National Perinatal Epidemiology Centre - Doctoral Theses
Permanent URI for this collection
Browse
Recent Submissions
Item Data quality in the evolving digitised health service(University College Cork, 2021) McKernan, Joye; Greene, Richard A.; Corcoran, PaulBackground/Objective: The research undertaken for this thesis focussed on data quality in the evolving digitised health service. In Ireland we all need to have our details on an electronic healthcare record. We need to have fully integrated systems documenting our health information across our whole life. We as patients need to be central to our care and have access to our data. EHRs can change healthcare by saving money, improving communication, and reducing errors. The introduction of an EHR is a substantial change management project that needs to include all stakeholders to ensure success. It requires vision, dedication, time, and patience. The power and importance of data cannot be overemphasised; we need to analyse what is required from data, using robust standard approaches, and ensure data is of high quality so that it can be used to improve patient outcomes and improve staff working conditions. The aim of this research project was to focus on aspects of digitisation that go towards achieving a high-quality data repository. We aimed to investigate the development and use of an EHR in the Irish healthcare system with specific consideration to the elements that impact data quality. We examined the experiences of the development team, patients, staff, service culture and the data collected. Methods: We used both quantitative and qualitative methods; this mixed method approach allowed for a deeper understanding of the issues. A document analysis of the closure report of the implementation of the EHR (MN-CMS) from the national project team was supported with discussions with team members. Patients at antenatal booking visits in an Irish maternity unit were invited to participate and complete a survey with respect to digitization of their health data. The survey was divided into three distinctive sections; participant information, regarding the staff encounters on their visits and questions about the new system. To engage with staff a pre- EHR implementation survey, a post EHR implementation survey and a post-implementation EHR documentation audit was carried out. A four-step approach was required when applying a national framework to a national data set. The four steps included a literature review, using elements of a data quality framework to develop the planning of an audit tool, data quality assessment of the Major Obstetric Haemorrhage (MOH) audit dataset. The fourth step assessed the data quality using the five dimensions of quality: (1) relevance, (2) accuracy and reliability, (3) timeliness and punctuality, (4) coherence and comparability, and (5) accessibility and clarity. To explore data quality in an EHR two phases were used; initially we examined the data from year 1 (2018); following analysis of the data set we found data quality issues. We then enacted an intervention and assessed the effect of a new data quality process. The intervention was to introduce a data quality resource to assess the datapoints within 1-2 days after documentation of the care by the healthcare professional. We assessed clinical data extracted from the MN-CMS national database for missing data and then examined the significance of the data issues. An ethnographic study approach was used to explore service culture around shift clinical handover, the process was divided into three components: an observational study, a short staff survey and a cause-and-effect assessment. Results: This project showed that several factors, need to be explored to fully understand data quality in healthcare. There is a growing need for high quality clinical ‘Big Data’ to measure, enhance and evaluate healthcare; clinical data systems need to be producing high quality complete and accurate data for primary and secondary use. Patients want to have access to their records and want to engage with healthcare professionals in their care. This engagement will lead to patients having more control over their health outcomes. EHRs are now becoming more and more widespread globally; in Ireland the Maternal & Newborn Clinical Management System (MN-CMS) has been implemented for four maternity units and is a pathfinder EHR project. It is a clinically led, patient centred EHR. Staff engagement is required for the implementation phase; they are a vital component to ensure a successful implementation. Staff may require additional training to ensure their documentation positively impacts data quality. There is a requirement to standardise terminology in relation to data quality and use data quality frameworks to assess the dimension of data quality. It is meaningful and useful to apply national data quality frameworks to data sets to investigate where improvements may be made. Capturing and ensuring quality data from an EHR takes time and resources; the data needs to be examined for accuracy and completeness. Resources in the form of staff are required to achieve this impact on data quality. They can improve data directly and more importantly they can engage with staff regarding their documentation, identify need for further training, technical solution changes and indeed review of data points and the value of recording them. Following the implementation of an EHR, workflows and practices might not change when they should have; it is important to explore why these changes may not occur and address the issues to identify the barriers and allow enablers to achieve appropriate change, engaging with staff in the process. Conclusion: This project aimed to explore the impacts of digitizing healthcare documentation on the quality of that data, examining the impact through patients, staff, and processes. This thesis has shown a need to move towards standardised terminology and methodologies to achieve these goals and the projects involved took a practical solutions approach. We have shown the importance of staff members and their role in the success of the project implementation. We have highlighted the importance of the use of frameworks to robustly assess data quality. There is growing literature regarding EHRs and data quality with the rapid expansion in digitization of healthcare data. This thesis adds to that literature, but significantly more work is needed in the areas of standardisation of data quality frameworks, the importance of staff in data quality, and co-designed patient portals.Item Study of methods, systems, recommendations and bereaved parents’ involvements in perinatal death reviews, inquiries and audits(University College Cork, 2021-12-10) Helps, Änne; O'Donoghue, Keelin; Leitao, Sara; Greene, Richard A.; National Perinatal Epidemiology Centre; University College