Lessons for the future: Reflections on a review of child death overview panels through a local lens in the United Kingdom

Abstract

Child death overview panels (CDOPs) were set up in the United Kingdom following the confidential enquiry into maternal and child health. Their scope is to identify learning points and modifiable factors that focus on improving services and prevent further deaths. In the light of UK national review and subsequent legislative changes to local safeguarding arrangements, we wanted to share the lessons learnt from our local network study during this time of transition. At times of system change, organizational memory can be eroded, which results in lost opportunities to further strengthen multi-agency working in practice. Overall, our local study highlighted key learning points which could be of use in emergent safeguarding partnerships. Professionals need to continue to actively pursue and create opportunities to collect and collate comprehensive data and promote collaborative multi-agency arrangements. Panels need to be responsive to all partners involved in the safeguarding process, which includes parents. A level of reciprocity needs to be nurtured for safeguarding panel members and acute care providers to work in ways which promote learning, consider emotional support systems and explore ways to define and mobilize knowledge that can inform the safeguarding process and prevent future avoidable child deaths.

Keywords

Child death reporting child death review child mortality modifiable factors

Introduction

Working together interprofessionally, socially and organizationally to safeguard the health and well-being of children from preconceptual care to adulthood is critical within the United Kingdom, since we have the highest child mortality rate among the 15 pre-2004 countries of the European Union (HM Government, 2016; Wood Report, 2016; Wolfe et al., 2011). Furthermore, avoidable death rates from illnesses following presentation to acute care services (i.e. meningococcal infection) are higher in the United Kingdom than in Sweden, France, Italy, Germany and the Netherlands (Wolfe et al., 2011). The Confidential Enquiry into Maternal and Child Health (Pearson, 2008) ‘Why Children Die’ identified that ‘avoidable factors in 26% of cases’ with ‘potentially avoidable factors in a further 43% of cases’ (p. 47): for example, prompt access to knowledgeable professionals able to recognize serious childhood illness.

Local authorities (LAs) in England hold statutory responsibility to work in partnership with clinical commissioning groups throughout the child death alerting, review and reporting process (HM Government, 2018; NHS Clinical Commissioners, n.d.). Local Safeguarding Children’s Boards (LSCBs), exist within LAs, their scope includes assessing help provided to families, quality assurance practices, monitoring and evaluation of training as well as reviewing all child deaths, that is, where a death certificate has been issued (note still births and late foetal loss are excluded), that occur in their area (HM Government, 2013, 2018). Child death review (CDR) is achieved through the work of a child death overview panel (CDOP), supported by the child death review manager (CDRm) or administrator. The CDOP is a multi-agency group of professionals who review information about each child death (Allen et al., 2014).

Panels have a core membership drawn from key organizations including a range of health service providers, police, school representatives, education welfare service, safeguarding, legal services, public health and lay representation. Other agencies and/or specialists may be co-opted to contribute to discussions of certain types of child death when they occur. Panels systematically review and analyse all documentation of pertinent child death information to identify missed opportunities to take protective action (Douglas, 2016) through lenses focused on prevention, shared learning, transparency and system improvements (Fraser et al., 2014).

While the number of panels fluctuates, likely due to organizational changes, the Department of Education (DfE, 2017) cite all 148 LA returned data on behalf of 90 CDOPs (DfE, 2017). Changes to CDOPs since 2008 include merger of panels, thereby merging geographical areas and adoption of national guidance documents (DfE, 2015a). More recently, the Wood review (HM Government, 2017) and HM Government, Putting Children First (2016) were pivotal in shifting governance, with Health taking the overall mandate for safeguarding children and the CDOP process (HM Government, 2017). New overarching systems of child safeguarding practices, review process and panels must be operational by September 2020 (HM Government, 2018).

Whichever organization has leadership responsibility, there remain core areas linked to process, quality and learning at local and national levels (HM Government, 2017 see Figure 4 p. 23 & Figure 6 p. 28 for proposed outline from 2019 onwards). Each local CDR chair and designated doctor are supported by the role of a CDRm. The manager follows protocol in gathering multi-professional CDR information from key review meetings, such as perinatal mortality review groups to facilitate the methodical collection, assimilation, and analysis of data (using standard forms for notification (Form A), reporting (Form B) and analysis (Form C)) to inform the panel process (HM Government, 2018). Ideally within six weeks from the CDOP receiving the child death report, they meet, reviewed and scrutinize the data presented to them on behalf of local CDR partners (HM Government, 2018). The CDRm generates a child death annual summative report which may be available to the public online. Recommendations are also disseminated to operational level teams in health and social care with a view to local implementation.

National Health Service (NHS) England established strategic clinical networks to focus on priority areas with a view to building collaborations and bringing about improvement in the quality and equity of care for current and future populations (NHS Commissioning Board, 2012). As a part of the wider work program of the Children’s Clinical Network (CCN), the study team was commissioned to explore the complexity of issues that were believed to surround how child death reporting systems for children up to 18 years of age were working in practice across a local geography.

