Potential fraud in state Medicaid programs is getting renewed attention, with a recent emphasis on home care, also known as personal care or in-home supportive services. Home care helps with self-care activities such as bathing, dressing, and eating for older adults and people with disabilities. KFF estimates that over 5 million people use Medicaid home care, which allows individuals to receive long-term care without moving into an institution. The Trump administration has recently pointed to Medicaid home care as a source of fraud. Medicaid home care is susceptible to fraud because services are provided in people’s homes to vulnerable individuals who may be less able to advocate for themselves, including some with Alzheimer’s and other dementias. However, there are also additional safeguards against fraud in Medicaid home care compared to other types of Medicaid services. This issue brief describes how Medicaid home care operates, including who is eligible, the various systems in place to promote program integrity in its delivery, and challenges using data newly released by the Centers for Medicare and Medicaid Services (CMS). Key takeaways include the following.

• All states provide optional home care services to people whose needs are sufficient to warrant institutionalization. An institutional level of care is generally beyond what family members are capable of providing.
• Recognizing the higher risk of fraud in Medicaid home care, federal and state governments have implemented additional tools to identify and detect home care fraud. States, along with the federal government, use provider credentialing and enrollment and data analytics to help prevent fraud. There has been new attention on fraud in Minnesota’s Medicaid program recently, but the fraud, and the state’s work to root it out, date back at least 18 months. 
• On February 14, 2026, CMS released a dataset with provider-level spending data that the agency suggests could be used to identify unusual billing patterns for specific services, states, or providers, but the limited data could result in mistaken conclusions. Home care is a major emphasis of the new dataset, which stems from the fact that second to hospital spending, long-term care is the second-largest source of Medicaid spending. Although Medicaid long-term care was historically provided primarily in nursing facilities, most enrollees now recieve home care.

Why does Medicaid cover home care and who is eligible for services?

All states provide optional home care services. Under Medicaid, states are required to cover long-term care provided in nursing facilities, but not home care, which has been referred to as the “institutional bias” in Medicaid. States may only provide home care if they can demonstrate that providing the services would cost no more than institutional care would cost for an individual. All states choose to provide optional home care to people who would otherwise require institutionalization. The increased availability of home care reflects people’s preferences to remain in their homes. Expansions of Medicaid home care services also followed the 1999 Supreme Court ruling in Olmstead v. L.C., which declared that unjustified institutionalization of people with disabilities by a public entity (including Medicaid) is a form of discrimination and not permissible under the 1990 Americans with Disabilities Act. Even though nearly all of the benefits are optional for states to provide, the majority of people who use long-term care now do so at home.

Medicaid home care use is limited by eligibility criteria that generally make it only available to people whose needs are sufficient to warrant institutionalization. To be eligible for Medicaid home care, applicants must meet both financial and “functional” eligibility criteria. Functional eligibility for Medicaid home care, which is evaluated by assessment tools developed by states, generally requires individuals to demonstrate that they need an institutional level of care. There are no recent data available about states’ specific definitions for an institutional level of care, but it generally indicates that people would require 24-hour services and assistance with multiple activities of daily living (ADLs), which include bathing, dressing, eating, toileting, continence, and transferring between bed and other settings.

An institutional level of care is generally beyond what family members are capable of providing. People who require an institutional level of care generally have complex needs that require both skilled and unskilled services and often require services to be provided around the clock. In some cases, family caregivers may not have the medical expertise to provide services, but there are also challenges related to the physical demands of the job and having time to provide such intensive services. Helping family members to bathe, dress, and toilet themselves often requires the strength to lift them, which not all family members have. The time required to provide such intensive services also makes it difficult for family caregivers to provide this level of care and maintain employment or take care of their own health needs. KFF’s focus groups with paid and unpaid family caregivers provide detail that caregiving is physically, mentally, and emotionally challenging; and that family caregivers cannot provide an institutional level of care without supports. To help people requiring an institutional level of care remain at home, Medicaid supports family caregivers by providing supplemental paid care and with direct supports, such as respite care, training, and in some cases payments to the family caregivers to reflect the fact that caregiving makes it impossible to maintain outside employment.

What program integrity tools for Medicaid home care exist?

Recognizing the higher risk of fraud in Medicaid home care, federal and state governments have implemented additional tools to identify and detect home care fraud. In 2016, Congress passed the 21^st Century Cures Act, which requires states to implement electronic visit verification for all Medicaid personal care and home health services if a visit is made to a person in the home. State’s electronic visit verification must include six data elements: member receiving the services, caregiver providing the service, type of service, location of the service delivery, date of the service, and time the service begins and ends. Electronic visit verification was established to help promote fiscal integrity for Medicaid home care, and states had until 2023 to fully implement the requirements. The Health and Human Services Office of Inspector General (HHS OIG) has an active project underway to evaluate the availability and completeness of the electronic visit verification data and how states are using the data to promote program integrity.

