Skip to content
A NSW Government website

Projects

Use of the CHeReL's linkage services has increased steadily since the unit was established in 2006. A list of completed projects and published research is available below.

Projects have been carried out for a broad range of purposes using data from a variety of sources. Some of the research applications of linked data have included:

  • Follow-up of cohorts of people with rare conditions or outcomes reported through population health datasets (eg. infective endocarditis, childhood cancer)
  • Morbidity and mortality associated with infectious diseases (eg. Hepatitis C, influenza) 
  • Follow up of researcher-supplied cohorts to obtain information on service utilisation or health-related outcomes (eg. 45 and Up study, the Australian Longitudinal Study on Women's Health and the HIV Observational Study Database)
  • Case control studies of cancer screening behaviour and outcomes
  • Validation of the accuracy of screening tests by linkage with outcome information (eg. antenatal serum screening and pregnancy outcomes)
  • Incidence of diseases and conditions by the identification of first-time events (eg. first admissions for stroke and heart failure)
  • Reporting of outcomes of health care adjusted for co-morbidity using historically linked data (eg. outcomes for stroke care in NSW hospitals)
  • Studies of health care safety, utilisation and costs (eg. adverse events in hospital, cancer patterns of care)
  • Validation studies of the accuracy of information held in population health datasets
  • Improved ascertainment of health information (eg. Aboriginality, diagnoses) using multiple data sources

Other purposes include population health surveillance and health system performance monitoring, where data linkage has enabled the development and monitoring of new performance indicators and strengthened the capacity of government to develop and evaluate policies and track performance over time.

Research publications using linked NSW data can be found on the PHRN website. Please select NSW under the jurisdiction filter. 

The following linkage projects were completed in 2011, 2012, 2013, 2014, 2015, 2016, 2017 2018 and 2019. A list of projects completed in previous years is also available.

 

Long term quality of life and nutritional outcomes following oesophagectomy: Ivor Lewis oesophagectomy vs minimally invasive oesophagectomy for oesophageal cancer, A/Prof Garett Smith, Royal North Shore Hospital

There continues a rising incidence of oesophageal cancers in Australia. Most patients present with worsening dysphagia and weight loss, however roughly a third of patients already have localised disease at time of diagnosis.

In oesophageal cancer patients, alongside traditional open surgical procedures, minimally invasive oesophagectomy has been increasingly used in recent years. The open procedures are associated with increased respiratory complications and other morbidities. Studies comparing open versus minimally invasive techniques have shown no major differences in survival or quality of oncological resection.

Despite improved oncological outcomes post-surgery, patients who survive still experience significant health-related quality of life (HRQoL) issues which may also contribute to overall poor long-term survival in patients. Studies have found short-term advantages of the minimally invasive technique, however, how these patients do in the long-term is still poorly understood and reported. Through this project, the researchers intend to assess the long-term HRQoL and nutritional outcomes in patients following open Ivor Lewis or minimally invasive oesophagectomy.

The researchers identified the cohort of patients who had an oesophagectomy at a major tertiary centre between 2001 and 2017 via review of hospital medical records and existing databases. Cohort data in the Royal North Shore Hospital oesophagectomy 2001-2017 dataset were linked by the CHeReL to mortality outcome data in the NSW RBDM Death Registrations and the Cause of Death Unit record Files (COD-URF). The researchers also plan to collect data prospectively via questionnaires, blood and stool tests.

The study findings are expected to help improve understanding of patients’ quality of life and needs postoperatively and inform decision making in doctors to benefit future oesophageal cancer patients undergoing surgery. The secondary purpose of this study is to investigate the impact of oesophagectomy on patients’ nutritional status and identify important nutritional deficiencies to improve patient care. 

Using linked poisoning data for toxicovigilance and to improve care of poisoned patients, Dr Rose Cairns, University of Sydney

Several studies have demonstrated significant cost benefits of poisons centres, primarily in reducing unnecessary hospital presentations and admissions, and in improving care and reducing  length of stay of hospitalised patients. However, there are no Australian studies that comprehensively describe poisoning epidemiology across all ages and exposure types.

The New South Wales Poisons Information Centre (NSWPIC) is Australia’s largest poisons centre, taking calls from healthcare professionals and members of the public. There is limited systematic information regarding NSW poisons centre service utilisation.

The researchers of this retrospective cohort study plan to use data linked via the CHeReL to analyse patterns in poisons centre service utilisation and factors affecting utilisation to help PIC guideline development and service planning.  A methodological study analysing poisoning case ascertainment in different health datasets is also planned. Geographic variation in poisonings will be studied for targeted prevention initiatives and deaths post-discharge for poisoning patients will be examined to provide guidance on the safety of driving after discharge, and guide mental health follow-up post overdose.

The cohort for this study includes NSW patients affected by poisoning, with poisoning-related records in any of the following: the NSW Poisons Information Centre (NSW PIC), Admitted Patient Data Collection (APDC), Emergency Department Data Collection (EDDC), Cause of Death Unit Record Files (COD-URF) and NSW Ambulance dataset. The CHeReL supported the study by identifying the cohort and linking cohort data from these datasets to poisoning data held by the NSW Poisons Information Centre.

Overall, the program of work is expected to provide a comprehensive overview of poisoning in NSW.

Determining the health and welfare use of New South Wales workers with long duration workers’ compensation claims (the Transitions study), Prof Alex Collie, Monash University

This study aims to examine the health care and welfare (social security) use of a group of more than 4000 injured workers in New South Wales with very long periods of time off work, whose income support through the NSW workers' compensation scheme ceased in late 2017 due to changes in the state workers' compensation legislation (the Section 39 cohort).

The retrospective case-controlled cohort study will examine the extent and nature of healthcare use and welfare use in the Section 39 cohort at the conclusion of their income support and for a 12- month period after withdrawal of income support in the workers’ compensation system. Health and welfare outcomes will be compared to an injured worker control group and a community control group.

The planned analysis dataset includes state and Commonwealth datasets which were linked and accessed via the CHeReL and AIHW respectively. The Section 39 cohort and injured worker comparison group were identified by the State Insurance Regulatory Authority of NSW (SIRA) and a potential community control group was identified by linkage to the Medicare Enrolment File by the AIHW. The CHeReL linked the cohort to NSW hospitalisation data from the Admitted Patient Data Collection (APDC) and the Emergency Department Data Collection (EDDC). The AIHW linked the Department of Social Security interactions of the cohort, medical benefits and pharmaceutical benefits from the Data Over Multiple Individual Occurrences (DOMINO) dataset, the Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Schedule (PBS) respectively.

The study is expected to provide new information about health and welfare outcomes of injured workers, on the impact of a major social policy change, and will for the first time examine interactions between Australian systems of income support and healthcare for injured workers.

The Antecedents of Renal Disease in Aboriginal Children and Young People (ARDAC): Understanding and moderating the trajectory towards chronic disease in young Aboriginal people, Professor Jonathan Craig, Flinders University

Addressing the gap in health between Aboriginal and non-Aboriginal Australians is a national health priority. Approximately 50% of this gap can be attributed to diseases linked by common risk factors, namely cardiovascular disease (CVD), diabetes mellitus (DM), and chronic kidney disease (CKD). As such, it is important to consider the interrelated nature of these conditions, their determinants, and their burden.

The ARDAC Study commenced in 2002 and is population-based study of 3758 participants recruited during early childhood with 57% of participants identifying as Aboriginal. The ARDAC study aims to identify early childhood factors associated with the emergence of chronic conditions over the life course. As the largest cohort study of its kind, with strong Aboriginal governance, the ARDAC Study investigates the effects of childhood hypertension, diabetes, cardiovascular, kidney health, and modifiable health risk behaviours and socio-demographic factors on overall life course outcomes between Aboriginal and non-Aboriginal participants in the ARDAC cohort.

To achieve its aims, the ARDAC study has linked its cohort data to over 20 statewide data collections via CHeReL, including the Perinatal Data Collection (PDC), Perinatal Death Reviews (PDRs), NSW APDC, NSW EDDC, Ambulance eMR/ PHCR, BDMRs, NAPLAN & ANZDATA. Commonwealth data, including the MBS and PBS data collections, was linked and provided by the Australian Institute for Health and Welfare.

With the ARDAC cohort now young adults, aged between 18 to 30 years of age, the study will address the current evidence gap for this age group for who data on chronic disease health and wellbeing are currently limited. Addressing this evidence gap is crucial because early adult life is a critical transition period for the development of chronic disease. As such, the ARDAC data linkage will generate definitive evidence on the life course trajectories for chronic conditions from childhood to early adult life among Aboriginal Australians.

Late postoperative complications after childhood hypospadias repair performed in three Children's hospitals in New South Wales between 1991 and 2006, Dr Peter Pockney, Dr Aniruddh Deshpande John Hunter Hospital

Hypospadias is the most common congenital penile anomaly. Most diagnosed children undergo surgical correction in early childhood as a single or two-stage procedure, with the aim to achieve both functional (urinary and sexual) and cosmetic normalcy.

There is currently limited knowledge on long-term outcomes of hypospadias repair, particularly the incidence and type of subsequent penile interventions. Filling this knowledge gap will help understand the risk factors of future complications requiring intervention and devise preventative strategies and improve the care of hypospadias patients of all ages.

This retrospective data linkage study included a cohort of children operated on for hypospadias in three NSW Children's hospitals between 1991 and 2005. The cohort was produced from the Medical Record Departments based on the pre-selected diagnostic and procedure codes. The CHeReL linked the cohort to the NSW APDC and EDDC to identify individuals who subsequently had penile surgery at any age between 2001 and 2020 in public or private hospitals in NSW.

The linked data will be used to analyse the incidence and type of the unplanned subsequent penile procedures and identify potential risk factors (e.g., severity of hypospadias, type of repair, presence of persistent chordee, division of the urethral plate etc.). The researchers also aim to conduct a time-to-event analysis of subsequent penile procedures. By ascertaining potential risk factors for future long-term complications and the researchers expect to be able to propose interventions to mitigate them.

Generating evidence for a strong foundation in the early years: using population health data for translational child health, healthcare and policy, Professor Natasha Nassar, University of Sydney

Long-term health, developmental and educational outcomes of many perinatal and early childhood conditions are largely unknown. These include infants born prematurely, or with other medical conditions, or born to mothers with disorders of immune or endocrine systems. Children experiencing acute or chronic health conditions or injuries throughout childhood, and undergoing subsequent procedures or interventions are also at increased risk of complications, including impacted neurodevelopment.

This data linkage project seeks to establish an Australian birth cohort of routinely collected and linked population perinatal, health, development, and education data of infants born in NSW from 1994-2015.  The proposed research will utilize the cohort data to investigate the impact of early life events on subsequent child health, development, and healthcare utilization.

The specific aims of this project, for each condition or intervention, are to assess the burden of disease, assess maternal and infant characteristics and risk factors of infants with each condition, examine health outcomes, health service utilisation and associated costs for infants with and without each condition/ intervention and investigate association between presence of condition/ intervention and developmental, educational, neurodevelopmental, and mental health outcomes.

The CHeReL identified the cohort for this retrospective study consisting of infants and their mothers from NSW PDC from 1994 onwards and from all APDC separations up to age 25 years old (0-24 years at admission). The CHeReL linked cohort data to hospital and emergency admissions in the NSW APDC and EDDC respectively, death data from the RBDM and COD-URF and to the Mental Health Ambulatory Collection data. Several other datasets such as out-patients dataset, the Neonatal Intensive Care Units Study (NICUS), Register of Congenital Conditions (RoCC), NSW Mental Health Outcomes & Assessment Tools Data Collection (MH-OAT data), NSW Board of Studies NAPLAN, school attendance and HSC, Department of Family & Community Services Disability Services and Operating Room Datasets were also linked to create a comprehensive dataset.

This cohort will provide the opportunity to investigate emerging issues in child and adolescent health on a population level.

Mortality and hospitalisation outcomes of young people in primary mental health care: a 10 year follow up study, Professor Ian Hickie, University of Sydney

Understanding the risk of death from suicide in people experiencing mental distress is essential for delivering targeted clinical interventions as well as interventions at a population level. Recent analyses provide support to a disproportionate increase in emergency department presentations related to self-harm and suicidal behaviour presentations.

The data linkage study aims to establish mortality and hospitalisation outcomes over a ten-year period in a population of young people engaged in primary mental health care services. The researchers aim to establish rates of death from suicide and other causes, such as accident and injury, and where substance use or intoxication was a contributing factor, as well as rates of presentation to emergency departments and hospital admissions. The secondary aim of this study is to establish rates of presentation to emergency departments and hospital admissions.

To achieve this, mortality and hospital outcomes were linked to data already collected as part of a longitudinal study of 7000 young people (aged 12-30 years old) conducted by Brain and Mind Centre. Young people- aged between 12 and 30 years at first presentation, who attended at least one visit to the service, and consented to participate in the Brain and Mind Research Patient Register constituted the study cohort. The cohort dataset included sociodemographic, clinical, and functional data that were collected at baseline and at subsequent visits to the health service. Baseline neuroimaging and neurocognitive assessments that were available for a subset of 1000 patients will be used to identify predictors of mortality and hospitalisation outcomes. The CHeReL facilitated the study by linking cohort data to hospital and emergency admissions in the NSW APDC and EDDC respectively, death data from the RBDM and COD-URF.

Health service utilisation and outcomes following surgery for Inflammatory Bowel Disease, Professor Jane Young, University of Sydney

 Australia has one of the highest incidences of inflammatory bowel disease (IBD) in the world. While many patients with IBD can be managed with medical therapy, a few require surgery. These surgeries include those performed as an emergency operation (for example, an emergency colectomy), or those performed electively (including ileal pouch surgery).

The first part of this retrospective data linkage study looks to assess outcomes of patients undergoing IBD surgery. All individuals undergoing surgery for IBD admitted to NSW hospitals from July 2001 to March 2019 constituted the study group. Previous Australian studies in this cohort have limited to hospital based / single-centre studies. This study looks to use NSW administrative (APDC) data, linked with mortality data, to assess trends, outcomes, and health service utilisation in patients undergoing surgery for IBD, with the overall aim of identifying potential areas for improvement and ways of minimising ‘unwanted’ variation in outcomes following this surgery.

The second part of this study aims to validate the APDC to investigate outcomes of IBD surgery. A prospective database of all individuals undergoing ileal pouch surgery for IBD by a single surgeon and admitted to Royal Prince Alfred Hospital (RPAH), Sydney, from July 2001 to March 2019 (Ileal Pouch database) will be supplemented by comprehensive medical records review for each patient within the database and used for validation. By comparing this data to the APDC data, the accuracy of the APDC in terms of case ascertainment and diagnosis of complications can be determined. This validation study is important to guide future use of the APDC in investigating patients undergoing IBD surgery.

The CHeReL facilitated this study by identifying the cohort of all individuals in the APDC who had been diagnosed with IBD or had undergone surgery for IBD and in the Ileal Pouch database to their hospitalisation data in the APDC and death data in the NSW RBDM and COD-URF.

Macquarie MINDS: Monitoring of Injury and psychosocial health outcomes, career trajectories and continuing educatioN, LiveD experiences and Social connectedness, Dr Reidar Lystad, Macquarie University

The period from late teens to adulthood can have significant impact on the future health and well-being of individuals. The Macquarie Minds study aims to establish a large, prospective longitudinal cohort of young adults and routinely capture information on their life experiences and exposures and assess their impact on education, employment, lifestyle, health-related quality of life, well-being, social support, life events, carer responsibilities, and use of social media technology. The study findings could potentially be instrumental in understanding the determinants of healthy and resilient individuals in our society.

The pilot cohort study aims to determine the feasibility of recruiting university graduates to establish the longitudinal cohort study. The specific objectives of linking administrative health data are to determine the proportion of consented participants that can be matched to the Master Linkage Key; and to determine the proportion of such matched participants who have relevant health records in each of the data collections.

The cohort consists of Macquarie University graduates in 2018 who consented to participation in surveys and data linkage. University graduates are used as a proxy for the population of interest (i.e., emerging adults) as it allows convenient recruitment of a clearly defined cohort (i.e. graduates from a single university in a single calendar year). Primary cohort data were collected through an online registration form, baseline, and follow-up surveys and linked by the CHeReL to hospital data from NSW APDC and NSW EDDC and to NSW eMR, NSW PHCR, NSW Central Cancer Registry, NSW RBDM Death Registrations and Cause of Death Unit Record File.

The data linkage will enable the researchers to investigate individual and social factors associated with health outcomes and service use (ambulance, emergency department, hospitalisation, cancer registry, and mortality records).

Mapping the care pathway of youth and adolescents that self-harm in the Australian Capital Territory, A/Prof Alice Richardson, Australian National University

Self-harm is relatively common among young people with prevalence rates reported in adolescent samples ranging from 7.3 to 16.1%. Studies in New South Wales and Victoria found that the number of children and adolescents engaging in self harm is increasing. There is a lack of comprehensive research within Australia regarding patients who have self-harmed or have suicide ideation, who use a hospital Emergency Department (ED) as the first point of contact with the health system for mental distress.   

