety was influenced by the international momentum to improve patient

safety and by the growing and increasingly informed debate about how to measure patient

safety. The resultant mixed methods approach is intended to provide rich and detailed data

about the pregnancy (antenatal) care provided by the service. Pregnancy care was

considered a useful measure of the service on the assumption that quality pregnancy care

reduces the risk of unexpected poor outcomes. Whilst an evaluation of intrapartum care, in

addition to pregnancy care, would provide a more comprehensive evaluation of the service

such an evaluation was beyond the scope of this PhD. At the outset, I had hoped that the

Page | iii

data would usefully inform the service about opportunities for future patient safety

initiatives in its maternity services and I am pleased to have been told that that has been so.

Since I commenced my doctoral research Victorian public maternity services have been

rocked by findings of a review of Djerriwarrh Health Service’s maternity service. While

the detailed findings of that review have not been made public the summary findings

echoed the need for all services to have a mature and embedded culture of safety. In that

regard, I hope that my findings may also offer useful insights to others responsible for

maternity services more broadly.

The research reported in the thesis is divided into quantitative and qualitative components.

The quantitative research involved the development and validation of an audit tool

(Chapter Three) that was then applied to measure actual health care delivery against those

standards (Chapter Four) as recorded in the health records of women attending the new

maternity service for their pregnancy care. At the research site, episodes of care were

documented in hard copy and retained in a health record. A copy of the health record is

scanned and stored electronically. While this facilitated access to records the inherent

challenges of legibility and missing information in an unregulated hardcopy medical

records system remained. In addition, the complex reporting required of clinicians likely

confounded data completion and veracity as clinicians, at least apparently, struggled to

ensure completion of duplicate information on various unconnected platforms. Of course,

this observation in itself was inherent in an overall assessment of safety and I refer to the

matter in the relevant Discussion in Chapter Four, providing recommendations for future

service design and development. Despite these challenges, my review of pregnancy care,

measured against a pre-defined template of standards of care, suggested that there was a

Page | iv

high level of compliance with organisational expectations reflective of a high quality

service (Chapter Four).

Current concepts of patient safety internationally also influenced my use of qualitative

research. Hospitals in the United State of America (USA) regularly conduct surveys of

hospital staff as a component of assessing the hospital’s culture of patient safety. This

survey was used to develop prompt questions for interviews with staff involved with

maternity care at the research site about their perceptions of the safety and quality of the

service in which they work. The semi-structured interviews yielded qualitative data for

thematic analysis, described in Chapter Five, and indicated that staff perceived their

service to be safe and of high quality. Nonetheless, staff, regardless of discipline or

seniority referred to numerous systems at various organizational levels to optimise patient

safety. Staff emphasised the influence of relationships and the importance of respect on

team performance. These and other service improvement opportunities identified by the

mixed method approach that I used are explored in depth across Chapters Four, Five, and

Six.

The results reported herein reassure that the service is safe, with lower error rates than

those reported in the literature for acute medical and surgical settings. However, the

usefulness of such comparisons is debatable, prompting the need for ongoing research

measuring patient safety, particularly in pregnancy care. The qualitative results detailed in

Chapter Five reaffirm the high standards of care, arguably attributable to the high levels of

trust and mutual respect among senior staff.

Page | v

Peer-reviewed publications arising from research

1. Sinni SV, Cross WM, Wallace EM. Designing a clinical audit tool to measure processes

of pregnancy care. Nursing: Research and Reviews. 2011; 1: 15-25,

(http://dx.doi.org/10.2147/NRR.S26543).

2. Sinni SV, Wallace EM, Cross WM. Patient safety: A literature review to inform an

evaluation of a maternity service. Midwifery. 2011; 27 (6): e274-8.

(doi:10.1016/j.midw.2010.11.001).

3. Sinni SV, Wallace EM, Cross WM. Perinatal staff perceptions of safety and quality in

their service. BMC Health Services Research. 2014; 14: 591 (doi:10.1186/s12913-014-

0591-4).

4. Sinni SV, Cross WM, Swanson AE, Wallace EM. Measuring Pregnancy Care: Towards

Better Maternal and Child Health. Australian and New Zealand Journal of Obstetrics

and Gynaecology. 2016; 56 (2): 142-7.

Page | vi

Oral presentations at learned scientific meetings during candidature 1. Sinni SV, Wallace EM, Cross W.

Measuring Patient Safety in Maternity Services.

Breathing New Life. Australian College of Midwives, Royal Australian and New

Zealand College of Obstetrics and Gynaecology Collaborative Conference. Canberra,

2008.

2. Sinni SV, Wallace E, Cross W. Patient Safety in Maternity Services. Perinatal Society

of Australia and New Zealand 10th Annual Congress. Gold Coast, 2008.

3. Sinni SV, Wallace EM, Cross W.

Patient Safety in Maternity Services.

Victorian Department of Health Stakeholders Forum. Melbourne, 2008.


Pregnancy care clinical audit: audit design, application and results.

Breathing New Life. Australian College of Midwives, Royal Australian and New

Zealand College of Obstetrics and Gynaecology, Australian College of Rural and

Remote Medicine, and Royal Australian College of General Practitioners

Collaborative Conference. Melbourne, 2012

5. Sinni SV, Cross W, Wallace EM.

What do providers of maternity care say about the safety of their service?

Breathing New Life. Collaborative Conference: Australian College of Midwives,

Royal Australian and New Zealand College of Obstetrics and Gynaecology, Australian

College of Rural and Remote Medicine, Royal Australian College of General

Practitioners. Melbourne, 2012


Maternity care. Findings from a clinical audit on pregnancy care and perceptions of

staff who provide that care.

Perinatal Society of Australia and New Zealand 14th Annual Congress. Sydney, 2012

Page | vii

Poster presentations at learned scientific meetings during candidature


Measuring patient safety – a mixed method evaluation of a maternity service.

Australian Association of Quality in Health Care Annual Scientific Meeting.

Melbourne, 2011.

2. Sinni SV, Wallace EM, Papacostas K, Yasmine M, Edwards A, Hand N, Stewart L.

Improving care processes for pregnant women requiring urgent assessment.

Australian Association of Quality in Health Care Annual Scientific Meeting.

Melbourne, 2011.


A mixed method evaluation of a maternity service.

Australian College of Midwives Annual Conference. Sydney 2011.


A mixed method evaluation of a maternity service – quantitative findings. Perinatal

Society of Australia and New Zealand 13th Annual Congress. Hobart, 2011.

5. Tippett C, Sinni SV, Dunn J, Malkoutzis V, Klyne R, Cliffe K, Hamilton B, Taylor K,

Alexander-Smith U.

Maternity care of women with complex medical needs.

Perinatal Society of Australia and New Zealand 14th Annual Congress. Sydney, 2012.

6. Sinni SV, Alston N, Shekleton P, Little J, Couper D, Deller K, O’Mara K, Hodges R.

Perinatal bereavement: a journey through women's experiences from diagnosis to

postnatal community support.


7. Sinni SV, Cox J, O’Connell A, Vermeulen M, Whitecross S, Papacostas K, Stewart L.

Transition to from midwifery student to graduate – perceptions from a tertiary setting.


Page | viii

8. Sinni SV, Beaves M, Cox J, McCulloch B, O’Connell A.

Fetal surveillance training and support program for graduate midwives – a Southern

Health experience.


9. Sinni SV, Thorne B, Flett F, Lievesley S, Stewart L, Papacostas K.

British midwives’ perspectives on working in an Australian tertiary perinatal centre.


Page | ix

Institutional oral presentations during candidature



Southern Health Inaugural Research Week. Monash Medical Centre, 2007.


Measuring the processes of care in a maternity service using quantitative and

qualitative methods.

Monash Institute of Medical Research Student Symposium, 2010.


Evaluation of a Maternity Service, PhD project update.

Casey Hospital Maternity Practice Improvement Committee, 2010.



Southern Health Internal In-service to midwives. Casey Hospital, 2011.

5. Sinni SV, Wallace E, Cross W.


Southern Health Internal In-service to midwives. Monash Medical Centre, 2011.

Page | x

General Declaration Declaration for thesis based or partially based on conjointly published or unpublished work In accordance with Monash University Doctorate Regulation 17/ Doctor of Philosophy and Master of Philosophy (MPhil) regulations the following declarations are made: I hereby declare that this thesis contains no material which has been accepted for the award of any other degree or diploma at any university or equivalent institution and that, to the best of my knowledge and belief, this thesis contains no material previously published or written by another person, except where due reference is made in the text of the thesis. This thesis by publication includes four original papers published in peer reviewed journals. The core theme of the thesis is measuring patient safety in a maternity service using quantitative and qualitative (mixed) methods. The ideas, development and writing up of all the papers in the thesis were the principal responsibility of myself, the candidate, working within the Department of Obstetrics and Gynaecology, The Ritchie Centre, Monash Institute of Medical Research, under the supervision of Professor Euan Wallace and Professor Wendy Cross. In the case of chapters 2 – 5, my contribution to the work involved the following:

Thesis chapter

Publication title Publication status*

Nature and extent of candidate’s contribution

2 Patient safety: A literature review to inform an evaluation of a maternity service.

Published 85%

3 Designing a clinical audit tool to measure processes of pregnancy care.

Published 85%

4 Measuring pregnancy care – towards better maternal and child health.

Published 85%

5 Perinatal staff perceptions of safety and quality in their service.

Published 85%

I have renumbered sections of submitted or published papers in order to generate a consistent presentation within the thesis. Signed: Date: 14/08/2015

Page | xi

Declaration for Thesis Chapter 2 Declaration by candidate In the case of Chapter 2, the nature and extent of my contribution to the work was the following:

Nature of contribution Extent of� contribution (%) Conception and study design, ethics approval, ongoing reporting requirements, literature review, manuscript write up.

85%

The following co-authors contributed to the work. Co-authors who are students at Monash University must also indicate the extent of their contribution in percentage terms:

Name Nature of contribution Extent of contribution (%) for student co-authors only

Euan M Wallace Contributed to project design, development of ideas and critical revision of the manuscript

Wendy M Cross Contributed to project design, development of ideas and critical revision of the manuscript

Candidate’s�Signature Date

4/10/2012

Declaration by co-authors The undersigned hereby certify that:

(1) the above declaration correctly reflects the nature and extent of the candidate’s contribution to this work, and the nature of the contribution of each of the co-authors.

(2) they meet the criteria for authorship in that they have participated in the conception, execution, or interpretation, of at least that part of the publication in their field of expertise;

(3) they take public responsibility for their part of the publication, except for the responsible author who accepts overall responsibility for the publication;

(4) there are no other authors of the publication according to these criteria; (5) potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or

publisher of journals or other publications, and (c) the head of the responsible academic unit; and

(6) the original data are stored at the following location(s) and will be held for at least five years from the date indicated below:

Location(s) Monash Institute of Medical Research

Euan M Wallace

Date 2/10/2012

Wendy M Cross

Page | xii

!


Nature of�contribution Extent of�contribution (%)

Conception and study design, ethics approval, ongoing reporting requirements, literature review, manuscript write up.

85%





Candidate’s�Signature Date

4/10/12 Declaration by co-authors The undersigned hereby certify that:







Location(s) Monash Institute of Medical research

Date Euan M Wallace 2/10/12

Wendy M Cross 2/10/12

Page | xiii

!


Nature of contribution Extent of contribution (%)


85%



Euan Wallace Contributed to project design, development of ideas and critical revision of the manuscript

Wendy Cross Contributed to project design, development of ideas and critical revision of the manuscript

Amy Swanson Data collection, manuscript revision

Ann Digby Data collection, manuscript revision

Candidate’s Signature

Date 21/8/15









Date Euan M Wallace 21/8/15 Wendy M Cross 21/8/15 Amy Swanson 6/9/12 Ann Digby 25/9/12

Page | xiv

-


Nature of contribution Extent of contribution (%)


85%





Candidate’s Signature

Date 4/10/12









Date Wendy M Cross 2/10/12

Euan M Wallace 2/10/12

Page | xv

Acknowledgements

Professor Euan Wallace was instrumental in this work. In 2004, we explored the possibility

of a PhD program to evaluate a new maternity service at Monash Health (then known as

Southern Health). Professor Wendy Cross joined us, providing nursing leadership and

qualitative research expertise. I have spent countless hours with Wendy and Euan for

whom I have enormous respect. I am privileged to have worked under their supervision.

I gratefully acknowledge Professor Don Campbell and staff at the Monash Institute of

Medical Research who provided valuable insights about the principles of patient safety and

how we might measure it. I acknowledge Dr Bill Shearer and Ms Patricia McGarrity who

taught me so much about patient safety.

I thank Ms Lisieux Jones for accommodating my endless requests including arranging

meetings, conference registrations, accommodation and travel.

I gratefully acknowledge my various employers who supported my ongoing candidature. I

especially thank my colleagues in Birth Suite at Monash Medical Centre who have

supported me with interest and enthusiasm, and have been very accommodating of

commitments associated with my candidature.

I am profoundly grateful to my colleagues at the Casey Maternity Service who were so

generous with their time, and who inspired me to return to direct clinical care.

Special thanks to Ms Kerrie Papacostas, who has unparalleled knowledge and passion for

the art and science of midwifery, and who has answered endless requests to provide

procedural documents and information about the various models of care at Monash Health.

I thank my dear friend Dr Louise Corben PhD who listened patiently and empathetically,

Page | xvi

providing helpful advice about life as a PhD candidate. Heartfelt thanks go to my extended

family and friends for their interest and support.

Most importantly I thank my children, Paul, Madeleine and Emily for their patience and

unwavering support, love and encouragement. They are my greatest source of pride and joy

and I dedicate this work to them.

Page | 17

CHAPTER ONE

General overview

Page | 18

1.1 Background

In 2002 Monash Health (then Southern Health) and the Victorian Department of Health

(the Department) began discussions about building a new hospital (Casey Hospital) in

Berwick that would include maternity services. At the time, the Department was preparing

the “Future Directions for Victoria’s Maternity Services” 2004 policy document that

included Government support to establish midwifery led models of care. While a

standalone midwifery birth centre was initially planned for Casey Hospital, a midwifery-

led collaborative maternity service model was finally agreed for the Casey service,

including on-call senior obstetric staff and on-site hospital medical officers. The

Department recommended that the new maternity service be evaluated. Two of the

researchers, the PhD candidate (a midwife) and her supervisor (an obstetrician), were

involved in these discussions in their professional roles in administration and governance

of maternity services. An evaluation methodology based on concepts of patient safety was

agreed. Translation to a PhD research project was agreed in October 2004 before the first

birth at Casey Hospital in March 2005. A second supervisor with experience in nursing and

qualitative research methods and with organisational experience brought the research team

to three.

Extensive consultation ensued with relevant stakeholders to agree research and evaluation

methodologies. Existing processes would be used where possible. An extensive literature

review was undertaken to identify existing evaluation methodologies and to inform the

development of new ones that could be applied to Casey Hospital’s maternity service. That

literature review was written and edited to be suitable for peer-reviewed publication, work

that comprises Chapter Two. Reflecting restricted word limits for journal review articles

Page | 19

much of the original literature review was necessarily edited. The edited material not in

Chapter Three is presented here.