An HHS OIG report finds that in fiscal year 2024, there were 298 fraud convictions “involving personal care service attendants” from the Medicaid fraud control units, which was 36% of all fraud convictions through the Medicaid fraud control unit, more than that of any other provider type. Although significant, the number of fraud convictions (total and as a percentage of all convictions) is notably lower than the average from 2015-2022 before electronic visit verification was fully in place. During the prior years, fraud convictions involving personal care service attendants averaged well over 400 each year and 43% of all convictions. The amount of money recovered from all convictions is small ($961 million in FY 2024 or $536 million on a 5-year average basis) relative to Medicaid spending.

States, along with the federal government, use provider credentialing and enrollment and data analytics to help prevent fraud. Providers must meet certain state and federal requirements to be eligible to participate in the Medicaid program. Additionally, states use data analytics to confirm that providers have not previously been convicted of committing Medicare fraud or fraud in a different state’s Medicaid program, and to identify unusual billing patterns for specific services or by specific providers. When Minnesota uncovered fraud in its Medicaid home care programs in 2024, the state undertook a series of actions to address that fraud, including targeting specific providers and specific types of services (Box 1).

What do newly released data about home care spending mean?

On February 14, 2026, CMS released a dataset with provider-level spending data that the agency suggests could be used to identify unusual billing patterns for specific services, states, or providers, but the limited data could result in mistaken conclusions. The data include seven fields, including the number of beneficiaries seen, counts of services, and the total spending for each procedure that is included in the data, but they omit significant elements important for pursuing meaningful analyses. With the data release, CMS posted figures to illustrate how the data could be used, with one of the figures displaying total spending among the top 20 procedures in the Medicaid data. The data show that personal care (the primary home care benefit) is the top procedure in terms of spending. However, the personal care “procedure” encompasses a wide range of services that may vary in complexity, difficulty, and length of visit (ranging from less than 30 minutes up to an entire day). In comparison, spending on emergency department visits is split among multiple procedure codes based on the complexity of the case and spending on psychotherapy is split based on the length of the visit (e.g., 30-minute visits and 45-minute visits are considered separate procedures). The data exclude all institutional records and all information about prescription drugs, which are significant shares of Medicaid spending, with hospital care accounting for 37% and being the single largest source of Medicaid spending.

Understanding the context for increases in home care spending is important context for interpreting spending data. Increased spending on Medicaid home care reflects state and federal policy choices to increase the availability of home care in lieu of institutional care when feasible. Analyses of federal spending on long-term care show that home care has grown from 1% of all long-term care spending in 1981 to 64% in 2023. Although the shift away from institutional care dates to the 1980s, the COVID-19 pandemic shone a new spotlight on the challenges of institutional care and illuminated the extent of unmet need for home care. In response, states expanded the availability of home care, increased payment rates for workers in home care settings, and made other efforts to help people remain at home rather than institutional settings. Between 2019 and 2023, the number of Medicaid home care users increased by over 750,000 people. In general, expansions of home care have garnered bipartisan support, with both 2024 presidential candidates expressing support for investments in family caregivers and more at-home services for people who need long-term care.

The Medicaid program finances about 4 in 10 births in the U.S. Federal law requires states to provide pregnancy-related Medicaid coverage through 60 days postpartum. After that period, some postpartum individuals may qualify for Medicaid through another pathway, but others may lose coverage, particularly in non-expansion states. To help improve maternal health and coverage stability and to help address racial disparities in maternal health, a provision in the American Rescue Plan Act of 2021 gave states a new option to extend Medicaid postpartum coverage to 12 months via a state plan amendment (SPA). This new option took effect on April 1, 2022 and was originally available for five years; however, the option was made permanent by the Consolidated Appropriations Act 2023. The Centers for Medicare and Medicaid Services (CMS) released guidance on December 7, 2021 on how states could implement this option.

States that sought to implement extended postpartum coverage prior to April 1, 2022 have done so through a section 1115 waiver or by using state funds. This page tracks state actions to implement extended Medicaid postpartum coverage, including states that have implemented a 12-month postpartum extension, states that are planning to implement a 12-month extension, states with pending legislation to seek federal approval through a SPA or 1115 waiver, and states that have proposed or received approval for a limited coverage extension.

Medicaid Postpartum Coverage Extensions: Approved and Pending State Action as of February 24, 2026

Medicaid Postpartum Coverage Extensions: Approved and Pending State Action as of February 24, 2026

Since the opioid epidemic was declared a public health emergency in 2017, it has claimed more than half a million lives. While the epidemic was initially driven by prescription opioids and heroin, it has evolved in recent years, to be dominated by illicit synthetic fentanyl—a substance significantly more potent than morphine. By 2023, most counterfeit opioid pills contained a deadly dose. As of 2022, nearly 1 in 3 adults reported in a KFF survey that they or a family member have been addicted to opioids (29%).

