In Ontario, administrative health databases are managed and stored by provincial data centers such as ICES, and independent organizations, such as CIHI. These organizations are designated as “prescribed entities” under PHIPA and are therefore authorized to collect individual-level personal health information from custodians without consent (30). Since individuals may access various health services across the healthcare system, their information may be recorded in multiple administrative health databases. Data are then anonymized by assigning facility-unique identifiers, such as those derived from Ontario’s universal health coverage plan (i.e., the Ontario Health Insurance Plan [OHIP]) (31, 32). This anonymization process ensures that individual records can be de-identified and linked across multiple databases, providing a more comprehensive view of an individual’s interactions with the healthcare system (5).

ICES and CIHI oversee numerous national and provincial administrative health databases, though the exact number of databases held at any given time varies as new databases are created and others are updated or consolidated. Together, they manage de-identified and linkable unregulated substance use-related health data (including diagnoses, harms, and treatment) for over 14 million people in Ontario covered under OHIP (24, 31, 32).

Accessing unregulated substance use-specific administrative health data from national and provincial administrative health databases involves navigating complex regulations. According to PHIPA, only prescribed entities like ICES and CIHI have legal authorization to access and disclose personal health information for purposes such as healthcare system planning, evaluation, and monitoring (30).

As custodians of administrative health data, prescribed entities must ensure the privacy and confidentiality of these data. They control access and use through data sharing agreements with the original data providers, and through policies and processes that have been approved by the Information and Privacy Commissioner of Ontario (32). External entities or researchers seeking access to these data for research purposes must submit formal data access requests to ICES or CIHI, providing detailed project plans that include research purposes, intended audiences, and relevant research ethics board approvals (32). These requests undergo rigorous and time-intensive reviews and evaluations based on methodology, feasibility, timeline, and ethical or privacy considerations (6).

Additionally, several online sources regularly access, analyze, and publish unregulated substance use-specific data for Ontario. Examples include ODPRN’s Opioid Indicator Tool, PHO’s Interactive Opioid Tool, the Public Health Agency of Canada’s Opioid-and Stimulant-Related Harms Interactive Tool, and the Canadian Substance Use Costs and Harms (CSUCH) Tool (1, 26, 27, 33). These sources present unregulated substance use-related health data on relevant indicators such as opioid-related ED visits, hospitalizations, OAT prescriptions, harm reduction services, substance use-related injuries, and deaths (1, 26, 27, 33). Moreover, some local and municipal health units also publish dashboards, such as Toronto Public Health’s Toronto Overdose Information System, which provides municipal-level health data on unregulated substance use in the city, including hospital visits and opioid toxicity deaths (34). See Supplementary Table 2 for a non-exhaustive list of select publishers of unregulated substance use-specific health data in Ontario, including a list of relevant data publishers, data sources, indicators, scope, and frequency of data updates.

Non-administrative unregulated substance use health data, typically obtained from surveys, are generally managed and stored by the organization conducting the survey. These entities hold legal authority to store and utilize the data they collect for research purposes, as well as to publicly share their analyses while ensuring compliance with privacy regulations (36). For example, Statistics Canada provides data access to external researchers upon request, often with an associated cost, subject to approval under the Statistics Act, a Canadian federal law that governs the collection, analysis, and dissemination of statistical information by the Government of Canada (37).

Organizations routinely distribute aggregated non-administrative health data on unregulated substance use after completing each data collection cycle. Researchers can then integrate data from multiple sources into their analyses (33).

Independent researchers, academic institutions, and organizations (e.g., harm reduction networks) also contribute to the collection and dissemination of health data on substance use. For example, the Canadian Research Initiative in Substance Matters (CRISM) collects and shares data on unregulated substance use prevalence, treatment and service utilization, access, and harms among people who use drugs (PWUD) both within the province and nationally (38, 39).

