Creating an 11-year longitudinal substance use harm cohort from linked health and census data to analyze social drivers of health
Bibliographic record
Abstract
Abstract Research on substance use harm in Saskatchewan has faced challenges due to an absence of linked data to analyze and report on the social drivers of substance use harm. This study uses the Canadian Census Health and Environment Cohorts (CanCHECs) 2006 to create, describe, and validate a cohort of Saskatchewan residents focusing on substance use. We achieved validation by comparing our descriptive findings with those from other Canadian studies on substance use. The second objective of this study was to underscore the potential CanCHEC holds in studying substance use, especially by bridging the gap in data concerning the linkage of social determinants of health and administrative health data. Additionally, to facilitate further research using this rich national data source, we share our Stata do-file, providing a detailed walkthrough for creating national or provincial substance use cohorts. About the Research Department The Saskatchewan Health Authority Research Department leads collaborative research to enhance Saskatchewan’s health and healthcare. We provide diverse research services to SHA staff, clinicians, and team members, including surveys, study design, database development, statistical analysis, and assistance with research funding. We also spearhead our own research programs to strengthen research and analytic capability and learning within Saskatchewan’s health system. Disclaimer This working paper is for discussion and comment purposes. It has not been peer-reviewed nor been subject to review by Research Department staff or executives. Any opinions expressed in this paper are those of the author(s) and not those of the Saskatchewan Health Authority. Suggested Citation Marouzi Anousheh, Plante Charles, and Fornssler Barbara. 2024. “Creating an 11-year longitudinal substance use harm cohort from linked health and census data to analyze social drivers of health.” MedRxiv. Extended Abstract Background Research on substance use harm in Saskatchewan has been hampered by an absence of linked data to analyze and report on the social drivers of substance use harm. This study aims to create, describe, and validate a cohort of Saskatchewan residents by linking their sociodemographic data to their health outcomes using line-level data made available by Statistics Canada’s Research Data Centres (RDC) program. Methods We used Canadian Census Health and Environment Cohorts (CanCHECs) 2006 to create a cohort of Saskatchewanians followed from 2006 to 2016. We linked sociodemographic information of the 2006 Census (long-form) respondents to their hospitalization data captured in the Discharge Abstract Database (DAD) (2006 to 2016) and their mortality records in the Canadian Vital Statistics Death Database (CVSD) (2006 to 2016.) We developed an algorithm to identify Saskatchewanians who experienced a substance use harm event. We validated the cohort by comparing our descriptive findings with those from other Canadian studies on substance use. Results We used CanCHEC, a national data resource, whereas most previous studies have used provincial data resources. Despite this difference in constructing the cohorts, our results showed trends consistent with previous studies, including an overrepresentation of individuals with lower socioeconomic status within the PESUH group. Similar to other Canadian studies, our results indicate an increasing rate of substance use harm from 2006 to 2016. To facilitate further research using CanCHEC, we share our Stata do-file, providing a detailed walkthrough so other researchers can create national or provincial substance use cohorts. Conclusion Using CanCHEC to create substance use cohorts will enable health researchers to provide a province-wide, population-level, and longitudinal perspective on substance use harm. This comprehensive view is crucial in effectively contextualizing smaller-scale and local studies, allowing us to disentangle the “fundamental causes” of health within the region. Key Messages CanCHEC provides researchers with an excellent opportunity to measure and examine health inequalities across socioeconomic and ethnocultural dimensions for different periods and locations in Canada. There has been a steady increase in people who experienced substance use harm in Saskatchewan, from 2006 to 2016. People who experienced substance use harm between 2006 and 2016 were overrepresented among individuals with an education level below high school, those in the lowest income quintile, residents of rural areas, and Indigenous population. This study provides a Stata do-file, including a detailed walkthrough for using CanCHEC to create national or provincial substance use cohorts.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.008 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.006 |
| Science and technology studies | 0.004 | 0.001 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.002 | 0.005 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".