Please Enter a Search Term
 
"Building a Healthy Tomorrow"  TM
AdmissionsCommunicationsDivision/DisciplineEducationFaculty/StaffLeadershipResearchStudents
eHealth Research Unit

Current Projects

Below is a complete list of our current projects.
logo of the eHealth Research Unit (eHRU)

Understanding the impact of the wait time between initial primary care referral of a patient to the time the patient is seen by a specialist, or Wait Time 1

Dr. Gerard Farrell, Dr. Susan MacDonald, Dr. Christopher Kovacs, Dr. Jaqueline Elliott, Dr. Clare Liddy, Dr. Erin Keely, and Ms. Ann Hollett

In NL, we have received funding from the Newfoundland & Labrador Medical Association and the Department of Health and Community Services to advance eConsult services in the province. Similar to the LHIN CEO’s question, we need to understand the needs of NL PCPs, and conducting this audit will both provide the base line data for the eConsult service introduction and contribute to the larger national study underway as health regions and provinces try to understand Wait Time 1 and its implications for healthcare delivery.

The objectives of this study are threefold: (1) to understand and test the feasibility of calculating wait times from the primary care perspective through a chart audit; (2) to calculate wait time 1 for primary care practices in Ontario and Newfoundland; and (3) to develop a chart abstraction manual for future studies examining wait times. The results of this study will be highly relevant for the province, health regions and the participating clinics, to better understand how long patients are waiting for access to specialist care.

Improving Requests to NLPDP by Better Forms Management

Dr. Aaron McKim, Dr. Gerard Farrell, Ms. Ann Hollett

The eHRU has secured $160, 000 in funding from the Newfoundland and Labrador Medical Association’s Clinical Stabilization Fund(August 2016) to explore the idea of forms management as an entity separate from the EMR. An electronic forms management warehouse could act as a central repository for patient care related forms. Forms generated would be sent to the warehouse for conversion to electronic format, storage and availability. Availability at this stage would be via the Web as well as through a tablet based app. Discussions could occur with EMR vendors to make the forms accessible from within EMRs, which would be the ideal solution.

For the purposes of this project, NLPDP (Newfoundland and Labrador Prescription Drug Program) has been chosen as a test case. Prescribers will be able to find the correct NLPDP form for the context of the patient they are dealing with. The forms accessed will be fillable pdf’s of NLPDP forms, completable online. They will then be able to print the completed form and fax it (as they do now; we will also explore allowing prescribers to fax the completed form directly to NLPDP from the form). A secondary piece of work will involve building logic into the interface to build the correct form from the information provided by the prescriber about the patient to ensure that the form is completely and correctly filled by the prescriber prior to submission. If NLPDP are prepared to work with us, we can also investigate having the data submitted securely directly to them electronically, without the need for faxing at all.

It is hoped that this project will begin a discussion over governance of forms. Now, any entity can generate a form and require its completion. Family Doctors have no input as to the propriety of yet another form. Regional Health Authorities each have their own version of some forms with different requirements for the same end. It is hoped that this work will cause the appropriate authorities to discuss what actions require a form, what information is required on the form, and who is best to fill in the form. It is also hoped that the idea of electronically prepared forms, filled in part from electronic records and transmitted digitally as a new way of doing business will be realized as a result of this work.

Reducing wait times for specialist care in NL through eConsultation

Dr. Gerard Farrell, Dr. Susan MacDonald, Dr. Christopher Kovacs, Dr. Jaqueline Elliott, Dr. Clare Liddy, Dr. Erin Keely, and Ms. Ann Hollett

The eHRU has secured $161, 700 in funding from the Newfoundland and Labrador Medical Association’s Clinical Stabilization Fund(February 2016) to pilot replicating the Champlain BASE eConsult service in Newfoundland and Labrador. Our research objectives include, but are not limited to:

1. Reduce specialist wait times in the participating specialties
2. Enhance the knowledge acquisition among primary care providers
3. Lessen the burden on remote and rural patients travelling for non-urgent referrals
4. Increase the satisfaction of both specialists and primary care providers

In Ontario, the Champlain BASE* (Building Access to Specialists through eConsultation) eConsult service has demonstrated excellent potential as a tool to help address the waitlist problem by minimizing unnecessary referrals, and ensuring a more effective referral when it is deemed necessary. This secure web-based service enables the family doctor or nurse practitioner to submit a clinical question to a specialist who replies within seven days (average 2 days). By providing advice directly to the provider, a referral may be avoided or at least enhanced while the patient remains on the specialists’ waitlist. Data from the fully operational eConsult service in Ontario consistently demonstrates that in 40 percent of all cases, a face-to-face referral, which was originally considered, was no longer necessary as a result of the service.

