As the U.S. coronavirus outbreak spreads beyond densely populated metropolitan areas, a new KFF analysis finds that rural areas typically have fewer intensive care hospital resources than their urban counterparts, and populations at greater risk of developing serious illness and complications from COVID-19.

While metro and non-metro areas have similar numbers of hospital beds per capita (23.5 vs 23.8 beds per 10,000 people), non-metro areas have fewer intensive care (ICU) beds – about 1.7 per 10,000 people, compared to 2.8 in urban areas. When adjusted for age, non-metro areas have only 1.6 ICU beds per 10,000 age-adjusted population, compared to 2.9 ICU beds in metro areas. Since older adults with COVID-19 are both more likely to require hospitalization and more likely to require intensive care while hospitalized, outbreaks in rural communities could strain the already limited capacity of their health systems.

The analysis includes interactive maps that enable users to explore estimates of hospital bed capacity by area, and to adjust those estimates based on the age of the population.

On average, residents of rural areas tend to be older and sicker than people who live in urban areas. 20% of people living in non-metro areas are age 65 or older, and 26% of residents under age 65 have health conditions that put them at a higher risk of developing serious COVID-19 complications. Many may already face challenges accessing health care services due to factors like provider shortages and longer travel times to reach hospitals.

The analysis is part of the Peterson-KFF Health System Tracker, an online information hub dedicated to monitoring and assessing the performance of the U.S. health system. For more data, analysis, polling and journalism on the COVID-19 pandemic, visit our special resource page on kff.org.

The coronavirus outbreak has hit densely populated urban areas of the United States first and hardest. Some health systems have experienced surges of patients, raising concerns that there are not enough hospital beds, staffing, and equipment. The novel coronavirus was slower to spread to rural areas in the U.S., but that appears to be changing, with new outbreaks becoming evident in less densely populated parts of the country.

A new issue brief looks at urban-rural differences in pandemic preparedness, and finds that non-metro typically have fewer intensive care hospital resources than their urban counterparts, and populations at greater risk of developing serious illness and complications from COVID-19.

The analysis is available in full on the Peterson-KFF Health System Tracker, an online information hub dedicated to monitoring and assessing the performance of the U.S. health system. For more data, analysis, polling and journalism on the COVID-19 pandemic, visit our special resource page on kff.org.

With President Trump and some Republican governors pushing to restart the nation’s economy, most of the public (80%) supports strict shelter-in-place policies to limit the coronavirus’ spread compared to just one in five (19%) who say these measures pose unnecessary burdens and cause more harm than good, the latest KFF Health Tracking Poll finds.

Majorities of Republicans (61%), independents (84%) and Democrats (94%) support the shelter-in-place orders, though four in ten Republicans (38%) say such orders do more harm than good. Even in states with Republican governors, more Republicans say the strict measures are worth it (63%) than say they do more harm than good (36%).

Republicans are now more optimistic than Democrats and independents that the crisis is getting better. A slight majority of Republicans (53%) now believe “the worst is behind us,” while most Democrats (64%) and independents (56%) believe the “worst is yet to come.”

About half the public overall (51%) now says the worst is yet to come, down from 74% three weeks earlier. Republicans’ growing optimism is a major reason for this shift.

Most adults say they have not left their home at all during the past week to either visit close friends or family (70%), go to work (67%), or exercise (57%). About one in five (20%) say they haven’t left their home even to shop for food, medicine, or other essential household items. One-third (34%) of all adults say they have been deemed an “essential worker,” meaning they are still required to work outside their home.

Three in four Americans (76%) say they bought or made a protective mask to wear in public, consistent with many social distancing guidelines issued by states. Those living in counties where more than 25 people have died from COVID-19 are more likely than those living in counties with 5 or fewer deaths to report using a mask (82% vs. 69%).

Eight in 10 say they could continue to shelter-in-place for at least another month, including a third (34%) who say they could do so for at least six months. Fewer Republicans say they can do this for at least a month (68%) or at least six months (25%), while more say they can do it for less than a month (26%) or not at all (4%).

