Health Systems
The Potential Impact of Artificial Intelligence on Healthcare Spending
Nikhil Sahni et al.
NBER Working Paper, January 2023
Abstract:
The potential of artificial intelligence (AI) to simplify existing healthcare processes and create new, more efficient ones is a major topic of discussion in the industry. Yet healthcare lags other industries in AI adoption. In this paper, we estimate that wider adoption of AI could lead to savings of 5 to 10 percent in US healthcare spending -- roughly $200 billion to $360 billion annually in 2019 dollars. These estimates are based on specific AI-enabled use cases that employ today's technologies, are attainable within the next five years, and would not sacrifice quality or access. These opportunities could also lead to non-financial benefits such as improved healthcare quality, increased access, better patient experience, and greater clinician satisfaction. We further present case studies and discuss how to overcome the challenges to AI deployments. We conclude with a review of recent market trends that may shift the AI adoption trajectory toward a more rapid pace.
The Effects of Competition on Physician Prescribing
Janet Currie, Anran Li & Molly Schnell
NBER Working Paper, January 2023
Abstract:
We ask how competition influences the prescribing practices of physicians. Law changes granting nurse practitioners (NPs) the ability to prescribe controlled substances without physician collaboration or oversight generate exogenous variation in competition. In response, we find that general practice physicians (GPs) significantly increase their prescribing of controlled substances such as opioids and controlled anti-anxiety medications. GPs also increase their co-prescribing of opioids and benzodiazepines, a practice that goes against prescribing guidelines. These effects are more pronounced in areas with more NPs per GP at baseline and are concentrated in physician specialties that compete most directly with NPs. Our findings are consistent with a simple model of physician behavior in which competition for patients leads physicians to move toward the preferences of marginal patients. These results demonstrate that more competition will not always lead to improvements in patient care and can instead lead to excessive service provision.
Provider turf wars and Medicare payment rules
Alice Chen et al.
Journal of Public Economics, February 2023
Abstract:
Barriers to entry and scope of work restrictions are common features of US labor markets, especially within healthcare industries. However, concerns over access and costs of care have encouraged some deregulation. We empirically explore relaxed occupational rules of anesthesia care whereby physician-trained anesthesiologists and certified registered nurse anesthetists (CRNAs) perform overlapping services but often engage in joint production. After CRNAs were granted practice independence, we find only modest (3%) reductions in anesthesiologist billing for CRNA supervision and no evidence of greater use of CRNAs. Our results caution policymakers against overestimating the downstream impact resulting from removing these provider regulations.
Unsupervised Machine Learning for Explainable Health Care Fraud Detection
Shubhranshu Shekhar, Jetson Leder-Luis & Leman Akoglu
NBER Working Paper, February 2023
Abstract:
The US spends more than 4 trillion dollars per year on health care, largely conducted by private providers and reimbursed by insurers. A major concern in this system is overbilling, waste and fraud by providers, who face incentives to misreport on their claims in order to receive higher payments. In this work, we develop novel machine learning tools to identify providers that overbill insurers. Using large-scale claims data from Medicare, the US federal health insurance program for elderly adults and the disabled, we identify patterns consistent with fraud or overbilling among inpatient hospitalizations. Our proposed approach for fraud detection is fully unsupervised, not relying on any labeled training data, and is explainable to end users, providing reasoning and interpretable insights into the potentially suspicious behavior of the flagged providers. Data from the Department of Justice on providers facing anti-fraud lawsuits and case studies of suspicious providers validate our approach and findings. We also perform a post-analysis to understand hospital characteristics, those not used for detection but associate with a high suspiciousness score. Our method provides an 8-fold lift over random targeting, and can be used to guide investigations and auditing of suspicious providers for both public and private health insurance systems.
Geographical Variations In Emergency Department Visits For Mental Health Conditions For Medicaid Beneficiaries
John McConnell et al.
