The President's Town Hall Meeting

The President held a town hall meeting at the White House yesterday that was given more than an hour of time on network television. This reflects the importance of the topic at hand--health care reform. There were a number of interesting comments made, although one observer pointed out that there was really nothing new and earth shaking that was discussed.

What did catch my attention, even if it was not so new, was the statement that people who are satisfied with their current insurance and their current doctor (which, for my nursing friends, will hopefully translate into a more general term, "provider") should not have to make a change. However, the President also used the cliche about the devil you know being better than the one you don't--in other words even if we are not thrilled with our current health insurance or provider, we know the problems with what we have and we fear what we may get instead. If we combine these two ideas, despite new choices being offered to those who have no other option, those with a choice may make the choice to stay in their current insurance plan with their current providers. If so, nothing about those people's health care utilization is likely to change.

I should say--at first. We will probably need everyone's health care utilization to change in some way if we are going to control costs in the long run. One way to get around the possibility that little change will occur for people who don't change insurers or providers is to change incentives at other levels. For instance, incentives for employers, insurers, providers, and facilities could change. This would be sort of a "behind the scenes" approach to making change.

With that in mind, politicians and the public will have to figure out whether it is ultimately preferable for the public to face new incentives (whatever they may be) head on or to face similar incentives for themselves but find that they will be asking their providers why things are changing around them as providers' incentives change.

Click here for an audio version of the blog.

Prevention of Blindness

I am all for the prevention of blindness and the provision of services for those who have uncorrectable vision impairments to make individuals with this situation as self-sufficient as possible. I have been involved in estimates of the costs of vision impairment around the world and in the United States. My colleagues have done great work in Australia, in the UK, and now in Canada. A description of the recently completed work in Canada (done by an excellent set of researchers from Australia) can be found at http://www.cbc.ca/health/story/2009/06/23/vision-loss-cost.html.

The story is similar to what has been found in other countries with aging societies. The number of blind individuals or individuals with vision impairment will increase as society ages. The costs per person are high, and the costs per person are distributed between direct costs of medical care, costs of support, and costs due to lost productivity.

The news article points out that Canada made a commitment in 2003 to develop a national vision plan by 2005 and implement by 2007. Interestingly, they have not completed that task. We, in the United States, are not the only ones who take a long time to implement policy.

Finally, the report called for a plan with four criteria, the first of which is "Cost-effective preventive measures". A key thing to recognize when academic types talk about cost-effective preventive measures is that they will not always be cost saving. Certainly, for the person for whom something is prevented--it saves them costs. Certainly, for society as a whole, we'll end up paying less for costs related to a given condition. However, the costs of prevention (particularly if it is not well targeted) may exceed the cost savings.

Does this mean we should only adopt measures that save money? I don't think so. I do think that we need to be aware of whether money will be saved and whether we think that spending more money is worth the benefits. And, before you think that I am picking on vision problems, I can assure you I am not. I would ask the same question about HIV prevention, flu prevention, or vision problem prevention.

Click here for an audio/text version.

The Health Plan No One is Debating

Today, I found a piece on the Wall Street Journal website that mentioned a health plan that we all have missed. You can read about it at http://www.cbpp.org/cms/?fa=view&id=674. It is a nice description of the proposed legislation. The proposal is bipartisan. The proposal has been estimated to be budget neutral through 2014. The plan would rely on group insurance and competition but through state level pooling rather than employers. It would even result in 99% of Americans being covered. It has a lot to offer.

However, it may be ignored for several reasons. First, neither of the sponsors are really "big name" senators. Second, it is a true overhaul of the current system. Third, despite being a real change, it is not single payer so those who are supporting that approach are not in favor. Even if this quiet little bill were "just right" for the country, it is probably "just wrong" for politics.

It shows how irrational our system of political debate can be.

Value Based Benefit Design

Here is an interesting link that describes value-based benefit design (http://www.commonwealthfund.org/Content/Newsletters/Purchasing-High-Performance/2009/June-18-2009/Feature-Articles/Its-All-About-Value-Based-Benefit-Design.aspx). In short, this means providing incentives to use specific types of care and not others. This type of benefit design could be used by a single payer or in a competitive market. in a competitive market there would be stronger incentives to find the best value-based insurance design to attract people to your plan to make the most money. If consumers understand what they are getting, this could create quite an array of choices that would make sense for different populations and allow people choices.

