Health and the COVID-19 Pandemic: July 2023

NAIR- Covid 2023_KKM JER (002)_Page_01
We would like to hear your thoughts on this report and how you use it. Please provide your email address before accessing the report - we will not use it for other purposes, and providing your email address is optional.
Download full Report

Executive Summary

The COVID-19 pandemic changed how autistic people accessed services and engaged in their communities, ultimately impacting their quality of life. Access to appropriate services and accommodations help autistic individuals in maintaining employment, pursuing education, caring for their health, and establishing independence. Changes in access to services result in long-term consequences, which can be dire for autistic people. In an effort to improve policies and programs for autistic individuals, documentation of disruptions in accessing services during the COVID-19 pandemic informs better evidence-based practices for future public health emergencies.

This report examines the impact of the COVID-19 pandemic on health and healthcare among autistic children and adults. To build a comprehensive picture, we included various data sources, including health care claims and administrative records. We explored the availability of services for autistic children based on caregiver report from the National Survey of Children’s Health (NSCH). To understand hospitalization covered by both private and public health insurance, we used national emergency hospitalization records (via the National Emergency Department Sample [NEDS]) and hospital admissions data (via the National Inpatient Sample [NIS]).  Finally, we used patient medical records from Kaiser Permanente Northern California (KPNC) to look at service utilization among adult autistic patients from Northern California over the same period. These data sources cover various populations, some of which provide nationally representative pictures of autistic children and adults, others cover specific or regional populations but cover diverse populations in terms of income and race and ethnicity.

Published 7/31/2023

https://doi.org/10.17918/COVIDNAIR2023

How to cite this report

Rast, Jessica E.; Koffer Miller, Kaitlin H.; Rava, Julianna; Ventimiglia, Jonas, C.; Tao, Sha; Bromberg, Jennifer; Ames, Jennifer L; Croen, Lisa A; Kuo, Alice; and Shea, Lindsay L. National Autism Indicators Report: Health Care Experiences During the COVID-19 Pandemic. Philadelphia: Policy and Analytics Center, A.J. Drexel Autism Institute, Drexel University, July 2023.

Acknowledgements

  • Design: Chisa Merriweather
  • Editing: Paul Cirillo
  • Web design: Kyle Chvasta

A Note about Language

For the purposes of this National Autism Indicators Report (NAIR), identity first language (i.e., autistic individual or individual on the spectrum vs. individual with autism spectrum disorder [ASD]) has been purposefully used per the preference and guidance from autistic communities. It is important to recognize that there is not a universal consensus in the autistic community about the use of either person-first or identity-first language as this community is not a monolith. However, in deference to documented stances on language-use from self-advocates, identify-language is the default for this report.

Chapter 1: Service environment changes

The COVID-19 pandemic resulted in restrictions to commonly accessible medical services for the U.S. population. For autistic children and adults, a lack of access to their usual health services may have led to an increase in crisis care utilization, which is more expensive to individuals and costly to the system. The purpose of this chapter is to examine changes in health services use among autistic children and adults. Information gleaned from this chapter may better prepare the U.S. healthcare system to support autistic people during a public health emergency.

Data used in this chapter: the National Survey of Children’s Health includes information about health and health care for a nationally representative sample of children in the U.S.

Insurance and policy changes during the COVID-19 pandemic:

  • There were many changes to the health policy landscape to protect people and enable them to maintain health insurance and health care during the COVID-19 pandemic. Because there was increased job loss and unemployment, it resulted in a need to rely on public safety nets, including Medicaid and Children's Health Insurance Program (CHIP) for health insurance.
  • The Families First Coronavirus Response Act (FFCRA) required continuous coverage in Medicaid and CHIP until the U.S. government lifts the COVID-19 pandemic classification. The FFCRA resulted in more and more people enrolling, and maintaining enrollment, in Medicaid during the COVID-19, without any disruptions that would limit their ability to access necessary health care services.
  • Between April 2020 (with the enactment of the FFCRA) and December 2021, there was a 21.8% increase in Medicaid and CHIP enrollment. In March 2023, the FFCRA’s continuous coverage requirement expired; states will resume pre-pandemic eligibility and enrollment procedures starting summer 2023.
  • The proportion of the U.S. population uninsured decreased during the COVID-19 In all states, excluding North Dakota, there were lower uninsured rates in 2021 compared to 2019. Much of this can be attributed to the increased Medicaid-enrollment and stipulations of the FFCRA, in addition to other state and federal policies.

