The New York Times partnered with Hilltop senior data scientist Morgan Henderson, PhD, and policy analyst Morgane Mouslim, DVM ScM, on an August 22, 2021 front-page article on hospital pricing. The article highlights variations in the prices some of the largest hospitals in the country charge different insurance companies for the same procedure. What is more, the “cash price” for uninsured patients can be higher or lower than the negotiated rates charged to insurers. A companion article shows just how difficult it can be for consumers to find hospital pricing information. This is despite a January 2021 federal rule requiring hospitals to make pricing information accessible to consumers by posting it on their websites. Research conducted by Dr. Henderson and Dr. Mouslim earlier this year and published in Health Affairs found a lack of compliance by many hospitals.
Some Historical Context
More than a decade ago, a series of exposés in the Wall Street Journal featured consumers with rapidly mounting hospital debts who were hounded for years by debt collectors hired by the hospitals where they or a family member received care. In some cases, the patient was deceased. Oftentimes the debts were initially just a few hundred or thousand dollars. Some of the patients were uninsured; others had deductibles or copayments they could not afford. Still others were billed after-the-fact for services by a provider affiliated with the hospital but not in their insurer’s provider network. This is now termed “surprise billing,” and Congress is working to address it.
Presently an estimated 79 million Americans are struggling to pay their medical bills and have accumulated medical debt. The 2010 Affordable Care Act (ACA) sought to address the crushing medical debt experienced by many consumers by expanding coverage through Medicaid and the marketplaces. The ACA also required hospitals to have financial assistance policies that are widely publicized. However, 12 states have not expanded Medicaid, plans on the ACA marketplaces remain unaffordable for many, and eligibility for hospital financial assistance policies varies widely across the states and from hospital to hospital. Hilltop’s review of hospital financial assistance policies found that 21 states have their own requirements, but only 9 of these mandate income-based thresholds for patient eligibility.
The U.S. spent $1.2 trillion on hospital services in 2019. To better understand the drivers behind hospital spending, health economists have turned their attention to the prices charged for hospital services, comparing commercial health insurance rates to Medicare and examining the rates that commercial insurers negotiate with hospitals. A study examining prices paid in 2018 by private health plans to 3,112 hospitals across the U.S. found that prices for hospital inpatient and outpatient services for commercially insured patients averaged 247% of what Medicare paid. In five states, hospital prices for commercially insured patients were in excess of 325% of Medicare prices. A study of hospital prices for patients enrolled in large employer plans found that prices can also vary substantially across and within smaller geographic regions. For example, in 2018 the average price for knee and hip replacements—a common surgical procedure—was $58,193 in the New York metro area, more than double the average price of $23,170 in Baltimore, Maryland.
Maryland is an exception, however. Maryland regulates hospital prices as part of its “all payer” rate setting system that dates back to the 1970s and operates under a waiver from CMS. Each hospital must charge the same prices to all payers, with public payers receiving a small discount. While variations in within-hospital pricing are minimal, price variations across hospitals do exist since rates are set for each hospital independently.
The Road Ahead
The new federal reporting requirements became effective on January 1, 2021, and we are learning more about hospital pricing. Hospitals are required to post online for each item and service they provide the gross charge or “chargemaster;” the discounted cash price for patients who pay cash; payer-specific negotiated charges negotiated with a third-party payer; and de-identified minimum and maximum charges. The cash price applies to self-pay patients (whether uninsured or not) and is intended to be “unrelated to any charity care or bill forgiveness that a hospital may choose or be required to apply to a particular individual’s bill.” The WSJ conducted a preliminary review of hospital prices posted online for 17 services often used by uninsured patients in an emergency. Half of the cash prices were at or above the 60th percentile of the rates disclosed by the hospital and 23% were at the 100th percentile, higher than the negotiated rates. This suggests that uninsured patients are paying some of the hospitals’ highest prices.
