Physician-administered drugs and biologics account for a growing share of pharmaceutical spending, but they also generate substantial revenues for the hospitals and physician practices that acquire the products at one price and are reimbursed at a higher price. This study analyzed insurer expenditures and provider margins with a focus on Keytruda, the world's largest biologic in terms of sales, using detailed insurer, provider, and market data on 59,717 patients treated in 5,008 hospitals and physician practices. Keytruda expenditures by private insurers increased by 142 percent from 2020 to 2024, while the number of patients using Keytruda increased by 110 percent. Price markups-the ratio of the reimbursement price charged to insurers compared with the acquisition price paid to the manufacturer-averaged 173 percent in hospitals eligible for 340(B) discounts, 78 percent in hospitals not eligible for those discounts, and 16 percent in community-based physician practices. For hospitals eligible for 340(B) discounts, each Keytruda patient was associated with $102,680 in annual revenue, measured through insurer reimbursement, after organizational, patient, and market factors were adjusted for. Each Keytruda patient was associated with $67,825 in revenue at noneligible hospitals and $3,094 in revenue at physician practices.
Whether US pharmaceutical markets reward novel therapies more than incremental follow-on drugs remains unclear. We examined trends in prescription drug revenues across three dimensions of pharmacological novelty-molecular structure, therapeutic target, and delivery properties-using a retrospective analysis of branded small-molecule drugs approved by the Food and Drug Administration and marketed between 2000 and 2019. Novelty measures derived from ChEMBL, a comprehensive database of bioactive molecules with druglike properties, were linked to nationally representative utilization and spending data from the Medical Expenditure Panel Survey, with rebate-adjusted net revenues estimated using SSR Health data. Beginning around 2013, gross and net revenues rose disproportionately for highly novel drugs across all three dimensions of novelty, whereas revenues for medium- and low-novelty drugs remained relatively stable. From 2013 to 2019, mean gross revenue more than tripled for drugs with high molecular or target novelty and more than doubled for those with high delivery novelty. These revenue gains were associated with increased prescription volume, rather than higher prices. The timing of this divergence coincided with the expansion of pharmacy benefit managers' formulary exclusion policies, which may have reduced the usage of drugs with close substitutes and increased market returns to pharmacological novelty.
The opioid crisis has remained a public health challenge in the US for more than two decades. After increasing from 2002 to 2012, opioid-related hospitalizations decreased from 2016 to 2019, but more recent trends are unknown. We sought to determine trends in incidence and outcomes of US opioid-related hospitalizations in recent years. We conducted a serial cross-sectional study of adult primary and secondary opioid-related hospitalizations, using the 2016-23 National Inpatient Survey and regression models, annual percent change, and z-tests to examine change during the study period. Measures included annual opioid-related hospitalizations, hospital stay outcomes, and sociodemographic characteristics. Among the 1.48 million opioid-related hospitalizations during 2016-23, survey-weighted to 7.42 million, 0.93 million were classified as primary and 6.49 million as secondary opioid-related hospitalizations. Hospitalizations decreased through 2023 for all groups except patients who were Hispanic or Native American, who reported Medicaid as their payer, or who were older than age sixty-five. The greatest decline occurred among people younger than age thirty-five. The proportion of hospitalizations ending in death and self-directed discharge increased by 0.58 percentage points and 2.68 percentage points, respectively. Overall, differential reductions in opioid-related hospitalizations and increasing mortality and self-directed discharge highlight opportunities for interventions.
Vertical integration between Medicare Advantage (MA) plans and hospital providers is increasingly common, but little is known about how vertically integrated plans pay affiliated providers for medical services. Providers may accept lower prices from affiliated plans to give those plans a competitive edge in the MA market. Conversely, plans may pay vertically integrated providers higher prices to increase their medical loss ratios-a measure of revenues spent directly on health care required by the Centers for Medicare and Medicaid Services. Using a novel data set of vertically integrated MA plans matched with negotiated hospital pricing data, we found that in 2024, 66-73 percent of hospitals charged similar prices to affiliated and unaffiliated plans, but prices differed at a sizeable minority of hospitals. On average, affiliated plan prices were 5.3 percent higher than unaffiliated plan prices for the same procedure at the same hospital. Affiliated plan prices were higher relative to unaffiliated plan prices in the inpatient (rather than outpatient) setting, in more concentrated MA markets, and where MA hospital prices were lower relative to traditional Medicare hospital prices. Our results suggest that vertically integrated MA plans behave differently from standard MA plans, but strategies vary by market and hospital characteristics.
