Using theory to design,implement and evaluate sustained nurse domestic violence screening and supportive care

Abstract

Domestic violence is prevalent and harmful for abused women and children and damages women’s capacity to maintain wellbeing and parent effectively. As women’s health is likely to be poor, greater identification and support by nurses for abused women is needed. The aim of this study is to highlight the importance of theory in nursing intervention research on domestic violence and how the application of implementation theory can be used to promote programme sustainability. In particular, we discuss the use of Normalization Process Theory (NPT) in the MOVE (improving maternal and child health care for vulnerable mothers) study, a randomised controlled trial which aimed to increase nurse inquiry, safety planning and referral of abused women. NPT was used in the participatory action research and design stage of the trial, in the implementation and process evaluation phases. In conclusion, the use of implementation theory in nursing research can enhance the uptake of complex interventions, address the knowledge–practice gap and potentially improve outcomes for women experiencing domestic violence.

Keywords

complex interventions domestic violence maternal and child health nursing Normalization Process Theory sustainability

Introduction

Domestic violence policy and sustainable care

One third of women across the globe have experienced some form of domestic and/or sexual violence in their lifetime. Exposure to domestic violence (DV) results in major mental and physical health damage for women and children (World Health Organization, 2013a). These health problems force victims to attend health care services frequently. The disease burden warrants attention by health care services to enhance the capacity of providers to support abused women and children (García-Moreno et al., 2014). This includes greater identification and support by nurses for victims of abuse.

Governments have become increasingly aware of the economic cost and detrimental effects of violence against women and have introduced policy and public health prevention programmes such as screening for DV (e.g. Council of Australian Governments, 2010). This policy assumes that women experiencing DV will be identified and receive the support they need. This is often not the case, with many health care providers, including nurses, encountering barriers to asking women about abuse and providing supportive care (Sprague et al., 2012). Although many screening programmes now exist in various settings such as emergency departments, antenatal clinics and primary care, evidence of the benefits for women of DV screening are limited (O’Doherty et al., 2015). According to the World Health Organization (2013b), current best practice suggests nurses screen only pregnant women and others in high risk groups (such as those seeking abortion services) for DV. It is also unclear whether the introduction of new clinical practices such as DV screening programmes are sustainable (O’Doherty et al., 2015), especially in complex or ‘real world’ environments such as hospitals and primary care settings. Very few rigorous studies have examined DV screening sustainability beyond 12 months (O’Campo et al., 2011; Taft et al., 2015). Improvements for abused women and children can only be achieved if nurse DV screening and care is sustained.

There is a significant gap between best practice and what is actually delivered in clinical care (Grol and Wensing, 2013). Researchers frequently develop and test clinical interventions which are shown to prevent and/or improve various conditions successfully, yet this evidence often does not translate to knowledge users such as nurses and other health care professionals. The outcome of this evidence–practice gap, between what we know works and what is currently delivered in clinical settings, is suboptimal patient care (Grol and Wensing, 2013).

How then, do we best ensure that evidence and new practices such as DV screening become effectively delivered within clinical practice? One strategy to bridge the knowledge–practice gap is to identify barriers (translational blocks) and enablers of integration by conducting implementation research (Grol and Wensing, 2013).

Implementation science

Implementation research or science is the study of methods that influence the integration of research findings into policy and practice. It attempts to understand individual behaviours and the wider contextual factors contributing to the sustainable uptake and implementation of evidence-based interventions (National Institutes of Health, 2015). This new knowledge is then used to optimise implementation activities. Research on implementation addresses the level to which health interventions can fit within ‘real world’ or pragmatic health service systems (Peters et al., 2013). When considering implementation of evidence into practice, terminology often differs between countries and disciplines, which may cause confusion for readers. Knowledge translation is a common term in Canada, whereas implementation science is preferred in Europe and in the USA (Grol and Wensing, 2013). Alternative terminology includes knowledge translation, knowledge transfer, knowledge exchange, research utilisation, diffusion and dissemination (McKibbon et al., 2010).