Building an understanding to inform the review

To better frame and more broadly understand the specific challenges faced by CDOPs and CDRms, we undertook a literature review that followed traditional PRISMA (Moher et al., 2009) guidelines using the search terms: Child death review, child death manage*, child death overview/panel, in the following electronic databases: Embase, Medline, Social Policy and Practice, Web of Knowledge, CINAHL and The Cochrane Library. Inclusion criteria were papers published in English in the United Kingdom and the United States, Australia and New Zealand between 2005 and 2015. The reason for including papers published in Countries other than the United Kingdom related to the notion that these systems have informed some UK changes (Bunting and Reid, 2005; Fraser et al., 2014). Figure 1 outlines the review process, and Table 1 presents key papers and findings that were selected for in-depth review.

Figure 1.

Flow diagram showing identified and included references.

Table 1.

CDOP literature review – Appraisal of included papers.

Authors/date	Aim/focus	Methods	Sample	Key findings	MMAT score (National Collaborating Centre, 2015)
Allen et al. (2014)	To establish how well panels work from the perspective of the paediatricians involved. To ascertain whether they deliver good value To identify areas for improvement.	Questionnaire survey: To every CDOP paediatrician in the country (n = 93).	84/93 paediatricians involved in CDOPs.	60 (71%) believe CDOPs are good value, 73 (87%) feel case discussions are rigorous and consistent, over 90% believe correct issues identified. Areas for improvement: 40 (48%) suggest devolving specialist deaths (e.g. neonates) to hospital-based review meetings or holding themed meetings with invited specialists, 11 (13%) suggest filtering out cases where learning is unlikely before full CDOP meetings and 13 (15%) called for national integration and analysis of data.	*
Devaney et al. (2010)	Evaluation of the process for reviewing child deaths where abuse of neglect is suspected, in one part of the United Kingdom.	Qualitative design: This was an evaluation of the case management review process using the Delphi technique.	Policy Delphi Technique’ with three phases. Phase 1, a semi-structured interview. Phase 2, a second iterative questionnaire constructed from the content analysis of phase 1. Phase 3, a third questionnaire modified to include individual and statistical responses from phase 2. Initial sample of 28 panel member ‘experts’ including Directors of Children’s Services, Senior officials from DOH, Social Services and DOE. 67% response rate (n = 20) overall.	A total of nine themes were evident in the data which overall indicated consensus among the expert panel that although the current system for reviewing non-accidental child deaths should be retained, more attention to process issues and the translation of learning into action is needed to improve the quality of the reviews.	***
Mazzola et al. (2013)	Sought to review local CDOP: How well they performed and how to improve effectiveness.	Mixed methods: Analysis of forms and data extraction to provide descriptive statistics. Semi-structured interviews with key stakeholders (appropriate to the process under review). Considered: Rapid response; admin/information gathering; panel itself; identifying modifiability and implantation of findings.	25 stakeholders were approached and 18 were interviewed. 208 notifications made between 1st April 2008 and 31st January 2011 were reviewed.	Main cause of death: perinatal and neonatal events (45) followed by chromosomal, genetic and congenital abnormalities (19), then acute medical or surgical. 69 – no modifiable factors, 28 modifiable actors. 4 distinct issues emerged: Issues with CDOP meetings themselves; problems with information sharing prior to meetings; long delays in completing cases; variability in classification from meeting to meeting; problems interpreting terminology; triage system needed to identify CDOP cases that might not need to come to panel. Implementing lessons learnt: Focus more on whether something modifiable or non modifiable rather than acting on lessons learned; current system has no clear guidance on whose responsibility it is to decide a functional or realistic plan for implementing change and how to measure it; no central guidance on prioritization of recommendations; need better sharing between CDOPs, regular seminars and updates. Logistical problems of gathering data: incomplete forms and duplication; neonatal information needs improving; need better liaison with coroner re-postmortems. CDR process as a whole; at time of writing, still a developmental process; good multi-agency review; greater commitment and awareness needed; time and resource demanding.	***
Pearson, ward-Platt, Kelly (2011) (Why children Die)	To describe the avoidable factors associate with child deaths identified by a confidential enquiry. To highlight examples of good practice.	Qualitative case study approach: Cases selected blindly, but in equal numbers from predetermined age bands.	119 of 126 reviews had sufficient information to determine avoidable factors. The cases were comparable with the whole CEMACH data set in terms of sex and cause of death (957 in total).	31 (26%) had avoidable factors that were predominantly related to individuals or agencies with direct responsibility to the child. 51 (43%) were defined as potentially avoidable. 130 factors identified as in relation to the above 82 cases, of which 64% were healthcare related. Panels disagreed with the medical certificate of cause of death in a third of the cases reviewed. The panel felt only 119 cases had sufficient information to reliably determine whether avoidable factors were present. They demonstrated a clear reticence to describe non-natural deaths as unavoidable. Conclusion: Confidential enquiries on a national scale provide epidemiological and qualitative evidence that will improve understanding of why children die and hence indicate strategies to reduce mortality.	**
Sidebotham et al. (2008)	To inform the introduction of the new child death review processes. Evaluated 4 components: basic systems, data collection, data analysis, outputs of the CDR process.	Mixed methods: Using elements of action research.	Phase I: Questionnaire to all LSCBNs in England 60/144 responses. Phase II: Audit tool development and audit of 9 sites by LSCBN chairs, Collection of protocols. Phase III: Non-participant observations of 7 CDOP meetings, semi-structured interviews with all CDOP chairs.	Extensive report showing variable practice. Only 3 areas had CDOPs already set up. Neonatal deaths reviewed in hospital services, replication not desired. Good working relationships identified as key to success of CDOP. Need to clarify relationship between CDO process and any other review processes identified. Identified need for CDR managers and administration teams. No audit or governance systems yet in place. No clearly developed systems for analysis of data. Concerns around data capture, storage and data sharing. Process should be focused on outcomes and a preventive agenda.	**
Sidebotham et al. (2011)	To observe and describe the experiences of CDOP in implementing new CDR processes and making prevention recommendations. To support the development of early starter CDOPs.	Mixed methods: Evaluation design incorporating a small initial, quantitative audit. The qualitative methods used were interviews, structured observations and an evaluation of documents.	Initial questionnaire was distributed to the chairs of all 144 LSCBs in England (42% response rate). Nine study sites selected from all LSCBs to reflect geography, population, ethnic composition and levels of deprivation. Following a preliminary audit to capture information about the existing status of the CDOP, observations took place at each CDOP and the chair of each CDOP took part in the interview. Documents collected were compared with the observational and interview data using a process of triangulation.	The initial audit described the demography of the sites. The results from the triangulation of the three main methods of data collection fell into nine core themes with two overarching categories; the systems and structures in place to support the CDOP process and the actual process and function of the CDOP. The importance of the CDR being a multidisciplinary process was emphasized. The process was time intensive in relation to panel meetings and preparation for panel meetings. The focus should be on learning lessons, not apportioning blame but clear structures and processes are needed to gather and use data effectively.	**
Vincent (2014)	To compare and contrast child death review processes in six countries (United States, England, Wales, Australia, New Zealand, Canada).	A qualitative, multiple case study approach, with a case being defined as a country (New Zealand, England, Wales), a state (Australia, United States) or a province (Canada). Each case comprised documentary analysis, semi-structured interviews and observation of CDR meetings (United States and England).	A multiple case study approach was taken with 18 case studies being selected across the six countries. The 18 case studies were made up of three countries, five states in Australia, eight states in the United States and two provinces in Canada. Interviews were conducted with over 100 key informants.	Variation within and across the six countries in terms of how CDR teams are organized and how/what is reviewed. Each case was analysed and cross-case similarities and differences, and universal, transcendent themes were identified. The structure of CDR processes may make little difference to how CDR findings influence prevention. Standardized data input processes and definitions facilitate effective CDR. Systematic data collection and reporting at both individual and national/state levels facilitates identification of major issues/trends to inform prevention initiatives. A public health model will also offer potential in terms of prevention.	***
Wirtz et al. (2010) USA	Aim to assess the quality of written recommendations in published CDRT reports. To provide guidelines for improving the quality and effectiveness of written recommendations. Getting the knowledge from CDRT in to practice.	A descriptive, non-experimental design analysed a set of 1093 recommendations from 21 randomly selected, publicly available state and local CDRT reports. An assessment instrument, modelled on the public health approach, was developed to score the quality of recommendations. It consists of three components divided into 10 dimensions.	214 reports (CDRT) – 79 met criteria and 21 chosen at random to review. Majority recommendation were preventative in nature.	Developed a guideline to help improve how written recommendations can be focused and level of authority is clear. Team leadership is critical to success.	***