During the unwinding of the continuous coverage requirements related to the COVID-19 public health emergency in 2023, as an unprecedented volume of consumers were transitioning out of Medicaid, California launched a facilitated enrollment program for people who lost Medicaid and were eligible for subsidized Marketplace coverage. We conducted a survey in 2023-24 to assess coverage outcomes among Californians who lost Medicaid coverage, were assigned to a default plan, and had thirty days to opt in to coverage under the program. Although national surveys indicate high rates of uninsurance after loss of Medicaid, we found that 85 percent of those who were eligible for an Affordable Care Act Marketplace subsidy had some form of health insurance. Among those without other available coverage, nearly two-thirds opted in to a Marketplace plan. People with a high school education or less were least likely to enroll in Covered California, as were those who rated their health status as poor. Even after we accounted for premium amounts, income, education, health status, and language preference, take-up was higher among Asian/Asian American (84 percent) and White (79 percent) people than among Hispanic/Latino (53 percent) and Black/African American (54 percent) people. Our findings indicate that facilitated enrollment interventions should be strengthened to overcome administrative and other burdens that persist among subpopulations.
The rapid growth of private equity (PE) and publicly traded corporation (PTC) ownership in hospice has raised concerns that investor-driven profit incentives may undermine care quality. Evidence on how these acquisitions affect care delivery remains limited. We linked a national PE and PTC acquisition database to Medicare claims for a beneficiary sample for the period 2010-21 and used a difference-in-differences event study to compare acquired versus nonacquired for-profit hospices on process-based quality measures and Medicare reimbursement. After PE acquisition, registered nurse, social worker, and home hospice aide minutes per thirty days declined 5.14 percent, 12.32 percent, and 6.62 percent, respectively; after PTC acquisition, registered nurse and home hospice aide minutes per thirty days declined 4.63 percent and 9.09 percent. Declines in visit minutes also were observed in the last seven days of life. Reductions in visit minutes were driven by four large acquirers. These findings highlight the need for increased transparency and oversight policies, as well as payment reforms that align reimbursement with care intensity and quality.
The 2021 expansion of the Child Tax Credit provided advance monthly cash transfers to most US households with children. Although prior evaluations documented overall improvements in household well-being, less is known about variation by household structure. This study examined policy-period changes in parental mental health and material hardship among female-headed households. Using data from 1.3 million respondents to the Census Bureau's Household Pulse Survey (2020-25) and an intersectional framework, we applied difference-in-differences and multilevel models to estimate changes in parental mental health and material hardship before, during, and after expanded Child Tax Credit implementation. Non-female-headed households showed improvements in depression, anxiety, and housing insecurity, whereas female-headed households, particularly those with low incomes and Black, Hispanic, and Asian households, showed smaller or no comparable gains. Food insecurity did not improve across groups. Intersectional analyses showed that the highest burdens and smallest improvements were concentrated among low-income, female-headed households across racial and ethnic groups. Findings suggest that the expanded Child Tax Credit did not fully close baseline disparities, underscoring the importance of household structure and intersecting social positions in policy design and evaluation.
Medicare Advantage (MA) plans receive capitated payments that could lead to greater efficiency compared with traditional Medicare but may also curtail provision of beneficial services. However, MA effects are not fully understood, as most research to date is cross-sectional and could be subject to selection bias. This study investigated a natural experiment in which five states (Alabama, Arizona, Colorado, Connecticut, and New Jersey) shifted health benefits for retired state employees from supplemental plans for traditional Medicare to MA plans during the period 2017-19, leading to 87 percent of the more than 220,000 state retirees in these states shifting to MA. We found increased use of certain outpatient services (annual wellness visits and home evaluation and management visits) and observation stays, along with reduced hospital and postacute care admissions. We found no effect on days spent at home or in the community, or on mortality.