Successful implementation of interventions (such as DV screening) into clinical practice can be a lengthy and complex process (Medical Research Council, 2008). Complex interventions have many interacting components that act both independently and interdependently. When developing complex interventions, substantial piloting work and consideration of the practical issues of implementation are needed to prevent the development of weak interventions that may be difficult to implement and evaluate (Medical Research Council, 2008). Greater awareness of the evidence–practice gap has led to calls for increased use of theory to explain implementation processes (McEvoy et al., 2014).

Use of theory

Many factors determine implementation success and can include features of the target group (knowledge, skill, attitude), client (behaviour, compliance), contextual setting (social norms, leadership) and economy (Grol et al., 2013). There are various theories and/or knowledge translation frameworks that can be used in implementation research (Grol et al., 2013; Nilsen, 2015). No single theory or implementation framework will address all the factors described above. Implementation theory can be used to provide a framework for thinking, to design interventions and as a framework for analysis to understand how and why interventions succeed or fail (McEvoy et al., 2014; Nilsen, 2015). Theory driven approaches are recommended in complex intervention research (De Silva et al., 2014).

Theoretical frameworks can be classified into process and/or impact theories (Grol et al., 2013). Examples of practical process frameworks include the precede–proceed model of Green and Krueter (1991) and the framework of Fixsen et al. (2009) on core implementation drivers (Moullin et al., 2015). Process frameworks often act as descriptive stepped guides to planning and implementing change.

Impact theories are more explanatory or predictive and aim to describe key determinants that facilitate change (Grol et al., 2013). These may include theories on factors related to individual psychology, such as the Theory of Planned Behavior (Ajzen, 1991) or social interaction and organisational context (Normalization Process Theory; NPT) (May and Finch, 2009). Sociological approaches focus on collective work and cooperation rather than solely emphasising individual behaviour and intentions. Researchers may use a combination of implementation theories in one intervention, for example using individual (patient or health care professional) behaviour theories with more complex sociological and contextual frameworks to gain a deeper understanding of the systems influencing clinician and implementation change.

Theory assists us to interpret our research findings. Sociological theories can help to make predictions regarding the sort of context most favourable for a successful trial. Sociological theories such as NPT (described in more detail below) are descriptive, explanatory and predictive (Moullin et al., 2015), derived from a complex intervention in clinical care (May et al., 2003) and thus may suit pragmatic interventions such as nurse DV screening research.

Implementation theory in DV intervention research

Improving the implementation of DV evidence into clinical practice may be an effective strategy to advance the care of abuse victims; however, very few knowledge translation initiatives regarding DV have been published and research is limited on best methods for uptake to inform policy and practice (Larrivée et al., 2012; MacGregor et al., 2014). Larrivée et al. (2012) suggest that the study of knowledge translation has not been extended to research on violence against women and children. A review of 62 papers on knowledge translation and implementation strategies in intimate partner violence and child maltreatment intervention research revealed significant evidence gaps (MacGregor et al., 2014). The review identified the underuse of implementation strategies and lack of evidence on the long-term effectiveness of implementation interventions.

There are some recent examples in which implementation theories have been used in health care service and nursing DV intervention research. These include the violence adapted transtheoretical model of behaviour change (TTM) (Frasier et al., 2001; Prochaska and Velicer, 1997) used in the development of the GP intervention trial of DV screening by Hegarty et al. (2013) and follow-up counselling for abused women. Systems thinking was also successfully utilised by the US medical health service Kaiser Permanente to develop multilevel system support for implementation of DV screening and care by health care providers (Miller et al., 2015).

Nursing-specific DV interventions have also been developed and implemented using randomised controlled designs. Jack and colleagues (2012) describe the development of a modified nurse family partnership (NFP)/DV home visiting intervention. Based on the successful NFP programme (Olds et al., 2007), this theory-based intervention used several models (including the TTM) to develop a nurse tailored intervention, aimed at enhancing implementation and optimising care for young, vulnerable first time mothers (Jack et al., 2012).

In this papers we discuss the use of NPT in the implementation of a complex intervention called MOVE (improving maternal and child health (MCH) care for vulnerable mothers). The MOVE trial of enhanced DV screening and supportive nursing care applied the NPT to several stages of a complex intervention, aimed at sustainable DV screening and improved safety planning and referrals for women (Hooker et al., 2015). The aim is to discuss the importance and use of implementation theory in nursing DV intervention research.

