Clinical decision support in emergency medicine : exploring the prerequisites
Author: Skyttberg, Niclas
Date: 2019-11-08
Location: Inghesalen, Widerströmska huset, plan 2, Tomtebodavägen 18 A, Karolinska Institutet, Solna
Time: 09.00
Department: Inst för lärande, informatik, management och etik / Dept of Learning, Informatics, Management and Ethics
View/ Open:
Thesis (1.106Mb)
Abstract
A clinical decision support system is a technical system that combines individual patient data and evidence-based clinical knowledge to give advice and support to clinicians. For quite a long time, the emergence of such systems has been predicted and expected to impact health care dramatically by improving both quality and productivity. Three factors make Swedish emergency medicine an interesting context which could be mature for the introduction of clinical decision support systems. Firstly, Sweden is a leader in the implementation of health care information technology, and the coverage of electronic health records is around 100% in the country. Secondly, emergency medicine is a field with high patient turnover, frequent decisions, and substantial impact on patient outcome. Thirdly, although there are abundant publications on clinical decision support system development and implementation in general, there is less knowledge of such systems in the urgent care context. Therefore, this doctoral project aimed to explore the prerequisites prior to implementation of clinical decision support systems in emergency medicine.
This thesis is based on a mixed-methods design and consists of four individual studies. Proctor’s conceptual model of implementation research was used as a framework for the project. Study I included semi-structured interviews with 16 medical doctors and nurses from nine Swedish emergency departments. Content analysis was used to describe factors affecting vital sign data quality in emergency care. Study II extracted vital signs from 330 000 emergency department visits to assess the effects of different documentation workflows on data quality. Study III prospectively explored 200 vital sign measurements from 50 emergency care visits to evaluate the impact of manual and automated documentation on vital sign data quality. Study III also used data from an adapted NASA TLX questionnaire to compare the workload of clinical staff (n=70) in manual and automatic documentation. Study IV used semi-structured interviews with 14 emergency medicine physicians from three different sites. Content analysis was used to explore participants’ expectations and concerns regarding clinical decision support systems.
There are three main results and conclusions from the research. Firstly, documentation of vital signs in the emergency department is still surprisingly paper-based, which makes vital sign data unfit for reuse in clinical decision support. Secondly, automation of vital sign documentation is feasible in emergency care and should improve data quality and reduce workload. Thirdly, enthusiasts towards decision support are at risk of disappointment with the level of innovation in the currently available decision support systems, and this may affect the implementation strategy negatively.
This thesis is based on a mixed-methods design and consists of four individual studies. Proctor’s conceptual model of implementation research was used as a framework for the project. Study I included semi-structured interviews with 16 medical doctors and nurses from nine Swedish emergency departments. Content analysis was used to describe factors affecting vital sign data quality in emergency care. Study II extracted vital signs from 330 000 emergency department visits to assess the effects of different documentation workflows on data quality. Study III prospectively explored 200 vital sign measurements from 50 emergency care visits to evaluate the impact of manual and automated documentation on vital sign data quality. Study III also used data from an adapted NASA TLX questionnaire to compare the workload of clinical staff (n=70) in manual and automatic documentation. Study IV used semi-structured interviews with 14 emergency medicine physicians from three different sites. Content analysis was used to explore participants’ expectations and concerns regarding clinical decision support systems.
There are three main results and conclusions from the research. Firstly, documentation of vital signs in the emergency department is still surprisingly paper-based, which makes vital sign data unfit for reuse in clinical decision support. Secondly, automation of vital sign documentation is feasible in emergency care and should improve data quality and reduce workload. Thirdly, enthusiasts towards decision support are at risk of disappointment with the level of innovation in the currently available decision support systems, and this may affect the implementation strategy negatively.
List of papers:
I. Skyttberg N, Vicente J, Chen R, Blomqvist H, Koch S. How to improve vital sign data quality for use in clinical decision support systems? A qualitative study in nine Swedish emergency departments. BMC Med Inform Decis Mak. 2016 Jun 4;16:61.
Fulltext (DOI)
Pubmed
View record in Web of Science®
II. Skyttberg N, Chen R, Blomqvist H, Koch S. Exploring Vital Sign Data Quality in Electronic Health Records with Focus on Emergency Care Warning Scores. Appl Clin Inform. 2017 Aug 30;8(3):880-892.
Fulltext (DOI)
Pubmed
View record in Web of Science®
III. Skyttberg N, Chen R, Koch S. Man vs Machine in emergency medicine – a study on effects of manual and automatic vital sign documentation on data quality and perceived workload, using observational paired sample data and questionnaires. BMC Emerg Med. 2018 Dec 13;18(1):54.
Fulltext (DOI)
Pubmed
View record in Web of Science®
IV. Skyttberg N, Kedra M, Chen R, Koch S. Building trust in Clinical Decision Support Systems – A qualitative semi-structured interview study with emergency medicine physicians. [Manuscript]
I. Skyttberg N, Vicente J, Chen R, Blomqvist H, Koch S. How to improve vital sign data quality for use in clinical decision support systems? A qualitative study in nine Swedish emergency departments. BMC Med Inform Decis Mak. 2016 Jun 4;16:61.
Fulltext (DOI)
Pubmed
View record in Web of Science®
II. Skyttberg N, Chen R, Blomqvist H, Koch S. Exploring Vital Sign Data Quality in Electronic Health Records with Focus on Emergency Care Warning Scores. Appl Clin Inform. 2017 Aug 30;8(3):880-892.
Fulltext (DOI)
Pubmed
View record in Web of Science®
III. Skyttberg N, Chen R, Koch S. Man vs Machine in emergency medicine – a study on effects of manual and automatic vital sign documentation on data quality and perceived workload, using observational paired sample data and questionnaires. BMC Emerg Med. 2018 Dec 13;18(1):54.
Fulltext (DOI)
Pubmed
View record in Web of Science®
IV. Skyttberg N, Kedra M, Chen R, Koch S. Building trust in Clinical Decision Support Systems – A qualitative semi-structured interview study with emergency medicine physicians. [Manuscript]
Institution: Karolinska Institutet
Supervisor: Koch, Sabine
Co-supervisor: Chen, Rong; Örnung, Göran; Blomqvist, Hans
Issue date: 2019-10-18
Rights:
Publication year: 2019
ISBN: 978-91-7831-563-5
Statistics
Total Visits
Views | |
---|---|
Clinical ... | 1196 |
Total Visits Per Month
March 2024 | April 2024 | May 2024 | June 2024 | July 2024 | August 2024 | September 2024 | |
---|---|---|---|---|---|---|---|
Clinical ... | 7 | 4 | 5 | 5 | 4 | 5 | 2 |
File Visits
Views | |
---|---|
Thesis_Niclas_Skyttberg.pdf | 1027 |
Top country views
Views | |
---|---|
Sweden | 583 |
United States | 205 |
China | 87 |
Germany | 53 |
Ireland | 28 |
United Kingdom | 27 |
Slovenia | 26 |
Netherlands | 21 |
Russia | 20 |
France | 10 |
Top cities views
Views | |
---|---|
Stockholm | 231 |
Karlstad | 130 |
Ashburn | 90 |
Bro | 34 |
Dublin | 26 |
Hangzhou | 23 |
Boardman | 12 |
Mountain View | 12 |
Shenzhen | 11 |
Beijing | 9 |