The interesting thing about value-based insurance design is that as we understand what really works (or doesn't work) for specific conditions we can design ever more specific benefits (inclusions and exclusions of what will be covered or what will be covered more generously) that are more specific to individuals. Some people may not find that appealing. Nevertheless, at a population level this could be very useful.

The link above gives a couple of examples. One is drug reimbursement to encourage generics. That is nothing new. Another is a set of incentives to encourage patients (and their providers who recommend procedures) to use minimally invasive surgery when the option is available. That would seem like an obvious choice for most patients. The key is that the doctors have to understand the incentives as well and review with the patients the options that they have and what each option will cost.

This example actually illustrates the value of either single payer or retention of something closer to our current system. If each plan or employer can pick and choose what it wants to encourage, that would allow lots of choice and allow people with the right information in easily understandable format to find the plan that is closest to their ideal. However, more options means more time that physicians would have to spend reviewing with each patient to make the right decision for the patient based on incentives. One set of rules might be easier.

Nevertheless, this is not something that has made the news as much but definitely something to watch for once we get beyond the debate of how many payers we will have.

Little Things Can Change the World

I was a coauthor on a paper recently published in the Bulletin of the World Health Organization (http://www.who.int/bulletin/volumes/87/6/08-055673.pdf) about the burden of uncorrected refractive error (i.e. needing glasses but not being able to get them or having an outdated prescription). The Johns Hopkins Bloomberg School of Public Health put out a press release (http://www.jhsph.edu/publichealthnews/press_releases/2009/frick_vision.html), so I don't need to restate either the whole article or the press coverage. The article described how many people around the world don't have needed glasses and projects what the loss of productivity associated with it might be.

What I do want to mention is how easy (and how relatively inexpensive) this would be to fix. Here in the United States we talk about spending $1 trillion over 10 years to insure an extra 16 million people. The team with whom I worked on the paper that in the Bulletin of the WHO has not yet done the "down to the penny" calculations of the expected cost of fixing the problem of people who do not have the glasses they need, but I'll do a back of the envelope calculation here and then update when we have real numbers.

In the United States we can get glasses for as little as $50 for a cheap pair of glasses or spend hundreds of dollars for a fancy pair with lots of bells and whistles. However, there are places around the world where people can get glasses for as little as $5 or $10 per pair--perhaps a total of $30 by the time you pay the labor to fit them. We estimated a total of 158.1 million people didn't have the right glasses--if they had any. Even if we had to spend $100 per person, this would be a total of only $16 billion. And, this would not need to be done every year. At lower costs per person it may be less than $1 billion to solve the entire problem that affects more people around the world than the total number of uninsured in the United States.

So, there are some problems that have the potential to be solved for relatively small amounts of money in comparison to what we have been discussing in the United States for the health care system. Sometimes we would do well to think about the small problems we can solve (but that may be less exciting) in addition to struggling to solve some really big problems.

"Crowd Out"

One concern with a lot of public plans for health care is referred to in health economics as "crowd out". If we look at the Congressional Budget Office's (CBO's) analysis of the Kennedy-Dodd bill, we see a great example of this.

The bill will end up spending $1 trillion over 10 years. Over 30 million people will be covered under the plan funded by the government. Only 16 million of those will be from those who are uninsured at the present time. That leaves 2/3 of the problem with the uninsured unsolved for spending $1 trillion dollars. We could give every uninsured person (and there are nearly 50 million at this point) $2,000 of care a year for 10 years and spend only $1 trillion. Maybe that would only partially solve the problem for 50 million individuals (i.e. they would still be underinsured), but it would at least provide some coverage for everyone.

However, that is not what the plan is doing. The plan solves about 1/3 of the uninsured problem and then is projected to attract people who have private insurance at present. That is the situation called crowd out. It would be one thing if the government took over all health insurance. However, having a simultaneous public and private system there is a risk that some things will be taken out of the private market's hands (which usually operates more efficiently and has those who benefit from the insurance paying for it) and shifting it to the government which often acts less efficiently and spreads the cost to the taxpayers.

A great example in the past was influenza vaccination--so this really can be said to apply not just to insurance but to health care in general. Before Medicare covered flu shots, about 50% of older adults were getting the shots and paying for it themselves. A decade and a half later we still are not vaccinating all older adults on Medicare. Thus, the majority of what has been spent by the government has been for services that would have been provided anyway.