Methodology

MOVE intervention

MOVE was a randomised controlled trial of an enhanced model of DV screening and supportive care by MCH nurses. It was conducted for 12 months across eight MCH nurse teams in the north west area of metropolitan Melbourne, Victoria, Australia. The primary aims of the original MOVE study were to compare whether more MCH nurses in the intervention than in the comparison arm screen for DV, have more mothers disclose DV and make safety plans, and refer abused women to specialist DV services 12 months after implementation of the MOVE model (Taft et al., 2015). Within an over-arching NPT theoretical framework, the MOVE design (detailed below) included a systematic review of evidence on community nurse DV practices (unpublished) and participatory action research for six months with four nurse consultants from the intervention group teams. The MOVE model was the result of this six-month practice consensus. It included nurse clinical practice guidelines, a clinical pathway and a maternal health and wellbeing checklist that combined self-completion DV screening questions with other questions regarding maternal physical and emotional health. In addition, each intervention team had several nurse mentors and was allocated a local DV liaison worker to provide secondary consultation and support for nurses including facilitated referrals to specialist DV services.

Study context

The Victorian MCH service is a community-based primary health care service for families located within local government. MCH nurse midwives deliver the predominantly postnatal service to families with children from newborn to school age. Their role is similar to the health visitor in the UK. Families are offered a home visit initially and are then encouraged to attend the MCH clinic for nine follow-up consultations until the child is 3.5 years. Additional consultations are offered to vulnerable families.

Earlier research with MCH nurses recognised that MCH nurses referred very few abused women for follow-up care and encountered both individual and structural barriers to asking about DV, despite prior DV training (Taft et al., 2011). At the early stages of the MOVE trial design, the Victorian state government introduced mandatory routine DV screening by all MCH nurses at the four-week postnatal visit and at any other time if needed. This led to a screening versus enhanced screening and supportive care trial (MOVE). If screening was to be introduced into MCH, it would need to be a sustainable clinical practice to maintain the benefits for women and children. To maximise sustainability, the NPT was used throughout the trial.

What is NPT?

Developed by Carl May and colleagues in the UK (May and Finch, 2009) from the Normalization Process Model (May et al., 2007b), NPT is a social implementation theory that provides a conceptual framework for understanding ‘the social organisation of the work (implementation), of making practices routine elements of everyday life (embedding), and of sustaining embedded practices in their social contexts (integration)’ (May and Finch, 2009: 538). NPT recognised the introduction of a new clinical practice as a social process and the action or work of implementation. It aids our understanding of how new and complex interventions become normalised into clinical practice (Finch et al., 2013). The theory can be used as a tool to design interventions, to organise and evaluate the work of individuals and groups (implementation processes) (Murray et al., 2010) and answered our research question: How can those factors that promote or inhibit normalisation be identified, conceptualised and evaluated so that evidence based programmes such as MOVE become successfully integrated into MCH nurse practice?

NPT constructs

The work of implementation and normalisation is operationalised through four main mechanisms or constructs – coherence, cognitive participation, collective action and reflexive monitoring (each with four sub-constructs) (May and Finch, 2009). These constructs are not linear but iterative and interrelated in a dynamic relationship when new practices are introduced (Murray et al., 2010).

Coherence or ‘sense making work’ refers to an individual and communal understanding of the intervention or process required. For an intervention such as DV screening to be normalised, all stakeholders need to have an understanding of and value the DV work and to demonstrate some investment in the meaning of the concept.

Cognitive participation, or ‘participation work’, states that normalisation of a new practice or model occurs if all stakeholders engage with the model and use it, and that there is a demonstration of commitment to the DV screening work required to implement the intervention.

Collective action, or ‘enacting the work’, refers to doing the work needed to enact a set of practices. This involves how new roles are organised, how people relate and are involved and the knowledge and skills needed to complete the work. Normalisation of a model involving DV screening occurs if all stakeholders work to operationalise it and demonstrate investment in effort. The four sub-constructs of collective action were the mechanisms underlying the earlier Normalisation Process Model (May et al., 2007a).

Reflexive monitoring, or ‘appraisal work’, is the individual or group assessment of the way a new practice affects participants and other stakeholders. Normalisation occurs if continual monitoring and evaluation of the screening work occurs. It refers to an investment in comprehension (May et al., 2011).

