In the first part of this series, I briefly outlined the character – the face if you will – of our health system. In this second part, I will touch upon the pre-reform finance of this system and why it is as it is. One of the issues to remember in this context is that some – by no means all – of the doom and gloom heard about health [insurance] reform and our health system is correct. What was passed in March was health insurance reform, not health care reform. There are many pluses to even insurance reform, but we must not make it out to be more than it actually is. Insisting that insurance companies spend 85 cents of each premium dollar on care is a plus, there are some data suggesting only a handful of companies actually do this (1); not being eliminated from your insurance plan because of pre-existing conditions (ie, because you are sick) is a plus; having a federal / state program to fall back upon if you haven’t been able to get insurance in the past is a plus; elimination of life-time benefit caps is a plus, especially if you become severely or chronically ill; the ability to keep ones adult children one one’s insurance plan until the child is 26 is a plus if the child is struggling through their first job without benefits; putting – at least verbal – emphasis on prevention is a plus. And there are more.
But this health insurance reform will likely not rein in costs to the extent promised unless more than verbal emphasis is placed upon prevention – exercise, diet, safety, as well as medical care. This assumes a measure of personal responsibility on the part of our fellow citizens as well as on the part of those of us in medicine. Additionally, there are estimates that the influx into the health system of previously uninsured people getting their preventive care will actually – at least in the short run – drive up health care costs. It is my best estimate that this is correct, but that the costs will decrease in the long term.
Having said this, and despite my preference for a single payer or “Medicare-for-all” option (with some alterations on the fee schedules, more on this later) I am a cautiously optimistic supporter of health insurance reform, also called the Accountable Care Act (ACA).
As mentioned in Part I of this series – and the formulation is not mine although I no longer remember where I first heard / read it:
In the US Health System, we episodically deliver extraordinary care, it is the routine provision of indicated care in average circumstances with which we have such trouble.
This plays out in many ways, one of which is in the realm of the financing of our system.
Let us begin to consider several issues, we have already noted the overall health expenditures and per capita expenditure indexed to GDP (Part I). We will look here more closely at why the amount consumed is as it is, and dissect more precisely where the resources go.
On first pass, health expenditures can be generally divided into two major categories: Public and Private Spending.
While the percentage of total health care expenditure financed from public sources has been and – to date – still is relatively low in comparison to other countries analyzed by the OECD (Organization of Economic Cooperation and Development), the absolute amounts are similar (2,3). For example, the percentage of GDP (gross domestic product) consumed by public spending for heath was 5.8% and 6.7% for the US in 2000 and 2005, respectively, while the median OECD values were 5.9% and 6.4%, respectively. In comparison to specific countries, rather than median values, in 2000 the UK, Italy, and Japan spent 5.9% each of their GDP on public expenditure for health; Canada spent 6.5%, not dramatically more than our 5.8%. In 2005, the percentage of our GDP consumed by public expenditures for health care was the same as for Switzerland, Japan, Italy, Canada, and New Zealand – 6.7% ( http://titania.sourceoecd.org/vl=3756626/cl=19/nw=1/rpsv/health2007/g5-2-01.htm , accessed 10/19/10) .
This is where we in the United States clearly “pull ahead” of our sister nations in the OECD. The definition used for Private Expenditure when the OECD compiles data falls into two broad categories: 1. premiums paid by individuals and families for private health insurance; 2. out of pocket spending for deductibles, co-insurance, and services not covered by health insurance (1). Private spending for health as a percentage of total health expenditure is shown (Figure 1). What is clear from this table is that our private expenditures are more than all of our sister countries, and we spent more than twice the median of the OECD for both 2000 and 2005. Total health expenditure – both public and private are shown for 1999 – 2000 (Figure 2) and 2005 (Figure 3); for each time period we are the top spenders. The per capita expenditures – private and public are shown as well (Figure 4). The question remains, as noted in Part I of this discussion, as to whether we get more value / benefit for the expenditures.
TOTAL SPENDING ON HEALTH CARE
One may reach a reasonable conclusion that we spend a significant amount more on health than our sister nations by looking at the above figures. However, to make this part of the discussion abundantly clear, our expenditures on a per capita basis, using purchasing power parity (PPP) dollars – to equalize the values of various currencies – were 134% and 132% higher than the OECD median for the years 2000 and 2005 (Figure 5). It is reasonable to ask why these figures are as they are. In Part I of this discussion, I offered a discussion showing that the quality outcomes for our system were not generally better than our sister nations, so we cannot argue, I think, that costs are higher because we provide higher quality medicine. But if the spending is not driven by higher quality, then what might be doing so ? To this we will now turn our attention.
