A Decade Lost: Primary Healthcare Performance Reporting across Canada under the Action Plan for Health System Renewal.

In 2004, Canada's First Ministers committed to reforms that would shape the future of the Canadian healthcare landscape. These agreements included commitments to improved performance reporting within the primary healthcare system. The aim of this paper was to review the state of primary healthcare performance reporting after the public reporting mandate agreed to a decade ago in the Action Plan for Health System Renewal of 2003 expired. A grey literature search was performed to identify reports released by the governmental and independent reporting bodies across Canada. No province, or the federal government, met their performance reporting obligations from the 2004 accords. Although the indicators required to report on in the 2004 Accord no longer reflect the priorities of patients, policy makers and physicians, provinces are also failing to report on these priorities. Canada needs better primary healthcare performance reporting to enable accountability and improvement within and across provinces. Despite the national mandate to improve public health system reporting, an opportunity to learn from the diverse primary healthcare reforms, underway across Canada for the past decade, has already been lost.

Introduction

More than a decade ago, Canada's provincial and territorial health ministers and the federal government produced the Action Plan for Health System Renewal (Canadian First Ministers 2003) and the First Ministers' Accords of 2004, which promised annual and comprehensive public reporting to Canadians using agreed-upon indicators of health status, outcomes and service quality. The goal of this agreement was to shape the future of the public health system with governments, providers and citizens working together towards reform. This mandate ended in 2014, and the Health Council of Canada, tasked with monitoring the implementation of the Accord, including annual public reporting, was disbanded in March 2014.

The rationale and impetus for health system performance reporting have not diminished since the First Ministers' Accord. Public performance reporting may increase accountability, enable public participation in healthcare (Ellins and McIver 2009; Powell et al. 2003), impact societal and professional values surrounding our healthcare decision-making, direct attention to issues not currently on the policy agenda (Oxman et al. 2009a, 2009b) and improve performance (Hibbard et al. 2012; Smith et al. 2012; The Commonwealth Fund 2011; Watson 2009).

While some sectors, such as hospitals, witnessed growing initiatives for public reporting over the past decade (Canadian Institute for Health Information [CIHI] 2014b), the primary healthcare (PHC) sector performance reporting continued to lag behind other health system sectors despite the significant reforms and investments during that time (CIHI 2009; Health Council of Canada 2012; Hutchison 2013). The First Minister's Communiqués and Accords from 2000, 2003 and 2004 required each province and the federal government to report on many elements of the health system but mandated only a few specific to the PHC system: access to care, the composition of the groups providing care, patient satisfaction with care and the degree to which technology is being incorporated into the primary care system (Table 1).

Table 1.

Reporting requirements mandated in the First Minister's Communiqués and Accords from 2000, 2003 and 2004, as well as whether Canada and its provinces met them (shaded)

Limited though the mandate for PHC reporting was, the national interest in ensuring high quality PHC across the country should be strong. Countries with a high-functioning PHC system have healthier populations, a more equitable distribution of health and lower healthcare expenditures (Starfield et al. 2005). Canada, however, does not have a single PHC system but rather 13 distinct provincial and territorial health systems linked by a set of guiding principles enshrined in the Canada Health Act (Government of Canada 1985). While this Act calls for universal public insurance of physician and hospital services, a great deal of variation exists across these systems. The last decade has seen a range of primary care reforms across the country in such areas as physician remuneration, team-based care and regional governance (Hutchison et al. 2011). In fact, as the end of the First Ministers' agreement on public reporting was drawing near, there were growing calls for public reporting on primary care performance to support quality improvement and accountability as a minimum requirement for continuous progress in achieving our goals for the PHC system (Aggarwal and Hutchison 2012; Health Quality Ontario 2014a).

We conducted a review of PHC reporting in Canada to identify the impact and legacy of the health accords. The aims of the work are to determine what performance attributes are being reported on, by whom and how, and what attributes of PHC are most important for reporting. The overall goal of the project was to ascertain what we could learn about the public reporting on the PHC system, which occurred while there was a national mandate for common reporting, to identify opportunities to improve performance reporting across the PHC systems serving Canadians and meet the growing demands for better evidence and information.