The lack of comprehensive data on substance use treatment in Ontario can be largely attributed to the fact that many substance use treatment facilities in the province are privately funded (16, 53, 54). Although some data on the availability and types of substance use treatment services are captured in administrative databases such as the Drug and Alcohol Treatment Information System (DATIS), these data are limited to publicly-funded treatment services and are therefore incomplete (55). Administrative health databases in Ontario do not capture diagnoses and services delivered outside of Ontario’s public healthcare insurance plans, as current provincial data privacy laws do not govern the collection, use, and disclosure of personal information by institutions and health providers outside the public sector (49, 50, 56). This includes any data collected from private health organizations or independent organizations, such as privately or donor-funded recovery treatments and support services, as well as community-level resources such as shelters and harm reduction services, leaving a significant information gap on the uptake of these treatments and services (56). Moreover, this gap is compounded by variability in definitions of treatment and treatment types (e.g., residential treatment, detoxification programs, outpatient programs), differences in surveillance indicators among programs, and inadequate efforts to disseminate these data where available (16, 54). For instance, current healthcare accreditation programs lack harm reduction and treatment performance measures, often limiting organizations’ knowledge of valuable surveillance indicators required to improve outcomes (10). Additionally, national efforts to collect and share publicly funded substance use treatment utilization data in Canada, including DATIS data from Ontario, have stalled. For example, the National Treatment Indicators project, has not been updated since its last published report in 2021, which covered data from 2016 to 2018 (57).

There are also information gaps in unregulated substance use surveillance and harm assessments for specific population subgroups. These include correctional populations, migrants/asylum seekers, people experiencing homelessness, veterans, and other vulnerable or hard-to-reach populations who are often excluded from national cross-sectional surveys due to sampling bias (e.g., telephone or permanent mailing address requirements) (28, 53, 58, 59). Specifically, data on substance use prevalence and estimates of substance use disorders among individuals who are incarcerated or unhoused are sparse, due to their underrepresentation in national surveys and administrative health datasets (49, 53, 59). Other data gaps among specific subpopulations may be related to limited resource and training capacity for the collection, management, and sharing of data among organizations who commonly work with these communities (16, 18, 60). For instance, organizations may lack the ability to collect comprehensive and inclusive data on their clients, resulting in the absence of data on these groups (16, 18, 60, 61).

In addition, addressing the causes and harms of unregulated substance use requires comprehensive information on the physical, mental, and social influences of health, including social determinants of health (2, 3, 62). These data are important for policymakers to gain a holistic understanding of the intersecting factors associated with unregulated substance use, and to effectively address the multifaceted needs of populations and specific subgroups (2, 3, 62). This typically involves collecting information from sectors, ministries, and organizations outside of the health sector, such as children and youth services, community and social services, correctional services, immigration, and housing (63). However, in Ontario, existing data collected across sectors and ministries are often overlooked in larger-scale provincial health data strategies, and are not consistently included in medical records or administrative health data (63). Current cross-sectoral data linkage efforts are typically ad-hoc and program-specific, missing opportunities for sustained integration, particularly with socio-demographic data (3, 50, 51, 53, 62, 63). The lack of an integrated approach to data collection and analysis across the public sector impedes the ability to draw important longitudinal connections and insights between health and social factors that influence substance use (63).

Additional issues that result in data gaps include instances where available data are broadly aggregated (2, 18, 44, 64). Limitations on how data are collected, organized, and presented can impede efforts to identify emerging issues and target interventions effectively and precisely (2, 18, 44, 64). For instance, there is a distinct need for disaggregated data which would allow for analyses that are more sensitive and would fill critical gaps in understanding potential inequities in impacts of unregulated substance use among different subgroups (2, 18, 44, 64). Additionally, the collection of more granular data can help address issues related to absent data in linkages, particularly where records cannot be matched due to insufficient identifiers, thereby reducing bias and improving the accuracy of population estimates (65). Strengthening these data linkages is critical for identifying service gaps, targeting interventions, and ensuring that harm reduction, treatment, and prevention strategies are reaching populations most at risk of overdose or drug-related harms (65).

There is also a notable lack of ecological drug use data in Ontario, including detailed information on the types of drugs in circulation over time and by location, as well as population-level prevalence estimates of people who use unregulated drugs. Reliable data on drug use profiles, such as substance type and source, mode of use (e.g., injection versus inhalation), duration, and frequency of use, remain limited. Currently, these data are primarily derived from population-based surveys, which rely on self-reports and are prone to several biases (66). Moreover, the rapidly evolving unregulated drug market introduces novel substances that are often not systematically captured in surveys (28). When they are captured, substances may be misclassified or classified under broad categories (e.g., fentanyl-related compounds), limiting disaggregation by drug type and mode of use (28). Additionally, inconsistent or outdated terminology (e.g., drug names) across surveys, which may not reflect street names used by PWUD, further compromises data accuracy and completeness (28).

