Value-based contracts between managed health care systems and school systems are feasible and potentially clinically and financially gainful.

The rise of value-based contracts in pediatrics and performance-based contracts in education has signaled the readiness of both sectors to adopt financial risk for high-cost groups of children.

Schools, especially those using a population-based (capitated) formula, are likely to be more interested in long-term outcomes and may fund some interventions. Health care organizations bearing financial risk and long-term commitments may do the same.

Value-based contracting between the 2 systems can initially focus on shared high-need, high-cost cases (eg, asthma, attention-deficit/hyperactivity disorder).

Value-based health systems, such as accountable care organizations (ACOs), have made progress in improving care value.

 However, provision of health care for children, the cornerstone of our future society, remains haphazard and subpar.

 We suggest partnerships between value-based health care systems and “capitated” special education systems as a novel relationship to drive high-value pediatric care.

Health care and education are tightly connected for the pediatric population—a healthier child is a better learner, and vice versa.

 Early childhood education, enrollment in higher education, and high school graduation are highly correlated with health.

 Fundamentally, the goals of improving child well-being make child health care and education inseparable, yielding a hypothesis that a partnership between health care and school systems has potential. Historically, however, child health care systems have been reimbursed for caring primarily for ill children in fee-for-service health settings, and schools were focused on providing education and limited health care services through special education for the children with the most needs in schools. Neither group had strong motivation to partner with the other.

The advent of alternative payment models like capitation in health care changed the incentives for health care systems, increasing the focus on prevention and savings. Interestingly, 8 states (eg, California, Massachusetts) also use a “capitated” system in special education, namely census-based education,

 which pays a school district a lump sum for special education according to the number of the whole population of students, regardless of the actual number of special education students. Both types of capitation shift the risks of overspending to those delivering services (health care organizations or school districts) and encourage the 2 systems to support children with highest needs and long-term chronic conditions across settings.

Educating a child with special needs is expensive, costing 2 to 3 times more than for a student with typical development. Using California as an example, the $26,000 average annual educational costs of a student in special education are almost triple those of a student in regular education.

 With more students needing special education, the total expenditure increased by 28% in the decade between 2007-2008 and 2017-2018.

 Likewise, the medical costs of these similar groups of children with chronic and complex conditions are 2 to 10 times more than those of children with typical development.

Some high-need, high-cost diseases may spark shared interests in the 2 systems (see the 

 for examples). For example, 1 in 59 children in the United States is affected by autism, a condition generally considered to be the most expensive developmental disorder. The total cost of care for individuals with autism is estimated at $236 billion to $262 billion annually in the United States.

 Although there is no effective cure for autism, early intervention (EI) before the age of 5 years is critical to improving cognitive, social, and language development.

 Because these services are rarely integrated between special education and health care, better coordination has the potential to substantially reduce the cost of care for autism for both systems. Initial estimates showed that 2 to 3 years of EI would save almost $20,000 in therapy-related health care costs per child in the year after EI

 and approximately $200,000 in educational costs per child across 18 years of education.

Despite shared motivation and new opportunities to align value, we know of no specific examples of shared, value-based contracting between health care and school systems. Yet, based on a number of emerging activities, we believe that such value-based contracts between the 2 systems are more than a pipe dream. At a national level, both HHS and the Department of Education strongly encourage embedding health supports in educational settings for children, setting the groundwork for more extensive clinical and financial collaboration.

At a local level, there are many school-based health projects, but at least 2 examples are associated with capitated pediatric populations. Partners for Kids (PFK), a pediatric ACO in Ohio, carries full financial risk for approximately 330,000 Medicaid-insured children. To ensure population health, PFK has introduced care in schools. PFK started 2 dozen school-based health centers and an asthma therapy program through which children can receive medication at school in order to decrease the percentage of emergency department (ED) visits related to asthma and to increase school attendance. Given the heavy burden of behavioral health on PFK, specialists from PFK also assisted more than 125 schools in implementing the PAX Good Behavior Game, an evidence-based preventive intervention that reduces drug dependence, suicidal ideation, drop-out rates, and special education rates.

 PFK has a cost growth rate 2 to 6 times less than that of fee-for-service.

 It is unclear, however, how much these school programs have contributed to PFK’s lower costs. A similar value-based program developed by Boston Children’s Hospital also partnered with schools and the larger community to deliver asthma support. This program reduced missed school days by 20% and ED visits by 50%.

 These value-based efforts are growing (eg, Colorado Medicaid Accountable Care Collaborative, Wisconsin Children’s Hospital). Although these initiatives do not involve value-based contracts between the 2 systems per se, they have highlighted the possibility of both systems pursuing aligned health- and-education-related quality metrics together.

