An Update on the Development and Feasibility Assessment of Canadian Quality Indicators for Atrial Fibrillation and Atrial Flutter.

BACKGROUND: In 2010, the Canadian Cardiovascular Society Atrial Fibrillation/Atrial Flutter (AF/AFL) quality indicator (QI) working group was established to develop QIs and assess feasibility of measurement. After extensive review, 3 priority QIs were selected. However, none were measurable at a national level.
METHODS: The working group reconvened in 2017 to review the relevance of previously proposed QIs, identify opportunities to develop new QIs, and propose an initial strategy for measuring and reporting.
RESULTS: Two additional priority QIs were added to the previous 3: proportion of patients with nonvalvular (NV) AF/AFL sorted by stroke risk stratum and annual rate of hospitalization for a new heart failure diagnosis. An environmental scan was undertaken to determine the potential of existing databases to provide national and provincial estimates. On the basis of validated administrative codes, the Canadian Institute for Health Information discharge abstract database can be used for inpatients. In collaboration with the Canadian Primary Care Sentinel Surveillance Network, 2 of the 5 QIs can be assessed in outpatients (patients with NVAF/AFL sorted by stroke risk stratum and high risk for stroke NVAF/AFL receiving oral anticoagulation). Stroke prevention therapy can be further measured in selected provinces with linked databases including prescriptions.
CONCLUSIONS: This first step could provide a better initial understanding of the quality of AF/AFL care in Canada, but important gaps in the meaningful measurement of QIs remain. The AF/AFL QI working group has limited capacity to make progress without national level leadership and the resources to support data aggregation, data analysis, and pan-Canadian reporting.

Atrial fibrillation (AF), the most common sustained cardiac rhythm disorder, is a critical public health issue. The prevalence of AF is rapidly rising2, 3 and is projected to double over the next 30 years because of an ageing population and increasing adverse lifestyle and cardiovascular risk factors that cause AF. Clinically, the consequences associated with AF are significant with a 5-fold increase risk for ischemic stroke, a 3-fold increase risk in developing heart failure (HF), and a near doubling in mortality.5, 6, 7 One of the most devastating sequelae of AF, stroke, is preventable with use of oral anticoagulation (OAC) therapy that is effective and safe, yet is underused.8, 9 The healthcare costs due to AF are high; an estimated $815 million Canadian dollars occur from hospitalizations alone annually. Thus, the importance of understanding the quality of AF care in Canada is critically important to ensure that all Canadians are receiving optimal care, that costs are contained, and that there is strong evidence to guide how resources are allocated.

To address this need, the Canadian Cardiovascular Society (CCS) established an AF/Atrial Flutter (AFL) quality indicator (QI) working group in 2010, tasked with developing and selecting QIs and performing a feasibility assessment. A detailed review of the process and findings has been published. Briefly, a total of 27 QIs were initially proposed in 3 categorical areas: (1) access, (2) therapy, and (3) outcomes. After internal review, application of multiple rating strategies, and feedback from external experts, stakeholders, and CCS membership, 5 priority QIs were initially selected. The CCS directed a final list of no more than 3 priority QIs per Working Group. After further internal discussion focusing on the most clinically relevant, evidence-based, and generalizable indicators, the following 3 QIs were selected:

Proportion of patients with a diagnosis of NVAF/ AFL at high risk of stroke (age ≥ 75 years or CHADS2 ≥ 2 [Congestive heart failure, Hypertension, Age ≥ 75, Diabetes, prior Stroke or systemic embolism]) receiving an OAC.

Annual rate of stroke in patients with NVAF/AFL.

Annual rate of major hemorrhage in patients with NVAF/AFL.

A feasibility assessment followed in which it was determined that none of the QIs as defined could be adequately measured using data sources available regionally, provincially, or at the national level across inpatient and outpatient healthcare settings.

