..Current Projects
DELPHI

Deliver Primary Healthcare Information Project


DELPHI Background:

Funder: Ontario Ministry of Health and Long-Term Care
Years: 5
Lead: Dr. Moira Stewart

The DELPHI project based at the Centre for Studies in Family Medicine at The University of Western Ontario contributed infrastructure facilitating the uptake of electronic medical records (EMRs) in participating Southwestern Ontario family practices to create a researchable database built from pooled EMR data. This database is now known as the DELPHI database.

The image below of SouthWestern Ontario indicates the relative location of our 10 participating primary care practices. Currently there are 23 primary care physicians and 3 nurse practitioners contributing data to the DELPHI database.



Introduction to Database:

This active database contains records beginning from October of 2005 to the present and includes records of approximately 30,000 patients and 285,000 encounters. In addition, DELPHI participants are asked to use the International Classification of Primary Care to code up to five Reasons for Encounter and Diagnoses per visit for a random 30% of their patients. ICPC is an internationally accepted classification system that is specifically suited to the unique practice environment of family physicians. The International Classification of Primary Care is unique to the primary health care setting because:

  1. It is patient oriented – it incorporates the patients’ experience of their illness through reasons for encounter;
  2. It enables a description of signs and symptoms to be made prior to a definitive diagnosis; and
  3. It allows an appreciation of the connectedness of the multiple problems patients bring and an understanding of how unorganized symptoms coalesce into diagnosable and treatable problems.

Thus, the DELPHI database contains an ICPC random subpopulation available for analyses on such occurrences as episodes of care, symptom progression and physician volume or "workload".

^back to top



DELPHI Project Team:

at the Centre for Studies in Family Medicine:

  1. Moira Stewart, Principal Investigator
  2. Amardeep Thind, Investigator
  3. Amanda Terry, Investigator
  4. Judith Belle Brown, Investigator
  5. Stewart Harris, Investigator
  6. Graham J. Reid, Investigator
  7. Sonny Cejic, Investigator
  8. Neil Marshall, Investigator
  9. Joan Mitchell, Investigator
  10. Louisa Bestard Denomme,DELPHI Program Coordinator
  11. Heather Maddocks, Data Analyst
  12. Vijaya Chevendra, Systems Manager

Institute of Clinical Evaluative Sciences Collaboration Team:

  1. Doug Manuel, Investigator
  2. Rick Glazier, Investigator
  3. Karen Tu, Investigator
  4. Liisa Jaakkimainen, Investigator
  5. Astrid Guttman, Investigator

^back to top



Studies and Projects:

 

PROJECT #1:

A PURPOSE-BUILT RESEARCH DATABASE
FOR INTERDISCIPLINARY PRIMARY HEALTH CARE

- Moira Stewart, Lead Investigator -

Canada Foundation for Innovation funds permitted the building of the PHC Researchable Database during 2004 with the co-operation of 25 rural and urban interdisciplinary practices throughout Southwestern Ontario. This proposal was approved in March 2003.

^back to top

PROJECT #2:

DELPHI - DELIVER PRIMARY HEALTHCARE INFORMATION
- Moira Stewart & Doug Manuel, Lead Investigators -

This initiative titled "Interdisciplinary Primary Health Care Quality & Comprehensiveness - Estimation, Validation and Generalization" from March 2004 to March 2006 was funded by the Primary Healthcare Transition Fund of the Ministry of Health and Long-Term Care and led to the creation of the DELPHI database.

Key outcomes and results arising from this project included:

  1. creation of a primary health care researchable database based on EHR data from independent practice sites in Southwestern Ontario all using one software, and
  2. development and validation of EMR measures of primary care through comparison with HA data, in which three objectives were met (description of PHC, test of validation of measures, and test of generalizability of the sampling).

The results of this project described primary health care practice quality, validated the measures and assessed their generalizability.  The project successfully linked two highly experienced research teams:

  1. the Centre for Studies in Family Medicine (CSFM) and its Canadian Foundation for Innovation (CFI)-funded EMR database; and
  2. the Institute for Clinical Evaluative Sciences (ICES) with its health administration (HA) databases. 

