When it comes to rooting out wasteful spending in federal entitlement programs, attention has long focused on preventing beneficiaries from gaming the system.

A new Stanford study identifies a fresh cause for concern: the for-profit companies that the U.S. government increasingly tasks with providing benefits to Americans who are often poor, elderly or both.

In a new working paper, Maria Polyakova, an assistant professor of medicine, finds that outsourcing public assistance services to third parties can lead to unanticipated effects on prices as well as on which beneficiaries gain the most from public dollars.

That’s because companies are in the business of making money. And when they know which of their consumers are likely to get certain levels of public support, they will try to use this information to maximize their profits, according to the research published this week by the National Bureau of Economic Research.

Polyakova shows that when companies act in their self-interest, unforeseen inequities and inefficiencies can arise that may hurt some consumers while helping others. At a time when governments in the United States and around the world are increasingly turning to the private sector to provide public benefits — namely in health care and in education — Polyakova says policymakers need to better understand how these intermediaries are affecting welfare programs.

“Policymakers have to be more careful about introducing intermediaries into public services,” says Polyakova, who is a faculty fellow at the Stanford Institute for Economic Policy Research (SIEPR), and teaches at the Stanford School of Medicine. She is also a core faculty member of Stanford Health Policy. “They may want to revisit how they think about outsourcing when research is showing that there are unintended consequences that may be positive or negative.”

Health Insurance Pricing under the Microscope

Intermediaries are central to a number of public services where the U.S. government provides subsidies to consumers, often based on income, age or employment status. Prominent examples include privately-managed Medicare Advantage Plans, drug benefits under Medicare Plan D, and charter schools in secondary education.

According to Polyakova, most research into wasteful spending within government subsidies has focused on consumers and how they try to trick the system by, for example, hiding income to qualify for a tax credit or cash assistance. Governments, though imperfect, have long been seen as benign players.

The increasing involvement of for-profit companies, she says, shows there’s a need to closely examine what’s happening on the supply side of public welfare.

To do that, Polyakova found an ideal setting: the federal health insurance marketplace created by the Affordable Care Act of 2010. Most consumers who shop for coverage through www.healthcare.gov receive a subsidy in the form of a tax credit that covers all or part of their insurance premium. The amount of their tax credit is tied to their household income.

The dollars at stake are significant. The Congressional Budget Office estimates that in 2019 the federal government will pay $560 billion in subsidies for privately-provided health insurance, including the spending on the Affordable Care Act marketplaces as well as other similarly designed programs. That figure is expected to hit $1.2 trillion over the next decade.

The Neighborhood Effect

Polyakova and her co-author — Stephen Ryan of Washington University’s Olin Business School — analyzed data from 2017 covering more than 9 million enrollees across some 2,570 counties around the country. They find that the presence of an intermediary significantly impacts insurance prices and key measures economists use to calculate the effects of a policy beyond a given benefit’s face value.

Specifically, they show that health insurance companies will have an incentive to raise premiums in markets where more consumers receive the higher tax credit because their incomes are low and the government is required to subsidize them.

On the flip side, insurers will charge lower prices in places where such subsidized consumers are less willing to buy coverage if they think it costs too much.

To illustrate the unintended consequences of the insurers’ actions, the researchers point out that, in the first instance where prices increase, consumers with incomes that are slightly higher than other community members will end up paying more for the same coverage. Under the second scenario, consumers who don’t qualify for the tax credit because their incomes are too high benefit from the lower premiums aimed at nearby residents.

“The price you pay for insurance will depend on who your neighbors are,” says Polyakova. “If you live near people who are poorer than you, you will be affected differently than if you live near people who are richer than you.”

Change the subsidy, change the calculation

Like with financial aid, tax credits for insurance coverage are calculated based on consumer income. But there is another type of subsidy that policymakers could use — flat vouchers, in which all members of a market receive the same benefit regardless of income, age or some other characteristic. For their research, Polyakova and Ryan also analyze how flat vouchers that only vary by age, but not by income, would hypothetically alter private health insurance prices in the federal Affordable Care Act marketplace.

Here, too, the scholars find different impacts on different types of consumers whether the subsidy is based on income or delivered as a flat voucher.

The analyses, says Polyakova, drive home the point that policymakers need to understand that there are trade-offs to relying on for-profit companies to provide government services and that the type of subsidy offered can alter how they calculate prices in disparate ways.

“There’s nothing wrong with companies trying to maximize their profits,” says Polyakova. “But sophisticated policymakers need to understand what happens when private markets get involved.”

Global warming and more days of extreme heat are exacerbating the health risks of pregnancy, particularly among African-American women, according to new Stanford-led research.

The maternal mortality rate among all women in the United States is already the worst of any industrialized nation. And black women are three to four times more likely to die from pregnancy-related problems than white women.

“It is truly a crisis that in America, one of the wealthiest countries in the world, more women are dying from pregnancy or childbirth complications than in any other developed country,” said Maya Rossin-Slater, a core faculty member at Stanford Health Policy and a faculty fellow at the Stanford Institute for Economic Policy Research.

In a new working paper published by the National Bureau of Economic Research, Rossin-Slater and two other health economists underscore how little research is out there about the impact of rising temperatures on the health of mothers and their newborns.