Innovating Online with Stepped Care: A Comprehensive Mental Health Care Model

Dr. Peter Cornish, Dr. Chan, Dr. Rick Audas, Dr. Gerard Farrell

We have developed a unique but untested model of stepped care for secondary mental health care settings. Patient intake is handled through a decentralized case management approach in which all providers assume responsibility for at least one half day of scheduled and walk-in intakes. Each provider is responsible for managing all cases that present during their scheduled intake/walk-in coverage times. Referrals to other providers or trainees are permitted but typically provider availability is scarce. As such, providers are motivated to refer patients to lower steps of care unless patient presentation severity warrants “stepping up.”

Funding application submitted to CIHR Operating Grant: eHealth Innovation Partnership Program, June 2015.

Secure Health Statistical Analysis Methods

Dr. Saeed Samet, Ahoora Sadeghi Boroujerdi, Dr. Shabnam Asghari, Oliver Hurley

Health informatics, using new information technology, provides a high demanding set of data resource and knowledge that is very useful for secondary data users and researchers in various health systems and applications

We propose a set of privacy-preserving methods and techniques for some popular health statistical analysis methods. Using this set of secure protocols health researchers and other data users are able to issue their requests as some queries, and receive only the results of their queries from the data owners, while each data custodian can keep their sensitive data private. Proposed methods have been tested using sample data to illustrate the performance of the results in terms of computational and communication complexities.

HearMe

Javad Rahimipour Anaraki, Ahoora Sadeghi Boroujerdi, Dr. Saeed Samet

HearMe is a web application which is capable of generating transcripts and keywords for YouTube videos in five different languages such as English and Spanish. HearMe is available online at hearme.mybluemix.net.

The main goal of this app is to help people with hearing problems to read what they’re not able to hear, and app developers who need keywords extraction as a pre-processing tool. The only requirement of the app is a link to the desired video on YouTube.

ePortfolio

Dr. Danielle O’Keefe, Dr. Gerard Farrell

The Discipline of Family Medicine has hired a programmer to develop an electronic version of their Portfolio, used in evaluation of Family Medicine Residents. The requirements are complete and we are in the design phase, hoping to roll the completed application out for the incoming residents in July. The resulting application could be modified for use by other residency programmes.

Simulation Storyboard App

Dr. Noel O’Regan, Dr. Gerard Farrell

We are exploring the development of an application that will allow instructors involved in patient simulation scenarios to storyboard the flow of the simulation, including timing and prompts. We are also discussing the development of a tablet app that will take the resulting storyboard and play it back at the time of simulation to help keep the instructor on time and in place with the simulation. DELTS are working on a budget for the work, the grant is being written for submission this spring.

Designing a Physician-Friendly Interface for an Electronic Medical Record System

Donald Craig, PhD and Gerard Farrell, MD
This EMR prototype addresses two long-standing problems among commercial EMRs. First, the prototype features flexible data storage. Such a feature provides a greater degree of freedom in upgrading or, indeed, changing vendors. Second, Dr Craig's prototype has involved practising physicians during its development and therefore it provides a superior user utility and experience. (since 2008) (paper)

COPD and CHF Apps

Eastern Health, Dr. Gerard Farrell

Dr. Farrell is working as a Medical Advisor to Eastern Health on this project. They have money from Infoway to implement apps for remote monitoring of patients with CHF and COPD to tailor intervention to minimize emergency room visits. Recognizing the experience of the eHRU, Eastern Health has requested our involvement once the provincial funding has been secured.

Anesthesia Resident Reflection App

Dr. Michael Bautista, Dr. Gerard Farrell

Develop an app that would encourage anesthesia residents to reflect on their daily experiences in the context of the CanMeds roles using as a guide the Ignation Pedagogical Paradigm. Grant proposal written. The resulting application could be generalized to other resident programmes.