^back to top

PROJECT #3:

IMPROVE - IMPROVING PRACTICE OUTCOMES
VIA ELECTRONIC HEALTH RECORDS

- Moira Stewart & Doug Manuel, Lead Investigators -

This initiative built upon previous work and was supported by the funding program Enhancing Quality in Primary Health Care of the Ministry of Health and Long-Term Care (April 2007 to May 2008).  This project had 3 specific objectives:

  1. To assess the validity and generalizability of six original measures over a one year period: 1) interdisciplinary care; 2) diagnosis; 3) volume; 4) referral; 5) quality of diabetes care; and 6) comprehensiveness of care.
  2. To derive two new measures of PHC and test their validity and generalizability over a one year period: childhood immunization; and preventive care for adults.
  3. To create and assess the impact on quality of care, of provider support tools (i.e. feedback
    on practices at the provider level) regarding diabetes care, childhood immunization and preventive care.

The latter objective was achieved by completing a randomized control trial in the summer of 2007.  Using the DELPHI database (pooled anonymized patient records from 25 physicians) a feedback tool to show each primary care physician the level of preventive care they had reached for their own patients for mammograms, pap smears, fecal occult blood tests, and microalbumin creatinine ratio tests was created.  In addition, educational materials and a hands on training program teaching physicians how to use software components to extract lists of eligible patients using specific criteria were completed.

With the IMPROVE educational intervention, physicians successfully learned to perform practice-level analysis using their EMR and extract basic information about their patient population.  The quality of care outcomes are currently being tracked and analyzed.

During this project, the DELPHI team held their first Inaugural conference entitled: Electronic Medical Records (EMR) in Primary Care Research: International Perspectives.  The conference was held at the Toronto Marriott Downtown Eaton Centre on March 25th 2008. Keynote speakers included Dr. Michael Kidd of Australia, Dr. Shmuel Reis of Israel, and Dr. Chris Van Weel of the Netherlands. 

Conference proceedings:
Inaugural Conference on Electronic Medical Records (EMR) in Primary Care Research International Perspectives Proceedings

^back to top

PROJECT #4:

THE 3S PROJECT - SUSTAIN, STRENGTHEN AND SCALE ELECTRONIC
HEALTH RECORD DATA TO EVALUATE AND IMPROVE PREVENTION AND CHRONIC DISEASE MANAGEMENT IN PRIMARY HEALTH CARE

- Moira Stewart & Doug Manuel, Lead Investigators -

Building upon the success of earlier projects, the DELPHI team began this initiative in April 2008.  Funded by the Enhancing Quality in Primary Health Care Round II of the Ministry of Health and Long-Term Care, the DELPHI team set out to complete the following objectives:

  1. To ascertain whether effects of an EMR-based education intervention, for family doctors aimed at improving their preventive care and diabetes care, are sustained after a six month period (sustain).

    A toolkit based on the IMPROVE educational intervention has been created for using EMRs to improve preventive care clinical practice.  It provides step by step instructions for primary care practitioners to generate summaries of Electronic Medical Record data in order to identify eligible patients for five preventive services.

    A User's Guide to Preventative Care Assessment: Using Your EHR to Improve Practice


  2. To strengthen (develop, test and validate) new measures/indicators of quality of PHC in one EMR software, especially for diabetes and congestive heart failure.

    Congestive Heart Failure:
    Resulting from the second objective, the DELPHI team has focused on congestive heart failure and is currently looking at the operationalization and measurement of the Ontario Ministry of Health and Long-term Care (MOHLTC) Heart Failure Patient Care Flow Sheet guidelines for management of CHF patients using EMR data from the DELPHI database. 

  3. To scale up (adapt) the existing and new indicators to a second EMR software.

    This third objective involves an expansion to 10 additional family practices and an additional software.
               