Leading up to and during the pandemic, opioid overdose deaths increased sharply. Deaths began to fall in mid-2023 and have continued to decline, though they remain above pre-pandemic levels. While it is not possible to identify a single driver of the decline, multiple policy actions may have contributed. These policies included efforts to expand access to treatment and overdose-reversal drugs and public awareness efforts about counterfeit opioid pills. They also included supply-side actions aimed at improving fentanyl detection at the ports and borders and limiting the flow of precursor chemicals used to manufacture illicit fentanyl abroad. These efforts coincided with indicators of shifting fentanyl supply, including DEA testing that suggested lower fentanyl potency in counterfeit pills.

Despite progress, a range of more recent federal policy actions may affect future trends, including federal budget cuts, federal staffing reductions, and cuts to federal grants that support state and local programs; reduced Medicaid and Marketplace coverage; and a shift toward a more enforcement-focused approach, including the designation of illicit fentanyl as a “Weapon of Mass Destruction.” This analysis examines opioid overdose deaths over time – including 2024 (the latest finalized data available through CDC WONDER data) – and trends across demographic groups and states. Additional data can be found on KFF’s State Health Facts.

Key Takeaways:

Overall trends: Opioid overdose deaths fell sharply from 2023 to 2024 (79,358 to 54,045), driven largely by decreases in fentanyl-involved deaths. Even after these declines, deaths remained above 2019, the year before opioid deaths increased sharply during the pandemic.

Demographic variation: In 2024, opioid death rates were the highest among those aged 26-64, AIAN people, Black people, and males. All demographic groups saw declines in opioid death rates from 2023 to 2024. However, most groups still had higher rates in 2024 than in 2019.

State variation: State rates in 2024 ranged from 3.3 per 100,000 in Nebraska to 38.6 per 100,000 in West Virginia. From 2023 to 2024, opioid death rates fell across all states, with the largest drops in Virginia (-44%), Wisconsin (-44%), and West Virginia (-46%). About half of states remained above 2019 levels, which may reflect differences in the timing of fentanyl’s spread and state policy.

What are the trends in opioid deaths?

Opioid overdose deaths fell sharply in 2024, nearing pre-pandemic levels but remaining above 2019 (Figure 1). Total drug overdose deaths dropped from 105,007 in 2023 to 79,384 in 2024 (-24%), while opioid deaths fell from 79,358 to 54,045. Provisional CDC data suggest opioid deaths have continued to decline through 2025. By 2024, opioid deaths were near but still above pre-pandemic (2019) levels, about 4,200 higher than in 2019, the year before the sharp pandemic-era rise in opioid deaths.

Fentanyl was involved in most opioid overdose deaths in 2024 (Figure 1). Declines in fentanyl-involved deaths drove the overall drop in opioid deaths. Deaths involving other opioids, including prescription opioids and heroin, also declined but to a lesser extent. Opioid overdose deaths include fatalities of unintentional, intentional (suicide or homicide), and unknown intent. Any drug overdose death involving opioids is counted as an opioid overdose death. Because more than one opioid can be involved in a single death, opioid subcategories do not sum to total opioid deaths.

How do opioid deaths vary across demographics?

In 2024, opioid death rates were highest among adults ages 26 to 64, American Indian/Alaska Native (AIAN) people, Black people, and males (Figure 2). Rates were highest among adults ages 26 to 44 and 45 to 64 (29.1 and 24.9 per 100,000, respectively), well above other age groups. By race and ethnicity, AIAN people had the highest opioid death rate (35.5 per 100,000), and Black people had somewhat higher rates than White people (22.8 vs. 17.5 per 100,000), a reversal from earlier in the opioid epidemic when rates were higher among White people. Because White people make up a much larger share of the population, the number of deaths was highest among White people (33,105), followed by Black people (10,202) and AIAN people (845), even though rates were higher among Black and AIAN people. Opioid death rates among males were more than double those of females.

From 2023 to 2024, opioid overdose death rates declined across all demographic groups (Figure 3). Young adults (ages 18 to 25) saw the largest decline (-42%), while adults ages 65+, saw the smallest (-20%). Declines by race and ethnicity ranged from -28% among AIAN people to -39% among Black people, and rates fell for both males and females. 

Most demographic groups continued to have higher opioid death rates in 2024 than in 2019 (Figure 3). Rates remained especially elevated among AIAN people (+101%) and adults ages 65+ (+63%). Slower declines among older adults may reflect that SUD can be harder to detect and treat and that few treatment programs are tailored to older adults. Two groups had lower rates in 2024 than in 2019: White people (-9%) and young adults ages 18-25 (-30%). Declines among White people relative to other race and ethnicity groups may partly reflect better access to opioid use disorder treatment.

How do opioid deaths vary across states?

Opioid overdose death rates varied across states in 2024, ranging from 3.3 per 100,000 in Nebraska to 38.6 in West Virginia (Figure 4). Nebraska, South Dakota, and Iowa had the lowest opioid death rates, (3.3, 5.4, and 5.8 deaths per 100,000 people, respectively). Rates were the highest in the District of Columbia (34.1 per 100,000), Alaska (37.0), and West Virginia (38.6).