When we have public policies that only partially solve problems, we should be very careful to look out for "crowd out", assess how much there is likely to be, and assess whether we really want our tax dollars to be used this way.

A few thoughts on the nursing labor market

In addition to having the honor of being on the radio last week, a few other health economists of my generation have been on the radio or TV lately--each locally. One of my colleagues--Joanne Spetz (you can see her faculty profile at http://nurseweb.ucsf.edu/www/ffspetj.htm) was asked to comment on the nurse labor market. I won't claim to represent exactly what she said, but I know this is a hot topic at schools of nursing and among people looking to get back into the work force. So, I'll offer a few comments building on her ideas--have to give credit where credit is due.

For the first time in a long time, some nursing students have been having a more difficult time finding a job. Does this mean the nursing shortage that we have heard about for years is over? No.

Reason #1: Health care is a normal good--excuse the economics jargon. A normal good is one for which more is purchased when incomes go up. Right, now some people are out of a job--income down. Some people are feeling like they have less discretionary income--even if their income is constant or growing a little, the prices of things like electricity and gasoline are going up--that also makes people feel like their incomes are going down. Also, even if people's incomes are constant, they are paying more for health benefits--less to spend after buying health insurance. All of this will lead to less spending on health care if people's health doesn't change. Less being spent on health means we need fewer health care providers, including nurses.

Reason #2: More people are uninsured or having their health insurance benefits cut. This makes people pay more out of pocket. When they pay more out of pocket, they purchase less medical care. Again, this leads to less demand for nurses.

Reason #3: Nurses in the labor force are staying in the workforce longer (i.e. avoiding retirement) or looking for more hours. Nurses who were out of the labor force are returning to the labor force that they were out of when times were flush. All of this leads to more nurses being in the market to provide more hours.

So, with the demand for health care and ultimately nurses going down and the supply of nursing hours going up, we have less of a shortage.

Should states then stop looking to expand nursing faculty and should those who have been planning to go into nursing pull out? Probably not. Here's why:

We expect that sooner or later the economy will turn around. That will result in incomes going up. Policy will likely lead to more people being insured--more on that in a day or two. And, eventually, even if more people were not insured and the economy did not get better, the generation before mine will create a bigger demand for health care than any that has gone away because of the economy. Sooner or later we will need all the nurses that are in the labor force, that are being trained right now, and that will be trained in the near future. While the economic situation has led to some temporary relief of the shortage, the demographic situation will change this in the near future regardless of what happens to the economy.

The Economic Importance of Words

The WHO declared an H1N1 influenza pandemic last week--on Thursday June 11. Did anyone take notice?

Why did they declare a pandemic? It is not a measure of severity. It is a measure of ease of spread. If it spreads easily it may affect a lot of people. If it spreads easily but is not very severe, then there will be a lot of sick people but not many deaths. Still a big economic consequence.

How well do we understand it? Enough that several manufacturers have the raw materials to begin production of the vaccine, but that they are not sure it will be ready for the fall. So, at least some possibility of limiting the economic consequence, although a vaccination program is not without its own economic costs.

How well do we understand its consequences? Not terribly well as we are still trying to figure out why the severity in different locations has varied so much. Perhaps despite all the current complaints about how expensive our health care system is, having an overabundant system in which people can get at least urgent care fairly readily does have its advantages.

Yet, all in all, it does not seem to many average observers that it poses much of a threat. I can't say why the pandemic announcement didn't seem to get so much attention, but maybe it is an issue of words really mattering. Kids yell across the playground (or at least YELLED across the playground 35 years ago when I was one of them), "Sticks and stones may break by bones, but names will never hurt me." And yet, we all know that words do matter in some pretty important ways.

The words that first introduced us to so called "swine flu" several months ago made people behave very differently. We heard all the dire predictions We had school districts closing their doors and telling parents to keep children at home. At a professional conference I attended, we were missing almost everyone from one country. It seemed like a problem when we heard about it all the time.