Pregnant women, for example, are not able to regulate body temperature as efficiently as non-pregnant individuals due to the physiological changes they undergo during gestation. Heat exposure can alter blood flow in the placenta, which can weaken the placenta and lead to complications. And high heat can lead to other pregnancy complications, such as hypertension, preeclampsia and prolonged premature rupture of membranes.

“All of these issues can translate into women needing to be hospitalized during pregnancy and experiencing complications during childbirth,” wrote Rossin-Slater, an assistant professor of health research and policy at Stanford Medicine. Her co-authors are Jiyoon Kim, assistant professor of economics at Elon University, and Ajin Lee, an assistant professor of economics at Michigan State University.

The researchers said most of the discussion about maternal health focuses on the health-care system, but that other determinants of poor maternal health and racial disparities are much less understood, particularly when it comes to how the environment is impacting pregnancy.

So they launched what they believe is the first study to identify the causal effects of prenatal exposures to extreme temperatures on the health of the mothers themselves.

As the Earth Warms, So Does Exposure to Extreme Heat

Their paper focuses on an environmental factor that is becoming increasingly relevant due to the growing consensus that climate change is contributing to a gradual warming of the earth: exposure to extreme heat.

The researchers studied the effects of exposure to extreme temperatures during pregnancy on maternal and child hospitalizations, using inpatient discharge records from three U.S. states with different climates: Arizona, New York and Washington. Their data comes from the State Inpatient Databases from the Healthcare Cost and Utilization Project, including 2.7 million inpatient records of 2.7 million infants and 2.2 million mothers in those three states.

And to measure temperature exposure, the researchers obtained data from the National Oceanic and Atmospheric Administration (NOAA).

For every county in their data, the researchers calculated the average temperature for every month. Then for every given day in a specific month in that county, they looked at the historic average for how high or low that day’s temperature was relative to the overall temperature in that month in that county.

For example, a 90-degree day in Arizona in September would not be classified as extreme heat since it’s relatively common. But a 90-degree day in New York would be, since temperatures that high are much less common. They classified “extreme heat” as a given day when the temperature is more than three standard deviations (3SD) above that historic county mean.

Then, they compared the outcomes of women who are of the same race giving birth in the same county and calendar month, but in different years. These women are likely similar in terms of their demographics and socioeconomic status, but may be exposed to different temperatures during pregnancy. For example, consider a black woman giving birth in November 2011 in Queens County, New York, and a black woman giving birth in November 2012 in the same county. If there were a heat wave in Queens in the August 2012, then the latter woman is exposed to more extreme heat during pregnancy than the former. 

The economists found that each additional day with heat that is at least 3SDs — or substantially higher than the historic county-month average — during the second trimester of pregnancy increases the likelihood that a newborn is diagnosed with dehydration by .008 percentage points.

“Our results provide new estimates of the health costs of climate change and identify environmental drivers of the black-white maternal health gap,” they wrote. “Understanding the health consequences of this increase in extreme heat is critical information for discussions about the costs of climate change and the possible benefits of mitigating policies.”

The researchers found that each additional day of extreme heat exposure during pregnancy increases black women’s likelihood of hospitalization during pregnancy. Since black women on average are exposed to more extreme heat than white women — due to different residence patterns and access to mitigating technologies like air conditioning — extreme heat may contribute to exacerbating the already large gap in maternal health between black and white women.

Detrimental Consequences of Rising Temperatures

Scientists predict global average temperatures will continue to rise over the next 50 to 100 years as greenhouse gases continue to trap more heat in the Earth’s atmosphere. The U.N. Intergovernmental Panel on Climate Change last year warned that nations worldwide must quickly reduce fossil fuel use to keep the rise in global temperatures below 1.5°C by 2050. 

The panel also said the number of days with mean temperatures above 32°C in the average American county is forecasted to increase from about 1 to 43 days per year by 2070-2099.

That could have detrimental consequences for babies and mothers alike.

“Overall, our findings on infant health suggest that exposure to extreme heat during the second trimester increases the likelihood of the baby being dehydrated at the time of birth,” the researchers wrote. “This, in turn, appears to increase the likelihood of subsequent readmission to the hospital many months later for causes linked to dehydration.”

And these impacts are typically missed when researchers only measure infant health using more standard variables, such as birth weight.

The authors note dehydration is one of the leading causes of morbidity and mortality in children. Studies show that children under 5 years old who have an average of two episodes of gastroenteritis associated with dehydration per year leads to 2 to 3 million pediatric office visits and accounts for 10% of all pediatric hospital admissions in the United States. 

Experts believe black women are three- to four-times more likely to die from pregnancy-related causes due to lack of access to and the poor quality of health care, as well as clinicians not monitoring black women as closely — or actually dismissing their symptoms altogether.

“The fact that the adverse impacts on health during pregnancy are larger for black than for white mothers suggests that climate change may exacerbate the already large racial gap in maternal health,” the researchers said.

Stanford Medicine has announced a merger of its Division of Health Services Research from the Department of Health Research and Policy with the Center for Health Policy and the Center for Primary Care and Outcomes Research — joining together the School of Medicine’s leading researchers in the field of health policy.

The three centers have all been under the Stanford Health Policy umbrella for several years — CHP is a part of the Freeman Spogli Institute for International Studies and PCOR is a division in the Department of Medicine. The faculty will now all sit under one roof at Stanford Health Policy’s newly renovated offices in Encina Commons.