WHAT DRIVES OUR EXPENDITURES AND COSTS ?
This issue has been studied rather nicely. Do we spend more because our health care use is significantly higher, our administrative issues – more on this one later – more complex, our population older, we practice more defensive medicine because of our fear of malpractice, and / or because our waiting lists are long (3) ? Two major causes were found: the higher GDP, which accounts for 56% in spending variation, and the higher prices charged (2). So a part of the answer to this is that if a county has more money to spend, it is spent on health. But a second part is that we in the US are simply charged more for the services we receive. Why is this and from where does this stem ?
The figure showing Medicare Per Capita Spending in five regions of our country, from 1992 through 2006 is informative (Figure 6). Is it possible that the costs of providing quality health care to the elderly in Miami are so much higher than are those in Salem, Oregon or San Francisco ? This seems hard to believe and, in a recent New Yorker piece and followup (4, 5), the reasons for at least some of this variation was laid bare.
There are at least three reasons I can find that helps to explain our expenditures: Administrative Cost; Pharmaceutical Costs, and End-of-Life Care.
Administrators are indispensable. In any system, it is imperative to have individuals to keep the books, make sure workers are paid, ensure regulations are followed, see to it that the supplies needed to perform are available, and make sure the facility – whatever the facility is – is kept up. But how much administration is enough ? Does having more administrative assistance ensure a higher quality of care ? For-profit health care institutions do tend to have more in the way of administrative assistance, but their quality of care is no higher than those institutions without this level of administrative assistance. Private insurers have a significant administrative load to keep track of a applicant’s eligibility, various co-payments , referral networks, and approval requirements.
From 1991 through 2005 total US health care administrative costs – private and public – were reported to have increased from 6.6% to 7% (Figures 7, 8, 9). Analyzed by payer, during this same time period, private sector administrative costs went from 16.8% to 14%, while public administrative costs went from approximately 3.2% to 5%. Even this, roughly, 3- to 5-fold difference between private and public insurance administrative costs may significantly underestimate the administrative load placed upon our health system by the privates (6).
It is the manner by which administrative costs are calculated that determines the amount spent on managing the health plan. According to the Centers for Medicaid and Medicare Service (CMS) this calculation is simply the direct health care expenditures divided by the administrative costs and net cost of the private health insurance plan (http://www.cms.hhs.gov/NationalHealthExpendData/02_NationalHealthAccountsHistorical.asp last accessed 10/17/10). Wollhandler and colleagues (6) used a more intense analysis of administrative cost to determine the actual, as opposed to the reported, percent of the health care dollar spent on administration. While the reported percent of total expenditure (public and private sectors) assigned to administration is 6.6% to 7% (see above), and that attributed to private insurance is 14% to 16.8%, these investigators suggest that the real percentage may approximate 30%. The discrepancy between these figures is accounted for because of excluded administrate costs: costs not normally added into the administrative cost bucket. These include physician time spent on administration, clerical time spent on administration – including billing / collections, payment variance, and post-rejection appeals, as well as the “legions” of insurance workers whose job it is to prevent “medical losses” (the term insurance companies use to refer to physician and hospital payment). For example, the percentage of the health work force – excluding insurance company workers – that are administrative / clerical increased from 18.2% to 27.3% between 1969 and 1999 (Figure 10). When one now looks at the number of employees per 10,000 enrollees hired on by selected insurance companies in the US and Canada, the results are significantly different: in the US it ranges between 12.2 and 31 employees per 10,000 enrollees, while in the Canadian Province plans the numbers are 1.2 to 1.4 (Figure 11). When the total costs of administration are added up, for US private companies, they are 31% of benefit dollars, compared to the Canadian system’s 16% (Figure 12).
Is this too much ? As stated above, we need administrators and – as the Division Chief of Critical Care Medicine I am one myself for a portion of my time – these individuals need to be funded. But is the $ 290 to $ 320 billion per year we spend on administration well spent ? Could these funds perhaps be better utilized ?
In general, pharmaceutical expenditures per capita and as a percentage of GDP are higher in the US than in other OECD nations (Figures 13 and 14); this despite significant funding provided by the US Government through its funding mechanisms – for example the National Institutes of Health – in the development of pharmaceutical agents. In 2006, under then-President GW Bush, Medicare Part D was launched, an attempt – in the best light – to provide drug coverage for our elderly citizens. There were, however, “issues” when the program was being developed: the administration would not allow the use of government purchasing power to negotiate less expensive prices for the drugs, nor would our leadership allow the importation of cheaper drugs from Canada. Additionally, the Conference process, where differences in House and Senate bills were ironed out, was held in secret, except that pharmaceutical industry members were allowed to attend (7).