The word pandemic, without supporting information to put it in context, can make people behave very differently. As far as economidw is concerned, it is not necessarily irrational behavior--that would be behavior that doesn't make sense in light of good information. Instead, the issue is whether people have the right information on which to make a decision in the first place. The Centers for Disease Control and Prevention in the United States have tried to provide information. I will be the first to admit that keeping up with all the information that may become available is a non-trivial, time consuming, and somewhat daunting task. Yet, if we do not do it, our behavior may end up seeming irrational and having some of the same consequences as purely irrational behavior if we don't get the information we need. I am certainly not immune to this--although if anything I could be described as taking the whole thing too lightly.

What Do We Need?

Still continuing with thoughts from Friday, Anthony McCarthy asked me to consider "what type of program would you build" before coming on the show.

From what I have discussed already, you can see that I can argue either side of whether we should have a single payer system. It is an open empirical question whether it would cost "less" to society to lose the choice of different financing options and possibly losing some treatment options or to gain the efficiency of having one set of rules and lower administrative costs.

However, the discussion does not end with the choice of financing. In fact, that is really only the beginning.

The second thing I would focus on is how the health care that is provided is organized. There was an interesting piece recently focused on variation in medical care practice around the country. One piece of the article was a discussion of ways of organizing care to get better outcomes at lower costs. The conclusion that was drawn that in care systems where one organization is responsible for all the care of an individual that usually promotes the kind of thinking that can lead to high quality efficient care. The key is that very few of the care systems in America have chosen to organize in this way. Could the government find ways to provide incentives for providers to organize care that way?

The third thing--in any system that ensures everyone (if we consider a mandate or consider health insurance as a right--which is different from considering health care or good health as a right since even those with health insurance doesn't guarantee access to care) there should be a combination of rights and responsibilities. While it is not always easy to establish what is a person's fault, some things (e.g. smoking) are obviously issues of choice and we should consider higher copayments or higher premiums for people with poor health that can be attributed to their behavior.

So, insure everyone, reorganize care, and provide incentives for people to be responsible. That would be my plan.

Insurance Exchange

One term that is floating around in the health care debate right now is an insurance exchange. The idea has been compared to the stock exchange in something I read. At the stock exchange you can trade all stocks. At an insurance exchange it would be an easy way to make available and, with the right information, compare different insurance products. It would also offer an opportunity for pooling risk.

What is interesting is how this is being portrayed on the two sides of the debate. On the one hand, some single payer advocates are saying that the for profit insurers will have an opportunity to exert too much control over the regulation of the exchange. They would set up a system by which they could market to and attract the best risks. That would leave the higher risks for the government. The government program would then become more expensive and in a few years the private insurers could make the claim that the government really can't manage a cost-efficient program.

On the other hand, you have a variety of parties making the argument that this is just a step to single payer. That the government will set rules to favor the government plan and that private industry will then be pushed out of the insurance market.

I'm sure that each is possible. If we don't go to a single payer plan immediately, it would be nice if we could actually create a plan by which we would have healthy competition. Mark Pauly in a recent editorial (NEJM, 2009; 360(22): 2271-3) suggested that if some people want to be insured by the government that should be a choice as long as the government plan plays by the same rules as the private plan. Similarly, as I mentioned yesterday, if different people want insurance plans that are more or less generous for certain types of care (e.g. not everyone wants a plan that covers chiropractic services) that should also be allowed.

Some Downsides of Single Payer?

So, today I had the pleasure of appearing on the Anthony McCarthy show (www.theanthonymccarthyshow.org) on WEAA (www.weaa.org) in Baltimore. I must say it was an interesting experience. My fellow guest, Kevin Zeese (imagine having two Kevins as guests on the same show) is a single payer advocate. Nothing wrong with that. As I have said beofre on this blog, there is a lot to be said for the single payer idea. Ther would be one set of rules. One set of paperwork. No advertisting among insurers. If it was run by the government and they could match the administrative cost rate for Medicare, we'd be looking at around 3%. In many ways it seems great. Proponents also point out that it would give ultimate choice of providers and facilities because everyone would accept the insurance.

That is all well and good. However, before we go too far down this path, let us assess what some reasons for being a little wary of this are. I am not advocating either side at this point. I am simply wanting to make sure that I (as I "think out loud") and you consider what this means.

First, if you don't like the benefits in the single payer plan you are stuck. Single payer proponents say that people really want choice of provider and are not so interested in choice of insurance product. In general, I would agree with the intuition behind that statement. However, some basic observation of the heterogeneity of preferences that people show suggests that we might want to allow for variable policies.