CorkBackground: An estimated 5.3 million perinatal deaths occur worldwide each year. In Ireland, there were 335 perinatal deaths reported in 2019. These deaths are devastating for the parents, families and, if unexpected, for the healthcare staff involved, with long-lasting emotional consequences. Some of these deaths are unavoidable, but many are preventable. To investigate these deaths and identify contributory factors, local hospital-based perinatal death reviews as well as national perinatal mortality audits are carried out. In certain circumstances, for example if a higher than expected intrapartum perinatal death rate is recorded, an external inquiry may be commissioned to investigate events of public concern. Reports with recommendations are published after local perinatal death reviews, perinatal audits and external inquiries. In Ireland, there is currently no standardised format to the recommendations or their implementation. Further, the involvement of bereaved parents in local maternity hospital-based perinatal death reviews is poorly explored. The aim of this thesis is to analyse the methodology and structure of perinatal mortality audits, local reviews and inquiries, as well as recurrent themes in the recommendations of the published reports and the inclusion of bereaved parents in reviews. Methodology: Both qualitative and quantitative methods were employed for this thesis. A topic can be explored with flexibility and in depth by using a mixed methods approach. An integrative literature search was carried out focussing on the types and evolution of perinatal mortality audits and reviews in high-income countries. Further, an integrative review using quantitative and qualitative methods to identify established national perinatal mortality audits in four high-income countries and national initiatives addressing recommendations from these audits was done. Content analysis of the audits’ recommendations was performed organising them into themes according to topics covered. Additionally, a service evaluation of the local maternity hospitals’ perinatal death reviews in Ireland was carried using an electronic survey. The quantitative and qualitative data collected from all 19 maternity units were analysed to identify and compare current local review processes. In the analyses of the ten Irish inquiry reports relating to perinatal deaths and pregnancy loss services in the maternity services quantitative and qualitative data were collected by two clinicians using a specifically designed review tool. Descriptive analyses of the main characteristics of the reports gave an overview of the terms of reference and inquiry review process, and identified recurring themes in the recommendations. Qualitative content analysis of the reports’ findings and recommendations was used to identify key domains. An inductive thematic analysis with a semantic approach following the steps of familiarising, coding, identifying, grouping and revising themes identified the main themes and subthemes for each domain. Lastly, purposeful sampling was used to recruit bereaved parents in Ireland to take part in semi-structured interviews to examine how parents may be appropriately involved in the local hospital-based review in a way that is beneficial to them and the review process itself. Reflexive thematic analysis using a five-phase process (familiarisation, open coding, generating themes, developing themes, refining themes) was carried out on the collected data by three researchers. Results: Internationally, differences in perinatal mortality classifications, audits and reviews, as well as barriers to the implementation of recommendations were noted. Common and recurrent themes of recommendations from four established national perinatal mortality audits suggested a lack of progression of recommendations that is shared between countries. These four countries have adopted varying national initiatives and programmes to address the audits’ recommendations. A lack of standardisation for the methods of local perinatal mortality reviews and external inquiries in Ireland was highlighted within this thesis. Recommendations from ten inquiry reports were numerous and repetitive suggesting a lack of clear ownership for the implementation process. An analysis of the findings of the ten inquiry reports showed that that elements of governance of Irish maternity services (workforce, leadership, management of risk, work environment) impacted negatively and directly on the management of perinatal deaths and bereavement services. Further, three elements (hospital oversight, national documents, data collection) identified from the inquiry reports in turn affected governance structures in the management of perinatal deaths. Examination of these inquiry reports highlighted shortcomings in the perinatal bereavement care and pregnancy loss services provided to families in the Irish maternity services and the short- and long-term effects this can have. Interviews with bereaved parents revealed that parents want a more inclusive and open process that allows them to be included in the local hospital perinatal mortality review. However, this parental involvement needs to be carefully considered, flexible and appropriately resourced. Conclusion: The culture in the maternity unit determines how bereaved families and hospital staff cope after an adverse event like an unexpected perinatal death. A lack of open disclosure can have negative effects on how bereaved parents process events and cope with their grief after the death of their baby. Recently many reports with multiple recommendations have been published to improve safety standards in the Irish maternity services; however, implementation thus far has been slow and incomplete. The focus is currently on collecting data and highlighting issues, and less on progressing recommendations to implement changes and prevent similar events recurring. To overcome barriers to successful recommendation implementation and advance perinatal mortality audits and reviews, suggestions based on examples from the international literature were identified and provided as part of this thesis. Perinatal mortality processes, including reviews, need to be standardised across the 19 maternity units. Suggestions to achieve this include the adaptation of the national Incident Management Framework specifically to the maternity setting, the implementation of an electronic review tool such as MERT (Maternity Event Review Tool) for perinatal deaths and an assessment of the feasibility of a national perinatal (and/or paediatric) Coroner for Ireland. The inclusion of parents in perinatal mortality reviews needs to be addressed urgently yet carefully considered and resourced, in order for it to be beneficial to them and the review process itself. A collaborative process between staff and parents can highlight