There were other concerns noted in the development process of this drug plan. Then Republican Representative from Louisiana Tauzin – now lobbyist for the Pharmaceutical Researchers and Manufacturers of America (PhRMA) – co-authored the Medicare Part D bill while negotiating a $ 2 million per year job as a lobbyist with PhRMA (7). Additionally, the head of CMS at the time, Thomas Scully understated the projected cost of the program by $ 134 billion (7); when the Chief CMS Actuary objected, this professional was reportedly threatened with termination if the more costly estimate was shared with Congress; after the legislation passed, Scully reportedly resumed his career as a health-care lobbyist.
Aside from the seemingly corrupt manner with which the bill was developed and implemented, and the cost to the national treasury, there are significant costs – the doughnut hole – that Medicare beneficiaries were left with until the passage of the ACA in March of 2010. The standard benefit for those citizens with an income greater than 150% of the federal poverty level had a $ 250 deductible and then a benefit “gap” – the doughnut hole – for drug spending between $ 2,250 and $ 5,100 (8). Thus the beneficiary was responsible for drug costs of about $ 2,850; for those with moderate and high drug costs, there was full-price payment out-of-pocket for months. While this has been corrected with the passage of the ACA in March, one marvels at the level of apparent corruption that was allowed / encouraged during the work-up of this bill by the GW Bush administration and the Republican congress.
Did the doughnut hole compromise patient care ? At least one review suggests that this is so. Among elderly with at least three chronic illnesses, significantly more prescriptions we left unfilled or were delayed amongst patients who had either no coverage or Medicare Part D – 35% and 25% – as compared to either employer coverage, VA coverage or “other” coverage (Figure 15) – 11.9%, 15.6% and 18.1% respectively. Additionally, the percentage of these individuals who spent more than $ 300 per month out of pocket on drugs was, for Part D and none, 11.1% and 13.6% respectively, as compared to 7.8%, , 6.6%, and 11.2% for employer, VA, or “other” coverage, respectively. Statistical significance was achieved as noted in the figure (Figure 15). There were, not-withstanding these problems, other data suggesting that, despite the corruption and the increased out of pocket expenses noted, there were some bright areas in Part D. In 2007, Medicare spending was $ 432 billion, Part D comprised $ 50.1 billion (11.6%). The Medicare Part D spending was 10% less than in 2006 and the number of prescriptions were up by 6.5%; compared to 2006, the patient out-of-pocket expenses decreased by 16%. These improvements were concentrated amongst low-income patients (9).
Not-withstanding this improvement, 60% of patients were unaware that a benefit gap existed; 36% of these patients, when faced with the gap, exhibited coping behavior that included switching to a cheaper drug or skipping or splitting pills. Another 15% decreased adherence to their physician’s recommendations by not filling the prescription or, again, by splitting or skipping pills. Seven percent of patients faced with this gap borrowed money to pay for their medications or went without some necessity so that they could obtain the drugs (10). A Canadian study showed that when the “market” was allowed to dictate – a co-pay based upon resources available – who would or would not get their essential drugs (defined as insulin, anticoagulants, antihypertensives, anticholesterol agents, diuretics, antidysrhythmics, aspirin, thyroid medication, antidepressants, and so forth), there was a significant decrease in the number of essential drugs / day taken, and an increase in adverse events – defined as first occurrence of acute care hospitalization, long-term care admission, or death – observed (11). In the year following the imposition of the co-pay, there was a 9% decrease in essential medication use by the elderly and a 4% decrement among those on welfare. Adverse events rose 117% in the elderly and 97% in those on welfare. Finally, emergency department visits rose 43% in the elderly and 78% in those on welfare (11). All of this suggests that when looked at in either a medical or economic manner, limiting essential medications as a cost saving method is short-sighted.
