Large Language Models (LLMs) are trained on a prodigious corpus of human writing and may reveal human preferences over characteristics of life courses, such as income, longevity, and working conditions. We present OpenAI's GPT-5.4 and a broadly representative sample of Americans with pairs of life stories and ask them to choose the life they would prefer for themselves. A person's choice is better predicted by the LLM's choice than by another person’s choice over the same stories, and LLM valuations of several life attributes are similar to those derived from human responses. Our results suggest that LLM responses offer a scalable and cost-effective complement to existing methods for studying human preferences. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

An analysis of FDA-approved drugs reveals uneven timing and concentration of revenue life cycles across modalities and therapeutic areas, offering evidence to guide innovation strategy and policy design in a post-Inflation Reduction Act world.

Regulators of new products confront a tradeoff between speeding a new product to market and collecting additional product quality information. The FDA's Breakthrough Therapy Designation (BTD) provides an opportunity to understand if a regulator can use new policy to innovate around this tradeoff—i.e., whether it improved regulator productivity by allowing products to come to market more quickly without compromising quality. We find that the BTD program shortened clinical development times by 23 percent and did not impact the ex post safety profile of drugs with the designation. In exploring mechanisms, we find that the BTD program had the greatest impact on less experienced firms and was associated with reduced BTD clinical trial design complexity. The results suggest that targeted regulatory innovation can shorten R&D periods without compromising the quality of new products.

This study compares the revenues earned by manufacturers 9 vs 13 years following US Food and Drug Administration approval for negotiation-eligible products from 2012 to 2022.

This JAMA Forum discusses some key considerations for addressing the question of whether and when prevention interventions are worth the cost.

We develop a framework to measure the role of hospital allocation in racial disparities in health care and use it to study Black and white heart attack patients. Black patients receive care at lower-performing hospitals than white patients. However, over two decades, the performance gap between hospitals treating Black and white patients shrank by over two-thirds. This progress is due to more rapid performance improvement at hospitals that tended to treat Black patients rather than reallocation of patients. Hospital improvement is correlated with adoption of a productivity-raising input, beta blockers. Our work highlights reallocation and performance improvement as future disparity-reduction levers.

Medicare Part D does not allow plans to exclude drugs in six protected classes from their formularies, which may limit plans' ability to negotiate rebates and lead to higher spending. We estimated the association between protected-class status, US-level estimated rebates, and formulary coverage during the period 2011-19. We found that protected classes indeed had significantly lower exclusion rates in Medicare Part D during this period relative to nonprotected classes, and this difference was larger than the corresponding difference in commercial plans. US-level average rebates grew 22.5 percentage points less in protected than in nonprotected classes during 2011-19, a period when formulary exclusions increased. Relative to nonprotected classes, US-level average rebates in protected classes were especially low among drugs with high Medicare market share. These results suggest that Medicare Part D protected-class policy may reduce rebates.

This JAMA Forum discusses the promise and pitfalls of regulating prices in the US health care system.

This JAMA Forum discusses treatment and health care delivery, innovative use of risk analytics, and spending and coverage priorities in the US health insurance system.

Importance:The Inflation Reduction Act (IRA) requires Medicare to negotiate prices for some high-spending drugs but exempts drugs approved solely for the treatment of a single rare disease. Objective:To estimate Medicare spending and global revenues for drugs that might have been exempt from negotiation from 2012 to 2021. Design, Setting, and Participants:This cross-sectional study analyzed drugs that met the IRA threshold for price negotiation (Medicare spending >$200 million/y) in any year from 2012 to 2021 and had an Orphan Drug Act designation. We stratified drugs into 4 mutually exclusive categories: approved for a single rare disease (sole orphan), approved for multiple rare diseases (multiorphan), initially approved for a rare disease and subsequently approved for a nonrare disease (orphan first), and initially approved for a nonrare disease and subsequently approved for a rare disease (non-orphan first). Outcomes:The primary outcomes were the number of sole orphan drugs, estimated Medicare spending on those drugs from 2012 to 2021, and global revenue since launch. Results:Among 282 drugs, 95 (34%) were approved to treat at least 1 rare disease, including 25 sole orphan drugs (26%), 20 multiorphan drugs (21%), 13 orphan first drugs (14%), and 37 non-orphan first drugs (39%). From 2012 to 2021, Medicare spending on sole orphan drugs increased from $3.4 billion to $10.0 billion. Each year, a median (IQR) of $2.5 ($1.9-$2.6) billion in Medicare spending would have been excluded from price negotiation because of the sole orphan exemption. The cumulative global revenue of the median (IQR) sole orphan drug was $11 ($6.6-$19.2) billion. Conclusions and Relevance:The sole orphan exemption will exclude billions of dollars of Medicare drug spending from price negotiation. The high level of global revenues achieved by these drugs, however, suggests that special exemption is unnecessary for them to achieve financial success. Congress could consider removing the sole orphan exemption to obtain additional savings for patients and taxpayers and to eliminate any potential disincentive for developing additional indications for these drugs.