Responding to patient-initiated portal messages is a growing source of work for physicians, yet little is known about how these messages are distributed across patients and physicians. Understanding the distribution is critical for developing interventions to reduce burden and for designing reimbursement policy. Using data from 487,442 patients at University of California San Francisco (UCSF) Health and national electronic health record (EHR) metadata from 224,068 ambulatory care physicians in the US, we examined the distribution of patient-initiated medical advice request messages across patients and physicians. The distribution was highly skewed at both the patient and physician levels. The top 5 percent of UCSF Health patients accounted for 52.8 percent of all messages. Among physicians nationally, primary care physicians received a median of 9.6 messages per week, whereas the top quartile received 53.3 messages per week. Physicians with higher visit volume received fewer messages per visit, and higher message volume was associated with greater EHR work outside of work. These findings highlight the need for targeted interventions and payment models that account for the uneven distribution of asynchronous portal-based care.
The Medicare Drug Price Negotiation Program exempts or delays negotiation for drugs with orphan indications. Whether these protections are necessary to sustain orphan drug development is unclear. We compared clinical trial costs, worldwide revenues, and time to cost recovery for 167 drugs projected to exceed $200 million in annual Medicare spending, classified by orphan status: exempt under the Inflation Reduction Act (IRA); exempt under the One Big Beautiful Bill Act (OBBBA); subject to delay under the OBBBA; and not orphan or not qualifying for exemption or delay. Overall, 94 percent recovered clinical trial costs within ten years of global launch. IRA-exempt drugs were younger, had lower trial costs but comparable revenues, and recovered costs more quickly than nonqualifying drugs. OBBBA-exempt and OBBBA-delayed drugs did not differ significantly from nonqualifying drugs. Drugs benefiting from orphan protections recovered development costs at rates comparable to or faster than those of nonqualifying peers, calling into question the need to protect blockbuster orphan drugs from negotiation.
Medicaid managed care organizations serve as the primary vehicle for delivering behavioral health services to low-income youth. However, market consolidation has shifted enrollment toward a handful of large national parent firms. Understanding the implications of this shift is critical for identifying and closing gaps in adolescent mental health services. Using national Medicaid claims data from 2022, we examined the associations between managed care organization ownership by five major parent firms and measures of youth mental health service use and provider composition. Managed care organizations owned by large national parent firms were characterized by lower mental health screening rates and, for most firms, higher use of emergency department, inpatient, and medication-based care, including psychotropic prescribing without psychotherapy. Provider composition also varied by firm, suggesting differences in how these organizations structure their behavioral health networks, with some relying more heavily on psychiatrists and psychologists and others depending more on master's-level therapists. The growing influence of large national parent firms in shaping behavioral health access for Medicaid-enrolled youth underscores the need for closer scrutiny of how managed care structures serve youth with significant mental health needs.
Medicare Advantage (MA) has become a major source of dental coverage for older US adults. However, little is known about enrollment in plans with dental coverage or about the scope and continued availability of coverage over time. Using Centers for Medicare and Medicaid Services Plan Benefit Package and enrollment data, we examined county-level variation in dental plan enrollment, generosity of benefits, and continuity of dental plan availability across conventional MA plans and MA Special Needs Plans (SNPs) during the period 2010-25. Enrollment in conventional MA plans with any dental benefits rose from 50.3 percent in 2010 to 97.1 percent in 2025. Enrollment in SNPs with dental benefits increased from 68.6 percent to 93.5 percent during this period. Among plans continuously available for at least five years, benefit generosity rose after 2019, but plan continuity declined after 2021 as a result of plan exits and the withdrawal of dental benefits. Net losses of insurers offering dental benefits in conventional MA plans were more likely in counties with relatively high percentages of Black and Hispanic residents, in dental Health Professional Shortage Areas, and in areas with midlevel social deprivation. These findings underscore the need for strong oversight of MA dental benefits and strategies to achieve sustained and equitable access to dental coverage.