This data linkage study is one component of The Better Care and Better Outcomes for youth experiencing self-harm study. This component of the study aims to comprehensively quantify the relationship between different paths of mental health care and repeated ED presentation using linked administrative data from Australian Capital Territory, Australia.

The primary goal of the data linkage study is to estimate the rates of re-presentation to EDs by adolescents for self-harm, and to identify the effect of treatment pathways for children and adolescents who self-harm on re-presentations to EDs. The study will use linked routinely collected data to estimate rates of readmission to emergency departments adjusting for sociodemographic characteristics such as age, sex, and severity of mental health condition.

Individuals were eligible for inclusion in the study if they were aged 12 – 25 and ever had an ED presentation or an admission to a hospital or other health facility, from 2012-2018, with a diagnosis identified by a list of ICD-10 codes related to self-harm. Cohort data was linked by the CHeReL to hospital data from ACT Admitted Patients Collection and ACT Emergency Department  Data Collection DDC and death data from ACT Registry of Births, Deaths, and Marriages and Cause of Death Unit Record File, as well as mental health data from community mental health Database (MAJICeR).The study findings will help quantify the burden of ED re-presentation by individual characteristics; investigate variation in hospital-community mental health care paths by individual characteristics and to identify the optimal MH care path in relation to reducing ED representation in the ACT.

Epidemiology of congenital heart defects in Australian children and impact on health, health service utilisation, health costs, neurodevelopment and educational attainment, Dr Samantha Lain, University of Sydney

Congenital heart disease (CHD) is a structural heart abnormality and is the most common birth defect. About 3,000 affected babies are born in Australia each year.  CHD is the leading cause of hospitalisation in the first decade of life, with a third of these individuals requiring surgical intervention, often open-heart surgery. Currently, ~85% of affected children survive to adulthood, however little is known about their pathways through the health system, and their health and neurodevelopmental outcomes. 

This retrospective, population-based cohort study will use linked data to examine accuracy of reporting cardiac defects using hospital data, determine the prevalence of CHDs, describe and quantify CHD patient outcomes, identify factors driving poor health, neurodevelopmental, educational and mental health outcomes for children with CHD, and determine the health service usage among children with CHDs, and estimate the cost to the health system.

To facilitate this study, the CHeReL identified the cohort from the Kids Heart Biobank by linking to hospital discharge data and further linked to outcome variables concerning health (hospitalisation, procedures through linkage to the APDC, EDDC, mental health,), early child development (AEDC), Social services (Dept of Family and Community Services), school test (NAPLAN, School attendance and HSc) and survival of individuals (RBDM) with CHD will be identified from administrative datasets.

Overall, the project is expected to provide valuable insights into the health service utilisation and costs, and ongoing health, neurodevelopmental and cognitive outcomes for children with a diagnosis of CHD.

OUTBREAK: The characterisation of antimicrobial utilisation across the Illawarra Shoalhaven and Hunter New England regions, Professor Steven Djordjevic, University of Technology Sydney, Janaye Fish, Illawarra Shoalhaven Local Health

Antimicrobial resistance (AMR) is a significant public health and economic concern. AMR is associated with poor clinical outcomes, in terms of mortality and morbidity, length of hospitalisation and health care costs. Inappropriate use of antimicrobial drugs is thought to be a major risk factor for the development of AMR.

This retrospective, observational data linkage project aims to characterise and geospatially map community antimicrobial utilisation across the Illawarra Shoalhaven and Hunter New England regions of New South Wales using integrated datasets. To achieve this the researchers identified cohort of participants from the Sax Institute 45 and Up study (a longitudinal study on healthy aging of participants across New South Wales). Cohort data was linked to Pharmaceutical Benefits Scheme (PBS) and Medicare Benefits Schedule (MBS) data by the AIHW, for the researchers to characterise antimicrobial utilisation across the two pilot regions. Further, the CHeReL linked hospital data from the NSW APDC and EDDC, mortality data from the NSW RBDM and COD-URF and their history of communicable diseases from the NCIMS to enable analysis of trends and changes in community antimicrobial use and healthcare service utilisation across these two regions.

The study findings will be used to determine feasibility of subsequent linkage to private and public pathology data, and data generated from the microbiome sampling of a subset of patients within the Illawarra Shoalhaven region.

Data on antimicrobial use and healthcare service utilisation is essential in understanding the factors that influence prescribing practices, and in monitoring the efficacy of interventions designed to improve antibiotic stewardship.

Assessing unmet needs in palliative care in Illawarra Shoalhaven, Prof Kathy Eagar, University of Wollongong

With an ageing population and growing burden of chronic disease, end-of-life care is an increasingly important issue. Palliative care is known to improve patient and family outcomes, and there is emerging evidence to suggest that it is also cost-effective. Accurate information about the demand and supply for palliative care and delivery of non-beneficial or poor-quality care at end of life is required at a regional level to enable population-based planning.  The proposed project aims to address the gap in knowledge of local patterns of morbidity and death and quality of end-of-life care in the Illawarra Shoalhaven region. Using health data, proposed project will assess the level of potential unmet need for palliative care at the end of life and the level of potential non-beneficial care at end of life among the residents.

This is population-based retrospective observational study of residents in the Illawarra Shoalhaven (Local Government Areas of Illawarra) aged 15 years and over who died from 2013-2017. Analysis will be inclusive of data in the 2 years prior to death.

The clinical and service information required for this analysis were sourced and linked by the researchers from the Illawarra Shoalhaven LHD datasets. Further, the CHeReL provided mortality data i.e.  death registration and cause of death data by linking to RBDM and ACR COD_URF.

By identifying any specific gaps or indications of suboptimal care by patient subgroups, the results will be readily translatable into strategic service planning, policy and practice.

COSMIC: Colorectal cancer Outcomes in people with Severe Mental illness, Professor Steve Kisely, University of Queensland

Cancer is one of the major causes of death among people with a psychiatric illness. Our previous research has shown that cancer incidence rates in people with severe mental illness (SMI) - i.e., those with schizophrenia or bipolar affective disorder, are similar to those in the general population, but that cancer mortality is higher in those with SMI than those in the general population. Possible reasons include poor cancer screening participation rates in those with mental illness, delays in diagnosis leading to more advanced disease at diagnosis, and sub-optimal post-diagnosis management.

The researchers of this project utilised data from the Australia’s National Bowel Cancer Screening Program (NBCSP) to determine where the major barriers to optimal cancer care for those with SMI occur. The researchers will test the hypothesis that people with SMI have lower screening rates and are more likely to present with more advanced cancer. They are also less likely to receive the appropriate specialist surgical procedures, chemotherapy or radiotherapy.

People with SMI were identified from the prescribing data in Pharmaceutical Benefits Scheme (PBS) for second generation antipsychotics, indicated solely for treatment of either schizophrenia or bipolar affective disorder or lithium prescriptions, indicated for bipolar affective disorder.

This data linkage study aims to use Commonwealth data (NBCSP, Medicare Enrolment File, Medicare Benefits Schedule, Pharmaceutical Benefits Scheme, Australian Cancer Database and the National Death Index) to compare bowel cancer screening participation in people with SMI to those from the general population.

Additionally, the researchers accessed linked NSW Cancer Registry and hospital data via the CHeReL to examine care pathways from diagnosis through treatment and end-of-life care.

A dashboard of predictive analytics and decision support to drive care quality and person-centred outcomes in aged care, Professor Johanna Westbrook, Macquarie University

Dashboards are a vehicle for data integration, predictive analytics, and decision support. Clinical dashboards provide visual representations of patient status and clinical environments that change dynamically in real-time.

The aim of this project is to develop a digital dashboard of integrated information, risk indicators and decision support, to better identify and support older adults at risk of poor outcomes in residential and community-based aged care with special focus on two priority aged-care challenges, namely falls and client wellbeing.

The CHeReL facilitated this project by linking , routinely collected retrospective health and aged care data from the NSW APDC, NSW EDDC and death data from the NSW RBDM to retrospective, non-identifiable data from the care management system of a large aged-care provider. Linked data will be used to develop and validate risk models that can identify aged care clients at risk of poor outcomes (i.e., fall-related hospitalisations and low wellbeing scores). The resulting aggregate risk models will be used in the dashboard to identify and improve the care of at-risk clients. Linked data via the CHeReL will also be used to evaluate the impact of a stepped-wedge cluster randomised controlled trial of the dashboard on the care and outcomes of aged care clients.

The dashboard is expected to help aged care managers monitor the quality of the care they provide; support general practitioners and care staff to obtain an overview of each client at a glance; and deliver information to aged care clients and their families.

Surgery for non-malignant polyps in the era of advanced endoscopic therapeutic techniques: A state-wide retrospective analysis, Professor Michael Bourke, Westmead Clinical School / Westmead Hospital

Australia has one of the highest rates of colorectal cancer (CRC) in the world. Early detection and definitive treatment have shown to reduce colorectal cancer morbidity and mortality. Screening programs have resulted in the identification of large pre-malignant lesions in asymptomatic individuals. Development of advanced polypectomy techniques has meant most of the non-malignant polyps can now be managed safely and cost-effectively through endoscopic removal. There is no published Australian data on the prevalence of traditional surgery for non-malignant polyps or local post-operative mortality and morbidity data for this indication.

This retrospective cross-sectional study aims to estimate the rates of surgery for non-malignant polyps compared to rates of surgery for malignant lesions. The study cohort of all patients in NSW who underwent colorectal surgery for non-malignant polyps between mid-2007 – 2017 were identified from the Admitted Patients Dataset Collection by the CHeReL using procedure and diagnosis ICD codes. Further, the CHeReL linked cohort data to the NSW Cancer Registry, the NSW EDDC and death data from NSW RBDM. Using linked data, the researchers aim to explore all aspects of cost and highlight key areas of cost saving and the potential to reduce strains on surgical wait lists and reduce poor outcomes for patients.

Using additional diagnoses recorded in the APDC the investigators aim to capture adverse events and determine risk factors for morbidity and mortality. Cost of admission and length of stay for admissions that contained adverse events compared to uneventful admissions will also be estimated.

Overall, the study findings will allow clinicians to better inform their patients on management options for complex polyps, highlight a key area of cost saving and will provide data to justify better referral pathways to expert endoscopists for management of these lesions.

Parenting and intergenerational disadvantage: A population trial of the Triple P system of parenting and family support, Professor Matthew Sanders. The University of Queensland

This research project is a large-scale study of the impact of enhanced parenting support on indicators of wellbeing, risk and disadvantage at a community level. The study aims to compare thirty-two Queensland communities (Intervention) with matched NSW comparison communities (Comparison) on a number of outcomes. Intervention involves a three-year implementation of the Triple P - Positive Parenting System.  Triple P incorporates five levels of intervention on a tiered continuum of increasing strength and narrowing population reach for parents of children from birth to age 16. Outcomes include child maltreatment, non-accidental injuries and emergency department visits, school attendance and achievement, family breakdown and community connectedness.

The research questions to be addressed will focus on child physical and mental health, and parent mental health. The study cohort included all babies in the Perinatal Data Collection born since 2005, their mothers and their other parents identified in the Registry of Births, Deaths and Marriages birth registrations. Population health information sourced from ongoing data collections via the CHeReL includes hospitalisation data from the NSW APDC and EDDC, birth and death data from the NSW RBDM, Perinatal Data Collection, Cancer Registry data and the Mental Health Ambulatory Data Collection.

The researchers will use cohort linked health data, multiple government administrative data-assets and a two time-point prospective survey data for their research and evaluation. Data will be used to understand and quantify aspects of physical and mental health of parents and children living in the intervention and comparison sites as part of the evaluation process. Linked data that focuses on the health and wellbeing of entire populations of children and parents will provide researchers with a comprehensive picture of health and wellbeing at the population level and enable them to monitor trends over time.

Variations in Care for Pancreaticoduodenectomy in NSW - An outcome and cost analyses, Professor Vincent Lam, Macquarie University

Pancreatic cancer is the fifth most common cause of cancer death in NSW. Pancreaticoduodenectomy provides the only chance of cure for cancers in the head of the pancreas. Pancreaticoduodenectomy is a high acuity procedure associated with significant morbidity, mortality, and financial costs. It has previously been shown that there is significant variation in patient outcomes following pancreaticoduodenectomies in NSW including large differences in cost, perioperative mortality, readmissions, and hospital stay.

This retrospective cohort study using linked data aims to examine the variation in costs for patients who underwent pancreaticoduodenectomies in NSW public hospitals, evaluate any hospital and patient specific variables contributing to the cost, and examine the association between cost and postoperative patient outcomes. The study will also use linked data to explore if high volume centres have lower patient mortality rates, lower readmission rates as well as cheaper in-hospital cost when compared to low volume centres. The CHeReL supported this study by extracting and linking de-identified cohort data from the NSW Central Cancer Registry, hospital data from Admitted Patients Dataset Collection, death data from NSW Registry of Births, Deaths and Marriages and the Cause of Death Unit Record File, healthcare cost data from Activity Based Management department of NSW Ministry of Health, and the NSW National Weighted Activity Unit (NWAU)

The resultant data analyses may strengthen the argument for the current centralisation of pancreaticoduodenectomy in NSW and help in further defining a best practice model for NSW.

The impact of specialist neonatal carer on long term child outcomes, Dr Serena Yu, University of Technology Sydney

There is a paucity of evidence in relation to the long-term outcomes of neonatal care - an area of intensive resource use and rapid medical advances. Significant data limitations and the infeasibility of randomising the care of very sick babies for a clinical trial, are barriers to understanding neonatal care costs and long terms gains.

This project aims to use well-established quasi-experimental methods to approximate the results of a randomised trial, and utilise population-based administrative data, including sociodemographic data, to compare the care, health outcomes and costs of different cohorts of newborn babies, and particularly at-risk babies – for the first five years of life.

Singleton babies born in NSW from 2004 - 2018, and NSW women who had a singleton pregnancy over the same period were included in this study. The CHeReL identified the cohort from the Perinatal data Collection and linked cohort data to NSW Admitted Patient Data Collection (APDC), NSW Emergency Department Data Collection (EDDC), Neonatal Intensive Care Units Data Collection (NICUS), Registry of Births, Deaths and Marriages (death and birth registration data), and the Australian Early Development Census. In addition, data from the Commonwealth Department of Human Services (Medicare Benefits Schedule and Pharmaceutical Benefits Scheme data) were linked to identify long-term healthcare cost and service use.

The research findings will address a significant evidence gap pertaining to the long-term outcomes following specialist neonatal care, and how this interacts with the social determinants of health. The outcomes of this project will be to inform decision-making by policymakers, administrators, clinicians and consumers, redress inequalities where they exist, and improve resource use.

NSW Benign Colorectal Diseases Study,Professor Vincent Lam, Macquarie University

Benign colorectal diseases such as diverticulitis, benign perianal disease and inflammatory bowel disease (IBD) account for a significant burden of disease to Australia’s hospital system. Despite the frequency of these diseases, there is limited up-to-date data on the epidemiology, disease burden and health care cost in Australia.

This retrospective cohort study explores the epidemiology of a range of both common and less common benign colorectal diseases. This project aims to utilise pre-existing databases collected by NSW and linked by the CHeReL to investigate the in-hospital management and outcomes of a range of benign colorectal diseases from 2005 onwards. All adults presenting to NSW hospitals from both APDC and EDDC databases with separation time from 2005 onwards with benign colorectal diseases will be included. The CHeReL identified the cohort and linked cohort data to NSW Admitted Patient Data Collection (APDC), Emergency Department Data Collection (EDDC), Cause of Death Unit Record File (COD-URF), Cancer Registry (CR) and Registry of Births, Deaths and Marriages (RBDM).

Linked longitudinal data will enable the researchers to study long term trends and outcomes of patients with benign colorectal diseases and help in identifying successful management strategies especially in areas of controversy eg. management of acute diverticulitis, pilonidal sinus, haemorrhoids. In addition, it is hoped the long-term data may confirm or refute disease associations such as between diverticulitis and bowel cancer. The study findings will provide an insight in disease epidemiology, disease progression, patient outcomes, management strategies and complications for a range of benign colorectal diseases in NSW hospitals.

The economic impact of providing precision medicine through whole genome sequencing – Mitochondrial disease cohort, Professor Deborah Schofield, Macquarie University

Clinical manifestations of mitochondrial disorders in adults include muscle weakness, fatigue, cardiomyopathy, stroke-like episodes, deafness, diabetes, and eye problems. There is currently no reliable diagnostic approach available. Recent advances in genomic diagnostic indicator testing and metabolomic tests shows potential for greater than 90% diagnostic sensitivity.

The aims of this cross-sectional observational study including a cohort of individuals with mitochondrial disease is to evaluate hospital resources used by mitochondrial disorder patients, assess comorbidities experienced by individuals affected by patients with mitochondrial disorders, and to evaluate the economic impact of genetic sequencing in mitochondrial disorders.

The EPIC-Mito study will recruit 250 individuals with mitochondrial disease and collect cross-sectional data through a survey questionnaire. The study will use Medicare and pharmaceutical claims data from Services Australia, and hospital admissions, emergency departments and outpatient records datasets held by NSW Ministry of Health linked to Survey data via the CHeReL. A microsimulation model will be developed from the data collected. The research will determine health burden due to mitochondrial diseases and the extent to which clinical genomics can reduce these impacts. 