Methods

We performed a scoping review of PHC performance reporting in Canada, accessing provincial health ministry reports and websites, as well as provincial quality councils, federal health organizations and national professional bodies. The search was performed between September 20 and November 16, 2012. See (available at: Appendix 1) for the complete list of sources reviewed. Each ministry's or organization's website was scanned using the site's embedded search function when applicable, incorporating terms such as “primary care” or “primary health care” along with “performance,” “measurement” or “indicators”. The websites were also scanned by parsing through sitemaps, uncovering sub-pages such as “reports,” “publications,” “resources” or the likes thereof. Publications intended to report on the performance or status of the healthcare system and which presented quantitative data on PHC-specific indicators were selected for further analysis. Documents without PHC-specific data were excluded.

We included reports dating back to 2004 when the provinces agreed to public reporting on the health system. However, we restricted our in-depth analysis to reporting between 2009 and 2012 to reflect the best available reporting, as performance measures and data collection systems have continued to improve since 2004, and provinces had access to CIHI's Pan-Canadian Primary Care Indicators since its 2006 release (CIHI 2006). We limited our focus in public reporting to PHC, adopting Starfield's definition of PHC as the “products or services designed to address acute and episodic health conditions and to manage chronic health conditions. [PHC] is also where health promotion and education efforts are undertaken, patients receive first care and where those in need of more specialized services are connected with other parts of the system” (Starfield 1998). This definition is broad enough to capture the diverse and emerging models of PHC delivery across Canada, though it is purposefully narrower than the WHO definition to narrow the scope of our search (WHO 1978). We focused on public PHC performance reporting at the provincial level, as this is what all provinces had agreed to.

We were guided by the Hogg et al. conceptual framework for the systematic evaluation of PHC performance (Hogg et al. 2008). This broad framework has been used for several Pan-Canadian research studies aiming to measure comprehensive PHC performance. It enabled consideration of the range of activities occurring within PHC beyond only those elements included in the First Ministers' agreements (Dahrouge et al. 2009). The framework integrates the health system and community context with the practice and recognizes that the organization of a practice also influences performance. Finally, this framework considers quality of care at the individual patient level, allowing an exploration of attributes of care of potentially greater interest to patients, a key stakeholder group for public reporting. Each PHC-specific indicator from the reports selected for further analysis was extracted and matched to an attribute from a modified Hogg framework to facilitate comparison of performance reporting where variations of a similar indicator might be used.

In order to identify PHC attributes important to the patient population for public reporting, we reviewed two Canadian studies which directly surveyed Canadians on their perceptions of the most important aspects of their PHC (Berta et al. 2008; Wong et al. 2008). The common PHC system attributes important to the participants in both studies included accessibility, responsiveness, interpersonal communication, technical effectiveness and whole-person care.

Provider and policy maker priorities were identified from a recent report by the CIHI, in which the members of the two stakeholder groups were recruited to participate in the focus groups and rank the importance of CIHI's previously published 105 Pan-Canadian Primary Health Care Indicators (CIHI 2012).

These stakeholder priorities were compiled for comparative purposes and are displayed in Figure 1. The two studies and the national consultation report by CIHI did not present participants with identical sets of attributes of care and offered varying levels of specificity at the indicator level. In order to compare, identify common interests and link these to reported information, we labelled the stakeholder priority features of PHC, linking each one to the relevant performance attribute in the Hogg framework. We then looked for alignment between the PHC performance information reported and the priority PHC features for different stakeholders. We adopted the attribute of the PHC from the Hogg framework such as access or care of chronic conditions as our unit of comparison because many different indicators, often varying slightly yet reflecting the same attribute of PHC, were reported.

Figure 1.

PHC reporting priorities for patients, policy makers and providers

Results

Who has reported on PHC performance over the last decade?

There is a great deal of variation in the method, quantity and quality of PHC reporting across the provinces. All provinces have a governmental ministry responsible for legislating and enacting health-related policy, and many have an arm's-length body, such as a provincial Quality Council, for overseeing and reporting on the quality of care and/or patient safety.

