Policy, Politics, and the Intensive Care Unit

FIGURE 1.1 Estimated effects of the Affordable Care Act (ACA) on the uninsured. (Data from Congressional Budget Office. Insurance coverage provisions of the Affordable Care Act. CBO’s March 2015 Baseline. Available at: https://www.cbo.gov/sites/default/files/cbofiles/attachments/43900–2015–03-ACAtables.pdf. Accessed May 2015.)

Underinsurance and the ICU

In another very important—and increasingly recognized—respect, the ACA will also not meet the standard of universal health care, and that is its failure to eliminate underinsurance. Though this is a problem no doubt predating the law, underinsurance is predicted to remain an important problem.

As long as there has been health insurance, there has been underinsurance. The concept, however, has been variably defined. For instance, in one study, individuals were defined as underinsured if they met one or more criteria for high financial exposure to health care costs (32). This included having out-of-pocket health care expenses 10% or more of income (or 5% for those with low income), or having out-of-pocket deductibles equal to or greater than 5% of income. Using these criteria, this study found that 20% of the nonelderly insured adult population—or 25 million people—were underinsured in 2007, which was up to 60% from 2003. The study further found that underinsurance affected both those with low income and those in the middle class. It found that the underinsured were more likely to avoid health care and to have increased likelihood of “financial stress” associated with health care, like putting health care costs on their credit card. In more recent years, underinsurance seems to have worsened. Indeed, a more recent study with similar methodology found that in 2014, 23% of nonelderly adults—or roughly 31 million people—were underinsured, which was approximately twice as many underinsured adults as 2003 (33).

One specific “cause” of underinsurance is health insurance plans that feature high levels of “cost sharing,” or out-of-pocket money paid for health care at the time of use. The typical forms of cost sharing include co-payments (fixed fees for services or drugs), co-insurance (a percentage of the service or drug that is paid out of pocket), and deductibles (an amount that must be paid out-of-pocket before insurance coverage begins). In employer plans, the past decade has seen a substantial increase in such cost sharing. Over the last 9 years, the percentage of workers with single coverage with an annual deductible increased from 55% to 80%, while the amount of the average deductible approximately doubled (34).

Underinsurance and high out-of-pocket financial exposure more generally is problematic insofar as it deters individuals and families from seeking needed health care, and because it may unfairly burden the sick and those with low incomes (35,36). But is underinsurance a problem for CCM? To answer this question, we can first return to the issue of how financial barriers to care might result in avoidance or delay in seeking care, which may have important ramifications on critical care outcomes. Though one might think that those with potentially emergent symptoms would be insensitive to the costs of care, this does not appear to always be the case. Let us imagine, for instance, that the hypothetical patient we discussed in the previous section was insured, but had a high-deductible or emergency department co-payment. Might this affect her decision to seek care?

There is evidence that it might. One recent study compared a cohort of individuals who were transitioned to high-deductible health plans with a cohort that remained in a traditional HMO health plan (37). Among those of low socioeconomic status, the switch to the higher cost-sharing program was associated with reductions in “high-severity” Emergency Department visits, which included diagnoses like acute asthma exacerbations and head wounds. In other words, high cost sharing led some to avoid going to the emergency department, even when they needed it. Additionally, we earlier discussed how the uninsured with a myocardial infarction were more likely to experience delays in hospital presentation (22). This study additionally found that those who were insured but who had “financial concerns” were also more likely to experience delays in presentation. In other words, financial out-of-pocket exposure may result in delayed or foregone care, even in the case of emergencies.

Cost sharing may also adversely affect individuals with chronic pulmonary disease. High cost sharing discourages adherence to medication, including in those with chronic disease (38,39). One study found that when inhaler medications were subject to deductibles and co-insurance, the rate of emergency hospitalizations for COPD, asthma, and emphysema increased significantly (40).

Financial harm is another potential problem with underinsurance. The high costs of ICU care may impose severe financial strain for families at a particularly vulnerable time. A patient admitted to the ICU, for instance, might expect that the cost of the hospitalization would rapidly reach his or her increasingly high deductible.

