Systematic continuous learning from data is vital to optimizing Canadian healthcare with improved patient outcomes and reduced costs. To improve real-time, patient and health system, data-informed healthcare quality improvement requires 1) improved capacity to dynamically add nationally collected hospital discharge data fields; 2) the creation of nationally coordinated strategic patient-level data collection and research for clinical problems and conditions of urgent health importance; and 3) the initiation and empowerment of perpetual national quality and research collectives, enabling platforms of standardized data collection, quality improvement and research.
A learning healthcare system that couples evidence generation to evidence application improves outcomes and reduces costs.73 Within the Canadian health care there exists a lack of nationally coordinated data collection. In fact, the lack of pre-existing, resource-appropriate standardized data collection for public health emergencies such as COVID-19 has produced marked regional and temporal variation in diagnostic and treatment protocols, the use of personal protective equipment, and the slower integration of data-informed protocols from hospital-to-hospital.74,75
The challenge to implementing data-driven quality improvement is the need for practice-specific and dynamically alterable data collection. First, we must improve the capacity to dynamically add nationally collected hospital discharge data fields. In this case the established Canadian Institutes of Health Information (CIHI) discharge database could institute a mechanism to add additional fields. Second, we must prioritize and enable nationally coordinated data collection and research when required. For example, the Short Period INcidence STudy for Severe Acute Respiratory Infections (SPRINT SARI) organized and supported by the Canadian Critical Care Trials Group (CCCTG), initiated to provide a mechanism to collect data on newly emerging infections, using an annual short data collection in hospitals to help sites be ‘research ready’. In March 2020, the CCCTG began to track all COVID-19 patients in participating critical care units. SPRINT SARI has been taken up around the world, with support from the WHO and with data curated at Oxford University, leading to the inclusion of over 118,000 individuals from 648 sites across 52 countries (figures growing daily).76 Canada is a founding member of this initiative that has been picked up around the world and one that offers a model of a more rapid, nimble learning health system in Canada.
Third, we suggest strategic support of perpetual national quality-focussed research networks. An example of a national quality network initiative comes from Canadian Association of Thoracic Surgeons (CATS). CATS has evolved a national clinical “learning network” that has led research and education of Canadian thoracic surgeons, recently establishing a data-driven approach, that uses harmonized data definition and collection practices, combined with data-informed local and national quality interventions to reduce post-surgical adverse events and practice variation through the generation of consensus recommendations.67,77,78 The combination of self-assessment, best evidence review and best practice review leading to consensus recommendations has already resulted in a reduction in major post-surgical adverse events at one hospital11, and findings of favourable impressions of the team-building, positivist and patient-centered aspects of this process10.
Implementation of such programs can all contribute to a rapidly learning health care system response, to help improve care in general as well as respond to threats such as COVID-19. All three programs are immediately implementable, capable of catalyzing Canada’s drive for a learning health care system. Now is the time for nationally coordinated data collection and quality-focussed learning networks.
Relevant papers:
- Ivanovic J, Al-Hussaini A, Al-Shehab D, Threader J, Villeneuve PJ, Ramsay T, Maziak DE, Gilbert S, Shamji FM, Sundaresan RS, Seely AJE. Evaluating the reliability and reproducibility of the Ottawa Thoracic Morbidity and Mortality classification system. Ann Thorac Surg 2013, 91(2): 387-393.
- Ivanovic J, Anstee C, Ramsay T, Gilbert S, Maziak DE, Shamji FM, Sundaresan RS, Villeneuve PJ, Seely AJ: Using Surgeon-Specific Outcome Reports and Positive Deviance for Continuous Quality Improvement. The Annals of thoracic surgery 2015, 100(4):1188-1194; discussion 1194-1185.
- Ivanovic J, Mostofian F, Anstee C, Gilbert S, Maziak DE, Shamji FM, Sundaresan RS, Villeneuve PJ, Seely AJE: Impact of Surgeon Self-evaluation and Positive Deviance on Postoperative Adverse Events After Non-cardiac Thoracic Surgery. J Healthc Qual 2018, 40(4):e62-e70.
