Prepared by: Grant Thornton A Product of the CMS Alliance to Modernize Healthcare Federally Funded Research and Development Center Centers for Medicare & Medicaid Services (CMS) Prepared For: U.S. Department of Veterans Affairs At the Request of: Veterans Access, Choice, and Accountability Act of 2014 Section 201: Independent Assessment of the Health Care Delivery Systems and Management Processes of the Department of Veterans Affairs Assessment G (Staffing/Productivity/Time Allocation) September 1, 2015 Prepared for CAMH under: Prime Contract No. HHS-M500-2012-00008I Prime Task Order No. VA118A14F0373 This document was prepared for authorized distribution only. It has not been approved for public release. ©2015 Grant Thornton. All rights reserved. Assessment G (Staffing/Productivity/Time Allocation) This page intentionally left blank The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. ii Assessment G (Staffing/Productivity/Time Allocation) This report is the result of the contributions of several team members, who conducted site visits, processed and analyzed data, managed team activities, and enabled the Grant Thornton team to execute a rigorous study over a short period of time. The team was led by Grant Thornton, with integral support from team member FTI Consulting, and three independent contractors. Hillary Peabody – Project Manager Dr. Peter Erwin – Project Director Sharif Ambrose - Quality Partner Erik Shannon - Program Partner Nicole Nelson – Lead Editor Contributors Craig Ahrens Katie Beringer Colin Boyle Marjorie Broffman Sheba Ehteshami Pam Gusmanos David Howard Steve Kleinglass Alex Margolis Kyong Pak John Podlenski Eddie Sanders Jake Schreiber Terrie Searles Dr. Artie Shelton Eliza Spencer Laura (Stone) Cole Dr. Edwin Zarling Yilin Zhang The Grant Thornton Team expresses appreciation to our internal reviewers, Tamara Reynolds, Robert Shea, Jeff Johnson, and Stacy Tselekis; Gloria Funes for graphics support; and Karen Dixon for administrative support. The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. iii Assessment G (Staffing/Productivity/Time Allocation) This page intentionally left blank. The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. iv Assessment G (Staffing/Productivity/Time Allocation) Preface Congress enacted and President Obama signed into law the Veterans Access, Choice, and Accountability Act of 2014 (Public Law 113-146) (“Veterans Choice Act”), as amended by the Department of Veterans Affairs (VA) Expiring Authorities Act of 2014 (Public Law 113-175), to improve access to timely, high-quality health care for Veterans. Under “Title II – Health Care Administrative Matters,” Section 201 calls for an Independent Assessment of 12 areas of VA’s health care delivery systems and management processes. VA engaged the Institute of Medicine of the National Academies to prepare an assessment of access standards and engaged the Centers for Medicare & Medicaid Services (CMS) Alliance to Modernize Healthcare (CAMH)1 to serve as the program integrator and as primary developer of the remaining 11 Veterans Choice Act independent assessments. CAMH subcontracted with Grant Thornton, McKinsey & Company, and the RAND Corporation to conduct 10 independent assessments as specified in Section 201, with MITRE conducting the 11th assessment. Drawing on the results of the 12 assessments, CAMH also produced the Integrated Report in this volume, which contains key findings and recommendations. CAMH is furnishing the complete set of reports to the Secretary of Veterans Affairs, the Committee on Veterans’ Affairs of the Senate, the Committee on Veterans’ Affairs of the House of Representatives, and the Commission on Care. The research addressed in this report was conducted by Grant Thornton LLP, under a subcontract with The MITRE Corporation. Grant Thornton also subcontracted with FTI Consulting and other independent contractors in the conduct of the assessment. 1 The CMS Alliance to Modernize Healthcare (CAMH), sponsored by the Centers for Medicare & Medicaid Services (CMS), is a federally funded research and development center (FFRDC) operated by The MITRE Corporation, a not-for-profit company chartered to work in the public interest. For additional information, see the CMS Alliance to Modernize Healthcare (CAMH) website (http://www.mitre.org/centers/cms-alliances-to-modernize- healthcare/who-we-are/the-camh-difference). The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. v Assessment G (Staffing/Productivity/Time Allocation) This page intentionally left blank. The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. vi Assessment G (Staffing/Productivity/Time Allocation) Executive Summary In a health system comprised of more than 150 hospitals and nearly 1,400 community-based outpatient clinics, Vet Centers and domiciliaries,2 determining the staffing levels, caseload, and productivity required of VHA providers to meet the needs of over nine million enrolled Veterans3 is a complex task. Yet, adequate provider staffing levels and a health care system that enables its clinicians to be productive in delivering VHA’s population health focused model of care are essential to meeting the goal of timely, high quality care for our Veterans. This report details an assessment of the staffing levels, caseload, and productivity of providers across the VHA health care delivery system, and the allocation of providers’ time between delivery of patient care and other tasks such as administration, education, and research. This assessment addresses section 201(G) of the Veterans Access, Choice and Accountability Act of 2014 (Veterans Choice Act). Congress enacted the Veterans Choice Act to improve Veterans’ access to timely, high-quality health care. It included a request for an independent assessment of several aspects of the VHA health care delivery system. Part G of Section 201 requires an independent assessment of “the staffing level at each medical facility of the Department and the productivity of each health care provider at such medical