The structure of partnerships between schools and health systems varies based on incentives. First, schools are likely to be more interested in long-term outcomes than most health plans because children do not leave school districts frequently,

 but they may shift health plans. School systems, especially those that have capitated special education, are motivated to pay for outcomes that are educationally and financially gainful, such as a reduced rate of special education and an improved rate of attendance. This motivation was evident in recent pay-for-performance contracts between school districts and early intervention agencies.

 For instance, school districts could set up pay-for-performance contracts with health systems to leverage health systems’ expertise in specialized early intervention to improve at-risk student outcomes and to reduce the rate of special education. Then, school districts could financially incentivize health systems if the rate of special education goes down and student outcomes are improved. These school-based service lines may become a new revenue stream for health systems. Second, capitated health plans, managed care organizations (MCOs), or ACOs are often more interested in short-term outcomes due to high churn rates and annual incentive contracts from state Medicaid agencies. Thus, these health plans are more likely to finance interventions that can generate a quick return on investment (eg, school-based asthma clinics to reduce ED visits, school-based adolescent wellness visits to receive Medicaid incentives). For instance, capitated health systems may set up upside sharing contracts with school-based health clinics/schools as a way to incentivize quality school-based care for asthma.

Value alignments between the 2 systems will necessitate some foundational elements and contracting mechanisms. First, health care and school systems would have to share coverage of similar populations geographically. They would develop trusting relationships, identify goals, and agree on metrics. Second, data sharing systems and agreements attuned to both the Health Insurance Portability and Accountability Act of 1996 and the Family Education Rights and Privacy Act would be set up. Third, the 2 systems would have to agree on the services through which to achieve the goals and then develop contracts. Indeed, value-based contracting will not solve all structural barriers, such as a shortage of behavioral health clinicians. To enable quality pediatric care at scale, health and school systems will need to continually advocate for system changes (eg, increased total funding for pediatric care).

The need for better coordination between the 2 systems is even more evident now as coronavirus disease 2019 (COVID-19) has put both health care systems and school systems under great stress. Primary care facilities for children are under tremendous financial pressure, with many sites serving low-income children expected to close. Massive delays in the receipt of routine health care have occurred, placing children at risk of vaccine-preventable disease as school resumes. Schools have also been subjected to intense pressures. They were primary places to provide health-related services for many school-age children, ranging from primary care at school-based health centers to behavioral/mental health services through individualized education programs before COVID-19. However, large-scale school closures have pushed off both education and its affiliated health care services, even when telehealth or distance learning is available. Even here, new opportunities exist to coordinate digital access for school-age children and their families for both health care and education, ensuring access to support at home, health care, and school. Treatment planning and progress monitoring can be done in a collaborative manner in order to maximize continuity of services during and after COVID-19.

Also, with nonurgent care being postponed or being converted to telehealth visits, health care systems may have difficulty reaching children and their families. Health care systems, especially MCOs that share a similar geography of children with particular school districts, may take advantage of school-based communication channels (which have frequent contacts with parents) to disseminate population-based health monitoring in order to inform population-based health strategies for children in the shared geography. Such population-based health monitoring many include measures on social determinants of health needs and unmet care needs. The likely repercussions of COVID-19 will continue to prompt creative, collaborative efforts to support children more efficiently. 

In sum, value alignment between health care and school systems has promise but is replete with challenges. A start could be made, however. MCOs and ACOs are likely to be the most fruitful sites to initiate the collaboration. Expanding on the few existing partnerships, such as PFK, and evaluating their effects more extensively on educational outcomes and school financing could be the next step.

The authors thank Dr Elliot Fisher, Dr Kolbe Lloyd, and Dr Bill Koski for their suggestions on this paper. They also thank Dr Arnold Milstein, Dr Nirav Shah, and Dr Sarah Rosenbaum for conversations that prompted the conceptualization of this paper.


 Clinical Excellence and Research Center, Stanford University (VW), Stanford, CA; Institute for Exceptional Care (HP), Indianapolis, IN; Nationwide Children’s Hospital (KK), Columbus, OH; The Ohio State University (KK), Columbus, OH.

 Dr Kelleher’s hospital is part-owner of a not-for-profit example listed in the manuscript (Partners for Kids). The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (VW, HP, KK); acquisition of data (VW, HP, KK); analysis and interpretation of data (VW, HP); drafting of the manuscript (VW, HP, KK); critical revision of the manuscript for important intellectual content (VW, HP, KK); provision of patients or study materials (VW); obtaining funding (VW); administrative, technical, or logistic support (VW); and supervision (VW, KK).

 Venus Wong, PhD, Clinical Excellence and Research Center, Stanford University, 365 Lasuen St, Stanford, CA 94305. Email: vwhwong@stanford.edu.