Falls risk associated with cataract and after first and second eye cataract surgery, Associate Professor Soufiane Boufous, The University of New South Wales

Falls incur over $1 billion in treatment, disability, lost output and mortality each year in Australia. People with cataract are at approximately 3-times increased risk, making fall prevention a critical public health issue.

The aims of the prospective, 24-month cohort study are to measure temporal changes in the rate of falls and falls requiring medical attention, before cataract surgery of first eye, between first and second eye and after second eye surgery.  A second aim is to characterise the visual function in older people in these time-periods and investigate vision-related mechanisms for increases in falls risk, adjusting for known predictors of falls. This study also aims to evaluate healthcare costs for each participant in each time period and compare the costs for those who fell and those who did not during the study period.

The cohort for this study includes patients aged 65 years or older with bilateral cataract presenting for surgery at public hospital eye clinics in Sydney, Melbourne, Perth and Adelaide. The investigators plan to use linked data to identify and capture falls requiring medical care, and cross-check against self-reported falls.  These will be obtained by linking study cohort data to hospital and emergency datasets in each state as well as the Medicare Benefits Schedule (MBS). For NSW, the CHEREL linked the Cataract Surgery Cohort Study data to the Admitted Patients Dataset Collection, Emergency Department Data Collection and death data from Registry of Births, Deaths and Marriages and Australian Coordinating Registry Cause of Death Unit Record Files.

With limited resources to shorten public waiting lists, there is a need to understand an individuals’ risk of fall injury or other negative consequences while waiting for surgery.  This important project will inform the development of strategies to reduce falls risk in the many older people with cataract.

Assessing risk and predicting harms of prescribed opioids, Professor Louisa Degenhardt, The University of New South Wales

Australian opioid prescribing increased fifteen-fold in the last two decades. Originally registered to manage cancer and acute pain, opioids are now approved to treat an increasing number of conditions, especially chronic noncancer pain which has traditionally been undertreated. However, in parallel to escalating use, there is also evidence of increasing opioid related hospitalisation, dependence and overdose, indicating that at least some of the growth in prescribing is not consistent with quality medicine use.

The aims of this study are to examine the magnitude of risk for adverse outcomes across different opioids and formulations for a range of known (e.g. overdose, dependence) and potential opioid risks (e.g. falls resulting in emergency department visits and hospitalisation), towards a comprehensive documentation of risks for all prescribed opioids in Australia. The study will also identify risk factors for those adverse outcomes, including clinical, prescriber and treatment-related characteristics, among different patient groups (such as cancer, CNCP and acute pain); and use machine learning methods to develop risk prediction models to identify those who are at highest risk of adverse outcomes

This retrospective observational data linkage study includes eight Commonwealth and NSW collections. The study cohort includes NSW residents who started a new opioid dispensing episode from 2002, derived from the Pharmaceutical Benefits Scheme (PBS) dataset held by the AIHW and then linked in-house to Medical Benefits Scheme, Australian Cancer Database, National Death Index. The CHeReL provided linked records from the NSW-based datasets (Admitted Patient Data Collection, Emergency Department Data Collection, Mental Health Ambulatory Data Collection and the Pharmaceutical Drugs of Addiction) and the ACT-based datasets (Admitted Patient Collection and Emergency Department Data Collection) to AIHW for linkage to the cohort.

The study is expected to provide valuable information about the drivers of long-term prescribed opioid use, nonadherence, dependence, overdose and other harms.

Prescribing medicine use and healthcare contacts before suicide: Detecting opportunities for intervention and prevention using population-based linkage of routinely collected data, Dr Kate Chitty, University of Sydney

In Australia, suicide is the leading cause of death for people aged 15–44 years. Health professionals deliver most of our key suicide prevention strategies via health services, but other efficacious population-level strategies include means restriction and public awareness campaigns. Currently, we have no population-level data allowing us to determine which individuals, in what parts of Australia, are likely to use our most promising interventions delivered by health services. The aims of this study are to describe: (1) health service utilisation rates in the year prior to death by suicide, and how this varies by individual case characteristics; (2) prescribed medicines use in the year prior to death by suicide, medicines used in suicide by poisoning and how this varies by individual case characteristics. The cohort for this population-based case series study included all intentional deaths that occurred in Australia between 2013 and 2019 as derived from the National Coronial Information System (NCIS). Cohort data were linked to routinely collected data from the Pharmaceutical Benefits Scheme (PBS), and the Medicare Benefits Scheme (MBS) by the AIHW and to NSW Health state-based information from the following collections: Admitted Patient Data Collection, Mental Health Ambulatory Data and Emergency Department Data Collection. Linking routinely collected data on prescribed psychotropic medicines and mental healthcare prior to suicide death offers an opportunity to comprehensively examine the use of these treatments before suicide.

Description and prediction outcome of drowning patients in New South Wales,Associate Professor Michael Dinh, NSW Agency for Clinical Innovation

Drowning is a preventable cause of death, with a high significance in young people. Drowning is one of the top five leading causes of death amongst those ages 1-14 years in Australia with the largest number of drowning deaths in Australia being from New South Wales. Patients who experience drowning, can have different outcome ranging from full recovery to death.

This study aims to explore the outcome of patients who are admitted to hospitals for drowning in the NSW and investigate prediction of patients’ outcome based on accessible data.

The cohort for this retrospective study was defined from drowning cases identified in the NSW Ambulance dataset (NSW Amb) and using select ICD codes in the NSW Admitted Patient Data Collection. The CHeReL then linked cohort data to all matching records from the APDC, NSW Amb, Emergency Department Data Collection and to death data from NSW Registry of Births, Deaths and Marriages and Cause of Death Unit Record File.

The study will provide outcome data for drowning patients which would contribute in producing better models of care for drowning patients across NSW. Ultimately, the results of the study will provide data to enable efficient and effective hospital care for drowning patients. The current pathways of care and guidelines could be updated based on the results of the study.

Impact of chronic health conditions and injury on school performance and health outcomes of children: pilot study, Associate Professor Rebecca Mitchell, Macquarie University

Different types of injuries can affect children in different ways and serious injuries can adversely impact on the child’s psychological and physical health, school performance, and their long-term quality of life. Chronic health conditions can also have an adverse impact on a child and their ability to perform well at school and to complete their schooling.

The overall objective of this longitudinal, population-based case-comparison study of injured or chronically ill children is to examine school completion, education performance and health outcomes of children who were hospitalised with an injury or a chronic health condition. Cases were identified by a principal diagnosis of injury or a principal or any diagnosis or some common chronic conditions while the comparison group consisted of children ≤ 18 years of age (identified through the NSW Registry of births) who did not have any of the diagnosis codes assigned to the cases.

Cases and comparator group were identified through data extraction by the CHeReL from NSW hospitalisation data (Admitted Patient Data Collection (APDC), Emergency Department Data Collection (EDDC)) and the NSW Registry of Births. The study cohort was linked to matching records from the APDC, EDDC , Mental Health ambulatory client contacts - to capture health service use and health outcomes; to mortality data from NSW Registry of Births, Deaths and Marriages and the Cause of Death Unit Record File; and to the National Assessment Plan for Literacy and Numeracy (NAPLAN) dataset, NSW school enrolment form information, high school completions- to capture scholastic performance and school enrolment.

The study findings will provide valuable information on factors influencing school performance of injured or chronically ill children compared to their matched peers, identify factors that either positively or negatively mediate young people completing high school and assess characteristics of long-term health service utilization and hospital treatment cost among injured or chronically ill children.

Variation in mortality of hepatocellular carcinoma in NSW: a population-based data linkage study,Associate Professor Amany Zekry, The University of New South Wales

The incidence of liver cancer in Australia is increasing. Survival rates in Australia, as in other nations are poor. Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer, representing 75-90% of liver cancers. Clear geographic differences in HCC incidence and mortality are seen throughout Australia. These variations may be due to differences in access to screening and preventative healthcare, as well as differences in the prevalence of risk factors between populations

This study aims to determine if there is geographic variation in HCC associated mortality across NSW and to identify factors associated with greater mortality. The role of patient location in terms of the prevalence of viral hepatitis and diabetes, social disadvantage and access to HCC care will be investigated. The study outcomes include estimation of overall and disease-specific survival after HCC diagnosis, patient and system characteristics associated with earlier mortality, geographical mapping of the prevalence of HCC and its aetiologies and variation in survival after HCC diagnosis.

The study cohort included persons diagnosed with HCC in NSW between July 2001 and December 2015 identified through the NSW Central Cancer Registry and NSW Admitted Patient Data Collection. Cohort data from these two datasets were linked by the CHeREL to mortality data from the NSW Registry of Births Deaths and Marriages and Australian Coordinating Registry Cause of Death Unit Record File and to NSW Notifiable Conditions Information Management System to capture history of infections. Aggregated non-identifying data on cases of diabetes by area within NSW from the National Diabetes Services Scheme (NDSS) registry and publicly available data on socioeconomic Index for Areas (SEIFA), national reports for the Hepatitis B and Hepatitis C Mapping Project will also be utilised for the analysis.

Findings from this study could potentially inform strategic planning of prevention programs, as well as research and health system resource allocation.

The economic and social impacts of genetic sequencing for intellectual disability (EPIC-ID, Professor Deborah Schofield, Macquarie University

Severe intellectual disability (ID) is an important unmet diagnostic and management challenge due to its high prevalence, life-long nature, and rate of recurrence within families. Despite its significant impact, there is relatively little data on the cost of ID in Australia, and thus the capacity to quantify the benefits of effectively treating or preventing ID is limited. Next Generation Sequencing (NGS) has increased knowledge of ID aetiology. Current evidence suggests that whole genome sequencing (WGS) will provide the capacity to quickly diagnose the cause of about 60% of severe ID, shortening the time to diagnosis.

This study will analyse the health services (including hospital admissions and emergency department presentations) used by patients with ID and assess comorbidities experienced by individuals affected by intellectual disability. The investigators will then determine the extent to which clinical genomics can contribute to ameliorating these impacts.

This data linkage study includes a retrospective and a prospective cohort of individuals who are the primary carers of a child or relative with ID. The child or relative(s) with ID in the prospective group will be offered genomic testing. Information will be collected from both groups using the baseline questionnaire and from two follow-up questionnaires at one month and 12 months after receiving the results of genomic testing from the prospective group only. Cohort data (EPIC-ID dataset) will be linked to NSW Admitted Patient Data Collection and Emergency Department Data Collection by the CHeReL and to Medicare Benefits Scheme (MBS) and Pharmaceutical Benefits Scheme (PBS) by the AIHW.  Survey and linked data will used to quantify the economic and psychosocial impacts of caring for children or relatives with ID and economic costs of ID to the Australian government including the health system and other government departments such as education and social security.

Quality in General Practice - trial of a funding model in primary care, Professor Andrew Bonney, University of Wollongong

This cluster randomised controlled trial aims to evaluate the impact of an outcomes-based funding model in primary care practice. The model provides targeted practice incentives for patient enrolment with a preferred provider, longer consultations, same day access and structured follow-up after hospitalisation.  The impact of the model on quality of care and health service utilisation for patients at increased risk of hospitalisation will be compared to usual care.

The primary data for this study (EQuIP data set) includes Survey data collected from staff and patients enrolled in the trial at 33 general practices across NSW (11 practices), Victoria (12 practices) and Tasmania (10 practices). The primary survey data will be linked to Admitted Patient Data Collection, Emergency Department Data Collection and to death registrations from the Registry of Births, Deaths and Marriages from NSW, Victoria and Tasmania by the respective state linkage agency and to Pharmaceutical Benefits Schedule (PBS) and Medicare Benefits Schedule (MBS) by the Commonwealth Department of Human Services. Additionally, the evaluation study will also use six-monthly data extracted from electronic patient records via Medicine Insight program and quarterly data extracted from electronic patient records by practice staff.

The study will primarily test the hypothesis that among high-risk patients and children attending general practices, the introduction of a practice-level service model incorporating continuous and graded quality improvement incentives, will improve patient-perceived relational continuity. Secondary outcome measures include rates of prescriptions, pathology, imaging, hospitalisation and downstream health system costs.

GuideLine-concordANt breast Cancer surgEry and Radiotherapy (the GLANCER study), Dr Marina van Leeuwen, The University of New South Wales 

The Optimal Care Pathway for Women with Breast Cancer together with The Cancer Australia Statement – Influencing Best Practice in Breast Cancer sets out guidelines for best practice in breast cancer care.

The aim of this retrospective cross-jurisdictional record linkage study is to use linked data to investigate unwarranted clinical variation in relation to current best-practice surgical and radiotherapy guidelines that were identified through preliminary studies. Specifically, the following aspects of guideline-concordant care will be examined: breast conserving surgery in women with early-stage breast cancer; hypofractionated radiotherapy in women with early-stage breast cancer; reoperation following breast conserving surgery; breast reconstruction following mastectomy; and sentinel lymph node biopsy in patients with preoperative ductal carcinoma insitu (DCIS).

The CHeREL facilitated this study by identifying a cohort of adult breast cancer cases on the NSW Central Cancer Registry (CCR) and/or the NSW Admitted Patient Data Collection (APDC) between July 2001 and December 2018 and by linking cohort data with data from BreastScreen NSW and the NSW Clinical Cancer Registry (ClinCR). The cohort data have also been linked with the National Death Index (NDI), the Medicare Benefits Schedule (MBS), and the Pharmaceutical Benefits Scheme (PBS) by Australian Institute of Health and Welfare (AIHW).

By identifying factors amenable to changes in policy and practice, this project has the potential to directly impact the delivery of breast cancer care in Australia.

Measuring medication adherence: application of group-based trajectory model for Australians with chronic health conditions requiring continuous use of medication, Mr. Kyu Hyung Park, Macquarie University 

Poor medication adherence (MA) for chronic conditions is known to increase morbidity, mortality and healthcare spending. Traditional measures of medication adherence estimate the amount of medicine taken relative to the total amount prescribed and do not consider longitudinal patterns of taking medications. The recently developed group-based trajectory model (GBTM) considers longitudinal patterns of medication consumption, allowing researchers to differentiate between alternative patterns or ‘types’ of non-adherence and to measure their effect on health outcomes. This allows clinicians to tailor interventions to the particular characteristics and medication needs of patient groups based on their pattern of medication consumption. 

The data linkage study aims to explore the use of GBTM to define medication adherence for Australians with three chronic health conditions – depression, cardiovascular disease and osteoporosis.

45 and Up Study participants who used medications within the specified ATC medication codes constituted the study cohort. The CHeReL linked cohort data (linked to Medicare Benefits Schedule and Pharmaceutical Benefits Schedule) to matching records from NSW Admitted Patient Data Collection, Emergency Department Data Collection, Australian Coordinating Registry Cause of Death Unit Record Files and the Mental Health Ambulant Data collection.

Linked data will be used to identify adherence types based on patterns of medication consumption; finding relationships between adherence type and patient characteristics, health outcomes and healthcare resource use; and evaluating the potential benefits of using the GBTM in measuring medication adherence, compared to the traditional measures.

The research findings are expected to inform the development of targeted public health interventions to help improve adherence, thereby leading to better health outcomes and more efficient healthcare.

Indigenous Medication Review Service (IMeRSe) – a feasibility study, Professor Amanda Wheeler, Griffith University

The Indigenous Medication Review Service (IMeRSe) is a Commonwealth funded feasibility study as part of the Pharmacy Trial Program, across 9 sites (in Queensland, Northern territory and New South Wales) in Australia. It involves the development and evaluation of a culturally responsive medication review service, to be undertaken by community pharmacists on a fee-for-service basis in conjunction with Aboriginal Health Services.

Linked data will be used in the pre-post observational study to investigate the feasibility of identifying ‘serious medication-related problems’ through the use of a pre-specified list of potentially preventable medication-related hospitalisations (PPMRHs), and to see if this list can be used to estimate the effect of the IMeRSe intervention on hospitalisation rates, and to investigate study enrolment rates and retention rates to inform a future RCT.

For NSW, the study cohort was defined as any participant in IMeRSe who was recruited in one of four NSW sites (including Gunnedah, Brewarrina, Nowra and Moree) and has provided consent to release hospital admission data. The CHeReL linked cohort data to hospital admissions (NSW Admitted Patient Data Collection, Emergency Department Data Collection) for one year prior to and six months after enrolment in IMeRSe. Similar linkages of hospitalisation data to the locally recruited cohorts will be carried out in Queensland and Northern Territory. Linked data from the Medicare Benefits Schedule and Pharmaceutical Benefits Schedule will be provided by the Commonwealth Department of Health.

The project will be carried out in partnership between Griffith University, the National Aboriginal Community Controlled Health Organisation (NACCHO) and the Pharmacy Guild of Australia and is expected to support Indigenous peoples, especially those with long-term conditions, to manage their medicines safely and improve their health and wellbeing.

Tracking patients who have been discharged from a community palliative care service, Dr Davinia Seah, University of Technology Sydney

The demand for palliative care services are projected to increase as the population ages and the population grows. As the number of deaths are expected to double by 2040 in Australia, having efficient use of a specialist community palliative care team becomes increasingly important to support people who want to remain at home. While most patients are discharged because of death, some patients are discharged from community palliative care service because they have stabilised and no longer require specialist palliative care. The aim of this project is to track the outcomes and health service utilisation of these patients.