1.2 Literature review and Casey Hospital

Literature was accessed to aid the development of the research design. Initially, the scope

grew with each consultation when various stakeholders offered valuable insights about

their area of expertise. As the body of knowledge from the literature grew, so the research

design was refined. The literature provided herein is a summary of the information, rather

than a detailed analysis of the contents. Where the literature relates directly to my research

observations I have commented.

1.2.1 Evaluation methodologies

Evaluation of health services is complex, multi-layered and often disconnected with the

focus usually being a retrospective analysis of outcomes. However, evaluations should

include how system improvements are designed and implemented, including threats or

barriers to uptake, with recommendations of how these were overcome. They should

include detailed descriptions of initiatives that had a positive impact, and those that had a

negative effect, offering lessons for future planning (Hagedorn, Hogan, Smith, et al, 2006).

Evaluation research methodology rarely integrates health outcomes and the analysis of

systems of care. A systems approach to research can minimise fragmentation and support

organisational change to improve patient safety, particularly where research designs are

integrated in the evaluation methodologies (Pang, Sadana, Hanney, et al, 2003).

An evaluation framework to systematically introduce research outcomes in routine clinical

practice was developed by the United States Veterans Health Administrators (VHA, 2006).

The principle lessons from the quality enhancement research initiative (QUERI) are that

Page | 20

formative evaluation methods help explain what worked and what did not. Formative

evaluation includes qualitative analysis of clinicians’ individual and collective views about

their experiences. Themes uncovered by this method of qualitative research can inform

subsequent strategies, simultaneously engaging clinicians. Quantitative and qualitative

evaluation methods used in combination can provide a more holistic view and enable

system-wide improvements in healthcare delivery (Giacomini 2001, Salmon 2007, Porter

2007). These reports heavily influenced the research design and informed our mixed-

methods approach.

The “Australia’s Health Workforce” report (Productivity Commission, 2005) provided

insights to explain the challenges of engaging a disparate workforce to embrace system

improvement initiatives. The report identified the complexity of responsibilities in

healthcare delivery, shared between providers, employers, governing and accreditation

bodies, professional bodies, education providers and society more broadly. While the

primary aim of the study was to assess workforce shortages, the findings also help to

explain some of the many and significant challenges in building safer health care systems.

The report notes that all too often the skills of many healthcare workers are not used to

their full advantage and that habitual practice inhibits multidisciplinary learning

opportunities. I too observed this in the emergent themes of my qualitative research and I

discuss this in the context of the literature in Chapter Five.

Persistent variation in care provision and, most likely, outcomes is linked to a lack of

systematic documentation, outcome assessment and appropriately resourced evaluation

methods (Mainz 2003). This was evident in my quantitative research findings reported in

Chapter Four. Where evidence based clinical care is implemented to minimise variation in

Page | 21

practice it is often seen by clinicians, particularly senior clinicians, as being overly

restrictive and unable to account for “necessary” freedom and variations in complex

decision-making. Not surprisingly, such views impede the uptake and sustainability of

such safety initiatives (Kemm 2006). Some of the personnel I interviewed at Casey

Hospital reported such views (Chapter Five). In conflict with this clinician desire for

freedom, the principles of reliability that have been shown to reduce variability and

minimise opportunities for system failures and that are used throughout other industries,

such as aviation and manufacturing are increasingly recommended for healthcare (Nolan,

Resar, Haraden, et al, 2004). In the study related in Chapter Four I specifically sought to

explore the complexities of hospital structures and processes that are thought to contribute

to care deficiencies.

Variation in outcomes is also attributed to the extent of uptake of research-based clinical

care. An audit of medical records to ascertain the implementation of research-based

preventative methods to improve health outcomes in paediatrics showed disparate uptake

(Young and Glade, 2006, p4). However, better outcomes were demonstrated where

collaborative working groups had the opportunity to learn about implementation successes,

failures and barriers together, in essence including clinicians in the theory behind

“constrained” or managed clinical care. In particular, factors noted to improve acceptance

by clinicians and subsequent implementation included identifying the goal of “reducing

deviation and variation from evidenced-based or generally agreed-on practice guidelines

for care” (Young and Glade, 2006, p4) and subsequent checking that the goal is achieved.

The challenges to acceptance and implementation included appropriate timing of review

processes relative to implementation, sustainability and the ability to generalise across

different settings (Young and Glade 2006). Our processes of evaluation are a typical

Page | 22

example of the time delay between implementation and evaluation. The service became

operational from early 2005 and the first publications of the evaluation process arose in

2010. Our quantitative results, described in Chapter Four, reflect the findings of Young

and Glade (2006), with wide variation in compliance with organisational standards. To be

effective evaluation needs to regular, timely and delivered locally – attributes that are all

too often missing from local, regional and national systems.

1.2.2 Application to the research design

The literature was imperative in defining the scope of the project. The evaluation literature

made it clear that the approach should be multifaceted and include quantitative and

qualitative data to provide information about how care was delivered in maternity services,

rather than focus on outcome data of which there are abundant reports – many with limited,

if any, traction. In an organisational climate with a sound reputation for its commitment to

patient safety and quality of care, this framed the research direction. Patient safety

literature not referred to in Chapter Two but important in shaping the research design is

reported below.

1.2.3 Patient safety

In the context of the history of modern medicine, the patient safety movement is relatively

new, gathering momentum really only since the early 1990s. A classic study in 1991

(Brennan, Leape, Laird, et al) found that almost 4% patients admitted to health care

facilities suffered an iatrogenic event. Of these, 14% were fatal, resulting in 13,805 deaths.

While that study was not the first to explore adverse events the methodology is considered

a cornerstone of contemporary patient safety initiatives (Baker 2004). In their seminal

work “To err is human” Kohn, Corrigan and Donaldson (1999) urged healthcare

Page | 23

organisations to build safer systems of care and avoid blaming individuals for medical

errors. Lessons that too many hospitals are yet to heed today.

At national and international levels the need for formal approaches to patient safety is well

established. The World Health Organisation launched the World Alliance for Patient

Safety in 2004 in response to a World Health Assembly Resolution urging the need for

attention to improving patient safety. The Joint Commission, responsible for accreditation

of health services in the USA, has 800 tools available online. Similar tools were developed

for application in the UK. The Australian Commission on Safety and Quality in Healthcare

launched a National Patient Safety Education Framework in 2005 and a toolkit in 2006.

Australian Government agencies have developed tools for local adaptation including the

Victorian Quality Council’s organisational checklist for safety and quality. Tools to help

find areas for improvement include specific triggers for case reviews (Brennan, et al, 1991)

including perinatal trigger tools launched by the Institute of Health Care Improvement

(IHI) in 2005.

At a health service level, clinical indicator reporting provides opportunities for

benchmarking between health services and for monitoring trends over time, both within

and between services. Such benchmarking has been shown to help identify improvement

opportunities (Collopy 2000). However, while analyses of clinical indicators are

unquestionably useful the methods to interpret data and share lessons among peers appear

are all too often ad hoc. For example, the USA Agency for Healthcare Research and

Quality (AHRQ) developed Failure to Rescue quality indicators (2007). The Failure to

Rescue indicator is intended to identify patients who die after developing a complication

that was failed to be addressed. The underlying assumptions were that these deaths were

Page | 24

avoidable and that good hospitals identify complications quickly and treat them

aggressively. However, while this tool is useful in retrospective analysis there is no

evidence that its implementation has reduced the number of events of failure to rescue.

Good in theory but no evidence of effectiveness in practice.

In the context of maternity care, a maternity audit tool to assess appropriateness of

intrapartum care was developed and described by Simpson (2005). The tool identifies

points along the continuum of care where diversions or omissions from expected care

might occur. The tool is proscriptive, requiring 100% compliance with practice principles

and is complemented with strategies to engage clinicians. Where variations arise, there is

an expectation of a multi-disciplinary review focussed on systems of care to identify

opportunities for improvement and make recommendations to prevent recurrences. The

tool provided consistent measures to assess perinatal patient safety (Simpson 2006).

Tools are available from the Perinatal Society of Australia and New Zealand (PSANZ) to

aid clinical review after stillbirth and neonatal death. The PSANZ tools differ from the

“failure to rescue” suite because they are not designed to prevent future events but to better

interrogate current ones. Recommendations of perinatal care based on the latest available

evidence accompany the tools but there is little to inform perinatal care providers regarding

methods of clinical practice and system review that uncover barriers to implementation of

best practice, as the “failure to rescue” package offers.

The National Health Service (NHS) in the UK had a more didactic, system wide approach

to clinical risk management via the Clinical Negligence Scheme for Trusts (1995). The

NHS Litigation Authority manages liabilities incurred for clinical negligence claims. The

Authority uses financial incentives to improve patient safety within a system-wide

Page | 25

framework. Compliance with standards is assessed at regular intervals and results

determine premiums paid. However, even such an apparently well structured safety

framework does not prevent major service failures, including failure of maternity services

such as at Furness General Hospital (Kirkup 2015).

An innovative partnership between a US healthcare organisation and its insurers in

addressing perinatal patient safety is the establishment of a dedicated perinatal patient

safety nurse (Brown Will, Hennicke, Jacobs, et al, 2006). The role resulted from concern

about medical liability and malpractice and is described as a proactive step to embed

evidence into clinical care, lead clinical audit reviews with a system based approach, focus

on shared lessons and service design improvements to minimise the opportunity for

adverse events to recur. The role is seen to contribute positively in building a patient safety

culture in obstetrics and includes clinical advice to clinicians concerned about risk, for

example in interpreting cardiotocography. The role promotes clinical audit as a function of

routine clinical care in which all clinicians can participate.

It is difficult to reconcile that these initiatives have impacted positively on patient care

when system failures contributing to maternal and fetal harm continue to be published

(Kirkup 2015). However, they provide important historical perspectives to improve patient

safety.

1.2.4 Casey Hospital maternity service

As detailed above, page 17, the origin of the work leading to this PhD was a government

desire to evaluate the state’s newest public maternity service being established in new

build general hospital – Casey hospital – in Berwick on the south-east fringes of

Page | 26

metropolitan Melbourne. To provide better context for the evaluation findings reported in

this thesis a brief history of the maternity service is detailed here.

Casey Hospital opened its doors in 2004. The first birth was in March 2005. On

commissioning, Casey Hospital was to provide birthing services for 500 low risk births per

annum. This was increased to 1000 births per annum in 2006 and thence to 1500 births per

annum in 2007-2013. In 2014, capacity was further increased to 2000 births per annum and

in 2017 Casey Hospital will provide care for 2500 births per annum. Figure 1.1

summarises birth numbers at Casey Hospital 2004-2015 inclusive.

Figure 1.1. Number of births (per month) at Casey Hospital 2004-2015.

Dotted lines denote planned births per month. Bars denote actual births per month.

At the outset, the birth suites at Casey Hospital were built to accommodate sustained

growth into the future with a longer term maternity service plan that would allow 3500-

4000 births per annum. This is important because while birth numbers at the hospital

steadily increased from 2004 onwards, the birthing capacity of the hospital was never

limited by the availability of birthing rooms. To illustrate this, Figure 1.2 details the

number of births per birth suite room per annum for all three of Monash Health’s maternity

0

20

40

60

80

100

120

140

160

180

200

Jul-0

4 O

ct-0

4 Ja

n-05

Ap

r-05

Jul-0

5 O

ct-0

5 Ja

n-06

Ap

r-06

Jul-0

6 O

ct-0

6 Ja

n-07

Ap

r-07

Jul-0

7 O

ct-0

7 Ja

n-08

Ap

r-08

Jul-0

8 O

ct-0

8 Ja

n-09

Ap

r-09

Jul-0

9 O

ct-0

9 Ja

n-10

Ap

r-10

Jul-1

0 O

ct-1

0 Ja

n-11

Ap

r-11

Jul-1

1 O

ct-1

1 Ja

n-12

Ap

r-12

Jul-1

2 O

ct-1

2 Ja

n-13

Ap

r-13

Jul-1

3 O

ct-1

3 Ja

n-14

Ap

r-14

Jul-1

4 O

ct-1

4 Ja

n-15

Ap

r-15

Jul-1

5 O

ct-1

5

Page | 27

facilities – Monash Medical Centre, Dandenong Hospital and Casey Hospital. On average,

Victorian public hospital birth suite rooms service 380 births per annum – slightly over one

per day. At Casey Hospital, even when it was providing care for 2000 births per annum,

the number of births per room only reached 265. By comparison, Dandenong Hospital,

providing care for 2500 births per annum, averages 380 births per room per annum and

Monash Medical Centre, providing care for about 4000 births per annum, averaged 520

births per room per day.

Figure 1.2. Average number of births per birth suite room per year, Monash Health

The “capacity” of a birthing service does not only relate to numbers of births. It also relates

to case complexity. In this regard, Casey Hospital maternity service was part of a single

integrated service, providing care for low risk women. The other two services provided

care for medium risk (Dandenong) and high risk (Monash) pregnancies. The clinical

governance supporting this single integrated service included booking inclusion/exclusion

and transfer criteria for each hospital. To provide a sense of the capability of Casey

Hospital when the evaluation was undertaken, these are detailed on the following pages.

0

100

200

300

400

500

600

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Monash Dandenong Casey

Page | 28

While a detailed description of perinatal outcomes for Casey Hospital is beyond the scope

of the studies in this PhD, the ultimate measure of quality of maternity service is perinatal

mortality. In Victoria, perinatal mortality is expressed as the Gestation Standardised

Perinatal Mortality Ratio (GSPMR). By definition, the state average is 100. (It is a ratio).

Figure 1.3 details the GSPMR for all three of the Monash Health maternity services 2003-

2015. It can be seen that the GSPMR at Casey Hospital (and Dandenong Hospital) were

significantly above the Victorian average but that the rate has been progressively falling.

Figure 1.3. GSPMR for Monash Health maternity services 2003-2015, inclusive

It is hoped that this brief summary of Casey Hospital maternity service provides a useful

backdrop to the studies and results that follow.

0

20

40

60

80

100

120

140

160

180

2003-7 2004-8 2005-9 2006-10 2007-11 2008-12 2009-13 2010-14* 2011-15*

Rat

io

MMC DH CH Victoria

Page | 29

Clinical Practice Protocol

Women’s & Children’s Program, Maternity, Casey Inclusion/exclusion criteria

page 1 of 2

Casey Hospital (Casey) – inclusion/exclusion criteria This protocol must be read in conjunction with Casey Transfer Out criteria protocol

Definition Criteria for selection of women booked to birth at Casey

Who Doctors, midwives

Expected outcome

Appropriate selection of women booked to birth at Casey

Booking Inclusion Criteria

Parity: ≥ 5 subject to consultant obstetrician review

Age: no limits

Stature: Weight should be gauged by body mass index (BMI) and be ≤35 at first presentation

Booking Exclusion Criteria

Women with any of the following conditions should not be booked for maternity care at Casey

Medical/Surgical History: • alcohol abuse • anaesthetic difficulties • anaemia (Hb<90g/l not responding to treatment) • antiphospholipid antibodies • asthma and/or chronic bronchitis requiring hospitalisation within the last 12

months • autoimmune disorders • bleeding disorder and/or haemolytic disease • BMI ≥ 35 • cardiac disease • coagulopathies • deep vein thrombosis • diabetes mellitus, types I and II • drug abuse • endocrine disorders • epilepsy requiring drug therapy in the past 12 months • haemoglobinopathies • HIV • hypertension • gastroenterological disorders • Infectious diseases (eg acute onset Hep C) • neurological disorders • pulmonary embolus • psychiatric illness (subject to psychiatric advice) • renal disease • respiratory disease • thyroid disease (uncontrolled) Obstetric/Gynaecological History/Findings: • absence of antenatal care in a woman who is 28 weeks gestation or more at first

presentation • birth weight < 2500gms or > 4500gms (previous baby, subject to obstetric review) • cervical anomalies • cervical incompetence

Page | 30


Women’s & Children’s Program, Maternity, Casey Inclusion/exclusion criteria

page 2 of 2

• congenital anomaly (subject to obstetric review) • cone biopsy (subject to obstetric review) • fetal abnormality requiring specialised neonatal care • mid-trimester abortion consistent with cervical incompetence • multiple pregnancy • other obstetric, medical, psychological or social factors (not listed formally herein)

may also constitute reason for exclusion. Decisions based on such other factors will be made only after suitable consultation.