1. Overview of research on ACO performance. National Association of ACOs. Accessed June 20, 2020. https://www.naacos.com/assets/docs/pdf/NAACOS-ACO-PerformanceResearchReport7.19.18edited.pdf

2. Davis K, Stremikis K, Squires D, Schoen C. Mirror, mirror on the wall, 2014 update: how the U.S. health care system compares internationally. The Commonwealth Fund. June 16, 2014. Accessed June 19, 2020. https://www.commonwealthfund.org/publications/fund-reports/2014/jun/mirror-mirror-wall-2014-update-how-us-health-care-system

3. Vital Signs. National Academy of Medicine. Accessed June 20, 2020. https://nam.edu/vital-signs-initiative/

4. Basch CE. Healthier students are better learners: a missing link in school reforms to close the achievement gap. 

. 2011;81(10):593-598. doi:10.1111/j.1746-1561.2011.00632.x

5. Lanza HI, Huang DYC. Is obesity associated with school dropout? key developmental and ethnic differences. 

. 2015;85(10):663-670. doi:10.1111/josh.12295

6. Parker E. 50-state comparison: K-12 special education funding. Education Commission of the States. March 20, 2019. Accessed June 11, 2020. https://www.ecs.org/50-state-comparison-k-12-special-education-funding/

7. Petek G. Overview of special education in California. California Legislative Analyst’s Office. November 6, 2019. Accessed June 11, 2020. https://lao.ca.gov/reports/2019/4110/overview-spec-ed-110619.pdf

8. Hoefgen ER, Andrews AL, Richardson T, et al. Health care expenditures and utilization for children with noncomplex chronic disease. 

. 2017;140(3):e20170492. doi:10.1542/peds.2017-0492

9. Asthma surveillance data. CDC. Updated January 28, 2020. Accessed May 8, 2020. https://www.cdc.gov/asthma/asthmadata.htm

10. Nurmagambetov T, Kuwahara R, Garbe P. The economic burden of asthma in the United States, 2008-2013. 

. 2018;15(3):348-356. doi:10.1513/AnnalsATS.201703-259OC

11. Data & statistics on autism spectrum disorder. CDC. Updated September 25, 2020. Accessed June 30, 2020. https://www.cdc.gov/ncbddd/autism/data.html

12. Lavelle TA, Weinstein MC, Newhouse JP, Munir K, Kuhlthau KA, Prosser LA. Economic burden of childhood autism spectrum disorders. 

. 2014;133(3):e520-e529. doi:10.1542/peds.2013-0763

13. Danielson ML, Bitsko RH, Ghandour RM, Holbrook JR, Kogan MD, Blumberg SJ. Prevalence of parent-reported ADHD diagnosis and associated treatment among U.S. children and adolescents, 2016. 

. 2018;47(2):199-212. doi:10.1080/15374416.2017.1417860

14. Gupte-Singh K, Singh RR, Lawson KA. Economic burden of attention-deficit/hyperactivity disorder among pediatric patients in the United States. 

. 2017;20(4):602-609. doi:10.1016/j.jval.2017.01.007

15. Robb JA, Sibley MH, Pelham WE Jr, et al. The estimated annual cost of ADHD to the U.S. education system. 

. 2011;3(3):169-177. doi:10.1007/s12310-011-9057-6

16. Buescher AVS, Cidav Z, Knapp M, Mandell DS. Costs of autism spectrum disorders in the United Kingdom and the United States. 

. 2014;168(8):721-728. doi:10.1001/jamapediatrics.2014.210

17. Dawson G, Rogers S, Munson J, et al. Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model. 

. 2010;125(1):e17-e23. doi:10.1542/peds.2009-0958

18. Cidav Z, Munson J, Estes A, Dawson G, Rogers S, Mandell D. Cost offset associated with Early Start Denver Model for children with autism. 

. 2017;56(9):777-783. doi:10.1016/j.jaac.2017.06.007

19. Chasson GS, Harris GE, Neely WJ. Cost comparison of early intensive behavioral intervention and special education for children with autism. 