Specifically, the researchers will use linked data to track ambulance call outs, emergency department (ED) presentations, hospital admissions and the length of stay of each admission, intensive care unit admissions and the length of stay of each admission, place of death after discharge.

The cohort for this retrospective cohort study included all patients who were discharged alive from the Sacred Heart Health Service community palliative care service between July 2010- July 2016 and were followed up until death or censored at the end of 2018. Cohort data was extracted from the Sacred Heart Palliative Care System database and linked by the CHeReL to hospitalisation data from  NSW Admitted Patient Data Collection and Emergency Department Data Collection, Death data from the Registry of Births , Deaths and Marriages and Australian Coordinating Registry Cause of Death Unit Record Files and to NSW Ambulance dataset.

The study findings will be used to improve discharging practices and resource planning for community palliative care services. Further, the results will be presented and disseminated at relevant department meetings and conferences to influence decision making at a local hospital level, but also statewide level.

Evaluating Healthy Ageing in Australia (New South Wales state-based datasets), Professor Steve Wesselingh, South Australian Health and Medical Research Institute

Australia is a country with an ageing population. Currently 7% of those aged 65+ are receiving residential aged care services, and they account for 40-50% of hospitalisations and days spent in hospital and also have high rates of ambulance service utilisation. There is an urgent need to better coordinate and integrate information about people receiving aged care services, so that decision-making and practices achieve quality, coordinated, efficient, innovative and age-friendly services and outcomes.

The primary aim of this population based observational cohort study is to describe the epidemiology of aged care recipients and their service utilisation (i.e. aged care and health care) from 2002 to 2017. The secondary aims of the study include an evaluation of the risk factors for death and other health-related events in this population, and to identify successful areas and areas for improvement.

The cohort for the national study is defined as older Australians (aged 65+ years or, 50+ years and of Aboriginal or Torres Strait Islander descent) who have been assessed for and/or received aged care services between 2002 and 2017, including those who received residential aged care, home care package programmes, transition care packages, and home and community care support during the study period. The National cohort includes records in the National Aged Care Data Clearinghouse (NACDC), linked to the National Death Index (NDI), Medicare Benefits Schedule (MBS), Pharmaceutical Benefits Scheme (PBS), and to state-based hospitalisation datasets for residents from each participating jurisdiction. For the NSW subset of the national cohort, The CHeReL provided linked records from the Emergency Department Data Collection, Admitted Patient Data Collection and Ambulance datasets to the AIHW for linkage for the Registry of Senior Australians (ROSA) research team.

The study findings are expected to guide evidence-driven decision-making across public and private aged care sectors in Australia.

Use of big data to investigate factors that promote, constrain or prevent the emergence of social and spatial inequities in health, behaviours and health services use across middle to older age, Professor Thomas Astell Burt, University of Wollongong

 Built and natural environment are now recognised as modifiable levers for improving the prevention and management of noncommunicable diseases through pathways such as physical activity and sedentary behaviour, diet, sleep, social relationships, and stress. Places where people live not only contribute to their health but are also likely to influence their use of healthcare and related economic costs and playing a role in whether people end up in hospital for potentially preventable reasons.

This project aims to examine change and stability in built and natural environment in Sydney and other urban areas and how these exposures affect trajectories and inequities in health outcomes, health-related behaviours, health service use and economic costs. The aim is to understand the effectiveness and cost-effectiveness of potential environmental interventions that may keep people healthy and out of hospital.

This study will use data from participants in Wave 1 and Wave 2 of the Sax Institute’s 45 and Up Study, as well as data from those who took part in Social Economic and Environmental Factors (SEEF) sub-study. This multilevel longitudinal study will utilise bespoke environmental indicators and routinely collected hospital, emergency department and mortality records linked to approximately 110,000 Wave 1 participants who lived in the major cities within NSW of Sydney, Wollongong and Newcastle.

The CHeREL facilitated this study by linking cohort data to records from the NSW Admitted Patient Data Collection, Emergency Department Data Collection, RBDM Death Registrations and the Australian Coordinating Registry Cause of Death Unit Record File

The planned overall outcome is to make high quality evidence available to decision-makers at all levels of government and city planning to bolster the business case for investing in healthier urban environments.

Perioperative Risk Assessment and Modification in Patients on Renal Replacement Therapy NSW/ACT, Dr Magid Fahim, Princess Alexandra Hospital, University of Queensland

Patients who suffer kidney failure need Renal Replacement Therapy (RRT) such as dialysis or a renal transplant to survive. Patients on (RRT) are hospitalised frequently including for surgical procedures to either save their life or improve their quality of life. However, patients with kidney failure are often considered ‘high risk’ for surgery due to their ill health. The aim of this project is to define the type and frequency of surgical intervention in Australasian dialysis and transplant patients, and also identify their outcomes and outcome determinants in order to facilitate doctor-patient discussions and ensuring that a well-informed decision is made when considering surgery.

The proposed research involves linkage of national surgical hospitalisation and outcome data from Admitted Patient Data Collection (ADPC) units (all states and territories and New Zealand) with the Australasian renal replacement therapy registry (ANZDATA) to develop a comprehensive and robust perioperative dataset of patients receiving renal replacement therapy who have undergone surgical intervention.

The NSW and ACT cohort of patients for this research included adult incident and prevalent RRT patients registered ANZDATA Registry from 1 July 2001 (for NSW) and I July 2004 (for ACT) who had a record in NSW/ACT Admitted Patient Dataset with a diagnosis or procedure code related to surgery from a predetermined list of codes. The CHeReL facilitated the study by linking the ANZDATA Registry data to the NSW APDC and ACT Admitted Patient Care dataset (APC). 

This study is expected to describe the surgical disease burden in patients on RRT and answer important questions around perioperative RRT outcomes by providing novel epidemiological, outcome and risk prediction data.

 

An evaluation study of the New South Wales Involuntary Drug and Alcohol Treatment (IDAT) Program, Professor Alison Ritter, The University of New South Wales

The NSW Involuntary Drug and Alcohol Treatment (IDAT) program started in NSW in May 2012 to provide involuntary treatment as the last resort option to people who are severely dependent on alcohol and/or drugs who may be at risk of serious harm and for whom no other treatments have worked. This data linkage study is part of part of a broader follow-up evaluation program that aims to examine the effectiveness of the IDAT program. A process evaluation, cost assessment, and an outcome evaluation were also carried out under the evaluation program.

The data linkage sub-study is focused on health service utilisation (HSU) and aims to measure the change in HSU outcomes for IDAT patients (12-month pre and 12-month post treatment), and to compare HSU outcomes between the IDAT patients and a control group.

The study cohort included all IDAT patients who received treatment at the IDAT program since the program’s commencement in May 2012 and the control group consisted of persons with alcohol or meth/amphetamine dependence, who were frequent attenders at hospitals and had received some form of treatment at least once, but not received IDAT. Both groups were identified by the CHeReL through data linkage and linked to routinely collected administrative data from five data collections: NSW Admitted Patient Data Collection (APDC), NSW Emergency Department Data Collection (EDDC), NSW Drug and Alcohol Minimum Data Set (NSW MDS DATS), NSW Mental Health Ambulatory Data Collection (MHADC) and NSW Mortality Data (RBDM).

This study will provide evidence to assist the NSW Ministry of Health and the Local Health Districts in their decision-making regarding the future of the IDAT program.

 

Physical and mental health, costs and health service use associated with obesity and bariatric surgery, Professor Alison Venn, University of Tasmania

There is considerable evidence from clinical trials supporting the effectiveness of obesity surgery (also known as bariatric surgery) for morbid obesity in the short-term. Relatively little is known about longer-term health outcomes, health service utilisation, medications and associated costs for people undergoing obesity surgery in Australia and how these outcomes and costs compare to those of patients who are potentially eligible but do not undergo obesity surgery.

In addition, there is Inequity of access to obesity surgery and long wait times in the public health system since most obesity surgery occurs in the private hospital system. Uninsured patients who are primarily of lower socioeconomic status and who have a higher risk of morbid obesity have more limited access to this form of treatment. Importantly, little is known about the health and cost implications of the long waiting periods for initial obesity surgery, or the impact of reducing these waiting times.

The retrospective study cohort comprised of an obesity surgery operated group and a matched control group of body mass index (BMI)-age-sex-propensity score-matched 45 and Up participants who have not had obesity surgery. The CheReL facilitated this study by linking the 45& Up cohort data, and linked MBS and PBS data to Admitted Patient Data Collection (APDC), Emergency Department Data Collection (EDDC), NSW Central Cancer Registry (CCR), NSW Mental Health Ambulatory Data (MH-Amb), NSW Registry of Births, Deaths and Marriages (RBDM) (deaths) and Cause Of Death Unit Record Files (COD-URF).

This project aims to assess the characteristics, physical and mental health outcomes, costs and health service use associated with obesity and obesity surgery. Outcomes, service use and costs will be compared, preoperatively and post-operatively in participants in the 45 and Up cohort who have had obesity surgery, and compared to the control group.

Findings from this study will inform obesity surgery management as well as resource allocation decisions regarding the provision of obesity surgery in the public and private health care systems.

 

Exploring the role of being an overweight or obese women and how accurate self-reporting of mammographic breast screening participation is in New South Wales, Dr Kate McBride, Western Sydney University

  Currently, breast cancer is a major health issue in Australia and places a substantial burden on the healthcare system. Early detection via mammography screens has been shown to reduce mortality due to breast cancer, however screening rates for free services are less than ideal in some at-risk population groups. Previous studies have shown that obesity is linked to a decreased rate of participation in mammography screens.

This study aims to establish a comprehensive and rigorous understanding of how Body Mass Index (BMI) affects participation in mammographic breast screening.

The CheReL will facilitate this study by linking self-reported screening data from the Australian Longitudinal Study of Women’s Health (ALSWH) with objectively collected screening data from the BreastScreen NSW screening program from 1996 onwards.

This study aims to capitalise on these two existing data sources by examining a randomly stratified group of 13,714 women to determine the association between different variables and mammography screening. Specifically, the association between BMI and mammography screening participation over time, and recommendations for the development of potential interventions to facilitate access to mammography screening among women in the target age group.  This project will also examine the accuracy of self-reported screening data in the ALSWH by comparing it to the BreastScreen routinely collected data.

Results of this study are expected to inform recommendations for future programs and development of treatments.

 

Evaluation of Community-based Mental Health Programs: Community Living Supports (CLS) and Housing & Accommodation Support Initiative (HASI), Associate Professor Karen R Fisher, University of New South Wales

NSW Health has implemented state-wide programs designed to support people with mental illness in their recovery from mental illness within the community. These include Community Living Supports (CLS) and Housing & Accommodation Support Initiative (HASI). The Social Policy Research Centre (SPRC) at University of New South Wales was commissioned by NSW Health to conduct an evaluation of CLS-HASI programs in relation to their objectives, including how effective and culturally appropriate these supports were for Aboriginal people.

The mixed method evaluation study included qualitative interviews with participants and stakeholders and secondary data analyses. The CHeReL facilitated the secondary data analyses by linking the CLS-HASI Minimum Data Set (individuals who entered CLS or HASI in the period 2002-2019 – ‘cohort data’) and that of a comparison group (sourced from the CLS-HASI program waitlists) with the following data sets: NSW Admitted Patient Data Collection (APDC),  NSW Emergency Department Data Collection (EDDC),  NSW Registry of Births, Deaths and Marriages (RBDM) (deaths),  NSW Mental Health Ambulatory Data Collection (MH-AMB),  Australian Coordinating Registry (ACR) Cause of Death Unit Record File (COD URF), and the Mental Health - Outcomes and Assessment Tool MH-OAT. The CHeReL also developed linkage with additional external datasets held by HASI program partner agencies including the NSW Department of Communities and Justice (DCJ) Housing, the NSW Bureau of Crime Statistics and Research (BOCSAR) and the NSW Corrective Services Offender Inmate Management (OIM) system.

The data linkage study will develop quantitative analysis of the linked program and outcomes data for CLS and HASI clients to measure change over time through time series analysis comparing before, during and after entry to the program. The quantitative analysis will be incorporated into health economic Markov modelling to assess program cost effectiveness. The findings from this social policy research study are expected to assess the quality, outcomes and benefits of CLS-HASI and policy implications.

 

Variations in surgical outcomes in New South Wales, Professor Louisa Jorm, University of New South Wales

Australia’s population is geographical dispersed and primarily located in major cities. Moreover, Australia has a unique mix of public-private hospital services with a complex funding model.  It is likely that these factors promote fragmentation of health care.

The aims of this data linkage study are to quantify and characterise variations in surgical outcomes (including complications, post-discharge care, ED presentation, readmission, mortality) in New South Wales patients in relation to public versus private hospitals and payers, hospital surgical volume and geographic remoteness and the adoption of newer technologies. Specific areas of focus are cardiac, orthopaedic, thoracic, vascular, abdominal, colorectal surgeries as well as urological, neurosurgical, and gynaecological interventions.

The cohort for this study (~4.5 million people) was identified by the CHeReL from NSW hospital records and included patients with a hospital separation in NSW from July 2001 to latest available, with any of the pre-defined surgical, interventional or a procedure code. Cohort data was linked by the CHeReL to the NSW Admitted Patient Data Collection (APDC), NSW Emergency Department Data Collection (EDDC), NSW Registry of Births, Deaths and Marriages (RBDM) (deaths), Australian Coordinating Registry (ACR) Cause of Death Unit Record File (COD URF) and the Australian Orthopaedic Association National Joint Replacement Registry. The Australian Institute of Health and Welfare further linked the cohort to Commonwealth datasets (MBS, PBS and National Death Index).

This retrospective population-based cohort study will explore care fragmentation, identify disparities in care and outcomes by subgroups including diagnoses, age, social class and indigenous status. By focusing on three key areas—the interplay of public and private sectors, geographic dispersion and centralisation of services, and adoption of new surgical technologies – the study will provide a comprehensive picture of surgical care outcomes in Australia.

 

Digital vs Screen-Film Mammography in Breast Cancer Screening: Impact on Screen-Detected and Interval Cancers’ Detection, Characteristics and Treatment, Dr Katy Bell, University of Sydney

Breast cancer is the most common cancer in women in the world and in Australia. Screening programs aim to reduce cancer mortality rates through earlier detection and early treatment of malignancies. To this end, population-wide mammography screening program was implemented in Australia in 1991. In recent years, full-field digital mammography (FFDM) has replaced screen-film mammography (SFM) throughout Australia. The effect of this move on health outcomes including screen-detection rates and interval cancer rates remains unclear.

This data linkage study aims to measure the effect of this change in practice on health outcomes through time trend analyses of individual patient data on breast cancer screen-detection rates and interval cancer rates before, during, and after the roll out of digital mammography in NSW.

Two cohorts were examined through this study; cohort A - all women ever screened and cohort B - all women ever diagnosed with breast cancer. The CHeReL facilitated this study by linking data from Breast Screen NSW, NSW Central Cancer Registry (CCR), NSW Admitted Patient Data Collection (APDC) and NSW Registry of Births, Deaths and Marriages (RBDM). Using linked data the researchers will capture women’s journeys from screened/unscreened to cancer/no cancer to treatment/no treatment to dead/alive and thus create a patient journey from screening to diagnosis to treatment to death.  An analysis of trends in the distribution of available tumour characteristics (such as stage, histology, and grade) is included in this study.

The project findings are expected to provide important information in relation to the incremental benefit and overdiagnosis attributable to changing from film to digital mammography in NSW.

 

Epidemiology, characteristics and management and outcomes of injury in individuals aged 0-25 in NSW, Associate Professor Julie Brown, Neuroscience Research Australia

Globally, traumatic injury is the leading cause of mortality, hospitalisation and ongoing disability in children over the age of one and in adults up to the age of 45. In Australia, national child injury hospitalisation and death rates have not decreased in the past ten years. The effects of traumatic injury can be devastating and long-lasting, with repercussions for the patient, their family and their community.

The investigators of this project aim to establish a program of population-based injury research focused on children and young people aged 0-25 years. Access to population-based injury data, and analysis will inform current and proposed injury prevention research, provide opportunities to analyse the effectiveness of injury prevention initiatives, allow evaluation of injury management and outcomes and to assess the impact of trauma management guidelines.

The study cohort includes all injured children and young people age 0-25 admitted to a NSW hospital during the study period or those who died as a result of injury sustained during the study period. The CHeReL facilitated this population-based retrospective cohort data linkage study by extracting cohort data from the Admitted Patient Data Collection (APDC) and linked to matching records in the APDC and NSW Registry of Births, Deaths and Marriages (RBDM) and Cause of Death Unit Record File (COD URF).

The study outcomes are expected to provide a much-needed baseline on which an evidence-based and strategic approach to reducing the burden of injury to children and young people in NSW can be developed. Importantly, this program of work will provide this for both injury prevention and injury management actions, and provide much needed resources for injury prevention regulators, policy makers and advocates, as well as health professional education, and trauma service planning.