Since this initial effort, a number of NVAF/AFL guideline documents have been updated,12, 13 and a report from the American College of Cardiology/American Heart Association on clinical performance and quality measures for patients with NVAF/AFL was published. Accordingly, in June 2017, the CCS AF/AFL QI working group reconvened with a strategic plan focused on revisiting the relevance of previously proposed QIs, to identify opportunities for new indicator development and to propose an initial strategy for measuring and reporting on selected priority QIs.

Methods

The renewed AF/AFL QI working group retained its multidisciplinary and pan-Canadian composition and has membership consisting of academic and community clinician content-area experts (primary care, cardiology, emergency medicine), as well as representatives from the Canadian Institute for Health Information (CIHI), CorHealth Ontario, and Institut National d’Excellence en Santeá et en Services Sociaux. Three subgroups of the larger CCS AF/AFL QI working group were formed to address the QI categorical areas (access, therapy, outcomes).

The first task for the AF/AFL QI working group members was to participate in a survey to evaluate the relevance of 27 QIs and determine whether each QI should be kept as a priority indicator, kept as a nonpriority indicator, altered, or removed entirely. The threshold for an indicator to be placed in each of these categories was > 50% of the votes. The results of the survey and any new proposed QIs were discussed among the members of the respective 3 categorical subgroups of the overall working group (access, therapy, outcomes) and then brought forth to the entire working group for review at the 2017 Canadian Cardiovascular Congress meeting. The priority indicators were also rated for relevance by attendees of the CCS AF and HF workshop in 2017. After the meeting, each subgroup was tasked with further narrowing down the list of priority QIs.

Finally, an environmental scan of existing registries, databases, and networks at the regional, provincial, and national levels capturing inpatient and outpatient populations to determine the feasibility of QI measurement was performed.

Results

Selection of QIs

The survey results for the 27 QIs according to the 3 categorical subgroups are shown in Table 1. Following the process outlined earlier, 5 QIs were selected as being most clinically important to determine standard quality of care for the broadest AF/AFL patient population briefly summarized are as follows:

Table 1

Survey results for the QIs in the access, therapies, and outcomes subgroups

	Access (n = 12)	Therapies (n = 8)	Outcomes (n = 7)
QI classification
Keep as priority	Confirmed diagnosis of NVAF/AFL and echocardiographic assessment	Diagnosis of NVAF/AFL and at high risk of stroke receiving an OACRisk stratification of patients with NVAF/AFL for stroke	Rate of stroke in patients with NVAF/AFLRate of major hemorrhage in patients with NVAF/AFL
Keep as nonpriority	Percentage of patients with HF diagnosis with an EKG within 3 mo of diagnosisPercentage of patients with hypertension with an EKG within 3 mo of diagnosisPercentage of patients with valvular heart disease with an EKG within 3 mo of diagnosisPercentage of patients with a stroke with an EKG within 3 mo of diagnosisPercentage of patients with hyperthyroidism with an EKG within 3 mo of diagnosisPercentage of patients with palpitations with an EKG within 3 mo of diagnosisPercentage of patients with new diagnosis of AF with a transthoracic echocardiogram at 3 moPercentage of patients with new diagnosis of AF with TSH at 3 moPercentage of patients with new diagnosis of AF screened for hypertension (documented blood pressure) at 3 mo	Percentage of patients with CHADS2 score of ≥ 2 maintained on OAC post-catheter ablation for NVAF at 1 y post-ablationAnticoagulation for valvular AFPercentage of patients with diagnosis of AF prescribed an antiarrhythmic drug with follow-up EKG within 3 moPercentage of patients with diagnosis of AF prescribed an antiarrhythmic with documentation of left ventricular function	Percentage of patients with major complications of catheter ablation for AF occurring within 30 d post-ablationPatients undergoing repeat catheter ablation(s) for AF within 2 y of the index procedurePopulation rate of diagnosis of AFPopulation rate of diagnosis of AFLPercentage of patients with diagnosis of AF prescribed an antiarrhythmic for > 1 y with an EKG within 1 y demonstrating normal sinus rhythm
Remove	Percentage of patients with new diagnosis of AF with chest x-ray at 3 moPercentage of patients with new diagnosis of AF screened for substance abuse
Alter
No clear majority		Quality of anticoagulation with warfarin in patients with AF/AFLPercentage of patients with new diagnosis of AF with documented resting heart rate < 100 beats/min at 3 mo

AF, atrial fibrillation; CHADS2, Congestive heart failure, Hypertension, Age ≥ 75, Diabetes, prior Stroke or systemic embolism; EKG, electrocardiogram; HF, heart failure; NVAF/AFL, nonvalvular atrial fibrillation/atrial flutter; QI, quality indicator.

