By KIM BELLARD

Among the many things that infuriate me about the U.S. healthcare system, health systems sending their patients to collections – or even suing them – is pretty high on the list (especially when they are “non-profit” and./or faith-based organizations, which we should expect to behave better).

There’s no doubt medical debt in the U.S. is a huge problem. Studies have found that more than 100 million people have medical debt, many of whom don’t think they’ll ever be able to pay it off. Kaiser Family Foundation estimates Americans owe some $220b in medical debt, with 3 million people owing more than $10,000. It’s oft cited that medical debts are the leading cause of bankruptcy, although it’s quite not clear that is actually true.

So you’d think that helping pay off that debt would be a good thing. But it turns out, it’s not that simple.

A new study from the National Bureau of Economic Research (NBER) by Raymond Kluender, et. alia, found that, whoops, paying off people’s medical debt didn’t improve their credit score or financial distress, made them less likely to pay future medical bills, and didn’t improve their mental health.

“We were disappointed,” said Professor Kluender told Sarah Kliff in The New York Times. “We don’t want to sugarcoat it.”

The researchers worked with R.I.P. Medical Debt, a non-profit that buys up medical debt “at pennies on the dollar,” to identify people with such debt, and then compared people whom R.I.P. Medical Debt had helped versus those it had not. One set of people had hospital debts that were at the point of being sold to a collection agency, and another had debts that had already been sent to collection. And, perhaps to highlight how little we understand our healthcare system, they asked experts in medical debt what their expectations for the experiment were.

Much to everyone’s surprise, having debt paid off made no difference between control and debt-relief groups. I.e.,

• “We find no average effects of medical debt relief on the financial outcomes in credit bureau data in either of our experiments.
• We similarly estimate economically small and statistically insignificant effects on other measures of financial distress, credit access, and credit utilization.
• We find that debt relief causes a statistically significant and economically meaningful reduction in payment of existing medical bills.
• We estimate statistically insignificant average effects of medical debt relief on measures of mental and physical health, healthcare utilization, and financial wellness, with “opposite-signed” point estimates for the mental health outcomes relative to our prior.”

In short: 

Our findings contrast with evidence on the effects of non-medical debt relief and evidence on the benefits of upstream relief of medical bills through hospital financial assistance programs. Our results are similarly at odds with views of the experts we surveyed, pronouncements by policymakers funding medical debt relief, and self-reported assessments of recipients of medical debt relief. 

Amy Finkelstein, a health economist at the MIT and a co-director of J-PAL North America, a nonprofit group that provided some funding for the study, told Ms. Kliff: “The idea that maybe we could get rid of medical debt, and it wouldn’t cost that much money but it would make a big difference, was appealing. What we learned, unfortunately, is that it doesn’t look like it has much of an impact.”

If only it was that easy.

To be clear, there were three key statistically significant effects:

• “small improvements in credit access for the subset of persons whose medical debt would have otherwise been reported to the credit bureaus,
• modest reduction in payments of future medical bills, and
• worsened mental health outcomes, concentrated among those who had the largest amount of debt relieved and those who received phone calls to raise awareness and salience of the intervention.”

The authors admitted they had not expected the mental health results and had no good explanation, but their “preferred interpretation is that recipients of the cash payments viewed the transfers as insufficient to close the gap between their resources and needs, raising the salience of their financial distress and harming their mental health.”

As Neale Mahoney, an economist at Stanford and a co-author of the study, told Ms. Kliff: “Many of these people have lots of other financial issues. Removing one red flag just doesn’t make them suddenly turn into a good risk, from a lending perspective.”

The authors concluded:

Nonetheless, our results are sobering; they demonstrate no improvements in financial well-being or mental health from medical debt relief, reduced repayment of medical bills, and, if anything, a perverse worsening of mental health. Moreover, other than modest impacts on credit access for those whose medical debt is reported, we are unable to identify ways to target relief to subpopulations who stand to experience meaningful benefits.