    Also during this project, the DELPHI team held their second conference entitled: Electronic Medical Records (EMR) in Primary Care Research: How to Improve Data Quality.  The conference was held at the Toronto Marriott Downtown Eaton Centre on February 9th, 2009. Keynote speakers included Dr. Simon de Lusignan of the UK and Dr. Karen Tu of Toronto. The conference was well attended and received by its participants. Please see the link below for the conference proceedings.

    Electronic Medical Records (EMR) in Primary Care Research: How to Improve Data Quality Proceedings

^back to top

PROJECT #5:

WAIT TIMES
- Amardeep Thind, Lead Investigator -

Using the DELPHI database, Dr. Thind is examining an aspect of wait times that has not been looked at in the Canadian literature – that of wait time to see a specialist after a family physician referral. He is currently examining specialist wait times and equity in these waits for a sample of 16,115 referrals made from October 2005 to March 2008. Future analyses includes:

  1. ascertaining the patient, physician, and contextual level determinants of these waits, and 
  2. comparing waits to established benchmarks, with the goal of
  3. assisting policymakers in decreasing these waits for all Ontarians. 

^back to top

PROJECT #6:

DATA QUALITY
- Amanda Terry, Lead Investigator -

Currently, a study is underway to examine data quality within the DELPHI database. There are two objectives of this study:

  1. to develop measures of data quality; and,
  2. to test the measures in the DELPHI database.

The domains which will be examined include:

  1. comparability (concordance among patient populations, for example the Canadian census population and the EMR population; and conditions, for example published disease prevalence figures compared with prevalence in the EMR data),
  2. completeness  (calculating sensitivity values for test conditions & assessing recording), and
  3. correctness (calculating positive predictive values for test conditions and unlikely combinations of age and sex-specific conditions).

Future steps include continuing to refine and test the data quality measures.

^back to top

PROJECT #7:

NATIONALLY-FOCUSED NEEDS, GAPS AND OPPORTUNITIES ASSESSMENTS IN HEALTH SCIENCES AND POLICY RELATED TO ELECTRONIC HEALTH RECORDS
- Amanda Terry, Lead Investigator -

This project is assessing the current state and future potential of primary health care electronic health record use for policy-making (including strategic planning and program implementation), clinical care, and research in Canada.  The research team has two main tasks to address the objectives of this project:

  1. conducting an environmental scan, and
  2. implementing a collaborative consultation process to both identify goals for primary health care electronic health record use, and creating strategies to address these goals. 

This project is contributing to the realization of the full potential of primary health care EMR use in Canada by identifying priority areas for improvement and change.

Primary Health Care Electronic Medical Records: Gaps in Knowledge and Research in Canada

 

^back to top

PROJECT #8:

CASE ASCERTAINMENT
- Doug Manuel (Ottawa Health Research Institute), Lead Investigator -

To date, ascertaining disease status (case ascertainment) in populations with electronic data uses the same clinical decision-making approach of diagnosing disease.  Although there are many similarities, electronic data provide unique opportunities not readily available to clinicians.    This project initiates the approach of population-based disease ascertainment that builds on the clinical approach, but takes advantage of existing data that is available for large populations.  In doing so, we address two long-standing two concerns with EMR and HA data, that:

  1. that EMR data for individual patients is sometimes incomplete;
  2. single “tests” to identify people using HA data are typically not very discriminating (i.e. low likelihood ratio), so have a limited ability to definitely identify people with diseases.

^back to top

PROJECT #9:

STUDENT THESES

"Episodes of Care for Back Pain using Electronic Medical Records: What Determines Episode Length and Management of Care in Family Practice?
- Megan McKeown

Megan McKeown, a graduate student in the department of Epidemiology and Biostatistics, is examining episodes of care for back pain. This is a concept of both health care utilization and physician management of a health condition. Using the DELPHI database, Megan’s objectives are two-fold: 1) to characterize an episode of care in family practice in terms of calendar length, number of physician visits, medications prescribed, referrals made, and investigations ordered and 2) to investigate if patient characteristics, including age, sex, history of physician visits for back pain, and presence of chronic, musculoskeletal or psychosocial conditions affect episode of care characteristics.