All states had declines in opioid overdose death rates from 2023 to 2024, but the size of the decline varied widely. Rates fell by 5% in South Dakota and 8% in Alaska, compared with larger drops of 44% in Wisconsin and Virginia and 46% in West Virginia (Figure 5). KFF State Health Facts provides opioid overdose rates by state for 1999-2024.

By 2024, about half of states continued to have opioid overdose rates above 2019 levels. Several states were close to pre-pandemic levels, including D.C. and Utah (+1%, +2%, respectively). Over one-third of states (39%) had rates that dipped below 2019 levels, with the largest declines in NJ (-42%), Ohio and Massachusetts (each -36%). Alaska and Oregon had the largest increases relative to 2019 (+239% and +226%) (Figure 5).

State differences in opioid-related policy and the timing of fentanyl spread may help explain variation in opioid overdose death trends. Differences in how states use state opioid response grants and settlement funds, and how states structure Medicaid coverage for substance use services, including whether they adopt federal opportunities to expand treatment access may affect outcomes. Fentanyl spread unevenly across states over time, generally moving from east to west. As a result, some states experienced earlier increases, often before the pandemic, while others saw their steepest growth later. These policy and timing differences can affect how states’ 2024 rates compare with 2019. 

From 2014 to 2024, over half a million lives (516,790) were lost to suicide, with 2022 marking the highest annual total on record. Since then, overall suicides have declined somewhat, but trends diverged by method: firearm suicides continued to rise, reaching a new high in 2024. As a result, firearms accounted for 57% of all suicides in 2024, up from 50% in 2014, while suicides by other methods fell. Some of the shift may also reflect undercounting if some suicides are recorded as unintentional drug overdose deaths. These shifts may have implications for prevention strategies including the capacity and design of crisis and treatment systems.

In July 2022, the 988 Suicide and Crisis Lifeline launched nationwide, replacing the prior 10-digit number with an easier to remember, three-digit option that connects people in distress to counselors at 200+ local crisis call centers and, when needed, other crisis services. Since launch through October 2025, 988 has received more than 19 million calls, texts, or chats nationally, alongside improved answer rates and shorter wait times.

The combination of 988, other benefit expansions and distance from the pandemic may be factors contributing to small declines in overall in suicides since the 2022 peak; however access to mental health and substance use disorder treatment gaps persist. In 2025, the Trump administration discontinued the LGBTQI+ 988 call line and advanced an array of federal policy actions that could limit access to care including projected coverage loss in Medicaid and the Marketplace.

Key takeaways from analysis of CDC WONDER data from 2014 to 2024, which represents the most recent and comprehensive data available include the following:

Overall death rate: The age-adjusted suicide death rate in 2024 was 13.7 per 100,000 people.

Overall trends: Suicide deaths fell slightly from their peak of 49,476 deaths in 2022 to 48,824 deaths in 2024, but trends by suicide method diverged: suicides by other means declined while firearm suicides reached their highest level and accounted for 57% of all suicides (up from 50% in 2014).

Demographic variation: In 2024, suicide death rates were highest among AIAN people and males (22.5 and 22.3 per 100,000, respectively). Over the past decade, rates increased the most for Black people, while rates were stable or declined somewhat for adults ages 45 to 64 and females.

State variation: State suicide death rates in 2024 ranged from 5.7 per 100,000 in Washington D.C. to 29.7 in Alaska. About four in ten states had stable or lower rates than in 2014, while rates increased in the remaining states, ranging from a 27% decrease in Washington D.C. to a 35% increase in Wyoming. Rates tended to be higher in many Western states, while lower rates were more common in parts of the Northeast and a few coastal states.

What are the trends in suicide deaths?

Firearm suicides reached their highest level in 2024, while suicides of other means decreased (Figure 1). Total suicide deaths peaked in 2022 and fell slightly by 2024 (about 600 fewer deaths). Even as overall suicides decreased, firearm suicides rose to their highest level in 2024, about 6,000 higher than in 2014, and accounted for 57% of all suicides in 2024 (up from 50% a decade ago). The increase has coincided with changes in gun ownership, including a surge in new buyers during the pandemic and greater racial and ethnic diversity among gun owners. Because firearms are highly lethal, greater access can reduce opportunities for intervention. Some state policies, including extreme risk protection orders (ERPOs) and other gun laws, have been linked to declines in firearm suicides.

How do suicide deaths vary across demographics?

Overall suicide death rates were highest among AIAN people and males in 2024 (Figure 2). AIAN people had the highest suicide death rate (22.5 per 100,000), higher than the rate among White people (17.2), while rates for other racial and ethnic groups were lower. Because White people make up a much larger share of the population, the total number of suicide deaths was higher among White people than AIAN people (36,560 vs. 545), even though AIAN people had a higher suicide death rate. While females are more likely to report mental illness and to attempt suicide, males had a suicide death rate about four times higher (22.3 versus 5.6 per 100,000). There are similar suicide rates across age groups in 2024 except that adolescent rates are lower (5.7 per 100,000) than other age groups.