“On behalf of CHP/PCOR, I want to say how delighted we are to be coming together as one health policy unit,” said Douglas K. Owens, the director of both centers. “We very much look forward to further strengthening our ongoing collaborations in research and training, and the merger will provide great opportunities for synergy.”

Loren Baker, who worked with Kate Bundorf to lead the efforts at the Division of Health Services Research, called the melding of the faculty “a real opportunity to build on the strengths Stanford has in health policy,” and said it would position Stanford Health Policy faculty for further success in their research and education programs.

“We’ve had a long history of working together and we hope that this will lead to more and better opportunities to do that going forward,” Baker said.

Owens said he wanted to thank Loren and Kate “for their terrific leadership as we have worked out how to bring our groups together,” as well as the Dean’s Office, the Department of Medicine, and FSI for their support and guidance.

I always find it hard to believe so many people are living in poverty: some 39.7 million Americans, or 12.3% of the population. It’s such a wealthy country, yet so many are poor.

In a twist that could be interpreted as good news — it doesn’t seem fair to say there is anything positive about living in poverty — I recently learned that older, low-income Americans tend to be healthier if they live in more affluent areas of the country.

Not only are they healthier, but their physical well-being is better across the board with a lower prevalence of dozens of chronic conditions, particularly if they live in rural communities. This, despite their income having less purchasing power in those better-resourced neighborhoods.

This was the key finding in new research published by Stanford Health Policy’s Maria Polyakova in the Annals of Internal Medicine.

While recent studies have reported that low-income adults living in more affluent areas of the United States have longer life expectancies, less has been known about the relationship between the affluence of a geographic area and morbidity of the low-income population.

“I was interested in figuring out whether the same relationship holds for morbidity: Are poorer people less sick in richer areas?” Polyakova told me. “And if so, are there any specific conditions that drive these differences that could be the target for policy-making?”

So Polyakova, a faculty fellow at the Stanford Institute for Economic Policy Research, and her co-author, Lynn M. Hua at the University of Pennsylvania, set out to evaluate the association between chronic conditions among low-income, older adults and the economic affluence of a local area. 

They focused on nearly 6.4 million Medicare beneficiaries in 2015 aged 66 to 100 years old who received low-income support under Medicare Part D, a prescription drug program for Medicare enrollees. They investigated the prevalence of 48 chronic conditions among these patients, including common chronic conditions such as hypertension, depression, diabetes and Alzheimer’s disease. They found the presence of all conditions is highly correlated: places, where the poor tend to have a high prevalence of one disease, are likely to have a high prevalence of all 48 conditions.

“While we cannot ascertain a causal relationship, our results clearly point towards the importance of further understanding why the socioeconomic environment of low-income, older adults is so tightly linked to such a broad measure of health,” the researchers wrote. 

The results, they said, were broadly consistent with the extensive literature on the social determinants of health. But their work takes that literature even further.

“Our study extends this research by providing measures of the prevalence of chronic conditions among low-income, older adults for a large national sample of the U.S. population,” Polyakova said. 

The researchers used clinical, rather than self-reported measures of diagnoses and reported this group’s variation in morbidity across local areas of the country, rather than nationally. 

“Our results raise the bar for researchers who are trying to find out what factors drive health disparities in the U.S.; these factors would have to be able to explain the differences in nearly 50 condition,” Polyakova said.

The study supported by the National Institute on Aging came to three key conclusions:

1. The health of low-income, older adults in the United States varies substantially across local geographic regions, and this variation cannot be attributed to one specific disease or a narrow set of conditions. 
2. Consistent with their original hypothesis, they found that more affluent local areas of the country have a lower prevalence of chronic conditions in the low-income, older adult population.
3. The researchers found that low-income, older adults have better health in rural areas of the country.

I wondered why these poor, older adults do particularly well in rural communities, as those regions often lack easy access to high-quality health care and state-of-the-art hospitals.

“We don’t know the exact answer, but there is a general sense that differences in the social fabric and lifestyle in rural areas — could contribute to this pattern,” Polyakova told me. “It appears that better health in these areas persists, despite challenges of accessing formal care.”

On August 26, Judge Thad Balkman delivered a $572 million judgment against pharmaceutical giant Johnson & Johnson for the company’s role in fueling the opioid epidemic in Oklahoma. In the discussion that follows, Stanford Law Professors Michelle Mello and Nora Freeman Engstrom discuss the decision and how other cases tied to the national opioid crisis are developing.

The Oklahoma decision took many onlookers by surprise. How did the case unfold? And what did Judge Balkman find? On Monday, Cleveland County District Judge Thad Balkman of Oklahoma issued a judgment that capped off a long and closely-scrutinized trial wherein the Oklahoma Attorney General faced off against Johnson & Johnson (J&J), claiming that J&J contributed to the opioid epidemic that has devastated the state of Oklahoma.