 

Improving personalised decision making for oncology patients using data mining and rapid learning techniques, Associate Professor Lois Holloway, Liverpool Hospital, NSW

The rapid development and use of novel technologies and treatments for cancer, the difficulties of randomised clinical trials (RCT) in this area and the relatively small numbers of RCT patients as a proportion of all patients (3%) have all contributed to a growing evidence gap in cancer treatment decisions, particularly in radiation oncology. However, there is a huge amount of routine electronic clinical data collected on cancer treatments that could provide further evidence, complementing Randomised Clinical Trials data, to help support optimised clinical decisions.

The overall objective of this retrospective study is to develop and assess a data mining approach to radiotherapy data, wherein data from previous treatments are stored at the local institutions. Combined models are developed based on parameters determined from each of the local institution datasets to better inform clinicians and future patients. The project aims to test, use and further develop the distributed learning approach, initially for lung, head and neck and prostate radiotherapy treatments.

Patients identified in one of the 5 local radiotherapy datasets (i.e. patients with lung, head and neck or prostate cancer who have received radiotherapy) from five identified clinical sites in NSW formed the cohort and were linked by the CHeReL to NSW Cancer registry (NSW CR), NSW Clinical Cancer registry (NSW ClinCR), NSW Admitted Patient Data Collection (APDC), and NSW Registry of Births, Deaths and Marriages (RBDM) and Cause of Death Unit Record File (COD URF).

For each of the clinical sites an outcome prediction model will be developed based on available patient, disease and treatment characteristics using distributed learning and then validated. These models are expected to provide clinicians with additional clinical evidence where clinical trial evidence is limited or lacking, and improve the treatment decision and prognosis information available to patients and clinicians.

 

 

Impact of donor, recipient and transplant-related factors on kidney transplant and patient outcomes in Australia: The REnal TRansplant-related factors and Outcomes (RETRO) study, Associate Professor Germaine Wong, The Children’s Hospital at Westmead

Newer ways of matching donor kidneys to recipients have been developed and have the potential to revolutionise kidney transplantation but the evidence supporting their clinical use remains limited. This data linkage project aims to create and use linked datasets to evaluate the evidence for an alternative efficient method of utilising donor kidneys in Australia. To provide this evidence, detailed immunological and molecular data on the donors, recipients and potential transplant candidates from the National Organ Matching System (NOMS) were linked to the data from the Australia and New Zealand Dialysis and Transplant Registry and other health related datasets.

All recipients of kidney transplants and their respective donors as well as patients on the transplant wait list constituted the study cohort for this retrospective study. The CHeReL facilitated this study by linking the two cohort datasets and administrative data from the NSW Cancer registry (NSW CR), NSW Admitted Patient Data Collection (APDC), and NSW Registry of Births, Deaths and Marriages (RBDM deaths) and Cause of Death Unit Record File (COD URF), ACT Admitted Patient Collection (APC), ACT Cancer Registry (ACT CR) and ACT BDM Death Registrations.

Using existing genetic, immunological and healthcare data, this project aims to determine whether novel donor kidney allocation model utilising newer technologies may lead to better health outcomes compared to the current allocation model.

 

An evidence-based approach to improving outcomes and reducing hospital-acquired complications in patients with blunt chest injury: Chest Injury bundle of care Protocol (ChIP), Professor Kate Curtis, University of Sydney

Blunt chest injury resulting in rib fractures and hospital admission are rising in Australia. Failure to effectively treat blunt chest injuries (e.g.broken ribs) leads to hospital acquired complications: pneumonia, respiratory failure and unplanned intensive care admissions. These outcomes result in patient suffering and unnecessarily high healthcare costs.

This pre-post data linkage study will firstly determine incidence and outcome of patients with blunt chest injury in six NSW hospitals. The impact of implementing an evidence-based, multidisciplinary early notification mechanism and a chest injury care bundle protocol (ChIP), on the clinical and health service outcomes for patients with isolated blunt chest injury will be evaluated across three intervention sites and compared to three non-intervention sites. The researchers will also determine the rate of uptake of Chest Injury Protocol and identify any facilitators and barriers associated with the uptake of the protocol to inform wider implementation. Rates of non-invasive ventilation pre-and post-implementing the ChIP protocol compared to contemporaneous rates with current standard practice will be estimated

The Study cohort includes adults admitted with blunt chest injury via Emergency Department. The CHEREL facilitated this study by extracting cohort data from the NSW Admitted Patients Dataset collection (APDC) using identified diagnosis/procedure codes. Site medical records and “Patient with Acute Condition for Escalation” (PACE) activation data for the cohort were linked by the CHeReL to matching records in the APDC and Activity Based Funding (ABF) data.

This translational research study is expected to test a sustainable and evidence-based bundle of care protocol to reduce hospital-acquired complications associated with poorly managed chest injuries and improve the quality of care and patient outcomes.

 

Informing women about overdiagnosis in breast cancer screening: Effects on screening program participation, Professor Kirsten McCaffery, University of Sydney

Mammography screening can reduce breast cancer mortality. However, inconsequential disease that may never present clinically in a woman’s lifetime can also be detected by screening, leading to overdiagnosis and overtreatment. This is considered a serious and important harm of screening. There is a lack of evidence regarding how information on overdiagnosis affects women’s views and decisions about breast screening.

The aim of this BreastScreen data linkage project is to assess the effects of information about overdiagnosis on participation in the BreastScreen program. Specific outcomes include informed decision making, breast screening participation, and psychosocial outcomes over a period of 2 years.

The Breast Cancer Screening Information Study is a community-based, parallel-group, randomised controlled trial in NSW. Participants were randomly assigned to either the intervention decision aid (comprising explanatory and quantitative information about overdiagnosis, breast cancer mortality reduction, and false positives) or a control decision aid (including information about breast cancer mortality reduction and false positives). Self-reported study outcomes were assessed during four follow-up telephone interviews over a period of two years post-intervention.

The data linkage substudy included trial participants who provided written consent for linkage to their personal data held by BreastScreen NSW. The CHeReL linked the participant consent dataset (a database including all study participants who provided written consent for this data linkage) to the mammography uptake data from the BreastScreen program records and thereby enhanced the reliability and validity of the data for this important study outcome.

The overall study findings will provide important information on the effects of including information about overdiagnosis of breast cancer in a decision aid (evidence-based information resource) for women aged around 50 years who are considering mammography screening.

 

Surgical Care of Older PEople (SCOPE), Dr Lara Harvey, Neuroscience Research Australia.

Over half of all hospitalisations involving surgery in Australia are for older people, and there is evidence that there are wide variation in rates of surgical procedures being performed across geographical locations. Despite this, little is known at the population level about the hospitalisation experience and outcomes for older people who are hospitalised for a surgical condition.

This retrospective cohort data linkage study will examine hospitalisations for people aged 50 year and over, admitted to a NSW hospital for a surgical condition, during the ten year period 2007- 2016. Specifically, the project will provide hospitalisation profiles, examine trends, and quantify health service use and outcomes for this cohort. The project also aims to explore clinical variation in the care of older people for four high-volume clinical conditions, common in older people - cholecystitis, pancreatitis, appendicitis, and bowel obstruction. Factors that may contribute to unwarranted clinical variations will be explored and costs of differing models of care for the four selected surgical conditions will be estimated as part of this study.

The CHeReL facilitated this study by identifying the cohort individuals in NSW Admitted Patients Dataset collection (APDC) using identified diagnosis/procedure codes and linking to NSW Admitted Patients Dataset collection (APDC), Registry of Births, Deaths and Marriages (RBDM Deaths) and the Cause of Death Unit Record File (CODURF).

The research findings are expected to provide better understanding of the changing surgical profile in older people in NSW and this information can be used to identify priority areas of research as well as inform policy and models of care in NSW and beyond.

 

Survey of High Impact Psychosis (SHIP), Professor Vera Morgan, University of Western Australia

Physical morbidity, especially cardiometabolic disease, and premature mortality are elevated in people with psychotic illness. However, few studies have examined longitudinally the relationship between comprehensive clinical and other characteristics and mortality/morbidity in a large, representative sample of people with a range of psychotic disorders. These data are essential for understanding risk for avoidable mortality/morbidity in this population and informing services on appropriate and timely interventions.

The Survey of High Impact Psychosis (SHIP) data linkage study aims to estimate rates of mortality and morbidity with people with psychotic conditions, overall and by different conditions. Additionally, risk factors for increased morbidity and mortality in this population, risk prediction instruments for cardiovascular disease and costs associated with the additional morbidity and mortality will be estimated.

The SHIP is a comprehensively characterised and nationally representative sample of people with psychotic disorders. This retrospective study involves linking data of consented individuals from the SHIP cohort to morbidity, ED and cancer registry datasets in multiple Australian states and to the National death Index. For the NSW component, the CHeReL linked the cohort data to NSW Admitted Patient Data Collection (APDC), NSW Emergency Department Data Collection (EDDC) and the NSW Central Cancer Registry (NSW CCR).

Findings from this important study are expected to fill the knowledge gap on mortality and morbidity in people with psychotic disorders and help to address health inequities in this population.

 

The data linkage sub-study of the Expanded PrEP Implementation in Communities study (EPIC NSW), Professor Andrew Grulich, The University of New South Wales

For Gay and Bisexual Men (GBM) who are at high risk of HIV infection, pre-exposure prophylaxis (PrEP) against HIV is likely to represent the most effective means of HIV prevention. PrEP involves a combination of antiretrovirals taken on an ongoing basis to prevent HIV infection.

Despite the proven very high individual-level efficacy of PrEP, little is known about the population-level effectiveness of the implementation of PrEP. The Expanded PrEP Implementation in Communities (EPIC-NSW) study was a non-randomised implementation research project that aimed to assess the incidence of HIV among study participants and measure the population-level impact of the rapid roll-out of PrEP on HIV diagnoses among GBM in NSW over a two-year period. 

The data linkage sub-study aims to identify notifications of HIV and Sexually Transmitted Infections (STI) in NSW to optimise the ascertainment of these diagnoses in EPIC-NSW study participants.

The cohort for this prospective data linkage sub-study includes EPIC-NSW study participants who have provided their written consent for data linkage. The CHeReL facilitated this study by linking cohort data from the EPIC-NSW database to the NSW HIV Register and the NSW Notifiable Conditions Information Management System.

 

This data linkage will optimise the ascertainment of HIV and STI diagnoses arising from EPIC-NSW study participants and thus contribute to the overall aims of the study.

 

Investigating cardiovascular morbidity after injury and potential therapeutic interventions to mitigate disease, Associate Professor Janine Duke, University of Western Australia

Injury and trauma triggers systemic responses and the impact of these systemic responses can last long after the injury, leading to poorer future health for these individuals. In particular, cardiovascular dysfunction appears to be more common after injury. There is limited evidence that beta-blockers (a class of medications commonly used to treat cardiovascular disease) may be useful in reducing cardiovascular risk after injury.

This study will use linked administrative data to investigate whether individuals have greater morbidity after injury and whether specific medications reduce this morbidity. The study will also estimate the economic cost of injury and injury associated morbidity.

Participants from the 45 and Up Study who were hospitalised with an injury between survey completion and 2018 were the cases and those who completed the survey but had no such hospitalisation formed the control group for the case-control analysis. The CHeReL linked cohort data from the 45 and Up survey (linked to Medicare Benefits Schedule and Pharmaceutical Benefits Schedule) to the NSW Admitted Patient Data Collection, and the NSW mortality data (NSW Registry of Births, Deaths and Marriages and Australian Coordinating Registry Cause of Death Unit Record File).  

The study will compare the risk of post-injury morbidity (particularly CVD morbidity) and hospitalisation rates between the cohort and control group after adjusting for other factors and assess the economic cost of injury and subsequent morbidity. Within those with injuries, the study will determine if beta blocker medication use reduces the risk of post-injury CVD hospitalisations.

The knowledge gained from this project will be translated into improved clinical management of those with injuries, improving the well-being of patients with the aim to ultimately reduce post-injury cardiovascular admissions.

 

Quantifying the health and economic impacts of maintaining/adopting a healthy lifestyle on stroke survivors, Dr Wenbo Peng, University of Technology Sydney.

In Australia, nearly one million people are predicted to live with stroke by 2050 and the total burden of disease cost is estimated to be $49.3 billion. Lifestyle risk factors for stroke, including obesity, smoking, alcohol-dependence, and physical inactivity, account for approximately 80% of the risk of recurrent stroke of all types. Clinical Guidelines for Stroke Management have recommended lifestyle modifications for recurrent stroke prevention.

The overall aim of this project is to quantify the use of health services, health outcomes, and medical costs of Australian stroke survivors who maintain or adopt a healthy lifestyle. The researchers will profile stroke survivors who maintain or adopt a healthy lifestyle, compare their health service use with those who do not, and ascertain differences in medical costs (including Medicare and out-of-pocket costs) and health outcomes (including mortality rate) between the two groups.

The Centre for Health Record Linkage (CHeReL) facilitated this study by linking retrospective cohort data (i.e. 45 and Up study participants who had self-reported a stroke experience) from the 45 and Up Study to three routinely collected datasets - the NSW Registry of Births, Death and Marriages, the Australian Coordinating Registry Cause of Death Unit Record File, and the NSW Admitted Patient Data Collection while the Australian Institute of Health and Welfare provided Medicare Benefits Schedule and Pharmaceutical Benefits Schedule data via the Sax Institute.

The findings of this project may help empower stroke survivors to make necessary lifestyle changes to reduce their risk of recurrent stroke, aid clinicians and policy-makers in determining whether lifestyle change is a cost-effective care option for stroke rehabilitation, and pave the way to a stronger focus on secondary stroke prevention via lifestyle modifications.

 

Understanding potentially preventable hospitalisation among older Australians with chronic conditions, Associate Professor Megan Passey, University of Sydney

The number of potentially preventable hospitalisations (PPH) among older Australians continues to grow both internationally and in Australia. The overall aim of this project is to generate an evidence base identifying modifiable factors driving PPH admissions for chronic conditions amongst older community dwelling people.

The cohort for this study included eligible (community dwellers aged ≥ 45 years) patients with selected chronic PPH unplanned admissions from three NSW Hospitals. The CHeReL linked identified cohort (DaPPHne dataset) to their records in the NSW Admitted Patient Data Collection, NSW Registry of Births, Death and Marriages (Death registrations), the Australian Coordinating Registry Cause of Death Unit Record File and NSW Emergency Department Data Collection.

The data linkage substudy aims to develop models to predict the preventability of individual admissions using routine administrative data; investigate how additional items (e.g. measures of self-rated health) can improve the prediction of preventability; and investigate factors that influence longer-term outcomes following admissions classified as preventable and non-preventable, to explore and understand patient help seeking processes, experience of disease(s) management, and the contributions of social and personal factors to preventable admissions. Clinician assessment of the preventability of each admission will enable comparisons between admissions classified as preventable and not preventable.

The research findings are expected to improve measures of health system performance and develop effective interventions to reduce preventable admissions.

 

Investigating vaccine safety via cross-jurisdictional data linkage of the Australian Childhood Immunisation Register (ACIR) with hospital morbidity data - New South Wales (NSW) component, Associate Professor Michael Gold, University of Adelaide, South Australia

The success of Australia’s immunisation programs relies not only on optimising immunisation coverage but also ensuring immunisation safety, which includes the recognition of adverse events following immunisation.

The overarching aim of this project is to establish whether linked health data from multiple Australian jurisdictions (immunisation data linked to outcomes data (hospital separations and emergency department presentations) can be used in the assessment of vaccine safety.

A specific objective of the vaccine safety surveillance project involves determining whether the linked data can identify recognised serious and life-threatening adverse events following immunisation (AEFI).

For the NSW Component, the CHeReL linked records from NSW Admitted Patient Data Collection and the Emergency Department Data Collection for all NSW children aged less than seven years at admission to a public hospital between July 2003 to June 2013 and/or presentation to an emergency department presentation between January 2005 to June 2013. The Australian Institute of Health and Welfare will link to matching records in the core dataset for linkage Australian Childhood Immunisation Register (ACIR). Similarly, ACIR data will be linked to health outcomes data from other jurisdictions and the researchers will apply conventional epidemiological methods to demonstrate whether vaccine safety issues can be confirmed or refuted.

The findings from this project could provide a mechanism to identify potential safety issues following immunisation and allow risk-benefit evaluation of each vaccine.

 

Parenthood in patients receiving dialysis or kidney transplantation: A national study of perinatal risks and outcomes through population record linkage, Associate Professor Shilpanjali Jesudasan, The University of Adelaide, South Australia  

End-stage kidney disease (ESKD) is a serious health problem that can affect people of childbearing age. Despite reduced fertility and often substantial co-morbidity, many patients receiving renal replacement therapy (RRT; dialysis or a kidney transplant) do have babies. Women with ESKD experience high rates of adverse pregnancy outcomes and over half of the babies born to these mothers are born prematurely. Little is known about the outcome of these babies in the immediate perinatal period and into early childhood.

The aim of this project is to utilise existing perinatal, hospital, birth, death and renal registry data, to explore the outcomes of babies born to parents receiving RRT compared to the general population.