On the other hand, Allison Sesso, R.I.P. Medical Debt’s executive director, told Ms. Kliff that study was at odds with what the group had regularly heard from those it had helped. “We’re hearing back from people who are thrilled,” she said.

As statisticians would say, anecdotes are not data.

————-

Removing medical debt seems like a can’t-lose idea. A number of states and local governments have passed programs to pay off medical debt (most working with R.I.P. Medical Debt) and a number of others are considering it.

Last fall the Consumer Financial Protection Bureau initiated rulemaking that would remove medical bills from credit reports. It has also, according to NPR, “penalized medical debt collectors, issued stern warnings to health care providers and lenders that target patients, and published reams of reports on how the health care system is undermining the financial security of Americans.”

Director Chopra admits: “Of course, there are broader things that we would probably want to fix about our health care system, but this is having a direct financial impact on so many Americans.”

If nothing else, the new study should remind us that our health system is best at putting band-aids on problems rather than solving them. The problems we should be addressing include: why are so many charges so high, why aren’t people better protected against them, and why don’t more Americans have enough resources to pay their bills, especially unpredictable ones like from health care services?

I’m glad R.I.P. Medical Debt is doing what it is doing, but let’s not kid ourselves that it is solving the problem.

Kim is a former emarketing exec at a major Blues plan, editor of the late & lamented Tincture.io, and now regular THCB contributor

According to an article in CNN, some doctors in the U.S. are going broke. Read it here . I feel sorry for anyone who goes broke, but this is not unexpected.

According to the author, some doctors have unsustainable debt and are facing reimbursement reductions by insurers. A good question is why the unsustainable debt?  After all, many of our institutions have become bloated and can’t adapt to the economic downturn.

The article mentions a cardiologist who is going broke. According to a StudentDoc survey, the average cardiologist makes $403,000 per year. The average cardiovascular surgeon makes $558,719 per year. That’s not bad in either case.

My point is a lot of very highly paid people in every walk of life end up in bankruptcy. No matter how much you make, you can always spend more.  I believe the health care industry has been as guilty of over-optimism as the real estate and finance sectors.

People have been saying for years that many clinics and hospitals will face a solvency crisis brought on by excessive borrowing to fund sleek new facilities and excessive purchases of uber-costly medical equipment. Some doctors became wealthy during the good years, and some of them made purchases not easily retracted now that the economy has turned down.

Let’s not blame health insurers.  They represent the voice of their policyholders who are hurting economically.

Hospitals have been buying doctors’ practices on borrowed money for years. The ones that paid exorbitant prices are soon going to be hurt badly. We are seeing the leading edge of that now. In five years large numbers of hospitals will be looking for bailouts.

Thomas G. Emerick is the president of Emerick Consulting, LLC. He has served on a variety of employer coalitions and associations including the board of the National Business Group on Health and the U. S. Chamber of Commerce Benefit Committee. He blogs at Cracking Health Costs.

The housing crisis that precipitated our ongoing recession began with the foreclosure of 15% of US mortgages. There remains substantial disagreement, however, about whether and how public health departments should specifically address health problems experienced by the people who lost their homes in this crisis. While poor housing quality and homelessness have been statistically correlated to illness for many years, some argue that the correlation merely represents the influence of other factors that are common among people with housing insecurity: indebtedness and inability to pay for medical services, unemployment and associated insurance loss, food insecurity, mental illness, substance abuse, or family instability resulting in poor healthcare seeking or inadequate medical adherence.

As a result, it’s not obvious whether having health departments improve housing availability or quality will necessarily improve health conditions among the groups who face foreclosure. If better housing is really directly linked to better health outcomes, then health departments should expect a return on their investment in housing programs for this group. But if the statistical finding is merely secondary to other factors like indebtedness, then the money might be better spent elsewhere, for example in debt repayment programs, or in preventing the type of predatory banking practices that lead to the foreclosures. In this post, we try to answer the question: is the foreclosure crisis making people sick? And if so, what interventions have been shown to work, if any?