. 2007;16(3):401-413. doi:10.1007/s10826-006-9094-1

20. Boyer KB, Chang DI. Case study: Nationwide Children’s Hospital: an accountable care organization going upstream to address population health. National Academy of Medicine. April 24, 2017. Accessed June 1, 2020. https://nam.edu/case-study-nationwide-childrens-hospital-an-accountable-care-organization-going-upstream-to-address-population-health/

21. Kelleher KJ, Cooper J, Deans K, et al. Cost saving and quality of care in a pediatric accountable care organization. 

. 2015;135(3):e582-e589. doi:10.1542/peds.2014-2725

22. Sommer SJ, Bhaumik U, Tsopelas L, et al. Boston Children’s Hospital Community Asthma Initiative replication manual: needs assessment, implementation and evaluation. Boston Children’s Hospital. 2013. Accessed June 1, 2020. https://www.childrenshospital.org/~/media/centers-and-services/programs/a_e/community-asthma-initiative/replicationmanual2cfinal2c92413.ashx?la=en

23. CPS historical migration/geographic mobility tables. United States Census Bureau. Accessed May 8, 2020. https://www.census.gov/data/tables/time-series/demo/geographic-mobility/historic.html

24. Carnoy M, Marachi R. Investing for “impact” or investing for profit? social impact bonds, pay for success, and the next wave of privatization of social services and education. National Education Policy Center. February 4, 2020. Accessed June 30, 2020. https://nepc.colorado.edu/publication/social-impact-bonds

Articles

COMMENTARY

This article argues that value-based health systems may contract with school districts engaged in capitated special education to achieve better patient outcomes and lower costs for the pediatric population.

Achieving Better Value in Pediatric Care: School Systems as Clinical and Financial Partners?

To address a possible “twindemic” of coronavirus disease 2019 and influenza and the ongoing health issues in vulnerable communities, and to help rebuild trust among these communities, the health care industry must directly and consistently engage with communities of color to foster authentic relationships with those who already have long-standing connections, visibility, and trust in these communities. In addition, it has to listen and act on recommendations from this coalition, as well as look at health care in terms of preventive social determinants of health—issues that determine better or worse health outcomes for individuals.

In September, the United States surpassed 200,000 deaths from the coronavirus disease 2019 (COVID-19) pandemic, with some experts predicting that the death toll could nearly double by the end of 2020. September also marked the start of fall and the beginning of the influenza season, which, although not as deadly as COVID-19, accounts for approximately 20,000 deaths annually. This year, the fall season is accompanied by an urgent warning of an impending “twindemic” of COVID-19 and influenza by the CDC.

COVID-19 has affected our lives in ways we could never have imagined and has made many of us more attuned to our overall health and the need for preventive measures. A recent poll

 conducted by EmblemHealth found that nearly half of New Yorkers surveyed are more likely to get the influenza vaccination this year than in years past, and health systems across the country are reporting unprecedented demand for influenza vaccinations. But, as the past few months have revealed, access to health care in our country is deeply unequal, especially for communities of color. Nowhere has this been more apparent than in New York, the original epicenter of the COVID-19 pandemic in the United States and a microcosm of the devastating impact that racial disparities have on the health of our communities.

At the height of the pandemic in April, Black New Yorkers were dying at a higher rate than any other racial and ethnic group, and at twice the rate of White New Yorkers.

 As the CDC warns of a surge this fall and winter, these rates collide with the sobering statistics around influenza vaccination adoption. Among the 1005 New Yorkers who responded to EmblemHealth’s survey, Black women were significantly less likely to seek access to the influenza vaccination this year than any other group. Only 35% of Black women stated that they have received the influenza vaccination consistently over the past 3 years, compared with 56% of New York City’s general population. Concerns about effectiveness and safety were listed as top reasons to not get the influenza vaccination, which foreshadows the challenges that we may also face when the COVID-19 vaccine becomes available.

As a physician and a Black woman, I view these findings as alarming but, unfortunately, not unexpected. Over several decades, the health care industry has sown mistrust among communities of color, for reasons ranging from the Tuskegee syphilis experiment in the 1930s to the underrepresentation of Black patients in research and vaccine trials that persists to this day.

 In many cases, the former has shaped the latter, with the historical legacy of exploitation fueling the ongoing distrust of research and health care systems in the United States. In the case of Black women and the twindemic, this has the potential to be a lethal combination.

Although COVID-19 has exhibited deadlier consequences among men, women often hold jobs that put them at a higher risk of infection, such as domestic workers, nurses, or home health aides. A 

 analysis of Census data earlier this year found that 1 in 3 jobs held by women have been categorized as “essential,”

 and data from the Center for Economic and Policy Research indicated that women of color were more likely to have essential jobs.

 Although many female essential workers struggle with taking time off for routine care, the primary deterrents to getting the influenza vaccination among Black women are concerns about effectiveness, safety, and adverse effects. All of these fears are heightened by the ongoing political interference in public health practices and messaging. Nevertheless, the remedy to the fear and uncertainties among underrepresented and disenfranchised communities of color remains within the health care industry’s power and capacity to change.