For this cross-jurisdictional study, state and territory-based births record data of babies and their mothers during 1991-2013 will be linked with other existing population-based datasets and with ANZ Dialysis And Transplant Registry (ANZDATA) dialysis or transplant (RRT) patients using local linkage organisations, to create a cross-jurisdictional dataset for analysis. The RRT patients in ANZDATA linked to perinatal or birth data constitute the RRT cohort and all non-linked perinatal data formed the non-RRT cohort.

For NSW, the CHeReL linked the ANZDATA Registry, the NSW Perinatal Data Collection, Neonatal Intensive Care Units' Data Collection, the ACT Maternal Perinatal Data Collection, NSW and ACT Admitted Patient Data Collection, ACT Emergency Department Data Collection and the NSW and ACT Births and Deaths registry.

The linked dataset will allow researchers to identify the RRT Cohort, assess birth rates in the cohort, and examine the association between RRT factors and pregnancy outcomes, preterm delivery, antenatal events, delivery complications and perinatal outcomes. The study will also assess longer term morbidity and socio-economic burden of mothers with end-stage renal failure, compared with non-RRT cohort and women with other chronic diseases.

Using data linkage approaches, this study will provide evidence to inform clinical care and improve management of patients who are considering or are faced with parenthood whilst receiving RRT therapy.

 

The Health Impact and Economic cost of Residential Fires in New South Wales, Dr Kathy Tannous, Western Sydney University

  In most western industrial countries, most of the fire deaths and a high percentage of fire-related injuries occur in the home. According to Fire & Rescue NSW (FRNSW), a total of 23,766 residential fires, were reported in New South Wales between 2010 and 2015, and survey data indicates that a larger number of fires are likely to have gone unreported.

This retrospective cohort study aims to determine the number of residential fires and their associated injuries and fatalities, both reported and unreported, in NSW during 2005-2015. This will be undertaken by examining reported fire incidents from FRNSW, ambulance service use, health service use (emergency and hospital admissions), and mortality data for people who had fire incident related injuries in that period. The study will utilise linked data to determine the number of injuries and fatalities, by mechanism, profile, and health service utilisation over this period and quantify the economic costs of residential fires to FRNSW, the healthcare system, and the wider community.

The study cohort comprised of residents who had experienced a residential fire identified in eight data sets: the Australian Incident and Report System (FRNSW AIRS); NSW Ambulance Computer Aided Dispatch; NSW Ambulance electronic Medical Records; Patient Health Care Record; NSW or ACT Admitted Patient Data Collection, NSW Statewide Burn Injury Service Data; NSW Register of Births, Deaths and Marriage; and Cause of Death records.  Demographic information was extracted from FRNSW AIRS Data Linkage was carried out by the CHeReL for residents of NSW that had experienced residential fire and linked to the for residents of NSW that had experienced residential fires and linked to matching records in the other NSW and ACT health-related datasets.

The research will provide a better understanding of number, mechanism, nature, spatial variation, and risk factors of injuries and fatalities following residential fires and their economic impact. This information will be used to identify priority stages for research as well as inform and influence policy and planning in NSW and beyond.

 

Recidivism, health and social functioning following release to the community of NSW prisoners with problematic drug use, an evaluation of the Connections Program, Professor Elizabeth Sullivan, University of Technology Sydney

The rising rate of adult incarceration in Australia is a major public health problem with high rates of return to custody, morbidity and mortality experienced by inmates, especially those with problematic drug use. The NSW Connections Program has provided targeted individualised support at all adult correctional centres across NSW for inmates with problematic drug use in their transition post release to the community.

This project will evaluate the Connections Program and determine if the model of care reduces recidivism, mortality and improves the health of patients on the Connections Program. Furthermore, it will examine the impact of the program on perinatal and infant outcomes for inmates who recently became a new mother and on their infants.

The cohort for this retrospective record linkage study included all patients in the Connections program database during the period 2008-2015. The CHeReL linked the Connections data to the NSW Admitted Patient Data Collection (APDC), Emergency Department Data Collection (EDDC), Mental Health Ambulatory Data Collection for hospital and emergency department admissions and ambulatory mental health care episodes, to NSW Bureau of Crime Statistics and Research Re-offending database for offence information and recidivism; to Health Pharmaceutical Drugs of Addiction System for opioid substitution therapy information; and to Registry of Births, Deaths and Marriages (RBDM) and Cause of Death Unit Record File to identify mortality outcomes.

For the parents sub-study, the Connections dataset was linked by the CHeReL to RBDM Births and the birth registrations were linked to NSW Perinatal Data Collection (PDC), Perinatal Death Record, APDC and EDDC to examine maternal outcomes for  women on the Connection Database and infant outcomes for all mothers referred to the Connections Program.

This study will provide a methodology for ongoing monitoring of the Connections Program; and provide evidence as to whether the Connections Program should be trialled more widely in the prison system.

 

Exploring the relationship between mental illness, traumatic brain injury and dementia and family and domestic violence using Text Mining and data linkage, Professor Tony Butler, University of New South Wales

This study addresses a major public health problem in Australia - Family and Domestic Violence (F&DV) – through the innovative use of Text Mining and data linkage to better understand the relationship between mental illness, traumatic brain injury and dementia among perpetrators and victims and FDV incidents, recidivism and judicial outcomes.

The researchers have earlier accessed the police's written text narratives of F&DV events which occurred between 2005 and 2016 in New South Wales (COPS data) and demonstrated the feasibility of Text Mining from COPS narratives to identify the mentions of mental illness, traumatic brain injury and dementia among the persons of interest (perpetrators) and victims.

The proposed study aims to examine the veracity of these ‘mentions’ extracted from the COPS narratives using Text Mining through analysis of a linked dataset. The CHeReL facilitated the study by linking the COPS data, NSW Health diagnostic data from Admitted Patient Data Collection and Emergency Department Data Collection, mortality data  from NSW Registry of Births Deaths and Marriages and ACR Cause of Death Unit Record File and records from NSW Bureau of Crime Statistics and Research Re-Offending data (BOCSAR ROD) data.  Linkage of the COPS and NSW Health diagnostic data also allows the researchers to examine the proportions of the F&DV victims and perpetrators who were diagnosed with mental illness, traumatic brain injury or dementia and compare the characteristics of perpetrators and victims and examine features of FDV events in those with and without a diagnosis of these conditions. 

Research findings are expected to offer direct benefits to the NSW mental health, police and corrective service systems in terms of identifying the health risks of domestic and family violence. 

 

The DRIVE Study: where are they now? 12 years of follow-up for 20,000 novice drivers in NSW, Professor Rebecca Ivers, The George Institute for Global Health

Reliable evidence about the causes of young driver injury crashes is required if the substantial burden imposed by these injuries on young people in Australia and internationally is to be reduced. The primary aim of the DRIVE Relinkage study is to identify the direction and size of associations between postulated risk factors and the incidence of young driver crash and injury. The secondary aim is to determine whether factors associated with increased risk of driver injury are comparable across different socioeconomic, ethnic and cultural groups.

The DRIVE Study recruited and surveyed 20,822 novice drivers, aged 17-24 years, during 2003-2004, with a follow-up resurvey of some participants after a year. In the DRIVE linkage study, survey data was linked with up to approximately 2 years of follow-up licence, offence, crash and health data (from participant’s first point of licensing to 31 December 2005). The original DRIVE study investigated the importance of potential determinants of motor vehicle crashes and injuries among young drivers. The DRIVE Relinkage study aims to extend the original study by examining linked hospital, crash and death data over a 12-year follow-up period enabling researchers to investigate long-term effects of novice driver behaviour and attitudes

To facilitate this study, the CheReL linked cohort data to records related to motor vehicle accident admissions from the NSW Admitted Patient Data Collection, mortality data from NSW Registry of Births Deaths and Marriages (Death registrations) and ABS Cause of Death of Unit Record File (COD-URF), and crash data from the Centre for Road safety’s CRASHLINK.   

Findings to-date from the DRIVE study have been fundamental in informing risk factors for young driver crashes in NSW and the extended study is expected to provide information on the long-term effects of novice driver behaviour and attitudes.

 

Cancer and Reproductive Outcomes for Women: a population-based cohort study, Professor Elizabeth Sullivan, University of Technology, Sydney

Cancer treatment has implications for subsequent pregnancies. Moreover, as women delay childbearing until a later age, an increasing number of women are diagnosed with cancer around the time of pregnancy and this presents unique challenges with clinical management and burden for both the mother and baby. Cancer is the second most common cause of death among women of childbearing age. However, there is a lack of epidemiological data on reproductive outcomes in cancer or the role of reproduction in cancer outcomes.

To address the evidence gap, this project investigates the inter-associations between cancer and pregnancy, to better understand the birth outcomes and survival of women with pregnancy associated cancer (PAC); and cancer survivors who subsequently give birth. This study will provide information on pregnancy complications, and the maternal, perinatal and infant outcomes of women giving birth with PAC and in women who conceived and gave birth subsequent to their cancer diagnosis. This study will also examine the impact of giving birth on survivorship and morbidities of women with cancer.

The project consists of two population-based cohort studies, linked by the CheReL to multiple routine data collections in NSW over time, including the NSW Cancer Registry , NSW Admitted Patient Data Collection, NSW Emergency Department Data Collection, Perinatal Data Collection and Perinatal Death Reviews database, Neonatal Intensive Care Units’ Data Collection, NSW Register of Congenital Conditions, NSW Mental Health Ambulatory Data Collection and NSW Register of Births, Deaths and Marriages (Death registrations) and ABS Cause of Death Unit Record Files. The characteristics and outcomes of women will be captured comprehensively through the linked datasets and compared between subgroups depending on whether they have a history of cancer or have given birth. Such understanding is the first step towards improving care and optimising outcomes associated with pregnancy associated cancer. 

 

The D-Health Trial: Data Linkage, Professor Rachel Neale, QIMR Berghofer Research Institute

Research suggests that Vitamin D may play a role in many diseases including cardiovascular disease, diabetes, and cancer risk and survival, additional to its role in bone health. However, large population-based clinical trials are needed to determine if supplementation has health benefits.

The D-Health trial was established in 2014 to meet the need for such trials. Over 21,000 adults from the general Australian population were recruited and randomised to receive vitamin D or placebo, once per month for 5 years. The primary/secondary outcomes of this trial are mortality and cancer, although there is a wide range of other outcomes.

The data linkage component of the D-Health Trial seeks to link the trial cohort with administrative datasets from multiple Australian Jurisdictions to capture information about death, cancer, and health outcomes that frequently result in hospitalisation. For NSW and ACT, the CHeReL linked cohort data to NSW Cancer Registry , NSW Admitted Patient Data Collection, NSW Registry of Births, Deaths and Marriages (Death registrations) and ABS Cause of Death Unit Record File, ACT Cancer Registry, ACT Admitted Patient Data Collection, ACT Births, Deaths and Marriages (Death registrations) and Cause of Death Unit Record File.

The linked data will enable the parent trial to assess the effect of supplementing Australians aged 60-84 years with vitamin D for 5 years on key health outcomes. The projects also aims to contribute to individual patient data meta-analyses to enable analysis of the effect of vitamin D supplementation on less frequently occurring health outcomes.

Findings from the analysis of linked data are expected to provide robust estimates of effect of Vitamin D supplementation for a range of health conditions.

 

Melanoma Patterns of Care Study 10 year follow up study, Professor Anne Cust, University of Sydney

The Melanoma Patterns of Care Study (MPOC) is a population-based, observational study based on doctors’ reported clinical management of melanoma patients diagnosed with an in-situ or invasive primary melanoma over a 12-month period from October 2006 to 2007.

The aim of the data linkage study is to collect 10-year follow-up outcomes for melanoma patients in the Melanoma Patterns of Care (MPOC) study to determine the risk of recurrence and death associated with different aspects of melanoma diagnosis, clinical management and patient and histological factors. The study also aims to assess the risk of new primary melanoma and other cancers and determine whether planned follow-up, patient and histopathological factors and compliance with clinical practice guidelines are related to the stage and the other characteristics of new primary melanoma, or occurrence of other type of cancer.

Data was sourced from the NSW Cancer Registry (NSWCR) master database and data variables extracted from pathology reports held by the NSWCR. Data linkage was performed by the CHeReL to the NSW Admitted Patient Data Collection, NSW and ACT Registry of Births, Deaths and Marriages (Death registrations), ABS Cause of Death Unit Record File and The Melanoma Institute of Australia Medical Record Database.

Obtaining follow-up data on recurrence and vital status for MPOC participants provides a unique opportunity to inform clinical practice guidelines and impact on patient care in Australia and internationally. The information will also add to ongoing research on the aetiology of melanoma and whether there are shared underlying risks for different cancer types, based on genetic and treatment related factors.

Medically Assisted Reproduction: The role of IVF, ovulation induction and subfertility on Reproductive and Infant Outcomes, Professor Georgina Chambers, University of New South Wales

One in 25 children is conceived in Australia using Assisted Reproductive Techniques (ART). A further unknown proportion is conceived using prescribed medications in ovulation induction. The lack of conclusive findings on the role of these medically assisted fertility treatments on reproductive outcomes and the health outcomes of the children has created an increasing important gap in evidence to guide clinical practice.

This project aims to determine the number and proportion of children conceived by medically assisted fertility treatments, thus providing evidence on the role and contribution of ART and non-ART treatments to clinical care, family formation and population demographics in Australia. The project will also estimate reproductive and obstetric outcomes of women following fertility treatments and quantify the risk of adverse health outcomes in children under two years conceived through fertility treatments and common variants of ART practice compared to spontaneously conceived children.

This retrospective cohort analysis was facilitated through data linkage of NSW and ACT administrative datasets carried out via the CHeReL followed by linking of Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Schedule datasets by the AIHW. The CHeReL linked cohort data from the Australian and New Zealand Assisted Reproductive Technology Database (ANZARD) to NSW and ACT Perinatal Data Collections (PDC), NSW & ACT Registries of Births, Death and Marriages death registrations, NSW Register of Congenital Conditions, NSW & ACT Admitted Patient Data Collection, Cause of Death Unit Record File.

Data Linkage will enable validating the ANZARD database against the MBS and the NSW/ACT PDC. Findings from this study are expected to provide important evidence for advising patients, clinicians, and policy makers on the role of medically assisted fertility treatments in clinical practice and their safety.

 Maternal and environmental risk factors for congenital conditions in New South Wales, Australia, Dr Catherine Chojenta, University of Newcastle, NSW

The aim of this project is to examine the maternal sociodemographic and physical environment in relation to congenital conditions in NSW and to examine hospital use for congenital conditions. For this project, one condition (gastroschisis) and one major contributing factor (maternal diabetes), both of increasing prevalence in NSW, were chosen as a case study by the researchers. Specifically, a matched case-control study to examine the risk factors for gastroschisis and the relationship between maternal diabetes and congenital conditions will be examined.

To achieve these aims, a population-based data linkage study will be conducted; five years of available data (2011-2015) on the Register of Congenital Conditions RoCC was linked to NSW Perinatal Data Collection the PDC by the CHeREL. All births recorded on the PDC for the five-year period (approximately 500,000 births) and all available information regarding the mother’s sociodemographic risk factors will be examined.

The proposed study will make a major contribution to knowledge by using large scale surveillance data to examine both the sociodemographic and physical environment in relation to congenital conditions in NSW. The project will examine any geographical clustering, providing critical information regarding the environmental risks for congenital conditions. Clustering of sociodemographic factors will also be considered in order to examine unique associations between factors and risk for congenital conditions.

Through this study, the researchers hope to develop evidence to emphasise the need for an optimal maternal and physical environment in order to reduce the rates of congenital conditions.

Genetic contributors, clinical course and pharmacogenomics of Bipolar Disorder using data linkage in NSW, Associate Professor Janice M Fullerton, Neuroscience Research Australia

Bipolar disorder (BD) is a major mood disorder characterised by periods of mania and depression and is ranked in the top 20 most disabling disorders globally.

The overarching aim of this project is to gain a greater understanding of the causes and treatment of bipolar disorder (BD), examine the clinical, genetic and pharmacogenomic aspects of BD risk and response, as well as the genetic determinants of general medical comorbidities.

In the first phase, individuals already participating in the Sax Institute’s “45 and Up” Study cohort were linked to NSW administrative health data (Admitted Patient Data Collection, Emergency Department Data Collection, Mortality Data (Registry of Births, Deaths and Marriages, and ABS Cause of Death Unit Record File), and Mental Health Ambulatory Data Collection) via the CheReL and to Commonwealth Pharmaceutical dispensing and Medicare data (PBS and MBS) through the Sax Institute to identify participants likely to have BD. In the next phase, 3,500 individuals with a prpbable BD diagnosis will be invited to participate in a BD substudy. The substudy participants will complete a questionnaire on lifestyle, clinical symptoms and treatment experience, and provide biological (blood) samples for genomic and biomarker analysis.

The researchers aim to characterise the genetic determinants of BD risk and severity, and identify key genes and pathways for treatment response, informing future biomarkers for personalised medicine. Finally, the study will examine the clinical and genetic relationships between BD and common comorbid conditions and quantify the long-term outcomes of people with BD.