facility, compared with health care industry performance metrics, which may include an assessment of the case load and number of patients treated by each health care provider, time spent by health providers on matters other than caseload, including time spent at an affiliate, conducting research, training, or supervising other health care professionals of the department.” To address this requirement, and under contract to the MITRE Corporation, the Assessment G team, led by Grant Thornton LLP, in partnership with FTI Consulting, and three independent contractors, conducted an assessment of current provider staffing levels, caseload, and productivity, in comparison to health care industry benchmarks. This included an in-depth assessment of nurse staff resource allocation, decision-making, and processes which impact provider productivity and efficiency. The Assessment G team’s approach involved both quantitative analyses (for example, benchmarking against nationally recognized industry benchmark surveys), as well as qualitative data analyses (root cause analysis review of data collected from over 700 interviews at 24 site visits, as well as data collected from VHA subject matter experts at VHA Central Office). The Assessment G team had several key findings and observations pertaining to the core assessment objectives: staffing, productivity, and time allocation. Staffing The Assessment G team analyzed VHA provider staffing levels and compared them to the private sector (using physician per population ratio industry comparisons) and identified some 2 Veterans Health Administration: About VHA. (2015). Retrieved from http://www.va.gov/health/aboutvha.asp 3 Bagalman, Erin. (2014) The Number of Veterans That Use VA Health Care Services: A Fact Sheet. p3. Congressional Research Service. Retrieved from https://www.fas.org/sgp/crs/misc/R43579.pdf The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. vii Assessment G (Staffing/Productivity/Time Allocation) of the challenges VHA faces in ensuring it has sufficient providers to meet demand. In summary, VHA’s provider staffing mix reflects VHA’s care model and the needs of the Veteran population, but conclusions from Assessment G about the adequacy of provider staffing levels and the impact of contract providers are difficult to make without consideration of the results of Assessment A (Demographics) and Assessment B (Capacity). VA medical centers face issues with provider vacancies, lengthy hiring processes, and competitive compensation, each of which can contribute to provider shortages. Key findings with respect to the VHA provider staffing levels are:  Finding 1: VHA specialties with the highest provider full time equivalent (FTE) levels include medicine specialties, mental health, and primary care, consistent with VHA’s care model and the needs of the Veteran population. Social Workers also represent a significant portion of provider FTEs. (See Section 2.2.2)  Finding 2: VHA does not systematically track fee-based provider productivity, and does not capture FTE level information for fee-based care providers. (See Section 2.2.3)  Finding 3: VHA physician staffing levels per population are, in most specialties, lower than industry ratios. These ratios are not sufficient to establish whether VHA is staffed to meet demand. One factor to consider is that even industry physician supply is not sufficient to meet demand in many specialties. Another factor to consider is that VHA uses Advanced Practice Providers (APPs) extensively, but APPs are not included in industry ratios. (See Section 2.2.6.) Productivity The Assessment G team assessed the productivity of VHA providers in comparison to providers in the private sector. This assessment used several common health care industry productivity measures: encounters (count of direct provider-patient interactions in which the provider diagnoses, evaluates, or treats the patient's condition), work relative value units (wRVUs—a measure of a provider’s output which takes into account the relative amount of time, skill, and intensity required to complete a given procedure), and primary care panel size (the number of unique patients for whom a care team is responsible). The Assessment G team considered VHA’s care model, benchmarked providers accordingly, and considered the barriers VHA faces in delivering care at a rate of productivity that matches health care systems in the private sector. In summary, we found that the average caseload or panel size of primary care providers is slightly below the level expected, but VHA’s target panel size is comparable to the private sector considering the type of patient population served and the findings described in the body of this report. VHA mental health providers are generally more productive than many of their peers in the private sector. VHA specialty providers on the other hand tend to lag the private sector in their productivity, although providers at high complexity VA facilities tend to have high productivity. There are several operational constraints or barriers which may explain these differences, such as: insufficient exam rooms and clinical or non-clinical support staff, and a lack of standard practices for managing daily staff absences. Based upon the Assessment G team’s observations and the findings of Assessment F (Clinical Workflow), we have concerns that providers may not be properly documenting all of their workload, which may explain some of the difference in The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. viii Assessment G (Staffing/Productivity/Time Allocation) productivity. The accuracy of documentation and coding shouldn’t be just considered for the sake of measuring wRVUs; coding is important to measuring whether clinical pathways are being appropriately followed and understanding care outcomes. Key findings with respect to the caseload and productivity of VHA providers are:  Finding 4: VHA measures the performance of its PCPs using panel size. VHA calculates a modeled panel size for providers based on a variety of factors at each facility. The model was developed based on research into the appropriate panel size for the unique needs of Veterans. (See Section 2.3.5.2)  Finding 5: In accordance with policy, VHA facilities establish a maximum panel size for each primary care provider which is often lower than the modeled panel size. The maximum figure takes into account specialized panel needs (for example, a geriatric population) and other factors deemed appropriate by the facility. (See Section 2.3.5.4).  