Note: CDOP: child death overview panel; CDR: child death report; CEMACH: confidential enquiry into maternal and child health. National Collaborating Centre for Methods and Tools (2015). Appraising qualitative, quantitative and mixed methods studies included in mixed studies reviews: The MMAT. Hamilton, ON: McMaster University. (Updated 01 September 2017); retrieved from http://www.nccmt.ca/resources/search/232.

* level of confidence

Two salient learning points were identified from the literature review:

Ambiguity surrounding where neonatal deaths discussions should be presented, that is, at acute provider or CDOP level.

CDOPs at times were conflicted about how to bring into focus meaningful implementation that resulted in collaborative changes in primary and acute health care practice.

Multi-agency working is considered paramount in improving conveyance of information. Effective translation of CDOP findings to clinical areas influences opportunities for health promotion and prevention messaging. In summary, the literature review highlighted variation in practice, training and funding within, and across, different countries. Such variability appeared at times to lead to inconsistencies that many authors believed needed to be addressed if accountability for implementation were to happen within a collaborative partnership (Mazzola et al., 2013; Shanley et al., 2010; Vincent, 2014).

Aim

Lessons learned from the practice of CDR (between 2011 and 2014) within two geographical areas within the North West of England.

Ethics

Ethical approval for the study was granted by the University of Central Lancashire’s Science, Technology, Engineering, Medicine and Health Ethics Committee. No further approvals were needed as the study did not involve interviewing NHS patients and documents shared by participants were within the public domain.

Methods and analysis

We adopted a qualitative methods approach, involving documentary analysis and semi-structured interviews with key stakeholders (Bowen, 2009) to explore existing practice across the two geographical areas (herein referred to as area ‘A’ and area ‘B’) served by CDOPs, District General Hospital and tertiary hospital providers. These stages ran in parallel so as to directly address documentation queries with participants who were using these in practice.