Discussion: Using NPT

NPT helps explain in a practical way the dynamics of how complex interventions are successfully implemented, embedded and integrated (May and Finch, 2009). Using NPT for a complex intervention like MOVE, we needed to consider three different, interacting components and that nurses work in the ‘context of complex and dynamic collective interactions’ (May et al., 2007b).

Actors: in this study, MCH nurses, nurse mentors, MCH nurse team leaders and DV liaison workers were the individuals and/or groups involved in screening implementation.

Objects: resources and procedures that enable adoption of the intervention included the maternal health and wellbeing checklist and clinical practice guidelines.

Contexts: physical or organisational structures that can facilitate or inhibit change. For the MOVE nurse teams this included state and local government funding, policy and specialist DV service support for MCH nurse DV work.

NPT offers a theoretical framework, which optimises the development of the trial intervention by ‘supporting intervention design, describing the context of a trial and supporting the interpretation of a trial’s results’ (May et al., 2010). Consequently, NPT was used in the design, implementation and evaluation stages of the MOVE complex intervention in order to enhance its sustainability.

Design of MOVE

Successful knowledge translation and exchange requires collaboration between researchers and knowledge users. If research is to be applied it needs to address knowledge users’ questions and be relevant to their needs (MacGregor et al., 2014). Hence MOVE researchers included a participatory action research (Baum et al., 2006) component to the design phase of the study, which included four MCH nurse consultants from the intervention group for six months. Participatory action research ‘is collective, self-reflective inquiry that researchers and participants undertake so they can understand and improve upon the practices in which they participate’ (Baum et al., 2006: 854). In iterative monthly cycles and using NPT concepts to frame questions, nurses consulted with their teams and researchers to explore DV practice barriers, propose answers and develop a workable model and relevant clinical resources. Nurses devised questions and returned to their teams to discuss methods of DV screening and supportive care that could be sustained in clinical practice. All NPT constructs – coherence, cognitive participation, collective action and reflexive monitoring (May and Finch, 2009) were considered in the design of the MCH nurse clinical resources. Researchers and nurses used NPT to think through issues of implementation when planning and designing the MOVE trial. For example:

Coherence: a model that nurses could understand and in which they could invest meaning. Is the MOVE screening model clearly distinguishable from usual routine screening at four weeks? Will it fit with the existing goals of the organisation?

Cognitive participation: Who are the main stakeholders? How can we encourage nurses to invest more time and energy to do the work? What are the barriers to engaging with the new DV screening practice and how can they be overcome?

Collective action: How compatible is it with existing practices and what further relationships do we need to develop with referral agencies to improve collaboration? Will staff require extensive training to use it?

Reflexive monitoring: How can we monitor what effects the MOVE intervention has had?

Combining NPT with MCH nurse consultants’ knowledge and experience during the trial design enhanced the trial’s credibility through the co-designed resources and practices aimed to enhance implementation and sustainability.

MOVE intervention

NPT and its constructs were used to conceptualise the development of actor roles and the intervention components of MOVE. Actors included intervention group nurse mentors, team leaders and DV liaison workers. Role and model-specific practice recommendations were included in the MOVE clinical practice guidelines under four sections: inquire and connect; assess and support; document and quality assurance; and routine practice.

Coherence: the nurse mentor acted as a practice champion, providing support to universal nurses in their DV work. Nurse mentors were encouraged to liaise with DV workers and services, keep the DV topic on the agenda at regular team meetings and provide training for new team members unfamiliar with MOVE. Nurse team leaders emphasised the DV work as important. These tasks aimed to improve nurse and team understanding of the DV screening work required.

Cognitive participation: to encourage engagement with the DV work, the action research generated a self-completion maternal health and wellbeing checklist for screening women at three to four months. This time was preferred by nurses and had a better fit with the existing MCH practice framework. Self-completion removed responsibility for nurse face-to-face screening and was preferred by both mothers and nurses. Inclusion of the DV liaison worker in the team aimed to encourage DV work, provide secondary consultation for nurses and facilitate warm referrals.

