Building a world-class health data system: A recommendation for the federal government

Introduction

The federal government recently announced an extra $196 billion in healthcare funding to provinces over the next 10 years. Part of this will be provided via bilateral agreements, conditional on provinces achieving four ‘shared health priorities,’ including an expanding the sharing and use of key health indicators and building a world-class health data system (as stated by the Minister for Health). Bilateral agreements have been agreed ‘in principle’ with nine provinces, although the details still require discussion, including any associated targets or conditions. This article proposes an approach for the federal government as it undertakes discussions with provinces.

Context

Whilst some comparable data exists across provinces, the consensus is that significant deficiencies remain, hindering patient care, planning by provinces and accountability to the public. Key issues are: (1) Fragmentation; (2) Insufficient comprehensiveness; (4) Slow aggregation and reluctance to data-sharing; (5) Lack of data access for patients and staff; (6) Future-proofing systems to expand virtual care. 

In light of the deficiencies above, the federal government announced four federal objectives for better data: (1) Give patients access to their own electronic health information (to enable informed decisions and ‘self-care’); (2) Provide timely data for healthcare staff (to support clinical decision-making); (3) To enable more comprehensive planning (e.g. to manage national emergencies); and (4) Promote greater transparency on results (i.e. performance accountability).

Key considerations for the federal government when seeking to address their established objectives

Federal-provincial dynamics are important. Provinces are primarily responsible for healthcare provision and their own data systems, but federal spending power generates some influence, most explicitly through bilateral funding agreements. As such, the federal government cannot simply set out data requirements, but rather it needs to discuss and negotiate with the provinces.

There is precedent for the federal government seeking accountability whilst respecting jurisdiction. It would not be unreasonable for the federal government to request healthcare metrics from provinces, in order to create accountability around performance. For example, in existing bilateral health agreements, the Canadian Institute for Health Information (CIHI) publishes a set of pan-Canadian indicators. Outside the health sector, National Housing Strategy bilateral agreements specify outcomes and accountability mechanisms. 

Data has real impacts. Independent experts found that stronger health data in Canada would have saved more lives during COVID-19. This is because it took time to link different datasets and therefore it took longer to identify local infection hotspots. 

Improving healthcare data is not cheap, nor straightforward. Electronic health record (EHR) systems are costly: Although Canadian expenditure on EHR to date is difficult to calculate, one helpful point of comparison is that both the UK and Australia have spent billions. EHR is also technical and complex, as have been identified from previous Government reviews. On the upside, some provinces have worked hard to develop their EHR systems, including additional investment. In addition,there may be efficiency savings from digitisation because it enables quicker access to data. 

Overall, the federal government would be advised to have modest short-run expectations for ‘better data’: Improving data systems are complex and time-intensive and cannot be rushed.

A proposed approach for the federal government

The four federal objectives (above) largely determine what better data involves in practice. Data for patients/staff (1 and 2) effectively requires patient-level EHR. For greater transparency (4), the federal government has already prescribed a set of metrics where data is required. Better planning (3) has the most options, ranging from a quick exercise to minimize major sources of interprovincial variation to a more detailed project to align the details of different provinces’ data templates. 

Therefore, the main question is ‘how’ the federal government should proceed. Below, a proposed approach is set out for each objective, prefaced by some cross-cutting themes. 

Overall approach. A highly ‘authoritarian’ or ‘prescriptive’ federal approach is not recommended. Provincial jurisdiction means that the federal government has limited ability to dictate terms. Instead, giving provinces space to lead, but offering federal support (e.g. via legislation if needed) could lead to a more balanced approach.

Objective 1. Staff access. History shows IT projects can often be high risk (e.g. UK, Canada), so flexibility is helpful to mitigate financial overruns and/or failure. It also reduces federal reputational risk if the federal government drives the initiative and the intervention is not fully successful. EHR is a long-term project and provinces are at different stages, so priorities are (i) a roadmap/project plan with trajectories; (ii) a guarantee that systems will be interoperable between providers and across provinces.

Objective 2. Patient access. Similar to (1) above, the roadmap should also include planning for a user interface and enhanced protection for data security. Patient access is less critical for the quality of healthcare services, but is necessary for health promotion (e.g. self-care), which will become increasingly important with population ageing, as this will put pressures on healthcare and long-term care services. 

Objective 3. Planning. A flexible approach can be offered as: (a) One of the benefits of national planning is to improve a nation’s resilience against pandemics, but in practice pandemics are rare; (b) Workforce planning can often be done by provinces. Improvements in inter-provincial comparability can be made in the short term by focusing on major differences in the ways that provinces collect and present data, with further standardization over time. 

Objective 4. Transparency. Accountability is important for future performance gains, and some quick progress is needed politically because the federal government has made accountability such an important issue in relation to additional healthcare funding. This should be feasible as many provinces already publish metrics, although provinces may be wary of reputational risk from benchmarking. 

In summary. the proposed course of action for the federal government is: (1) For each objective, to offer considerable provincial flexibility, but asserting (or strongly requesting) requirements in a few priority areas; (2) To seek short-term gains in transparency and planning via ‘quick wins’; (3) While simultaneously targeting data infrastructure improvements over the medium-to-long term, via by ensuring that provinces develop a robust roadmap and implementation plan for improving data.