End of Life Care
End-of-life care and the problems that go with it are not uniquely American issues, although it does sometimes seem like it. In France in 2003, a 20 year old fire-fighter was involved in a motor vehicular crash that left him quadriplegic, blind, and mute. He was able to communicate only with thumb movements; through this method of communication, he asked his mother and the physician caring for him to stop further treatment and withdraw therapies that were being provided. As withdrawal of therapy was, at least at that time, illegal, euthanasia was performed, utilizing intravenous barbiturates and potassium chloride injections; charges were brought against both the physician and the patient’s mother (12). While I am not suggesting that euthanasia is the manner in which such problems should be dealt with, it is clear that if society does not deal with end-of-life – and quality of life – issues, individuals will find a way to handle them, for better or worse, themselves. Although there are many examples we could draw upon to make the point about this issue in the United States, the Schiavo issue (see this Blog, August entries for details) of several years ago makes the point nicely that we have difficulty with death. Further, the outrageous misrepresentation by some in the Republican Party and on the political right of provisions in the ACA, in this context the provision that would have funded patient – family – physician discussions about end-of-life issues, only lays bare the extreme difficulty we have in having this discussion. The political right famously called this intimate discussion “death panels”; of course, nothing was farther from the truth. As an intensive care physician, I have these discussions almost daily with patients and families. People want to know what is happening to them, the chances for recovery, and what, in this context, are their options. Sometimes, the best option is to fight hard for recovery; sometimes, the best option is to provide comfort care and allow death to take control. However, without discussion there is no way that family and patient can know the options; without discussion, there is no way that the physician can know the patient’s and family’s wishes, fears and desires.
Many, but not all, hospital deaths occur in our ICUs. A decade ago, in a six-state study (13) involving Florida, Massachusetts, New Jersey, New York, Virginia and Washington there were 552,157 deaths over a one year period. These states were chosen for the study as they represent 22% of the U.S. population and had high-quality hospital discharge data. Of these deaths, 38% occurred in the hospital, and 22% – 59% of all hospital deaths – occurred after an ICU stay. About 80% of hospital end-of-life (terminal) spending is consumed by in the ICU. The Non-ICU terminal median cost was about $ 7,919, while the length of stay was about 6 days; terminal hospitalization associated with an ICU stay had a median cost of $ 12,392 and a median length of stay of 7 days. Only in four states were the ICU LOS data separated out; these were a median LOS of 4 days. In has been noted in polling data (14) that about 90% of Americans would prefer to die at home surrounded, one assumes, by loved ones. Yet about 22% of us – if these data are extrapolated to the entire country – will instead die receiving the highest technology medicine available. To some extent this is no doubt due to the fact that we can’t always tell that an individual is going to die “this time” and thus hold off of all the high technology treatments. But there is some component, as well, that results from our unwillingness or inability to confront death and discuss it.
There were many things mishandled by our leadership in Congress and the White House during the health care reform debate. But one of the most disappointing was the manner in which the end-of-life “discussion” was caricatured by turning it into shrill warnings about “death panels”. The people who did this knew they were lying, and they did us all a great disservice.
It is recognized that 26% of federal Medicare spending occurs in the final year of life (15). It is stated – tongue on cheek – by those of us who care for the most seriously ill, that if we KNEW when the final 30 or 60 or 90 days of life were coming around, we wouldn’t do this. While we take pride in the United States that we provide cutting-edge technology for our patients, there is clearly a point where this is no longer appropriate or desirable, a point where quality of life issues should take precedence over extended quantity. While a difficult issue, it is one, as mentioned above, that mandates discussion.
And by not discussing, by not dealing with this issue – among others – what are the consequences ? Who says this is not sustainable ?
BY THE NUMBERS
In 2005 there were 45 million uninsured citizens in our country (16) and about two-thirds of these were employed. This is only part of the story, as there is another larger population of us that are underinsured – estimated at about 50 million people (17). These underinsured have multiple coping mechanisms including skipping appointments, leaving prescriptions unfilled, postponing laboratory studies, and so forth. Allowing “market forces” to handle this issue have not worked; the price of insurance premiums has risen 2 to 4 fold more than either the inflation rate or working people’s wages (Figure 16). To assert that “the market” must be allowed to resolve the issue of the uninsured and underinsured is, at least to this writer’s eye, an outstanding display of moral cowardice and inhumanity.
Uninsured people receive too little medical care, too late (18). Even adjusting for age, gender, tobacco use, and education, the uninsured are sicker when they present and die earlier, and they receive poorer care when they are hospitalized; the lack of insurance alone increases risk of death by about 25%, all other things being equal (18). Amongst uninsured people with medical illnesses, there was a significantly lower level of diabetic care – recommended eye and foot examinations, immunization, checks of hemoglobin A1c and cholesterol; renal failure patients started dialysis at a later stage of the disease; those with HIV disease were less likely to receive effective drug therapy; and there was less likely to be screening for hypertension or use of prescribed medication for blood pressure control if hypertension was diagnosed.