The health of our nation is at a critical crossroads: one where Black women stand to lose the most as a consequence of historical and systemic failures and their overrepresentation on the front lines of the pandemic. But it’s not too late to enact change that will save the lives of thousands of Black women now and in the future. We can do this by engaging those who are influential in the Black community, from church leaders to hairdressers and barbers; by addressing the shortage of Black physicians,

 whose voices and experiences represent the communities they serve; and by recruiting more Black men and women to participate in research and vaccine trials—not only for the COVID-19 vaccine, but for conditions such as cardiovascular disease and cancer, of which Black individuals have the highest death rate

 and the shortest survival compared with any other racial and ethnic group in the United States.

The consequence of populations of color not being represented in research, testing, and vaccinations is that, ultimately, we will not have equitable access to cures, and without community empowerment, we are doomed to repeat history’s mistakes. Black women and our families deserve better, and our lives depend on it.


 The author reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis; provision of patients or study materials; administrative, technical, or logistic support; and supervision.

 Diane M. Ashton, MD, MPH, EmblemHealth, 55 Water St, New York, NY 10041. Email: dashton@emblemhealth.com.

1. EmblemHealth study finds New Yorkers overall more likely to seek flu vaccine this year amid COVID-19. News release. EmblemHealth; October 1, 2020. Accessed October 21, 2020. https://www.emblemhealth.com/news/press-releases/EmblemHealth-Study-Finds-NewYorkers-Overall-More-Likely-to-Seek-Flu-Vaccine

2. Black people in NYC twice as likely to die from COVID as white people: data. ABC News. April 17, 2020. Accessed October 21, 2020. https://abcnews.go.com/US/black-people-nyc-die-covid-white-people-data/story?id=70208362

3. Frederick WAI, Montgomery Rice V, Carlisle DM, Hildreth JEK. We need to recruit more Black Americans in vaccine trials. 

. September 11, 2020. Accessed October 21, 2020. https://www.nytimes.com/2020/09/11/opinion/vaccine-testing-black-americans.html

4. Robertson C, Gebeloff R. How millions of women became the most essential workers in America. 

. April 18, 2020. Accessed October 21, 2020. https://www.nytimes.com/2020/04/18/us/coronavirus-women-essential-workers.html

5. Brown H, Rho HJ, Fremstad S. Mothers in frontline industries deserve better. Center for Economic and Policy Research. May 8, 2020. Accessed October 21, 2020. https://cepr.net/mothers-in-frontline-industries-deserve-better/

6. Sobowale K. We need more Black physicians. 

. July 17, 2020. Accessed October 21, 2020. https://www.scientificamerican.com/article/we-need-more-black-physicians/

Cancer Facts & Figures for African Americans 2019-2021

. American Cancer Society. 2019. Accessed October 21, 2020. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/cancer-facts-and-figures-for-african-americans/cancer-facts-and-figures-for-african-americans-2019-2021.pdf

WEB EXCLUSIVE

This year, the fall season is accompanied by an urgent warning from the CDC of an impending “twindemic” of coronavirus disease 2019 and influenza. Despite the warnings, Black women are not lining up for vaccinations.

Why Even the Threat of a Twindemic Isn’t Swaying the Most Vulnerable: Black Women

 To describe real-time changes in medical visits (MVs), visit mode, and patient-reported visit experience associated with rapidly deployed care reorganization during the coronavirus disease 2019 (COVID-19) pandemic.

 Cross-sectional time series from September 29, 2019, through June 20, 2020.

 Responding to official public health and clinical guidance, team-based systematic structural changes were implemented in a large, integrated health system to reorganize and transition delivery of care from office-based to virtual care platforms. Overall and discipline-specific weekly MVs, visit mode (office-based, telephone, or video), and associated aggregate measures of patient-reported visit experience were reported. A 38-week time-series analysis with March 8, 2020, and May 3, 2020, as the interruption dates was performed.

 After the first interruption, there was a decreased weekly visit trend for all visits (β

 < .05), an immediate decrease in office-based visits (β

 < .01), increase in telephone-based visits (β

 < .01), and increased video-based visit trend (β

 < .01). After the second interruption, there was an increased visit trend for all visits (β

 < .01), immediate increase in video-based visits (β

 < .05), increased office-based visit trend (β

 < .01), and decreased trend in video-based visits (β

 < .01). After the second interruption, there were increased weekly long-term visit trends for the proportion of patients reporting “excellent” as to how well their visit needs were met for all visits (β

 < .01). Video-based visits had the highest proportion of respondents rating “excellent” as to how well their scheduling and visit needs were met.

 COVID-19 required prompt organizational transformation to optimize the patient experience.

Large, integrated group practices have the essential framework and infrastructure to allow for effective coordination of care, whether in person or virtually, to effectively function during crises. During the coronavirus disease 2019 pandemic, the following were observed:

There were significant immediate decreases in weekly visit trends for all visits, primary care, and specialty care, with a significant increase in weekly visit trend for behavioral health. Indicators of longer-term recovery for overall visits were observed.