Documentary analysis has been used within a framework of comparative analysis to explore policy implementation (Shaw et al., 2004). In trying to amass relevant documents used within child death reporting processes, we requested blank documents from a range of key stakeholders across areas A and B. It is worth noting that within England, all causes of child deaths (i.e. neonatal, accidental, deaths overseas of children normally resident in England, those in custody) are reviewed not just those linked to child protection systems, as happens in some other countries. At the micro level, collecting blank baseline forms served as a primary approach, review of local amendments to the standard DfE (2015b) forms helped the team to review, organize and present information using standardized, evidence-based approaches. At the macro level, policy documents can provide a level of public and social narrative (Fitzgerald, 2007). In combination, these documents were examined to explore the extent practice, at an organizational level, was being enacted. We gathered retrospective public data through connecting with those organizations considered key partners working with our two CDOPs for the period 2011–2014 and from open access sources. Of note, while CDOPs must prepare and publish reports, the timely nature or access to these may vary (HM Government, 2018). Analysis of documentary data was viewed as an empirical exercise that involved mapping the sources of information used, key outcomes and form recommendations identified variations along with the rationale for change or omission, into a matrix developed by the study team.

Audio-recorded telephone or face-to-face interviews using a semi-structured interview guide (Figure 2) was completed with professionals involved in the CDOP process. These included CDRms or administrators and public health consultants, who were asked to share their experiences in collecting and reporting information (Table 2). Interview transcripts were coded in QSR NVivo 10, a qualitative data analysis software programme, using a coding frame developed and agreed by the study team following reading and re-reading of the transcripts. One member added additional codes as new themes emerged during coding; these were checked and agreed with the remaining study team members. Constant comparative analysis was used to compare data within and across interview transcripts and interpreted within five emergent themes (Fram, 2013). The following section reports findings from the five themes: data collection, panel membership, family support, process challenges and panel impact and the summative documentary analysis.

Figure 2.

Participant semi-structured interview guide.

Table 2.

Characteristics of participants.

Participant number	Job	Role within CDOP	Length of experience in CDOP	Professional background
1	Consultant in public health	Chair	4 years	Public health
2	CDOP manager	Area manager	4 years	Child protection/safeguarding
3	LSCB administrator	Administrator	11 months	Children’s services administration
4	LSCB business manager	Panel member	4 years	Unknown

Note: CDOP: child death overview panel; LSCB: Local safeguarding children’s board.

Findings

The findings are separated into two areas. Firstly, we report our interview findings linked to data collection and policy. Subsequently, we briefly report on the technical knowledge from documentary analysis linked to child death characteristics and modifiable risk factors. Such factors inform how those involved in the CDR process situate their knowledge and preserve memory which supports networking and partnership building.

Interview findings

Data were collected from four participants despite numerous attempts to connect with others involved in the CDOP process. Interpretation of data led to five themes being developed:

data collection methods,

panel membership,

family support,

challenges in the child death overview process and

impact on child health outcomes.

The following section presents the findings in each of the five themes. At times participant’s own words are used, especially when conveying the emotive aspect of this work; names have been changed to maintain anonymity.

Theme 1: Data collection methods

Interviewees who had worked within the CDR process for many years recalled the change from using locally contextualized and developed documentation to national standardized forms. After using the new forms, participants recognized omissions that appeared to constrain discussion at CDOPs, resulting in additions to the national form. Participants reported a need to include the following: antenatal care, infant-feeding methods, maternal body mass index (BMI) for neonatal deaths and bereavement support (later added to the national form; see Table 3). A shift from paper to electronic forms made collation of the reports easier for the CDRms. Yet despite online systems, the CDRm reported that considerable effort was needed to remind child death reporters (all agencies requested to complete the forms) to submit documents in a timely manner. Time for the completed report to reach the panel was further constrained when incomplete documentation was returned. Since the CDR report is the collation of reports from various professionals, the CDRm would often spend considerable time ensuring that reports were sufficiently detailed to minimize the subjective nature of reports presented to the CDOP. Each report was said to take between 2 and 18 months to complete, depending on whether any internal or serious case reviews, criminal investigations, complex investigations or an inquest was required; all of which need to be completed before the CDOP report is collated. Four months was reported to be the average time taken to generate a report. No national standard has been set for time to report to panel.

Table 3.

Availability and amendments to CDOP standardized forms.