Dr. Henderson and Dr. Mouslim will further investigate hospital pricing in the coming months. With a Hilltop Challenge award, they are collecting pricing data for “shoppable services” from the websites of hospitals around the country in order to explore the correlates of hospital pricing. They will link pricing data to hospital- and area-level characteristics to determine the extent to which factors such as the patient population, the hospital’s size, and its competitive environment are associated with variations in pricing at both the hospital and procedure level. Findings should help policymakers and ultimately consumers interpret the abundance of hospital pricing data that is now publicly available.
Cynthia H. Woodcock
|Morgan Henderson||Morgane Mouslim|
Maryland has made substantial progress in “rebalancing” Medicaid expenditures for long-term services and supports from institutional care to home and community-based services (HCBS). In FY 2019, Maryland Medicaid delivered HCBS to an estimated 19,440 adults aged 18 and older with chronic conditions and physical disabilities at a cost of $485 million. From FY 2013 to FY 2019, the number of Medicaid participants using HCBS increased by 39%, and the number of nursing home residents declined by 3%. Yet the registry of individuals seeking admission to Community Options, a §1915(c) Medicaid waiver program that offers a comprehensive array of community-based services, remains long.
Community First Choice and Community Personal Assistance Services provide personal assistance services and a limited number of other supports. These are Medicaid state plan benefits and thus available to all who meet financial and level-of-care eligibility requirements. About 10,500 Marylanders are enrolled in these two programs. Some enrollees are also on the registry for Community Options, hoping to benefit from the more extensive services available through this program.
Currently about 5,000 individuals receive HCBS through Community Options. Like many states, Maryland uses enrollment caps to limit the number of enrollees in §1915(c) waiver programs and thus maintain more control over Medicaid spending. The Community Options registry currently tops more than 19,000 individuals. Many wait years for admission to this coveted program.
To give greater priority to those on the Community Options registry who are at highest risk for nursing home admission, Hilltop developed the Pre-AI model to predict an individual’s risk of avoidable institutionalization based on demographics, clinical acuity, functional status, and health services utilization. Now, as openings become available in Community Options, the Maryland Department of Health is using Hilltop’s model to reserve 80% of openings for registrants with the highest risk scores. The remaining 20% of openings are made available on a first come, first served basis. Previously, all registrants were prioritized based on first come, first served.
In 2020, at the behest of advocates and other stakeholders, the Joint Chairmen of the Maryland General Assembly requested a report estimating the cost of making Medicaid HCBS programs available to more Marylanders. The Maryland Department of Health asked Hilltop to conduct this study, which was submitted to the Joint Chairmen on May 13, 2021, by Secretary Dennis R. Schrader.
For the study, Hilltop first reviewed the most recent literature on the costs and benefits of HCBS. We followed this with a review of Maryland’s experience with rebalancing and an analysis of the Community Options registry to better understand unmet need in the state. Finally, we estimated the costs to Medicaid of increasing enrollment in the Community Options program, taking into account possible offsets from decreased nursing home utilization. Key take-aways from our study include the following:
- The evidence in the literature continues to be mixed on whether states realize cost savings from expanding Medicaid HCBS, but HCBS participants report a higher quality of life and reduced caregiver burden.
- We estimated that only about 3,088 (16%) of the 19,804 individuals on the Community Options registry would meet both financial and level-of-care eligibility requirements for the program. Nearly half of those on the registry were already enrolled in Medicaid and, of those, one-third were receiving HCBS though other Medicaid programs.
- Nine percent of individuals on the Community Options registry had nursing home stays in FY 2019 but were not fast-tracked into the program as they could have been under Maryland’s Money Follows the Individual Act.
- The cost to Maryland of providing Community Options services to the 3,088 individuals on the registry who we predict would meet eligibility requirements would be about $31 to $39 million annually. The cost to Maryland for each additional Community Options participant would be about $10,000 to $12,500 annually. These estimates include nursing home costs avoided.
On March 31, 2021, Maryland Governor Larry Hogan submitted an amendment to the FY 2022 state budget that includes supplemental appropriations totaling $10.6 million for Community Options: $5 million from state general funds and $5.6 million from the anticipated federal Medicaid match. With these funds, Community Options will be able to serve about 400 more Marylanders. Hilltop’s study informed deliberations by policymakers and contributed to this important outcome.