Over the past decade, suicide death rates increased faster for younger than older adults and more for people of color than White people, while declining somewhat for adults ages 45 to 64 and females (Figure 3). From 2014 to 2024, rates increased 17% among adults ages 18 to 25 (from 13.4 to 15.7 per 100,000) and 13% among those ages 26 to 44 (from 15.9 to 18.0), while rates were mostly flat or declined among older age groups. Rates increased more among people of color than among White people, with the largest increase among Black people (up 53%, from 5.7 to 8.7), followed by Hispanic people (up 27%, from 6.3 to 8.0). In contrast, suicide rates among White people were more stable, rising 5% over the same period (from 16.4 to 17.2). Increasing suicide rates among people of color may reflect differences in diagnosis and access to mental health care, as well as stigma and discrimination. Trends may also be influenced by shifts in firearm access and potential racial and ethnic differences in 988 awareness, use, or perceived helpfulness.

How do suicide deaths vary across states?

Suicide death rates varied widely across states in 2024 (Figure 4). Rates ranged from 5.7 deaths per 100,000 people in the District of Columbia (D.C.) to 29.7 in Alaska, with a median death rate of 15.4 per 100,000. Rates tended to be higher in many Western states, while lower rates were more common in parts of the Northeast and a few coastal states. The suicide rate may vary by state due to factors such as demographics, firearm availability, mental health status, and access to mental health and crisis services.

About four in ten states had stable or lower suicide death rates than a decade ago, while suicide death rates increased in other states (Figure 5). Between 2014 and 2024, suicide death rates decreased or were relatively stable in 22 states. The largest declines were in D.C. (-27%, from 7.8 to 5.7 per 100,000), Vermont (-21%, from 18.7 to 14.7), and New Jersey (-19%, from 8.3 to 6.7). Rates increased by 15% or more in 10 states, with the largest increases in Wyoming (35% increase, from 20.6 to 27.8), Alaska (34% increase, from 22.1 to 29.7), and Iowa (33% increase, from 12.9 to 17.1).

Alcohol use disorder (AUD) is the most prevalent non-tobacco substance use disorder in the United States and over half of US adults (54%) say that someone in their family has struggled with an alcohol use disorder. Federal data show that 1 in 10 Americans (ages 12+) had an AUD in the past year and over 40% of drinkers reported binge drinking in the past month, yet only one-third of adults view alcohol addiction as a “crisis,” compared to over half who see opioids as such.

In early January 2026, the Department of Health and Human Services (HHS) released the updated 2025-2030 Dietary Guidelines for Americans (DGA). The report marks a departure from decades of guidelines that set recommendations to limit intake to specific daily caps (formerly one drink for women and two for men), instead advising people to “drink less for better overall health.” Without clear thresholds, it may be harder for individuals and clinicians to identify when clinical screening or treatment is warranted. This challenge is compounded by low public awareness of alcohol’s health risks. For example, fewer than 40% of US adults are aware that alcohol is a carcinogen, compared to over 90% awareness of tobacco’s link to cancer. In addition, clinical screening for AUD is inconsistent and treatment rates for AUDs are low. For those receiving or seeking treatment, a range of recent federal policy actions may affect future treatment access, including substantial coverage losses in Medicaid and Marketplace coverage.

This analysis largely focuses on the narrowest definition of alcohol deaths known as “alcohol-induced deaths” (referred to as “alcohol deaths” throughout the brief). These alcohol deaths are caused by conditions directly attributable to alcohol consumption, such as alcohol-associated liver diseases. Broader definitions of alcohol deaths extend this definition to also encompass cases where an alcohol-induced condition was a contributing factor, but not the underlying cause of death. Key takeaways from this analysis of CDC WONDER data from 2014 to 2024 include the following:

Key takeaways:

Overall trends: Alcohol deaths increased gradually before the pandemic, jumped in 2020 and 2021, and have fallen somewhat since then. Even after these declines, deaths remained above 2019, the year before the pandemic.

Demographic variation: In 2024, alcohol deaths were highest among adults ages 45 to 64, American Indian and Alaska Native (AIAN) people, and males. By 2024, alcohol death rates remained above 2019 levels for several groups; the groups most above their 2019 rates were adults ages 26 to 44, 65+, White people, and females.

State variation: State rates in 2024 ranged from 6.1 per 100,000 in New Jersey to 35.9 per 100,000 in New Mexico. In 2024, alcohol death rates remained higher than 2019 rates for most states. Changes ranged from declines in New Jersey (-9%) and West Virginia (-6%) to an 80% increase in Mississippi.

What are the trends in alcohol deaths?

Alcohol deaths are down from their peak but still above pre-pandemic levels (Figure 1). Alcohol deaths increased gradually before the pandemic, jumped in early pandemic years, and have declined somewhat since then. From 2014 to 2024, the year-over-year rise in alcohol deaths averaged about 5% per year, with the largest single-year jump from 2019 to 2020 (+26%). Deaths peaked in 2021 (54,258) and have fallen since. Even with declines after 2021, in 2024 alcohol deaths were still about 50% higher than a decade ago and about 20% higher than in 2019, the year before the pandemic.