This tendency also holds for surgical illness. While the insured meet their goals for cervical, breast, and colorectal cancer screening 90%, 70%, and 50% of the time, respectively, for the uninsured these numbers are – respectively – 77%, 52%, and 29%. Uninsured and Medicaid patients have more advanced cancers at the time of diagnosis, and / or lower survival rates for breast, colon, cervical, and prostate malignancies as well as for melanomas. The uninsured are more likely to present with perforation of their appendix at the time of diagnosis of appendicitis, and those uninsured patients with an abdominal aortic aneurysm are likely than the insured to suffer rupture (18).
The risks of being without access to high quality preventative health care are, thus, significant.
The question continues to be: Is this the best we can do ?
We will touch on this question in Part III.
1. Iglehart JK Defining Medical Expenses, N Engl J Med 2010; 363:999-1001 http://www.nejm.org/doi/full/10.1056/NEJMp1008571 accessed 6 October 2010
2. Anderson GF, et al: It’s the Prices, Stupid: Why the United States Is So Different from Other Countries. Health Affairs, 2003; 22 (3):89-105
3. Anderson GF, et al: Health Spending in OECD Countries in 2004. Health Affairs, 2007; 26 (5):1481-1489
4. Gawande A: The Cost conundrum – What a Texas town can teach us about health care. The New Yorker, June 1, 2009 http://www.newyorker.com/reporting/2009/06/01/090601fa_fact_gawande accessed 6 October, 2010
5. Gawande A: The Cost Conundrum – Redux. The New Yorker, June 23, 2009 http://www.newyorker.com/online/blogs/newsdesk/2009/06/atul-gawande-the-cost-conundrum-redux.html accessed 6 October, 2010
6. Woolhandler S, Campbell T, Himmelstein DU: Costs of health care administration in the United States and Canada. N Engl J Med. 2003;349:768-75. (Pus errata)http://www.nejm.org/doi/pdf/10.1056/NEJMsa022033 accessed 6 October, 2010
7. Slaughter LM: Medicare Part D: The product of a broken process. N Engl J Med, 2006;354:2314 – 2315 http://www.nejm.org/doi/full/10.1056/NEJMp068116 accessed 6 October, 2010
8. Stuart B, Simoni-Wastila L, Chauncey D: Assessing the impact of coverage gaps in the Medicare part D drug benefit. Health Affairs, 2005 (web exclusive) W5-167 – W5-179 http://content.healthaffairs.org/cgi/reprint/hlthaff.w5.167v1 accessed 6 October, 2010
9. Goldman DP, Joyce GF: Medicare Part D – A successful start with room for improvement . JAMA, 2008;299:1954 – 1955 http://jama.ama-assn.org/cgi/content/full/299/16/1954 accessed 7 October, 2010
10. Hsu J, Fung V, Price M, et al: Medicare beneficiaries knowledge of Part D prescription drug benefit benefits and responses to drug costs. JAMA, 2008;299:1929-1936 http://jama.ama-assn.org/cgi/content/full/299/16/1929 accessed 7 October, 2010
11. Tamblyn R, Laprise R, Hanley JA, et al: Adverse events associated with prescription drug cost-sharing among poor and elderly persons. JAMA, 2001;285:421-429 http://jama.ama-assn.org/cgi/content/full/285/4/421 accessed 7 October, 2010
12. Lemaire FJP: A law for end of life care in France ? Intensive Care Medicine, 2004;30:2120
13. Angus DC, et al: Use of intensive care at the end of life in the US. Crit Care Med, 2004;32:638
14. Fields MJ, Casel CK: Approaching death – Improving care at the end of life. Washington, DC, National Academy Press, 1997
15. Hoover DR, Crystal S, Kumar R, et al: Medical Expenditures during the Last Year of Life: Findings from the 1992–1996 Medicare Current Beneficiary Survey. Health Serv Res. 2002;37:1625-42
16. US Census Bureau. Income, Poverty, and Health Insurance Coverage in the United States: 2005. Suitland, MD, US Census Bureau, 2006
17. Schoen C, et al: Insured but not protected. Health Affairs, 2005;(supplement web exclusive):W5-289 – W5-302
18. Sarpel U, Vladeck BC, Divino CM, Klotman PE: Fact and fiction: debunking myths in the US healthcare system. Annals of Surgery. 2008;247:563-569