NWT RESEARCH DATABASE

Regions: North Slave Region

Tags: chronic disease, electronic health record

Objective(s): To create a longitudinal data repository for primary care chronic disease that will be a resource for surveillance and research in primary care to answer questions about the treatment and management of chronic diseases.

Project Description: The aim is to create a longitudinal data repository for primary care chronic disease that will form the basis of surveillance reports, research projects and feedback to participating practices. A regional and a central data repository will be a resource for surveillance and research in primary care to answer questions about the treatment and management of chronic diseases. This data will be used to explore the prevalence and incidence as well as risk factors related to chronic diseases such as; prevalence of co-morbidity and the effect of co-morbidity on management and disease control; drug utilization in chronic disease; and comparison of regions of Canada in relation to the above. At a local level, the regional data may also be used to provide clinicians with confidential individual feedback on their own practices. The project is separated into two phases. Phase 1 will last one year after which - subject to evaluation - a decision whether or not to expand will be made. Expansion of the research project beyond the data of residents of Yellowknife will depend upon two factors: •A positive evaluation of phase 1 •The expansion of Electronic Medical Record (EMR) functionality to communities outside of Yellowknife The territorial EMR project is projecting the expansion of digital charting to Hay River and Fort Smith, with further expansion projected the following year. DATA COLLECTION De-identified health data and information such as demographic information, conditions, medications, encounters, and other data relevant to chronic and acute conditions (such as laboratory data, physical exam data, referrals data) will be extracted from patients’ medical records and entered into the CPCSSN database. This data will only be identifiable by a unique CPCSSN case number. The practice level data will be de-identified and data from each local participating practice will be combined to create a local CPCSSN database that will be housed on the Department of Family Medicine’s regional server. The regional server is housed at the High Performance Computing Virtual Laboratory at Queens University, Kingston, Ontario. This data will then be merged with regional data collected from the other networks into a central database, the Central Repository. The Central Repository is a secure server located at the High Performance Computing Virtual Laboratory (HPCVL) at Queen’s University in Kingston, Ontario. INFORMING PATIENTS: CPCSSN study information is advertised to patients through posters and handouts and patients are able to refuse to participate. DATA EXTRACTED: The CPCSSN case verification number - will consist of two variables. The first variable is the CPCSSN Patient ID. The CPCSSN Patient ID is a number used to identify each unique patient in the CPCSSN dataset. The CPCSSN Patient ID is generated at the time of extraction, however the complexity of the CPCSSN Patient ID is not in its generation, but in ensuring that this assigned ID stays consistent across each extraction cycle. In order for CPCSSN to perform proper surveillance using the patient data, this CPCSSN Patient ID must represent the exact same patient in each data pull (every 3 months). In order to do this, a second variable, the CPCSSN Case Verification Number, is required. A secure, unique value to use for this second variable is the EMR Patient ID assigned to the patient. This ID is invisible to anyone who does not have direct access to the EMR database, and provides a unique identifier for each patient in the EMR. With each data extraction, the local data manager would be able to use the EMR Patient ID as a mapping mechanism during the de-identification and ID assignment processes to ensure that the same patient receives the same CPCSSN Patient ID each and every time data is extracted. The Case Verification Number (CVN) maintains the privacy of each patient in a number of ways. Alone, the CVN provides no usable information in patient identification. The EMR Patient ID can only be used to identify a patient if a person also has direct access to the EMR database. The CPCSSN Patient ID also contains no patient information and is used to uniquely identify each patient in the CPCSSN database. To further increase security, the CVN will be stored and encrypted in a different location that the actual CPCSSN database. Only those with specified access rights (Network Director, Data Manager) will have the passwords and knowledge as to where the CVN is stored. The CVN is stored between cycles in this location and only accessed when data needs to be extracted. The extracted health data is provided with a unique CPCSSN identifier, reviewed to ensure all identifiers are removed, and entered to the regional CPCSSN server. This data is then further processed through software to reduce the risk of re-identification prior to sending the data to the Central Repository to be merged with regional data. The Central Repository is a secure server located at the High Performance Computing Virtual Laboratory at Queen’s University in Kingston, Ontario. PROCESSES TO PROTECT PRIVACY: The following processes are in place to protect privacy: - research agreements/data sharing agreements that are obtained between custodians of the data and the researchers; -a unique EMR sign-on is obtained for the data manager to access EMR data; -EMR data is de-identified data and then processed through software to minimize the risk of re-identification; -confidentiality agreements are signed by everyone involved in the project; and -standard operating procedures and a privacy policy have been developed. 1) Primary care providers using electronic medical records will receive feedback on their practices regarding chronic disease management and prevention that can help improve care provided to patients. This will consist of quarterly feedback reports on the quality of care with comparisons to local peers, territorial, provincial and national Sentinels. Other information may also be provided which will help to improve the data in the electronic medical records. For example, the information can help primary care providers identify patients in the practice with certain chronic conditions to develop registries aimed to provide better care to their patients with those conditions. 2) Anonymized aggregate date will help inform the regional policy and decision makers about patterns of chronic diseases and the incidence and prevalence of chronic diseases. This information can help inform policy decisions aimed to improve the health care of the people in the Northwest Territories. 3) The anonymized regional and national data will be a resource to clinicians, policy makers and researchers who seek understandings and answers to other questions regarding chronic conditions. Information is needed about factors that impact chronic disease prevention and management since over 30% of Canadians suffer from one or more of the eight conditions that are under surveillance within the CPCSSN project. 4) The lessons learned in achieving a regional and national searchable health data repositories for longitudinal health care surveillance and research, will contribute important knowledge to the evolving field of health data informatics, including transformation, coding, automated de-identification, de-identification of free text, harmonizing data from different EMR systems – all of which assists further research using EMR systems while safeguarding the privacy rights of research subjects. The researcher agrees to provide the custodian a quarterly report (PDF and Excel format) comparing their practice results and the results for all regional and national practices combined as per the sample provided entitled CPCSSN Sentinel Feedback Report 2013-Q1, CSPC (Kingston). Until the Data Presentation Tool becomes functional for custodian use, the researcher and custodian agree to negotiate in good faith the feasibility of adding additional data elements to quarterly reports. The quarterly reports will be distributed by the Yellowknife Health and Social Service Authority senior management to participating management and health providers in order to: -establish a quality assurance accountability system; -promote quality of care through enhanced chronic disease registry capability; -improve data quality in the EMR by highlighting areas of deficit. Further the reports will be shared with the Department of Health and Social Services in order to inform the regional policy and decision makers about patterns of chronic diseases and the incidence and prevalence of chronic diseases in order to inform policy decisions aimed to improve the health care of the people in the Northwest Territories. The fieldwork for this study will be conducted from July 17, 2014 to December 31, 2014.