Therapies Indicators

Proportion of patients with NVAF/AFL according to stroke risk strata.

Proportion of patients with a diagnosis of NVAF/AFL at high risk of stroke (age ≥ 75 years or CHADS2 ≥ 2) receiving an OAC.

Outcomes Indicators

Annual rate of stroke in NVAF/AFL.

Annual rate of major haemorrhage in NVAF/AFL.

Annual rate of hospitalization for a new diagnosis of HF.

The details (definitions of numerator, denominator, calculation method, and rationale) of each QI are summarized in Table 2.

Table 2

Summary of selected CCS AF/AFL QIs

Therapies indicatorsA. Risk stratification of subjects with NVAF/AFL for strokePercentage of patients with a new diagnosis of NVAF/AFL who have a stroke risk prediction (CHADS2, CHA2DS2VASc, CHADS-65) score documented in their medical record or have the relevant elements of such scores recorded such that they can be readily and automatically calculatedNumerator: All patients with a diagnosis of NVAF/AFL who have a CHADS2 or CHA2DS2VASc score or CHADS-65 or the elements of these scores (stroke/TIA/SE, hypertension, heart failure, age ≥ 75 y, diabetes, atherosclerotic disease, age 65-74 y, female sex) documented in their medical recordDenominator: All patients with a diagnosis of NVAF/AFLPeriod of assessment: AnnuallyRationale: Will measure the proportion of patients with AF/AFL stratified for stroke risk using a recommended objective tool (CHADS2, CHA2DS2VASc, or CHADS-65)B. Diagnosis of NVAF/AFL and at high risk of stroke (age ≥ 75 y or CHADS2 ≥ 2) receiving an OACPercentage of patients with a diagnosis NVAF/AFL ≥ 75 y of age OR < 75 y of age with a CHADS2 score ≥ 2, and without a contraindication for anticoagulation, who are receiving a prescription for an OAC (warfarin [or other VKA], apixaban, dabigatran, rivaroxaban)Numerator: Primary analysis: All patients with NVAF/AFL ≥ 75 y of age OR < 75 y of age and a CHADS2 score ≥ 2, and without a contraindication for OAC, who are receiving a prescription for an OAC (warfarin [or other VKA] apixaban, dabigatran, rivaroxaban)Secondary analysis: Include the possibility of reporting according to CHADS-65 and CHA2DS2VASc ≥ 2 in men and ≥ 3 in womenDenominator: All patients with NVAF/AFL ≥ 75 y of age OR < 75 y of age and a CHADS2 score ≥ 2Period of assessment: AnnuallyRationale: Patients who are at high risk for stroke should be on an OAC for stroke preventionOutcome indicatorsC. Rate of stroke in patients with NVAF/AFLNumerator: Primary analysis: The number of patients with NVAF/AFL who have a stroke (within 1 y)Denominator: The number of patients with NVAF/AFLPeriod of assessment: AnnuallyRationale: Will measure the rate of stroke in patients with AF/AFL (according to risk score and antithrombotic use)D. Rate of major haemorrhage in patients with NVAFAnnual rate of major haemorrhage in patients with diagnosis of NVAF/AFL receiving an OAC (warfarin [or VKA]), dabigatran, rivaroxaban, apixaban)Numerator: Primary analysis: The number of patients with NVAF/AFL who are hospitalized for haemorrhage of any kind (an arbitrary definition of major bleeding) within a calendar year while taking an OACSecondary analysis: Possibility of reporting according to type of OAC (warfarin [or other VKA], apixaban, dabigatran, rivaroxaban)Denominator: All patients with NVAF/AFLPeriod of assessment: AnnuallyRationale: Will measure the proportion of patients with NVAF/AFL who experience complication of anticoagulation (according to type of anticoagulantE. Rate of hospitalization for new HFAnnual rate of new HF in patients with AF/AFLNumerator: Primary analysis: The number of patients with NVAF/AFL who are hospitalized for new HFDenominator: All patients with NVAF/AFLPeriod of assessment: AnnuallyRationale: Will measure the proportion of patients with NVAF/AFL who experience a new HF diagnosis