Since the final instalment of the First Minister's Health Accords was convened in 2004, no province has met the requirement to report annually on the performance of their PHC system. Based on the study search strategy, which sought to identify and retrieve publicly available reports containing PHC-related performance indicator data, seven of the 10 provinces were identified as having reported at least once on some element of the performance of the PHC system. Manitoba and British Columbia's Health Ministry annual reports have included one (total number of general practitioners [GPs]) and two (percentage of physicians implementing electronic medical record [EMR] systems, and percentage of GPs providing chronic disease management) PHC performance indicators, respectively, until 2012 when Manitoba incorporated an additional six indicators. Newfoundland and Labrador, Nova Scotia, Prince Edward Island and Saskatchewan did not report on the performance of their primary healthcare systems between 2008 and 2012.

In addition to the provinces, national reports on PHC performance have been released by the Federal Department of Health (Health Canada), by the federally funded independent Health Council of Canada; the federally funded, independent Canadian Foundation for Healthcare Improvement; the College of Family Physicians of Canada; and as joint ventures between the federally funded independent CIHI and Statistics Canada.

What is being reported?

The indicators contained in each report released from 2009 to 2012 were matched to PHC performance attributes and are presented in Table 2. While there were a few common PHC attributes frequently reported on, often many different indicators were reported for a single attribute. Access to care was the most commonly reported attribute; however, over 20 different indicators (some examples include: percentage of Albertans enrolled in a primary care network, patient-reported perception of access to healthcare, and wait times at community health centres) were reported for this attribute. The majority of provinces releasing performance reports in the last four years also reported on elements of the technical quality of PHC, such as immunization rates, as well as the outcomes of PHC, relating to patient safety and satisfaction.

Table 2.

Performance indicator contents of PHC reports released from 2009 to 2012, as matched to the Hogg framework*