 

 Linked futures: maternal mental health and children’s development through linked longitudinal data, Professor Rosa Alati, School of Public Health, Curtin University

Existing evidence suggests that exposure to prenatal depression, anxiety, and stress increases the risk of behavioural problems and reduces cognitive ability in early childhood. There is, however, limited literature on the longer-term behavioural and cognitive outcomes among offspring prenatally exposed to depression, anxiety, and stress. Further evidence of the potential for such relationships is needed to inform best practice decisions about treatment.

The aim of this project is therefore to assess the impact of maternal prenatal mental health problems and psychotropic medications on behavioural and educational trajectories of adolescent offspring in the general population.

The longitudinal cohort study is based on administrative data and includes all individuals born (live births) in NSW in 2003-2005, and their mothers and fathers. The CHeREL facilitated the study by linking multiple NSW data collections to create a birth cohort to which the Commonwealth datasets Medicare Benefits Schedule and Pharmaceutical benefits Schedule were linked by the AIHW for the period 2003-2017. The cohort data was linked by the CheReL to Perinatal Data Collection, Admitted Patient Data Collection, Mental Health Ambulatory Data Collection, NSW Registry of Births, Deaths and Marriages Birth and Death registrations, the National Assessment Program – Literacy and Numeracy, Australian Early Development Census (2009), Emergency Department Data Collection, and the Neonatal Intensive Care Unit Data Collection. 

By utilising administrative health data, the researchers have created the first big data asset of its type in this field, with the capacity to clearly demonstrate the broader and longer-term neurodevelopmental outcomes resulting from such exposure in terms understood by strategic health planners, clinicians, and the public.

 

The burden, costs and outcomes of firearm injury in New South Wales, Prof Natasha Nassar, The University of Sydney

Firearm injuries have a substantial impact on the health and safety of populations. These injuries can be regarded as unintentional (or accidental) or intentional (due to self-harm, or assault). There is limited data on healthcare use and outcomes, especially long-term outcomes of firearm-related injuries in Australia. This information is important for those affected by firearm injuries and their families, and for policy makers.

This population-based record-linkage cohort study aims to investigate the burden of firearm injuries and describe the extent, and longer-term outcomes associated with firearm-related injury in NSW. 

The study cohort was defined as registrations with codes indicating firearm injury in hospital admission data, emergency department, and mortality data since 2002. The cohort was identified by the CHeReL from NSW Admitted Patient Data collection (APDC), Emergency Department Data Collection (EDDC) and death registrations (Registry of Births, Deaths and Marriages (RBDM) and Cause of Death Unit Record File), and linked to matching records in the APDC, EDDC, RBDM and COD-URF and also to mental health data (Mental Health Ambulatory Data Collection and Mental Health Outcomes & Assessment Tools Data Collection) for each member of the cohort.

It is expected that this study will demonstrate that the immediate and long-term outcomes of firearm injuries, places a significant burden on individuals, communities and healthcare resources in NSW.

 

Long term follow up of participants in the ONTRAC skin cancer chemoprevention study (Oral Nicotinamide To Reduce Actinic Cancer), Professor Diona Damian, The University of Sydney

Skin cancer is Australia’s most common cancer; more than half of all Australians are likely develop skin cancer by age of seventy years. Nicotinamide (vitamin B3) has been shown to be highly effective in preventing skin cancers in animal models. Invitro and animal studies suggest that nicotinamide may have also have protective effects against melanoma and against internal malignancy respectively, however, it is not yet known whether nicotinamide will have the same protective effects in humans.

The researchers of this study earlier completed a double-blinded randomised controlled trial of oral nicotinamide in otherwise healthy volunteers at high risk of nonmelanoma skin cancer (NMSC; BCC and SCC). The ONTRAC study (Oral Nicotinamide To Reduce Actinic Cancer) showed a 23% relative reduction in the rate of new NMSCs in patients randomized to receive nicotinamide tables 500mg twice daily for 12 months within the period July 2012- June 2014 as compared to the placebo group.

The aim of this retrospective cohort follow-up study is to estimate the five-year cumulative incidence of melanoma, of internal malignancy and of mortality using linked hospital and mortality data within the cohort of participants in the ONTRAC skin cancer chemoprevention study. The researchers will also assess the legacy effects of oral nicotinamide on these outcomes.

The CheReL facilitated this study by providing 5-year follow-up data on ONTRAC participants using routinely collected data available via health record data linkage. To obtain the most recent data, the CheReL linked ONTRAC cohort data to the NSW Admitted Patients data Collection (APDC) and the NSW Registry of Births, Deaths and marriages (RBDM). The follow-up linked data is also expected to help inform the design of subsequent randomised trials of nicotinamide.

 

Air pollution and environmental exposures and mortality and morbidity in adult Australians: a large population-based cohort study, Professor Guy Marks, The University of New South Wales

Exposure to ambient air pollution is a leading contributor to the global burden of disease. Studies in moderate to high pollution environments in North America, Europe and Asia have established an association between long-term exposure to air pollution and an increased risk of Cardiovascular Disease, respiratory diseases, lung cancer, stroke, and all-cause mortality.

This important study of Australian aged 45+ years will advance knowledge about the health effects of exposure to environmental risk factors such as air pollution, green space, noise, and influences such as urban environments, energy transitions, social connectedness and walkability. The study will use air pollution as a priority exposure for analysis, but also aims to explore the health impact of the other environmental factors. Specifically, this study will quantify the mortality and morbidity in adult Australians which is associated with long-term exposure to air pollution. Innovative statistical methods will be used to assign air pollution exposures to the large, established 45 and Up Study cohort to determine the impact of air pollution on their health and mortality.

The study cohort includes all participants in each wave of data collection in the 45 and Up Study including the SEEF sub-study and will include prospective data from the 3rd follow-up wave. Cohort data was linked via the CheReL to hospitalisation, mortality and cancer data from NSW Admitted Patients data Collection, NSW Emergency Department Data Collection, NSW Registry of Births, Deaths and Marriages Death registrations and ABS Cause of Death Unit record File and the NSW Cancer Registry.

The overall aim of this work is to enable cost-benefit analyses for environmental planning and control by establishing valid and precise estimates for the position and shape of the air pollution exposure concentration and adverse health effect response functions for particulates and gaseous air pollutants in the low to moderate exposure range.

 

Adult Literacy and Aboriginal Community Wellbeing Study, Associate Professor Robert Boughton, The University of New England

Adult Indigenous Australians have low literacy levels comparable to those in the world’s lowest income countries, including the poorest of sub-Saharan Africa. Over 35% of the Indigenous adult population are reported to have minimal English language literacy, with the figure rising significantly in remote areas. The ‘social and economic costs’ to those with very low literacy are formidable; poverty, unemployment, poor health and crime.

The aim of this study is to analyse the impact of improved adult literacy on a range of wellbeing indicators among Indigenous Australians. The study aims to compare changes in health outcomes, health service utilisation, and contacts with crime and justice before and after the implementation of the literacy program in graduates (participants) to the non-graduates (low literacy individuals who declined participation) to determine the effect of the literacy program.

Participants in the Literacy for Life Foundations adult literacy campaign and matched control group of non-participants from the same region form the study cohort. Linked data will be used to validate self-reported responses from questionnaires, and to analyse longer term changes to health and social well-being outcomes. The cohort data from Adult Literacy and Aboriginal Community Wellbeing Study dataset was linked to hospitalisation data from the NSW Admitted Patient and Emergency Department Data Collections, mortality data from the NSW Registry of Births, Deaths, and Marriages and ABS Cause of Death Unit Record Files, crime-related data from the Bureau of Crime Statistics and Research data (BOCSAR),  and other health system related data - NSW Breastscreen , NSW Pap Test registry, NSW Ambulance and Notifiable Conditions Information Management System. 

This novel project is the first in Australia to investigate the effect of implementing an adult literacy campaign in low literacy Aboriginal adults.

 

A contemporary population study on short and long-term impact on morbidity and mortality in cardiology patients at a metropolitan tertiary hospital, Professor Leonard Kritharides, The University of Sydney

Rapid access chest pain clinics are increasing used in New South Wales to risk stratify patients who present to the emergency department with chest pain. Patients are reviewed in clinic often after further cardiac testing. Data on chest pain clinics in Australia is limited and there is no mortality data at a state level for rapid access chest pain clinics in Australia.

Using linked data, this retrospective cohort study aims to investigate the morbidity and mortality outcomes for a single-centre cohort of patients seen in the cardiology clinics at a metropolitan tertiary hospital and patients who underwent a non-contrast CT chest. The study will also investigate the incidence of fatal cardiovascular disease, and hospitalisation for non-fatal coronary disease, non-fatal stroke, atrial flutter/fibrillation according to a history of chest pain and incidental coronary calcium in non-contrast CT chest scans. Healthcare resource utilisation of patients seen in the rapid access chest pain clinics will be assessed by observing their hospital readmission rates. Potential healthcare savings by preventing hospital admissions and readmissions due to access to the rapid access clinic will be estimated.

All patients who have attended the Concord Repatriation General Hospital rapid access chest pain clinic since 2008 and hospital inpatients who had a non-contrast CT scan since 2011 (excluding those history of coronary artery diseases) constituted the study cohort. To facilitate this study, the CHeReL linked cohort data to hospital and mortality data from NSW Admitted Patient Data Collection, the NSW Emergency Department Data Collection, NSW Registry of Births, Deaths and Marriages Death registrations and ABS Cause of Death Unit Record Files datasets.

Findings from this study will provide valuable information on chest pain clinics and health outcomes of patients who are reviewed in these clinics.

 

Prudent advice on prevention of chronic diseases in Australia – Translating best evidence, using the 45 and up study cohort data (Program Linkage), Dr Martin McNamara, The Sax Institute

In Australia, chronic diseases (CDs) are the leading cause of ill-health and death, with 73% of deaths in 2013 due to one of 8 CDs. This retrospective cohort study series is a program of work that aims to use existing 45 and Up survey data linked to administrative datasets to examine various lifestyle factors identified in the literature as being associated with chronic diseases. To understand the burden of disease attributed to those risk factors, linked data will be analysed for risk factors, their role in the mortality and morbidity rates for the participants presenting them and their pattern of healthcare usage will be examined.

In the first instance, the researchers aim to derive evidence on the relative contribution of different risk factors to dementia prevention and progression and the impact of modifying these risk factors on dementia outcome. Dementia, a major CD, is now the leading cause of death for Australian women and the leading cause of disability in older Australians. The body of evidence for dementia is less robust, and the lack of awareness and understanding of dementia can result in barriers to its prevention and care.

For this study, the CHeReL linked the 45 and Up data to the NSW Admitted Patient Data Collection, Emergency Department Data Collection, NSW Cancer Registry, NSW Register of Births, Deaths & Marriages Death Registrations, ABS Cause of Death Unit Record File and NSW Mental Health Ambulatory Data Collection. Commonwealth data from Medicare Benefits Scheme, Pharmaceutical Benefits Scheme, National Death Index and National Aged Care Data Clearing house were linked by the AIHW.

This program of work aims to generate and translate evidence to create awareness and knowledge, to inform public health policy makers and interventionalists for a range of CDs. The researchers plan to develop a series of easy-to-use evidence-based resources to support and emphasize interventions to provide an effective and sustainable healthcare system for Australians.

 

Mapping correlates and predictors of youth perpetrated domestic and family violence on the Mid to North Coast of NSW, Professor Marie Hutchinson, Southern Cross University, NSW

This study forms the first part of a larger multi-study thesis on Domestic and Family Violence (D&FV) in NSW. The objective of the first phase of this study is to identify antecedents and pathways that contribute towards children and young people engaging in D&FV on the mid to north coast of NSW. Any act of violence, abuse or intimidation committed by members of a household including violence between children and their parents or carers and siblings, as well as intimate partners constitutes D&FV.

This study will examine the factors that contribute towards young people engaging in D&FV by linking data from the NSW Police Force and the NSW Ministry of Health. The researchers plan to use linked administrative data to identify the sociodemographic, family and individual characteristics of young people issued with a legal action for a D&FV offence and the association between these characteristics and severity of first offence, the risk and frequency of further offences and the risk of the young person committing other recorded non-D&FV offences in conjunction with D&FV offences. The study will also examine temporal processes and pathways for children and young people issued a legal action by the NSW Police Force.

Data pertaining to young residents (11-18 years) of mid to north coast of NSW, who were charged with a domestic violence related offence in the period July 2008-June 2018 was extracted from the COPs dataset and provided by the NSW Police to the CHeReL. The CHeREL linked the cohort data to NSW Admitted Patient Data Collection, NSW Emergency Department Data Collection and NSW Mental Health Ambulatory Data Collection.

The findings from this study will be used to inform local action, service developments and inter-agency service delivery and strategy. The overall aim is to build the capacity of services to better support vulnerable families and children residing on the mid to north coast of NSW. 

 

Reducing the evidence practice gap in preventing rehospitalisation and recurrence following stroke, Professor Amanda Thrift, Monash University

Stroke confers significant morbidity, mortality and economic costs. Approximately 50,000 Australians suffer a stroke each year. However, the uptake of proven treatments such as blood pressure lowering agents is poor.  There is also limited evidence of the effectiveness of the Chronic Disease Management (CDM) plans, which are aimed at providing better care in the management of risk factors in those with a chronic disease such as stroke.

The aims of this data linkage study are to (i) describe the uptake of CDM plans and evaluate their effectiveness, (ii) evaluate the effectiveness of guideline-based prescribed secondary prevention medications that are recommended in guidelines for stroke care, (iii) identify any variations in uptake, use and effectiveness of CDM plans and prescribed secondary prevention medications between urban and rural regions, and (iv) evaluate the cost‐effectiveness of CDM plans compared to other models of GP care. These endpoints will be assessed based on health outcomes at 1, 2 and 5 years after the stroke event.

The cohort for this study comprises of all patients presenting to the Emergency Department and admitted to hospital with stroke in NSW (and Victoria) over the period 2006-2012. Data will be extracted from Victoria and NSW’s hospital and emergency department data sets and submitted to AIHW for linkage with identified patient records in the National Death Index, Medicare Enrolment File, Medicare Benefit Scheme and Pharmaceutical Benefits Scheme. For NSW, the CheReL extracted data from the NSW Admitted Patient Data Collection, the NSW Emergency Department Data Collection.

Besides providing valuable evidence on the effectiveness of CDM plans for the management of stroke, the data linkage project will also demonstrate a pathway to achieve reliable cross‐jurisdictional data linkage between two states’ (Victoria and New South Wales) and national AIHW databases and contribute new knowledge in this field.

 

 

Health outcomes and service utilisation in a cohort of people who inject drugs, sex workers and ‘at risk’ young people - a record linkage study, Dr Bradley Mathers, The Kirby Institute, University of New South Wales

People who inject drugs, sex workers and young people engaged in street-based lifestyles (‘at risk’ young people) represent three overlapping groups who experience poorer health outcomes and a disproportionate burden of disease compared to the general population.

The aim of this project is to identify demographic, behavioural, and clinical factors that predict health and social outcomes in this socially marginalised urban population with high prevalence of injecting drug use and sex work. The study will also determine patterns and predictors of health service utilisation and identify predictors of uptake and outcomes of treatment including opioid substitution therapy and treatment for HIV and viral hepatitis.

 This project examines a cohort of clients attending a Medically Supervised Injecting Centre (MSIC) in inner city and those attending a primary health care service very close to the injecting centre. CHeReL linked the cohort to data from NSW Admitted Patient Data Collection and ACT Admitted Patient Collection, NSW Emergency Department Data Collection and ACT Emergency Department Information System, NSW and ACT Cancer Registry, NSW Mental Health Ambulatory Data Collection, Pharmaceutical Drugs of Addiction System NSW, NSW Notifiable Conditions Information Management System and ACT Notifiable Diseases Management System, NSW and ACT Perinatal Data Collection, National HIV Database (NSW Component), National AIDS Registry (NSW Component), NSW Bureau of Crime Statistics and Research Reoffending Database, NSW Offender Management Information System (OIMS), NSW and ACT RBDM Death Registrations, NSW Cause of Death Unit Record File and ACT coded cause of death.

The study is expected to provide a better understanding of the complex health and service needs of these groups towards improving the health of members of these vulnerable populations and reducing the associated burden of disease.

 

ADjunctive coRticosteroid trEatment iN criticAlly ilL patients with septic shock Health Economic & Outcome Research (ADRENAL Study), Dr Colman Taylor, The George Institute

In NSW, the estimated average cost per day in the intensive care unit is more than $4000 AUD. The costs of treating patients with sepsis and septic shock are thought to be considerably higher than the costs for treating patients without sepsis, in both intensive care and in the hospital ward due to prolonged stay.

The ADRENAL study is an NH&MRC funded randomised, blinded, placebo-controlled clinical trial assessing the effect of corticosteroids in treating patients with septic shock.  The primary outcome of is all-cause mortality at 90 days after randomisation. ADRENAL is recruiting from 69 intensive care units (ICUs) across Australia, New Zealand, Denmark, Saudi Arabia and the United Kingdom. 