Public Split on Willingness to Use Smart Phone Apps to Trace Potential Contacts with Infected People

Public health officials say that tracing and monitoring the contacts of infected people is a key element of any plan to contain the COVID-19 pandemic once people return to work, school and everyday life, and some have discussed using people’s smart phones to aid that effort.

The poll finds the public is divided on whether they would download an app on their phone to notify them when they come into close contact with an infected person (50% would, 47% wouldn’t) or to share their close contacts with public health authorities to aid tracking efforts (45% would, 53% wouldn’t).

There are strong partisan differences on these questions. About a third of Republicans (35%) would be willing to download an app to alert them if they come into contact with someone who is infected, and three in ten (29%) would be willing download an app to help public health officials track the spread of the outbreak. Much larger shares of most Democrats say they are willing to download an app for both of those purposes (63% and 58% respectively).

Younger adults generally are more willing to use apps for contract tracing than are older adults. For example, more than half (53%) of 18-29 year olds and just 36% of those ages 65 and over are willing to download an app to share contact information with public health officials.

“The mixed receptivity to using voluntary apps for contact tracing means that they can be an important tool to combat the pandemic but will not be a substitute for old-fashioned contact tracing”, KFF President and CEO Drew Altman said.

The poll finds that arguments that connect the use of such contact-tracing apps to people’s ability to return to work and allow businesses to reopen can increase people’s willingness to use them. Two-thirds (66%) say they would be willing to use them after hearing such arguments. In contrast, an argument about the risk of such information being hacked lowers the share willing to use such apps to 28%.

People are about twice as likely to say they would be willing to download an app for these purposes if it is managed by their state health department (63%), the federal Centers for Disease Control and Prevention (62%), or their local health department (62%) than if it was managed by a private tech company (31%).

While majorities of Democrats and independents and about half of Republicans are willing to download a contract tracing app if the data was managed by the CDC or their state or local health department, fewer than four in ten Democrats (37%), one-third of independents, and one-fourth of Republicans (26%) say they are willing to download such an app if the data was managed by a private tech company.

Other findings include:

• About four in 10 Americans (39%) say they personally know someone who has “tested positive for coronavirus” (24%) or who thinks they had or have coronavirus, but couldn’t get tested (29%). About one in 10 (9%) say they personally know someone who died as a result.
• Seven in 10 adults (72%) say they would be likely to use a coronavirus testing kit that they could use at their home and then send to a lab to find out if they have the virus. This includes majorities across age groups and partisan identification.
• Most Americans give themselves and the people in their household either an “A” (53% and 52%, respectively) or “B” (37% and 31%, respectively) for following local social distancing guidelines over the past two weeks. A smaller majority give their neighbors an “A” (35%) or “B” (35%).

Designed and analyzed by public opinion researchers at KFF, the poll was conducted April 15-20, 2020 among a nationally representative random digit dial telephone sample of 1,202 adults. Interviews were conducted in English and Spanish by landline (261) and cell phone (941). The margin of sampling error is plus or minus 3 percentage points for the full sample. For results based on subgroups, the margin of sampling error may be higher.

KFF will release additional findings from this poll related to the crisis’ impact on Americans’ mental health and finances in the coming days.

NOTE: More recent data are available in the “Demographics” section of the “COVID-19: Confirmed Cases & Deaths by State” dashboard found here.

In a recent analysis, we highlighted the higher risks COVID-19 poses for communities of color due to underlying health, social, and economic disparities. When we released that analysis, only a handful of states were reporting racial and ethnic data for confirmed coronavirus cases and deaths, but those data were already showing stark, disproportionate impacts for some groups of color. The Centers for Disease Control and Prevention (CDC) began reporting national data on confirmed coronavirus cases by race and ethnicity as of April 17, 2020. Similar to earlier state data, they suggest that the virus is having disproportionate effects, with Black people accounting for 34% of confirmed cases with known race/ethnicity compared to 13% of the total population as of April 20, 2020. However, race and ethnicity is missing or unspecified for nearly two-thirds (65%) of the CDC-reported cases, limiting the ability to interpret the data. In addition to the CDC data, a growing number of states have started reporting racial and ethnic data for cases and deaths, which provide further insight into how the virus is affecting communities across the country:

As of April 15, 2020, 33 states, including DC, were reporting data on distribution of confirmed coronavirus cases and/or deaths by race/ethnicity. Our analysis of these data finds that they continue to paint a sobering picture of how the virus is disproportionately affecting communities of color, as described and illustrated below (Figure 1). These data will continue to evolve as states update their data and additional states begin reporting data by race and ethnicity. Going forward, we will update these data on a regular basis and add them to our State Data and Policy Actions to Address Coronavirus dashboard.

In the majority of states reporting data, Black people accounted for a higher share of confirmed cases (in 20 of 31 states) and deaths (in 19 of 24 states) compared to their share of the total population. These disparities were particularly large in Wisconsin, where Black people made up a four-times higher share of confirmed cases (25% vs. 6%) and an over six-times higher share of deaths (39% vs. 6%) compared to their share of the total population. Similarly, in Kansas, Black people accounted for a three-times higher share of cases (17% vs. 6%) and an over five times higher share of deaths (33% vs. 6%) than their share of the total population. Other states where the share of deaths among Black people was at least twice as high as their share of the total population included Illinois, Michigan, Missouri, Arkansas and Indiana. Moreover, Black people accounted for over half of all deaths in DC (75%), Mississippi (66%), Louisiana (59%), Alabama (52%), and Georgia (51%).

 We also observed disparate impacts for Hispanic and Asian individuals in some states. In 6 of 26 states reporting data, Hispanic individuals made up a greater share of confirmed cases compared to their share of the total population, with the largest relative differences in Iowa (17% vs. 6%) and Wisconsin (12% vs. 7%). Asian people made up a higher share of cases or deaths relative to their share of the total population in a few states, although the differences generally are small. In Alabama, Asian people accounted for 4% of deaths compared to 1% of the total population. Although we identified fewer disparities for these groups compared to Black people, less states report data for these groups and states differ in how they report these data. For example, states vary in whether they include or exclude Hispanic individuals from racial categories and some report data for Asian people alone, while others combine Asian people with another racial group. Moreover, states do not provide data for subgroups of Asian people, which can mask disparities for subgroups who are at higher risk.

Data remain largely unavailable for smaller groups, including people who are American Indian or Alaska Native (AIAN) and Native Hawaiian or Other Pacific Islander (NHOPI), limiting the ability to identify impacts for them. These groups are at high risk given large pre-existing disparities in health, social, and economic factors, and there are large disparities in some of the states where data are available. For example, AIAN people make up a larger share of confirmed cases compared to their share of the total population in New Mexico (37% vs. 9%), and AIAN individuals make up five times more deaths compared to their share of the total population in Arizona (21% vs. 4%). The Indian Health Service (IHS) also reports confirmed cases among IHS patients. However, not all AIAN people are able to access services through IHS, and IHS has historically been underfunded to meet the needs of AIAN people, so these data do not provide for a complete understanding of impacts for this group.

Comprehensive nationwide data by race and ethnicity will be key to understanding how COVID-19 is affecting communities as well as shaping and targeting response efforts. While the majority of states are reporting racial and ethnic data, in many states, race and/or ethnicity is unknown for a significant share of cases and deaths. The unknown race share exceeds 20% for cases in 14 states and for deaths in 4 states. Moreover, as noted earlier, there are inconsistencies in how states report data that limit comparability across states. As such, the availability of comprehensive, consistent nationwide data disaggregated by race and ethnicity remains important for understanding the impact of COVID-19 across communities. Moreover, going forward, these data will be important to broader efforts to advance equity and address disparities that existed prior to COVID-19 and that will likely widen due to COVID-19.

Nearly half (45%) of adults across the country say that worry and stress related to the coronavirus (COVID-19) pandemic are hurting their mental health, an early sign that the health and economic crises is likely to increase mental health problems and further stretch the system’s capacity.