The foreclosure crisis and health outcomes

There have been many, possibly hundreds, of cross-sectional studies reporting statistically significant associations between poor housing conditions and poor health. The most common are associations between poor housing quality and asthma exacerbations (from dust and mold), particularly in substandard public housing facilities; associations between crowding and infectious diseases and mental health problems have also been researched for decades.

Lack of stable housing, not just poor housing quality, has been statistically related to poorer health as well; a recent interview-based study in the American Journal of Public Health, for example, revealed that crowding and multiple moves among low-income American families during the decade prior to the recession was associated with poor health, lower weight, and developmental delay risks among young children after correcting for a number of other potential confounders (see the table below).

Only a few studies, however, have specifically looked into the impact of the recent foreclosure crisis on health. The foreclosure crisis is not merely a matter of people moving to potentially substandard low-income tenements after foreclosing on single family homes; the recent crisis also involved involuntary moves, unanticipated homelessness and debt repayment problems among those who are not usually in the category of chronically housing-insecure people.

A survey of 388 residents in Alameda County, California (which contains the city of Oakland) reported that between 2006 and 2009, one in four mortgages in Oakland (14,941 property owners) entered into foreclosure, and among them almost one-third of tenants said they were living in sub-standard conditions with health risks due to mold, pests and utility shut-offs. It’s not clear how this rate related to earlier exposures in this group or to rates among those who did not foreclose. But the prevalence of stress, depression and anxiety were more than two times higher among foreclosed residents than comparison groups. Nearly 90% of respondents reported struggling to pay debts or pills, and many reported having to choose whether to pay for food, healthcare or utility bills. It’s not hard to imagine this resulted in decisions to forgo doctor’s visits, prescription refills or high-cost food purchasing (e.g., produce).

The National Bureau of Economic Research also issued a report in August 2011 investigating the relationship between foreclosure activity and the health of residents using zip code-level longitudinal data from Arizona, California, Florida, and New Jersey (four states that have been among the hardest hit by the foreclosure crisis). The researchers at the Bureau paired foreclosure data from 2005 to 2009 from realty databases with data on emergency room visits and hospital discharges. The zip code-level data allowed the researchers to control for many potential confounding factors associated with living in a particular area at a given time, such as changes in poverty rates from the recession, in an attempt to isolate the specific impact of foreclosures (however, they did not have individual-level data to see whether the foreclosed people were necessarily the same people going to the ER). The study found that an increase in the number of foreclosures was significantly associated with increases in medical visits for mental health problems (anxiety and suicide attempts) in a given zip code, was associated with an increase in visits for stress-associated medical conditions such as hypertension. As a control variable to ensure they weren’t finding spurious correlations, the researchers confirmed that foreclosures were not related to visits for cancer treatment, an indicator that should not respond to stress from housing loss or housing quality changes in the short term. Foreclosures also had a zero or negative effect on elective procedures, as one might expect. Age-specific results suggested that the foreclosure crisis was having its largest health effects on individuals 20 to 49 years old, and had larger effects among Blacks and Latinos than among whites.

In a separate study, researchers at the National Poverty Center used data from the Michigan Recession and Recovery Study to examine four measures of health: self-rated health, depression, anxiety attack, and problematic alcohol use, each assessed using standardized screening surveys. The Michigan Study is a stratified random sample of non-institutionalized English-speaking adults aged 19 to 64 who lived in Southeastern Michigan; it excludes individuals living in shelters, and oversamples Blacks to reflect the residential composition of the area. The researchers reported that nearly one-third of 894 participants recently experienced some type of housing instability. Many, but not all types of instability were associated with a greater likelihood of reporting fair or poor self-rated health, depression, and anxiety. Being behind on rent was associated with poorer self-rated health and more depression and anxiety, while being behind on a mortgage or in foreclosure was associated with a greater likelihood of anxiety and lower likelihood of problematic alcohol use, even after adjusting for socio-demographic characteristics, health history and financial problems and dropping individuals with a history of previous housing problems from the statistical analysis (it’s a common finding that in times of recession, drinkers cut down on alcohol, given the expense). Additionally, foreclosure increased the risk of depression, while a recent experience of homelessness increased the risk of both depression and anxiety. However, frequent moves were not associated with poorer health if they were not made for cost reasons, and “doubling up” with other family members was not associated with poorer health either, again suggesting that stress-inducing factors and stress-mitigating factors may be mediators of the relationship between foreclosures and housing.