This additional funding, together with needed improvements in nursing home discharge practices and registry screening and management, will enable the state to make significant progress in helping older adults and individuals with disabilities to remain in their own homes and communities.
Cynthia H. Woodcock
Many thanks to Hilltoppers Christin Diehl, Morgan Henderson, Roberto Millar, and Ian Stockwell for their contributions to this study.
The Hilltop Institute’s “Predicting Avoidable Hospital Events” model—otherwise known as the Hilltop Pre-AH Model™—originally began with a lunch. In mid-2018, Ian Stockwell, PhD—head of the analytics & research team here at Hilltop—and Howard Haft, MD—executive director of the Maryland Primary Care Program (MDPCP)—met over lunch to discuss the newly established MDPCP. This program is a key element of Maryland’s Total Cost of Care Model and is intended to help primary care providers practice advanced primary care. One of the central ideas of advanced primary care is that primary care providers can (and should) be pro-active, transformational players in a patient’s health care ecosystem, helping patients with disease prevention, chronic disease management, and even the avoidance of unnecessary hospitalizations. Dr. Stockwell, having already developed a predictive model for the Maryland Department of Health on nursing facility admissions, told Dr. Haft that it sounded like a predictive algorithm could be useful in triaging the large patient panels assigned to each primary care practice. Hilltop could develop a model to predict which patients are likely to incur an avoidable hospitalization in the near future, and providers could incorporate these risk predictions into their care processes in order to proactively care for high-risk individuals. Dr. Haft concurred.
The model-building process took about eight months. The first step, which took about three months, was a comprehensive literature review. We sought to identify, based on previously published research, which risk factors appeared to be predictive of avoidable hospital events. At Hilltop, we view this as a crucial element of predictive model development. Predictive models can, at one extreme, be “black boxes.” It is entirely possible to skip the literature review and feed raw claims into a complex neural network and allow the algorithm to find relationships between (for example) claim-level diagnosis codes and the outcome of choice (in this case, the occurrence of an avoidable hospital event). However, a black-box method of model development makes it difficult to meaningfully determine why certain patients have high (or low) risk scores. We knew that, for this model, we needed maximum transparency. We needed to make a model that identified high-risk patients in a way that allowed care providers to know why a certain patient was high-risk.
The literature review generated a list of about 200 risk factors that have been shown to be predictive of avoidable hospital events. We knew it was important for these risk factors to be operationalizable in administrative claims. The model was set to use Medicare fee-for-service claims, and even though a certain claim (an individual’s current blood glucose level, for example) might be predictive of avoidable hospitalization, this information is not available in administrative claims, so we couldn’t use this risk factor. We also knew that we would also have patient ZIP codes, so we took the time to create a library of about 40 social determinants of health, with the intention of capturing elements of an individual’s environment that may determine their risk of avoidable hospitalizations. We used various publicly available data sources to create these risk factors at the ZIP code level.
The next step was coding. This was the “needle-in-a-haystack” phase of model development. We were swimming in data—tens of millions of claims for hundreds of thousands of patients in practices participating in the MDPCP—but we needed to turn the data into the meaningful risk factors we had identified in the previous step. We used a combination of existing code, further research, and internal Hilltop expertise to turn the risk factors from words on a page to lines of code in a script. We also developed a dual-version coding system (one version in SAS, one version in Stata) to debug and cross-check. This took about three and a half months.
Next, we were ready to develop the model. We had the data and the risk factors and had to decide which model to use to best identify individuals at a high risk of incurring avoidable hospital events. There are many different models that can accomplish this, each with its particular strengths and weaknesses. We knew that we wanted something transparent, stable, and flexible. We were designing a model that could be in production for years, so we wanted to be able to add risk factors. We settled on a discrete time survival model that is operationalized as a person-month multivariable logistic regression. Once we had decided on this modeling strategy, it took a couple of weeks to generate the first internal estimates, and another month to fine-tune the modeling process.