Image

Stanford Law Professors Michelle Mello and Nora Freeman Engstrom

To understand the verdict, a bit of background is helpful. When Oklahoma initially sued, it cast the net broadly, asserting claims against several defendants under several causes of action.  Certain defendants (namely, Purdue and Teva) chose to settle rather than roll the dice at trial. (Purdue, the maker of OxyContin, agreed to pay Oklahoma $270 million and Teva, one of the world’s leading providers of generic drugs, $85 million; neither admitted wrongdoing.)  Further, over time, Oklahoma’s various causes of action got winnowed down to the singular claim that J&J had created a public nuisance by aggressively and deceptively marketing opioid products to Oklahoma’s doctors and patients.  This posture meant that Oklahoma’s victory at trial was far from a foregone conclusion, as public nuisance claims can be very hard to prove, particularly in cases that relate to dangerous products.

With that table set, the trial began on May 28, 2019.  In a crowded courtroom in Cleveland County, it stretched on for nearly seven weeks and featured dozens of witnesses and more than 800 exhibits. The trial was a bench trial, meaning there was no jury, but there was a written opinion explaining the judge’s decision.  Judge Balkman’s 42-page opinion offers a cogent summary of the evidence and governing law and, broadly, vindicates Oklahoma’s litigation strategy. The opinion finds that J&J engaged in a deceptive marketing campaign designed to convince Oklahoma doctors and the public that opioids were safe and effective for the long-term treatment of chronic, non-malignant pain. Further, this “false, misleading, and dangerous marketing” caused “exponentially increasing rates of addiction [and] overdose death,” which ravaged the Sooner State. The picture Judge Balkman draws is stark and, for J&J, devastating.

Are individuals suing drug companies too? Are there class action cases that are relevant?

There are some suits by individuals, but we don’t believe that’s where the big money damages—and the real social impact of the litigation—will be.  More important is the pending federal multi-district litigation (MDL), which consolidates nearly 2,000 individual federal lawsuits brought by cities, counties, municipalities, and tribal governments in a single action before Judge Dan Polster in Cleveland, Ohio. Additionally, 48 states have initiated separate litigation, with a lineup of claims and defendants similar to the MDL.

Does this win for Oklahoma mean these other plaintiffs have an easy road ahead?

Not easy, but potentially easier. The Oklahoma case is what we call a bellwether. Like the ram that leads the other sheep this way or that, the bellwether trial doesn’t control the path of future litigation. But it does go first, and it helps to indicate trends.

As a bellwether, the big verdict here is very reassuring to the many states, counties, municipalities, and tribes suing opioid makers, distributors, and retailers, and it is, correspondingly, very disturbing for those who made and sold opioids to the American public.  The verdict suggests that this litigation has legs, and that judges and juries may be willing to pin blame not just on Purdue, the maker of OxyContin, but on others who played an arguably less central role in fueling this public health crisis.

What is striking is how damning Judge Balkman’s factual conclusions about J&J’s conduct are, and how similar they are to the allegations made against other opioid manufacturers in other cases.  All the things he objected to regarding J&J’s marketing practices are things that others, too, allegedly have done. Some of them are things that multiple companies banded together to do. Plaintiffs’ attorneys should be feeling pretty confident about their chances of persuading other courts that those practices are problematic.

Is Oklahoma free to use the award as it wishes? Will the state share some of the award with the people who died or suffered in the opioid crisis (if the decision is upheld on appeal)?

The damages, in this case, are intended to fund Oklahoma’s “nuisance abatement plan.”  That’s the remedy in a public nuisance case: The defendant has to pay to clean up the mess it made. In this case, Oklahoma provided a detailed plan laying out what would be needed to abate the opioid problem in the state. The costs added up to $572 million for the first year, and that’s what the judge awarded—not the $17 billion Oklahoma sought for a multi-year abatement effort.

The plan specifies that the money will be used for opioid use disorder screening, prevention and treatment ($292 million), housing and other services for those in recovery ($32 million), continuing medical education programs ($108 million), a pain management benefit program ($103 million), treatment of neonatal abstinence syndrome ($21 million), and other services.  Individuals won’t be direct recipients of the funds, though they may receive the services funded.

Legally, what happens next?

J&J has vowed to appeal the “flawed” Oklahoma judgment, and we expect that the judgment will be appealed, first to Oklahoma’s intermediate, and then, likely, to its supreme, court.  More immediately, though, attention will turn from Oklahoma to Ohio.  The first bellwether trial in the MDL, involving claims from Ohio’s Cuyahoga and Summit counties, is scheduled to begin on October 21.

Even as they prepare for trial, however, lawyers for both plaintiffs and defendants are also, no doubt, continuing to work toward reaching a broad and encompassing settlement.  When Judge Polster was first assigned the MDL back in January 2018, he made no bones about his desire to do “something meaningful to abate this crisis”—and to do it quickly.  It hasn’t been easy to execute on that, which isn’t surprising given the unprecedented magnitude and complexity of the litigation.

Still, we expect that, sooner or later, the opioid litigation will settle.  Indeed, even as we write, news is breaking that Purdue and the Sacklers may be in the midst of a negotiation whereby Purdue would declare bankruptcy and the Sacklers would contribute a cash payment of roughly $4.5 billion-plus relinquish ownership of the company, in return for peace with plaintiffs.