CCS, Canadian Cardiovascular Society; CHADS2, Congestive heart failure, Hypertension, Age ≥ 75, Diabetes, prior Stroke or systemic embolism; CHA2DS2VASc, Congestive heart failure, Hypertension, Age ≥75, Diabetes, prior Stroke or systemic embolism, Vascular disease, Age 65-74, Sex (female); HF, heart failure; NVAF/AFL, nonvalvular atrial fibrillation/atrial flutter; OAC, oral anticoagulant; QI, quality indicator; SE, systemic embolus; TIA, transient ischemic attack; VKA, vitamin K antagonist.

Strategy for measuring and reporting QIs

Potential data sources

Inpatient setting

To provide national and provincial estimates on the proportion of patients with NVAF/AFL according to stroke risk stratum and determine rates of stroke, hemorrhage, and HF hospitalizations, the CIHI discharge abstract database can be used. The discharge abstract database contains information for all inpatient hospitalizations for all provinces and territories except Quebec. Data from Quebec are submitted to CIHI directly by the ministère de la Santé et des Services sociaux du Québec. Linked databases that include the Pharmacy Information Network for all available individuals in Alberta and British Columbia can provide inpatient estimates regarding the proportion of patients at high risk for stroke receiving OAC therapy.

Emergency department setting

The National Ambulatory Care Reporting System (NACRs) can be used to calculate stroke risk strata for patients with NVAF/AFL seen in the emergency department (ED) in 2 provinces with 100% mandatory reporting (Alberta and Ontario). The proportion of patients at high risk for stroke receiving OAC therapy can be estimated in provinces with linked databases that include prescription data.

Outpatient setting

The Canadian Primary Care Sentinel Surveillance Network (CPCSSN) is the first pan-Canadian primary care electronic medical record (EMR) surveillance system. There are currently 1180 participating primary care practices and 600 primary care practitioners with 1,800,000 patients in 8 provinces and territories. The EMR surveillance system captures comprehensive data regarding health conditions, encounter diagnoses, billing diagnoses, laboratory results, medications, physical signs, medical procedures, and referrals. With the use of the CPCSSN dataset, an EMR case definition for NVAF/AFL can be developed, and then the proportion of patients with NVAF/AFL according to stroke risk strata and the proportion of patients at high risk for stroke receiving OAC therapy can be measured. NACRs can be used to calculate stroke risk strata for patients with NVAF/AFL seen at hospital-based physician offices in Alberta and Ontario, and provinces with linked databases with prescription data can calculate the proportion of patients at high risk for stroke receiving OAC therapy.

Cohort definition and reporting period

Patients ≥ 20 years can be identified using International Classification of Diseases, 9th Revision Clinical Modification (codes 427.3) or International Classification of Diseases, 10th Revision code I48 for AF/AFL from April 2, 2006, 2008, to March 31, 2016 (fiscal years), with follow-up to March 31, 2017. Incident AF/AFL can be defined by at least a 2-year washout period. The inclusion criteria should be (i) ≥ 1 hospital discharge diagnosis in any position in the discharge record, (ii) ≥ 1 ED visit or hospital-based physician office visit in the 6 coding conditions, and (iii) for the analysis on OAC use, patients who survived 90 days after initial diagnosis. The exclusion criteria include (i) valvular AF because these patients may carry a different prognosis, (ii) death during index hospitalization, (iii) incident AF the same day or after diagnosis of ischemic stroke only for the QI measure for incidence stroke rates, and (iv) history of HF only for the QI measure for rates of new HF hospitalization. The validated administrative codes corresponding to cohort, outcomes, bleeding risk, and comorbidity definitions are shown in Table 3.16, 17, 18, 19, 20, 21