How do alcohol death rates vary across demographics?

Alcohol deaths in 2024 were highest among adults ages 45 to 64, American Indian and Alaska Native (AIAN) people, and males (Figure 2). By age, alcohol death rates peaked among adults ages 45 to 64 (28.9 per 100,000) and were next highest among adults 65 and older (21.5 per 100,000). AIAN people had the highest alcohol death rate across all demographic groups (57.9 per 100,000), more than four times the rate among White people, the racial group with the next highest rate. Because White people make up a much larger share of the population, the total number of deaths was higher among White people (32,849 vs. 1,424), even though AIAN people had the higher death rate. Alcohol death rates among males (17.3 per 100,000) were more than double those among females.

Alcohol death rates remained above pre-pandemic (2019) levels in 2024, especially for adults ages 26-44, 65+, White people, and females (Figure 3). Rates increased gradually before the pandemic, rose sharply in 2020 and 2021, and then declined from their peak, but not enough to return to 2019 levels. Adults ages 26 to 44 had the largest continued increases, with 2024 rates 38% higher than in 2019. Adults ages 65 and older also remained elevated, with 2024 rates about 22% higher than in 2019. Rates among White people remained about 20% higher than pre-pandemic levels and higher than rates among other race and ethnicity groups. Female alcohol death rates were still about 20% above 2019 levels, while male rates were closer to pre-pandemic levels.

How do alcohol death rates vary across states?

Alcohol death rates varied widely across states in 2024 (Figure 4). Alcohol-induced death rates ranged from 6.1 per 100,000 in New Jersey to a 35.9 per 100,000 in New Mexico. Higher rates tended to be concentrated in the West, particularly in the Mountain West (and Alaska), while lower rates were more common across parts of the South and the Northeast. Many factors may contribute to the differences in alcohol mortality rates across states, some of which may include differences in alcohol consumption, cultural attitudes, state-specific alcohol policies, and treatment rates.

In 2024, alcohol death rates remained higher than pre-pandemic (2019) levels in most states, though the magnitude varied widely (Figure 5). Changes from 2019 to 2024 ranged from declines in New Jersey (-9%) and West Virginia (-6%) to large increases in Mississippi (+80%) and South Dakota (+63%). Leading up to the pandemic, most states had modest steady increases, though growth was faster in some more rural states, including Wyoming, Montana, Arkansas, and North Dakota. Consistent with national patterns described above, many states experienced their sharpest increases in 2020 and 2021 and then declined somewhat from their peak, but in most states, 2024 rates remained above 2019 levels. South Dakota and Mississippi saw especially large early-pandemic increases and only modest declines afterwards, leaving them with the largest percent increases in alcohol death rates from 2019 to 2024.

What factors may contribute to alcohol deaths?

Alcohol is linked to far more deaths when broader definitions are used (Figure 6). Many alcohol-related conditions develop over time, and alcohol can also worsen other health problems or contribute to injuries, which complicates how alcohol deaths are counted. This analysis uses a narrow definition, counting only deaths where alcohol is listed as the underlying cause on the death certificate, such as alcohol-related liver disease. When deaths are also counted where alcohol is listed as a contributing cause on the death certificate (“alcohol-related deaths”), the total number of deaths nearly doubles, reaching 93,118 in 2024 (Figure 5). Under this broader definition, alcohol-related deaths exceeded opioid overdose deaths (55,535 opioid deaths when underlying and contributing cause are included), and opioid deaths change little when moving from the narrow to broader definition (54,045 under the narrow definition). Others methods can produce even higher estimates of alcohol deaths by accounting for deaths that alcohol increases the risk of, even when alcohol is unlikely to be recorded on the death certificate, such as certain cancers.

Alcohol treatment rates are low, reflecting a mix of provider, patient, and financial barriers. In 2022, only 7.6% of people ages 12 and older with a past-year alcohol use disorder (AUD) received any treatment, and fewer (2.1%) received evidence-based AUD medication. Providers may lack confidence or knowledge in treating AUD and prescribing AUD medication, which can reduce treatment initiation or referrals. On the patient side, limited understanding of what constitutes problematic drinking and attitudes towards seeking treatment can hinder recognition of need for help. Among adults who meet the criteria for SUD—which may include symptoms like increased tolerance, repeated attempts to quit or control use, or social problems related to use–95% did not seek treatment and didn’t think they needed it. Even when people want care, practical constraints such as coverage limits, treatment availability, paid leave, and out-of-pocket costs can affect decisions about whether they start or stay in treatment.

The Centers for Medicare & Medicaid Services (CMS) released the latest Medicare Advantage enrollment data on February 13, 2026. These data provide the first look into Medicare Advantage enrollment for 2026 following a statement published by CMS last fall that Medicare Advantage insurers projected total enrollment would be lower in 2026 than in 2025. The industry projections came at the same time insurers announced a drop in in the total number of Medicare Advantage plans that would be available for general enrollment (individual plans) in 2026, along with an increase the number of special needs plans (SNPs), which limit enrollment to beneficiaries with specialized health needs or who are eligible for both Medicare and Medicaid.