But even forging a settlement involving just those two entities is tricky—and forging a broader settlement will be exponentially harder for a number of reasons.  One is that any truly global agreement needs to pass muster with a range of defendants, some of whom have comparatively shallow pockets, and all of whom sold (or made or distributed) different products, at different times, in different quantities, in different states.  And, on the other side of the table, any settlement agreement needs to get buy-in from both those plaintiffs in the MDL and also state attorneys’ general, who have their own distinct set of priorities and interests relating to their separate lawsuits.  Further, because only a small proportion of eligible cities and counties have joined the MDL to date, any global settlement needs to somehow—equitably but firmly—close the courthouse door on those potential future plaintiffs.  None of this will be easy to accomplish.  But whenever new information reduces uncertainty about how courts would resolve a legal dispute, settlement becomes more likely—and, here, the Oklahoma verdict makes a significant contribution.

 

Nora Freeman Engstrom, Professor of Law and Deane F. Johnson Faculty Scholar, is a nationally-recognized expert in tort law, legal ethics, and complex litigation. Her work explores the day-to-day operation of the tort system—particularly its interaction with alternative compensation mechanisms. Michelle Mello, Professor of Law and Professor of Health Research and Policy (School of Medicine), is a leading empirical health scholar and the author of more than 150 book chapters and articles, including “Drug Companies’ Liability for the Opioid Epidemic,” recently published in the New England Journal of Medicine.  

A task force of national health experts has released a draft recommendation to screen all adults 18 to 79 years for the hepatitis C virus (HCV), noting the opioid epidemic has fueled what has become the most common chronic bloodborne pathogen in the United States.

Cases of acute HCV have increased 3.5-fold over the last decade, particularly among young, white, injection drug users who live in rural areas. Women aged 15 to 44 have also been hit hard by the virus that is spread through contaminated blood.

The U.S. Preventive Services Task Force, which makes recommendations followed by primary care clinicians nationwide, has until now recommended that people who are at high risk be tested for hepatitis C, as well as “baby boomers” born between 1945 and 1965.

“Unfortunately, HCV now affects a broader age range than previously with three times as many new infections per year,” said Stanford Health Policy’s Douglas K. Owens, chair of the independent, voluntary panel of national experts in prevention and evidence-based medicine.

The Task Force now recommends that clinicians encourage all their adult patients, even those with no symptoms or known liver disease, get a blood test for the virus. Pregnant women should also be screened; from 2009 to 2014, the prevalence of HCV infection among women giving birth has nearly doubled.

“The explosive growth in HCV has been fueled by the opioid epidemic, with the spread of HCV into younger populations,” said Owens, director of the Center for Health Policy and the Center for Primary Care and Outcomes Research. “HCV now kills more Americans than all other reportable infectious diseases combined, including HIV.”

An estimated 4.1 million people in the United States are carrying HCV antibodies; about 2.4 million are living with the virus, according to the Task Force. The HCV infection becomes chronic in 75% to 85% of cases and some of those people develop symptoms such as chronic fatigue and depression, and liver diseases that can range from cirrhosis to liver cancer.

Approximately one-third of people ages 18 to 30 who inject drugs are infected with the virus; 70% to 90% of older injection-drug users are infected.

There currently is no vaccine for hepatitis C although research in the development of a vaccine is underway. But there are effective oral direct-acting antiviral (DAA) medications that can clear the virus from the body, particularly if caught early.

“The good news is that treatment for HCV is far better, and far better tolerated than in the past, offering a cure to most people,” Owens said. “Early identification of HCV is important to prevent long-term complications of HCV including liver failure, liver cancer, and death.”

The Task Force said in a release that there are several key research gaps that could inform the benefit of screening for HCV infection:

1. ·     Research is needed on the yield of repeat vs. one-time screening for HCV.
2. ·     Research is needed to identify labor management practices and treatment of HCV infection prior to pregnancy to reduce the risk of mother-to-child transmission.
3. ·     Trials and cohort studies that measure effects on quality of life, function, and extrahepatic effects of HCV infection (such as renal function, cardiovascular effects or diabetes) would be helpful for evaluating the impact of DAA regimens on short-term health outcomes.
4. ·     Additional studies are needed to examine the epidemiology of HCV infection and the effectiveness of DAA regimens in adolescents.

Surprise, out-of-network charges billed to privately insured Americans who received inpatient care at in-network hospitals more than doubled between 2010 and 2016, according to a study by researchers at the Stanford University School of Medicine.

Many have heard the stories of patients hit with massive, unexpected medical bills after an emergency-room visit or routine surgery, even if that hospital is in their insurance network. News organizations, including Vox.com, have reported on the problem, and both houses of Congress will continue debating draft legislation to tackle unexpected bills after members return from summer recess.

Yet there has been little data-driven research about the financial consequences of out-of-network billing on patients.

 “Our paper is one of the first to spell out just how big the problem really is,” said Eric Sun, MD, PhD, an assistant professor of anesthesiology.

 he paper was published Aug. 12 by JAMA Internal Medicine. Sun is the lead author. The senior author is Laurence Baker, PhD, professor of health research and policy and core faculty member at Stanford Health Policy.

Co-author Michelle Mello, PhD, JD, professor of law and of health research and policy, said news coverage has tended to focus on the real horror stories: patients who got surprise bills for tens or hundreds of thousands of dollars.

“Our study asked what a typical patient who goes to a hospital they’re supposed to go to can expect,” said Mello, who is also a core faculty member at Stanford Health Policy. “And what we found was that although typical out-of-network bills are lower than the horror stories, they’re still high enough to create enormous stress for families — and they’re also really common.”