Though the ACA provides some important protections (like outlawing caps on benefits, capping out-of-pocket costs, and preventing discrimination against those with pre-existing conditions), it does not eliminate the problem of underinsurance, for several reasons. First, it will do little to contain ongoing trends toward higher cost sharing in the employer insurance market. Furthermore, it may actually encourage this trend through the so-called Cadillac tax (41). The Cadillac tax was meant to address longstanding concerns that the tax exemption for employer-provided health insurance has encouraged the growth of overly generous plans. The tax, which was initially meant to go into effect in 2018 but has now been delayed, takes the form of a 40% excise tax—payable either by the insurer or the employer—on insurance plans that exceed a specified threshold (41). As a result, some employers have already moved to limit benefits in an attempt to evade hitting that threshold (41,42). Because the threshold is tied to overall inflation (and not to health care inflation, which is higher), one group has predicted that the number of those affected by the Cadillac tax will “grow rapidly over time,” reaching 75.8% of all insurance plans for families by 2029 (43). Higher levels of cost sharing will be one way for insurers and employers to avoid paying this tax.

Second, the individual market insurance plans available through the ACA exchanges also incorporate substantial levels of cost sharing. The plans are divided into four metallic tiers—bronze, silver, gold, and platinum—that have actuarial values of 60%, 70%, 80%, and 90%, respectively. The out-of-pocket maximum for these plans can be as high as $13,200 for a family plan (though this depends on income), therefore requiring potentially high levels of cost sharing in the form of co-payments, co-insurance, and deductibles (44). For instance, in one recent survey, the average Bronze Plan had a medical deductible of $5,372 and a prescription drug deductible of $465 (44). Though there are cost-sharing subsidies for those making less than 250% of the federal poverty level, the burden of cost sharing may remain high for many middle-class families. Underinsurance will persevere; with financial barriers to care still in place, the United States will continue to lack a fully universal health care system.

Long-Term Care

The ordeal of surviving critical illness can, in some instances, leave patients profoundly disabled. Survivors of critical illness may be left with deficits in pulmonary function, neuromuscular strength, and cognitive capacity (45). They may face significant psychiatric sequelae including posttraumatic stress disorder, anxiety, and depression (45). Finally, they may have substantial, ongoing requirements for long-term care, whether in a facility or at home (45). Therefore, policy considerations around the question of long-term care should be central to practitioners and investigators in the intensive care field.

Simply stated, there is no “system” of long-term care in the United States. This reflects the historical development of long-term care in the United States, as recounted by Smith and Feng (46). In the 19th century, long-term care in the United States was provided in the tradition of the repressive English Poor Law (a centuries old system of relief for the poor), which essentially meant coupling public disgrace with the provision of benefits (so as to deter others from receiving such care) (46). In the early 20th century, however, long-term care recipients were progressively transferred to private boarding homes, which became the precursors for today’s for-profit skilled nursing facilities (and which were sometimes part of publicly traded corporations) (46). The passage of Medicaid was another crucial pivot in this history. By providing a huge influx of resources for long-term care, the law resulted in a large increase in the number of nursing home beds, while simultaneously fostering the “medicalization and institutionalization” of long-term care (46).

Today, Medicaid remains the only real safety net for long-term care expenses for all but the very wealthy. According to one review, for instance, the cost for a nursing home for those paying out-of-pocket averages $75,000 a year (47). Those who are neither impoverished nor rich must, therefore, “spend down” their resources if they require long-term care services—until they reach a state of impoverishment—at which point Medicaid can be accessed. The ACA originally included a provision—the CLASS Act—which would have established a limited long-term care benefit. However, this would have been a voluntary program that provided only a limited set of benefits to a small section of the population (48). The benefits would have only partially reduced the costs of long-term care, and so would not have changed the fundamental dynamic of the “spend down” (48). Particularly as a result of its “voluntary” structure, the underlying finances were found to be profoundly flawed; given the impossibility of repairing the provision legislatively, the act was dropped altogether by the administration (49).

In contrast to the US experience, however, other nations have moved in the direction of providing long-term care as a universal benefit, including Germany (47,50–53), Japan (47,51), and the Netherlands (47). Despite the fact that these countries, unlike the United States, have a universal benefit for long-term care, they have public spending on long-term care that is only either slightly more (Japan) or slightly less (Germany) than the United States (51). These programs often also aim to provide long-term care services to patients in the setting of their home and community. Germany, for instance, encourages and financially supports care provision by family members (53).