A significant immediate decrease occurred in weekly visits and trends for office-based visits, and immediate increases occurred in telephone-based visits and weekly visit trend for video-based visits.

Significant increases occurred in long-term trends for the proportion of patients rating “excellent” as to how well their visit needs were met for telephone-based, video-based, and all visits combined.

Video-based visits had the highest proportion of respondents rating “excellent” as to how well their scheduling and visit needs were met before and throughout the pandemic.

The coronavirus disease 2019 (COVID-19) pandemic has placed unprecedented strain on health systems. Prudent federal and state public health policies that include varying degrees of physical and social isolation and distancing have required rapid changes in the usual delivery of patient care.

 Among CDC recommendations are the rescheduling of nonurgent care and increased use of alternative approaches to in-person visits.

 The Coronavirus Aid, Relief, and Economic Security Act accelerated expansion and flexibility of telehealth activities.

 Physical and social distancing will likely continue for the foreseeable future and have a negative impact on what is typically a recession-proof health care economy.

Little is known about how the COVID-19 pandemic will ultimately affect practice and patient experience. An April 2020 report noted that 97% of physician practices experienced a direct or indirect negative financial impact, as evidenced by a 60% decrease in volume and corresponding 55% decrease in revenue.

 Others have reported a 30% to 70% reduction in independent primary care practice visit volume.

 Recently, a news report cited Department of Commerce data estimating an 18% annualized decline in health care spending based on the first 3 months of 2020.

 Negative financial impacts are likely exacerbated with the expansion of telehealth technology

 and purchase of personal protective equipment.

 One health system reported increasing telemedicine visits from 0 to 70,000 visits in a 1-month period, although it did not describe the delivery mode (ie, telephone or video).

 Medical specialties with a track record of delivering virtual care are not exempt from the pandemic’s impact, as another recent report suggested that radiology practices will likely experience a 50% to 70% decrease in volume lasting at least 3 to 4 months.

Managing immediate pandemic hazards while continuing to provide routine medical and preventive care that fulfills the patient care experience, population health, and per capita cost of care dimensions of the Triple Aim

 is challenging. Preparing and responding to the medical caseload surge and implementation of necessary public health measures to manage the COVID-19 pandemic in large, integrated health systems requires rapid and carefully choreographed organizational transformation. Health systems and medical practices unable to respond promptly to these demands face enhanced risk of failure and insolvency.

This report describes real-time changes in medical visits (MVs), visit mode, and patient visit experience prior to and during the COVID-19 pandemic in Kaiser Permanente Mid-Atlantic States (KPMAS), a large integrated health system, in accordance with the Standards for Quality Improvement Reporting Excellence guideline.

Effective January 2020, KPMAS had 767,163 members across Maryland, District of Columbia (DC), and Northern Virginia geographic regions. Membership reflects diverse regional demographics and encompasses public (Medicaid, Medicare, Affordable Care Act health insurance exchange plans), private, individual, and charity care lines of business.

 Within KPMAS, medical care is provided by the Mid-Atlantic Permanente Medical Group (MAPMG), an integrated medical practice of more than 1700 physicians that includes more than 50 services and specialties. Medical care is provided at 34 medical centers and 12 partner hospitals. Medical centers primarily provide outpatient and observational care. KPMAS contracts with local hospitals to provide inpatient care while using its own providers. These 12 “premier” hospitals, many with dedicated KPMAS units exclusively for KPMAS members, provide medical, surgical, and intensive care and rehabilitation services to KPMAS members, most often by MAPMG physicians and staff.

COVID-19 case accrual in KPMAS began on March 1, 2020. Between March 1, 2020, and June 20, 2020, KPMAS had 9256 patients with a confirmed COVID-19 infection (ie, clinical diagnosis or positive laboratory test [either within or external to KPMAS]).

In response to external public health and clinical guidance, KPMAS made rapid structural changes to effectively reorganize care and transition from office-based to virtual care platforms beginning in late February 2020.

 Medical centers were reorganized for in-person care to include screening for fever, cough, and shortness of breath at every entrance; triaging and testing patients outside of the centers; and colocation of patients at risk for COVID-19. Elective procedures were postponed. Of the 34 medical office buildings (MOBs), only 15 were designated to provide continued clinical care (consistent with Maryland regulations) and 8 provided urgent care. Those MOBs not providing clinical care remained open for pharmacy, laboratory, and radiology services. Six MOBs maintained 24-hour operations, with the remainder having adjusted hours. Drive-up testing for COVID-19 was available during limited times at 6 MOBs in DC, Maryland, and Virginia. Although core contract hospitals were not able to accept all KPMAS patients because of the dramatic increase in COVID-19–related admissions, emergency hospital licensure was obtained, and 163 hospital beds were added in 3 locations to alleviate the burden on community hospital partners providing focused care for critically ill patients with COVID-19.