Collective action: nurse team leader roles included providing skills sessions for nurses to practice role playing DV scenarios with staff. They also provided a management link to the organisational context, ensuring resource support. Team leaders ensured that local government policy addressed safety issues for nurses attending home visits and those in sole practitioner clinics. Nurse mentors were asked to update and distribute information about specific agencies such as those for migrant and refugee or indigenous communities.

Reflexive monitoring: to facilitate self and communal reflective practice on the DV work, quality assurance measures were incorporated into the clinical guidelines. These included team leader or nurse mentor chart audits, client surveys and provision of access to clinical supervision. Team performance on the DV work was to be discussed at regular team meetings.

MOVE process evaluation

Process evaluation is essential in complex intervention research to assess implementation quality, highlight contextual factors and help researchers explain their results. Quantitative measures of trial success do not provide information on how to replicate studies in a differing context. Furthermore, negative trials may be due, not to poor interventions, but to poor programme fidelity (the intervention not delivered as intended) (Moore et al., 2015).

Results from the MOVE study showed that the intervention was successful at increasing nurse identification and particularly completing safety plans with women. Sustainability was also achieved using NPT, as two years after the intervention, MOVE nurse teams had increased the likelihood of completing safety plans with women from three to four times than were MCH nurses in the comparison groups (Taft et al., 2015). NPT was used extensively throughout the evaluation stage to design data collection tools and to use as a guide to coding and a lens for interpreting qualitative results (May and Finch, 2009).

All NPT constructs were incorporated into the nurse survey questions and key stakeholder interviews used for process evaluation. For example, coherence measures in online MCH nurse survey questions included multiple scale responses to items such as: how important is it to screen all women for DV; MCH nurse interventions can make a difference to the lives of women and children experiencing DV; and it is part of my job to have the time to support women experiencing DV. Further information on survey questions within NPT constructs are reported in the MOVE/MOVE 2 trial process evaluation papers (Hooker et al., 2015, 2016).

Process evaluation involved analytical triangulation or multiple ways of seeing the data using NPT. All qualitative data from interviews and open ended text in surveys were coded inductively initially, then deductively to proposition statements derived from NPT. Transcript data were then coded to propositions/constructs that facilitated the deductive coding. For example, to describe coherence – the proposition that ‘MOVE will be normalised if all stakeholders have an understanding of and value DV work’ was coded from all data sources. The MOVE process paper provides more detail on proposition statements (Hooker et al., 2015).

The NPT website and toolkit provided useful guidance on coding and analysis (May et al., 2011). In addition, the NPT toolkit can display a visual diagram of theory constructs in your intervention. Answering a series of NPT construct questions generates useful radar plots, which give a visual interpretation of the analysis and highlight implementation strengths and weaknesses within each construct.

Lessons learnt from using NPT

From the research described, it is evident that NPT may be a valuable tool to assist nurse researchers in their planning, implementation and evaluation of complex interventions. NPT guides our understanding of how new clinical practices become part of routine care. The theory’s strength lies in its practical value in real world settings and of ‘understanding how new ways of thinking, acting and organising become embedded in healthcare systems’ (May et al., 2010). The extensive use of NPT in the MOVE trial facilitated our understanding of the individual and collective nurse DV work required and contributed to the sustained success of nurse DV practice change (Taft et al., 2015). Although there are clear benefits to using NPT, understanding and applying the theory to the research context of interest may be a challenge. This was especially evident when coding qualitative data, when, to ensure accuracy, a robust understanding of all theory constructs and sub-constructs is needed. NPT is a mid-range theory, which is becoming more recognised within implementation research. More use of NPT in nurse DV intervention research is needed to validate further its suitability in this context.

Conclusion

There is often a significant translation gap between evidence, policy and practice (Grol and Wensing, 2013). Implementation research allows us to address this gap and understand the processes involved in a complex nursing intervention such as MOVE. There are many types of frameworks and theories (Nilsen, 2015) that can be and have been applied to implementation and nurse DV intervention research. NPT is especially useful in complex settings and helps us to understand how a new and complex intervention becomes part of everyday sustained clinical practice (May and Finch, 2009). This paper has described the use of NPT in the design, intervention and evaluation stages of the MOVE trial, which enhanced both DV screening and sustainability (Taft et al., 2015). MOVE success in part is due to the established partnership with knowledge users and the participatory action research, evidence-based, strong design and the incorporation of an implementation theory into all stages of the trial (Hooker et al., 2015). The use of implementation theory in nursing DV research can enhance the uptake of complex interventions, address the knowledge–practice gap and potentially improve outcomes for women and children experiencing DV.