Table 3

Administrative codes

	ICD-9-CM	ICD-10-CM
Populations
AF/AFL	427.3427.31, 427.32	I48
Mitral or aortic valve disease	394, 395, 396, 424.0, 424.1	I05, I06, I08.0, I08.1, I085.2, I08.3, I34, I35
Tricuspid or pulmonary valvular disease	397, 424.2, 424.3	I07, I08.1, I08.2, I08.8, I08.9, I36, I37
Valve surgery and procedures	Procedure codes: 35.0, 35.1, 35.2, 35.96, 35.97, 35.99	CCI procedure codes: 1.HS.80,1.HS.90, 1.HT.80, 1.HT.89, 1.HT.90, 1.HU.80, 1.HU.90, 1.HV.80, 1.HV.90
Outcomes
Stroke and embolic events
Ischemic stroke	362.3, 33.x1, 434.x1,	H34.1, I63, I64
Systemic embolism	444	I74
Hemorrhagic stroke	430, 431, 432	I60, I61, I62
Transient ischemic attack	435	G45
Bleeding
Major bleeding (includes intracranial hemorrhage and GI bleed)	362.81, 379.23, 430, 431, 432, 456.0,456.20, 459.0, 530.21, 530.7, 530.82,531.0, 531.2, 531.4, 431.6, 532.0, 532.2, 532.4, 532.6, 533.0, 533.2, 533.4, 533.6, 534.0, 534.2, 534.4, 534.6, 535.x1,569.3, 569.85, 578, 596.7, 599.7, 719.1, 770.3, 784.7, 784.8, 786.3	H35.6, H43.1, I60, I61, I62, I85.x1, K22.11, K22.6, K25.0, K25.2, K25.4, K25.6, K26.0, K26.0,K26.2, K26.4, K26.6, K27.0, , K27.2,K27.4, K27.6, K28.0, K28.2, K28.4,K28.6, K29.x1, K31.80, K55.21, K62.5, K66.1, K92.0, K92.1, K92.2, M25.0,N02, R04, R31, R58I31.2, J94.2
Embolic risk
Heart failure	398.91, 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, 404.13, 404.91, 404.93, 425.4-425.9, 428	I25.5, I42.0, 142.6-I42.9, I43.x, I50.x
Hypertension	401-405	I10-I13, I15
Diabetes mellitus	250	E10-E14
Myocardial infarction	410, 412	I21, I22, I25.2
Coronary revascularization (CABG/PCI)	360, 361	1.IJ.50.ˆˆ, 1.IJ.76.ˆˆ
Chronic coronary artery disease	410, 411, 412, 413, 414, 429.2, V45.81, procedure codes 36.xx	I25.0-I25.2, I25.5, I25.8, I25.9
Peripheral vascular disease (including aortic plaque)	093.0, 437.3, 440.x, 441.x,443.1-443.9, 447.1, 557.1, 557.9V43.4 (procedure)	I70.x, I71.x, I73.1, I73.8, I73.9, I77.1, I79.0, I79.2K55.1, K55.8, K55.9, Z95.8, Z95.9
Bleeding risk
Alcohol misuse	265.2, 291.1-291.3, 291.5-291.9, 303.0, 303.9, 305.0, 357.5, 425.5, 535.3, 571.0-571.3, 980, V11.3	E52, F10, G62.1, I42.6, K29.2, K70.0, K70.3, K70.9, T51, Z50.2, Z71.4, Z72.1
Anemia	280-285	D50-D64
Excessive falls	E880-E886, E888	W00-W19
Hepatic disease	070.22, 070.23, 070.32, 070.33,070.44, 070.54, 070.6, 070.9,456.0-456.2, 570.x, 571.x,572.2-572.8, 573.3, 573.4,573.8, 573.9, V42.7	B18.x, I85.x, I86.4, I98.2, K70.x, K71.1, K71.3-K71.5, K71.7, K72.x-K74.x, K76.0, K76.2-K76.9, Z94.4
Cancer	140.x-172.x, 174.x-195.8, 200.x-208.x, 238.6 (malignancy)196.x-199.x (metastatic solid tumor)	C00.x-C26.x, C30.x-C34.x, C37.x-C41.x, C43.x, C45.x-C58.x, C60.x-C76.x, C81.x-C85.x, C88.x, C90.x-C97.x (malignancy)C77.x-C80.x (metastatic solid tumor)
Coagulation platelet defect	287	D69
Comorbidities
Chronic kidney disease	583, 584, 585, 586, 592, 593.9	N00-N23
Hypothyroidism	243.x, 244.x	E01.x, E02.x, E03.x, E89.0
Hyperthyroidism	242.x	E05.x
Pulmonary embolism		I26
Pneumonia		J09, J10, J11, J12, J13, J14, J15, J16, J17, J18