	National					AB			BC	MB	NB	NL	NS	ON	PE	QC	SK
	Health Canada	Health Council of Canada	CHSRF	CIHI	CCFP	Alberta Health and Wellness	Alberta Health Services	Health Quality Council of Alberta	BC Ministry of Health	Manitoba Health	NB Health Council	NL Centre for Health Information		Health Quality Ontario		CSBE
Healthcare system	0	0	0	0	1	0	0	0	0	0	0	0		0		6
Clinical accountability	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Clinical quality improvement process	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Resources and technical provisions	0	0	0	0	1	0	0	0	0	0	0	0		0		6
Provider remuneration	0	0	0	0	1	0	0	0	0	0	0	0		0		6
Funding	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Regional context	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Surrounding medical and social services	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Population and community characteristics	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Community integration	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Organization of the practice	0	2	5	0	14	11	6	2	8	4	9	0		10		59
Health human resources	0	0	0	0	0	3	4	1	0	4	5	0		0		4
Group composition	0	0	0	0	0	3	4	1	0	4	5	0		0		4
Training	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Office infrastructure	0	1	4	0	6	7	2	0	4	0	2	0		7		23
Information technology	0	1	4	0	6	7	2	0	4	0	2	0		7		23
Medical technology	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Office space design	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Organizational structure and dynamics	0	1	1	0	8	1	0	1	4	0	2	0		3		31
Job descriptions and team functioning	0	0	0	0	1	1	0	1	0	0	2	0		0		9
Management and practice governance	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Clinical information management	0	1	1	0	5	0	0	0	4	0	0	0		1		18
Organizational adaptiveness	0	0	0	0	2	0	0	0	0	0	0	0		2		4
Organizational culture	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Practice integration	0	0	0	0	0	0	0	0	0	0	0	0		0		1
Healthcare service delivery	5	48	19	12	11	2	11	26	0	3	40	1		43		86
Access	5	20	11	8	4	2	11	11	0	2	32	1		27		42
First-contact accessibility	2	9	4	5	0	0	10	6	0	1	20	0		17		15
Availability	2	2	3	2	4	2	1	4	0	1	7	1		8		9
Accommodation	0	2	0	1	0	0	0	1	0	0	3	0		1		7
Economic accessibility	1	7	4	0	0	0	0	0	0	0	2	0		1		11
Patient–provider relationship	0	7	6	3	1	0	0	10	0	0	4	0		3		7
Interpersonal communication	0	6	4	2	1	0	0	6	0	0	2	0		2		5
Respectfulness	0	0	0	0	0	0	0	3	0	0	0	0		0		1
Trust	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Whole-person care	0	0	1	1	0	0	0	1	0	0	2	0		0		1
Cultural sensitivity	0	0	0	0	0	0	0	0	0	0	0	0		1		0
Family-centred care	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Advocacy	0	1	1	0	0	0	0	0	0	0	0	0		0		0
Continuity	0	9	1	1	0	0	0	2	0	1	2	0		9		12
Continuity-relational	0	1	0	1	0	0	0	1	0	1	2	0		0		3
Continuity-information	0	8	1	0	0	0	0	1	0	0	0	0		9		9
Cooperative care	0	12	1	0	0	0	0	3	0	0	1	0		4		24
Coordination	0	12	1	0	0	0	0	3	0	0	0	0		4		24
Collaboration	0	0	0	0	0	0	0	0	0	0	1	0		0		0
Service delivery	0	0	0	0	1	0	0	0	0	0	1	0		0		0
Services offered	0	0	0	0	1	0	0	0	0	0	1	0		0		0
Services provided	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Population orientation	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Provider satisfaction	0	0	0	0	5	0	0	0	0	0	0	0		0		1
Technical quality of care	1	0	13	4	2	6	19	4	4	4	19	10		27		34
Health promotion and primary prevention	1	0	5	1	0	4	13	4	0	1	5	1		6		4
Secondary prevention	0	0	3	0	0	2	6	0	0	2	6	7		15		8
Care of chronic conditions	0	0	5	3	2	0	0	0	4	1	8	2		6		22
Care of acute conditions	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Immediate and intermediate outcomes of care	2	8	10	3	1	6	3	25	0	0	7	5		11		22
Adverse events/patient safety	0	0	2	0	0	3	0	2	0	0	1	0		0		9
Patient self-efficacy or activation	0	1	3	0	1	0	0	1	0	0	0	1		1		6
Acceptability of health services	0	0	0	0	0	0	0	0	0	0	0	0		0		0
Confidence in the health system	1	1	3	0	0	0	0	1	0	0	0	0		0		4
Patient-reported outcomes	1	2	1	3	0	3	3	20	0	0	6	4		5		3
Unmet needs for care	0	4	1	0	0	0	0	1	0	0	0	0		5		0

Numbers in group headings (bold) and subgroup headings (italics) represent the sum of the categories held within those groups and subgroups.

Are provinces reporting on indicators important to stakeholders?

The priority PHC features for patients, policy makers and physicians included 29 PHC features of priority to physicians, 26 of priority to policy makers and 13 of priority to patients. The three groups only had 16 priority features common between at least two of them (Figure 1). Of the 52 PHC features prioritized by the three groups, only two, access to a regular PHC provider and wait time for immediate care for a minor health problem, were shared by all three. There were 11 features that overlapped between policy makers and providers. Comparing the reporting obligations mandated by the First Minister's Accord (Table 1) with the more recent stakeholder priorities (Figure 1), only five overlap. These include access to a regular family doctor, the number of multidisciplinary PHC organizations or teams, access to after-hours care, patient satisfaction with care and the progress on information systems. Seven of the 12 indicators mandated for annual reporting in the First Ministers' agreements did not match any of the remaining 47 stakeholder priority features of the PHC system.

To determine if recent PHC reporting efforts matched current stakeholder priority features of PHC, we analyzed the degree to which the features listed in Figure 1 were reported on in federal and provincial reports from 2009 to 2012 (Table 3). As Alberta, New Brunswick, Ontario and Quebec performed the most consistent and thorough PHC reporting between 2009 and 2012, we focused solely on their performance for additional analysis. The actual reporting of the priority indicators for the four provinces examined decreased with the strength of the indicator priority. For the two indicators, which were priorities to all three stakeholder groups, the four provinces had 100% coverage. For the 13 indicators that were shared priorities for two of the three stakeholder groups, the four provinces reported on an average of 69% of them. In the set of 38 indicators uniquely prioritized by a single stakeholder group, the four provinces reported on an average of 22% of them.