The authors pointed to a recent survey that found 4 in 10 Americans would not be able to pay an unexpected expense of $400 without selling something or borrowing money. 

Yet the average amount of balance bills sent to patients in the study exceeded that: $2,040 for inpatient admissions and $628 for emergency visits in 2016. Further, 10% of the patients in the study faced an average of  $4,112 for inpatient admissions and $1,364 for ED visits.

Why is this happening?

Mello said there are two potential explanations behind the significant rise in out-of-network bills: insurers may be excluding more physicians from their provider networks, even when they essentially have no choice but to include the hospital; and hospitals may be increasingly tempted to make use of physician staffing groups that have market power and can resist pressure to come in-network at lower rates. 

Balance billing may also occur because insurers pay doctors too little to practice at one given hospital, so to make ends meet, they bill patients directly for additional amounts, Sun said.

“It certainly may be the case that consolidation among insurers has resulted in reduced payments to physicians, and that out-of-network billing is a way to compensate for this,” he said.

The classic example, Sun said, of how surprise billing works can be found at community hospitals that are not run by a multispecialty organization, such as Stanford Health Care or Kaiser Permanente. In that community hospital, everyone is in business for themselves. So, for example, when a patient comes to the emergency department and needs an operation, even if the patient’s chosen surgeon is in-network, all the specialists helping the patient through the surgery may not be. It is often difficult for patients who are in- or out-of-network to walk away if he or she doesn’t like the arrangement, particularly in emergencies.

“The bottom line is this: In a lot of situations, patients can’t choose the doctor, and the doctors can’t choose the patient,” Sun said. “It’s an especially big problem for hospital-based physicians, such as emergency room physicians, anesthesiologists, and radiologists.”

Even if a patient schedules a procedure, such as knee-replacement surgery or the delivery of a baby, accidents or the unknown often occur.

“What if something goes horribly wrong and you have to go to the ICU?” Sun said. “Now everyone seeing you, from the cardiologist or a neurologist, could be out-of-network.”

Crunching the numbers

The researchers examined 5.5 million inpatient admissions and 13.6 million emergency room admissions between 2010 and 2016, using data from the Clinformatics Data Mart, which comprises private health insurance claims from all 50 states from a large commercial insurer. The individuals in the database represent about 19% of all American who have commercial health insurance.

The researchers found the percentage of emergency department visits resulting in one or more out-of-network bills increased from 32.3% to 42.8% and that the average amount billed nearly tripled from $220 to $628. A whopping 85.6% of ambulance rides to emergency departments resulted in an out-of-network bill..

The proportion of inpatient admissions with out-of-network bills increased by almost 60% over that period, from 26.3% to 42%, with the average out-of-network bill more than doubling from $804 to $2,040.

The study found out-of-network billing was common among ambulance services and hospital-based physicians — such as emergency physicians, radiologists and anesthesiologists — providing care at in-network hospitals. “In some circumstances, patients could easily assume that the entire hospital team is in-network and thus the balance billing may come as a surprise,” the authors wrote. “Further, in these contexts, patients may have limited ability to choose.”

The authors note that the findings support current efforts in Congress and the states to limit balance billing in contexts where surprise is likely to be involved. They also note that patients may succeed in getting out-of-network bills reduced in some circumstances.

Still, Mello added: “This is not the way to run a health-care system. Avoiding outrageous bills shouldn’t require a PhD in health policy.” 

The other co-author of the study is Jasmin Moshfegh, a graduate student in the Health Policy PhD program.

Breast cancer is the second-most common cancer in the United States in women after nonmelanoma skin cancer — and the second leading cause of cancer deaths, after lung. 

Of the estimated 252,710 women who were diagnosed in 2017 with breast cancer, 40,610 died of the disease, or about 16%.

The odds of death from breast and ovarian cancer grow even greater when women have genetic mutations known as BRCA1and BRCA2. The mutations — which can make it harder for tumor-suppressing proteins to attack cancer cells — occur in 1-in-300 women and account for 5% to 10% of breast cancer cases and 15% of ovarian cancer cases.

The risks of ovarian and breast cancer are as high as 45% and 70%, respectively, in women carrying the genetic mutations. So the U.S. Preventive Services Task Force in 2005 and 2013 recommended genetic counseling for women with a high-risk family history of breast or ovarian cancer.

The Task Force is updating its recommendation for primary-care clinicians: They should now consider women with previous breast cancer or ovarian cancer who are considered cancer-free for genetic counseling — and more explicitly include ancestry as a risk factor.

The recommendations were published on Aug. 20 in the Journal of the American Medical Association.

“Some women can benefit from risk assessment, genetic counseling, and genetic testing, but not all women need these services. We suggest women talk to their clinicians and decide on best next steps together,” said Stanford Health Policy’s Douglas K. Owens, chair of the independent, voluntary panel of national experts in prevention and evidence-based medicine.

The Task Force notes that some ancestries are associated with increased risk of BRCA1/2mutations. An accompanying editorial in JAMA by Susan Domchek, MD, and Mark Robson, MD, notes that 1-in-40 random individuals of Ashkenazi Jewish descent have 1 of 3 specific BRCA1or BRCA2founder mutations, compared with 1-in-300 in the general population.