In addition, existing unregulated substance use health data in Ontario often suffer from broad categorization of socio-demographic identifiers, such as racial or ethnic classifications. For instance, data on racial and ethnic origin are often grouped into non-specific categories such as ‘Asian’, ‘Indigenous’, or ‘Visible Minority’ which can obfuscate potential differences, disparities, and inequities within these populations (58, 67). The use of aggregate socio-demographic and race-based data is largely due to the limited collection of these data within the health system more broadly (68). Currently, Ontario mandates the collection of standardized race-based data through the Anti-Racism Act for certain public sector organizations (e.g., child and youth services, and correctional services), but not universally across services within the health sector (67, 69). Moreover, the collection and use of data on historically-racialized groups such as Indigenous or Black communities must occur within the context of the data sovereignty principles established by these communities (e.g., OCAP and EGAP, as described above), and ensure their equitable involvement in any data activities that involve their communities (40–42). As such, the meaningful engagement of these groups, or lack thereof, in the design of data initiatives can lead to varying capacities to collect and use data on these sub-populations (40–42, 61). This can lead to inconsistent identifiers and indicators that may not accurately reflect their realities, effectively undermining the representation of these groups in population-level data (40, 61).

Standardized data collection and reporting measures are essential for generating reliable, accurate, and high-quality information that accurately reflects the unregulated substance use landscape. These measures can enhance the efficiency of data processes and enable comparability across jurisdictions within a national surveillance system. Our paper identified inconsistencies in terminology and metrics used by different organizations and sectors within Ontario, as well as across other provinces, territories, and international contexts (8). To address this issue, a national data dictionary should be developed to define terms, specify categories, standardize variable names and diagnostic codes, and align with industry and international reporting standards. This would improve the accuracy of national-level substance use data and enhance comparability with other countries (2, 16).

This data dictionary can ensure consistency in how key terms and outcomes, such as “treatment,” “adverse event,” “length of stay,” “substance use” (e.g., unregulated drugs vs. prescription medication), or “polysubstance use” (e.g., specifying which types/combinations of substances used), are defined and measured (16, 28). It should harmonize code systems, data element names, and value sets, clarifying which codes and values represent variables in electronic health reporting systems (7, 45). For example, identifying which codes/substances are included under the definition of “substance use,” along with specifying the mode of use (e.g., injection, inhalation, etc.) will promote shared meaning of concepts across systems, and allow for better disaggregation of data. In addition, guidelines for designing surveys that examine unregulated substance use behaviors would help reduce variations in research design, as well as reduce survey-related errors that impact data accuracy (28). Including street-level terminology (e.g., common names and forms of drugs) in these guidelines can improve the consistency and detail of the information recorded (28). Regular reviews and updates of terminology and indicators should be conducted to stay aligned with the evolving overdose epidemic in Canada, including changes in co-occurring opioid and non-opioid substance use over time.

The overdose crisis is multidimensional, involving rapidly changing dynamics across a wide range of interacting individuals, agencies, markets, and other societal entities and stakeholders. As such, to build a more comprehensive system that can effectively monitor and address emerging drug trends and harms, it is essential to broaden the surveillance scope to include the complex social and structural determinants of unregulated substance use. This includes strategies to streamline the integration of data from diverse sources, including across sectors, non-publicly funded organizations, and non-administrative health data sources (e.g., surveys). Leveraging information already routinely collected by these systems would require minimal investments, and would improve the identification of marginalized or hard-to-reach populations who are often socially excluded and underrepresented in administrative health data (e.g., people experiencing incarceration or homelessness, asylum-seekers, or sex workers) (63). For Ontario specifically, modernizing privacy legislation - such as PHIPA for health data and the Freedom of Information and Protection of Privacy Act (FIPPA) for data from other sectors - to include formal policies or data sharing agreements can help enable comprehensive data integration (63). Additionally, customizing data governance and privacy laws to include private health organizations that are not currently regulated under PHIPA would enhance the breadth of data collected (56, 63). Furthermore, provincial policies mandating or incentivizing standardized data practices (e.g., mandatory fields, public reporting requirements) can also accelerate alignment across Ontario’s health and social sectors. These strategies should be applied across other provinces and territories to ensure consistent reporting and harmonization of robust and comparable data for national monitoring and evaluation efforts.