• placental abruption • postpartum haemorrhage x 2 or significant postpartum haemorrhage x 1(subject to

obstetric review) • pre-eclampsia in the previous pregnancy • puerperal psychosis • rhesus isoimmunisation • stillbirth (not due to congenital malformation and subject to obstetric review) • thyroid disease • uterine surgery including Caesarean section, myomectomy and tubal reanastomosis

(involving the cornua)

• uterine myoma or malformation, abdominal or adnexal masses (subject to obstetric specialist review)

Postnatal transfers

Where appropriate postnatal transfers will occur back to the hospital / service where antenatal care was delivered when site admission criteria allows, and for extended postnatal care -postnatal domiciliary

If an adverse event occurs, document details in the health record and complete an incident report, notify site and program DONs. If a sentinel event occurs, notify the Southern Health Patient Safety Officer. See Quality Intranet site:

http://webserver/quality/Patient%20Safety/safety_frame.htm

This hard copy is only current at the time of printing

Strategic policy Patient care ACHS Continuum of care

Reviewer Maternity Guideline Development Group Last review date Mar 2006

Authoriser Maternity Executive Committee Next review date Apr 2007

Page | 31


Women’s & Children’s Program, Maternity, Casey Transfer out criteria

page 1 of 1

Casey Hospital (Casey) – transfer out criteria This protocol should be read in conjunction with Casey booking criteria protocol

Definition Criteria for women to be transferred from Casey to another SH maternity site

Who Doctors, midwives

Expected outcome

Women will be transferred to an appropriate model of care within Southern Health as clinically indicated

All transfers necessitating admission to another hospital requires a Casey obstetrician to speak directly with an obstetrician at the accepting hospital (consultant to consultant).

Indication for Antenatal Transfer Out

• APH • breech presentation in women who wish to attempt vaginal birth • Fetal abnormality requiring specialised neonatal care • Gestational diabetes requiring insulin • Intrauternie Growth Restriction (IUGR) (<32 weeks) • Medical or surgical conditions requiring treatment that may interfere with

pregnancy or increase the risk of preterm labour >23 weeks and <34 weeks • Preterm prelabour rupture of membranes • Rhesus isoimmunisation • Other obstetric, medical, psychological or social factors (not listed formally herein)

may also constitute reason for transfer. Decisions based on other such factors will be made only after suitable consultation

Indication for Intrapartum Transfer Out

Intrapartum transfer to a tertiary hospital should be considered for any of the

following after discussion with the obstetric consultant:

• APH (where it is safe to do so)

• Preterm labour >23 weeks

• Premature Rupture of Membranes >23 weeks

• Any situation where specialist neonatal or genetic involvement is necessary >23weeks

Postnatal transfers

Where appropriate postnatal transfers will occur back to the hospital / service where antenatal care was delivered when site admission criteria allows, and for extended postnatal care -postnatal domiciliary

If an adverse event occurs, document details in the health record and complete an incident report, notify site and program DONs. If a sentinel event occurs, notify the Southern Health Patient Safety Officer. See Quality Intranet site:

http://webserver/quality/Patient%20Safety/safety_frame.htm

This hard copy is only current at the time of printing

Strategic policy Patient care ACHS Continuum of care

Reviewer Maternity Guideline Development Group Last review date Mar 2006

Authoriser Maternity Executive Committee Next review date Apr 2007

Page | 32

1.2.5 References

Agency for Healthcare Research and Quality (AHRQ) quality indicators. Patient Safety.

Failure to Rescue. 2007; (available from:

www.qualitymeasures.ahrq.gov/summary/summary.aspx?doc_id=10684).

Australian Commission on Safety and Quality in Health Care, Measurement for

improvement toolkit. 2006; (available from:

http://www.safetyandquality.gov.au/internet/safety/publishing.nsf/Content/703C98BF3752

4DFDCA25729600128BD2/$File/ Toolkit_PartA.pdf).

Australian Commission on Safety and Quality in Health Care, downloaded 12/10/08

(available from http://www.safetyandquality.gov.au/publications/national-patient-safety-

education-framework-2005/) and (http://www.safetyandquality.gov.au/publications-

resources/publications/?yr=2006).

Baker G. Harvard Medical Practice Study. Quality and Safety in Healthcare. 2004; 13 (2):

151-2.

Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in

hospitalized patients - results of the Harvard Medical Practice Study I. The New England

Journal of Medicine. 1991; 324: 370-6.

Will SB, Hennicke KP, Jacobs LS, et al. The perinatal safety nurse: a new role to promote

safe care for mothers and babies. Journal of Obstetric and Gynaecological and Neonatal

Nursing. 2006; 35 (3): 417-23.

Collopy BT. Clinical Indicators in accreditation: an effective stimulus to improve patient

care. International Journal for Quality in Health Care. 2000; 12: 211-6.

Page | 33

Giacomini MK. The rocky road: qualitative research as evidence. Evidence Based

Medicine. 2001; 6: 4-6.

Hagedorn H, Hogan M, Smith JL, et al. Lessons learned about implementing research

evidence into clinical practice: experiences from VA QUERI. Journal of General Internal

Medicine 2006; 21 (2): 21-4.

Institute for Healthcare Improvement. (available from http://www.ihi.org/ihi/about).

Institute for Healthcare Improvement. Perinatal Trigger Tool. July 2006;

(available from:

http://www.ihi.org/IHI/Topics/PerinatalCare/PerinatalCareGeneral/EmergingContent/Perin

atalTriggerTool.htm).

Joint Commission (available from: http://www.jointcommission.org/PatientSafety/).

Kemm J. The limitations of ‘evidence-based’ public health. Journal of Evaluation in

Clinical Practice. 2006; 12 (3): 319-24.

Kirkup B. The report on the Morecombe Bay Investigation. March 2015; (available from:

www.gov.uk/government/publications).

Kohn L, Corrigan J, Donaldson M. To Err is Human. Building a Safer Health System.

Washington, DC, USA. National Academies Press. 1999.

Mainz J. Defining and classifying clinical indicators for quality improvement.

International Journal for Quality in Healthcare. 2003; 15 (6): 523-30.

Page | 34

National Patient Safety Agency, United Kingdom. Patient Safety Toolkit

(available from: http://www.npsa.nhs.uk/patientsafety/alerts-and-directives/patient-safety-

toolkits-e-learning-packages/).

NHS Litigation Authority Clinical Negligence Scheme for Trusts. (available from:

http://www.nhsla.com/RiskManagement/).

Nolan T, Resar R, Haraden C, et al. Improving the reliability of healthcare. Institute for

Healthcare Improvement. 2004.

Pang T, Sadana R, Hanney S, et al. Knowledge for better health – a conceptual framework

and foundation for health research systems. Bulletin of the World Health Organization.

2003; 81 (11): 815-20.

Perinatal Society of Australia and New Zealand. Perinatal mortality audit package.

(Available from: http://www.psanz.com.au/).

Porter S. Research methodology. Validity, trustworthiness and rigour: reasserting realism

in qualitative research. Journal of Advanced Nursing. 2007; 60 (1): 79-86.

Productivity Commission 2005, Australia’s Health Workforce. Research Report, Canberra

(Available from: www.pc.gov.au)

Salmon A. Walking the talk: how participatory interview methods can democratize

research. Qualitative Health Research. 2007; 17 (7): 982-93.

Simpson KR. Measuring perinatal patient safety: review of current methods. Journal of

Obstetric, Gynecologic and Neonatal Nursing. 2006; 35 (3): 432-42.

Page | 35

Simpson KR. Failure to rescue: implications for evaluating quality of care during labour

and birth. Journal of Perinatal and Neonatal Nursing. 2005; 19 (1): 24-34.

Veterans Health Administrators (VHA) quality enhancement research initiative (QUERI)

(available from: http://www.hsrd.research.va.gov/queri/)

Victorian Department of Health. Victorian Maternity Services Performance Indicators.

(available from: www.health.vic.gov/maternitycare).

Victorian Quality Council organisational checklist for safety and quality processes

(available from:

http://www.health.vic.gov.au/qualitycouncil/downloads/checklist_safety_quality.pdf).

World Health Organization. World Alliance for Patient Safety (available from:

http://www.who.int/patientsafety/worldalliance/en/).

World Health Organisation, Making Pregnancy Safer (Available from:

http://www.who.int/making_pregnancy_safer/en/).

Young PC, Glade GB, Stoddard GJ, et al. Evaluation of a learning collaborative to

improve the delivery of preventive services by pediatric practices. Pediatrics. 2006; 117

(5): 1469-76.

Page | 36

1.3 Hypotheses

Three key hypotheses underlie the research undertaken towards this doctoral thesis. These

are that:

1. variation in outcomes reflects variation in practice;

2. not all variation in practice is reflected in outcome data; and

3. outcome measurement alone provides limited information about the overall quality

of care.

1.4 Aims

The broad aim of my research was to undertake an evaluation of a new maternity service at

Casey Hospital and to identify opportunities for safety enhancement. My literature review,

undertaken in preparation for and to inform the evaluation, led me to plan a mixed-

methods approach to the evaluation underpinned by patient safety concepts. I also decided

to focus on pregnancy (antenatal) care because there appeared to be a gap in the literature

addressing this aspect of care. In contrast, intrapartum care is well represented and covered

by ample publications, both routine government reports and peer-reviewed research papers,

reporting rates of stillbirth, caesarean section, assisted vaginal births and perineal trauma.

Accordingly, my specific aims were to measure safety and quality of pregnancy (antenatal)

care in a new maternity service by:

1. developing a tool to audit the process of maternity care;

2. undertaking an audit of how maternity care is delivered;

3. assessing overall validity and utility of the audit tool; and

4. exploring staff perceptions about the safety and quality of maternity care at their

service

Page | 37

The results will be used within the organisation to inform ongoing quality improvement

activities.

To initially address these aims, I searched the literature to:

1. identify initiatives to improve patient safety, and their application to or adaptation

for maternity services;

2. identify tools that measure patient safety in maternity services that could be

applied; or

3. use information from the literature to develop tools to measure pregnancy care if

there were none that could be directly applied from the literature.

1.5 Thesis structure

The thesis is presented in the sequence that the research was undertaken. The provision of

care in any system is unique, probably due in part to inadequate systematic reporting

capabilities, and compounded by individual clinician preferences as previously reported

and explored in detail throughout the various chapters in the thesis. The Casey Hospital

maternity service is no different. It is a unique service within a greater organisational

system that itself is nested within a state-wide maternity system and public health system.

As previously noted, an exploration of the literature and wide consultation provided

abundant possibilities for the research design. The information was distilled to provide

clarity and direction to both the research and the evaluation. The literature review is

summarised in Chapter One: General Overview and was published as a peer-reviewed

paper as detailed in Chapter Two: Patient safety – a literature review to inform an

evaluation of a maternity service.

Page | 38

Clinicians were extensively consulted and engaged to participate and refine the research

methodology, agreeing that the quantitative component should focus on pregnancy care

allowing an opportunity to compare the various models of multidisciplinary care. Chapter

Three: Designing a clinical audit tool to measure processes of pregnancy care describes

why and how and clinical audit tool was developed to measure processes, as opposed to

outcomes, of care during pregnancy. The paper emphasises the importance of clinician

involvement throughout any audit cycle and details the many iterations of consultation to

ensure ongoing engagement. Pleasingly, the Australian Council on Healthcare Standards

advised the PhD candidate of their reference to the published material in their education

programme about developing clinical audits.

Chapter Four: Measuring pregnancy care – towards better maternal and child health

describes the results of the application of the clinical audit tool to the medical records of

women receiving their care at Casey Hospital. While quality of care overall was good,

wide variation in compliance was evident in our results, as has been reported elsewhere by

others. Recommendations for future practice and system design are offered.

Chapter Five: Perinatal staff perceptions of safety and quality in their service reports the

qualitative research component. The importance of clinician engagement and participation

is emphasised and described in detail. Our qualitative data led to the development of

emergent themes, most of which overwhelmingly affirmed that the maternity service is

safe and that the quality of care is of a high standard supported by comprehensive support

systems locally and at organisational levels. An emergent theme of concern was the

pervasive divisiveness between the various disciplines responsible for maternity care,

particularly between midwifery and obstetrics (junior medical staff). While not surprising –

Page | 39

similar tensions have been reported by others – it was disappointing to find professional

dichotomies in such a new service. The organisation recognised some of these issues early

and put mechanisms in place to redress them. These remedial initiatives are referred to.

Chapter Six: Discussion and closing remarks summarises the work and its implications,

with recommendations for further research.

Page | 40

CHAPTER TWO

Patient safety: a literature review to inform an evaluation of a

maternity service

Page | 41

2.1 Preamble

The publication, “Patient safety: A literature review to inform an evaluation of maternity

service” was introduced in detail in Chapter One (page 17). As noted, the research team

invested much of the first 12 months consulting widely about how to approach an

evaluation that would add value to the organisation, meet the needs of the various

stakeholders, and the requirements of PhD candidature. Considerable amounts of literature

were accessed, the most relevant of which have been include throughout the thesis. This

published literature review summarises our decision making in this regard.

Page | 42

2.2 Patient safety: a literature review to inform an evaluation of a maternity service

Page | 43

Page | 44

Page | 45

Page | 46

Page | 47

CHAPTER THREE

Designing a clinical audit tool to measure processes of pregnancy

care

Page | 48

3.1 Preamble

The literature was resounding in the messages about the gaps in patient safety. In

particular, a number of key messages were identified that critically informed the design of

evaluation and the overarching research reported in this thesis. These messages were that:

• health care is complex;

• there are no agreed measures of safety, particularly in relation to measuring safety

through a patient episode; and

• clinician engagement is essential.

In the publication Designing a clinical audit tool to measure processes of pregnancy care I

describe the methodology of the quantitative component of my mixed-methods research,

which I hope addresses gaps that I identified in the literature.