Health Affairs, February 2023, Pages 172-181
Abstract:
Despite Medicaid's importance as a payer and source of coverage for mental health care, relatively little is known about how prevalence, access, and quality might vary among Medicaid beneficiaries. This study used national Medicaid data from 2018 to assess regional variations in emergency department (ED) visits for mental health conditions, a measure that may reflect unmet needs for behavioral health care. We found substantial variations, with rates in the region with the highest visit rates eight times higher than those in the region with the lowest rates. Many regions with high rates of ED visits for mental health conditions also had high rates of outpatient mental health use. Regional patterns differed substantially, with some regions exhibiting high rates of ED visits related to anxiety but low rates for schizophrenia and vice versa. The presence of large variations in ED visits for mental health conditions, with substantial differences in the composition across regions, suggests a need for context-specific solutions, including assessments of the ways in which mental health benefits are structured at the state Medicaid agency level and of differences in provider accessibility and an understanding of the types of mental illness underlying high rates of use.
The Impact of Vertical Integration on Physician Behavior and Healthcare Delivery: Evidence from Gastroenterology Practices
Soroush Saghafian et al.
NBER Working Paper, February 2023
Abstract:
US healthcare is undergoing a period of substantial change, with many hospitals vertically integrating with physician practices. Such integration could improve quality by promoting care coordination, but could also worsen it by impacting care delivery. Evidence on how physicians alter their behavior from the changes in financial ownership and the incentive structures of the integrated organizations is scant. We examine Medicare patients treated by gastroenterologists, a specialty with a recent increase in vertical integration. Using a causal model and large-scale patient-level national panel data that include 2.6 million patient visits across 5,488 physicians, we examine changes in various measures of care delivery. We find that physicians significantly alter care processes (e.g., in using anesthesia with deep sedation) after they vertically integrate, and that patients' post-procedure complications increase substantially. We provide evidence that the financial incentive structure of the integrated practices is the main reason for the changes in physician behavior, since it discourages the integrated practices from allocating expensive resources to relatively unprofitable procedures. Although integration improves operational efficiency measured by physicians' throughput, it negatively affects quality and overall spending. We note some potential policy levers through which policymakers could mitigate the negative consequences of vertical integration.
The Value of Improving Insurance Quality: Evidence from Long-Run Medicaid Attrition
Ajin Lee & Boris Vabson
Harvard Working Paper, December 2022
Abstract:
The US government increasingly provides public health insurance coverage through private firms. We examine associated welfare implications for beneficiaries, using a 'revealed preference' framework based on beneficiaries' program attrition rates. Focusing on the Medicaid program in New York State, we exploit quasi-random variation in the initial assignment at birth to public versus private Medicaid based on birth weight. We find that infants assigned to private Medicaid at birth are less likely to subsequently leave Medicaid. We provide suggestive evidence that reduced attrition reflects beneficiary responses to improved program quality, rather than alternative mechanisms such as private Medicaid plans reducing re-enrollment barriers.
Nonparametric Estimates of Demand in the California Health Insurance Exchange
Pietro Tebaldi, Alexander Torgovitsky & Hanbin Yang
Econometrica, January 2023, Pages 107-146
Abstract:
We develop a new nonparametric approach for discrete choice and use it to analyze the demand for health insurance in the California Affordable Care Act marketplace. The model allows for endogenous prices and instrumental variables, while avoiding parametric functional form assumptions about the unobserved components of utility. We use the approach to estimate bounds on the effects of changing premiums or subsidies on coverage choices, consumer surplus, and government spending on subsidies. We find that a $10 decrease in monthly premium subsidies would cause a decline of between 1.8% and 6.7% in the proportion of subsidized adults with coverage. The reduction in total annual consumer surplus would be between $62 and $74 million, while the savings in yearly subsidy outlays would be between $207 and $602 million. We estimate the demand impacts of linking subsidies to age, finding that shifting subsidies from older to younger buyers would increase average consumer surplus, with potentially large impacts on enrollment. We also estimate the consumer surplus impact of removing the highly-subsidized plans in the Silver metal tier, where we find that a nonparametric model is consistent with a wide range of possibilities. We find that comparable mixed logit models tend to yield price sensitivity estimates toward the lower end of the nonparametric bounds, while producing consumer surplus impacts that can be both higher and lower than the nonparametric bounds depending on the specification of random coefficients.