 

"Intensification in the treatment of type 2 Diabetes in the family practices of South Western Ontario: A longitudinal study." – Arunmozhi Dominic

Clinical inertia is a new dimension in the assessment of quality of care in Clinical practice. Clinical inertia is defined as the lack of treatment intensification in a patient not at evidence-based goals for care. Clinical inertia is a major factor that contributes to inadequate chronic disease care in patients with diabetes mellitus, hypertension, Dyslipidemia, depression, coronary heart disease, and other conditions. Recent work suggests that clinical inertia related to the management of diabetes, hypertension, and lipid disorders may contribute to up to 80 percent of heart attacks and stroke. Although there are many research studies dealing with Clinical inertia in Diabetes practice  based on cross sectional data, there is no research done using longitudinal data till date. This research thesis is a novel venture to define and assess the time to event on intensification of treatment in Diabetes and the various factors which play their role in intensification, in family practices. For this study, the longitudinal data from the DELPHI database is analyzed using advanced statistical methods. The results of this study would be an important evidence for policy makers who are interested in setting clinical practice standards for the management of Diabetes and its complications in primary health care.

^back to top

PROJECT #10:

IMPLEMENTATION - A QUALITATIVE STUDY I & II
- Judith Belle Brown & Amanda Terry, Lead Investigators -

A qualitative study was conducted in 2006 to explore DELPHI project participants’ experiences, ideas, and perspectives regarding the adoption of electronic medical records (EMRs) into their primary health care practices and to examine their perceived barriers and facilitators to the EMR adoption.

This descriptive qualitative study explored the experiences of the 30 participants. Semi-structured interviews were conducted with two key themes emerging:

  1. barriers (i.e. level of computer literacy, training required, and time) and facilitators (i.e. having an “in-house” problem solver and the EMR’s integrated messaging system); and,
  2. a continuum of EMR adoption (i.e. levels of knowledge ranging from novice to advanced, and responses to the EMR that included participants’ reflections on their personal journey across the adoption continuum and that of their practice sites).  See link below for publication.

Currently, a follow-up study is being conducted to again explore project participants' experiences and perspectives two years following EMR implementation. Questions are focused on the feelings and experiences in relation to both the usage and impacts of the EMR within their practice.

^back to top

 



Collaborations:

CPCSSN - The Canadian Primary Care Sentinel Surveillance Network
– is a collaboration of a national primary care sentinel surveillance system using regionally linked practice based / primary care research networks (PBRNs or PCRNs) which use electronic medical records (EMRs). 
This network is designed to provide a robust and accurate data source for chronic disease surveillance over time. The project will coordinate the efforts of all PCRNs in conducting surveillance for the following chronic diseases: hypertension, diabetes, osteoarthritis, COPD, and depression.
The CPCSSN initiative is funded by the Public Health Agency under a contribution agreement with the College of Family Physicians of Canada on behalf of 9 practice based research networks (PBRNs) associated with departments of Family Medicine across Canada.

CIHI- The Canadian Institute for Health Information's Primary
Health Care (PHC) Electronic Medical Record Content Standards Project.

The DELPHI project is contributing to this CIHI initiative to help establish the extent of EMR data representative of 112 data elements relating to 12 primary health care quality of care indicators.

^back to top



Innovation:

The DELPHI database is unique in Ontario and Canada. Three aspects of DELPHI are unique:

  • its connection with a high performance research team;
  • its linkage with Health Administration (HA) data through an active collaboration with the Institute for Clinical Evaluative Sciences; and
  • its coding of multiple symptoms and tentative diagnoses per encounter


More Information:

For more information about the DELPHI Project, please contact:

Louisa Bestard Denomme
Research Program Coordinator
Centre for Studies in Family Medicine
The University of Western Ontario
Toll Free: 1-866-858-9929
London Area: (519) 661-2111 Ext. 22038
lbestar@uwo.ca

^back to top