Page | 49

3.2 Designing a clinical audit tool to measure processes of pregnancy care

Page | 50

Page | 51

Page | 52

Page | 53

Page | 54

Page | 55

Page | 56

Page | 57

Page | 58

Page | 59

Page | 60

CHAPTER FOUR

Measuring pregnancy care: towards better maternal and child

health

Page | 61

4.1 Preamble

This paper related in this chapter provides the results and associated discussion of the

findings of the clinical audit undertaken using the audit tool I designed, as described in

Chapter Three (page 47). As discussed already in Chapter One (page 17) and Chapter Two

(page 40), the complexity of health care provision hampers the ability to effectively and

efficiently measure health care delivery. The well-described and oft-reported problems of

gaps in information and illegibility of records were, not surprisingly, observed by me in my

Measuring pregnancy care. Nonetheless, the audit tool was developed to capture holistic

information about pregnancy care and provided abundant data for analysis. My

observations prompted recommendations for improvement, not just at Casey Hospital but

across the service systems.

Page | 62

4.2 Measuring pregnancy care: towards better maternal and child

health

Original Article

Measuring pregnancy care: towards better maternal and child health

Suzanne V. SINNI,1,2 Wendy M. CROSS,3 Amy E. SWANSON2 and Euan M. WALLACE1,2

1Department of Obstetrics and Gynaecology, School of Clinical Sciences, Monash University, Clayton, Victoria, Australia, 2MonashWomen’s Services, Monash Health, Clayton, Victoria, Australia and 3School of Nursing and Midwifery, Monash University, Clayton,Victoria, Australia

Background: Obstetrics remains the largest medico-legal liability in healthcare. Neither an increasing awareness of patientsafety nor a long tradition of reporting obstetric outcomes have reduced either rates of medical error or obstetric litigation.International debate continues about the best approaches to measuring and improving patient safety. In this study, we setout to assess the feasibility and utility of measuring the process of maternity care provision rather than care outcomes.Aims: To report the development, application and results of a tool designed to measure the process of maternity care.Materials and Methods: A dedicated audit tool was developed, informed by local, national and international standardsguiding best practice and then applied to a convenience sample of individual healthcare records as proof of function.Omissions of care were rated in order of severity (low, medium or high) based on the likelihood of serious consequenceson patient safety and outcome.Results: The rate of high severity omissions of care was less that 2%. However, overall rates of all omissions varied from0 to 99%, highlighting key areas for clinical practice improvement.Conclusions: Measuring process of care provision, rather than pregnancy outcomes, is feasible and insightful, effectivelyidentifying gaps in care provision and affording opportunities for targeted care improvement. This approach to improvingpatient safety, and potentially reducing litigation burden, promises to be a useful adjunct to the measurement of outcomes.

Key words: clinical audit, clinical error, patient safety, pregnancy care.

IntroductionImproved maternal and child health outcomes are keygoals for those providing maternity care.1,2 Measuring andreporting these outcomes have long been a cornerstone ofhigh-quality maternity care,3–6 identifying opportunities fortargeted improvements. This approach has underpinnedregional and national review processes in Australia andelsewhere.3–6

While reporting outcomes provides insightful compar-isons between hospitals, regions or nations, it does notaddress why outcomes differ and how healthcare may beimproved. Thus, measuring outcomes alone is unlikely tobe ever able to provide a complete picture of the qualityand safety of healthcare. Accordingly, it is increasinglyapparent that measuring and reporting healthcare processwill deliver better outcomes.7,8

Access to antenatal care by trained professionals isfundamental to ensuring optimal maternal and child healthoutcomes.1,2,9–11 Recommended components of routinepregnancy care are well described and have been used toinform regulatory, professional and organisationalexpectations of care providers.1,2,12–17 We used theserecommendations, applying principles of patient safetyand clinical audit methodology to measure pregnancy care,assessing whether such an approach would be feasible andinformative.

Materials and MethodsWe developed a clinical audit tool to measure pre-defined elements of pregnancy care.18 Briefly, we usedlocal, national and international clinical practice guide-lines12,13,15–17 to develop minimum ‘standards’ of careagainst which provision of care was audited18 (Table 1).We divided the audit tool into five sections (demographicinformation, first pregnancy visit, antenatal screening,antenatal visits summary and documentation) and applied itto the health records of a random convenience sample ofwomen who had given birth in an outer metropolitanmaternity unit with about 1500 births per annum.Individual patient health records were hand searched, and

Correspondence: Suzanne Sinni, Department of Obstetrics andGynaecology, Monash University, Monash Medical Centre,246 Clayton Road, Clayton, Vic., Australia.E-mail: [email protected]

Received 9 September 2015; accepted 29 December 2015.

© 2016 The Authors Australian and New Zealand Journal of Obstetrics and Gynaecology published by John Wiley & Sons Australia, Ltd

on behalf of Royal Australian and New Zealand College of Obstetricians and Gynaecologists.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any

medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

1

Australian and New Zealand Journal of Obstetrics and Gynaecology 2016 DOI: 10.1111/ajo.12441

Th e Australian and New Zealand Journal of Obstetrics and Gynaecology

Page | 63

data were entered and stored in a custom-made database(Filemaker Pro 9.0 v3; Filemaker Inc., Santa Clara, CA,USA). Data missing from a health record were recorded asabsent or unavailable and the care assumed not to havebeen provided. Maternal or fetal conditions or clinicalassessments that required further investigation or action andwhere no such action was recorded were interpreted as‘abnormal not referred’.18 Such results were consideredomissions of care with the potential to adversely affectpatient safety, the latter being a higher severity omission.We acknowledge that missing information may not reflectdeficiencies in care delivery but rather a deficiency in therecording of that care. However, in a healthcare system thatdepends upon multiple providers, a lack of documentationpresents risks of harm. The number of women withomissions of care is reported as a percentage of the healthrecords audited. The severity of the omission (low,medium, high) was determined by the degree to which careprovision was within the control of the care provider, and/orits potential to cause harm. Specifically, a pragmaticapproach to severity scoring was agreed between two of the

authors (SS, EMW) depending on either the likelihood ofthe error to potentially cause significant harm or, in the caseof omission of recording of care, the likelihood to expose thehealth service to successful litigation. Departures from careexpectations as detailed in the hospital’s own clinicalpractice guidelines17 were also accorded a higher severityrating because these were considered to reflect explicit localdeficiencies. Data analyses were performed using Prism !

software (Prism 2008; GraphPad Software Inc., v5.0d,SanDiego, CA, USA).Ethics approval for the study was granted by Monash

Health Human Research and Ethics Committee and bythe Monash University Research Ethics Committee.

ResultsA total of 354 health records, 24% of all women givingbirth at the hospital, were analysed. To confirm validity, arandom selection of 90 (25%) records was assessedindependently. Inter-rater reliability yielded a kappa scoreof 0.88.

Table 1 Details of standards of care (* organisational documentation requirements are unreferenced)

No. Standard References

1 If an early ultrasound (<14 weeks) result is available, the EDD is calculated from that ultrasound rather thanthe last known menstrual period (LMP). If an ultrasound is performed after 14 weeks gestation, the EDD iscalculated from the last menstrual period (if known, monthly and regular) unless the ultrasound differs bymore than one week. An ‘agreed’ EDD is to be made as early as possible in pregnancy and documented in themedical record and hand-held card.

18, 19, 22

2 Women should have their visit pregnancy care visit, excluding a GP visit solely to confirm pregnancy, between7 and 12 weeks’ gestation.

21

3 At the first visit, measure maternal weight and height and calculate body mass index (BMI) 18, 19, 22

4a A full medical and family history and a physical examination are to be completed at the first visit, andappropriate referral is made where indicated

21

4b At the first visit, discuss smoking behaviour/cessation, recreational drug use and alcohol consumptionimplications. At every subsequent visit, smoking behaviour enquiry and cessation advice and support isprovided if indicated.

22

5a At the first pregnancy visit, women are to be offered the following investigations: blood group and antibodies,full blood examination (including screening for haemoglobinopathy); serology for syphilis, hepatitis B, HIV andrubella; midstream specimen of urine (for culture and sensitivity).

18, 19, 21, 22

5b All women should be offered Down Syndrome screening. 18, 19, 22

6 At the 2nd visit (12–15 weeks) all women should be offered a 19–21 week morphology ultrasound to assessgestational age, placental position and fetal morphology.

18, 19, 22

7 All women should be offered screening for diabetes during pregnancy. 18, 19, 22

8a An assessment of a woman’s blood pressure must be made at every visit. 18, 19, 22

8b An assessment of symphyseo-fundal height (SFH) should be measured at every visit after 20 weeks gestation. 18, 19, 22

8c Abdominal palpation to determine fetal lie and presentation should occur at every visit after 30 weeksgestation.

18, 19, 22

8d Women should be asked about fetal movements at every visit. 22

8e Neonatal vitamin K administration and Hepatitis B vaccination should be discussed during pregnancy. 22

8f Anti-D administration to Rhesus (Rh) negative women 22

9a All pages in a health care record should have the patient’s unique identifier. *9b All entries in a health care record should be legible, dated and signed, with each signature accompanied by the

author’s legible name.*

10 Visit schedule: 10–12 weeks – hospital midwife assessment and confirmation of maternity booking, 12–15 weeks (obstetric review), 20–22 weeks (after mid-trimester ultrasound performed), 26–28 weeks, 31 weeks,34 weeks, 36 weeks, 38 weeks, 39 weeks (nullipara only), 40 weeks, 41 weeks.

22

2 © 2016 The Authors Australian and New Zealand Journal of Obstetrics and Gynaecology published by John Wiley & Sons Australia, Ltd


S. V. Sinni et al.

Page | 64

Demographic data were collected for descriptivepurposes. The 354 women were born in 52 differentcountries – 200 (55%) in Australia, 23 (6.5%) in India, 15(4%) in New Zealand, 15 (4%) in Sri Lanka, 11 (3%) inAfghanistan and the remaining 81 in 46 other countries.Twenty (6%) women required an interpreter. The mean(range) age of the women was 28 (17–43) years.

Standards of care

Overall rates of compliance are detailed in Table 2. Therate of high severity errors was 1.7%.

First pregnancy care visitStandard 1: Calculation and recording of estimated duedate (EDD). Of the 354 records, one record (0.3%) didnot have an EDD recorded. In 318 (90%) records, a firsttrimester ultrasound scan had been used to determine theEDD. Of the remaining 35 records, various methods havebeen used to calculate the EDD, only one of which wascompliant with agreed organisational procedure. In total,35 (10%) records were non-compliant with standard 1.Error severity: LOW.

Standard 2: Gestation at first visit. Fifty (14%) womenhad had their first antenatal visit by 12 weeks’ gestation(including those at 12 weeks gestation). The majority ofwomen (198, 56%) were first seen at 13–18 weeksgestation (Fig. 1). Overall, the care for 304 (86%) womenwas outside the standard. Error severity: MODERATE.Standard 3: Calculation and recording of maternalbody mass index (BMI) at first visit. Thirty-two (9%)women did not have their BMI calculated at their firstvisit. Error severity: LOW.Standard 4a: Medical and family history and physicalexamination. There were 224 (63%) records that werenon-compliant with this standard. Ten (3%) women hadno record of a medical history being taken, and in nine(2%) women, there was no record of a family historybeing taken. There was no record of a physicalexamination in 133 women (38%). Error severity:MODERATE.Standard 4b: Smoking, recreational drug use andalcohol consumption. Eleven (3%) women were notasked about smoking. Of the 45 women who reported thatthey smoked, 39 (87%) had no evidence of follow-upquestions during pregnancy. The health records of five

Table 2 Rate of non-compliance with Standards of Care (** denotes HIGH risk item)

Standard of careRate (%)

non-compliance Standard of careRate (%)

non-compliance

First Visit (Standards 1–4)EDD recorded 0.3 Physical examination recorded 38.0EDD method recorded 9.9 Smoking status recorded 3.1BMI recorded at first visit 9.1 Smoking follow-up recorded 11BMI>35 0.3 Drug use status recorded 1.4Medical history completed 2.8 Alcohol use follow-up recorded 2.3Family history completed 2.5Screening (Standards 5–7)Anti D to Rh neg woman recorded** 5.6 Positive MSU follow-up recorded ** 2.0FBE recorded 2.5 Offer of aneuploidy screening recorded 41.8FBE follow-up action recorded** 0.3 11–14 week ultrasound result recorded 1.7Syphilis result recorded 5.4 Abnormal 11–14 wk ultrasound follow-up** 0.3Hepatitis B result recorded 1.7 11–14 week ultrasound not available 26.8Positive Hep B follow-up recorded ** 1.1 Fetal anatomy ultrasound recorded 12.4Hep C recorded 14.1 Abn ultrasound follow-up recorded ** 1.1Rubella recorded 2 Fetal anatomy ultrasound not available 29.1Non-immune Rubella follow-up recorded ** 4.5 Offer of gestational diabetes testing recorded 14.4HIV screen recorded 17.2 Abnormal gestational diabetes follow-up** 0.3MSU recorded 24.9 Gestational diabetes result not available 14.7Pregnancy Care (Standard 8)BP measurement at each visit recorded 11 Abdominal palpation recorded at each visit 4.0Abnormal BP follow-up** 1.1 Abnormal abdominal palpation follow-up** 0.7Symphyseal-fundal height result recorded at each visit 7.9 Fetal movements recorded at each visit 7.6Appropriate follow-up for abnormal SFH** 1.1 Neonatal drugs discussions 40.1Documentation (Standards 9-10)Unique patient identifier completed 1.7 All entries have a signature 50.3All entries dated 2.3 All signatures appear with a legible name 98.9All entries are legible 16.7



3

Measuring pregnancy care

Page | 65

women (1%) who indicated they had used recreationaldrugs showed no evidence of referral to support services.Seventeen (5%) women indicated alcohol use duringpregnancy of >2 standard drinks per day, of whom only11 (65%) were referred. Error severity: MODERATE.Standard 5a: Early pregnancy investigations. Onehundred and seventy-four (49%) records were notcompliant with this standard. In 9 women (2%), there wasno evidence of a full blood examination (FBE). Twenty-two health records indicated abnormal FBE results, one ofwhich was not referred (4% of abnormal FBE results –HIGH severity). There was no evidence of syphilisscreening in 19 women (5%) (MODERATE severity).Hepatitis B screening was not evident in 6 records (2%),(MODERATE severity). Of 8 women with positivehepatitis B serology, 4 (50%) were not referred (HIGHseverity). Assessment of Rubella immunity was not evidentin seven health records (2%) (LOW severity). In the 56women with no Rubella immunity, there was no evidenceof follow-up in 16 (29%) (LOW severity). In 61 healthrecords, there was no evidence of HIV screening (17%)(HIGH severity). In 88 health records (25%), there wasno evidence of a midstream urinalysis (MSU) (LOWseverity). Of the 17 positive MSU results, seven were notacted upon (41%) (HIGH severity).Standard 5b: Aneuploidy screening. In 148 (42%) healthrecords, there was no evidence of either the offer orperformance of Down syndrome screening. Error severity:HIGH.Standard 6: Midpregnancy ultrasound. There was noevidence that a midpregnancy (18–22 weeks) ultrasoundscan had been performed in 44 (12%) women. Twenty-nine women had a fetal abnormality recorded, of which 4(14%) were not referred appropriately. Error severity:HIGH.Standard 7: Screening for gestational diabetes. Evi-dence of the offer of testing for gestational diabetes(GDM) was absent in the records of 51 (14%) women.The result of GDM testing was abnormal in 61 women,of whom one was not acted upon (2%). The results ofGDM testing were not available for 52 (17%) women.Error severity: MODERATE.