Genetic risk assessment and BRCA1/2mutation testing is a multi-step process that begins with identifying patients with a family or personal histories of breast, ovarian, tubal, or peritoneal cancer; family members with known harmful BRCA1/2mutations; or ancestry associated with harmful BRCA1/2 mutations. 

Clinicians should begin with a family and personal history for all women, Owens said.

“If this raises concerns about increased risk, the Task Force recommends that primary care clinicians use one of the available validated risk-assessment instruments to further assess the risk for BRCA1/2mutations.”

If the instrument indicates increased risk for BRCA1/2mutations, clinicians should refer patients for genetic counseling. Women found to be at increased risk from genetic counseling should then be referred for mutation testing, Owens said.

Listen to this podcast about the new USPSTF recommendations.

[[{"fid":"238783","view_mode":"crop_870xauto","fields":{"format":"crop_870xauto","field_file_image_description[und][0][value]":"","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_credit[und][0][value]":"","field_caption[und][0][value]":"","thumbnails":"crop_870xauto"},"link_text":null,"type":"media","field_deltas":{"1":{"format":"crop_870xauto","field_file_image_description[und][0][value]":"","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_credit[und][0][value]":"","field_caption[und][0][value]":"","thumbnails":"crop_870xauto"}},"attributes":{"class":"media-element file-crop-870xauto","data-delta":"1"}}]]

In another accompanying editorial published by JAMA Network Open, researchers note that since the new recommendations add ethnicities and ancestries associated with BRCA1or BRCA2gene variants, the number of testable patients has now been expanded.

“With the increase in women eligible for counseling and testing under these recommendations and the explicit directive for primary care practitioners to consider clinical genetics training … the oncology community should welcome the opportunity to better integrate comprehensive cancer risk assessment and genetic testing for BRCA1and BRCA2into routine preventive medicine,” they write.

Population-based risk assessment may increase insurance coverage and clinician-directed access to cancer genetic testing for up to 50% more women in the primary care setting than family history-based risk assessment alone, writes Olufunmilayo I. Olopade, MD, an expert in cancer risk assessment at the Center for Clinical Cancer Genetics and Global Health at the University of Chicago.

She and her co-authors note that researchers are also finding higher BRCA1and BRCA2 mutation frequencies across more diverse populations than previously realized. The genetic mutations have been identified in 12% to 18% of African-American patients with breast cancer; Hispanic high-risk patients living in the southwest of the United States, with a personal family history of breast or ovarian cancer, were found to have mutations rates as high as 25%.

Yet not as many African-American and Hispanic women get tested for the BRCA gene compared with white women, due to less awareness about genetic testing, distrust in the medical system, lower education levels and lack of physician referrals. 

And then there’s the cost.

“The cost of testing, now at $250 out of pocket if not covered by third-party payers, continues to remain an understandable point of concern for patients and the health-care system, but can be consciously addressed to reduce disparities in testing,” they write.

Olopade and her colleagues note there are fewer than 700 cancer-specific genetic counselors in the United States, relative to the more than 300,000 primary care physicians.

“Expanding access to genetic counseling and testing across the medical spectrum of care to primary care is not only appropriate, but also critical,” they said.

Americans have witnessed repeated mass shootings. The carnage in Texas and Ohio last weekend claimed another 31 lives and has left the nation stunned and angry.

Many are demanding that members of Congress pass tougher gun-control laws; others blame mental health and violent video games for the rampant shootings.

Stanford Health Policy’s David Studdert — an expert on the public health epidemic of firearms violence — acknowledges that mass shootings are on the rise in the United States.

“It’s been a horrific weekend,” said Studdert, a professor of law at Stanford Law School and professor of medicine at Stanford School of Medicine. “Experts now generally agree that mass shootings are becoming more common — and that a common thread is disaffected young men who have access to high-caliber, high-capacity weapons.”

Both suspects in the Dayton and El Paso shootings fit this profile.

Studdert notes, however, that while mass shootings have become the public face of gun violence, they account for less than 1% of the 40,000 firearm deaths each year.

“So as a public health researcher, I do care about mass shootings and I am interested in understanding and their causes — but the focus of my ongoing research is the other 99 percent.”

Largest investment in firearms research in two decades

It’s that focus the Studdert will be pursuing in a recently-awarded $668,000 grant from the National Collaboration on Gun Violence Research. The private collaborative’s mission is to fund nonpartisan, scientific research that offers the public and policymakers a factual basis for developing fair and effective gun policies.

Studdert, Yifan Zhang, a statistician with Stanford Health Policy, and Stanford political scientist Jonathan Rodden are working with colleagues at UC Davis, Northeastern University and Erasmus University Rotterdam on the Study of Handgun Ownership and Transfer, or LongSHOT.

The team is following several million Californians over a 12-year period to better understand the causal relationship between firearm ownership and mortality. They launched in 2016 with the initial goal of assessing the risks and benefits of ownership for firearm owners.

“The implications of firearm ownership for owners is important because they usually are the ones making the decision to purchase and own,” Studdert said. “But we knew from the beginning that this was only part of the picture. The presence of a firearm in the home may also have health implications for the owners’ family members.”

In the new study, the researchers will identify the cohort of adults in California who live with firearm owners but are not themselves gun owners, and then compare their risks of mortality to a group who neither own weapons, nor live with others who do.

