Exploring the data collection process for a regional trauma registry

Utilising a regional web-based trauma registry to support trauma quality
improvement and injury prevention in the Midland region - a research
Bridget Kool, University of Auckland
November 2014
Trauma care systems
Organised and coordinated efforts in a defined
geographic area to deliver the full spectrum of care to
injured patients:
time of injury
rehabilitative care
Key elements of the Acute phase of trauma care:
Prehospital Care
Trauma System Coordination
Data Collection
Project Phases
1. A survey of people responsible for the collection of trauma registry
data regarding their perceptions of the training and support they
2. A survey of clinicians involved in the management of trauma
patients regarding the management of trauma patients in their
institution and the region
3. A study exploring the feasibility of linking St John prehospital data
with trauma registry and ACC data
Trauma Registry Data Collection
Trauma registries
• Essential component of any trauma system/service
• Source of information for injury epidemiology, performance improvement, efficacy of
clinical management for trauma patients, evaluating prevention efforts, conducting
outcomes research
• Types of data collected: demographic, injury circumstances, nature of injuries,
physiological parameters, processes of care, outcomes
• Time stamped data
• Data collection forms an important part of trauma registries
• Poor data collection results in missing data, false positives, false negatives, inconsistent
and incongruent coding, duplicate entries, redundant data and suppositious data
Midland Regional Trauma System: trauma registry
• Captures ~ 5000 people per year
• 85% single organ injuries
• Patients are selected prospectively in ED, according to the MRTS
inclusion and exclusion criteria
• Information collected from prehospital and hospital stay
• Data: demographic, injury circumstances, nature of injuries,
physiological parameters, processes of care, outcomes, quality filters
Project Overview
To gain a better understanding of the data collection process for the MRTS
Anonymous online survey (March 2014)
Participants: Staff responsible for the collection of data used in the MRTS trauma registry
Ethics approval: University of Auckland Human Participants Ethics Committee (Protocol number
010706), and the Waikato District Health Board (RD013071).
Results: Participants
• 5/7 took part - one from each of the 5 hospitals
• Mean length of time in the role 26 months (range 3-48 months)
• Spent on average 58% of their time completing TDFs (range 25-80%)
• Training and support:
• 5/5/had received training, 2/5 felt the training was inadequate
• 4/5 had received Abbreviated Injury Scale (AIS) training
• Most (3/5) felt that they received adequate support and guidance; 2/5 reported no regular
reviews, updates or guidance sessions to ensure consistency and accuracy of data collection
Results: Completion of Trauma Data forms
• TDF for minor traumas take 5 to 25 minutes to complete, major trauma TDFs
took 20 to 50 minutes
• 4/5 TDF was easy to work with
• 2/5 time allocated for data collection was adequate. Reasons for lack of
adequate time:
• lack of leave cover
• patient notes being unavailable (e.g. mornings)
• difficulty juggling their roles
Results: Trauma Data forms - demographics
• Ethnicity data sources: patient admission sheets (‘front sheets’), ACC forms, electronic patient
records, and from patients
• 2/5 had received ‘ethnicity data collection’ training
• Level of comfort collecting ethnicity data (1-very uncomfortable, 10-very comfortable): mean
rating 7.6 (range 2- 10)
• If more than one ethnicity located:
‘the major one’
‘the most obvious’
‘Sometimes people circle NZ Maori and Tongan, so I revert to where they were born’
‘the one the patient most identifies with’
Results: Trauma Data forms – clinical information
• ‘Major trauma’: 4/5 use “trauma call” and also the patients ISS
• ‘ISS’:
• 2/5 used ISS ≥15, 1/5 ISS >15
• 3/5 could calculate ISS (1 had an ISS calculator computer app)
• ‘ICU days’: inconsistencies when more than one period of stay in ICU (total days recorded,
or recorded in ‘issues’ section)
• ‘iatrogenic pneumothorax’: 2/5 – ‘adverse effects’, 1/5 – ‘complications’, 1/5 – in both
places, 1/5 – record one but location NS
Data collectors recommendations for improvements to data collection process
Key findings:
Overall data collection is of a high standard
Areas that could benefit from additional training and support have been identified
Some form modifications required and clarification in the data dictionary
Reassess role scope and funding
Quality and completeness of data recorded in medical record and ongoing issue
Findings consistent with international literature:
Collecting and recording prehospital events (Jones, 1995)
Under-reporting of complications (Zehtabchi, 2011)
Variability in definitions between centres (Owen, 1999)
Inconsistency or invalid records– 25% of NTDB records excluded (Hemmila, 2007)
Clinician survey
Project Overview
To explore clinician’s perceptions of the quality of trauma care delivered in the five hospitals of the
Anonymous online survey (Mid 2014)
Clinicians invited by email to take part
Domains of enquiry included: what currently guides clinical practice, clinician’s perceptions of the
quality of trauma care and presence or absence of inequities in trauma care in their region.