Daily internal emails were generated to update physicians and staff about the rapidly evolving status of the COVID-19 pandemic within KPMAS. Ongoing COVID-19 testing and diagnosis education was provided for physicians and staff via emails and webinars. All operations documents, including workflows, process, and organization, were made available on a SharePoint website accessible to all providers and staff at any time. Weekly COVID-19 lunchtime interactive seminars were provided to address physician and staff questions about testing guidelines, coding, resource management and KPMAS electronic health record (EHR), Epic-based HealthConnect, and any other related COVID-19 issues. Training sessions were also archived for later viewing.

 Appointment rescheduling from in person to telehealth was implemented. All face-to-face appointments, including behavioral health care visits, were given the option to convert to video visits as the goal. Because CMS waived guidelines requiring a license to practice in-state, it became possible to provide care to patients outside of DC, Maryland, and Virginia through telephone and video visits. Existing telemedicine infrastructure made it possible to scale up, including the availability of a telehealth support team for member and physician support for HealthCare Anywhere and integrated video visits (IVV). The HealthCare Anywhere app was used to provide patient care appointments on the fly (ie, not requiring prior appointments or scheduling). Members could be invited via phone number or telephoned by care providers for appointments if the KP.org website was temporarily unavailable or if members were not registered on KP.org. IVVs could be conducted via Enterprise Clinician Connect (eCC) for members already registered on KP.org. Epic-based HealthConnect could be used by providers to switch appointments from face-to-face to video visits. eCC sends alerts to physicians to update them when a member has joined for a video visit. Cortext Out of Network is a Health Insurance Portability and Accountability Act–compliant app that can be used to send text messages from physicians to members to update appointment status if needed. Instructions for conducting video visits can be sent to members via HealthConnect. Member support included interpreter services for the diverse population served by KPMAS.

Common operational metrics were used to describe medical practice impact, response, and patient experience associated with the organizational changes implemented. Specifically, MVs (ie, visits per week) by medical discipline (ie, primary care, specialty care, and behavioral health care) and visit mode (ie, office-based, telephone, and video) were measured using weekly visit counts.

Patient visit experience was measured by patient response to the following: (1) How well your needs and schedule were taken into consideration when this [office, telephone, video] appointment was scheduled and (2) How well your needs were met in this [office, telephone, or video] appointment/visit. Response sets were “no experience,” “poor,” “fair,” “good,” “very good,” and “excellent.”

Patient experience surveys were administered in accordance with an external vendor contract. Although there are well-recognized challenges with measuring and reporting sampling and response rates for patient experience surveys,

 they are commonly used to ascertain patient perceptions of their health care experience. The 77,403 responses for the first question and 82,762 responses for the second question provide a robust sense of patient experience during the observation period.

 (September 29, 2019, through June 20, 2020) was performed using March 8, 2020, and May 3, 2020, interruption dates to describe real-time changes in both effect and trend of the previously defined operational metrics during the immediate phase (post–COVID-19 period 1) and long-term phase (post–COVID-19 period 2) of the pandemic. The pre–COVID-19 period included 23 weekly time intervals beginning September 29, 2019, followed by 8-week and 7-week sequential post–COVID-19 periods. The first interruption (March 8, 2020) for post–COVID-19 period 1 allowed for sufficient lag after the CDC released official guidance for health care facilities on February 29, 2020,

 and followed the week after KPMAS experienced its first COVID-19 case. Post–COVID-19 period 2 began May 3, 2020, coinciding with rapidly evolving federal, state, and local policy changes.

Mean weekly MVs were described overall, by discipline, and by visit mode for the pre–COVID-19 period and post–COVID-19 periods 1 and 2. Patient visit experience was described overall, stratified by visit mode, and by the pre–COVID-19 period and post–COVID-19 periods 1 and 2. For patient visit experience, the mean proportion of respondents reporting an excellent response (ie, top-box analysis) was used. Top-box analysis is commonly used for the Consumer Assessment of Healthcare Providers and Systems

 and is consistent with operational reporting practices in KPMAS. Aggregate differences among the pre–COVID-19 period and post–COVID-19 periods 1 and 2 were evaluated using independent-groups, 1-way analysis of variance with Bonferroni correction for multiple comparisons.