Overall, the data show that total Medicare Advantage enrollment continued to increase, although at a slower rate of growth than in prior years. The increase in 2026 was largely driven by increased enrollment in SNPs. Decisions made by insurers to expand SNP offerings have translated into enrollment growth in that segment. Enrollment in individual plans increased, but more slowly than in any year between 2007 and 2025. Changes in enrollment varied across the private insurers that sponsor Medicare Advantage plans, with some plans experiencing more rapid growth, while others saw a drop in enrollment.

These patterns suggest that the Medicare Advantage market remains an attractive choice for Medicare beneficiaries. In 2026, the average Medicare beneficiary can choose from among 32 Medicare Advantage plans with prescription drug coverage, most of which have no premium (other than the standard Part B premium) and the vast majority of which offer dental, vision, and hearing benefits, in addition to reduced cost sharing compared to traditional Medicare without a supplement.

Medicare Advantage enrollment reaches 35 million, increasing by 1.1 million since February 2025.

Just over 35 million people are enrolled in Medicare Advantage as of February 1, 2026 (Figure 1). That reflects an increase of 1.1 million people since February 2025, which translates into 3% growth year-over-year. Enrollment in Medicare Advantage has increased steadily over the last two decades, rising from 8 million people (19% of eligible beneficiaries) in 2007 to 34 million people (54% of eligible beneficiaries) in 2025, but the pace of enrollment growth has recently slowed. In 2025, enrollment increased 4%, which was a slower rate of growth than any year between 2007 and 2024 when the increase in enrollment averaged 9% a year.

Medicare Advantage is the private plan alternative to traditional Medicare and provides coverage of Medicare Part A and Part B benefits. In most cases, Medicare Advantage plans also offer reduced cost sharing compared to traditional Medicare without supplemental insurance, coverage of non-Medicare services, such as vision, dental and hearing, and Part D benefits, usually for no additional premium (other than the Part B premium).

Special needs plans comprised 83% of the increase in enrollment over the last year.

In February 2026, more than 8 million people are enrolled in a SNP, an increase of nearly 900,000 enrollees since February 2025 (Figure 2), comprising 83% of total Medicare Advantage enrollment growth over the last year. The increase in enrollment in individual plans was much smaller, rising by 224,000 people compared to a year ago. Enrollment in employer- and union-sponsored group plans declined slightly, falling by about 40,000 enrollees compared to February 2025; the number of beneficiaries enrolled in group MA-PDs declined by about 1.2 million, which was mostly offset by a 1.1 million increase in enrollment in employer MA-only plans.

The share of Medicare Advantage enrollees in SNPs increased from 21% in 2025 to 23% in 2026. Enrollment growth in SNPs has increased steadily since 2018 (13% of Medicare Advantage enrollment), when these plans were made a permanent part of the Medicare program. (See Appendix Table 1 for detailed data on enrollment by prescription drug coverage and plan type.)

Humana and Kaiser Permanente were the only large insurers to increase enrollment.

Across the five largest Medicare Advantage insurers, only Humana Inc. and Kaiser Foundation Health Plan, Inc. saw an increase in total Medicare Advantage enrollment, with Humana boosting its enrollment by 1.2 million enrollees, and Kaiser Permanente adding a smaller number with 64,000 additional enrollees. (Figure 3). Humana and Kaiser Foundation Health Plan saw enrollment increase across all plan types, that is individual plans, employer- and union sponsored group plans, and special needs plans. In contrast, UnitedHealth Group, Inc., the largest Medicare Advantage insurer, lost over 530,000 enrollees compared to February 2025. That change reflects a decline in enrollment in both individual (-582,000) and group (-219,000) plans that were partially offset by an increase in SNP enrollment (+267,000). CVS Health Corporation has 29,000 fewer enrollees this month than a year ago, reflecting a decline in individual plan enrollment (-81,000) that was partially offset by increases in SNP (+40,000) and group plan (+12,000) enrollment. Elevance Health Inc., which has 368,000 fewer enrollees this year than last, was the only one of the five largest insurers to see a decline in SNP enrollment (-18,000).

Enrollment in plans offered by the 150 insurers with a relatively small number of enrollees (fewer than 1 million enrollees) increased by 734,000 people in 2026. That increase reflects growth in SNPs (+388,000), individual plans (+331,000), and group plans (+15,000). Across small insurers who offered plans in both 2025 and 2026, more than three-quarters saw enrollment increase compared to a year ago.

