Article Text

Download PDFPDF

P195 Creation of an HIV case definition to assess the quality of HIV care in the primary healthcare setting in canada
  1. Shabnam Asghari1,
  2. Sarah Boyd1,
  3. Jillian Blackmore2,
  4. Tao Chen3,
  5. Marissa Becker4,
  6. Claire Kendall5,
  7. Alexander Singer6
  1. 1Memorial University Centre for Rural Health Studies, St. John’s, Canada
  2. 2Former Research Assistant at Memorial University Centre for Rural Health Studies, St. John’s, Canada
  3. 3Canadian Primary Care Sentinel Surveillance Network, Kingston, Canada
  4. 4University of Manitoba, Centre for Global Public Health, Department Of Community Health Sciences, Winnipeg, Canada
  5. 5University of Ottawa, Ottawa, Canada
  6. 6University of Manitoba, Winnipeg, Canada


Background People living with the human immunodeficiency virus (PLHIV) have ongoing healthcare needs as HIV has become a chronic condition for those in treatment. With the success of antiretroviral (ARV) medications, AIDS-related illnesses are no longer the biggest threat to PLHIV, rather, emerging complications and ARV toxicities are of concern. For this reason, HIV care is now transitioning to primary care. To be able to assess the quality of HIV care in these settings a valid case definition is required. Our objective was to develop and validate a case definition for HIV applicable to the Canadian Primary Care Sentinel Surveillance Network (CPCSSN) database.

Methods Electronic Medical Record data from CPCSSN (exclusively primary care data) was used to develop a retrospective cohort between 2009 and 2016. We identified all possible records of HIV in the dataset based on the presence of HIV codes, keywords and ARVs. Every combination of codes, keywords and ARVs were analyzed to see which resulted in the most accurate definition of PLHIV. To assess the validity, we linked the data to a LHIV cohort (external reference standard) in Newfoundland and Labrador; and, a random sample of the CPCSSN database which was reviewed by two experts to confirm HIV status (internal reference standard). Sensitivity, specificity and predictive values were measured.

Results It was determined that the presence of an HIV keyword in the EMR along with either an ICD code or taking 3 or more ARV drugs was the most accurate algorithm for predicting PLHIV. Compared to internal and external references, the algorithm showed (97.1% and 95.0% sensitivity, 100% and 80% specificity), respectively.

Conclusion This is the first Canada-wide study investigating the utilization of primary healthcare by PLHIV. This case definition will contribute to future research and improvements in providing care to PLHIV in a primary care setting.

Disclosure No significant relationships.

  • HIV
  • quality of care

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.