Rationing Medicine Through Bureaucracy: Authorization Restrictions in Medicare
Zarek Brot-Goldberg et al.
NBER Working Paper, January 2023
Abstract:
High administrative costs in U.S. health care have provoked concern among policymakers over potential waste, but many of these costs are generated by managed care policies that trade off bureaucratic costs against reductions in moral hazard. We study this trade-off for prior authorization restriction policies in Medicare Part D, where low-income beneficiaries are randomly assigned to default plans. Beneficiaries who face restrictions on a drug reduce their use of it by 26.8%. Approximately half of marginal beneficiaries are diverted to another related drug, while the other half are diverted to no drug. These policies generated net financial savings, reducing drug spending by $96 per beneficiary-year (3.6% of drug spending), while only generating approximately $10 in paperwork costs. Revealed preference approaches suggest that the cost savings likely exceed beneficiaries' willingness to pay for foregone drugs.
The value of an additional day of post-acute care in a skilled nursing facility
Rachel Werner et al.
American Journal of Health Economics, Winter 2023, Pages 1-21
Abstract:
With ongoing efforts to improve the value of health care in the United States and reduce wasteful spending, we examine empirically the value trade-offs involved in an additional day in a skilled nursing facility (SNF) after hospital discharge. To control for potential endogeneity, we use the percentage of Medicare beneficiaries enrolled in Medicare Advantage in each county-year as an instrument for individuals' SNF length of stay among Traditional Medicare beneficiaries, as local Medicare Advantage penetration puts downward pressure on SNF length of stay for all SNF patients but does not directly affect utilization management of those enrolled in Traditional Medicare. We also test for heterogeneity in treatment effect across patients by clinical complexity and two non-health-related factors, marital status and nursing home profit status. We find that one additional day in a SNF lowers short-term readmission rates, but this effect is small and heterogeneous across patient types. The most clinically complex patients (those with the longest predicted SNF stays) benefit the most from an additional SNF day, as do patients whose stays are shorter because of non-health-related factors. The cost savings from reduced readmission rates are small and do not offset the additional SNF costs.
Dynamic Pricing Regulation and Welfare in Insurance Markets
Naoki Aizawa & Ami Ko
NBER Working Paper, February 2023
Abstract:
While the traditional role of insurers is to provide protection against idiosyncratic risks of individuals, insurers themselves face substantial uncertainties due to aggregate shocks. To prevent insurers from passing through aggregate risks to consumers, governments have increasingly adopted dynamic pricing regulations that limit insurers' ability to change premiums over time. This paper develops and estimates an equilibrium model with dynamic pricing and firm entry and uses it to evaluate the design of dynamic pricing regulations in the U.S. long-term care insurance (LTCI) market. We find that stricter dynamic pricing regulation lowers social welfare as the benefit from improved premium stability is outweighed by the cost of reduced insurer participation. The welfare loss from stricter dynamic pricing regulation could be mitigated if the government also expands public LTCI through Medicaid.
Ground Ambulance Billing And Prices Differ By Ownership Structure
Loren Adler et al.
Health Affairs, February 2023, Pages 227-236
Abstract:
The No Surprises Act prohibits most surprise billing but notably does not apply to ground ambulance services. In this study we created a novel data set that identifies the ownership structure of ground ambulance organizations to compare pricing and billing between private- and public-sector ambulances, with a specific focus on organizations owned by private equity or publicly traded companies. Overall, we found that 28 percent of commercially insured emergency ground ambulance transports during the period 2014-17 resulted in a potential surprise bill. Our analysis illustrates that being transported by a private-sector ambulance in an emergency comes with substantially higher allowed amounts, patient cost sharing, and potential surprise bills compared with being transported by a public-sector ambulance. Further, allowed amounts and cost sharing tended to be higher for private equity- or publicly traded company-owned ambulances than other private-sector ambulances. These findings highlight substantial patient liability and important differences in pricing and billing patterns between public- and private-sector ground ambulance organizations.