The ADRENAL-CEA project is a national data-linkage based cost-effective analysis to determine the short- and long-term cost-effectiveness, economic and clinical outcomes of using hydrocortisone compared to placebo for treatment of critically ill patients with septic shock. The cohort includes all consented patients randomised to ADRENAL in jurisdictions where data registries are available to obtain resource use costing data. The ADRENAL study database includes information pertaining to the hospital admission including ICU resource use. The project included record linkage of the Adrenal study cohort to admitted hospital data and emergency data across multiple jurisdiction and national mortality data from the AIHW. For NSW data, CHeReL linked the cohort to the NSW Admitted Patient Data Collection and NSW Emergency Department Data Collection and the NSW RBDM death registrations.

The ADRENAL-CEA project provides an opportunity to look beyond the traditional 90-day mortality window and to consider the broader societal and economic implications of surviving septic shock.

 

Tracking the progression, outcomes and service use of people with hospital and community acquired pressure injury in NSW Australia. Professor Kim Usher, The University of New England

In hospital and community settings, Pressure Injury (PI) also commonly known as pressureulcers or pressure sores, are a leading cause of preventable patient harm and an important patient safety issue. PIs are adverse events associated with health care that can cause significant patient harm and discomfort, and have been identified as a cause of death. While there are point prevalence studies available, there is no current literature on the state-wide prevalence of PI or the economic burden of PI on the NSW health system.

 This retrospective audit study aims to estimate the annual prevalence of hospital- and community-acquired PI in NSW from 2010 to 2015, and to analyse the progression of injury stages, surgery, and/or mortality and to quantify the health-related costs for patients with a hospital- or community-acquired PI.

 The study cohort includes admissions for adult individuals with diagnosis codes relevant to PIs from the NSW Admitted Patients Dataset Collection. APDC data will be analysed to estimate the annual prevalence of NSW hospital acquired PIs.  Survival analyses and differences in survival among participant sub-groups will be conducted using linked NSW RBDM death data and NSW COD-URF (cause of death) records. Predictors of hospital-acquired PI progression and mortality will be identified. Similar analyses among community dwelling adults in NSW and variances between rural, regional and metropolitan areas and nursing home admission are also included in this study.

 PI prevention is an important aspect of health care to prevent the discomfort and morbidity associated with these injuries, as well as reducing health care costs. This important audit will provide more comprehensive and updated picture on the prevalence of hospital and community acquired pressure injury (PI) in NSW and progression, healing or mortality using population based data.

 

REDUcing the burden of dialysis Catheter ComplicaTIOns: a National approach (The REDUCCTION partnership project)

Associate Professor Martin Gallagher, The George Institute for Global Health

Healthcare Associated Infections (HAI) threaten the safety of patients, cause significant and life threatening harm to patients, and bring major additional costs to already stretched health budgets. Patients with kidney disease are especially susceptible to HAI, due to the wide use of central venous haemodialysis catheters. These catheters, while essential to the delivery of the life sustaining dialysis treatment, are a major driver of blood stream infection and contribute to the increased mortality seen in dialysis patients.

 The aims of the REDUCCTION data linkage project are to estimate the costs to the health systems associated with dialysis catheter exposure, estimate excess costs (using markers of health service use such as hospitalisation, inter hospital transfers, length of hospital stay) that arise from dialysis catheter infections, study the association between clinical outcomes such as in-hospital and long-term mortality and dialysis catheter exposure and infection.

 The REDUCCTION partnership project involves renal units across Australia. All patients receiving incident vascular access catheters for dialysis within the participating renal units during the study period defined the cohort. For the NSW & ACT components, standardised cohort data was linked to the NSW and ACT Admitted Patient Data Collections, NSW RBDM and the ACT Mortality data collection and the Australia and New Zealand Dialysis and Transplant Registry.

The findings of this study will further medical knowledge and may improve how renal physicians care for kidney disease patients requiring catheters in the future.

 

Examining utilisation of cardiac devices in New South Wales: a data linkage project

A/Prof Jodie Ingles, University of Sydney

Automatic implantable cardioverter defibrillators (AICDs) and transcatheter aortic valve insertions (TAVI) are two cardiac procedures that involve surgically inserting a cardiac device to prevent heart failure and/or sudden death. However, there is anecdotal evidence suggesting these devices may be overused in specific patient populations that receive limited clinical benefit from the therapy.

The aims of this retrospective study include characterising the AICD/TAVI population, reporting procedure trends over observation period and examining the association between patient sociodemographic, patterns of care and patient outcomes. This project will utilise population-level, routinely collected data to examine real world patterns of care and patient outcomes in persons receiving AICD or SAVR (standard aortic valve replacement) or TAVI implantation therapy in New South Wales.

For this project persons with an AICD or TAVI procedure code recorded in New South Wales (NSW) hospitalisation data from 2001 to 2017 were linked with NSW Admitted Patient Data Collection, NSW Emergency Department Data Collection and where relevant, NSW Mortality Data (New South Wales Registry of Births, Death and Marriages [NSW RBDM]; Cause of Death Unit Record File [COD URF]) mortality records. The AICD cohort was also linked with NSW Perinatal Data Collection records.

Evidence generated through this project is expected to assist clinicians in determining whether these therapies are suitable for their patients.

 

The Sydney Triage to Admission Risk Tool - 2 (START2): refining a disposition decision support tool to improve patient flow in Emergency Departments.

Associate Professor Michael Dinh, Royal Prince Alfred Hospital, NSW

Emergency Departments (ED) in NSW and Australia continue to face enormous challenges with respect to increasing population demand for health care, endemic overcrowding and access block. Novel solutions that are clinician focused and data driven are required to sustainably improve senior clinical decision making in ED, and these have not been described previously in the scientific literature.

This data Linkage project intends to identify risk factors for in-patient admission, define the most appropriate admitting clinical service, as well as flag patients who might receive more appropriate care in an outpatient setting or short stay medical unit.

The Sydney Triage to Admission Risk Tool - 2 (START2) project will refine will previously derived and internally validated a risk prediction tool for ED disposition called the Sydney Triage Admission Risk Tool (START). Data linkage between Emergency Department Data Collection (EDDC), Admitted Patient Data Collection (APDC), and Deaths Registry (NSW RBDM) to outcomes such as length of stay, re-admissions and medical diagnosis in ICD codes is expected to enhance the predictive capabilities of this clinical decision support tool to further streamline the patient’s journey.

 

Quality of Care in the Management of Cancer

Associate Professor Shalini Vinod, Liverpool Cancer Therapy Centre

There are little data in the Australian population on receipt of best practice care and clinical outcomes. This data linkage study is the first and largest population based study in Australia assessing the use of quality of care indicators and adherence to Clinical Practice Guidelines (CPG) in cancer management.

The data linkage project aims to investigate variations in cancer care by modality (surgery, radiotherapy, chemotherapy, palliative care) and cancer type, identify patient populations where there is a disparity, evaluate patient, tumour and health services factors and the correlation between receipt of quality of care and CPG adherence with clinical outcomes.

A retrospective cohort of breast, prostate, lung, colorectal, cervix or uterine cancer patients diagnosed since 2005 were linked by the CHeReL to the NSW Cancer Registry (NSW CCR), NSW Clinical Cancer Registry data (ClinCR), the NSW Admitted Patient Data Collection (APDC), NSW Registry of Birth Deaths and Marriages (RBDM) death registrations, and Cause of Death Unit Record File(COD-URF). The Australian Institute of Health and Welfare (AIHW) will link the cohort to MBS and PBS datasets.

Multiple comparisons of the quality of care indicators based on clinical evidence and major consensus guidelines or evidence summaries and some specific comparisons depending on the site of cancer are expected from this research. Findings are expected to provide a better understanding on the use of Quality of care indicators and translation of Clinical Practice Guidelines in cancer management in NSW.

 

The evaluation of the impact of a world first state-wide program in reducing cardiac arrests and other adverse events in Emergency Departments in New South Wales

Associate professor Jack Chen, The Ingham Institute, University of New South Wales

In order to prevent unexpected cardiac arrests and unexpected deaths in hospitals, the NSW government has taken a state-wide proactive approach by developing standardised policies and guidelines around the deteriorating patient and mandating that every NSW state public hospital implement a rapid response system or RRS (namely, the Between the Flags (BTF) program, Clinical Excellence Commission (CEC) NSW Health, 2008).

The data linkage project aims to examine the effectiveness of the implementation of a two-tiered RRS across every NSW Emergency Department; and to understand the incremental value of the current two-tier RRS system versus the one-tier Medical Emergency team (MET) system in the ED setting. 

Adult public hospital patients from four population administrative databases (NSW Emergency Department Data Collection (EDDS) database; NSW Admitted Patients Data Collection (APDC) database; the New South Wales part of the Australia and New Zealand Intensive Care Society (ANZICS) database managed by the Centre of Outcome Research(CORE); and the NSW Registry of Births, Deaths and Marriages (RBDM) from 1 Jan 2005 onwards to June 2017 were linked by the CHeReL, in order to assess if the hospital-wide RRS system across all NSW hospitals reduced unexpected cardiac arrests in the ED and hospital mortality and improved other patient safety indicators.

The Study will analyse these primary and secondary outcomes and their changes before, during and after implementation of the program.

 

 Towards Optimising Hospitalised Older Adults Medications

Professor Sarah Hilmer, Royal North Shore Hospital

Medicines use in older people is a complex balance between managing disease and avoiding medication-related problems. The prevalence of polypharmacy (concurrent use of five or more different medicines) is increasing, especially in people approaching the end-of-life and living with dementia. Polypharmacy increases the risk of medication errors, adverse drug reactions, falls, confusion, hospitalisation and death. One in five medicines taken by older people is harmful or unnecessary (inappropriate).

This data linkage project aims to define the prevalence and management of inappropriate polypharmacy and their clinical variation in hospitalised older patients (with and without dementia). An electronic decision support system prototype to support deprescribing practices will be developed and piloted followed by an evaluation for feasibility for use across all NSW public hospitals. Further, an economic analyses will be conducted to estimate the potential clinical and economic impact of the intervention to reduce inappropriate polypharmacy in older patients

This retrospective multi-centre cohort study consists of 2000 participants aged ≥75 years and admitted to six hospitals in New South Wales admitted under general medicine, geriatric medicine and rehabilitation. Longitudinal data on the cohorts’ clinical outcomes and medication use were obtained through data linkage via the CHeReL. Linked datasets included the NSW Admitted Patient Data Collection (NSW APDC), the NSW Emergency Department Data Collection (NSW EDDC), Registry of Births, Deaths and Marriages (RBDM) and the Australian Coordinating Registry Cause of Death Unit Record (COD-URF). Additionally data from the Pharmaceutical Benefits Scheme (PBS) and the Repatriation Pharmaceutical Benefits Scheme (RPBS) were linked by the Australian Institute of Health and Welfare.

The overall aim of this research is to identify and prioritise inappropriate polypharmacy and inform the process of implementing deprescribing interventions in the hospital setting.

  

Accuracy of multi-parameter first trimester screening in the identification of pre-eclampsia and fetal growth restriction in an Australian population.

Associate Professor Andrew Mclennan, University of Sydney

Pre-eclampsia (PE) is a multisystem disorder that affects 2-5% of all pregnant women and is the leading cause of maternal and fetal morbidity and mortality. Relying on obstetric, medical and family history to identify women at high risk of developing PE only identifies approximately 30-40% of cases and has a high false positive rate.  After identification, an effective intervention to reduce the prevalence of the disease remains a challenge.

The proposed data linkage study aims to validate first trimester multi-parameter screening test for the identification of pre-eclampsia in a large mixed-risk Australian obstetric population and, provide contemporary assessment of pre-eclampsia incidence. The study will also evaluate the use of a multi-parameter risk algorithm for the prediction of pre-term fetal growth restriction (FGR).

This retrospective cohort comprises of singleton pregnancies between 10 and 14 weeks gestation in women referred to a specialist prenatal screening and diagnostic ultrasound practice in metropolitan Sydney to undertake first trimester screening for adverse pregnancy outcome. For the data linkage project, the CHeReL linked cohort data (i.e. the first trimester screening database, from two sources- Sydney Ultrasound For Women (SUFW) and Royal Prince Alfred Hospital (for women screened in the first trimester at RPAH but not delivered at RPAH)) with matching mothers’ and babies’ records in the NSW Perinatal Data Collection.

The findings from this study are expected to provide important pregnancy and birth outcome information from the largest population of pregnant Australian women undertaking first trimester screening for pre-eclampsia. This will include valuable contemporary information regarding the true incidence of PE and gestational hypertension in Australia, the accuracy of first trimester screening for PE and the downstream benefits in terms of both health outcomes and costs from appropriately targeted interventions in a population identified at high PE risk.

 

Advance Care Planning for Patients with advanced illnesses attending hospital outpatient clinics: Data linkage

Professor Andrew Hayen, University of Technology, Sydney

Advance Care Planning (ACP) is a process of reflection, discussion and communication that enables a person to plan for their future medical treatment and other care, for a time when they are not competent to make or communicate decisions for themselves. ACP could significantly improve the quality of care provided to patients with advanced illnesses by allowing patients to receive patient-centred care and avoid unwanted or inappropriate hospital admissions and treatment.

The data linkage project is part of a larger study and randomised trial looking to see whether a facilitated ACP intervention provided to patients with advanced illnesses having supportive and palliative care needs and attending a hospital outpatient clinic can reduce unplanned hospital admissions at six months, improve patient care and encourage health professionals to incorporate ACP into routine care.

The Centre for Health Record Linkage (CHeReL) performed data linkage between the ACP project dataset and NSW Ministry of Health data from the NSW Ambulance dataset (NSW Amb), NSW Emergency Department Data Collection (EDDC), NSW Admitted Patient Dataset Collection (APDC) and NSW Registry of Births, Deaths and Marriages (RBDM) and the Australian Coordinating Registry Cause of Death Unit Record (COD-URF). The Linked data will be used to compare the intervention and control groups to measure key outcomes; unplanned hospital admissions, acute health services utilisation and health outcomes and to carry out a health economics evaluation.

The project findings will help determine if having facilitated ACP discussions influences patient outcomes or reduces acute health resources utilisation, to inform future practices and hopefully encourage health professionals to partake in ACP discussions.

 

Using linked data to evaluate and improve hip fracture care

Dr Lara Harvey, Neuroscience Research Australia

Hip fracture (HF) is the most serious and costly fall-related injury suffered by older people. Delivery of evidence based care for HF can reduce morbidity and mortality, improve function and optimise discharge destination and from a health service perspective, can reduce length of stay, complications and readmission rates.

Australia has an established Guideline for Hip Fracture Care, a Clinical Care Standard with Quality Indicators and the Australian and New Zealand Hip Fracture Registry (ANZHFR). The ANZHFR collects variables that are acute-care focused and provides hospitals with real-time data on performance against a number of quality indicators. The registry, however, does not collect data beyond the acute episode and cannot assist with understanding the long term impact and ongoing cost of HF.

This data linkage study aims to measure longer-term outcomes for HF patients at various health service levels, explore clinical variation in selected processes and short term outcome measures for HF patients.

Demographic and clinical information were extracted for all individuals aged >=50 years admitted to a NSW hospital with a fall related HF from 1 January 2010 from the NSW Admitted Patient Data Collection (APDC) and linked to the Registry of Births, Deaths and Marriage (RBDM), District and Network Return dataset (NDR), Sub-Acute Non-Acute Patient data (SNAP) and ANZHFR by the Centre for Health Record Linkage (CHeReL). The AIHW undertook further linkage to Commonwealth data from National Death Index (NDI), National Aged Care Data Clearing House (NACDC), the Pharmaceutical Benefit Scheme (PBS) and Australasian Rehabilitation Outcomes Centre Database (AROC).

The project will identify factors (patient/process/ system level) that may contribute to any identified unwarranted clinical variation and provide baseline trend data against which to calculate the clinical and cost-effectiveness of the introduction of the Guideline/Standard/Registry combination in NSW.

 

SAFety, Effectiveness of care, and Resource use among Australian Hospitals (SAFER Hospitals)

Dr Isuru Ranasinghe, Basil Hetzel Institute and University of Adelaide

In Australia, more than 10 million hospitalisations occur annually for treatment of a range of acute and elective conditions and this care is becoming increasingly expensive. However, very few outcomes of hospital care are measured or reported nationally despite increasing local and global concerns about variation, quality, appropriateness and escalating costs of hospital care.

The SAFety, Effectiveness of Care, and Resource Use among Australian Hospitals (SAFER Hospitals) study is a multidisciplinary research collaboration to evaluate the safety and effectiveness of hospital care, and the impact on cost and resource utilisation in Australian hospitals. The goal of this study is to estimate the hospital-wide incidence of serious adverse events, unplanned readmissions, mortality and costs following hospital care using existing linked data and advanced statistical and machine-learning based computational methods. It seeks to assess the variation in these outcomes among hospitals and estimate the proportion of outcomes that may be preventable.

The SAFER hospitals study is part of a larger national observational cohort study and includes data from multiple Australian jurisdictions. The data linkage for NSW and ACT data was carried out by the CHeReL. Cohort data included hospital admissions from 1 January 2012 from NSW Admitted Patient Data Collection (APDC) and ACT Admitted Patient Collection (APC). This data was linked to the NSW Emergency Department Data Collection (EDDC), and Registry of Births, Deaths and Marriages (NSW RBDM), and the ACT Emergency Department Information System (EDC), Re