Pregnancy careStandard 8a: Measurement of blood pressure. Bloodpressure (BP) measurement was not recorded at everyvisit in the health records of 39 (11%) women(MODERATE severity). New onset hypertension wasrecorded in the records of 21 women, of whom 4 (19%)were not acted upon (HIGH severity).Standard 8b: Assessment of fetal growth. Symphyseal-fundal height (SFH) was not measured at every visit after20 weeks in 28 (8%) women. Twenty assessments of SFHwere recorded as abnormal, of which four (20%) were notacted upon. Error severity: MODERATE.Standard 8c: Abdominal palpation. Abdominal palpa-tion at every visit after 30 weeks was not evident in thehealth records of 14 (4%) women. There were 20abnormal abdominal palpation results recorded, with two(10%) not acted upon. Error severity: LOW.Standard 8d: Fetal movements. For 27 (8%) women,there was no evidence that she was asked about fetalmovements. Error severity: LOW.Standard 8e: Neonatal vitamin K administration andHepatitis B vaccination. This was not evident in 144(40%) health records. Error severity: LOW.Standard 8f: Rhesus prophylaxis program. Thirty-sixwomen were Rh negative. In 2 (6%) of these women,there was no record of anti-D being administered. Errorseverity: HIGH.

DocumentationStandard 9a: Use of patient’s unique identifier. Six(1.7%) health records did not have a unique patientidentifier on every page. Error severity: LOW.Standard 9b: Signing of medical record entries. In eight(2%) health records, there were entries that were notdated. There were missing signatures in the records of 178women (50%) and missing legible name in those of 350(99%). In 59 health records (17%), not all entries in therecord were legible. Error severity: MODERATE.Standard 10: Schedule of visits. There was poorcompliance with the recommended schedule of visits,particularly in the third trimester when a large number ofhealthy women with a normal pregnancy had anapparently excessive number of visits.

DiscussionIn this study, we have assessed the feasibility of using abespoke clinical audit tool to assess pregnancy care, asdefined by compliance with hospital policies, as anadjuvant approach to improving safety and quality ofhealthcare. We found that deficiencies in care were evidentin 85% of case-records but that deficiencies that couldhave significantly compromised pregnancy outcomes wereevident in less than 2%. We believe that this approachcould be readily embedded in maternity services to beused with established measurement of perinatal outcomesto guide improvements.

Figure 1 Gestation at first antenatal visit.



S. V. Sinni et al.

Page | 66

While patient safety is paramount in health caredelivery, there remains little agreement on how best tomeasure it.7–10,19,20 Many diverse approaches and toolshave been reported.21–23 However, we were unable toidentify any that assessed the continuum of pregnancycare. Accordingly, we developed the tool to audit careagainst a number of standards.18 That the rate of a‘high severity’ avoidable errors was less than 2%suggests that overall care was of a reasonable standard.However, a reasonable or acceptable error rate was notagreed upon prior to the study and there is not broadagreement on this in the literature. Reported medicalerror rates range from 2 to 40%, depending onassociated patient risk factors and co-morbidities.8,20,24,25

Reports about patient safety in an ambulatory settinghave been published,26–29 although none relate tomaternity care, and analyses have been limited to onlythose records where adverse events were known to haveoccurred. Thus, as far as we are aware, this is the firstattempt to measure the rate of care process complianceas an outcome in itself.Having established a baseline error rate (85%) and a

critical error rate (2%) for our service, we believe that thenext step is to determine acceptable error rates. We wouldsuggest that the acceptable error rate will differ for severeerrors – likely to be a zero – and minor-to-moderate errors.We are now engaged with the service to set these rates.The high rates of unavailable data (up to 85%) from the

health records should be a major concern for bothindividual clinicians and the organisation. Documentationof care is fundamental to the transfer of clinicalinformation between care providers, particularly whenlarge components of care, such as in pregnancy, may beprovided in the community by non-hospital staff. In theseinstances, access to hospital information technologysystems can be inconsistent or non-existent. Effectivereporting relies on transcription from files retained by themedical practitioner to a woman’s handheld record, whichultimately is filed in the health record. There is also theissue of woman’s choice of pathology provider andmedical provider preference of pathology services. In caseswhere there is no interface between the pathology providerand hospital information systems, data collection againrelies on manual transcription. This study identifiedcompelling system failures that are likely to be replicatedin other maternity services but that could be readilyovercome with integrated information managementsystems such as afforded by an electronic medical record.Indeed, an electronic medical record that was configuredto prompt the user to complete the summary of careprovision, in effect to manage care, would be expected todramatically improve compliance, to improve the flow ofinformation between clinicians and to improve quality ofcare. One outcome from this study is to highlight for thehealth service the potential value of an integratedelectronic maternity record.In addition to highlighting critical areas of risk exposure,

the audit tool identified opportunities to improve

organisational systems of care and deliver efficiencies. Forexample, the majority of women received their firstantenatal visit after 13 weeks gestation, principally due tolack of sufficient appointments. However, fifty womenwho were assessed as being at low risk of developingcomplications were over-serviced by having too manyantenatal visits. For example, one woman was seen 16times after 28 weeks of pregnancy without any apparentreason or pregnancy complication. These insightsprovided opportunities for the health service to improvescheduling and thereby free up appointment times forbetter use in early pregnancy. These changes would beexpected to save costs and improve care. Our toolprovides a means to assess whether improvements andcompliance are then achieved.Our interpretation that omissions in the health record

reflect omissions of care may be perceived as extreme. Weacknowledge that omissions in documentation do not provethat care was not provided. However, it is difficult to arguethat care was offered or provided if there is no documentedevidence for it. For example, in the case of aneuploidyscreening, it would be difficult to defend care where awoman gives birth to a child with Down syndrome andclaims wrongful birth stating that she was never offeredtesting and would have terminated the pregnancy had sheknown, if there was no record that counselling had takenplace. Similarly, the lack of evidence of HIV screening or ofappropriate referral for abnormal hepatitis serology or anabnormal fetal anatomy ultrasound are of great concern.Highlighting these deficiencies enabled changes to resultsmanagement to improve patient care.

LimitationsThe key limitation in our work is that to date, this exercisehas been necessarily limited to measuring process. Whethermeasurement and reporting of the results will lead tochange and ultimately to improved process and betteroutcomes has yet to be demonstrated. Ideally, such future‘intervention’ studies should be designed and reportfindings in accord with the standards for qualityimprovement reporting excellence (SQUIRE) guidelines.30

Further, the study was undertaken within a single hospitaland the results may not be generalisable beyond the siteunder review. In particular, the grading of error severity islikely to vary depending on an individual hospitalguidelines. For example, if routine screening for gestationaldiabetes is not required in a given hospital, then the lack ofprovision of such screening would not be recorded as anerror as it would be in our service. The development ofregional or national guidelines and standards, such as thosepublished by NHMRC12 or NICE,13,14 would assist in thedevelopment of uniform error severity scoring.

ConclusionsWe believe that this study usefully contributes to theknowledge of measuring safety in pregnancy care by



5

Measuring pregnancy care

Page | 67

providing a broad assessment of the entirety of antenatalcare, identifying important opportunities for careimprovement. Most importantly, we have shown that it isboth feasible and insightful to measure process of care inaddition to outcomes of care as an adjuvant approach toimproving quality and safety.

AcknowledgementsThe authors wish to acknowledge the support andengagement of Health Service where the research wasbased and the Victorian Government’s Department ofHealth and Operational Infrastructure Support Programfor financial support.

Competing interestsThe authors declare that they have no competing interests.

References1 Royal College of Obstetricians and Gynaecologists. Safer

Childbirth: MINIMUM Standards for the Organization andDelivery of Care in Labour. London, UK: RCOG Press, 2007.

2 Commonwealth Government of Australia. ImprovingMaternity Services in Australia. The report of the maternityservices review. Commonwealth of Australia, Barton ACT2009. (Publications No. P3-4946).

3 Knight M, Kenyon S, Brocklehurst P et al. (eds) on behalf ofMBRRACE-UK. Saving Lives, Improving Mother’s Care –Lessons learned to inform future maternity care from the UKand Ireland Confidential Enquiries into Maternal Deaths andMorbidity 2009-12. Oxford: National Perinatal EpidemiologyUnit, University of Oxford 2014.

4 Li Z, Zeki R, Hilder L, Sullivan E. Australia’s mothers andbabies 2011. Perinatal Statistic Series Number 28. Cat. No.PER 59. Canberra: AIHW National Perinatal Epidemiologyand Statistics Unit.

5 Consultative Council on Obstetric and Paediatric Mortalityand Morbidity. 2010/2011 Victoria’s Mothers and Babies.Victoria’s Maternal, Perinatal, Child and Adolescent Mortality.CCOPMM, Melbourne 2014.

6 Department of Health and Human Services. VictorianPerinatal Services Performance Indicators 2012–13.Melbourne: Victorian Government, 2015.

7 Pronovost P, Goeschel C. Viewing health care delivery asscience: challenges, benefits, and policy implications. HealthServ Res 2010; 45: 1508–1522.

8 Pronovost PJ, Colantuoni E. Measuring preventable harm:helping science keep pace with policy. JAMA 2009; 301:1273–1275.

9 Petrou S, Kupek E, Vause S, Maresh M. Antenatal visits andadverse perinatal outcomes: results from a British population-based study. Eur J Obstet Gynecol Reprod Biol 2003; 106: 40–49.

10 Barros A, Ronsmans C, Axelson H et al. Equity in maternal,newborn and child health interventions in Countdown to2015: a retrospective review of survey data from 54 countries.Lancet 2012; 379: 1225–1233.

11 Goldenberg R, McClure E. Disparities in interventions forchild and maternal mortality. Lancet 2012; 379: 1178–1180.

12 Australian Health Ministers’ Advisory Council. Clinical PracticeGuidelines: Antenatal Care – Module 1. Australian GovernmentDepartment of Health and Ageing, Canberra 2012. Availablefrom http://www.health.gov.au/antenatal (accessed 20/8/15).

13 National Collaborating Centre for Women’s and Children’sHealth. Antenatal Care: Routine Care for Healthy PregnantWomen. London, UK: RCOG Press, 2009.

14 National Institute for Health and Care Excellence. AntenatalCare. NICE quality standard 22. Guidance.nice.org.uk/qs22(accessed 17/11/15).

15 Nursing and Midwifery Board of Australia. Nationalcompetency standards for the midwife. Available from: http://http://www.nursingmidwiferyboard.gov.au/Codes-Guidelines-Statements/Professional-standards.aspx (accessed 20/8/15).

16 Royal Australian and New Zealand College of Obstetriciansand Gynaecologists. College Statements: Obstetrics. Availablefrom: http://www.ranzcog.edu.au/college-statements-guidelines.html#obstetrics (Accessed 20/8/15).

17 Monash Health. Pregnancy tests, investigations and key visitinformation. Clinical Guideline. Available from http://system.prompt.org.au/download/document.aspx?id=3205821&code=34D92259041C9CB2E5A483A97E9BB586 (Accessed 20/8/15).

18 Sinni S, Cross W, Wallace E. Designing a clinical audit toolto measure processes of pregnancy care. Nurs Res and Rev2011; 1: 15–25.

19 Sinni S, Wallace E, Cross W. Patient safety: a literaturereview to inform an evaluation of a maternity service.Midwifery 2011; 27: 274–278.

20 Brennan T, Leape L, Laird N et al. Incidence of adverseevents and negligence in hospitalized patients: results of theHarvard Medical Practice Study I. Qual Saf Health Care2004; 13: 145–151.

21 Victorian Department of Health. Clinical Risk Management.Available from http://health.vic.gov.au/clinrisk/ (Accessed 20/8/15).

22 Institute for Healthcare Improvement. Tools. Available fromhttp://www.ihi.org/resources/Pages/Tools/default.aspx(Accessed 20/8/15).

23 Graham WJ. Criterion-based clinical audit in obstetrics,bridging the quality gap? Best Pract Res Clin Obstet Gynaecol2009; 23: 375–388.

24 Weingart S, Wilson R, Gibberd R, Harrison B. Epidemiologyof medical error. BMJ 2000; 320: 774–777.

25 de Vries E, Ramrattan M, Smorenburg S et al. The incidenceand nature of in-hospital adverse events: a systematic review.Qual Saf Health Care 2008; 17: 216–223.

26 Hammons T, Piland N, Small S et al. Ambulatory patientsafety what we know and need to know. J Ambul Care Manage2003; 26: 63–82.

27 Gandhi T, Seger A, Overhage J et al. Outpatient adverse drugevents identified by screening electronic health records. JPatient Saf 2010; 6: 91–96.

28 Schnall R, Bakken S. Reporting of hazards and near-misses in theambulatory care setting. J Nurs Care Qual 2011; 26: 328–334.

29 Poon E, Gandhi T, Sequist T et al. “I wish I had seen thistest result earlier!”: dissatisfaction with test result managementsystems in primary care. Arch Int Med 2004; 164: 2223–2228.

30 Ogrinc G, Mooney SE, Estrada C et al. The SQUIRE(Standards for Quality Improvement Reporting Excellence)guidelines for quality improvement: explanation and elaboration.Qual Saf Health Care 2008; 17(Suppl 1): i13–i32.



S. V. Sinni et al.

Page | 68

CHAPTER FIVE

Perinatal staff perceptions of safety and quality in their service

Page | 69

5.1 Preamble

In Chapter Five I describe the qualitative component of my research. In the quest to

improve healthcare safety scholars have increasing looked to team behaviours and their

effect on patient safety. I have done likewise. I generated qualitative data from in-depth

interviews with staff about their perceptions of the safety and quality of their service,

exploring themes derived from the literature about team behaviour and patient safety. I was

surprised to find traditional dichotomies and professional conflicts existing between the

disciplines involved in maternity care in such a new service. Reassuringly, this was not a

dominant theme. Nonetheless, it led me to further exploration of the abundant literature on

tensions between providers of maternity care.

Page | 70

5.2 Perinatal staff perceptions of safety and quality in their service

RESEARCH ARTICLE Open Access

Perinatal staff perceptions of safety and quality intheir serviceSuzanne V Sinni1*, Euan M Wallace1 and Wendy M Cross2

Abstract

Background: Ensuring safe and appropriate service delivery is central to a high quality maternity service. With thisin mind, over recent years much attention has been given to the development of evidence-based clinicalguidelines, staff education and risk reporting systems. Less attention has been given to assessing staff perceptions of aservice’s safety and quality and what factors may influence that. In this study we set out to assess staff perceptions ofsafety and quality of a maternity service and to explore potential influences on service safety.

Methods: The study was undertaken within a new low risk metropolitan maternity service in Victoria, Australia with astaffing profile comprising midwives (including students), neonatal nurses, specialist obstetricians, junior medical staffand clerical staff. In depth open-ended interviews using a semi-structured questionnaire were conducted with 23 staffinvolved in the delivery of perinatal care, including doctors, midwives, nurses, nursing and midwifery students, andclerical staff. Data were analyzed using naturalistic interpretive inquiry to identify emergent themes.