Preparation for Clerkship App

Dr. Jenny Harris, Dr. Gerard Farrell

We are attempting to turn the information contained in the Preparation for Clerkship course into an information resource app. The Preparation for Clerkship course occurs at the end of Phase three in the spring of the year. It is information intense, out of context (it occurs in a classroom and online, not on the floors) and months before the students arrive on the wards. By making it into an app, we are hoping that, even if the students don't retain the knowledge from the course, they will be able to access in context and in time. Work is ongoing.

Medical App Review Website

Ms. Ann Hollett

This initiative was started in 2012. This website has been launched to help medical clerks navigate through the thousands of apps available on the internet.

The purpose of this website is to gather app reviews from medical clerks and post them online to share with their classmates. The site is a resource to learn about the web tools that are available to the medical community and how colleagues rate them and put them to the best use.

Telegerontology: A Novel Approach to Optimize Health and Safety and to “Age in Place” Among People with Dementia in Newfoundland and Labrador.

Dr. Roger Butler, Dr. Michelle Ploughman, Dr. Susan Mercer, Dr. Gerard Farrell, Ms. Ann Hollett, Ms. Elizabeth Wallack.

This project was developed out of a concept for a teledementia project initiated by the late Dr. Maxwell House in 2011. The goal of this project is to test various dementia assessment tools via Skype and using a remote geriatric specialty team provide real time support to caregivers at the point of care in their homes.

Three software apps developed for the iOS platform are being used in the project. These apps will be completed by caregivers and transmitted to a central server. This allows the geriatric specialty team to remotely monitor the progress of these persons. The process begins with a home visit and evaluation of the circumstances of the patient and the caregiver. The caregiver is trained to use the apps. Dr. Butler follows the patient and caregiver using the data generated by the apps and “meets” with them weekly via Skype. If he feels an intervention is necessary, he conveys that to the patient's primary care giver with advice as to the problem and its solution.

In 2012 we applied for and received seed money from the Government of Newfoundland and Labrador in the amount of $20,000. Additional bids for funding in 2013 were also successful from the following funders: NLCAHR ($40,000), the Alzheimer Society of Canada ($26,160), and the NLMA ($10,000).

The team has developed a working partnership with the Alzheimer’s Society of Canada, Newfoundland Chapter in order to access their First Link Program which communicates with Alzheimer’s patients and caregivers via Skype. A working partnership has also been established Dr. Anne Snowdon and her team at the IVEY Business School in London, Ontario to avail of their expertise in economic analysis.

The project team started recruitment of physicians and patients for the first set of test and control groups in late 2013. They conducted the home site visits with these groups in Clarenville and Burin in June 2014. Active recruitment of physicians and patients for the second set of test and control groups is now underway. The home site visits are scheduled for March 2015 in Placentia, Whitbourne, Carbonear, and Harbour Grace.

In 2014 the eHealth Research Unit was also successful in securing additional funding for this project from funders of previous projects housed at the unit. The Healthcare Foundation agreed to re-assign the remaining $4.485 from the Impact of Medical Residents’ Exposure to Electronic Medical Records Project to be used in this project. The Lawson Foundation agreed to allow the unit transfer $7,568 remaining in a research account held by Dr. House to also be used for this project.
This project has been presented at PriFor 2014 and at a Research in Progress session, December 2014.

This project is scheduled to run until June 2017. The next phase will be to apply for further, more extensive funding and to take this program of healthcare delivery across the province.

The Effectiveness of the Edmonton Symptom Assessment System (ESAS) in Monitoring Palliative Care Patients in the Community. Phase 2

Dr. Susan MacDonald, Dr. Gerard Farrell, Ms. Carmel Collins, Ms. Ann Hollett.

In January 2012 preparations began for the second phase of the project which was to develop and test a remote method of monitoring the symptoms of palliative patients to prevent unnecessary hospital and clinic visits and for early identification of developing crises. A software app to be used on the iOS platform was developed by DELTS here at Memorial. This app would permit patients to log on and complete their ESAS regardless of their location. App development was completed in 2013.

In 2012 and 2013 Dr. Susan MacDonald, co-investigator on the project was successful in competing for the Eastern Health Lighthouse Grant for Innovations to help fund the development of the web app and implement the app in a second phase trial. These grants were in the amounts of $7,100 and $10,000 respectively.