Ethics approval: University of Auckland Human Participants Ethics Committee (Protocol number
010706), and the Waikato District Health Board (RD013071).
Results: trauma care management
Overall clinicians felt their hospital and the region managed trauma well (mode 7;
1= very poor, 10=very good)
The majority felt comfortable managing major trauma (mode 8; (1- not very
comfortable, 10 very comfortable)
Perceived barriers: lack of knowledge regarding diagnosis and management:
Spinal trauma 29% (agreed or strongly agreed)
Chest trauma 24% (agreed or strongly agreed)
Abdominal trauma : 20% (agreed or strongly agreed)
Head injuries 14% (agreed or strongly agreed)
93% used the internet to aid clinical decision making ; 96% of those used the site
‘UpToDate’ (evidence-based physician-authored clinical decision support resource)
Results: quality and equity of care
• Over half (56%) the clinicians ‘agreed/strongly agreed’ that there was
consistency in the quality of trauma care delivered across their hospital
• Concern was expressed over: lack of consistency in trauma calls, reduced
senior management out of hours, and variability within clinicians
• 93% of clinicians who responded felt that there was consistency in the quality
of trauma care received by different ethnic groups across their hospital
• Organisation treats people unfairly based on:
How well they speak English: 19% ‘somewhat often’
How well educated they are: 11% ‘somewhat often’
Record Linkage
Project Overview
To establish the feasibility of linking data from prehospital emergency service providers to hospital trauma care
providers and ACC at the individual level.
200 St John records of acutely injured adults (≥ 15 year old) transferred to Waikato hospital during a 12
month period (Jan2012 to Feb 2013) were randomly selected. Patient status = 1 (immediate threat to life) or 2
(potential threat to life)
These records were probabilistically linked to the MRTS using the following fields: response date, patient
name, patient DOB/age
Once the records were linked, the person’s ACC claim number and National Health Index (NHI) number were
abstracted from the MRTS trauma registry
These records were then probabilistically linked to ACC claims data using the NHI and ACC numbers
• Of the 200 patients identified by St John, 43 were registered with the MRTS
• Of these 41 matched with an ACC record (95% match)
• Of the 2 records that did not have a positive match – both had differing dates of injury to
the MRTS and St John data
• Additional analyses will look at what factors made a positive match less or more likely
These small projects are helping us to identify opportunities to improve the
reliability of the data to inform quality improvement activities and monitor
the effectiveness of injury prevention activities
Other initiatives in the pipeline:
5 journal articles under development
A study looking at the reliability of ethnicity data in the MRTS trauma registry
A trial of an e-form for trauma registry data collection
HRC grant application ‘Trauma Track’
University of Auckland: Faculty Research Development Grant
Coinvestigators: Grant Christey, Shanthi Ameratunga, Ross Lawrenson, and Nina Scott
The survey participants
MRTS staff
[email protected]

similar documents