Table 3.

Degree of reporting on performance indicators prioritized by patients, policy makers and providers by AB, NB, ON and QC from 2009 to 2012

Priority indicators	AB	NB	ON	QC	Average	Percentage coverage
Shared by all three stakeholder groups	2	2	2	2	2	100%
Shared by two of three stakeholder groups	6	9	11	10	9	69%
Unique to a single stakeholder group	8	7	8	11	8.5	22%

What are provinces reporting on?

The number of indicators reported on from 2009 to 2012 by the four provinces leading the field in quantity and quality of reporting were classified by the PHC attribute they measure. Only eight performance domains were represented: access, services provided, continuity or care coordination, patient satisfaction, information technology and information management, group composition or roles, physician services provided and medical errors. All of these provinces reported on access to care and elements of the technical services provided, such as cancer-screening rates or immunizations and care of chronic conditions. Beyond those two attributes, different provinces had very different reporting patterns. Alberta focused on patient satisfaction in their reports, while Quebec presented a large number of indicators on continuity/coordination of care. New Brunswick had almost none of these.

Where is the information coming from?

Many different strategies were used by the provinces to obtain and present data on a given attribute of PHC performance. Among the two attributes most frequently reported by the leading provinces, access to care and technical services provided, Alberta obtained its information mostly from a combination of provincial administrative databases and provincial surveys, New Brunswick obtained its data from provincial surveys and national/international surveys, Ontario relied on provincial administrative databases and national/international surveys and Quebec reported entirely using data from an international survey – specifically the Commonwealth Fund International Survey of PHC Providers. The Canadian Community Health Survey (CCHS) was the source of data for only some provinces on indicators of the technical quality of care such as mammography rates, and New Brunswick was the only province that used the CCHS data to report on access to PHC. The provinces which developed their own surveys used different indicators, suggesting they were not standardized or shared instruments. Finally, public reporting based on data extracted from EMRs, as is done in the Quality and Outcomes Framework in the UK, was notably absent (Prescribing and Primary Care Services 2014).

Discussion

As the demands on and investments in the health system increase, the need for accountability and good-quality data to track progress and guide investment only grows. Most provinces have an arm's-length organisation tasked with public reporting on healthcare quality, such as a quality council, and all provinces and the federal government have ministries or departments capable of reporting on the health system. Nonetheless, at the end of the decade-long mandate of the Action Plan for Health System Renewal, there is limited PHC system performance reporting and no systematic comparative capacity across the country. Further public reporting priorities need to be updated to ensure they match information needs for policy makers, patients and the public and providers.

This retrospective search for public PHC performance reporting identified reporting efforts by most provinces; however, a number of public reports from the past decade are no longer publicly accessible. The major overviews of health system performance reporting covering our search period, issued by the Health Council of Canada (2012) and the Conference Board of Canada (The Conference Board of Canada 2013), did not reference any public reporting, which we had not retrieved, suggesting that the major publicly accessible information available to decision-makers and the general public was captured in our search.

Current reporting by the four leading provinces matched poorly against mandated reporting. While some priorities for stakeholders will shift over time, provincial reporting also only partially covered more recent priority features for PHC identified by patients, physicians and policy makers. Despite the call for comprehensive performance reporting to drive progress in the PHC system reform (Lester and Roland 2009), provinces continue to use different data sources and non-standardized indicators, resulting in a lack of comparable indicators on most features of interest in PHC performance.

The current diversity in reporting across the provinces may reflect different provincial priorities in the decentralized governance model of Canadian healthcare, with differing short- and medium-term goals between provinces. While the provinces are each unique, they face many similar challenges and must work within broadly similar resources and resource constraints. The inability to systematically compare PHC performance over time across the provinces through the past decade, due to a dearth of common publicly reported indicators, is a vital missed opportunity for the Canadian health system. Further, regular reporting and benchmarking within each province is still important for effective evaluation of reform efforts and accountability of policy makers and the healthcare system (Aggarwal and Hutchison 2012; Best et al. 2012; Smith 2009) and citizen engagement (Ellins and McIver 2009; Powell et al. 2003). Comparing the effect of different strategies within and also across the provinces is likely to yield more relevant solutions and lessons for all Canadians than most other sources of guidance. There are only so many solutions which make sense for similar problems (Blumenthal and Osborn 2013).