We have now started recruiting patients to the study of how this will impact on their care.

Lucas: a machine learning package

Javad Rahimipour Anaraki, Dr. Saeed Samet
Machine Learning (ML) is a set of methods that empowers computer programs to learn by investigating and exploiting relations between data. Then the outcome can be applied to further and unseen data.

At eHRU, we are working on several ML methods to integrate them into a single, easy to use, and concise package called Lucas that is aimed to process health datasets, efficiently and securely.
Our goals are as follows:

1.Preparing an easy to use ML package
2.Secure processing of sensitive data such as health datasets
3.Involving in variety real world problems in Newfoundland and Labrador
4.Proposing new and widely available data processing methods

Lucas is a machine learning tool for performing pre-processing (such as feature selection, feature-sample selection, and sample selection based on fuzzy-rough sets), classification and visualization of datasets. Lucas is cross platform open source software under GNU GPL for Windows, Linux and Mac based systems.

eHRU OSTIS

Dr. Saeed Samet,,Mohammad Kazem Poozesh, Dr. Shabnam Asghari, Navid Shekoufa, Oliver Hurley

In recent years, there has been a growing interest among health scientists, health authorities and researchers on the role of health geography. Health geography is the application of geographical information, perspectives, and methods to the study of health, disease, epidemiology, and healthcare. These systems can summarize a large amount of datasets into visual maps that can be very intuitive for the planning, management, evaluation of public programs, and for engaging the attention of policy makers, authorities and the public in the process.

One of these systems that are being developed is the Online Spatiotemporal Information System (OSTIS) for Chronic Disease in Newfoundland and Labrador. They are currently using datasets from NL component of Canadian Chronic Disease Surveillance System (CCDSS), Newfoundland and Labrador Centre for Health Information (NLCHI), and NLCHI Mortality System. They will integrate processed information from these datasets with the Digital Newfoundland map available through Memorial University’s (MUN’s) Department of Geography and MUN’s Library to show the information visually on Newfoundland map.

At EHRU we are going to develop a mobile-based version for the OSTIS. Our application will connect through a secure channel (HTTPS) to the web servers of the OSTIS then it will receive data in a readable format like XML or JSON. After that it will parse this data based on user requests, extract the corresponding information and shows them on the visual maps and charts with related factors like age, sex, mortality, diseases prevalence.

RDC Ignite R&D Granted Project: A Common Framework of Privacy-Preserving Data Analysis for Health Research. 2013-2015

Dr. Saeed Samet

Since 2000, privacy-preserving techniques for data mining and machine learning methods have been proposed for various applications in two main approaches. In the first approach, Secure Multi-party Computation (SMC) methods using cryptography are utilized, and in the second one Randomization and perturbation are used to protect the underlying data. Some of the existing protocols are client-server, while in the others two or more parties privately own a portion of the whole data and want to jointly and securely reach to aggregate knowledge.

This research proposal will extend those techniques in the direction of statistical analysis methods widely used in health research, such as calculation of confidence intervals, regression analysis, survival and proportional hazard analysis, chi-square test, odds ratio, relative risk, and so on.

Creating a common framework, which can perform secure statistical computations on encrypted data, will definitely play a significant role on data privacy applications. By using this framework, many existing privacy issues, such as original data leakage, direct access to the plain data, risk of data re-identification, as well as accuracy issues such as lower precision of the results introduced by generalization and suppression, and also the necessity of getting patient consent, will no longer exist. Rectifying the above disadvantages will allow us to use the framework for health services research which strongly needs both high accuracy and privacy. Also, working in a common framework will enable the clinical units such as hospitals working together to provide data to a wide range of data users, by having a set of unique methods and standards to apply on their private data. Applications in other disciplines can also utilize the framework in their applications to enforce privacy-preserving techniques when needed.