Second, the government is not known for paying a lot for health care. Providers may find that they are not making enough. Proponents of a single payer system note that many providers state that they approve of a single payer system when polled. They are probably frustrated with the administration and paperwork. They may not appreciate the payment they end up with.

Third, consider comparative effectiveness research. If the government uses it in a single payer system to prescribe the care that must be given that limits choice of approaches to treatment. It could be argued that this would also be true for private plans. However, if there were multiple private plans some may more or less strictly adhere to the evidence and consumers would have their choice among different plans with different levels of adherence.

So, if we were to go down this path we would have to think very carefully about whether people value choice of payments, whether people value choice of treatment, and how well providers would be paid. As a society, we may decide that a single payer system (because of the potential savings) does make sense. However, we may decide that the limits that are imposed cause problems that outweigh the savings.

Over the next two days I'll talk about some thoughts on the public option idea and what elements I think we need to combine to have a good plan--although it may make no one particularly happy.

Problems with the Definition of Comparative Effectiveness

The quote below is from a news posting on February 27. I received it in an email to AcademyHealth members (www.academyhealth.org). This is a professional organization of people interested in jhealth policy and health services research.

The quote I put below is from the summary of the budget and part of a document titled “A New Era in Responsibility”. It has been sitting and waiting for me to think about ever since then. Back on March 12, I offered a comment on the definition of comparative effectiveness. I am returning to this, as it is even more difficult to believe what was printed in light of our nation’s continued discussions of which of the several options we have for health care reform will be of greatest benefit to society and control costs best. Here is what was printed nearly four months ago:

"Building on the unprecedented $1.1billion included in the recovery Act for comparative effectiveness research, the Administration will continue efforts to produce state-of-the-science information on what medical treatments work best for a given condition. When coupled with electronic health records, these findings can form the basis for clinical decision support tools-distilling all available evidence on the outcomes of different treatment options into user-friendly pop-up alerts for physicians at the point of care. These findings can thereby enhance medical decision-making by patients and their physicians."

Note what this says. We are going to spend $1 billion of taxpayer money to further study which treatment works best. We really do still have some questions about what works best for some conditions, but we also know a lot about what works best for some things already. Do we need to spend $1.1 billion on this? Maybe.

In theory, we are going to combine this with electronic health records to make better decisions. Well, we still don’t have an electronic health records system. So far, what has developed has not been coordinated. I know that a lot of people are concerned about centralized decision making, but if we have multiple electronic health records systems that have to talk to each other we may find that we are little better off than we are now. So, spending $1.1 billion to gain information to combine with an as yet non-existent system seems problematic.

Finally, the last sentence talks only about enhancing patients’ and physicians’ decision making. It doesn’t say anything about the constraints that both will face. Supposedly “better” medical decisions that ignore constraints may lead to higher costs.

One might be left with the impression that a lot of what we are doing is for show and not actually going to save us money and not going to make a big difference in our ability to make the best decisions with constraints. To be “kind” to the statement from the budget we could say, “It is just politics to leave cost out for now.” However, this is not a time for subtlety. The best way to allow fellow politicians, bureaucrats and citizens to properly evaluate policy recommendations is to clearly state exactly what is meant.

Centralization of decision making?

Something to think about. What value is there to centralization of decision making--or at least of a part of the process?

Here is a short report in a statement from the Commonwealth Fund--a nonprofit foundation that aims to promote a high quality health care system.

"Despite overwhelming evidence that experiences during the early childhood years contribute greatly to later health and educational attainment, the United States lacks a clearly articulated policy to address the needs of young children and their families during this crucial period.

The U.S., however, is not the only country to struggle with the direction of early childhood policy, according to An International Comparison of Early Childhood Initiatives: From Services to Systems, a new Commonwealth Fund report by Neal Halfon, M.D., M.P.H., and colleagues. England, Canada, and Australia all started with similarly fragmented early childhood services, and families in these countries are facing similar pressures resulting from long hours at work, irregular work schedules, and limited child care options.

In the report, Halfon, the founding director of the University of California, Los Angeles (UCLA) Center for Healthier Children, Families and Communities, and his team detail each country's efforts to develop policies that they hope will produce lasting gains for their youngest citizens. The authors also consider the implications of these nations' experiences for the development of early childhood policy in the U.S.