A new issue brief explores how the crises and related measures to protect public health, including social distancing, business and school closures, and shelter-in-place orders, are likely to affect Americans in different circumstances, including those already living with, or at risk for, mental illness or substance use disorder.

Drawing on data from the KFF Health Tracking Poll and other relevant research, the brief finds:

• People who said they were “sheltering in place” in late March were more likely to report negative effects on their mental health than those who were not. Since then, more states have imposed and extended stay-at-home orders so they affect nearly the entire country.
• Some populations, including older adults, adolescents, and parents with children younger than 18, may be particularly at risk for poor mental health consequences of social distancing policies. A broad body of research links social isolation and loneliness to both poor mental and physical health.
• Building on research showing poor mental health outcomes associated with economic stress, people who have experienced job or income loss due to the coronavirus crisis are more likely than others to say that the current crisis is harming their mental health.

As more coronavirus cases and deaths are confirmed in jails and prisons, states can utilize programs and other policy actions to mitigate the risk and spread of the disease within facilities and the broader communities in which they are located, according to a new issue brief from KFF.

More than two million individuals are incarcerated in prison or jail each year, while millions more move into and out of incarceration over the course of the year and others regularly interact with the correctional system, including staff, health care workers, vendors and visitors. Given the close quarters, the coronavirus may spread easily among people in correctional facilities and extend into nearby communities.

As of April 14, 2020, there were 694 confirmed coronavirus cases in federal prisons, including 446 incarcerated individuals and 248 staff, as well as 14 reported deaths among inmates. Reports indicated that there were many more coronavirus cases in state and local correctional systems. People in jails and prisons also are at increased risk for serious illness from coronavirus due to high rates of underlying health problems.

Correctional systems can take a range of actions to mitigate the spread of the coronavirus include reducing admissions, increasing the number of people released from jails and prisons, and reducing visits and other unnecessary contacts within facilities and for people under community supervision. Some examples:

• All 50 states have implemented some form of restriction on visitation to correctional facilities, with 15 states suspending all visitation and 37 states suspending all visitation except for legal visits.
• California is preparing to expedite the release of up to 3,500 inmates from state prisons who have less than 60 days left on their sentence and were convicted of non-violent offenses.
• Los Angeles County released approximately 1,700 inmates from local jails, or roughly 10 percent of the inmate population. All had been convicted of non-violent crimes and had less than 30 days left on their sentences.
• The Iowa Department of Corrections has expedited the release of 700 inmates who were previously determined to be eligible for parole. Other states, including Illinois, Colorado, and North Dakota, have taken steps to expedite or ease restrictions for granting parole.

States can also use Medicaid to enhance coronavirus response efforts for the justice-involved population. Medicaid facilitates access to care for eligible individuals moving into and out of justice the system, particularly in states that have adopted the ACA Medicaid expansion. However, Medicaid reimbursement for incarcerated individuals is limited to inpatient care provided in facilities subject to certain requirements. Medicaid coverage among incarcerated individuals will be increasingly important if hospital care needs grow among incarcerated individuals due to the COVID-19 crisis.

At least 40 states already facilitate access to Medicaid by suspending inmates’ Medicaid coverage rather than terminating it when they are incarcerated, allowing them to get coverage immediately upon release and facilitating access to reimbursement if they receive inpatient care while incarcerated. States also can seek waivers to expand the services and facilities for which federal Medicaid reimbursement is available for incarcerated individuals. In Illinois and California, for example, officials have asked the Centers for Medicare and Medicaid Services to allow the states to claim reimbursement for Medicaid services provided in jails and prisons for COVID-19-relatedservices.

The full analysis, COVID-19 Response for the Justice-Involved Population, provides an overview of the COVID-19-related health risks for the justice-involved population, discusses the role Medicaid can play in response efforts for justice-involved individuals, and identifies other steps states and localities can take to mitigate risk and spread of COVID-19 for this population.

For other KFF data and analyses related to coronavirus and the COVID-19 pandemic, visit kff.org.