IT technology and networking to improve quality of care in high risk preterm infants
One to 2% of all births are very preterm born before 32 weeks of gestation. This group of infants contribute disproportionally to the burden of mortality and morbidity related to prematurity.
The spectacular development of neonatal intensive care allowed a drop in neonatal mortality of very preterm infants from 50% to less than 15% in the last decades. However, 15 to 25 % of the surviving infants will present neurodevelopment or physical impairment. The main complications are cerebral palsy, cognitive impairment, neurosensory function alterations, behaviour, attention deficit and hyperactivity disorders and bronchopulmonary insufficiency.
Furthermore an increasing number of diseases in adulthood like cardiovascular diseases or obesity have their origin in the pre- and perinatal period. Prematurity combined with growth restriction favours appearance of those diseases in later life.
Continuous improvement of neonatal outcomes is a major issue in public health and all our efforts should be dedicated to quality of care. Different approaches contribute to our aims: clinical and health services research, new treatments, organisation of care and health politics. At the level of an individual neonatal intensive care unit and in order to evaluate local practices, one of the strongest tools for improvement is benchmarking with other units. Establishing networks of neonatal intensive care units for registration of relevant items in large databases makes comparisons possible. Benchmarking and feedback to the units helps them in understanding their performance. These networks can be regional, national or international.
Networking platforms: evolving IT technology
The major problem encountered with patient data collection is the multiple data entry in local, regional, national and international databases which are not interconnected. Feedback to the neonatal units is often absent or confusing and therefore frustrating for the caregivers.
In Belgium the College of Physicians is responsible for quality control with electronic databases. The introduction of new web-based software "eNewborn" in 2008 increased voluntary participation of the neonatal intensive care units from 65 % to 100% in 3 years and this is maintained since. In the UK, the Royal College of Paediatrics and Child Health includes all 200 neonatal units in England, Wales and Scotland in a National Neonatal Audit Programme and the Neonatal Data Analysis Unit at Imperial College London similarly offers these neonatal units opportunity to participate in national and international research studies.
In 2011 eNewborn software was proposed to enrich the first European platform EuroNeoNet. Unfortunately the collaboration was interrupted because the leader of EuroNeoNet Professor Adolf Valls i Soler suddenly and unexpectedly died in 2013 and the network had to close their activities in 2014.
Since then Dominique Haumont, Neena Modi and Mark Turner put their efforts together to move forward and create a new benchmarking and evaluation programme: the eNewborn Project. Internationally defined neonatal items of patients characteristics are anonymously registered on a secure web-based platform.
The choice to federate networks with new IT software was the challenge undertaken. The demo describes how to benchmark interactively: htpps://www.enewborn.org
Expected outputs from the eNewborn Project:
- Improve quality of care by benchmarking and feedback to the units: comparisons are made by country/network and internationally
- Use the database for clinical research including randomised controlled trials
- Make links between neonatal care and later outcome
- Plan new research and define best therapeutic strategies
- Give young investigators opportunity to use IT technology to stimulate their clinical questions
- Monitor the health status of preterm infants over time
- Help decision makers to improve health care systems
The platform has data from Belgium, United Kingdom, France, and the Czech Republic. New collaborations with several countries will be discussed after the eNewborn meeting.
eNewborn is a partner of the International Neonatal Consortium (INC). INC is a coalition of industry, academia, patient representatives (EFCNI), the U.S. Food and Drug Administration (FDA), the European Medicines Agency (EMA), other governmental agencies, professional organizations, foundations and the Critical Path Institute that will focus on accelerating the development of safe and effective treatments for neonates.
eNewborn project leaders (pilot phase):
- Professor Dominique Haumont (emeritus) Neonatology Saint-Pierre University Hospital, Brussels Database expert at the College of Physicians, Federal Public Health Service, Belgium
- Professor Neena Modi Director, Neonatal Data Analysis Unit Professor of Neonatal Medicine,Imperial College London President, UK Royal College of Paediatrics and Child Health
- Dr Mark Turner Senior Lecturer in Neonatal Medicine, University of Liverpool Chair, European Network of Paediatric Research at the EMA
IT and biostatistics:
- Cuong Nguyen BA, MS IT consultant in HIS (Hospital Information System) at Saint-Pierre University Hospital, Brussels Collaborator at the College of Physicians, Federal Public Health Service, Belgium
- Eugene Statnikov BSc Senior Data Analyst Neonatal Data Analysis Unit Imperial College London
- Agueda Azpeitia Garcia, MS Biostatistician and data manager Bilbao, Spain Lan Phan Ngoc, MS IT project coordinator Brussels Administration (UK)
- Richard Colquhoun Manager Neonatal Data Analysis Unit Imperial College London
eNewborn software description Goals:
- Online interactive benchmarking and user friendly charting dedicated to neonatal networks
- Simultaneous access to benchmark statistics at different levels: region, country, international and customized networks
- Reporting on annual or pluriannual basis in comparison with the different network reference populations
- Database available for in depth epidemiological studies and research trials
The major highlights of eNewborn are:
- Intuitive, fast and easy to use
- Can be adapted very quickly to new items, functionalities and projects
- User can access database output instantly(validation to be defined)
- User can navigate in charts from top down to detail of population in drill down manner
- "State of the art" in Web application development
- Items can be registered on line or by uploading existing files 4 Data entry: Items which are registered
- EuroNeoNet General Data form
- On demand: Infections and patient characteristics according NeoKiss definitions Follow-up characteristics
- Harmonized international neonatal items
- Customized items for specific projects Statistical output:
- Mean (SD), median (range), whisker box plot, incidence and survivalin graphs for each item
- Control confidence interval (CI) compared with the reference network population
- Refining of the target population is possible with drill down using all combinations of the selected items over different time periods
- The selected population can be displayed in one click on a excel file for further statistical analysis for each individual neonatal unit
- Patient records are stored in Oracle Database as XML data
- Copies of each patient record are stored also in XML file
- Intensive use of statistical and analytical functions from Oracle
- Web server is running under Apache/PHP
- Client interface use
- Database Oracle (11g V 2)
- Php (v5.x) Apache (v2.2)
- Google API tools
- The Web site (eNewborn) is hosted in England (Enciva Solutions Ltd Tech Support)