Key points for policy, practice and/or research

DV is prevalent and harmful.

The health sector response needs to be sustainable.

Implementation theory can facilitate sustainable nursing practice change.

NPT is useful in complex DV intervention research.

Footnotes

Declaration of Conflicting Interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Leesa Hooker, PhD, is a maternal and child health nurse, midwife and lecturer in nursing and midwifery at La Trobe University. She has been working in the area of women’s and children’s health for the past 20 years and has recently completed a PhD on improving maternal and child health nurse practice for families experiencing domestic violence.

Angela Taft, PhD, is professor/director at the Judith Lumley Centre, La Trobe University, and a social scientist using rigorous combinations of qualitative and epidemiological methods to answer complex questions about women’s health. Over the last 13 years she has led a major programme of research on intimate partner/gender-based violence.

References

Ajzen

(1991) The theory of planned behavior. Organizational Behavior and Human Decision Processes 50(2): 179–211.

Baum

MacDougall

Smith

(2006) Participatory action research. Journal of Epidemiology and Community Health 60(10): 854–857.

Council of Australian Governments (2010) National Action Plan to reduce violence against women and their children. Canberra: COAG.

De Silva

Breuer

Lee

(2014) Theory of change: A theory-driven approach to enhance the Medical Research Council’s framework for complex interventions. Trials 15: 267.

Finch

Rapley

Girling

(2013) Improving the normalization of complex interventions: Measure development based on normalization process theory (NoMAD): Study protocol. Implementation Science 8: 43.

Fixsen

Blase

Naoom

(2009) Core implementation components. Research on Social Work Practice 19(5): 531–540.

Frasier

Slatt

Kowlowitz

(2001) Using the stages of change model to counsel victims of intimate partner violence. Patient Education and Counseling 43(2): 211–217.

García-Moreno

Hegarty

d’Oliveira

AFL

(2014) The health-systems response to violence against women. The Lancet 385(9977): 1567–1579.

Green

Kreuter

(1991) Health Promotion Planning: An Educational and Environmental Approach, Palo Alto, CA: Mayfield Publishing Company.

10.

Grol

Wensing

(2013) Implementation of change in healthcare: A complex problem. In: Grol

Wensing

Eccles

(eds) Improving Patient Care: The Implementation of Change in Health Care, 2nd edn. UK: Wiley Blackwell, pp. 3–17.

11.

Grol

Wensing

Bosch

(2013) Theories on implementation of change in healthcare. In: Grol

Wensing

Eccles

(eds) Improving Patient Care: The Implementation of Change in Health Care, 2nd edn. UK: Wiley Blackwell, pp. 18–39.

12.

Hegarty

O’Doherty

Taft

(2013) Screening and counselling in the primary care setting for women who have experienced intimate partner violence (WEAVE): A cluster randomised controlled trial. The Lancet 382(9888): 249–258.

13.

Hooker

Small

Humphreys

(2015) Applying normalization process theory to understand implementation of a family violence screening and care model in maternal and child health nursing practice: A mixed method process evaluation of a randomised controlled trial. Implementation Science 10: 39. DOI: 10.1186/s13012-015-0230-4.

14.

Hooker L, Small R and Taft A (2016) Understanding sustained domestic violence identification in maternal and child health nurse care: process evaluation from a 2-year follow-up of the MOVE trial. Journal of Advanced Nursing 72(3): 534–544. doi:10.1111/jan.12851.

15.

Jack

Ford-Gilboe

Wathen

(2012) Development of a nurse home visitation intervention for intimate partner violence. BMC Health Services Research 12: 50.

16.

Larrivée

M-C

Hamelin-Brabant

Lessard

(2012) Knowledge translation in the field of violence against women and children: An assessment of the state of knowledge. Children and Youth Services Review 34(12): 2381–2391.

17.

McEvoy

Ballini

Maltoni

(2014) A qualitative systematic review of studies using the normalization process theory to research implementation processes. Implementation Science 9: 2.