The findings reveal the importance of distinguishing different types of housing instability, considering separately the instability experience of being behind on housing payments even when a move has not yet taken place, and comparing the way a range of housing instability experiences are associated with health. A key problem with all three of these studies, however, is that they’re not designed to examine actual medical outcomes from foreclosure; they rely on self-reported indexes of stress rather than actually tracking real morbidity and mortality. Unfortunately this tends to overemphasize healthcare utilization measures (which could be confounded by loss of insurance) or stress-related health conditions that people almost always agree to on surveys (no one says they’re not stressed, even in good times, and separating the stress of poverty from the stress of foreclosure is splitting academic hairs). Such studies therefore risk ignoring other individual-level health risk exposures and actual medical outcomes (e.g., should doctor’s offices also anticipate higher asthma rates from poorer housing quality among those who have moved?) at the expense of overemphasizing self-perceived mental health outcomes.

Intervention data

While these preliminary surveys give a glimpse into the potential health consequences of recession, can the health problems experienced by those who have faced foreclosure be improved by essentially reversing foreclosure’s effects through housing programs?

A recent systematic review revealed 19 housing intervention studies that measured quantitative health outcomes from a number of housing support programs: medical priority rehousing (i.e., prioritizing public housing for those with chronic medical problems), rehousing people in more affordable areas, and subsidizing the refurbishment of existing units to improve the lower-quality tenements to which people have moved. Eleven of the studies were prospective, of which six had control groups.

The three studies of medical priority rehousing reported improvements in self-reported physical and mental health; however only one was a prospective study (and was small), and none of them controlled for the effects of possible confounding variables. The studies of rehousing and refurbishment were less impressive. The results showed that housing improvements may improve residents’ mental health, but generally only small improvements in physical health were observed. In several cases, particularly those involving refurbishment, increases in residents’ subsequent housing costs (as landlords raised rents after renovations) actually produced adverse effects, causing people to forgo important expenditures like out-of-pocket healthcare payments, medication co-payments and quality food to pay rents. In many cases, people moved away as subsidized refurbishment of low-income housing made such housing too costly, as the refurbishments were not typically paired with rent-control laws.

In total, eight of nine studies that measured mental health showed improvements and one study demonstrated a “dose-response” relationship between the degree of housing support and mental health outcomes. Effects on specific physical outcomes (i.e., general health, respiratory symptoms) were less clearly recorded, with similar studies showing effects in opposite directions among small samples of people. While the evidence suggests that housing interventions could likely improve mental health, the poor quality of the studies forbids us from being able to assess long-term physical health outcomes from housing improvements; what is clear is that if housing interventions are not paired with economic regulations to maintain low rents, they may have little benefit and potentially even do harm.

An agenda for action

In light of this data, facilitating access to rent-controlled or medical priority housing might be key areas of attention for public housing agencies who have been tasked with the management of many foreclosed families during the recession period. Several families who have faced foreclosure have entered into public housing. There are 3,200 public housing agencies nationwide, but 40% of the public housing structures they administer are more than four decades old. The 1937 Housing Act established that is was a federal priority to create safe, decent, and sanitary housing throughout the country. Yet more than 70 years later, almost 13 million low-income people pay more than half of their monthly income for rent, live in severely substandard housing, or both.  Most public housing residents have limited space and are isolated from core social and health services. People of color are also twice as likely as their white counterparts to live in moderately or severely deficient homes.