A time-series analysis was performed using the 

 command (Stata/MP 16.1 for Windows [StataCorp LLC]) and assumed the following form:

 is the summary outcome (ie, medical visits, patient experience rating) at each time point. 

 is the dichotomous indicator variable representing the March 8, 2020, interruption; 

 is the interaction between time and the indicator for the first interruption; 

 is the dichotomous indicator variable representing the May 3, 2020, interruption; and 

 is the interaction between time and the indicator for the second interruption. 

 is the trend for the pre–COVID-19 period; 

 is the effect (ie, change in the level of MVs and patient experience rating) immediately after post–COVID-19 period 1; 

 is the trend difference between the pre–COVID-19 period and post–COVID-19 period 1; 

 is the effect immediately after post–COVID-19 period 2; and 

 is the trend difference between post–COVID-19 period 1 and post–COVID-19 period 2. Additional posttrend analyses were performed to estimate the post–COVID-19 period 1 (

) trend, and trend difference between the pre–COVID-19 period and post–COVID-19 period 2 (

All models used the Newey-West method for estimating standard errors and a 6-lag order model to account for potential autocorrelation and heteroskedasticity. The Cumby-Huizinga post hoc test was used to evaluate the autocorrelation structure. Alpha was set to < .05.

The KPMAS Institutional Review Board determined that this activity was not human subjects research.

More than 2.2 million MVs and more than 77,000 patient visit experience responses were used to describe MVs, visit mode, and patient-reported visit experience during the study period.

For all visits, there was an immediate decrease in mean weekly MVs from the pre–COVID-19 period to post–COVID-19 period 1 (63,415.13 to 49,317.88; 

). A decrease in weekly primary care visits was also observed (34,396.87 to 29,948.13; 

 < .05) for post–COVID-19 period 1. For specialty care, there was a decrease in mean weekly MVs from 25,143.87 in the pre–COVID-19 period to 15,115.38 in post–COVID-19 period 1 and 16,870.43 in post–COVID-19 period 2 (

 < .01). An increase in mean weekly MVs for behavioral health care from the pre–COVID-19 period to post–COVID-19 period 2 was also observed (3874.39 to 4606.43; 

From the pre–COVID-19 period to post–COVID-19 period 1, there was an immediate decrease in specialty care visits (β

). Also, there was a decreased trend for all visits (β

 < .05), and an increased trend in behavioral health care (β

 < .01). From post–COVID-19 period 1 to post–COVID-19 period 2, there was an immediate increase in all visits (β

 < .01). In addition, there was an increased trend for all visits (β

 < .01) and specialty care visits (β

For office-based visits, there was a decrease in mean weekly MVs from 54,479.35 in the pre–COVID-19 period to 12,848.25 in post–COVID-19 period 1 and 10,246.86 in post–COVID-19 period 2 (

 < .01) (Table 1). In contrast, there was an increase in the mean weekly MVs from 7897.26 to 27,177.25 and 25,018.14 (

 < .01) for telephone-based visits and from 1038.52 to 9292.38 and 20,157.86 (

 < .01) for video-based visits during post–COVID-19 periods 1 and 2, respectively. Mean weekly video-based visits were also significantly higher in post–COVID-19 period 2 compared with post–COVID-19 period 1 (

From the pre–COVID-19 period to post–COVID-19 period 1, there was an immediate decrease in office-based visits (β

 < .01) and an immediate increase in telephone-based visits (β

 < .01), along with a decreased trend for office-based visits (β

 < .05) and an increased trend for video-based visits (β

). From post–COVID-19 period 1 to post–COVID-19 period 2, there was an immediate increase in video-based visits (β

 < .01), an increased trend in office-based visits (β

 < .01), and a decreased trend in video-based visits (β

Patient Visit Experience: Scheduling Needs Met

Overall, the mean proportion of respondents rating how well their scheduling needs were met as “excellent” significantly increased from 57.83% in the pre–COVID-19 period to 64.45% in post–COVID-19 period 1 and 67.71% in post–COVID-19 period 2 (

 < .01) (Table 1). When stratified by visit mode, similar significant patterns were observed for office-based visits. Video-based visits consistently had the highest proportion of respondents rating “excellent” as to how well their scheduling needs were met, with 72.14%, 71.57%, and 71.26% for the pre–COVID-19 period, post–COVID-19 period 1, and post–COVID-19 period 2, respectively.

From the pre–COVID-19 period to post–COVID-19 period 1, there was an immediate increase in the proportion of patients reporting “excellent” as to how well their scheduling needs were met (β

 < .01) for all visits along with an increased trend (β

Coronavirus disease 2019 (COVID-19) was associated with immediate weekly visit trend decreases for overall, primary care, and specialty care with long-term recovery trends; transformation to virtual visits; and increasing long-term trends for meeting patient scheduling and visit needs.

February 2021

Achieving Better Value in Pediatric Care: School Systems as Clinical and Financial Partners?

Why Even the Threat of a Twindemic Isn’t Swaying the Most Vulnerable: Black Women

Patient Experience After Modifying Visit Delivery During the COVID-19 Pandemic