Finding 6: The actual panel size of VHA primary care providers is lower than internal and external benchmarks. (See Section 2.3.5.5)  Finding 7: When compared to the private sector using wRVUs, there is a productivity gap in VHA specialty care. (See Section 2.3.6.3)  Finding 8: When encounters (visits) are used as a measure, the gap shrinks and VHA specialty care compares more favorably to the private sector. (See Section 2.3.6.4).  Finding 9: VHA mental health providers are more productive than academic medical center (American Medical Group Management Association [AMGMA]) benchmarks, as measured by both wRVUs and encounters. (See Section 2.3.6.5)  Finding 10: Overall, VHA specialty care providers are producing fewer wRVUs than private sector benchmarks; however, VHA specialty care providers at the highest complexity facilities are more productive than their peers. Further, the most productive VHA providers (those at the 75th percentile of VHA providers) are often more productive than the private sector. (See Section 2.3.6.6)  Finding 11: Productivity and access are important measures in population based health models like VHA that focus on patient outcomes, rather than volume. VHA’s Office of Productivity, Efficiency, and Staffing (OPES) reports on productivity and access offer tools for use by medical facilities. With some improvements to expedite adoption and regular use by medical centers, these tools could become key resources in optimizing productivity and maximizing access to care. (See Section 2.3.6.8)  Finding 12: VHA dentists see fewer patients on average than private sector benchmarks, but serve a population with special needs. The dentistry patient population of VHA generally has a compensable service-connected dental disability, is older, has more complex injuries, and may present for dental care following years of dental neglect. (See Section 2.3.7.4). Key findings with respect to the barriers VHA faces in delivering care that is equally as productive as the private sector are: The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. ix Assessment G (Staffing/Productivity/Time Allocation)  Finding 13: Insufficient exam rooms and poor configuration of space limits providers’ productivity, ability to maximize patient throughput, and reduces patient access. (See Section 2.3.8.3)  Finding 14: Clinical and administrative support staff ratios are insufficient and may limit provider productivity. (See Section 2.3.8.4)  Finding 15: Insufficient clinical and administrative support staff results in providers and clinical support staff not working to the top of their licensure. (See Section 2.3.8.4.1).  Finding 16: While there has been widespread implementation of the Patient Aligned Care Team (PACT) model in primary care clinics and the National Nurse Staffing Methodology in many areas of inpatient care, there are no current VHA standards for staffing levels and/or mix in specialty clinics, with the exception of eye clinics. Furthermore, VHA OPES has developed state of the art tools for managing staffing and productivity, but these tools will require improvements for leaders to more effectively leverage them in resource decisions. (See Section 2.3.8.4.2)  Finding 17: Organizational siloes and separate reporting lines exist for physicians, nurses and medical service administrators at a majority of VA Medical Centers (VAMCs). As a result, service chiefs do not have control over the resourcing and performance of their clinical support staff (nurses) or clerical and administrative support staff. (See Section 2.3.8.4.3)  Finding 18: Many facilities do not have a centralized staffing office or nurse float pool to address daily staff variances or absences. (See Section 2.3.8.4.4)  Finding 19: During site visits and interviews with VHA Central Office leaders, we consistently heard concerns that providers do not fully document and accurately code all of their clinical workload. (See Section 2.3.8.5) Provider Time Allocation The Assessment G team assessed how VHA providers spend their time, to include the time that VHA providers spend on non-patient care activities, particularly time spent on education and research activities, as well as time spent overseeing residents in a clinical setting, and time spent at academic affiliate medical centers. We compared VHA providers’ clinical time to private sector data, as well. In summary, we found that VHA providers spend approximately the same proportion of their time on clinical care activities as the private sector, despite a rich research output. Key findings with respect to VHA providers’ time allocation are:  Finding 20: VHA physicians spend a comparable proportion of total time devoted to clinical activities as private sector physicians. There is some potential difference in the definition of direct patient care used by the private sector, specifically with respect to training, teaching and research, but we believe this represents only a small proportion of a provider’s direct patient care time. (See Section 2.4.2)  Finding 21: Across all VHA providers, less than two percent of time is devoted to research. Since provider time spent devoted to clinical care activities is comparable to the private sector, it does not appear that research activities reduce providers’ time spent treating patients. Despite the overall low proportion of time spent on research, the The views, opinions, and/or findings contained in this report are those of Grant Thornton and should not be construed as an official government position, policy, or decision. x