Multiple states have established benchmarks for health care spending growth. Since 2021, spending growth has exceeded most states' benchmarks, prompting concerns about unsustainable growth. However, these benchmarks largely do not adjust when economywide inflation changes. I collected data on states' benchmark-setting processes, targets, and reported health care spending and identified a set of six states that reported data on per capita spending growth in both 2022 and 2023. Meeting benchmark spending targets for these years would have required real (inflation-adjusted) per capita health care spending to decline by an average of 1.6 percent per year. The same nominal spending benchmark targets would have allowed real spending growth of 1.9 percent per year if inflation had stayed at its historical average. Actual real spending growth was only 0.7 percent per year. Consistent with this, health care as a share of gross domestic product for these states remained stable, at 10.9 percent in 2021 and 10.7 percent in 2023. These findings suggest that nominal spending benchmark designs can generate misleading performance signals, which can be reduced by adopting inflation-shock adjustment protocols and routinely reporting both nominal and real spending performance.
By 2034, national health spending is projected to total nearly $9.0 trillion and to represent 20.6 percent of the economy, compared with $5.3 trillion and 18.0 percent in 2024. The rate of national health spending growth during this period is influenced by continued elevated use of medical services and goods through 2026; major legislative changes that affect insurance coverage and spending through 2028; and continued demographic shifts toward public programs, mainly Medicare. The insured share of the population is expected to be 90.5 percent in 2034, compared with 91.8 percent in 2024.
The hospital wage index standardizes Medicare hospital payments for labor cost differences, paying otherwise equivalent hospitals more when they operate in areas with higher labor costs than in areas with lower labor costs. However, because of a plethora of exceptions, labor costs are commonly disconnected from the originally assigned wage index, and policy makers have expressed concerns that exceptions are not justified. Using publicly available wage index and Centers for Medicare and Medicaid Services impact files, we found that wage index exceptions increased by nearly 60 percent from 2016 to 2024 and were highly prevalent, with more than 70 percent of hospitals receiving exceptions by 2024 (compared with 46 percent in 2016). Growth was disproportionate across states and hospital types. Two exceptions-geographic reclassifications and rural floor adjustments-increased annual hospital revenues by an average of $650,000 and $930,000, respectively. Growth in costly exceptions distorts wage index accuracy and impedes the policy's intended goal of calibrating payments to actual labor costs.
Medicare beneficiaries frequently visit the emergency department (ED) at the end of life, but little is known about the epidemiology of patients admitted to hospice from the ED. We used 100 percent Medicare fee-for-service claims from the period 2018-20 to describe the frequency of direct ED-to-hospice enrollments, associated patient and hospice agency characteristics, and patient outcomes. In this sample, 4.3 percent of initial enrollments in hospice originated from the ED. ED-to-hospice admissions featured short lengths-of-stay (21.7 percent were two days or less) and high rates of general inpatient level of care at the time of enrollment (23.6 percent). The 10 percent of hospice agencies with the highest proportion of ED-to-hospice enrollments were less often for-profit than agencies ranked below the fiftieth percentile in respect to proportion of ED-to-hospice enrollments. Further research is needed to increase understanding of how much patients benefit from ED-to-hospice transfers when their hospice stays before death are very short, and what drivers lead to these ED-to-hospice transfers.
Medicaid covers Medicare premiums and out-of-pocket expenses for low-income older adults, who make up a growing segment of the US population. Using administrative data, we characterized the change in the number of beneficiaries dually enrolled in Medicare and Medicaid and corresponding changes in Medicaid spending on Medicare Part A and Part B premiums during the period 2013-24. The number of dually enrolled beneficiaries increased by 27 percent, while Medicaid spending on Medicare premiums nearly doubled, primarily as a result of growing Part B premium spending. We estimated that at least half of the growth in Part B premium spending was attributed to the increasing per beneficiary Part B premium cost over time, with a smaller portion attributed to rising enrollment. Dual enrollment per capita increased in all but nine states. State Medicaid programs face a greater burden from rising Part B premium costs, yet they have little opportunity to control this spending without federal action.