CABG, coronary artery bypass grafting; GI, gastrointestinal; ICD-9-CM, International Classification of Diseases, 9th Revision Clinical Modification; ICD-10-CM, International Classification of Diseases, 10th Revision Clinical Modification; MI, myocardial infarction; PCI, percutaneous coronary intervention.

Discussion

The development and assessment of AF/AFL QIs are essential for monitoring and benchmarking care in Canada. This initial work was undertaken by the AF/AFL QI working group in 2010 and resulted in the selection of 3 priority QIs; however, none were measurable at a national level. Since that time, the working group reconvened and reviewed QIs in the areas of access, therapy, and outcomes. In addition to the previously selected 3 priority QIs, 2 other indicators were added with respect to therapies and outcomes. An environmental scan was undertaken, and it was determined that the 5 QIs could be measured in a limited capacity using national and selected provincial databases.

Among the 3 identified categorical areas for QI development, there was 1 QI related to “access”—confirmed diagnosis of AF/AFL and echocardiographic assessment. Although the consensus was to maintain this QI as a priority, challenges in measurement remain, particularly given that electrocardiograms and echocardiograms are mainly done in outpatient facilities. In certain provinces where echocardiograms are performed uniformly in hospitals because of strict government control, wait times are approaching several years, whereas in other provinces wait times are measured in days. For this reason, it was considered a lower priority compared with other QIs in the therapy and outcome areas. However, it is important to note that had we been able to measure this QI, it would have provided important information regarding the extent to which variations exist in access times across Canada for such a fundamental procedure as an echocardiogram.

The access subgroup then focused on whether physicians taking care of patients with AF/AFL could access subspecialty care (AF clinic) from the ED or outpatient settings. A prior nationwide report had found 14 AF clinics using various models of care. Since that time, an environmental scan done estimates the presence of 19 AF clinics across Canada. There was no specific registry to access the number and location of AF clinics. The majority of clinics were within academic centers, and no standardized method for collection of data was occurring across the AF clinics. Given these observations, it was thought that a QI could not be developed at this time.

The assessment of stroke risk for patients with AF/AFL has been universally recommended among guidelines12, 13 and included as a performance measure among the various professional societies; however, the specific stroke risk scheme and subsequent definition of eligibility for OAC use vary depending on the country. We maintained both priority quality outcome indicators as first proposed by the AF/AFL working group to measure the rates of stroke and major hemorrhage because of their clinical relevance.

A new outcomes QI was developed to measure rates of hospitalization for new HF among patients with NVAF/AFL. HF and AF often coexist, and each condition worsens the prognosis of the other. After hypertension, HF is now the second most common comorbid condition. Furthermore, HF hospitalization (or rehospitalization) is recognized as a marker of quality of care, and if there is timely access to medical care and appropriate management of rate or rhythm, an HF episode may be circumvented.