Since the Surgeon General’s Call to Action in 2009 to promote healthy homes among low-income residents, several agencies have banded together to create a federal interagency working group on healthy housing, with a focus on improving public housing options. HUD unveiled its Healthy Homes Strategic Plan, declaring the necessity to “work across communities and the nation to ensure our homes are sited, designed, built, renovated, and maintained in ways that support the health of residents”. Legislative proposals have been introduced in both the House and Senate. Senator Jack Reed (D-RI) and Congressman Robert Brady’s (D-NJ) Healthy and Safe Housing Act, for example, supported the creation of an independent interagency Council on Healthy Housing in the Executive Branch.

All of this sounds like political gesturing. So the Kresge Foundation, the National Center for Healthy Housing and others have recently assembled lists of specific actions these political groups can take on to actually implement healthier housing policy. Among the suggestions were to:

(1) Officially develop national standards for housing and building codes, particularly for public housing developments. While available studies of the impact of foreclosure suggest primarily mental health risks from moving to lower-quality housing or becoming homeless, these studies have not been designed to collect information on long-term relationships between poorer housing among foreclosed families and physician ailments that emerge after years or manifest in ways that are not captured by stress-focused surveys. The historical literature on housing and health suggests, for example, that detecting the relationship between asthma exacerbations and worsening housing conditions can often take years. Currently, housing and building codes are local, varied, and sporadically enforced. Setting and branding a recognizable national standard is an important step. But both scientific and practical expertise will be needed to achieve a feasible, evidence-based set of standards. Part of this will involve setting standards for dust and mold in the context of respiratory infections and asthma, as well as reforming the nation’s chemical use policy to avoid hazardous substances in building materials (e.g., the Toxic Substances Control Act Reform Bill) and continue to address the persistent problem of lead exposure, which still affects more than 240,000 children nationwide.

(2) Create direct economic buffers for those with housing insecurity, not just indirect subsidies. We previously found that spending on economic “buffer” programs was a key defense against morbidity and mortality increases during recessions. Data from previous recessions suggests that rapid transitions people have not been able to prepare for—sudden job loss or unexpected housing loss—are the key modes through which recessions trigger increased heart disease deaths, mental illness, and substance abuse during recessions, particularly among those with limited social support from family and friends. One effective buffering policy appears to mitigate these problems more than any other: active labor market programs, which are programs designed to rapidly re-employ people who have just become unemployed, or to keep some elements of work available for those who would otherwise be laid-off. Such programs appear more effective at preventing downstream health problems than indirect social safety net operations that individually address each consequence of unemployment (e.g., food or housing insecurity) rather than directly addressing the unemployment itself. That is, from a medical perspective, having a job that can pay the rent (when that rent is set to a reasonable level) appears to be better for your health than being unemployed but having free housing.

(3) Implement model programs through public health departments. Healthy housing has been called a “multidisciplinary issue”—a term that makes it sound like many people are coming together to tackle the problem creatively. In reality, this phrase means that no single agency or organization is taking responsibility for the problem. Public health departments are uniquely poised to implement healthy housing policies, because they have the incentive to avoid downstream county hospital and emergency room utilization costs, because they have the technical expertise to address medically complex issues arising among public housing residents, and because the evidence base exists in the public health literature to implement specific cost-saving and life-saving programs through health departments. A recent series of discussions sponsored by the National Heart, Lung and Blood Institute (a division of the NIH) reported on specific programs ranging from a Boston-area asthma relief initiative that provides practical intervention protocols to reduce housing-related asthma exacerbations (diagram below), to the San Francisco Housing and Urban Health program, which implements medical priority rehousing in single-room-occupancy hotels of the Tenderloin district.

Many of these model programs, however, are understandably designed for the chronically homeless and for folks who have traditionally occupied public housing. The new wave of foreclosures offers different challenges, as it has affected families that have been securely housed in single-family homes until recently, and who will likely not be familiar with state-based safety-net resources. In the meantime, better epidemiological studies are needed in the public health literature to examine what if any physical health outcomes really result from foreclosure and movement of these people to alternative housing arrangements or homelessness (e.g., should we plan for more tuberculosis transmission, or will mental health problems truly dominate?), and to prospectively evaluate the comparative effectiveness of different housing interventions in a manner that corrects for confounding factors to provide us with a clear sense of the impact of alternative programs.