	Area A	Area B	Comment
Notification of Child Death Form A	Yes	Yes Additions: gender, ethnicity, travelling family, asylum status, legal status, child protection plan, section 20/section 31. Designation of reporting agency, has the death been STEIS reported	Form A (consist of three sub-sections) Child Death Notification Form – incorporates the child’s details, referral details and details of the death
Agency Report Form B	Yes – however instructions to users were omitted re supplementary forms. Addition: Agency report requesting specific details on bereavement support	Yes Additions: Child’s unique CDOP number, residential ward, bereavement support offered to family. Booking in, antenatal care, feeding Hx, Mothers BMI, Updates from formal review/internal review, MPMN Form and upload to the MPMN portal	Form B (six sub-sections) Agency Report Form – incorporates the child’s details, provider agency details, a summary of the case and the circumstances leading to the death, in-depth child’s details (e.g. birth weight, disabilities, ethnicity), family and environment details, parenting capacity and details of any service provision
CDOP reports Quarterly reports	Yes Eight: April 2013 to March 2014 and April 2014 to March 2015	Yes Eight: April 2013 to March 2014 and April 2014 to March 2015
Annual reports	One annual CDOP report (April 2013 to March 2014) (while 2 annual reports pre 2013 were collected no quarterly reports were available therefore these were excluded as they did not represent all areas post 2013)	One annual CDOP report (April 2013-March 2014) (Earlier report available but no corresponding quarterly reports; resulted in exclusion from comparison)	Public domain
Supplementary Forms	Not available	Yes	Forms B2–B12 Supplementary Forms – these forms are to be completed where appropriate and provide details on the type of death (e.g. neonatal death, sudden unexpected death in infancy, road traffic accident, drowning)
Form C	Not available	Yes	Form C Analysis Proforma – this form is completed by the CDOP to evaluate the information received regarding the child’s death, categorize the cause of death and identify any lessons and modifiable factors. This form also facilitates understanding at a national level

Note: CDOP: child death overview panel; STEIS: Strategic Executive Information System; BMI: body mass index; MPMN: maternal and perinatal mortality notification.

Theme 2: Panel members

Participants reported that CDOPs aimed to include representation from all professional groups so that experts in relevant fields reviewed all aspects of the child’s death. Combining smaller panels meant more consistent specialist representation, since professionals used a rota system to allocate attendance. The professional background of the CDOP chair varied, with one area valuing the diversity a public health background could bring. Furthermore, the role of the chair could also include running briefing sessions and training events. The CDRm role was diverse that is, generating reports, progressing actions, outcome mapping to identify trends over time and working with sub-groups such as the critical incidents group. One participant celebrated how fostering collaboration between national CDRms and CDOPs had resulted in early dialogues about trends as either localized or generalized issues.

Theme 3: Family support

The involvement of the CDOP team in providing support to bereaved families was variable with one participant suggesting ‘that’s not the particular focus for one of the chairs’ (ID02). Although access to support was believed to be important, variation meant that this could range from provision of a designated bereavement support resource to providing a leaflet. Families were informed about the CDOP process by letter, which included a reply slip providing options for feedback. In one area, only five families were said to have returned these slips during the time the CDRm had been in her role. This experienced CDRm remained proactive in seeking parent inclusivity and reported that she had:

Gone out to see probably about four or five, six families over [all these] years because that’s all that’s requested it. (ID02)

A few families were also reported to have contacted the CDRm by phone. Families who complained were sometimes thought to:

Have a rough ride because they think if they start complaining about things they are viewed as stroppy parents rather than devastated parents as a result of their bereavement. (ID02)

Theme 4: Challenges in the CDR process

Participants identified facilitators and inhibitors to the collection of comprehensive data. Facilitators included the introduction of an electronic system, briefing sessions for reporting agencies, sharing of good practice such as examples of anonymized completed reports, staff training and the development of a tracking log for communications with reporting agencies and professionals. Inhibitors were primarily linked to missing data, which include fear of litigation which was thought to hinder the sharing of agencies internal reviews into child deaths, worries about confidentiality from General Practitioners and the police, missing case notes particularly from maternity services or for families who had moved or lived out of the area at the time of the death. Inevitably, in this time of financial constraints, participants talked about the impact of service reorganization on the ability to ensure comprehensive representation on the CDOP as the demands on individuals’ time within their jobs had increased. Maintaining consistent staffing of the CDR service itself was also reported to be difficult at times, creating additional strain on the system.

Theme 5: Impact on child health outcomes

One respondent talked about the role of the Office for Standards in Education (Ofsted), an independent inspection agency in the United Kingdom, in challenging CDOPs to identify the impact of their review processes on children and families. This was interpreted as meaning the actions taken as a result of the review process to reduce future child deaths, such as running campaigns on safe sleeping, ‘training health visitors and children’s centre staff’ (ID01). Further examples included publications such as preventing children swallowing batteries (modifiable factor) in agency newsletters and requesting audit data around prevalence of Caesarean deliveries. At times, panels were reported to struggle to specifically identify single learning points. This happened when there was uncertainty but a likely link to the child death resulting in outcomes being labelled as ‘an issue identified’ (ID02) rather than as a modifiable factor. Sometimes, the issues identified concerned physical safety in local areas and in these cases LA were contacted to raise awareness in the hope that action would be taken to remove the risk.

Documentary findings

Secondary documentary analysis from data across the participating sites is summarized in Table 3. Data spanned a three-year period during which 104 child deaths occurred. Twenty-two percent of deaths (n = 23) were reported as having modifiable factors which included population health messaging, that is, safe sleeping, environmental signage; early access to health providers, equipment and care pathways, that is, timely diagnosis, correct equipment use; and comprehensive risk assessment, that is, substance use, violence.