Insurance coverage in autistic children from the National Survey of Children’s Health

Nationally, more than half of autistic children have public health insurance, including Medicaid. The proportion of autistic children with public health insurance has remained relatively consistent over the last decade, and mostly supported by the federal Children’s Health Insurance Program (CHIP). From 2019 to 2021, the number of autistic children enrolled in a public insurance program increased slightly and few autistic children were uninsured in either year. In comparison to CYSHCN, public health insurance enrollment was more common among autistic children.

Figure 1.1
Figure 1.2

There were notable differences in insurance type among autistic children by race and ethnicity. In 2021, more than three-quarters of Black non-Hispanic children were covered by public health insurance and Asian non-Hispanic children were the least likely to have public health insurance (44%).

Chapter 2: direct COVID-19 service utilization

Most autistic children and adults have co-occurring health conditions that put them at increased risk of severe COVID-19 infection. The aim of this chapter is to describe emergency department and inpatient hospital visits for COVID-19 infection among autistic children and adults. Autistic children and adults, as well as people with developmental disabilities and underlying health issues, were at increased risk of hospitalizations for COVID-19.

Data used in this chapter: The National Emergency Department Sample (NEDS) captures a nationally representative sample of emergency department visits in the U.S. for children and adults, regardless of how their visits is paid for (referred to as all-payer as it captures private insurance, public insurance, and uninsured patients). The National Inpatient Sample (NIS) is an all-payer database that captures a nationally representative sample of all inpatient hospitalizations in the U.S. for children and adults.

Emergency department visits for COVID-19

The following graphs show the rate of emergency department (ED) visits in 2020 by age group. We separated this data by age because older adults were at greater risk for severe complications as a result of COVID-19 than younger people. In autistic children (ages 0-17), 3% of emergency department visits in December 2020 listed COVID-19 as the primary reason for the visit; December was the month with the most COVID-19 visits for this age group. Similarly, for autistic adults ages 18-44, the most COVID-19 ED visits were in December 2020 (6% of all ED visits were for COVID-19). Among older autistic adults (ages 45+), 11% of ED visits in December 2020 were for COVID-19, but, April 2020 saw the most COVID-19 ED visits among older autistic adults (13%) 2020. Overall, December 2020 saw the most COVID-19 ED visits for autistic children and adults. While an increase in ED visits for COVID-19 may seem surprising nine months after the start of the COVID-19 pandemic, a surge in ED visits correlates with inconsistent COVID-19 policies among states and a surge in the Delta variant across the country during that time frame.

Figure 2.1 (1)

Among children and adults ages 18-44, rates of ED visits for COVID-19 were similar for autistic versus all other people. However, ED visits for COVID-19 were much more common in April 2020 among older autistic adults than older adults broadly. The significant increase in ED visits among older autistic adults compared to other older adults in April 2020 demonstrates an area in which public health and healthcare policies need to consider strategies for maintaining continuation of usual health services for autistic individuals during public health emergencies.

What do we mean by “ED visits for COVID-19”? When a person has a visit to the ED, doctors record the main reason for the visit to the ED (like asthma, heart attack, stroke). ED visits for COVID-19 are those visits where the primary reason for was COVID-19.

Most COVID-19 hospitalizations for autistic people were covered by public insurance. Among autistic children, two-thirds of ED visits for COVID-19 in 2020 were paid for by Medicaid and about one-third were covered by private insurance. 

Among autistic adults (18-44), just under half of COVID-19 ED visits were paid for by Medicaid and just over one-quarter were paid for by Medicare. Among older autistic adults (45+), most COVID-19 ED visits were covered by Medicare and very few were covered by private insurance (5%). Autistic people often have more than one type of insurance, for example both private health insurance and Medicaid. These results only report the insurance that was billed first at the hospital, referred to as the primary payer.

Inpatient Hospitalizations for COVID-19

Like the previous findings on ED visits, we examined the rate of inpatient hospitalizations for COVID-19 in each month of 2020. Examples of inpatient hospitalization include being treated for severe illnesses (like an injury or a heart attack) or being in the hospital to deliver a baby. Inpatient hospitalization means a person was admitted to the hospital, whereas a vit to the ED may not result in a hospital admission.