NSERC Discovery Research Project: Distributed and Scalable Privacy-Preserving Data Mining Techniques for Big Data. 2015-2019

Dr. Saeed Samet

The long-term objective of this research is to develop scalable privacy-preserving methods and protocols for data mining algorithms on Big Data with the long-term vision of integrating the proposed methods and protocols into the existing tools on Big Data, such as Hadoop and NOSQL. The research will focus on practical methods and techniques for privacy-preserving protocols on both simulated and real data (genome-environment-wide associations in type 2 diabetes to uncover gene-environment interactions associated with this highly common disease). Utilizing type 2 diabetes data provides a complex but manageable dataset on which to test our privacy-preserving techniques and will be useful for comparison in similarly complex applications in health, business and government. Type 2 diabetes is reported to affect 2.5 million Canadians, at a cost of over $15B a year [15]. In the US it is estimated that one-third of all children born since the year 2000 will be affected with this disease at some point in their lives reducing their life expectancies [16]. Worldwide, type 2 diabetes accounts for 90 % of cases of diabetes. Understanding the multiple aspects of this complex disease including genetic, environmental and demographic factors is essential [17], and the associations that emerge should be validated across different big datasets while preserving the individual’s privacy, which will limit the access to such data in various forms. In order to utilize these datasets it will be necessary to preserve the individual’s privacy while allowing meaningful data mining and computational operations. The results will extend the set of secure protocols to cover statistical analysis and data mining methods, and the proposed techniques will be applicable in other areas of health, business, and government where Big Data are used. Therefore, the focus of the short-term objectives will be to propose, design and implement efficient privacy-preserving tools, using new and existing privacy-preserving techniques applied to simulated and type 2 diabetes data. The specific short-term objectives of this research are to (1) indicate which steps (from data gathering to dissemination of results) of Big Data require privacy protection and where currently available privacy-preserving techniques are used; (2) identify the statistical and data-mining techniques that are currently used on genetic data; (3) develop privacy- protected data-mining and computational procedures and algorithms for these applications and (4) test these privacy-protected algorithms on simulated Big Data and type 2 diabetes datasets.

Electronic Medical Records

Dr. Donald Craig, Dr. Gerard Farrell
The benefits of Electronic Medical Records (EMRs) are well documented. EMRs have the potential to offer physicians access to patients' aggregate health information in a timely and coherent manner. With a more complete patient picture, EMR's can allow physician's to make a more accurate diagnosis thereby reducing errors due to incomplete information.

However, merely providing physicians with the means to access a patient's health record is insufficient. The interface through which the physician accesses the record must not place unnecessary cognitive load upon the physician during his or her assessment of a patient. Unfortunately, many of the commercially available EMRs have User Interfaces (UI) that were cumbersome to use or difficult to navigate. Such interfaces do not work well with a physicians' traditional non-linear workflow and can result in significant frustration or even increased errors. Ultimately, these barriers have impeded adoption of EMRs in clinics despite the potential benefits.

Project Goals
With this situation in place, the eHRU undertook a research project to design and implement an EMR UI which would be more amenable to a physician's workflow. Such a design would have to offer physicians many of the benefits of a paper chart while at the same time providing them with helpful ways of organizing and viewing a patient's health record that would not be possible with a traditional chart. When designing the interface, the decision was made that the interface should adopt the following broad characteristics:

1. The interface should eschew pop-up windows or dialog boxes. Such windows tend to be viewed as 'interruptions' by user and can consequently disrupt to a physician's workflow. When overused, they contribute to the cognitive load required when navigating an interface: the physician has to open, move and resize numerous windows that litter the desktop when attempting to retrieve a desired portion of a patient's record. The windows may have to be manually closed before moving to another part of the record.

2. The interface should offer the physician a means of quickly 'flipping' through a patient's record. Being able to quickly thumb-through a patient's paper chart and zero-in on relevant portions allows a physician to quickly get an overview of a patient's current condition. This can also be useful when assessing a new patient or when the physician needs to be quickly brought up to speed of a patient's recent history. This feature in an EMR would provide physicians a means of acquiring a helpful 'gestalt' of a patient.

3. When recording information, the interface should give physicians the same flexibility as a blank piece of paper. While rigid forms consisting of checkboxes and selecting items from pull-down lists of limited choices can be unavoidable in some cases and may be desirable in other cases, such form elements should not be the sole mechanism available to a physician when recording patient information. He or she must be able to freely record his or her assessment of a patient using either free-form or semi-structured text. By adopting some modest restrictions and conventions on the way physicians record patient information, the EMR should be able to intelligently extract relevant information such as objective quantifiable data, orders and prescriptions.
 
Copyright © 2017 Memorial University of Newfoundland.   Newfoundland and Labrador, Canada.