Thesis Format
Integrated Article
Degree
Doctor of Philosophy
Program
Nursing
Supervisor
Donelle, Lorie
Abstract
BACKGROUND: In response to a global focus on sustainable healthcare, less is known about the role of palliative care for “persons with ALS” (PALS) across the disease trajectory. With one’s home, the preferred place to manage complex disease conditions and plan end-of-life care by palliative care patients/caregivers, it is important to identify risk factors that may signal early institutionalization. With the proliferation of electronic health records (EHRs), the intensity of care requirements associated with fatal neurological conditions (ALS) is becoming more visible. Data science originating from the fields of computer science, statistics and information science, an emerging field within nursing research is creating new avenues to discover knowledge about patient experiences previously unanswerable using traditional methods. An understanding of home healthcare experiences for PALS will provide opportunities to re-design sustainable, health service delivery programs that align with payor mandates and meet societal expectations.
AIM: In Canada where nursing data science research is less common, this study applied data science methods to homecare EHRs to explore factors as a means of determining PALS’ capacities and preferences to remain cared for at home.
METHOD: A nursing informatics framework (data, information, knowledge, wisdom [DIKW]) integrated with a data science methodology, (knowledge discovery in databases [KDD]) was employed in a retrospective analysis of an ALS homecare EHR dataset. Logistic regression modelling was used to evaluate risks for PALS’ institutionalization.
RESULTS: Five significant logistic regression models were generated. The final model, with six variables, offered the greatest clinical utility. Four risk factors characterized ALS clinical manifestations of disease decline, and two were electronic assessment algorithm-based outcome scales, recognized as powerful predictors of PALS’ institutionalization and informal caregivers’ burden.
CONCLUSION: Study findings highlight the utility of applying big data science to EHR datasets to introduce new knowledge into homecare practice settings. Homecare providers who understand electronic assessment by-products (factors) associated with clinical documentation practices will be well positioned to be proactive and mitigate institutionalization risks of PALS under their care. This study's data science approach fosters a new path in Canadian homecare research to advance health service delivery, education, and policy.
Summary for Lay Audience
Amyotrophic Lateral Sclerosis (ALS), a rare and fatal neurological disease with a limited life expectancy of 2–5 years, is estimated to grow significantly worldwide. In response to the global focus on sustainable healthcare, less is known about the palliative care needs for “persons with ALS” (PALS), the most common cause of neurological death in Canada and a designated priority neurological condition associated with an increased burden of care for patients, their families, and caregivers (CGs). The intensive “caring” supports required by PALS who progressively lose the capacity to care for themselves are also known to cause serious economic and societal consequences, garnering the attention of policy advisors and payors. With “home” remaining the preferred place to manage complex disease conditions and plan end-of-life care by many palliative care patients, such as PALS and CGs, it is important to identify care-related risks of institutionalization.
With the proliferation of electronic health records (EHRs) worldwide, care requirements and other factors (symptom progression) associated with conditions such as ALS are becoming more visible. Data science originating from the fields of computer science, statistics and information science is an emerging field within nursing research. It allows us to ‘interrogate’ healthcare EHR data, uncovering new knowledge about patient experiences that was previously beyond the reach of traditional research methods.
This study, rooted in Canada where nursing data science research is less common, analyzed an electronic homecare dataset (EHR) using data science methods to better understand institutionalization risk factors associated with PALS’s home-based healthcare experiences.
An improved understanding of healthcare experiences for highly complex patients will provide opportunities to re-design sustainable health service delivery programs that align with payor mandates and meet societal expectations. This study is believed to be the first nursing research project in Canada to apply data science methods to home health EHRs to examine care factors as a means of determining PALS’s capacities/preferences to remain home. The findings may inform homecare provider practices to mitigate institutionalization risks of PALS under their care. Finally, the study fosters a new path in Canadian homecare research using data science methods to advance nursing knowledge development.
Recommended Citation
Remus, Sally E., "HARNESSING DATA SCIENCE IN HOMECARE TO ANTICIPATE CARE FOR “PERSONS WITH AMYOTROPHIC LATERAL SCLEROSIS” (PALS)" (2024). Electronic Thesis and Dissertation Repository. 10187.
https://ir.lib.uwo.ca/etd/10187
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 4.0 License.