To conclude, it is entirely true that the ACA may help many to avoid financial ruin because of an ICU admission. However, it may do little to help our patients who survive the ICU and go on to require long-term care. For those individuals, having to spend down the entirety of one’s life savings to a state of pennilessness can perhaps fairly be compared with the stigma and scorn heaped on the unfortunate in the era of Poor Law–inspired long-term care.


The cost of health care is by no means a novel concern in the United States. Indeed, in 1927, a “Committee on the Cost of Medical Care” was formed to study this very problem (54). However, there can be no doubting that the problem of overall health care expenditure has reached a new state of urgency in more recent decades. From the comparative perspective, for instance, overall health care expenditures in the United States remain profoundly higher than other developed countries: in 2012, 16.9% of GDP was spent on health care in the United States, as compared to 11.6% in France, 10.9% in Canada, 11.3% in Germany, 10.3% in Japan, and 9.3% in the United Kingdom (55). These differences are particularly striking when we consider that these other nations spend less while covering the entirety of their populations.

However, it is true that the past 5 years has seen growth in US health care expenditures that is very low by historical standards (56). This slowdown in growth has been attributed in large part to the so-called Great Recession, though some provisions of the ACA have also been credited (56). In the long term, however, the ACA lacks potent cost-saving measures. Perhaps the ACA’s greatest inadequacy in this regard is its failure to reduce our exceptionally wasteful spending on health care administration. Overall, data point to a rising proportion of health care dollars devoted to administrative costs, itself the result of our uniquely fragmented, multipayer health care system. For instance, from 1969 to 1999, the percentage of the US health care labor force devoted to administrative activities rose from 18.2% to 27.3% (57). By 1999, expenditures on health administration were $1,059 per capita in the United States, as opposed to the $307 per capita spent by Canada. Similarly, the proportion of total hospital costs spent on administrative activities increased from 23.5% in 2000 to 25.3% in 2011, even while they remained roughly stable in Canada (Fig. 1.2 gives greater detail on international comparisons) (58). Perhaps, more remarkable is the difference in administrative costs among nations: 1.43% of GDP goes to hospital administration in the United States, as compared to 0.41% and 0.51% in Canada and Scotland, respectively (58). Canada and Scotland both have single-payer systems with globally budgeted hospitals (in which hospitals receive a single payment to cover all operating expenses), allowing them to avoid not only the expense of interacting with multiple payers, but also eliminating patient-by-patient billing entirely (58). Nothing in the ACA will increase the administrative efficiency of the US health care system; doing so would require eliminating the problem of multiple payers and, like Canada and Scotland, globally budgeting hospitals.

FIGURE 1.2 Total hospital administrative costs and spending, 2010. As per study authors, the numbers for Scotland and Wales are based on National Health Service hospitals only. Additionally, they note that the data for England, Scotland, and Wales are for April 1, 2010, to March 31, 2011. Data for France and Germany are not presented here given that data on these two specific metrics were not available for these two countries. The data for English share of GDP are inclusive of both NHS Trusts and Acute NHS Foundation Trusts. (Data from Himmelstein DU Jun M, Busse R, et al. A comparison of hospital administrative costs in eight nations: US costs exceed all others by far. Health Aff (Millwood) 2014;33:1586–1594.)

The ACA does include a variety of other provisions that, it is hoped, will result in major savings. However, whatever their benefits and downsides may be, there is no strong evidence that many of these changes—for example, the expansion of the electronic medical record or the promotion of preventive medicine—will produce major cost savings (59,60).

As we will argue, an NHP could, conversely, facilitate substantial savings that could then be redirected to providing full coverage to both the uninsured and the underinsured, as well as the creation of new benefits (e.g., long-term care), without changing overall health care expenditures. Before we turn to the details of such a program, however, we will briefly discuss the health care system problems faced by other nations, and then explore some important policy issues specific to the ICU.

International Considerations

A comparative discussion of the health policy problems faced by other nations is well beyond the scope of this chapter. However, it does seem important to briefly emphasize that while the problems of the US health care system, as we have outlined, are to some extent unique to the country, they have important parallels, in both high- and low-income nations, throughout the globe.