Results: Staff unanimously reported that there were robust systems and processes in place to maintain safety andquality. Three major themes were apparent: (1) clinical governance, (2) dominance of midwives, (3) inter-professionalrelationships. Overall, there was a strong sense that, at least in this midwifery-led service, midwives had the greatestopportunity to be an influence, both positively and negatively, on the safe delivery of perinatal care. The importance ofunderstanding team dynamics, particularly mutual respect, trust and staff cohesion, were identified as key issues forpotential future service improvement.

Conclusions: Senior staff, particularly midwives and neonatal nurses, play central roles in shaping team behaviors andattitudes that may affect the safety and quality of service delivery. We suggest that strategies targeting senior staff toenhance their performance in their roles, particularly in the training and teamwork role-modeling of the transitoryjunior workforce, are important for the development and maintenance of a high quality and safe maternity service.

Keywords: Maternity care, Clinical governance, Midwifery dominance, Safety, Qualitative research

BackgroundDelivering safe, effective and appropriate care is clearlythe aim of all maternity services. Over recent decades,with the aim of improving the quality and safety ofdelivered care, much attention has been paid to thedevelopment of evidence-based guidelines, staff educa-tion, risk management and reporting systems. However,perhaps less attention has been afforded to assessingand influencing staff attitudes and behaviours. This isimportant because patient safety is strongly influenced

by the attitudes and behaviours of the clinicians deliveringthe care [1]. Teams displaying high levels of mutual respect,trust and cohesion are significantly more likely to consist-ently provide high quality care with lower rates of adverseevents [2,3]. Outside of healthcare, private industry acrossdiverse sectors, including manufacturing, finance, retail andhospitality, have long applied an awareness of the influenceof team dynamics on productivity and safety to buildcapability and enhance outcomes [4-10].In healthcare while the past few decades have seen

increasing attention to safety and quality innovation mucheffort remain focussed on measuring health outcomes them-selves [11-15]. While health outcome data are certainlyimportant, and readily measurable, outcomes on their own

* Correspondence: [email protected] Ritchie Centre, Department of Obstetrics and Gynaecology, MonashUniversity and MIMR-PHI Institute of Medical Research, 246 Clayton Road,Clayton, Victoria 3168, AustraliaFull list of author information is available at the end of the article

© 2014 Sinni et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.

Sinni et al. BMC Health Services Research 2014, 14:591http://www.biomedcentral.com/1472-6963/14/591

Page | 71

do not provide information about the factors that mayhinder optimal care delivery and thereby compromiseoutcomes [11,12]. In short, they do not provide insightsinto the processes that might underlie poor outcomes butrather simply identify that poor outcomes have occurred.Such an outcome-focused approach cannot directly informwhat changes are required to effect improved care andbetter outcomes. Indeed, despite all the efforts aroundpatient safety over decades of study overall rates of medicalerror have essentially remained constant [16-18].However, more recently, clinical team behaviour has

been investigated as a possible contributor to compro-mised patient safety and a focus on improving teamdynamics developed as part of the pathway to improvedoutcomes [1-3]. It has been estimated that ineffectiveteam communication is an important contributory factorin about 8 out of 10 adverse events [19]. Of all health-care, there is a risk that the provision of maternity care,in particular, is structured in a manner that may under-mine effective team behaviour and thereby impair out-comes. Specifically, maternity care involves two separateprofessional groups – midwives and obstetricians – that,at least in some countries such as Australia, NewZealand and the UK, are both independent and able toexert differing ideological practices, often for the sameindividual patient [20-28]. For example, in general, mid-wives are trained as skilled birth attendants to facilitate anormal physiological process whereas obstetricians aremedically trained to expedite a condition fraught with riskrequiring high levels of intervention [20-28]. We won-dered whether the issues and potential tensions that thesemight bring to the dynamics of the attendant clinical teammight impact upon safety. Accordingly, we set out toexamine staff perceptions of safety and quality usingestablished qualitative methodology, exploring issues ofinfluence on team behaviours. Specifically, we aimed toassess how the workforce perceived the level of safety oftheir service, to explore perceived interactions betweenprofessional groups, and to identify issues that mightundermine team performance and thereby put safety atrisk.

MethodsSettingThe study was conducted within a new maternity serviceestablished for women with a low risk pregnancy in arecently built hospital. The service was a planned expan-sion of an existing maternity program that already pro-vided secondary and tertiary maternity services at thetwo other established hospitals within the organization(Monash Health). At the time of the study there were1000 births per annum in the low risk service that wasthe focus of the study. The service offered two models ofcare: midwifery-led or shared antenatal care between

midwives and community based General Practitioners(GPs). In both models of care, intrapartum care wasprimarily provided by midwives with the support ofon-site junior resident medical officers and on-call spe-cialist (consultant) obstetricians. All staff were expectedto follow existing organizational protocols relating themanagement of pregnancy and labour with clear androbust criteria for consultation and referral to medicalcare when deviations from normal were present.

ParticipantsAll hospital-based staff involved in perinatal care and/orits administration were invited to participate. Invitationsto participate in the study were extended at a regularlyscheduled work meeting then distributed via electronicmail. Twenty-three staff were interviewed, comprising aself-selected but representative cross section of theworkforce. Table 1 summarizes the participants and theirlevels of experience. Data saturation, as evidenced by nonew emerging themes, was achieved after 15 interviews.However, all who agreed to participate were offered aninterview and data were extracted from all 23 interviewtranscripts. No participant withdrew.

Data collectionWe used naturalistic interpretive design to inform adeep understanding of the subject matter [29]. This ap-proach was applicable because two of the researchers(SVS, EMW) were involved in the planning and imple-mentation of the maternity service and were responsiblefor overseeing quality and safety of the service and sowere known to stakeholders. In the first phase of thestudy, staff focus groups [30] were established to engagethe staff and to appraise them of the purpose of thestudy. Questions from a validated patient safety culturesurvey [31] were modified to form the basis of a semi-structured questionnaire that was then used to gener-ate open-ended discussions during in-depth interviewsabout the safety and quality of the service (Table 2).Semi-structured interviews included funnelling by begin-ning with open ended questions and drilling down to points

Table 1 The role and numbers of participantsStaff role Number interviewed

Midwives 10 (8 senior midwives*, 1 junior midwife,1 postgraduate midwifery student)

Obstetric staff 6 (5 consultant obstetricians, 1 junior medical officer)

Nurses 4 (2 senior neonatal nurses*, 2 nurse coordinators)

Paediatricians 2 (1 consultant paediatrician, 1 in a paediatrictraining program)

Clerical staff 1

*Senior midwives and neonatal nurses include managers, associate managersand clinical midwife/nurse specialists.

Sinni et al. BMC Health Services Research 2014, 14:591 Page 2 of 9http://www.biomedcentral.com/1472-6963/14/591

Page | 72

of interest, story telling to encourage the interviewee toelaborate by recounting their personal experiences, probingwhere the interviewer sought clarification, and paraphrasingby repeating back what the interviewee said to ensure com-mon understanding to prompt further discussion [30]. Theinterview protocol was designed to maximize consistencyincluding details about the interview such as time, place andperson being interviewed, instructions for the interviewerand questions that have a logical sequence with ampleopportunity for the interviewee to elaborate [32].All interviews were conducted by the same researcher

(SVS), mostly during scheduled work hours. Two specialistmedical staff were interviewed outside of work hours toavoid interruptions. Participants were asked to read thequestions immediately before the interview and seekclarification where necessary. During each interview, theprocess of participant checks was employed. At appropriatetimes, the researcher’s understanding of the points made bythe participants were rephrased by the researcher and theresearcher’s understanding verified by the participant by hisor her confirmation of the interpretation either verbally(“that’s right”) or non-verbally (“nodding”). Research biaswas managed through the use of prolonged engagement[29]. This method involved the establishment of the

interview and allowing the participants sufficient time todiscuss all the pertinent issues from their own perspectivewith the researcher. There were no time constraints in-volved with the interview. Themes that arose from the in-terviews were able to be explored at the time of interviewand were also able to be explored with other participants insubsequent interviews. Interviews lasted between 20 and90 minutes, most lasting 60 minutes. Discussions wereaudio recorded and transcribed by an independent source.Transcripts were checked for accuracy by the interviewerand de-identified then forwarded to other members of theresearch team.

Peer debriefingThe method of peer debriefing involved frequent discus-sions of the research at length with colleagues not in-volved in the research with methodological expertise.They offered support and guidance about the methodologyand subject matter. In addition the researcher enlistedanother person to conduct an audit of the transcriptions.Through external auditing of the tapes the accuracy ofdocumenting the transcription was verified and the correctmeaning of the data was derived.

Data analysisQualitative data were explored using NVivo9® software(QSR International Pty Ltd, Melbourne, Victoria, Australia).Thematic analysis was used to code the data and identifyemergent themes. Data were initially coded into 54 categor-ies, with 1–124 entries in each category (Table 3). Thesewere recoded and distilled [32,33] yielding three emergentthemes. Data analysis and interpretation was checked inde-pendently by the other two authors. Multiple strategieswere used to maximize trustworthiness of interpretationsincluding triangulation, member checking and peer debrief-ing, providing rich, thick descriptions of the data [32]. Theindependence of one of the three-member research

Table 2 Interview Questions1 What are your experiences of the safety and quality of the maternity

service?

2 Does evidence inform clinical practice decisions at the maternity service?

3 What happens when a near miss or adverse event occurs? Considerwhether you feel supported in reporting such an event, whether youbelieve your report will be followed up and how lessons are shared.

4 To what extent does mutual respect exist among professional groups(clerical staff, midwifery, neonatal nursing, obstetrics, paediatrics,anaesthetics, including trainees/students)?

5 What is your experience of the cohesion or otherwise among staffduring an obstetric or paediatric emergency?

Table 3 Codes and emergent themes from the dataInitial codes Clusters Themes

Acceptance, Accountability, Audit, Balance, Barriers, Blame,Boundaries, Capabilities, Caution, Clinical judgment, ClinicalReview, Collaboration, Communication, Confidence, Consistency,Culture, Evidence, Expectations, Experience, Fear, Goals to aspireto, Governance, History, Intimidation, Judgmental, Lack of support,Leadership, Maturity, Midwifery versus medical (non-interventionversus intervention), Organizational structures, Ownership,Perceptions, Personalities, Power, Priorities, Recognizing risk,Relationships between staff, Reporting adverse events, Resistance,Respect, Rules, Safety, Silos, Size, Staffing, Supervision, Support,Systems, Tension, Training, Trust, Women’s choice, Work-force,Work-load.

Accountability, Audit, Balance, Capabilities,Clinical Review, Culture, Goals to aspire to,Governance, History, Leadership, Organizationalstructures, Safety, Size, Support, Systems, Training

Clinical governance

Acceptance, Blame Caution, Judgmental, Maturity,Respect, Trust

Acceptance, Trust,Respect

Collaboration, Communication, Fear, Perceptions,Personalities, Relationships between staff, Tension

Relationships

Barriers, Boundaries, Confidence, Experience,Intimidation, Lack of support, Midwifery versusmedical (non-intervention versus intervention),Ownership, Power, Resistance, Silos, Women’s choice

Midwifery dominance,power, protection

Clinical judgment, Consistency, Evidence, Expectations,Priorities, Recognizing risk, Reporting adverse events,Rules, Staffing, Supervision, Work-force, Work-load

Rules


Page | 73

team (WC), with qualitative research expertise, minimizedresearcher bias, although this is an inherent attribute ofnaturalistic interpretive inquiry.

TriangulationData triangulation comprises the use of many datasources with matching foci to obtain different viewsabout a topic [34]. Data may be gathered at differenttimes, at different sites or from different people [34]. Inthis study data triangulation was undertaken betweeninterviewees and themes were cross-referenced from oneto another. In this way themes that were unique to asingle participant as well as themes that were common toall participants were identified. These emergent themeswere synthesized and the process was repeated until therewas minimal overlap in the remaining categories.

Ethics considerationsApproval to undertake this study was granted by SouthernHealth Human Research Ethics Committee (HREC) andMonash University HREC. Informed voluntary consentwas obtained from each participant.

ResultsThree principle themes arose from analysis: clinicalgovernance, dominance of midwives, and inter-professionalrelationships. Two subthemes were also evident: rules(within the clinical governance theme) and acceptance,trust and respect (within the inter-professional relationshipstheme).

Clinical governanceThe existence and importance of a robust and visibleclinical governance framework was identified as a dominanttheme. Specifically, participants were able to describe theinstitution’s numerous mechanisms, processes and systemsat the local, site and organizational level to improve patientsafety and minimize patient harm. These included the exist-ence and widespread use of evidence-based clinical practiceguidelines, the roles and activity of a local PracticeImprovement Committee (a clinical governance over-sight committee), and the routine use of regular andformal practice review. Senior staff demonstrated a highlevel of understanding of these various components of theclinical governance framework, while junior staff demon-strated that they had an awareness of the existence of thevarious activities even if they could not describe the com-ponents in detail. All staff were confident that breaches ofsafety would be acted upon swiftly and appropriately withthe focus being on improving systems rather than taking apunitive approach.Example and illustrative staff quotes are detailed below.

specialist obstetrician: “… our safety is alwaysparamount … because we are low risk … we alwaysmake sure that … we’re very careful to practice within… those guidelines … so that safety of the womencomes first, … there is a work-force awareness of im-portance of safety and quality … there is engagementin a safety and quality framework”.nurse administrator: “an incredibly proactive team,very open to learning, and trusting that mistakes willbe learned from, we’ll learn from them, so always,always reported, always explored and investigatedwell locally, reported through the organizationalchannels through the quality coordinator and theDirectors of Nursing and also via the site, throughthe PIC (practice improvement committee) forinterdisciplinary conversation, where they reallyexplore and try to tease out what the issues are, tounderstand because it’s usually always about systems.It’s very rarely about individuals and reallyunderstanding you know where the incident hascome from and where the changes need to be made”.

RulesParticipant adherence to organizational policies and pro-cedures was strongly apparent. Midwives were perceivedto have good knowledge of organizational proceduresand to encourage adherence, which participants agreedenhanced the safety of the service. This is exemplified bythe following quote from a specialist obstetrician “…midwives here … they really follow the protocol”.In contrast, while obstetricians acknowledged the exist-

ence of organizational policies and procedures, and wereaware of their content, there was a perception amongsenior medical staff that professional opinion was some-times preferable:

specialist obstetrician: “… guidelines and protocolsmay not strictly apply to each and every situationbecause clinical scenarios can be quite different attimes so you have to use your own judgment”.