Effective data integration requires interoperable systems, compatible data formats, and standardized privacy protocols to support seamless exchange, integration, and storage across sectors (60, 63). In Ontario, data entities such as ICES have the technical capacity to link diverse datasets across provincial agencies with high linkage rates and extended coverage periods (50, 53, 81, 82). However, ICES’ designation as a prescribed entity under PHIPA restricts its scope to health information and to Ontario, limiting its ability to automatically link with data from other sectors, jurisdictions, and independent organizations (6, 83, 84). This constraint hinders cross-sectoral data integration and delays access to high-quality de-identified data (6). While ICES is expanding its partnerships to improve linkage capacity, policy changes that broaden its mandate to include sectors such as corrections, immigration, education, and children and youth services would enhance integration efforts (83). Enabling broader linkage would allow for more efficient data dissemination and provide policymakers with real-time comprehensive insights to design more responsive interventions, particularly for marginalized groups. For instance, linking opioid prescribing data with overdose and drug seizure records could support earlier detection of unregulated drug market shifts. Similarly, integrating social services or justice system data with treatment data could help identify high-risk groups and improve outreach strategies (22). Investing in real-time data integration, and strengthening surveillance infrastructure and data systems at the provincial level can serve as a scalable model for national infrastructure investments, contributing to a more cohesive pan-Canadian surveillance system.

Investing in local data capacity and workforce development, with a focus on data linkage is also key (60). These processes require skilled personnel with expertise in data technologies, research methods, and privacy legislation. Expanding training, technical support, and resourcing for data professionals is essential for the development of a national surveillance system (6, 60). Ontario’s Digital Health Playbook is one such initiative, guiding Ontario Health Teams in adopting digital health tools and training staff in data entry, reporting, and system navigation (85). Building equitable data capacity across the province will not only improve data quality and foster more timely data collection, but also enhance the granularity and representativeness of data to strengthen national surveillance capacity (85).

Accurate data on unregulated substance use are essential for effective monitoring and surveillance efforts. As such, indicators and data collection methods related to unregulated substance use must be meaningfully-designed with significant input from PWUD communities (61). Participatory collaboration with PWUD in planning and development would help ensure that questions and data elements are conceptualized and interpreted consistently, and that the information captured reflects their lived experiences (28, 61). For instance, to reduce errors in survey design and improve efforts to identify new or emerging substances that may pose a public health threat, PWUD can provide insights on the terminology used in their communities and assist in designing questions that reflect diverse experiences with less commonly known substances. This involvement can substantially improve the accuracy and relevance of ecological and prevalence data used to monitor trends in unregulated substance use.

The adoption of standardized socio-demographic data collection in routine clinical and health data processes - and across organizations and services involved in unregulated substance use care, both in Ontario and nationwide - can improve the ability to capture racial, ethnic, and socio-economic classifications. This, in turn, enhances the disaggregation of health data, enabling better characterization and identification of diverse population groups in both provincial and federal surveillance efforts. Policymakers and stakeholders should collaborate with provincial and national data-collecting organizations, such as Statistics Canada, to incorporate key indicators such as housing status, immigration status, socio-economic and linguistic categories, as well as specific categories representing all race, ethnicity, and sex and gender identities (3, 18, 67). To support this, guidance and training should be provided to care providers to ensure the safe and appropriate collection of race-based and identity data (68, 86).

However, it is crucial that the collection of any data adheres to community data sovereignty and governance processes established under OCAP or EGAP (18). Efforts to meaningfully engage with historically marginalized or racialized groups, such as Indigenous communities, should be prioritized when collecting health data concerning unregulated substance use among these groups, who are often underrepresented in data despite being disproportionately impacted by unregulated substance use (18, 40, 87). Organizations or researchers interested in collecting or analyzing data on these groups should ensure their equitable involvement throughout data collection, analysis, and reporting processes (40, 41, 61, 67, 87). These efforts can accurately capture their realities and social determinants of health, ensuring that surveillance initiatives reflect culturally relevant indicators and approaches, and promote data ownership (40, 41, 61, 67, 87). For example, tailored surveys like the Indigenous Peoples Survey and the First Nations Regional Health Surveys, developed with significant input from Indigenous communities and administered by Indigenous interviewers, capture comprehensive health and substance use data among Indigenous populations both on and off-reserve (40, 43, 88). These initiatives are critical for informing evidence-based policies and targeted interventions to mitigate the impact of unregulated substance use across diverse communities, thereby advancing health equity.