However, before briefly discussing areas of overlap, we should first recognize the obvious divergence between the United States and other high-income nations with respect to health care: many industrialized nations have universal systems of health that are inclusive of either all or almost all of the population. A wide variety of explanations have been offered for why the United States failed to follow the developed world’s lead in enacting a system of universal health care, and include such features of American political economy as the politics of race and the strength of powerful health care “stakeholders” (4). Navarro (61) emphasizes the importance of the labor movement (both through its unions and mass political parties) in the genesis of universalist national health programs in Europe. In any event, whether through the “Bismarck” model of health care reform (e.g., the social insurance model in Germany), the “Beveridge” model (e.g., the national health service in the United Kingdom), or some combination or variation thereof, systems of universal health care emerged in Europe and elsewhere in the industrialized world, mostly during the post–World War II period. To some extent, these developments represented a legal commitment to the notion of a right to health care that was proclaimed in such postwar documents as the constitution of the World Health Organization and the United Nations Universal Declaration of Human Rights.

At the same time, however, the universality of these systems has never been complete, and in some instances it has come under significant threat in recent years. For instance, though these systems have been very successful in facilitating universal access to care, they have not always eliminated the problem of underinsurance. Israel, for instance, has raised co-payments to such an extent that they now represent “a major bar to accessibility” to health care, especially for those with low incomes (62). In the Netherlands, meanwhile, concerns have been raised that health system privatization represents a turn away from universalism and a step toward tiered, inequitable access to health care (63). Similarly, some have argued that the 2012 Health and Social Care Act, passed by a Conservative-led government in the United Kingdom, is furthering the privatization of the English National Health Service and taking it away from its universalist foundations (64).

In the wake of the economic crisis of 2008, under the economic policy of austerity, the universality of the health care systems of many European nations has been the object of a more direct attack. Especially in the nations of Southern Europe (but also elsewhere, as in Ireland), the policy of health care austerity has meant reductions in insurance coverage, disrupted access to health care, and increases in co-payments (65). “The erosion of health coverage in a time of economic crisis across hard-hit countries is worrying both in terms of population health and for the future of the welfare state,” as one observer puts it, also noting that “the universal nature of health systems has been consistently undermined, while demands for such publicly provided services are heightened” (65). Such changes have translated into a substantial rise in “unmet medical need” in nations like Greece (66). The onset of austerity policies has also been associated with increases in the suicide rate in Greece (67).

Low- and middle-income countries also face a range of challenges of their own. In particular, lack of access to quality health care remains a grave problem. The Indian health care system, for instance, is characterized by high out-of-pocket expenditures, regressive financing, low rates of insurance coverage, and deep inequalities in access along the lines of class, caste, and rural/urban geography (68). China, meanwhile, “turned its health system on its head” during the years of free-market reform in the 1980s: public funding for hospitals and rural health professionals evaporated as health care became increasingly an entrepreneurial endeavor (69). Out-of-pocket health care expenses became unaffordable for many, resulting in avoidance of health care (70,71). Recent reforms (69) may improve access for many Chinese, but comprehensive universal health care, without financial barriers to care, remains an aspiration in both nations.

In short, progress toward universal health care has proceeded very unevenly; even when it is achieved, it is by no means infallible. Moreover, the problems of being uninsured and underinsured, though particularly marked (among developed countries) in the United States, represent common weaknesses in the health systems of countries throughout the globe.

We now turn to health policy issues specific to the ICU.


In 1952, Denmark was in the throes of a terrible epidemic of poliomyelitis. By late August, Blegdam Hospital had lost 27 patients to respiratory failure resulting from neuromuscular weakness (72). The hospital’s chief physician, faced with the looming death of yet another patient—a 12-year-old girl—requested assistance from the anesthesiologist Björn Ibsen. Ibsen famously proceeded to recommend a tracheostomy and to carry out manual ventilation, thereby aborting her death (72,73). After this historic success, the protocol was soon repeated on all patients dying of respiratory failure from polio in the hospital (73). The life-saving capacity of this intervention was undeniable: the mortality rate fell from 87% to less than 40% (72).