For autistic children, hospitalization for COVID-19 was uncommon. In most months fewer than 1% of hospitalizations for autistic children were for COVID-19. Among autistic adults (18-44), COVID-19 hospitalizations were the most common in April (20%) and December (5%).

Among children and adults ages 18-44, rates of hospitalizations for COVID-19 were similar to the rest of the population. However, hospitalizations for COVID-19 were much more common in April 2020 for older autistic adults than other older adults.

Most COVID-19 hospitalizations in 2020 were covered by public insurance. Among autistic children, two-thirds of COVID-19 hospitalizations were covered by Medicaid. 

Figure 2.2
Figure 2.3
Figure 2.4

Among autistic adults (18-44), most hospitalizations were covered by either Medicaid or Medicare. Among older autistic adults, more than three-quarters were covered by Medicare.

Chapter conclusions and implications. Visits to the emergency department and scheduled inpatient admissions were reduced in the general population during the beginning of the COVID-19 pandemic. Staffing shortages, medical equipment shortages, and transmission reduction measures caused people to delay care, or not seek care for non-COVID-19 conditions. The findings above highlight increases in ED and inpatient hospitalizations for everyone, but especially the oldest people. Autistic older adults saw bigger decreases in hospitalizations than their same-age peers. This may be reflective of increased risk factors for severe COVID-19 that autistic people face, including higher rates of co-occurring health and mental health conditions and worse access to preventive health care than the general population. During emergency measures, policies should ensure equal access to appropriate care for all people, especially those who are the most vulnerable.

Chapter 3: Other health service utilization

Most autistic children and adults have co-occurring health conditions that put them at increased risk of severe COVID-19 infection. The aim of this chapter is to describe emergency department and inpatient hospital visits for COVID-19 infection among autistic children and adults. Autistic children and adults, as well as people with developmental disabilities and underlying health issues, were at increased risk of hospitalizations for COVID-19.

Data used in this chapter: the National Survey of Children’s Health (NSCH) includes information about health and health care for a nationally representative sample of children in the U.S. The National Emergency Department Sample (NEDS) captures a nationally representative sample of emergency department visits in the U.S. for children and adults, regardless of how their visits is paid for (referred to as all-payer as it captures private insurance, public insurance, and uninsured patients). The National Inpatient Sample (NIS) captures a nationally representative sample of all inpatient hospitalizations in the U.S. for children and adults and is also all-payer. Kaiser Permanente Northern California (KPNC) includes information on adults covered by KPNC, an insurance provider and health care system in Northern California.

Changes in health care receipt for children from the National Survey of Children's Health

The COVID-19 pandemic interrupted access to health care services. We examined changes in use of services by children from 2019-2021. From 2019 to 2021, there were small decreases in the percentage of autistic children and children and youth with special health care needs (CYSHCN) who had any health care services, preventive services, dental services, and emergency department visits in the past 12 months.

Figure 3.1

Health services. The National Survey of Children’s Health (NSCH) asks parents/guardians about their child’s health care experiences in 1) any health care visits, 2) preventive health services, 3) emergency department visits, and 4) dental health services.

Any health care visits: “During the past 12 months, did the child see a doctor, nurse, or other health care professional for sick-child care, well-child check-ups, physical exams, hospitalizations or any other kind of medical care?”

Preventive health services: “During the past 12 months, how many times did the child visit a doctor, nurse, or other health care professional to receive a preventive check-up? A preventive check-up is when the child was not sick or injured, such as an annual or sports physical, or well-child visit.”

Emergency department visits: “During the past 12 months, how many times did this child visit a hospital emergency room?”

Dental health services: “During the past 12 months, did the child see a dentist or other oral health care provider for any kind of dental or oral health care?”

From 2019 to 2021, there were small decreases in the percentage of autistic children and CYSHCN who had any health care services, preventive services, dental services, and emergency department visits in the past 12 months

There were differences in the use of health care services among autistic children by race and ethnicity in 2021. Asian autistic children were the least likely to have had any health care service or any preventive services in the past 12 months (reported in 2021) while white children were the most likely to have used services during the same period.

Of all services, the biggest differences were in receipt of health care services: 68% of Asian children had a health care visit in the past 12 months (reported in 2021) compared to 79% of Black children, 80% of children of another reported race, 83% of Hispanic children, and 92% of white children.

