Population Data BC: Supporting population data science in British Columbia

Background

Population Data BC (PopData) was established as a multi-university data and education resource to support training and education, data linkage, and access to individual level, de-identified data for research in a wide variety of areas including human and community development and well-being.

Approach

A combination of deterministic and probabilistic linkage is conducted based on the quality and availability of identifiers for data linkage. PopData utilizes a harmonized data request and approval process for data stewards and researchers to increase efficiency and ease of access to linked data. Researchers access linked data through a secure research environment (SRE) that is equipped with a wide variety of tools for analysis. The SRE also allows for ongoing management and control of data. PopData continues to expand its data holdings and to evolve its services as well as governance and data access process.

Discussion

PopData has provided efficient and cost-effective access to linked data sets for research. After two decades of learning, future planned developments for the organization include, but are not limited to, policies to facilitate programs of research, access to reusable datasets, evaluation and use of new data linkage techniques such as privacy preserving record linkage (PPRL).

Conclusion

PopData continues to maintain and grow the number and type of data holdings available for research. Its existing models support a number of large-scale research projects and demonstrate the benefits of having a third-party data linkage and provisioning center for research purposes. Building further connections with existing data holders and governing bodies will be important to ensure ongoing access to data and changes in policy exist to facilitate access for researchers

Quality of initial HIV care in Canada: extension of a composite programmatic assessment tool for HIV therapy

OBJECTIVES

To document the quality of initial HIV care in Canada using the Programmatic Compliance Score (PCS), to explore the association of the PCS with mortality, and to identify factors associated with higher quality of care.

METHODS

We analysed data from the Canadian Observational Cohort Collaboration (CANOC), a multisite Canadian cohort of HIV-positive adults initiating combination antiretroviral therapy (ART) from 2000 to 2011. PCS indicators of noncompliance with HIV treatment guidelines include: fewer than three CD4 count tests in the first year of ART; fewer than three viral load tests in the first year of ART; no drug resistance testing before initiation; baseline CD4 count < 200 cells/mm³ ; starting a nonrecommended ART regimen; and not achieving viral suppression within 6 months of initiation. Indicators are summed for a score from 0 to 6; higher scores indicate poorer care. Cox regression was used to assess the association between PCS and mortality and ordinal logistic regression was used to explore factors associated with higher quality of care.

RESULTS

Of the 7460 participants (18% female), the median score was 1.0 (Q1-Q3 1.0-2.0); 21% scored 0 and 8% scored ≥ 4. In multivariable analysis, compared with a score of 0, poorer PCS was associated with mortality for scores > 1 [score = 2: adjusted hazard ratio (AHR) 1.64; 95% confidence interval (CI) 1.13-2.36; score = 3: AHR 2.02; 95% CI 1.38-2.97; score ≥ 4: AHR 2.14; 95% CI 1.43-3.21], after adjustments for age, sex, province, ART start year, hepatitis C virus (HCV) coinfection, and baseline viral load. Women, individuals with HCV coinfection, younger people, and individuals starting ART earlier (2000-2003) had poorer scores.

CONCLUSIONS

Our findings further validate the PCS as a predictor of all-cause mortality. Disparities identified suggest that further efforts are needed to ensure that care is equitably accessible.

Declines in highly active antiretroviral therapy initiation at CD4 cell counts ≤ 200 cells/μL and the contribution of diagnosis of HIV at CD4 cell counts ≤ 200 cells/μL in British Columbia, Canada

OBJECTIVES

The aim of the study was to examine trends in initiating highly active antiretroviral therapy (HAART) with a CD4 count ≤ 200 cells/μL and the contribution of having a CD4 count ≤ 200 cells/μL at the time of diagnosis to these trends, in British Columbia (BC), Canada.

METHODS

We included in the analysis treatment-naïve BC residents aged ≥ 19 years who initiated HAART from 2003 to 2012. Participants were classified as follows: Group 1: diagnosed and initiated HAART with a CD4 count > 200 cells/μL; Group 2: diagnosed with a CD4 count > 200 cells/μL and initiated HAART with a CD4 count ≤ 200 cells/μL; and Group 3: diagnosed and initiated HAART with a CD4 count ≤ 200 cells/μL. We measured trends in initiating HAART with a CD4 count ≤ 200 cells/μL and used logistic regression models to measure factors associated with initiating HAART with a CD4 count ≤ 200 cells/μL, stratified by having a CD4 count ≤ 200 cells/μL or > 200 cells/μL at the time of diagnosis.

RESULTS

Between 2003 and 2012, 3506 BC residents initiated HAART. Of these, 44% (1558 of 3506) initiated HAART with a CD4 count ≤ 200 cells/μL. This proportion declined from 69% (198 of 287) in 2003 to 21% (81 of 330) in 2012 (P < 0.001). The proportion of those in Group 3 increased from 49% (97 of 198) in 2003 to 69% (56 of 81) in 2012 (P < 0.001). Overall, 56% (1948), 22% (776) and 22% (782) made up Groups 1, 2, and 3, respectively. In adjusted analyses, seeing a specialist was significantly associated with being in Group 3. Using injection drugs and seeing a specialist were associated with being in Group 2.

CONCLUSIONS

In recent years, among individuals who ever initiated HAART in BC, being diagnosed with low CD4 cell counts has become a greater contributor to initiating HAART with low CD4 cell counts.