To some extent, the “birth” of intensive care in Blegdam Hospital might be conceived as a sort of golden model of the potential beneficial effects of intensive care medicine. Patients were facing almost certain death from a condition that might be entirely reversible; with the temporary application of invasive, intensive medicine, lives were saved.

Juxtaposed to this inspiring story, however, is a more critical viewpoint that has emerged in the intervening decades. For instance, as one of us has previously studied, the rapid proliferation of critical care units in the United States was to a significant extent driven not by perceived needs and benefits, but by profit-driven corporate enterprises (74,75). Even putting that issue aside, it is increasingly recognized that the ICU might actually be harmful for some patients. To some extent, a glimpse of that reality may have already been apparent at the time of the 1952 polio epidemic. Bion and Bennet (76), for instance, have noted that even at the time, it was recognized that though “the new system of management produced many more survivors, it delayed death amongst patients destined not to survive”. Some now see the ICU as representative of the worst of modern medicine: it is conceived as a highly invasive, enormously expensive site where dying is prolonged at the price of profound and unnecessary suffering. One can invoke the (sadly not unfamiliar) sight of a frail, elderly patient with advanced Alzheimer dementia, dying of multi-organ failure, chained to a respirator, invasive tubes, and central venous lines.

An essential goal of ICU policy is to steer the system away from the nonbeneficial interventions toward the beneficial. Achieving this goal is not always possible: Ibsen (76) could not, for instance, have predicted who was “destined not to survive”; he had to try and save all. But as CCM consumes increasingly large proportions of both health care and overall societal resources, it is clear that much more can and should be done to improve the rationale and appropriate use of these services.

Halpern et al. (77,78), for instance, have documented the rising resources devoted to intensive care medicine in two important papers. First, they noted that between 1985 and 2000, even though there was an 8.9% decrease in the number of hospitals and a 26.4% decrease in the number of hospitals beds, CCM beds actually increased by 26.2% (77). During this period, overall CCM costs rose to 190.4%, exceeding overall economic growth, resulting in about a half a percent of total GDP being devoted to critical care by 2000 (77). Between 2000 and 2005, similar trends continued: hospital beds continued to contract (though hospital days increased), while CCM beds continued to grow (78). The percentage of GDP devoted toward CCM rose to 0.66% (which did not account for physician costs that were privately billed) (78). “We surmise that in the current climate,” the investigators noted, “the number and type of CCM beds will continue to increase without a ‘plan,’ simply as a response to increasing hospital admissions and days” (78). And in light of these trends, some have described an “impending critical care crisis,” characterized by massive shortfalls in critical care physician staffing (79).

Are such trends problematic? We and others believe that they are. ICU care is one instance in which evidence points strongly to supply-driven demand, as convincingly argued by Gooch and Kahn (80). This situation is not necessarily driven by nefarious motives: we all want the best for our patients, and how could better observation and a greater intensity of nursing care be harmful, particularly when ample ICU beds are available? From the cost perspective, however, the phenomenon of supply-driven ICU demand is highly problematic, because it means that patients who will not benefit from ICU care may receive unnecessary care. The reality of supply-driven ICU demand was suggested by one cohort study (cited by Gooch and Kahn) conducted in Calgary, Alberta, involving patients whose clinical condition warranted activation of a rapid response system (81). In the multivariate analysis, if a rapid response was called when there were more empty beds, more patients were transferred to the ICU and fewer patients had a change in their goals of care from aggressive to “comfort” measures (81). Despite these differences, however, an absence of empty beds was not associated with greater hospital mortality (81). This finding suggests that having more beds available resulted both in greater use and greater intensity of care, without a corresponding improvement in overall outcomes.

FIGURE 1.3 Use of the ICU at the end of life: US vs. England. US figures are based on data from seven states. Data from England is based on the 56% of ICU bed sample analyzed. (Data from Wunsch H, Linde-Zwirble WT, Harrison DA, et al. Use of intensive care services during terminal hospitalizations in England and the United States. Am J Respir Crit Care Med 2009;180:875–880.)

Only gold members can continue reading. Log In or Register to continue

Feb 26, 2020 | Posted by in CRITICAL CARE | Comments Off on Policy, Politics, and the Intensive Care Unit

Full access? Get Clinical Tree

Get Clinical Tree app for offline access