While CDR reports highlighted modifiable factors, difficulties in analysis lay in variation of the scope and depth of information presented, comparability between LSCBs and a lack of evidence of how learning from CDR had translated to clinical contexts. For example, the need for consistent clinical reporting of fluctuations in mother’s BMI in perinatal care was a modifiable factor linked to risk of child death and co-sleeping. Indeed, maternal weight, family status and lifestyle, that is travelling family status, were additional data points added to the mandatory reporting forms. Understanding how these translated from the CDR report to dialogues with providers was not evident in evaluation of the CDOP process. However, it is the interview data that provide insight since the CDRms spoke about the importance of panel interactions, building networks and relationships as mechanisms to disseminate knowledge.

Discussion

The need to acknowledge collective accountability in the CDOP process is imperative, given the imminent shift in organizational and operational responsibility (Wood, 2016). Lessons learnt from our small review of local practice are echoed by the recommendations from the Wood (2016) report. In the light of the planned changes, we recognized we had a responsibility to highlight the ongoing challenges that must be addressed to strengthen the CDOP process as it moves forward in the next decade.

Our small review highlighted the complexity of reporting and documentation, noting the variation in how baseline data collection forms had been adapted, with additions, for local use. The additions were not meant to complicate nor duplicate work but tried to ensure that timely data were collected at the first point of contact to limit repeated multi-agency requests. While the amount of time needed to complete analysis varied, the outputs were highly valued as tangible ways to direct future knowledge mobilization and education efforts (Featherstone et al., 2018). Existing barriers that inhibit timely form completion must be addressed as the CDR process changes. For example, the Child Safeguarding Practice Review Panel may wish to undertake national review (HM Government, 2017, 2018; Wood, 2016) which will require discussion with the LSCBs, who will need to have completed data to share. Such pooling of data has the capacity to inform deeper understanding linked to child death (Garstang, 2018). Therefore, comprehensive, trustworthy and contextualized data must be linked to CDOP information processes (Allen et al., 2014; Mazzola et al., 2013). A recent study by Gijzen et al. (2016) identified that time and resource required in reporting could be significant, which has limited the adoption of CDOP in the Netherlands. Our study identified that time was a constraint for all those involved in the CDOP process.

The form additions (i.e. linked to ethnicity factors) resulted in information which, once analysed identified trends and risks, thereby enhancing the quality of the reporting process as evidenced by Firth et al. (2018). Therefore, ongoing consideration as to the quality, transparency and consistency in reporting and documentation will support future safeguarding partners to consider previous data inclusion and decisions processes (Children Act, 2004; Children and Social Work Act, 2017).

The need to build reciprocity between professionals and parents/caregivers is critical to the success of the CDR process, as it transitions. In acknowledging that outcomes should be the primary focus in the process of managing and mapping child death the mechanisms of how this happens must not be forgotten. Representation, joint-agency collaboration, professional/disciplinary knowledge and skills are pivotal in how the CDR process has functioned at local and Pan-local levels. Such knowledge should not be obscured at this time of transition. However, the position, role and place families had in the CDR process was limited within our study. Hopefully, early support following bereavement has taken place to avoid parents becoming ‘lost’ in what has been described as the ‘parallel investigations following their child’s death’ (HM Government, 2017: 26). The time between the child’s death and CDOP review can be lengthy which could leave parents with uncertainty about the process. Via existing CDR notification processes, the primary care team, that is, health visiting/school health, could be made aware that the child’s death was to be reviewed at the CDOP. This distanced objective nature of the CDR process, although conducted by professionals in a professional manner, may have left some parents feeling the system was impersonal or that (re)engagement was emotionally difficult. As such, the emotional impact of child death ‘work’ cannot be underestimated for either parents or professionals within this process (Gijzen et al., 2016). The visiting in the family home and telephone follow-up approach identified by one CDRm was admirable, which happened as a result of the participant working in this area for many years. Although not reported in the findings, one of the authors of this article is aware of locally funded psychological support and supervision sessions available to staff working in CDR.

In the light of the new changes, the purpose of the CDR process will remain: to identify and disseminate factors that are potentially avoidable or modifiable thereby positively influencing future child health outcomes (DfE, 2015a). Similar to Covington and Johnson (2011), our review identified times when knowledge translation and implementation (i.e. linked to health promotion and prevention) was difficult. Therefore, it was unclear whose responsibility it was to translate recommendations from CDOPs into policy and practice and then to audit impact. Infant and child deaths are sensitive metrics that reflect population health as well as social learning (Seske et al., 2016); therefore, audit provides an indicator of quality in this process. Challenges in sharing CDOP messages that could influence child health outcomes with professionals, organizations, agencies as well as individuals continue to be difficult despite technology and the perceived level of connectedness this has brought, see for example, Box 1. Within a framework of keeping our children safe (NSPCC, 2016), using outputs from CDOP could inform those working directly with families no matter what their individual affiliated professional stance, agency or charitable position. It remains unknown whether local charity representation at CDOP or collaborative working with such agencies, as suggested in the King’s Fund Patients as Partners model (2016), could be a platform for shared messaging.

Box 1.

Example of the how collecting and sharing metrics could raise patient safety issues.