Appendix

The Centers for Medicare and Medicaid Services (CMS) is focused on addressing “fraud, waste and abuse” in health programs including Medicaid. Efforts span across different types of health coverage and across all provider and service types. In November 2025, CMS issued a letter to states describing opportunities for federal and state governments to work collaboratively. Efforts to address fraud, waste and abuse are not new. The Center for Program Integrity (CPI), within CMS, was established in 2010 to coordinate program integrity efforts and move from a “pay and chase” model to higher reliance on data analytics to detect and prevent fraud. CPI has worked with states to provide training through the Medicaid Integrity Institute and access to broad sets of complete data to help promote program integrity efforts.

On February 14, 2026, CMS released a dataset with provider-level spending data that the agency suggests could be used to identify unusual billing patterns for specific services, states, or providers. This policy watch describes what the data include, what they exclude, and how they could potentially lead to mistaken conclusions given the limitations of the data.

What do the data include and exclude?

The new dataset includes seven types of data:

•  The national provider identifier (NPI) for the billing provider,
• The NPI of the servicing provider (which may be an individual or an organizational entity),
• The procedure code (also known as the healthcare common procedure coding system or HCPCS code),
• The month and year,
• The number of beneficiaries seen,
• The number of procedures delivered (the count of claims), and
• The total amount paid for the services.

The totals include records of outpatient services paid for by Medicaid directly (“fee-for-service”) and those paid for by Medicaid managed care organizations on behalf of enrollees between 2018 and 2024.

The data exclude all institutional records and all information about prescription drugs, which are significant shares of Medicaid spending, with hospital care accounting for 37% and being the single largest source of Medicaid spending. Beyond excluding entire categories of services, the data omit several types of information that are important in evaluating the reasonableness of service volume and spending:

• Enrollment. The amounts of services used depend on how many people are eligible to receive the services, which varies based on state policies, the economy, and people’s demographics. Differences in service use over time or between geographic locations may not be comparable without accounting for the number of Medicaid enrollees and their age and health status.
• Benefits and Coverage. The volume of services used also depends on what services states elect to offer and how they determine who is eligible to use those services, features that may change over time.
• Payment rates. Spending on services depends on how much states are paying for each service, which could depend on the local cost of living as well as state decisions about how much payment rates should be to ensure reasonable access to care.
• Diagnoses. The data do not include any information to indicate what condition the procedures are used to treat.
• Place of service and other modifiers. The data exclude information about where the services were performed (including whether they were provided in-person or remotely) and other modifiers that are used to identify characteristics of the services.

How might the data lead to mistaken conclusions?

Although data analytics can be a powerful tool to identify potentially problematic patterns, they could lead to mistaken conclusions if used in isolation. A few specific shortcomings of the new Medicaid data stand out.

• The reported procedures are not always comparable to each other. The data provide counts of patients, services, and Medicaid spending for each procedure, but the procedures are not in all cases comparable to each other. Some are very narrowly defined whereas others encompass a wide array of procedures. In CMS’ example of how the data could be used, spending on personal care appears to be a significant outlier and the single largest source of Medicaid spending. However, the “procedure” for personal care includes a range of possible services lasting from 15 minutes up to an entire day. In contrast, there are multiple separate procedures for psychotherapy that each correspond to visit length (30 minutes, 45 minutes, 60 minutes, etc.). Similarly, there are multiple procedures for emergency department visits and office visits that correspond to the complexity of the case on a scale from one to five. Personal care would not be the largest category of spending if institutional spending were also included.
• Providers are not always comparable to each other. The data provide counts of patients, services, and Medicaid spending for each provider, but some of the providers are individuals whereas others are group practices, clinics, or even entire county and state health departments. In CMS’ example of how the data could be used to look at spending by provider, 10 of the 20 largest “providers” are state or local government agencies that both administer and deliver Medicaid benefits rather than health care providers. States’ approaches to delivering Medicaid benefits vary broadly, but in most cases, state or local health departments contribute by paying providers to deliver services and by providing services directly (particularly in the case of services for people with behavioral health needs and developmental disabilities).
• There is limited information about the methods used to create the dataset or the underlying data quality. The data do not indicate how underlying Medicaid data were aggregated to create the summary file. They also do not address issues related to the quality of the underlying data, which come from the Transformed Medicaid Statistical Information System or T-MSIS. T-MSIS is a rich source of data, but sometimes there are issues with the data in specific states for specific topics. CMS maintains a rigorous “data quality atlas” that supports “insightful, methodologically sound analysis” of the data. The atlas provides a wealth of information about potential data problems. It’s unclear how CMS handled those issues when generating the recently-released data. For example, CMS reports that in the 2024 data, there were six states with unusable information about total spending for the services included in the data; and an additional 16 states where the data were of high concern. It is unknown whether CMS included the unusable data in the public file or whether the public data reflect a different version of T-MSIS.

Beyond those data issues, the data are missing context about how Medicaid spending and use of care changed between 2018 and 2024. The COVID-19 pandemic, which began in 2020, resulted in major changes to Medicaid spending that resulted from increased enrollment during the continuous enrollment period and increased awareness of unmet needs for behavioral health and long-term care. As states increased access to those services, use of those services and associated spending grew due to changes in states’ policies regarding coverage, eligibility, and provider payment rates.