Dominance of midwivesParticipants reflected on the strong influence that mid-wives had in how maternity care was provided. Theycommented on the dominant role of the midwife, par-ticularly relative to junior medical staff, and the effect onthe balance of power among members of the clinicalteam on any given shift. While midwifery dominancewas perceived to be generally beneficial in ensuringadherence to organizational policies and procedures,concern was also expressed that the dominant role ofthe midwife may, at times, adversely affect the way somemidwives interacted with the junior medical workforce.In particular, there was concern that, at times, midwifery


Page | 74

dominance led to the exclusion of junior staff, both med-ical and midwifery, that limited the clinical experience andlearning opportunities of the junior staff. Midwifery domin-ance was also reported to influence clinical care deliveryand intervention or non-intervention. For example, severalrespondents made reference to a midwifery preferenceagainst the use of epidural regional anaesthesia for painmanagement in labour that, due to perceived dominance ofsenior midwives over junior medical staff, prevailed, oftenirrespective of patient preference. This perception is sup-ported by the observation that, of the three Monash Healthmaternity services, this service has the lowest epiduraluptake rate [35].

specialist obstetrician: “the reason the midwifery andthe medical workforce get on well is that actually theyboth want the same thing; the midwifery workforcewants to get on, do their work in an independentmanner, in a stand alone manner, and only call forobstetric assistance if it’s required. And they makethat judgment; and the medical staff … actually don’twant to be involved in the service. They want to beinvolved as little as possible because they’ve got busyprivate practices and are happy to come in to do theodd you know lift out forceps or caesarean section”.

senior midwife: “students and residents (juniormedical officers) feel that the midwives are far toobossy in their approach towards them”.

Participants described dichotomies between non-interventionist midwifery paradigms and medical in-terventions deemed unnecessary that might impact onprocesses of labour and birth. A number of cases wereprovided where midwives denied women requesting epi-dural analgesia in early established labour, as examplesthat some midwives worked according to their personalviews of promoting normal non-interventionist labourand birth rather than what an individual woman mightchoose. Forty eight percent of interviewees commentedon the behaviour of a few midwives, during the initialmonths of operation, of managing the care of labouringwomen in isolation and according to their personal mid-wifery practice principles rather than as per organizationalprocedures. In extreme cases, medical staff had beenrefused access to women by midwives perceived to beprotecting women from unnecessary intervention, such asroutine vaginal examination. References were made tonulliparous women having prolonged second stage oflabour without appropriate assessment of fetal well-beingor the need to expedite birth, the assumption being thatbirth is a natural process that should occur without inter-vention. The extent to which any individual woman mighthave been complicit with such behaviour from a midwife

could not be explored by the study design and so was notinvestigated as part of this study. That almost half the par-ticipants made reference to these past practices suggestslingering negative perceptions about a midwifery practicethat had supported an individual midwife’s beliefs ratherthan support women’s choice inclusive of all members ofthe healthcare team. These actions were perceived as con-tributing to potential breaches of safety, specifically wherethere was inadequate fetal monitoring or assessment ofneed for assisted delivery in women with a prolongedsecond stage of labour. Participants who made referenceto such behaviours also pointed out that organizationalstrategies, such as refinement and reinforcement of prac-tice protocols, had been implemented to ensure practiceconformity and had resulted in midwives either modifyingtheir behaviour or leaving the service. This outcome wasperceived as improving relationships between remainingsenior members of the clinical team.

midwife in charge of shift (associate midwiferymanager): “In the early days, the midwives were veryprotective of their women and we … weren’t allowedin those rooms by those midwives, whereas these daysnow I know exactly what’s going on in all thoserooms” and “there were midwives who were verypossessive of these birthing women and werereluctant to let people in”.

specialist obstetrician: “in the early days, that’s wherewe had the midwives who used to want to oversteptheir boundaries and didn’t want to talk to themanager you know, but these days now we don’t haveany problem with our midwifery staff, not at all”.

Nonetheless, overall midwives were seen, both bythemselves and by senior and junior medical staff, asthe principle professional group making care manage-ment plan decisions, as illustrated by the followingquotes:

junior medical officer: “It is a midwife led unit sothey, they are technically running this place, but itwas unlike any other departments that I’ve workedin where we don’t have to ask the nurses, you know,any permission. Just the patient’s permission to goand speak to the patient, get a history, do anexamination and everything. But here it felt like itwas the other way around; where we must askpermission of, either the nurse in charge or themidwife to actually go and speak to the patient and Ifound that difficult, being the only doctor on thefloor with a responsibility for the welfare of all thepatients if something goes wrong ..… often they’revery protective of their women”.


Page | 75

Inter-professional relationshipsThere was evidence that, at times, relationships betweenstaff could be tense and could potentially impact on servicedelivery. A senior midwife reflected on the importance thatrelationships has on service delivery. “… you obviously havepersonality conflicts so you just have to get on with that, tostep in where you need to …” and “…we’re very clear thatthough it is a midwifery led model we have certain guide-lines that we have to follow, and we have to involve ourconsultants and our doctors and we get medical studentscoming through now you know, so we have to reallyinvolve them in the woman’s care as well, it’s not just mid-wives you know. So there’s not that same boundary pushanymore as there, in fact we don’t have any of it anymore”.Similarly, an obstetrician noted the importance of rela-

tionships and interdependence between obstetriciansand midwives in maternity care: “you can’t do midwiferyon your own, nor can you do private obstetrics on yourown; you have to have midwife support”.

Acceptance, trust, respectParticipants believed that, overall, there were goodinter-professional relationships between the varioushealth professions. A specialist obstetrician recountedthe interdependence of midwifery and obstetrics andthe need for mutual respect to run a safe and efficientservice.

specialist obstetrician: “I have a lot of good things tosay as far as mutual respect is concerned. Being asmaller place everyone’s quite friendly and they makeyou feel very comfortable to be with and then I think,we do respect each other for what we are, for eachother’s opinions and yes there isn’t anything negative Ican think. I would say everyone’s respectful”.

A quote from a senior midwife corroborates this percep-tion: “they’re really good the obstetricians here, they willlisten to what you say if you are disputing what they’re say-ing. But it’s not often, you know. They’re the ones callingthe shots too, and so if you’ve given them the clinical infor-mation, they’re acting on what you’re saying because theyrespect you, they have respect for your knowledge aboutwhat’s meant to happen, about when you need them”.

However, among junior staff there was a sense thattrust and respect needed to be earned. Two junior staffreported having to earn trust leading to an improvedworking relationship with midwifery staff.

“… at the very beginning to feel accepted wasimpossible, it was quite a challenge, emotionallydraining and … they needed to gain trust in me aswell”. This was corroborated by a specialist

obstetrician in this quote: “midwifery staff … youknow …. you really have to earn your stripes withthem”.

DiscussionIn this study we set out to assess staff perceptions of safetyand quality in a new low risk maternity service and toexplore staff knowledge and involvement in clinical gov-ernance, including professional relationships within andbetween the different craft groups. While, reassuringly,staff reported that their maternity service was one of safetyand quality, their perceptions identified opportunities forservice improvement that may have broader applicabilityto other maternity services. Specifically, three key themesemerged – issues related to staff acceptance and involve-ment in clinical governance, including the uptake of andadherence to uniform policies and procedures, the domin-ance and influence of midwives within the service, andinter-professional relationships, particularly between se-nior and junior staff. These themes are consistent with theexisting literature on the effect of team behaviours onpatient safety [1-4,36,37] and have clear implications for thedevelopment and maintenance of a high quality and safematernity service. Of course, our observations of the rela-tionships between professional groups and the dominanceof the midwifery workforce are really only of relevance tohealthcare systems that have midwifery as an independentprofession such as in Australia, New Zealand, and theUnited Kingdom.Variation in clinical practice that is not justified is known

to threaten patient safety [37] such that high performinghealthcare systems seek to standardize clinical practiceusing evidence, where available, to guide that practice. Thisis the rationale behind the development and implementa-tion of evidence-based clinical guidelines, or policies andprocedures. However, the effectiveness of policies and pro-cedures is dependent on clinician uptake [37], which inturn is influenced by clinician ego [37], and on a robustframework to monitor uptake and effect change. Indeed,one of the most common disruptive behaviours in the clin-ical workforce relates to non-compliance with organizationalprotocols [2,14,18,37]. In this regard, it is well recognizedthat policies and procedures cannot account for all clinicalscenarios and that expert clinical judgment remains animportant “overlay” to ensure that, where appropriate, careis individualized, even if this is apparently outside of the rele-vant policy [37]. In this study we found that, in the main,midwifery staff adhered to organizational policies and proce-dures more closely than medical staff, particularly seniormedical staff. Indeed, only senior medical staff voiced astrong desire for individual clinician freedom. Unfortunately,we were unable to assess whether lack of adherence to pro-tocols by senior medical staff was more often appropriateor inappropriate but we suggest that, at least in this


Page | 76

setting, protocol violation is more likely to be a feature ofobstetricians rather than midwives. This observation hasrelevance to effective implementation strategies for futureprotocols. Certainly, an audit of protocol adherence withmaternity services would be a worthwhile routine qualitymeasure, recording adherence by professional group andseniority.That said, we also uncovered a concern among staff,

both medical and midwifery, that when the service firstopened some midwives did not follow organizationalprotocols. There was a perception that if the attendingmidwife was to be the woman’s guardian and advocatethen she would need to breach hospital policy and thatthis was acceptable [24-27]. Participants reported that, atthat time, the hospital identified this behaviour andimplemented strategies to improve protocol compliance,and thereby build a perception of safety. Overall, thiswas seen as a positive response by the hospital. There isa clear challenge here for the hospital and its staff. It isrecognized that there is a risk that rigid hospital policiescan disempower women to the point of oppression [27],resulting in a service that is not “woman-centered”. Yetthe desire for uniform agreed evidence-based practice isunderstandable. In this study, the staff reported that theenforcement of policy adherence led to the departure ofmidwifery staff who felt that they could no longerpractice the type of midwifery that they thought theyshould be practicing [25-27]. For example, hospitalpolicy required regular maternal and fetal assessmentin labour, in accord with accepted clinical practiceguidelines [3,38,39]. Some midwives believed that itwas their role to protect women from such “unneces-sary” intervention with some even viewing 4–6 hourlyvaginal assessments in labour as assault. Study participantsreported that these non-adherent behaviours promptedhospital interventions, including strict monitoring of adher-ence to protocol that in turn led some midwives to leavethe hospital’s employment. Whether this enhanced ordiminished service quality could not be assessed by ourstudy but would be an important consideration for futurework.Nearly three quarters of adverse events in maternity

care are related to failure in team processes and commu-nication [2,19]. This emphasizes the importance of ahighly functioning team as the foundation of a safe andeffective service. The influence of teamwork on perform-ance was demonstrated in a study of emergency responseclinical teams [36] where the complexities in healthcarewas shown to lead to team fragmentation, and therebyreduce team efficiency and productivity, and increasedisharmony. To counter this, “artifacts” such as uniforms,equipment and tools to share information, have beensuggested by some to be useful at creating cohesion betweenclinical team members [36]. Indeed, strong organizational

hierarchy can be created with just uniforms, and how theywere worn, with resulting impact on team dynamics, bothpositive and negative [36]. While traditional hierarchies werenot apparent in our findings, separate midwifery and obstet-ric stereotypes were apparent and could be described as arti-facts that influence cohesion or division among the team.Evidence of a workforce hierarchy was also apparent throughthe quite different responses given by senior and junior staff,regardless of their discipline. In particular, junior staffreported that they felt the need to earn trust and respectbefore being accepted and did not feel supported when theyfirst join the clinical team. This was a subtle but importantfinding that made visible the perception that the seniorworkforce was less than accepting of inexperienced staff,only valuing junior staff once they were able to contribute tothe team’s workload. This is important because it suggeststhat the senior workforce did not display an accurate under-standing of the role of the trainee – that is to be trained –and the need to embrace and support junior staff in thistraining role rather than devaluing their contribution. Webelieve that this apparent lack of understanding is an import-ant deficiency in our senior workforce that requires specificand targeted response. Implicit in the roles of both seniormidwifery and medical staff is the requirement to train jun-ior staff, both midwifery and medical. We believe that thefindings of this study suggest that this expectation should bemore explicit and that there are enhanced opportunities forsenior staff of both craft groups to be involved in the formaland informal training of junior staff of both craft groups.In that regard, we recently introduced a simulation-basedtraining program for undergraduate medical and midwiferystudents taught by senior obstetricians and midwives thathas been highly regarded by all involved [40].Applying group theory [9] to our findings suggests that

this perceived disharmony and perhaps hostility withinthe team is likely to lead to disruptive behaviour, whichin turn, would be expected to make adverse events morelikely [21]. Targeted education of senior staff might beexpected to improve the senior-junior dynamic andthereby improve team function, reduce disharmony andimprove patient safety. This, of course, would need to betested but is a promising opportunity to enhance patientsafety. Certainly, future assessment of safety of a servicewould usefully explore the functionality of junior-seniorinteractions. Unfortunately, we were unable to explorewhether the senior-junior tension was equally apparentin the midwifery and medical workforces and so itwould be worth exploring the two workforces separately,particularly if one workforce could offer insights to theother.An extension to, but quite separate from, the senior-

junior tension was the theme of midwifery dominance. Wehad abundant references from junior medical staffreporting frustration at being ordered around by


Page | 77

senior midwives, leaving them with feelings that theircontribution to the workplace was irrelevant. This isnot a finding unique to our service. Relationshipsbetween midwives and medical staff have been ex-plored by others [21], with medical staff commonlyreporting that midwives were disrespectful and argu-mentative, particularly towards junior staff, and thatmidwives often obstructed medical staff from gainingexperience [21]. Nonetheless, until this study, wethought that relationships between medical and midwif-ery staff in our service were good. The tension betweenjunior medical staff and midwives only became apparentthrough our open-ended, explorative, themed interviews.Of course, inter-professional tensions are not even limitedto medical versus midwifery. In the US, tension betweenobstetric nurses and midwives has been explored [22].Tensions between these two groups is very similar to thatbetween midwives and obstetricians, where nurses’ and ob-stetricians’ practice principles are seen by midwives to med-icalise what they perceive to be a normal physiologicalprocess that should not be interfered with unless medicallyindicated. Certainly our findings highlight the import-ance of exploring inter-professional relationships as acomponent of building and sustaining a safe and effect-ive maternity service. Ensuring multidisciplinary in-volvement at all levels of the governance framework,including protocol development and implementation, casereview, and outcome audits, would be expected to engagethe craft groups equally and enhance relationships. Thiswould be expected to all and to ensuring good collabora-tive relationships between the professions so that patientcomplications are dealt with if and when they arise, and yetintervention is not applied in the absence of complications[25]. “Authentic mutual regard and trust creates environ-ments where care is not only of high quality, but continu-ally improving, and where staff enjoy coming to work andworking together” (Downe [25] p 291). Indeed, in work-places with high levels of mistrust the opposite occurs.Negativity breeds mistrust and fuels disharmony amongteam members whose interests become focused on in-dividual survival rather than optimising patient care[10,20-22,36]. Thus, our observations suggest that solu-tions to disharmony between midwifery and medicalstaff – a common observation – are likely to need con-sideration of levels of medical intervention, ensuringthat this is appropriate. We believe that this would helpgrow a confident and trusting medical and midwiferyworkforce.This might be difficult. Intervention without clinical

indication is common [28]. In particular, obstetriciansaged 40 years or younger appear to favour medicalisedcare, including elective caesarean section, compared totheir older counterparts [28]. Most were in favour ofincreased midwifery care during labour and reduced

induction of labour rates for women free of complicationsbut most were opposed to homebirth. Some obstetriciansreported that midwives withheld important clinical infor-mation to delay the obstetricians’ attendance; midwivesreported using delay tactics to minimize unnecessarymedical intervention. Such behaviour was evident in ourstudy, including exclusion of senior midwifery input in theassessment of women in labour during the early years ofthe service. Participants stressed that these issues have sincebeen resolved.