Improving PHC performance reporting to fuel a smarter PHC system with greater capacity to learn and improve requires regular and comparable sources of data, which can be relied on by stakeholders to enable timely assessments of performance. As several provinces are currently building their PHC performance reporting strategies, such as Ontario's Health Quality Council's PHC performance framework and measurement and reporting strategy (Health Quality Ontario 2014b), a shared focus on developing high-quality and cost-effective data collection that enables provinces to further analyze and report on their priorities is needed. In some instances, existing national and international sources of data can play that role. The CCHS, administered and analyzed by Statistics Canada, is deployed annually, yet had limited and variable uptake by provinces over the past decade. This is despite the fact that it offers comparable provincial and health region information such as access to care provided by a regular family doctor or PHC provider. Regularly deployed international surveys can also offer valuable provincial-level reporting. In January 2016, CIHI publicly reported the results of The Commonwealth Fund 2015 International Health Policy Survey of Primary Care Physicians, presenting comparative data for each province on a range of measures from timely access to primary care appointments to proportion of primary care doctors who wait more than 15 days to receive a report from the hospital after a patient has been discharged (CIHI 2016). These data were possible because CIHI, the Canadian Institutes for Health Research and Canada Health Infoway, as well as the Provinces of Ontario and Quebec, agreed to pay extra for larger samples of providers in each province, enabling not only Pan-Canadian comparisons but also comparisons with nine other countries.

In 2013, CIHI launched a national toolkit of PHC performance measurement surveys, including a patient survey, provider survey and PHC organization survey (CIHI 2013a, 2013b, 2013c). These tools could also represent an efficient cost saver for provinces, eliminating the cost of developing new surveys in each province, while promoting standardized data collection.

EMRs are another source of patient and organizational data. The CIHI explored the potential for data extracted from EMRs to contribute to health system performance reporting as it currently does in the UK's Quality and Outcomes Framework program. However, that project concluded that EMR penetrance still does not cover most of the population, and usage patterns vary widely across healthcare providers, making attempts to extract standardized data labour-intensive and EMRs currently inadequate as a data source for comprehensive public reporting (CIHI 2014a). The focus on developing better EMR standards and the increasing uptake of EMRs may lead to EMRs as a valuable data source in the future. While Canada lost a decade of reporting and the learning opportunities that might have ensued, the UK, one of the most advanced countries for public PHC performance reporting, has just completed a review of its first 10 years of major reporting efforts (Dixon et al. 2015). Some lessons from the UK may offer guidance to Canada's next efforts; specifically, more is not always better. Rather than multiple initiatives reporting on the health system, a single site for public reporting is recommended with different interfaces aimed at specific audiences such as the public, providers and decision-makers. The different interfaces should be linked, ensuring full transparency, but the one aimed at the public should contain fewer indicators focusing on those most important to the public, such as access and satisfaction. These experiences can help guide our approach to PHC reporting. In 2013, CIHI conducted a large consultation with Canadians to identify their priorities for health system reporting in building their ourhealthsystem.ca website. Only a single PHC system indicator, access to a regular family physician, is reported on that website. The wealth of information available from existing sources ranging from the CCHS to the Commonwealth Fund's primary care physician survey should enable a richer assessment of the performance of the PHC system across Canada.

The promise of a new health accord between the Federal Government and the Provinces (Liberal Party of Canada 2015a, 2015b) offers an opportunity to accelerate ongoing learning from national innovations through publicly reported performance information. As the Federal and Provincial Governments look forward to the next decade of health system reform and accountability, they should aim for a shared commitment to updating priorities for public reporting, developing a small common core set for the public and a broader set for decision-makers and ensuring comparability of data across the country using stable and accessible data sources including existing national and international sources. PHC is too important to this country to continue to lag behind in accountability and capacity for improvement.