To further address data gaps and minimize the underrepresentation of particular subgroups in data collection, efforts are required to build data collection and management capacity among individuals from these communities and organizations that actively work with these groups (18). Bolstering training on data stewardship, privacy laws, research methods, and data analysis, among these communities would allow individuals to develop health indicators that accurately reflect their cultural backgrounds, lived experiences, and social determinants (40). This approach could foster trust among historically marginalized groups and promote autonomy and oversight in how their data are collected and utilized (40, 61). In addition, enhancing data infrastructure and equipping community organizations, such as harm reduction and treatment services, with basic data collection systems, can support the routine collection of quality data and linkage to other data, such as administrative health data (18, 60, 61). Investments in long-term, multi-disciplinary, collaborative initiatives involving government agencies, research institutes, and local/community-level organizations and researchers can improve infrastructure and provide the skilled personnel required for complex data linkage processes (60, 89). Inclusive and participatory collaboration can enhance data collection among marginalized populations, address crucial information gaps, and ensure that comprehensive and high-quality health data can be routinely collected, managed, and accessed efficiently (40, 41, 60). This can facilitate the usability and accessibility of disaggregated data, enabling a comprehensive understanding of health outcomes and social determinants of health. Together, these partnerships can provide frameworks and best practices that inform national engagement and reconciliation strategies in substance use surveillance efforts.

Addressing the lack of timely and comprehensive information to effectively inform policy and public health decisions requires a coordinated effort to enhance the capacity of stakeholders to collect, access, and report high-quality substance use-related health data. Policymakers should explore legislative or regulatory mechanisms to facilitate quicker and more comprehensive reporting of reliable data. For example, in Arizona, United States, first responders and healthcare facilities are required to report overdose-related information within five days, enabling near real-time updates to their public dashboard (71).

Efforts should also focus on streamlining data access and integration, clarifying governance processes, and ensuring standardized reporting across jurisdictions. For researchers in Ontario, this includes clearer, and more transparent data access procedures within key data-holding organizations such as ICES, CIHI, and Statistics Canada, to reduce approval delays and support more responsive surveillance (6). For regulatory stakeholders, enhanced training in privacy and security protocols could improve the efficiency and consistency of data reviews, easing administrative bottlenecks that impede urgent research and policy analysis. Additionally, role-specific training for frontline healthcare and data management staff on standardized data entry and system protocols is essential to minimizing data entry errors, improving data quality, and enabling faster integration and analysis across platforms (6, 67, 90).

The establishment of a national drug surveillance system, and the ability to generate near real-time insights, fundamentally hinges on sustained collaboration across sectors and jurisdictions, and strong partnerships among federal, provincial, and territorial stakeholders (11). This includes researchers, data custodians, policymakers, frontline service providers, and communities most affected by unregulated drug use. Such a system must be grounded in a realistic understanding of what types of policies or programs require more timely, granular data (60, 63). For instance, evaluating the long-term impact of OAT or wraparound care models necessitates longitudinal data linkage across health, social, and justice sectors. Achieving these goals will require interoperable systems and data-sharing protocols that facilitate seamless integration, while maintaining strong privacy protections, both provincially and nationally. As an example, during the COVID-19 pandemic, the Ontario Health Data Platform enabled rapid data-sharing and collaboration across ministries, researchers, and healthcare institutions. This model demonstrated that collaborative data access to enable timely research and analysis can be achieved at the provincial level without compromising privacy standards. Applying similar principles to the overdose crisis could facilitate timely insights into emerging drug trends and intervention outcomes, while also allowing researchers across regions to replicate and build upon analyses, improving standardization and generalizability across provinces (3).

Investments into innovative data strategies and modern infrastructure are also essential to fill critical gaps in existing surveillance systems. While promising developments, such as wastewater analysis, have expanded the toolkit for monitoring drug use trends, traditional analysis and monitoring methods remain slow, resource-intensive, and insufficient to meet the urgency and complexity of the substance use crisis (71). New strategies are needed to enhance the speed, reach, and depth of data collection and analysis nationally. For example, emerging approaches such as applying machine-learning techniques to analyze large-scale social media data, and deploying internet-based surveys, offer scalable and complementary tools to accelerate data collection and analysis, reach underrepresented populations, and deepen understanding of trends in unregulated drug markets, drug use, and their broader social, emotional, and cultural contexts (66, 91).

To meet the informational needs required for effective public health response, investments must support new infrastructure and approaches that prioritize timely and representative data, improved linkages across systems, and robust privacy protections (71). However, striking the right balance between data availability, efficiency, and confidentiality has long been a barrier to timely and comprehensive surveillance, and will only grow more complex with advancing technologies and increasing volumes of data. As novel opportunities to improve data processes continue to emerge, sustained innovation and cross-sector collaboration among public health and technology experts, data scientists, researchers, and policy makers, will be critical. These partnerships are essential to explore, validate, and responsibly integrate and operationalize these innovations into broader data strategies.

By enabling faster data collection and reporting, strengthening research infrastructure, fostering collaborative efforts, and integrating and scaling innovative approaches into national data surveillance strategies, these efforts can help support more agile, informed, and effective public health interventions at all levels (6).