For example, some children had had contact with agencies during their lives, others had not, reflecting the findings from Stallard et al.’s (2016) study focusing on suicide in youth. The addition of previous suicide attempts within the data gathering stage, while informative, needs to recognize that not all youth attempting suicide will be NHS users; Stallard et al. (2016) identified that 56% of these youth were unknown to specialist child and mental health services. Bringing these UK data sets together highlights the need to integrate work in this area, perhaps in a repository that can be used as an informative data source or linked to national drivers such as Clinical Networks (NHS Commissioning Board, 2012)

Conclusions and lessons for the future

In thinking about the future for CDOP and what lessons could be taken from our review, we make the following recommendations:

Additions to standard forms that help to contextualize local issues may have utility in wider practice and research settings. Early reporting of data as an interim report could be agreed by CDOPs when there are risk factors that warrant urgent attention, that is, may result in child injury.

Panels and LSCBs also have a responsibility to promote the use of standardization of bereavement support such as the Lullaby Trust website (http://www.lullabytrust.org.uk/child-death-review) and Child Bereavement Trust (http://www.childbereavementuk.org/).

A level of reciprocity needs to be nurtured for local safeguarding panel members and acute care providers to work in ways which promote learning, consider emotional support systems and explore ways to define and mobilize knowledge that can inform the safeguarding process and prevent future avoidable child deaths.

LSCBs provide leadership locally and have a national voice. Therefore, these boards have a responsibility to help advocate for change using guidance from the CDOPs.

CDR partners need to remain engaged during this time of transition, actively pursuing new opportunities that promote collaborative multi-agency arrangements, without the loss of organizational learning or wisdom. This is especially important since we were not assured that findings from panels actually brought about the hoped-for systematic changes in practice and policy locally, regionally or nationally.

Adequate input from families or family representatives such as charities needs to be considered. Panels could explore the feasibility and potential benefit of patient and public involvement and/or relevant local charitable representation.

There are some examples of good local practice, for example, collaborative working, that is, ‘safe sleep’ but duplication is common. In times of austerity, national strategies should be developed, coordinated and shared with CDOPs as well as other organizational structures such as Health and Wellbeing Boards.

Footnotes

Declaration of Conflicting Interests

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Cheshire and Merseyside Maternity, Children and Young People Strategic Clinical Network.

ORCID iD

Caroline Sanders

References

Allen

Lenton

Fraser

et al. (2014) Improving the practice of child death overview panels: a paediatric perspective. Archives of Disease in Childhood 99(3): 193–196.

Bunting

Reid

(2005) Reviewing child deaths: learning from the American experience. Child Abuse Review 14(2): 82–96.

Bowen

(2009) Document analysis as a qualitative research method. Qualitative Research Journal 9(2): 27–40.

CEMACH (2008) Confidential Enquiry into Maternal and Child Health: Why children die? A Pilot Study. England (South West, North East & West Midlands), Wales and Northern Ireland. London: National Children’s Bureau. Available at: http://www.rcpch.ac.uk/news/confidential-enquiry-maternal-and-child-health-cemach-why-children-die-pilot-study-2006-14-may (accessed 5 April 2019).

Children Act (2004) Children Act 2004. Legistaion.gov.uk Retrieved from https://www.legislation.gov.uk/ukpga/2004/31/contents (accessed 5th April 2019).

Children and Social Work Act (2017) Children and Social Work Act 2017. Legistaltion.gov.uk. Available at: http://www.legislation.gov.uk/ukpga/2017/16/contents/enacted (accessed 5 April 2019).

Covington

Johnston

(2011) A misdirected assessment of progress in child death review. American Journal of Preventive Medicine 40(5): e31.

Department of Education (DfE) (2017) Child death reviews: year ending 31st March 2017. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/627206/SFR36_2017_Text.pdf (accessed 5 April 2019).

Devaney

Lazenbatt

Bunting

(2010). Inquiring into non-accidental child deaths: reviewing the review process. The British Journal of Social Work 41(2): 242–260. https://doi.org/10.1093/bjsw/bcq069

10.

DfE (2015a) Working together to safeguard children. A guide to inter-agency working to safeguard and promote the welfare of children. Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/419595/Working_Together_to_Safeguard_Children.pdf

11.

DfE (2015b) Child Death Reviews: forms for reporting child deaths. Available at: https://www.gov.uk/government/publications/child-death-reviews-forms-for-reporting-child-deaths

12.

Douglas

(2016) Child Death Review Teams. Chapter 5. In Douglas

Emily M

(ed). Child Maltreatment Fatalities in the United States. Netherlands: Springer, pp. 71–89.

13.

Featherstone

Leggett

Knisley

et al. (2018) Creation of an integrated knowledge translation process to improve pediatric emergency care in Canada. Health Communication 33(8): 980–987.

14.

Firth

Petherick

Oddie

(2018) Infant deaths from congenital anomalies: novel use of the child death overview panel data. Archives of Disease in Childhood 103: 1027–1032.

15.

Fitzgerald

(2007) Documents and documentary analysis: reading between the lines. In: Briggs

ARJ

Coleman

(eds), Research Methods in Educational Leadership and Management, Chapter 17. London: Sage, pp. 278–294.