The Inflation Reduction Act (IRA) requires Medicare to negotiate lower prices for some medicines with high Medicare spending. Using historical data from public and proprietary sources to apply the IRA’s negotiation criteria retrospectively, we identify all drugs that met the eligibility criteria from 2012 to 2021 to classify drugs that would have had a negotiated price in effect in 2022 and to calculate associated decreases in industry revenues. Our results suggest that the IRA’s reduction in overall industry revenue will be modest, will not affect most top-selling drugs and will not likely result in large-scale defunding of research and development. Changes in the net present value of drug-development projects will be concentrated in medicines where Medicare is a notable purchaser and where the ratio between expected revenue and development costs was only marginally positive before the IRA. Policymakers considering narrowing or expanding the scope of Medicare negotiation should carefully consider the tradeoffs across medicines with diverse characteristics.

What happens when patients suddenly stop their medications? We study the health consequences of drug interruptions caused by large, abrupt, and arbitrary changes in price. Medicare's prescription drug benefit as-if-randomly assigns 65-year-olds a drug budget as a function of their birth month, beyond which out-of-pocket costs suddenly increase. Those facing smaller budgets consume fewer drugs and die more: mortality increases 0.0164 percentage points per month (13.9%) for each $100 per month budget decrease (24.4%). This estimate is robust to a range of falsification checks and lies in the 97.8th percentile of 544 placebo estimates from similar populations that lack the same idiosyncratic budget policy. Several facts help make sense of this large effect. First, patients stop taking drugs that are both high value and suspected to cause life-threatening withdrawal syndromes when stopped. Second, using machine learning, we identify patients at the highest risk of drug-preventable adverse events. Contrary to the predictions of standard economic models, high-risk patients (e.g., those most likely to have a heart attack) cut back more than low-risk patients on exactly those drugs that would benefit them the most (e.g., statins). Finally, patients appear unaware of these risks. In a survey of 65-year-olds, only one-third believe that stopping their drugs for up to a month could have any serious consequences. We conclude that far from curbing waste, cost sharing is itself highly inefficient, resulting in missed opportunities to buy health at very low cost ($11,321 per life-year).

Discover the world’s best science and medicine | Nature.com

Amitabh Chandra of Harvard University reviews “We’ve Got You Covered: Rebooting American Health Care” by Liran Einav and Amy Finkelstein. The Econlit abstract of this book begins: “Considers the objective and key elements of universal health insurance, presenting a blueprint for overhauling the US health-care system to guarantee essential medical coverage for everyone.”

Hospitals play a key role in patient outcomes and spending, but efforts to improve their quality are hindered because we do not know whether hospital quality indicators are causal or biased.We evaluate the validity of commonly used quality indicators, such as mortality, readmissions, inpatient costs, and length-of-stay, using a quasi-experimental design where hospital closures reallocate large numbers of patients to hospitals of different quality.This setting allows us to measure whether patient outcomes improve as much as quality indicators predict when a relatively low-quality hospital closes, or decline as predicted when a relatively high-quality hospital closes.Using more than 20 years of Medicare claims for over 30 million patients admitted with five common diagnoses, we find that hospital quality indicators overstate differences in the causal impact of hospitals on mortality and readmission rates by 7 percent or less, but overstate differences in the causal impact of hospitals on inpatient cost and length-ofstay measures by closer to 40 percent.On average, hospital closures reduce patient mortality by shifting patients to higher quality hospitals, but the but the effect varies widely depending on the relative quality of the closing hospital.

This JAMA Forum discusses alternative ways to achieve universal coverage in the US such as administrative simplification in the Affordable Care Act plans to increase enrollment, having a basic policy that would be available to everyone, and options for supplemental coverage.

A survey of the NEJM Catalyst Insights Council finds strong support globally for government regulation of biopharmaceutical prices.

The debate around prescription drug measures in the recently passed U.S. Inflation Reduction Act (IRA), which limit some patients' out-of-pocket costs, has not fully addressed their effect on physicians and patients via their effect on payers. Reducing patients' costs for prescription drugs under Medicare Part D will improve their adherence, health, and financial security. However, payers use cost sharing to negotiate lower prices, and a payer that reduces cost sharing without increasing utilization management may not be able to keep premiums low enough to keep its business viable. Therefore, the IRA provisions will increase Part D payers' reliance on other ways to manage drug costs, such as prior authorization and/or requirements to try less expensive drugs first. These methods impose paperwork burdens on physicians and may delay or change the care received by patients. The authors discuss these features of the IRA and outline priorities for leaders who aim to further value-based care more generally.

Alzheimer's disease (AD) affects one in ten people aged 65 or older and is the most expensive disease in the United States. We describe the central economic questions raised by AD. Although there is overlap with the economics of aging and health, the defining feature of the "economics of Alzheimer's disease" is an emphasis on choice by cognitively impaired patients that affects health and financial well-being, and situations in which dynamic contracts between patients and caregivers are useful but difficult to enforce. A focus on innovation in AD prevention, treatment, and care is also critical given the enormous social cost of AD and present lack of understanding of its causes, which raises questions of optimal resource allocation and alignment of private and social incentives. The enormous scope for economists to contribute to our understanding of AD-related issues including drug development, efficient care delivery, dynamic contracting, long-term care risk, financial decision-making, and the design of public programs for AD suggests a rich research program for many areas of economics.