Hispanic and white children were also the most likely to have an ED visit. There was no notable difference in receipt of health care service by insurance status.

Access to health care services for children

Use of health care services may have decreased for many reasons like closed medical offices, restrictions on office visits for non-urgent conditions, lack of transportation options, or fewer people getting exposed to other germs and getting sick during the periods.

Findings from the NSCH also highlight changes in access to care that may have impacted how people were getting services during the pandemic. Fewer autistic children received care coordination in 2021 as compared to 2019 (37% versus 42%). There was also more reported difficulty getting needed referrals though fewer people reported needing referrals for their child (41% in 2021 versus 45% in 2019). Finally, the level of unmet health care was largely the same between 2019 and 2021 (14% in 2019 and 11% in 2021).

These patterns of change from 2019 to 2021 were largely the same between autistic children and CYSHCN, though need for referrals was more common in autistic children than in CYSHCN.

Receipt of care coordination by parents of autistic children varied by race and ethnicity in 2021. One-quarter of Black children received care coordination, compared to two-thirds of Asian children. Use of referrals was also variable by race and ethnicity. Half of Hispanic children needed a referral as compared to 29% of Asian children. Difficulty getting needed referrals was greatest in children of another reported race or those who reported multiple races.

Figure 3.2
Figure 3.4
Figure 3.3

Receipt of care coordination was more common in children with public health insurance (43% of children with any public health insurance compared to 30% of children with only private health insurance). Referrals did not vary by insurance status.

Impact of COVID-19 on health care and childcare in the NSCH

About half of autistic children and CYSHCN had a virtual health care visit in 2021. Among autistic children, virtual visits varied by race and ethnicity. Black children were the most likely to have a virtual visit (64%) and Hispanic children, the least (38%).

Just over one-third of autistic children missed (or delayed) a preventive checkup because of the COVID-19 pandemic, as did 36% of CYSHCN.

Figure 3.5
Figure 3.6

Among autistic children, Asian and Black children were less likely to miss a visit than children of another race or ethnicity. There was no notable difference by insurance status.

Finally, 42% of parents of autistic children (3-11) had a childcare arrangement closed due to the COVID-19 pandemic. 

This varied by age with half parents of younger children (3-7) had a closure, compared to 30% of parents of older children (8-11). Closed childcare also varied by race and ethnicity of autistic children. Parents of Hispanic children and white children were the least likely to experience closures of childcare arrangements as compared to parents of children of other races. Closed childcare also varied by race and ethnicity of autistic children. Parents of Hispanic children and white children were the least likely to experience closures of childcare arrangements as compared to parents of children of other races.

Figure 3.7

Changes in national emergency department usage in 2020

There was a large decrease in emergency department (ED) use in the middle of 2020. Starting in April 2020, we see decreases in ED use for autistic children (0-17), adults (18-44), and older autistic adults (45+).

The number of ED visits returned to pre-pandemic levels in the fall of 2020 among older autistic adults with visits remaining low for autistic children. We saw greater decreases in ED use for non-autistic adults as compared to autistic adults.

Figure 3.8
Figure 3.9

Changes in national inpatient hospitalizations in 2020

Like trends seen in ED usage, there were sharp decreases in April 2020 in the number of inpatient hospitalizations among autistic individuals of all ages.

For autistic children, the number of inpatient hospitalizations did not increase to pre-pandemic levels in the rest of 2020 but did for autistic adults and older adults. There were smaller decreases in inpatient hospitalizations for all non-autistic individuals as compared to autistic individuals.

Service changes in autistic adults insured by Kaiser Permanente Northern California

Finally, we focused on services received by autistic adults (ages 18+) enrolled in Kaiser Permanente Northern California (KPNC) (n=7750).

From March 2019 through February 2021, there was a decline in the number of in-person primary care visits and a corresponding increase in the number of telehealth services received. In March-April 2020, the number of primary care visits (either in-person or via telehealth) declined, but visits started to increase again beginning in May 2020.

Figure 3.10
Figure 3.11

Nearly three-quarters of autistic adults had a primary care visit in the year preceding the COVID-19 pandemic (March 2019 to February 2020).

In the following year (March 2020 to February 2021), 58% of autistic adults had a primary care visit.

