A systems approach for modeling health information complexity
Introduction
Health care processes such as decision making, communication, and collaboration are all dependant on effective dissemination of information. Information complexity, caused by factors such as poor data integration and quality and timely access to information, can create problems that prevent optimal delivery of health care services (Ash, Berg, & Coiera, 2004; Gagnon et al., 2015). Information complexity issues are amplified in delivery models such as chronic care delivery where it is necessary to manage patient care over time and across multiple providers and settings (Sturmberg, O’Halloran, & Martin, 2012). Patient management is often facilitated using Clinical practice guidelines (CPGs), defined as evidence-based recommendations to guide physician decision making related to disease diagnosis, prognosis and treatment (Woolf, Grol, Hutchinson, Eccles, & Grimshaw, 1999). CPGs are a clinical decision support tool that provides a plan for how to manage a patient’s condition such as diabetes or to screen for diseases such as cancer. Canada’s Northwest Territories (NWT) has a higher incidence of colorectal cancer compared to the rest of Canada and patients often present with advanced disease by the time it is detected (Northwest Territories Health and Social Services, 2014; Yellowknife, NWT, & Services, 2015). Colorectal cancer incidence in the NWT is 1.6x higher than the Canadian average in males, and approximately double the rate compared to Canadian females. Only 20% of eligible individuals in the NWT underwent screening in 2011 and 2012, while the reported national average was 55.4% (Singh, Bernstein, Samadder, & Ahmed, 2015; Northwest Territories Health and Social Services, 2014). While CPGs for colorectal cancer screening can significantly improve cancer outcomes in rural and remote settings like the NWT, they remain underutilized, with the NWT having lower screening rates compared to other areas of Canada (Honein-Abouhadar, 2013).
We believe that the overarching cause of poor CPG implementation is information complexity and to address that issue we need modeling approaches to understand information complexity. However, to date there is no approach for identifying and modeling the information complexity factors contributing to poor CPG implementation. This paper fulfills the above need and uses a case study of colorectal cancer screening in remote and rural Northern Canada to develop a systems approach for modeling health information complexity. Our paper is presented in 5 sections. In Section 2 we provide background literature on CPGs, system complexity, and strategies for addressing complexity. In Section 3 we describe our case study, data sources and data analysis approaches. Section 4 is our results consisting of the two parts of our system based modeling approach. Part one is a set of health information system components and interrelationships, and part two is a method for system mapping using the system components and interrelationships. The results also provide examples of exploratory system models from our case study as well as an illustration of how we used the models to characterize and understand health information complexity. Section 5 is a discussion of our results. We conclude with limitations and next steps from our research.
Section snippets
Health systems and complexity
Health systems are complex entities that can be broadly defined as a system of systems that includes information, organization, finance, human resource and policy components. ‘Health systems’ is a broad term that refers to health systems in general. A ‘health system’ is a single representation or specific example of a system (e.g. the Canadian Health Care system). This terminology is consistently used, for example the World Health Organization’s definition of health systems (//www.who.int/topics/health
Case study
We studied colorectal cancer screening in the North West Territories (NWT) of Canada from September-December 2015. The NWT is a rural and remote region of Canada with a population of approximately 45 000 people spread across 33 communities within a geographic area of 1.6 million square kilometers (Yellowknife et al., 2015). NWT has a higher incidence of colorectal cancer and lower screening rates compared to the rest of Canada (Northwest Territories Health and Social Services, 2014; Yellowknife
Results
Our results present our systems approach for modeling health information complexity. First we describe the two parts to our systems modeling approach. Section 4.1 describes the system components and interrelationships that emerged from our data analysis while Section 4.2 describes how we used the system components and interrelationships to develop a system mapping method for health information complexity. In Section 4.3 we then use our mapping method to develop exploratory system models that
Discussion
While public health evidence such as colonoscopy screening clinical practice guidelines (CPGs) are quite explicit on how screening initiation and follow-up needs to occur, health information complexity can impair how information artifacts such as CPGs are implemented in actual health care settings. To manage health information complexity we must understand the relationship between it and other complexity factors such as process complexity. Health information complexity is a by-product of the
Conclusion
Information complexity is caused by factors such as poor data integration, interactions across multiple agents, information quality and timeliness of information access. Information complexity can impair the implementation of information artifacts in healthcare. To address information complexity issues, we first need to map, model and understand how information complexity occurs. This paper used a case study of colorectal cancer screening in remote and rural Northern Canada to develop a systems
Acknowledgments
This work was made possible by financial support and encouragement from the University of Ottawa Division of General Surgery Surgeon Scientist Program and University of Ottawa Clinician Investigator Program. We also acknowledge funding from a CIHR Frederick Banting and Charles Best Canadian Graduate Scholarship and a Student Research Grant from the Telfer School of Management Research Fund. We also acknowledge funding from a Discovery Grant from the Natural Sciences and Engineering Research
References (60)
- et al.
Rural vs urban residence affects risk-appropriate colorectal cancer screening
Clinical Gastroenterology and Hepatology
(2013) - et al.
Some unintended consequences of information technology in health care: The nature of patient care information system-related errors
Journal of the American Medical Informatics Association
(2004) - et al.
A systems perspective on rural and remote colorectal cancer screening access
Journal of Cancer Policy
(2017) - et al.
Dynamic modeling approaches to characterize the functioning of health systems: A systematic review of the literature
Social Science & Medicine
(2017) - et al.
Health system modelling for policy development and evaluation: Using qualitative methods to capture the whole-system perspective
Operations Research for Health Care
(2015) - et al.
Beyond technology: Identifying local government challenges for using digital platforms for citizen engagement
International Journal of Information Management
(2018) - et al.
Evidence-based medicine: Is it a bridge too far?
Health Research Policy and Systems
(2015) - et al.
Opportunistic decision making and complexity in emergency care
Journal of Biomedical Informatics
(2011) - et al.
A learning organization in the service of knowledge management among nurses: A case study
International Journal of Information Management
(2015) - et al.
Current status of screening for colorectal cancer
Annals of Oncology
(2013)