Finally, in September 2019, we moved the model into production. Using three years of data, we trained the model so it would estimate the relationships between our risk factors and the occurrence of future avoidable hospital events. Then, we applied these relationships to the most recent data in order to generate risk scores for each individual in the MDPCP cohort. The first risk scores were released through Maryland’s health information exchange, Chesapeake Regional Information System for our Patients (CRISP), on October 11, 2019. We have been updating, re-training, re-scoring, and releasing new risk scores monthly since then. We monitor model performance each month, and we’re proud to report that the model works well: the top 10% riskiest patients experience about half of all avoidable hospital events in a given month. This means we’re helping primary care providers identify the patients that truly are at risk of future avoidable hospitalizations.
Of course, we didn’t just build the model. We also developed extensive documentation in order to be as transparent as possible with stakeholders and end-users; met many times with our partners at the MDPCP project management office (PMO), CRISP, hMetrix, and other organizations to incorporate their feedback; conducted extensive internal testing in order to make sure that our estimates are as accurate as possible; delivered many trainings and presentations on the model; developed automated processes to conduct data quality checks; and more. Since the model has been in production, we’ve added a feature to let providers know the particular reasons for an individual’s risk, we’ve added risk factors, and we’ve adapted our internal procedures to ensure model quality during the COVID-19 pandemic. Going forward, we plan to use more detailed patient address data in order to better model the social determinants of health. We’re grateful for the support from our partners at the PMO and within Hilltop, and we look forward to helping providers practice precision primary care with this and other models in the coming years.
Morgan Henderson, PhD
Senior Data Scientist
Analytics & Research Team
As a public university, UMBC is dedicated to integrating teaching, research, and service to benefit the citizens of Maryland. The Hilltop Institute—through our longstanding partnership with the Maryland Medicaid program and our mission to advance the health and wellbeing of people and communities through research and analysis—exemplifies UMBC’s commitment to public service.
Hilltop’s partnership with the Maryland Medicaid program is built on a foundation of mutual trust, a spirit of collaboration, and a shared commitment to increasing access to affordable and high-quality health care for all Marylanders. Since 1994, Hilltop has performed data-driven research and policy analysis for the Medicaid program to inform policymaking and support program development, operations, and evaluation. Today we are actively engaged in furthering health care financing and delivery system reform in Maryland and supporting implementation of the state’s Total Cost of Care Model.
A number of other partnerships between state Medicaid programs and public universities exist across the country. While similar to Hilltop, each is unique in its own way. What these partnerships have in common is a shared vision for a more equitable and robust publicly funded health care system. Hilltop has been privileged to work with state agency and university representatives from partnerships in 24 other states through our work as a founding member of AcademyHealth’s State-University Partnership Learning Network (SUPLN). Our annual meetings, collaboration calls, and frequent communications have engendered collegial relationships that have resulted in timely joint research investigations aimed at addressing salient state health policy issues such as the opioid crisis and the response to the COVID-19 pandemic. SUPLN’s Medicaid Outcomes Distributed Research Network (MODRN) is enabling us to produce powerful cross-state Medicaid data analyses that were not previously possible.
In working with other SUPLN colleagues, we have come to better understand the attributes that distinguish a successful state-university partnership and the benefits that can accrue to both the state and the university. To capture these learnings, we have prepared briefs that articulate the value proposition for both the state agency and the public university. In partnership with the Millbank Memorial Fund and AcademyHealth, we are pleased to share the value propositions. Our hope is that these briefs will encourage additional states to pursue formation of partnerships. SUPLN stands by, ready to assist. The briefs can also provide guidance to existing partnerships on strengthening their roles and relationships and solidifying support within their states. Partnerships such as these have never been more important as we as a nation endeavor to ensure access to affordable health care and ensure all Americans good health and prosperity.
Cynthia H. Woodcock
The Hilltop Institute at UMBC
Chair, Steering Committee
AcademyHealth State-University Partnership Learning Network (SUPLN)