16.

Fram

(2013) The constant comparative analysis method outside of grounded theory. The Qualitative Report 18(1): 1. Available at http://nsuworks.nova.edu/cgi/viewcontent.cgi?article=1569&context=tqr (accessed 5 April 2019).

17.

Fraser

Sidebotham

Frederick

et al. (2014) Learning from child death review in the USA, England, Australia, and New Zealand. The Lancet 384(9946): 894–903.

18.

Garstang

(2018) Challenges o research using data from child death overview panels: Archives of Disease in Childhood 103(11). DOI: 10.1136/archdischild-2018-314821.

19.

Gijzen

L’Hoir

Boere-Boonekamp

et al. (2016) Stakeholders’ opinions on the implementation of Child Death Review in the Netherlands. BMC Research Notes 9(1): 1.

20.

HM Government (2017) Child Death Review. Statutory Guidance. Available at: https://consult.education.gov.uk/child-protection-safeguarding-and-family-law/working-together-to-safeguard-children-revisions-t/supporting_documents/Child_death_review_stat_guidance.pdf (accessed 5 April 2019).

21.

HM Government (2018) Child Death Review. Statutory and Operational Guidance (England). Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/747995/Child_death_review_statutory_and_operational_guidance_England.pdf (accessed 5 April 2019).

22.

HM Government (2016) Putting Children first: delivering our vision for excellent children’s social care. Available at: https://www.gov.uk/government/publications/putting-children-first-our-vision-for-childrens-social-care (accessed 5 April 2019).

23.

HM Government (2013) Working together to safeguard children. Available at: https://www.gov.uk/government/publications/working-together-to-safeguard-children–2 (accessed 5 April 2019).

24.

Mazzola

Abdu Mohiddin

Ward

et al. (2013) How useful are child death reviews: a local areas perspective. BMC Research Notes 6: 295.

25.

Moher

Liberati

Tetzlaff

et al. The PRISMA Group (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med 6(7): e1000097. Available at: https://doi.org/10.1371/journal.pmed.1000097 (accessed 5 April 2019).

26.

NHS Clinical Commissioners (n.d.) About CCG’s. Available at: https://www.nhscc.org/ccgs/ (accessed 5 April 2019).

27.

NHS Commissioning Board: A special health authority (2012) The Way Forward; Strategic Clinical Networks. Available at: https://www.england.nhs.uk/wp-content/uploads/2012/07/way-forward-scn.pdf (accessed 5 April 2019).

28.

NSPCC (2016) How safe are our children. Available at: https://www.nspcc.org.uk/globalassets/documents/research-reports/how-safe-children-2016-report.pdf (accessed 5 April 2019).

29.

Pearson

(Ed) (2008) Why Children Die: A Pilot Study 2006; England (South West, North East and West Midlands), Wales and Northern Ireland. London: CEMACH. Available at: https://www.publichealth.hscni.net/sites/default/files/Why%20Children%20Die%20-%20a%20pilot%20study%202006.pdf (accessed 5 April 2019).

30.

Pearson

Ward-Platt

Harnden

et al. (2011). Why children die: avoidable factors associated with child deaths. Archives of disease in childhood 96(10): 927–931. DOI: 10.1136/adc.2009.177071.

31.

Seske

Muglia

Hall

et al. (2016) Infant mortality, cause of death, and vital records reporting in Ohio, United States. Maternal and Child Health Journal 21(4): 727–733.

32.

Shanley

Risch

Bonner

(2010) U.S. child death review programs: assessing progress toward a standard review process. Am J Prev Med 39(6): 522–528.

33.

Shaw

Elston

Abbott

(2004) Comparative analysis of health policy implementation: the use of documentary analysis. Policy Studies 25(4): 259–266. DOI: 10.1080/0144287042000288451.

34.

Sidebotham

Fox

Horwath

et al (2008). Preventing childhood deaths: a study of ‘early starter’ child death overview panels in England. Retrieved from https://webarchive.nationalarchives.gov.uk/20081105160515/; https://webarchive.nationalarchives.gov.uk/20081105160515/http://www.dcsf.gov.uk/research/data/uploadfiles/DCSF-RR036.pdf (accessed 5th April 2019).

35.

Sidebotham

Fox

Horwath

et al (2011). Developing effective child death review: a study of ‘early starter’ child death overview panels in England. Injury Prevention 17(Suppl I): i55–i63. Doi: 10.1136/ip.2010.027169.

36.

Vincent

(2014) Child death review processes: a six-country comparison. Child Abuse Review 23(2): 116–129.

37.

Wirtz

Foster

Lenart

(2011) Assessing and improving child death review team recommendations. Injury Prevention 17(Suppl I): i64–i70. Retrieved from https://injuryprevention.bmj.com/content/17/Suppl_I/i64.info (accessed 5th April 2019).

38.

Wolfe

Cass

Thompson

et al. (2011) Improving child health services in the UK: insights from Europe and their implications for the NHS reforms. BMJ (Clinical Research ed.) 342: d1277.

39.

Wood

(2016) Wood Report of local safeguarding children boards. Department of Education. DOI: org/10.1080/0144287042000288451.