ConclusionsExploring staff perceptions of the safety and quality oftheir service has afforded important and unexpectedinsights into mechanisms of a less safe service and op-portunities for a safer service. It is certainly obvious thatthere is a need to develop a better understanding of howteamwork impacts patient safety. Our study shows thatwhile there was resounding affirmation of a safe andquality service there was clear evidence of impairedteamwork and team tensions – factors known to com-promise patient safety. In particular, we found thatsenior staff appeared to have an unrealistic view of therole of junior staff and that senior midwifery staff wereparticularly hostile to junior medical staff. In a maternitysetting midwives have arguably the greatest potential toinfluence staff morale and thereby team performance. Wesuggest that each of the emergent themes reported hereoffer opportunities for service enhancement. Specifically,practice improvement initiatives, such as increasing theinvolvement of all professional groups in service planning,review, and governance, might be expected to improverelationships and better balance authority. Future researchcould address such initiatives as a template for otherservices.

Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsSS co-designed the study, undertook the interviews, analysed the data andwrote the manuscript. EMW developed the original idea, co-designed thestudy, advised on data analysis and co-wrote the manuscript. WC co-designedthe study, assisted with data analysis and co-wrote the manuscript. All authorsread and approved the final manuscript.

Author details1The Ritchie Centre, Department of Obstetrics and Gynaecology, MonashUniversity and MIMR-PHI Institute of Medical Research, 246 Clayton Road,Clayton, Victoria 3168, Australia. 2School of Nursing and Midwifery, MonashUniversity, Clayton, Victoria, Australia.

Received: 16 September 2013 Accepted: 10 November 2014

References1. Braithwaite J, Iedema R, Jorm C: Trust, communication, theory of mind

and the social brain hypothesis. Deep explanations for what goes wrongin health care. J Health Organ Manage 2007, 21(4/5):353–367. doi:10.1108/14777260710778899.


Page | 78

2. Pronovost P, Holzmueller C, Ennen C, Fox H: Overview of progress inpatient safety. Am J Obstet Gynecol 2011, 204(1):5–10.

3. Pronovost P, Freischlag J: Improving teamwork to reduce surgical risk.J Am Med Assoc 2010, 304(15):1721–2722.

4. Van Der Vegt G, Bunderson J: Learning and performance inmultidisciplinary teams. Acad Manag J 2005, 48(3):532–547.

5. Argyris C: On Organizational Learning. 2nd edition. USA: BlackwellPublishing; 1999.

6. Argyris C: Teaching smart people how to learn. Harv Bus Rev 1991, 69:99–109.7. Manz C, Neck C: Teamthink: beyond the groupthink syndrome in

self-managing teams. J Manag Psychol 1995, 10(1):7.8. Kolb D, Rubin I, McIntyre J: Organizational Psychology: An Experimental Approach

to Organizational Behavior. 4th edition. New Jersey: Prentice-Hall Inc; 1984.9. Horwitz M, Cartwright D: A projective method for the diagnosis of group

properties. Hum Relations 1953, 6:397–410. doi:10.1177/001872675300600406.10. Cartwright D: Achieving change in people: some applications of

group dynamics theory. Hum Relations 1951, 4:387–391. doi:10.1177/001872675100400404.

11. Pronovost P, Goeschel C: Viewing health care delivery as science:challenges, benefits, and policy implications. Health Serv Res 2010,45(5.2):11508–11522. doi:10.1111/j.1475-6773.2010.01144.x.

12. Pronovost PJ, Colantuoni E: Measuring preventable harm: helping sciencekeep pace with policy. JAMA 2009, 301(12):1273–1275.

13. Agency for Health Care Research and Quality (AHRQ): National HealthcareQuality Report 2007. Rockville, MD: Agency for Healthcare Research andQuality; 2008.

14. Pronovost P, Nolan T, Zeger S, Miller M, Rubin H: How can clinicians measuresafety and quality in acute care? Lancet 2004, 363(9414):1061–1067.

15. Australian Commission on Safety and Quality in Health Care: Measurementfor improvement toolkit. 2006. Downloaded 22 November 2011 from:http://www.safetyandquality.gov.au/publications/measurement-for-improvement-toolkit-a/.

16. Brennan T, Leape L, Laird N, Hebert L, Localio A, Lawthers A, Newhouse J,Weiler P, Hiatt H: Incidence of adverse events and negligence inhospitalized patients: results of the Harvard Medical Practice Study I.Qual Saf Health Care 2004, 13:145–151. doi:10.1136/qshc.2002.003822.Downloaded from qhc.bmj.com on 8 April 2007.

17. Weingart S, Wilson R, Gibberd R, Harrison B: Epidemiology of medicalerror. BMJ 2000, 320(7237):774–777. Downloaded from http://www.amsa.org/AMSA/Libraries/Initiative_Docs/epidemiology_of_medical_error_BMJ.sflb.ashx 15 April 2007.

18. de Vries E, Ramrattan M, Smorenburg S, Gouma J, Boermeester M: The incidenceand nature of in-hospital adverse events: a systematic review. Quality andSafety in Health Care 2008, 17:216–223. doi:10.1136/qshc.2007.023622downloaded from qualitysafety.bmj.com on 12 December 2011.

19. Scalise D: Clinical communication and patient safety. Hospital HealthNetwork 2006, 80:49–54. 2.

20. Pinki P, Sayasneh A, Lindow S: The working relationship betweenmidwives and junior doctors: A questionnaire survey of Yorkshiretrainees. J Obstet Gynaecol 2007, 27:365–367.

21. Veltman L: Disruptive behavior in obstetrics: a hidden threat to patientsafety. Am J Obstet Gynecol 2007,587e1–587e5. doi:10.1016/j.ajog.2007.03.011.

22. Kennedy H, Lyndon A: Tensions and teamwork in nursing and midwiferyrelationships. J Obstet Gynecol Neonatal Nurs 2008, 37:426–435. doi:10.1111/j.1552-6909.2008.00256.x http://jognn.awhonn.org.

23. Raftopoulos V, Savva N, Papadopoulou M: Safety culture in the maternityunits: a census survey using the safety attitudes questionnaire. BioMed CentHealth Serv Res 2011, 11:238. http://www.biomedcentral.com/1472-6963/11/238 accessed 31/05/2012.

24. Weik E: “Birthing” versus “being delivered”: of bodies, ideologies, andinstitutions. Res Sociol Organ 2011, 27:171–201. Downloaded 14/07/2011.doi:10.1108/S0733-558X(2009)0000027008.

25. Downe S: Towards Salutogenic Birth in the 21st Century. In EssentialMidwifery Practice: Intrapartum Care. Edited by Walsh D, Downe S. Oxford,UK: Wiley-Blackwell; 2010. doi:10.1002/9781444317701.ch16.

26. Kennedy H, Anderson T, Leap N: Midwifery Presence: Philosophy, Scienceand Art. In Essential Midwifery Practice: Intrapartum Care. Edited by Walsh D,Downe S. Oxford, UK: Wiley-Blackwell; 2010. doi:10.1002/9781444317701.ch7.

27. Nolan M: Childbirth Education: Politics, Equality and Relevance. InEssential Midwifery Practice: Intrapartum Care. Edited by Walsh D, Downe S.Oxford, UK: Wiley-Blackwell; 2010. doi:10.1002/9781444317701.ch3.

28. Klein M, Liston R, Fraser W, Baradaran N, Hearps S, Tomkinson J, KaczorowskiJ, Brant R, The Maternity Care Research Group: Attitudes of the newgeneration of Canadian obstetricians: how do they differ from theirpredecessors? Birth 2011, 38:129–139.

29. Lincoln Y, Guba E: Naturalistic Inquiry. USA: Sage Publications; 1985.30. Schneider Z, Whitehead D, Elliot D: Nursing and Midwifery Research. Methods

and Appraisal for Evidence Based Practice. 3rd edition. Sydney: Mosby; 2007.31. Agency for Healthcare Research and Quality: Hospital survey on patient

safety culture. http://www.ahrq.gov/professionals/quality-patient-safety/patientsafetyculture/index.html (accessed 16/1/14).

32. Creswell J: Research Design. Qualitative, Quantitative and Mixed MethodsApproaches. 3rd edition. London: Sage; 2009.

33. Richards L: Handling Qualitative Data. A Practical Guide. 2nd edition. USA:Sage Publications; 2009.

34. Thurmond VA: The point of triangulation. J Nurs Scholarship 2004,doi:10.1111/j.1547-5069.2001.00253.x.

35. Biro MA, Knight M, Wallace E, Papacostas K, East C: Is place of birthassociated with mode of birth? The effect of hospital on caesareansection rates in a public metropolitan health service. Aust N Z J ObstetGynaecol 2014, 54(1):64–70. doi: 10.1111/ajo.1214.

36. Berlin J, Carlström E: From artifact to effect: the organizing effects ofartefacts on teams. J Health Organ Manage 2010, 24(4):412–427.doi:10.1108/14777261011065011.

37. Walker S, Brett S: Oiling the wheels of intensive care to reduce “machinefriction”: The best way to improve outcomes? Crit Care Med 2010,38(10):s642–s648. (Supplement) doi:10.1097/CCM.0b013e3181f20691.

38. National Institute for Health and Clinical Excellence: Intrapartum care. Careof healthy women and their babies during childbirth. 2007. http://www.nice.org.uk/search?q=intrapartum+care accessed 03/09/2013.

39. The Royal Australian and New Zealand College of Obstetricians andGynaecologists: Intrapartum fetal surveillance clinical guidelines. 2006.http://www.fsep.edu.au/index.php?option=com_content&view=article&id=80&Itemid=97 accessed 03/09/2013.

40. Kumar A, Carole Gilmour C, Debra Nestel D, Robyn Aldridge R, Gayle MCLellandG, Wallace E: Can we teach core clinical obstetrics and gynaecology skillsusing low fidelity simulation in an interprofessional setting? Aust NZ J ObstetGynaecol epub ahead of publication doi:10.1111/ajo.12252.

doi:10.1186/s12913-014-0591-4Cite this article as: Sinni et al.: Perinatal staff perceptions of safety andquality in their service. BMC Health Services Research 2014 14:591.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit


Page | 79

CHAPTER SIX

Discussion and closing marks

Page | 80

6.1 Summary of the findings

I have used mixed methods research to evaluate the newest maternity service in Victoria’s

largest healthcare organisation. Preliminary discussions about an evaluation commenced in

2002, a full three years before the service opened. My final paper was published in 2016,

making this a 14 year odyssey – four year’s longer than Homer’s epic travails of Odysseus.

Initially, at the direction of the Victorian Department of Health, the evaluation was to

encompass all aspects of the new maternity service but, as is not uncommon with

governmental directives, this was to be without commensurate funding to undertake such a

large project. The desire to undertake “deep dive” evaluation and to incorporate the

evaluation into a PhD program forced the evaluation team to limit the size and scope of the

project. In hindsight this was the correct decision because enlisting the local clinicians to

develop a methodology that would be meaningful to them, as described in Chapter One

and Chapter Three, afforded the findings much more traction in effecting change. Such

engagement and involvement of the clinical staff would not have been possible across all

domains of care. Instead, the evaluation was limited to antenatal care, Chapter Four, and

to personnel perceptions of service safety and quality, Chapter Five.

The results from the clinical audit are reported in Chapter Four. High levels of compliance

with clinical standards were demonstrated overall. However, documentation of the care

provided was generally poor, regardless of the discipline – medical or midwifery. Indeed,

overall there was no difference in compliance to standards between medical and midwifery

care providers in the low-risk maternity model provided at the site at the time. I also

highlighted the ongoing problems with multiple data entry, incompatible electronic data

interfaces and inaccessibility of data across the various service providers involved in

pregnancy care. This is probably reflective of healthcare data capture and storage more

Page | 81

generally across Victoria. The finding did cause fragmentation of medical records to be

added as a priority to the Risk Register for Monash Health’s maternity service and

contributed to the business case for the purchase, in 2015, of a new electronic maternity

record (K2 Systems).

My qualitative observations on staff perceptions were also reassuring (Chapter Five). Staff

perceptions of the safety and quality of the service were high. I was surprised to find that

junior staff, particularly junior medical staff on rotation through obstetrics, felt

undervalued and that it took time for all junior staff, medical and midwifery, to feel

accepted as part of the clinical team. This insight was fed back to the senior medical and

midwifery staff at Casey Hospital and, I hope, led to some changes in what they expected

of their junior colleagues. Whether this has resulted into changes in junior staff perceptions

has not yet been re-evaluated.

The association between midwifery dominance, power and protecting women is well

documented in the literature and, not surprisingly, was evident in my thematic analyses.

That the health service had already identified some of this issue and made steps towards

improvements, sometime necessitating making individuals redundant, reflects, I believe, a

mature and confident health service with a clear vision of who they are and want they

want.

6.2 Limitations

The research was designed to address the needs of the one organisation where the

evaluation took place and to address care provision in the very early days of the service.

Whether the findings hold true today for the service, or indeed hold true for other services,

has not been tested. Further, time constraints did not allow evaluation of care provision in

Page | 82

labour or postpartum. That would also be worthwhile. In this regard, it would be relatively

straightforward to develop audit tools to assess the quality of intrapartum and postnatal

care, as was done for pregnancy care. Last, it was not possible to correlate poor outcomes

with care provision. I have argued that simply measuring outcomes is not sufficient

(Chapter Four) but that measuring the processes that lead to the outcomes is also

necessary. Whether inadequate care provision underlies the majority of adverse outcomes

has not actually been explored by anyone. Rather, individual case reviews – such as Root

Cause Analyses – are undertaken following an adverse outcome, often identifying care

deficiencies. Whether such deficiencies occur with similar frequency or severity when

there is not an adverse outcome has not been explored. The audit tool that I developed and

validated (Chapter Three) would allow efficient identification of cases deficient or

sufficient in process to undertake such an analysis.

6.3 Directions for future research

The research has identified multiple future projects worthy of exploration at any service

where maternity care is provided. These are identified within each of the chapters. The

clinical audit was extensive and detailed and therefore time consuming. Any of the

individual indices measured is worthy of detailed exploration as separate research projects,

whether locally or across a state-wide or national system.

Similarly, the qualitative research generated further research possibilities. Although the

results were positive, the lack of support for junior staff was unexpected and worthy of

exploration at the local level to ascertain if it is an ongoing issue.

Page | 83

6.4 Conclusions

Using a dedicated clinical audit tool designed to measure patient safety in maternity

services, I have shown that Victoria’s newest maternity service is highly compliant with

clinical standards of pregnancy care. However, there were low levels of compliance with

documentation requirements, which probably reflect inadequate information capture and

storage issues more than clinician intent to be non-compliant. Nonetheless, I identified

many opportunities for service improvement that, if implemented, would be expected to

translate to better outcomes over time. I hope that the high level of involvement of the

clinicians and the detailed feedback of the results to them will lead to the implementation

of various local service improvement strategies that, in turn, may help every woman at

Casey Hospital to receive the highest standard of maternity care regardless of her level of

risk.

patient safety in maternity care: towards better outcomes · 2017. 4. 4. · patient safety in...

Documents