Older autistic adults were more likely to access primary care than younger adults. In the year before the pandemic, 66% of autistic young adults (18-24) had a primary care visit as compared to 95% of adults (60+). 

In 2020, 52% of autistic young adults had a primary care visit as compared to 87% of older adults. While all age groups experienced a decline in primary care use between the pre-pandemic and early pandemic periods, the youngest group saw the largest decline

Figure 3.12
Figure 3.13

Changes in primary care use by autistic adults from 2019 to 2020 varied by race and ethnicity.

Asian autistic adults had the biggest decrease in primary care use from 2019 to 2022(74% versus 51%) followed by Black adults (76% vs. 58%).

Use of urgent and emergency care also decreased from 2019 to 2020.

There were large decreases in the number of ED and urgent care visits in April-May 2020 and then subsequent increases through February 2021.

From March 2019 to February 2020, 16% of autistic adults had an ED visit as compared to 11% from March 2020 to February 2021.

Figure 3.14
Figure 3.15

Chapter conclusions and implications. Efforts to reduce COVID-19 transmission and staffing shortages reduced access to health care services during the COVID-19 pandemic. People with chronic health conditions may have suffered from worsening health due to reduced access which can lead to a need for more costly emergency care. We found decreases access to service areas such as dental care, emergency department visits, and inpatient hospitalizations in 2020 and 2021 as compared to pre-pandemic years. Younger people experienced greater reductions in care, perhaps younger people often have less need for maintenance and acute care than older people. Black and Hispanic autistic children experienced larger decreases in care access as compared to white autistic children. Children with public health insurance, like CHIP and Medicaid, had better access to care, including virtual care and mental health care, than those with private insurance.

Chapter 4: Mental health service utilization

The COVID-19 pandemic had a detrimental impact on the mental health of the entire U.S. population. Prior to the pandemic, autistic people were already at increased likelihood of having co-occurring mental health conditions. Increased rates of mental health services use among autistic children and adults may have other effects on quality of life, such as an increased financial burden or decreased social participation. The purpose of this chapter is to understand the use of mental health services by autistic children and adults during the COVID-19 pandemic.

Data used in this chapter: the National Survey of Children’s Health (NSCH) includes information about health and health care for a nationally representative sample of children in the U.S. The National Emergency Department Sample (NEDS) captures a nationally representative sample of emergency department visits in the U.S. for children and adults, regardless of how their visits is paid for (referred to as all-payer as it captures private insurance, public insurance, and uninsured patients). The National Inpatient Sample (NIS) captures a nationally representative sample of all inpatient hospitalizations in the U.S. for children and adults and is also all-payer.

Mental health care in autistic children from the NSCH

In 2019 and 2021, just under half of autistic children (3-17) received mental health care. Parents of autistic children reported unmet need for mental health care; 13% in 2021 and 9% in 2019. Fewer CYSHCN received mental health care, but unmet need was similar to children with autism.

Figure 4.1
Figure 4.2

Among autistic children, there were differences in receipt of mental health care by race and ethnicity in 2021. Only about 20% of Black children received mental health care in 2021, compared to nearly 50% of children of all other races.

Asian children had the lowest reported unmet need for mental health care whereas children of another reported race had the highest. There were also differences in mental health care by health insurance status: nearly half of children with public insurance received mental health care, while one-fifth of children with private insurance received mental health care. Unmet need was also greater in children with private health insurance in 2021.

Mental health care was more common in autistic children with public insurance than those with private insurance.

More children with private health insurance reported mental health care as an unmet need in 2021.

Figure 4.3
Figure 4.4

National emergency department visits for mental health conditions

In chapter 3, we found that visits to the emergency department (ED) decreased in April and May 2020. Now we consider the reason for ED visits that did occur.

Among autistic children (0-17), visits for mental health conditions comprised 13-15% of all visits from January-March 2020, and 16-18% of all visits from April-December 2020. Among autistic adults (18-44), visits for mental health conditions comprised about one-quarter of ED visits across 2020. ED visits for mental health conditions were less common in older autistic adults than autistic individuals of other ages.

Among autistic children (0-17), there was a 5% increase in ED visits for mental health conditions from April 2019 to April 2020.

There was a smaller increase in autistic adults (18-44) (2%) and a decrease in older autistic adults (-1%) for ED visits for mental health conditions.