Surprisingly little is known about the “secondhand” effects of having guns in the home.

“Existing studies don’t differentiate between owners and non-owners within households, and that is something we have the ability to look at,” Studdert said. “And a very large proportion of non-gun-owners who are living in homes with guns are women — so this is a group that has really been understudied.”

There is already substantial evidence that a gun in the home is associated with increased risks of suicide. But it is not clear how particular subgroups, such as women who don’t own guns, are affected.

“Because our cohort is so large,” Studdert said, “we will also be able to explore whether gun ownership confers certain benefits, as gun-rights advocates often claim, such as enhanced safety in dangerous neighborhoods.”

Studdert said a better accounting of the risks and benefits that firearm ownership poses for non-owners could help inform decisions regarding gun ownership and storage, as well as policies aimed at improving gun safety.

The politics of federal funding for firearms research

The National Collaboration on Gun Violence Research is funded through private philanthropic donations. It was seeded with a $20 million gift from Arnold Ventures and intends to raise another $30 million in private funding for firearms research.

“It’s the biggest investment in firearms research since the late 1990s,” Studdert said.

Research on the impact and causes of firearm violence was dealt a huge blow in 1996 when the so-called Dickey Amendment was passed by Congress. The law has been interpreted as prohibiting the National Institutes of Health and the Centers for Disease Control and Prevention from conducting firearms research.

Studdert said that the growth of research funding from philanthropies like the Arnold Foundation and Joyce Foundation is a welcome development, but that it will take a large and sustained investment to move the science of firearm violence forward.  

“The core funder of large-scale research essentially vacated the space for 20 years,” he said.  “It’s going to take some time to recover. Developing a generation of researchers with expertise will take give to 10 years. But it has to be done — the size of the social problem demands it.”

Many primary-care physicians continue to join multispecialty group practices, such as the Palo Alto Medical Foundation and Stanford Health Care, instead of working in their own solo practices or in practices with only other primary care doctors.

Physicians in practices of nine or fewer dropped from 40% in 2013 to 35% in 2015; the rate of those in practices of 100 or more increased from 30% to 35% during the same period.

But is this growing trend having a positive impact on health-care use and spending?

Stanford Health Policy’s Loren Baker and Kate Bundorf set out to find a few answers.

In their new working paper published by the National Bureau of Economic Research, they focused on Medicare beneficiaries who changed their primary care physician when they moved from one area to another. They focused on people who switched from a doctor in a primary-care-only practice to one in a multispecialty practice, and those who made the opposite switch.

They then compared changes in health-care use and spending before and after the move — and among patients who switch practice types and those who do not.

“We wanted to look at people who experienced an abrupt change in their primary care physician for reasons other than an explicit choice to change their physician,” Bundorf said.

“A weakness of this approach is that they may differ from people who don’t move,” said Bundorf, an associate professor of health research and policy at Stanford Medicine and a senior fellow at the Stanford Institute for Economic Policy Research (SIEPR). “But our descriptive statistics suggest that we capture a broad cross-section of types of people.”

They identified 119,272 patients who moved from one area to another and used data from Medicare claims from 1999 to 2010.

The results were striking

The results of the analysis show that multispecialty practice decreased annual medical spending by $1,600 per Medicare beneficiary — a 35% reduction in spending.

“The size of the result is very intriguing,” said Baker, a professor of health research and policy at Stanford Medicine and also a SIEPR senior fellow. “We are often happy when a change in health-care delivery can help achieve savings of even a few percentage points, but our results suggest the potential here is much larger than that.”

With their co-author Anne B. Royalty of Indiana University-Purdue University Indianapolis, they found the results were driven primarily by changes in hospital expenditures and are concentrated among patients with two or more chronic condition. This suggests multispecialty practices improve care delivery by reducing hospitalizations among relatively sick patients.

“The results imply that, while research has shown the potential for physician consolidation to increase prices in some settings, large multispecialty groups also have the potential to lower costs,” the authors wrote.

Baker said when they set out to do this research, they were not sure what they would find. 

“The results suggest that the way care is organized can be important for health-care delivery and that there may be organizational changes that could help us better manage spending.”

Larger practices don’t always lead to better health outcomes

While the pervasive view underlying the move toward consolidated practices is that larger, more integrated organizations provide care more efficiently through greater coordination of care, the literature on physician integration largely does not support this view. 

The authors point to a study that looked at data from the 1970s to 2013, which found little evidence that these larger practices improve the quality and lower the cost of health care. 

“The hope of many policymakers was that the larger, multispecialty organizations could organize care more efficiently,” Bundorf said. “Yet there is little evidence to date that larger practices have generated these types of benefits for patients. Since the literature has not taken a hard look at the single or multispecialty side of things, we wanted to investigate that here.”

While the trend is toward these big multispecialty practices, many physicians choose to remain in smaller or solo ones.

The Kaiser Family Foundation estimates that there are some 139,000 primary care physicians in the United States. Despite the transition to multispecialty practices, some 40% of physicians continue to work in solo practice or a small group of two-to-nine physicians.

It’s important, they said, to continue examining the health-care spending of all practice sizes.

“We are very curious to continue work on ways to learn more about how organizations achieve savings,” Baker said. “We are also excited to continue thinking about opportunities for policies that would encourage the formation of effective practices, for example through payment policy reforms.