Study limitations

Since the last feasibility assessment, the working group was able to determine that selected priority therapy and outcome QIs could be measured with modified definitions using national and select provincial databases. We recognize that there are several limitations and potential challenges to performing these analyses.

First, to develop a nationally representative dataset that allows fair interprovincial comparison, the cohort is mainly restricted to hospitalized patients. Confinement to inpatients may result in higher estimates of annual incidence of target outcomes, because hospitalized patients are older and have more comorbidities. A collaboration with CPCSSN would provide some data on the quality of AF/AFL care from the outpatient setting but will be cross-sectional in nature.

Second, fatal stroke or bleeding events that occur out of hospital would not be counted as outcomes. This would result in a reduced sensitivity and bias toward lower absolute event rates. However, most strokes and major bleeding events do result in hospitalization, and this limitation would not influence trends over time or inter-provincial comparisons. The components of bleeding risk stratification scores contain key components that cannot be calculated using administrative data.

Third, CIHI does not possess data on out-of-hospital mortality. Therefore, we anticipate this will result in analytic issues with censoring patients and potentially lead to an overestimate of the incidence of the outcomes of interest.

Fourth, NACRs is only available with 100% completeness in Alberta and Ontario, and subsequently data regarding stroke risk stratification in the ED and hospital-based physician offices will be limited. Future initiatives to aid other provinces in complete reporting are needed.

Fifth, data on OAC therapy are currently limited to 2 provinces, which would provide a limited sense of geographical differences. The working group would like to work with CIHI to use the National Prescription Drug Utilization Information System database, which contains claims and formulary data for public drug programs from 10 provinces/territories and formulary information from 1 federal drug program.

Sixth, other data sources do exist in the context of clinical trials or province-specific disease or procedure registries. However, the extent to which the clinical trial data in particular are truly reflective of their disease populations, as opposed to being convenience samples, may not always be apparent. Accessing clinical trial or registry data can be a challenge. Data quality, especially in regard to registries, may vary between different registries from different locations, even within a given province. Finally, the ability to link such information sources to other datasets, such as those pertaining to healthcare visits or drug prescriptions, will differ.

Last, administrative data can be subject to misclassification; however, many of the administrative codes have been previously validated.

The Future

The AF/AFL working group’s environmental scan of available data illustrates incremental progress in our ability to assess the quality of AF/AFL care and drive evidence-based improvements. However, the obvious fragments in Canada’s 13 health systems, lack of national leadership, and lack of resources for quality measurement and improvement limit progress.

In Canada, the Canadian Foundation for Healthcare Improvement (CFHI) is 1 of 10 pan-Canadian Health Organizations that operate at arm’s length of the federal government. The CFHI is mandated to respond to disparate health policy issues and, more specifically, to support healthcare organizations to adapt, implement, and measure improvements in patient care, population health, and value for money. The obvious alignment between the efforts of the CCS AF/AFL quality working group and the CFHI’s mandate suggests this fundamental health gap be filled by a national government entity, like the CFHI, or by a newly formed pan-Canadian Health Organization.

Of equal importance to national leadership, resources to support data aggregation, data analysis, and pan-Canadian reporting are urgently needed. Ideally, this fundamental component for ongoing improvement in quality of care should be embedded into the health system. Furthermore, there is an evident need to address the existing gaps cited in tracking QIs to make more meaningful strides in measurement and improvements in quality of care. For these reasons, investment in infrastructure at the provincial and federal level is imperative to facilitate measurement and reporting of QIs within the current health care context and to enable better and more streamlined reporting systems in the future.

Identifying gaps in care is essential for targeting improvement efforts. Arguably, to the extent that much of Canadian health care is publicly funded, the federal and provincial governments have a shared responsibility to provide the necessary resources and infrastructure for data acquisition in an effort to support evidence-based improvement efforts. Just as important, they have an obligation to report it. Canadian taxpayers are required to make significant financial contributions to the health care sector; as such, they deserve to know whether they are getting the access, equity, and quality that they expect in return.