Figure 4.5

What do we mean by “ED visits for mental health conditions”? When a person has an ED visit, doctors record the primary reason for the visit to the ED. Visits for mental health conditions are those visits where the primary reason for was recorded as any mental health disorder or condition.

Figure 4.6

National inpatient hospital admissions for mental health conditions

In 2020, one-quarter to one-third of hospital admissions in autistic children (0-17) were for mental health conditions. This was largely consistent over the year with slightly higher admission rates in March, April, and May.

Admissions for mental health conditions were even more common in autistic adults (18-44). In May 2020, 48% of admissions were for mental health conditions. Older autistic adults had the fewest admissions for mental health conditions (ranging from 7-16%).

There were not significant increases in the percentage of admissions for mental health conditions among autistic children and autistic adults (18-44) from April 2019 to April 2020. However, in older autistic adults, there was a 10% decrease in admissions for mental health conditions during the same period.

Figure 4.7

Chapter conclusions and implications. Emerging research on the impact of the COVID-19 pandemic in the general population highlight increases in mental heath concerns among children and adults, including anxiety and depression. In this study, there were no apparent changes in mental health concerns or receipt of care among autistic children identified through 2021 (the specified period of this report). Future work will be needed to examine changes in mental health in the years following the COVID-19 pandemic. There are noted differences in mental health care access by an individual’s health insurance status.  Nearly half of children with public insurance received mental health care, while only 1/5 of children with private insurance received mental health care. Unmet need for mental health care were also greater in children with private health insurance in 2021. To accommodate mental health care needs during the COVID-19 pandemic, many states changed regulations around use of telehealth for mental health care and eased restrictions that limit clinicians from practicing across states lines. Making these changes permanent allows for increased access to appropriate mental health care and also could increase access to the most qualified professionals, as few mental health practitioners feel equipped to provide care for autistic patients. Increased accessibility of mental health care through public health insurance is of importance to examine how private insurance may improve care access.

Conclusions and recommendations for research and policy

Takeaways from our findings:

  • The COVID-19 pandemic significantly impacted the access to services and supports for autistic people, affecting their employment, education, health, and independence.
  • Autistic adults had a higher rate of visits to the emergency department (ED) and inpatient hospitalizations for COVID-19 compared to non-autistic adults.
  • Certain risk factors for severe COVID-19 infection, such as obesity, chronic lung disease, diabetes, and chronic kidney disease, are more common in autistic people, putting them at increased risk.
  • Access to care decreased during the COVID-19 pandemic, including dental care, emergency department visits, and inpatient hospitalizations.
  • Younger people experienced greater reductions in care, and Black and Hispanic autistic children experienced larger decreases compared to white children.

Research recommendations:

  • Further research is needed to examine the long-term mental health changes among autistic children in the coming years following the conclusion of the official PHE.
  • Future studies should explore the effectiveness and impact of telehealth for mental health care among autistic individuals.
  • We need to investigate disparities in mental health care access and utilization among different racial/ethnic groups and health insurance statuses.

Policy actions:

  • Healthcare providers and virtual platforms should ensure accessibility and high-quality services for all patients, including those on the autism spectrum. This can include creating physical and virtual environments that are sensory-friendly, minimizing overwhelming stimuli such as bright lights, loud noises, and crowded waiting areas. Provide options for individuals to request accommodations, such as quiet spaces or extended appointment times. Additionally, implementing communication strategies that cater to diverse communication preferences of including the varying autistic individuals. This may include using visual aids, written instructions, or simplified language to enhance understanding and ensuring any online platforms are user-friendly and accessible for individuals with diverse sensory and cognitive profiles.
  • Health care policies should continue to support telehealth and virtual care options for mental health services, including easing portability restrictions and expanding coverage. This includes the need to ease portability restrictions, allowing mental health providers to deliver care across state lines, and expanding coverage to ensure equitable access for individuals in diverse geographic areas. By supporting telehealth and virtual care, individuals on the autism spectrum and other neurodivergent individuals can benefit from increased accessibility and convenience. It enables them to receive care from the comfort of their own homes, reducing barriers such as transportation challenges or sensory overload in clinical settings. Moreover, expanding coverage for telehealth services can enhance affordability and affordability and make mental health care more accessible to individuals with limited financial resources or those residing in underserved areas.

Appendix: Methods