clinical areas in need of change, enhance lessons learned, and may prevent future perinatal deaths.Item An exploration of miscarriage in the Republic of Ireland: incidence, management, risk factors, interventions, and populations’ knowledge(University College Cork, 2019-08) San Lázaro Campillo, Indra; O'Donoghue, Keelin; Meaney, SarahBackground: Miscarriage is one of the most common complications in early pregnancy. It is estimated that approximately one out of five women will have a miscarriage throughout their reproductive lives. Despite the high prevalence of miscarriage and the biopsychological burden associated with experiencing miscarriage, there are several gaps in the literature. For example, there is a lack of standardisation of definitions and types of miscarriage worldwide. This high heterogeneity in cut-offs for defining miscarriage is limiting international comparisons of the evidence available. This is distorting the recording of data related to miscarriage in national and international health databases. Furthermore, little is known about the trends of hospital admissions for miscarriage and the non-fatal complications associated with it. In fact, there is no sufficient evidence of the validity of diagnosis of miscarriage in routinely collected health databases. In addition, although approximately 50% of miscarriages are linked to chromosomal abnormalities, the underlying causes of miscarriage are still unclear for the remainer. Therefore, it is imperative to understand and identify causes and risk factors of unexplained miscarriage in order to develop effective treatments and promote healthy behaviours among the population. The most well-established risk factors for miscarriage are advanced maternal age, previous pregnancy loss and parity. However, there is a need to identify risk factors in order to be able to prevent the likelihood of experiencing miscarriage. It is accepted that women who experience miscarriage suffer from psychological morbidity after the loss and in subsequent pregnancies. Nevertheless, further research is needed in order to obtain robust evidence on what specific group of women are more susceptible to develop psychological morbidity after miscarriage, what are the psychological and emotional changes during pregnancy after a miscarriage, and what are the effective non-pharmacological interventions to improve psychological wellbeing as well as future pregnancy outcome. Outline and aims: In this thesis, I explored several dimensions surrounding the event of miscarriage. To do that, I firstly reviewed the published evidence to date about miscarriage in order to find gaps in the literature. This thesis encompassed a total of six research studies to contribute to the existing body of knowledge about miscarriage. The main objective of the first study was to determine national trends in incidence and management of inpatient admissions for early miscarriage in the Republic of Ireland. After this study, it was essential to validate the diagnosis of miscarriage in the national health system used to obtain these trends. Consequently, the aim of the second study was to compare agreement for the diagnosis of miscarriage between three types of routinely collected hospital-based health records. This thesis includes three research studies that explored several gaps in the literature about pregnant women with a history of miscarriage. The first study explored the risk factors associated with miscarriage among women attending an early pregnancy assessment unit (EPAU). The second study aimed to determine barriers and facilitators when designing large-scale longitudinal studies; and the third study was a systematic review, which aimed to identify randomised controlled trials that assessed the effect of interventions to reduce stress, anxiety and depression in pregnant women with a history of miscarriage. Finally, this thesis includes a cross-sectional study that was designed to assess university student’s knowledge of basic reproductive health information about miscarriage. Findings and clinical implications: This thesis provides additional evidence to the growing body of work focusing on miscarriage. This thesis highlights the need for unifying inpatient and outpatient data in order to estimate the total burden of miscarriage at a national level. Furthermore, it is crucial to standardise the diagnosis of the type of miscarriage at a national level. The results presented in this thesis also emphasise the misunderstanding of causes, signs and symptoms of miscarriage, which shows it is essential to inform the public about miscarriage in general, as well as its treatments and the scientific evidence available to date. In addition, reproductive health information about miscarriage should be disseminated to a younger stratum of the population, who are at early stages in their reproductive life. Indeed, this would enable better informed decision-making about their reproductive behaviour and lifestyle by helping them to be aware of risk factors for miscarriage, identifying signs and symptoms of miscarriage and learning what to expect when experiencing a miscarriage. Moreover, providing reproductive health information about miscarriage will help the population to be aware of when and where seek for help. In this thesis, I suggest University settings as the ideal scenario to reach and promote reproductive health information about miscarriage in this targeted group. Efforts to satisfy the population’s needs on reproductive health and pregnancy loss should be made by healthcare professionals and researchers, and should also include public health advocates and policymakers. As a result of the findings from this thesis, I suggest further research in the area of miscarriage, and I outline a number of recommendations in relation to clinical practice and public policy. It is essential to obtain robust evidence on the association of poor mental health and adverse pregnancy outcomes that may lead to targeted interventions for women who are at higher risk of developing stress or mental disorder before, during and after pregnancy. The need for targeted interventions to reduce stress and increase mental wellbeing among pregnant women with a history of miscarriage is also warranted. An effort should be made to design and implement high quality, appropriately powered, RCTs that can provide reliable and non-biased evidence on preventable risk factors and effective psychological and behavioural interventions that may improve outcomes in future pregnancies. To achieve this goal, research funders need to acknowledge the burden of miscarriage at national and international level and support well-designed and large-scale RCTs. Funding RCTs in this area will lead to increase the understanding of potential interventions that might improve women´s psychological wellbeing after pregnancy loss.