As the report shows, the building blocks for the early childhood system of the future are already emerging. Approaches that align strategies across multiple levels of government—local, state, and national—and that integrate health, education, and family support services are proving to be the most successful. "

This is the type of argument that I have been making for years in a variety of contexts in the research and policy related work that I do. It started with work on the Healthy Start project in Baltimore City. If we take it as a given that we are likely to have some form of government public health intervention (and I realize that not everyone even agrees with that premise) then the even more difficult question is how to get multiple levels of government or even multiple agencies within the same level of government to cooperate. To share resources to use for a program. To share savings from a program that works.

In some cases we have to ask whether all the resources used by all levels of government are less under a new system. Unfortunately, our system is so fractured that we have great difficulty to align everyone's incentives. The report from the Commonwealth Fund mentions this as a key issue. I find it hard to believe that we can do that in the United States. I find it particularly hard to believe since there are so many different interests aiming for political power at each level and so many vested interests in not sharing this way. I find it particularly hard to believe since Americans--in general--are not big fans of centralization. However, we may consider at least some additional centralization--even if we only centralize discussion--to stop having to focus so much on individual winners and losers and to think harder about whether society wins or loses when we make decisions.

Social Entrepreneurship

This evening, I received an invitation to speak at a conference next April. Aside from thinking to myself--Gosh, that is a long time away--I have been thinking about the organization. I spoke at this organization's conference once before. I will probably accept the invitation again. The organization is called Unite for Sight (www.uniteforsight.org). I want to spend a little time thinking about social entrepreneurship.

A lot of the comments on my blog (at least on FB) have essentially been questions of whether we should leave things to private industry (or private charity) or regulate through the government. Social entrepreneurship is an interesting "in between".

Social entrepreneurship is using either charity or government dollars (most often charity) in ways that do not involve purchases to provide goods and services once. Rather, it represents money that is spent to build up capacity in a market. In the case of Unite for Sight, it is spending a money to provide incentives to develop the supply side of the market for some vision care services. For vision screening. For glasses--both for distance and for reading glasses. The key is that a market has not developed on its own but can exist. Resources need to be made available to bring the market into being to begin with.

Total "free marketeers" may still say--if the market is not developing on its own then we should just let things be. However, there are real limitations to markets operating efficiently all over the world. In the case of the mostly developing countries in which Unite for Sight and other similar organizations operate, there are failures in the market for credit to start businesses that charity can help to alleviate--but then will not be expected to continue providing. There are also failures in the provision of education for running facilities that charity can overcome.

We should consider for developing countries whether there are small market failures that small amounts of one time investment might help to alleviate and bring a private, self-sustaining market into being. If those who give money to charity want to make a sustainable difference, this type of intervention may be better than other similar interventions.

The $1000 those of us who are insured pay for the uninsured

I'm sure many of you saw a few weeks ago when there was a report on how much those of us with insurance pay on average (through premiums) for people without insurance. If not, here is a link to a story about the study
(http://news.yahoo.com/s/ap/20090528/ap_on_bi_ge/us_health_overhaul_uninsured). The key question is what should policy makers do about this.

One answer may be nothing. Sure it not great to have millions of uninsured people who have problems getting care but we as a society have tolerated that for a long, long time. In that case, there are still ways for those who are uninsured to get at least some care, we can say that we will still leave things to the market (which some favor), and we know how much it is costing people with insurance to subsidize. Some people fear that if we do anything other than leave it alone, those of us who are already insured will end up paying more. Maybe--but if we could pay more and have better protection from ever being uninsured some people might find that worthwhile?

Other than leaving things as they are, we can think about requiring everyone to be insured. That would probably relieve some of the burden on us. However even if everyone paid their own way through a premium, we would still have to subsidize some poor families. In that case, we would not see the $1000 through our premiums but I bet we'd see at least a part of the $1000 through our taxes. And, it might even be the full $1000--some people could pay at least part of the premium but the overall use of health care may go up as people had to pay less out of pocket.

We could also switch to a single payer system in which case the entire system would be tax based.

Not clear which of these would be best for those already with insurance--although people in general don't like change. Nevertheless, the more we stay with exactly what we have the more we will go on paying for others' care in ways that are not obvious and be at risk of loosing our own health insurance some day.

No easy choices.