18.

MacGregor

Wathen

Kothari

(2014) Strategies to promote uptake and use of intimate partner violence and child maltreatment knowledge: An integrative review. BMC Public Health 14: 862.

19.

McKibbon

Lokker

Wilczynski

(2010) A cross-sectional study of the number and frequency of terms used to refer to knowledge translation in a body of health literature in 2006: A Tower of Babel? Implementation Science 5: 16.

20.

May

Finch

(2009) Implementing, embedding, and integrating practices: An outline of normalization process theory. Sociology 43(3): 535–554.

21.

May

Finch

Mair

(2007a) Understanding the implementation of complex interventions in health care: The normalization process model. BMC Health Services Research 7: 148.

22.

May

Mair

Dowrick

(2007b) Process evaluation for complex interventions in primary care: Understanding trials using the normalization process model. BMC Family Practice 8: 42.

23.

May

Mort

Williams

(2003) Health technology assessment in its local contexts: Studies of telehealthcare. Social Science & Medicine 57(4): 697–710.

24.

May C, Murray E, Finch T, et al. (2010) Normalization Process Theory. Online users’ manual and toolkit. Available at: http://www.normalizationprocess.org.

25.

May

Finch

Ballini

(2011) Evaluating complex interventions and health technologies using normalization process theory: Development of a simplified approach and web-enabled toolkit. BMC Health Services Research 11: 245.

26.

Medical Research Council (2008) Developing and evaluating complex interventions: new guidance. London: MRC.

27.

Miller

McCaw

Humphreys

(2015) Integrating intimate partner violence assessment and intervention into healthcare in the United States: A systems approach. Journal of Women's Health 24(1): 92–99.

28.

Moore

Audrey

Barker

(2015) Process evaluation of complex interventions: Medical Research Council guidance. BMJ 350: h1258.

29.

Moullin

Sabater-Hernandez

Fernandez-Llimos

(2015) A systematic review of implementation frameworks of innovations in healthcare and resulting generic implementation framework. Health Research Policy and Systems 13: 16.

30.

Murray

Treweek

Pope

(2010) Normalisation process theory: A framework for developing, evaluating and implementing complex interventions. BMC Medicine 8: 63.

31.

National Institutes of Health (2015) Implementation Science Information and Resources. Available at: http://www.fic.nih.gov/researchtopics/pages/implementationscience.aspx.

32.

Nilsen

(2015) Making sense of implementation theories, models and frameworks. Implementation Science 10: 53.

33.

O’Campo

Kirst

Tsamis

(2011) Implementing successful intimate partner violence screening programs in health care settings: Evidence generated from a realist-informed systematic review. Social Science & Medicine 72(6): 855–866.

34.

O’Doherty

Hegarty

Ramsay

(2015) Screening women for intimate partner violence in health-care settings. Cochrane Database of Systematic Reviews 4: CD007007. DOI: 10.1002/14651858. C D007007.pub2.

35.

Olds

Sadler

Kitzman

(2007) Programs for parents of infants and toddlers: Recent evidence from randomized trials. Journal of Child Psychology and Psychiatry 48(3–4): 355–391.

36.

Peters

Adam

Alonge

(2013) Implementation research: What it is and how to do it. BMJ 347: f6753.

37.

Prochaska

Velicer

(1997) The transtheoretical model of health behavior change. American Journal of Health Promotion 12(1): 38–48.

38.

Sprague

Madden

Simunovic

(2012) Barriers to screening for intimate partner violence. Women & Health 52(6): 587–605.

39.

Taft

Hooker

Humphreys

(2015) Maternal and child health nurse screening and care for mothers experiencing domestic violence (MOVE): A cluster randomised trial. BMC Medicine 13: 150.

40.

Taft

Small

Hegarty

(2011) Mothers’ AdvocateS In the Community (MOSAIC) – non-professional mentor support to reduce intimate partner violence and depression in mothers: A cluster randomised trial in primary care. BMC Public Health 11: 178.

41.

World Health Organization (2013a) Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non partner sexual violence , Geneva: WHO.

42.

World Health Organization (2013b) Responding to intimate partner violence and sexual violence against women: WHO clinical and policy guidelines, Geneva: WHO.