(Editor’s note: The following is the minority dissent to the Colorado Commission on Affordable Health Care Reports and Recommendations to the General Assembly. This dissent is authored by Linda Gorman, a member of the Commission and is based on version 4 of the final draft given to commissioners. The final report was released on June 30, 2017).
The Reports issued by the Colorado Commission on Affordable Health Care reflect a specific, viewpoint on health care in Colorado, the United States, and the rest of the world. It was enforced by majority vote. Not all members of the Commission share that viewpoint. Readers interested in crafting effective policy are strongly encouraged to conduct their own investigation, and to pay close attention to weaknesses in the theoretical and empirical underpinnings that plague many of the policies the Reports recommend.
Many Report recommendations reflect an implicit belief that central planning can lower health care costs. This belief is reflected in recommendations that the state control the health care workforce, determine the value of treatments offered to patients, and set prices paid for services. Euphemistic labels for price controls occur throughout the Reports. Price controls are disguised as proper “incentive structures” and “value-based” payments. A great deal of evidence dating back over 100 years suggests that price controls and central planning controls like certificate of need increase health care costs by retarding innovation and limiting competition. This evidence is not addressed in the Reports.
The bias against innovation without state permission is evident in the recommendation to study increased regulation for freestanding emergency departments (FSEDs). One FSED in Colorado is decades old. In the past few years, others have been opened because physicians and investors believe that they can provide better service for the same price. Colorado is one of the few states in which regulations allow for such potentially beneficial innovations. The Commission heard testimony from the Department of Health Care Policy and Financing suggesting that there were no serious continuing problems with the operations of FSEDs. It heard testimony from the Colorado Hospital Association suggesting that their data showed that the new facilities seemed to reduce the wait at existing emergency rooms but were not associated with an observable increase in patient emergency department use.
The 2017 Report does not discuss how FSEDs may benefit patients by lowering access costs. It focuses instead on the cost differences between emergency facilities and urgent care facilities. It says that had “patients sought treatment in urgent care centers instead of FSEDs, health care spending would have declined.” It does not examine whether symptom severity, hours of operation, or range of treatment options might rationally cause patients to choose one over the other. Instead it simply sows doubt with meaningless claims like “it is difficult to conclude that all of these patients definitively should not have gone to an urgent care center instead of an FSED.” It also fails to describe facility fees, explain why they exist, or show how they are a problem.
The malign effects of central planning are illustrated by the success of consumer directed health insurance and health savings accounts in reducing spending trends in the general population. They do this by getting rid of central planners and putting people in charge of health spending on discretionary services. The Reports do not discuss the success of consumer directed health insurance in reducing health spending or how Colorado Medicaid’s Consumer Directed Attendant Services Support program has increased patient satisfaction for the same expenditure. They also ignore how poor incentive structures and underpayment for drugs, doctors, and hospitals by public programs may increase private costs.
In general, the theoretical and empirical literature in economics and health policy suggest that price controls and central planning make people worse off by wasting resources, retarding innovation, treating patients poorly, and enriching those with political power. The Reports fail to mention the results from this literature, opting instead to recommend increased research on such health care topics as an analysis of overhead costs and the causes of high health care utilization in rural Colorado. The fallacy inherent in this point of view was ably summed up by Alfred E. Kahn in The Economics of Regulation, a book published in the early 1970s:
Effective regulation of operating expenses and capital outlays would require a detailed, day-by-day, transaction-by-transaction, and decision-by-decision review of every aspect of the company’s operation. Commissions could do so only if they were prepared completely to duplicate the role of management itself. [p. 30, 1995 edition]
While much of the analysis in the Reports is concerned with pricing variation in poorly defined service bundles, it is surprisingly unconcerned with how variations in the underlying health of the people who live in different parts of the state might affect costs. It ignores the significance of known cost differentials in providing food, housing, and other essentials. It fails to make clear distinctions between the operational results of for profit and non-profit health providers, an especially important task given the difference in taxes, access to capital, and malpractice expenses.
Testimony to the Commission suggested that the private sector has made significant strides in providing patients with data on the cost of care. The Reports make little mention of this, instead recommending that more public money be spent on subsidizing vaguely worded “transparency” efforts and supporting All Payer Claims Database efforts to create more complete provider claims data and make the data public, supposedly in support of enabling providers in the business of providing care to “better understand how their rates compare to those of other providers.” Insurers already have these data. The All Payer Claims Database results are problematic due to partial claims coverage and the fact that there is no objective way to distinguish between the actual cost of the same kind of care provided to a patient on time after a next day appointment or two hours late after waiting two months for an appointment.
Most of the Report recommendations call for more public spending, more public control over provision of health care by the private sector, and increased spending on public programs. This was summed up in the last Commission meeting as having to spend money to make money. In view of the waste and incompetency that characterizes so many government programs, the Report recommendations claiming that health care costs can be reduced by expanding public spending on housing and education deserve special scrutiny.
The state already spends large sums on housing and education. History suggests that many government social programs have a limited record of success. They often generate costly unintended consequences. It is well known that better health and lower health care costs are correlated with higher income. It is unknown whether better health is caused by higher income or is a result of personality traits like perseverance, self-denial, and impulse control, traits that may also generate higher income.
Report recommendations that extend research on results from small numbers of particularly disadvantaged people to conclusions about effects in the general population are fundamentally unconvincing. Program quality often declines when program size increases. Initial work often neglects to accurately determine total program costs. Recommending preschool for Medicaid eligible children is a case in point. Authorities differ on the academic effects of preschool, likely because measured outcomes apparently depend upon a child’s family background and his subsequent schooling.
The 2017 Report claims the 1972 Abecedarian (ABC) Project provides evidence of the effectiveness of preschool in reducing health care costs. This claim, in the words of Grover J. Whitehurst writing for the Brookings Institution in 2014, is equivalent to believing that “an expansion of the number of U.S. post offices today will spur economic development because there is some evidence that constructing post offices 50 years ago had that effect.“ There were 111 total children in the Abecedarian Project, just 57 in the group enrolled in preschool, and the study suffered from a compromised random assignment. More recent studies of large scale preschool programs in Tennessee and Quebec suggest that state funded preschool has little lasting effect on academic achievement. It may have negative effects on non-cognitive “soft” skills like self-control.
The description of the Affordable Care Act (ACA) in the 2017 Report provides another example of how the things left out of the Reports limit the examination of policy options. In its description of the Affordable Care Act and its impact on Colorado’s health care costs the 2017 Report says:
The Affordable Care Act had many goals, among those was the desire to reform how health insurance operates in the private market, and thus to make coverage more available to the population. Arguably, this goal was achieved. More people have coverage than ever before. As an aside, the ACA also provided of [sic] millions of dollars to the state from the Prevention and Public Health Fund, funding programs to prevent diabetes, heart disease, stroke, tobacco use and cancer. [page 57, Final Report Draft v4.]
An analyst relying on this report would have no idea that the (ACA) effort to “reform how health insurance operates in the private market” has had a disastrous effect on Colorado’s individual insurance market. Premiums for individual health insurance have doubled in less than 5 years and are likely higher than those that would have been charged by CoverColorado, the high risk pool that the state closed in 2013. Whether additional state funding has prevented a single case of diabetes, heart disease, stroke, or cancer is an open question.
The discussion of the ACA fails to mention that it has caused people who have individual coverage to experience unprecedented difficulty with narrow networks that block their access to medical care and generate large unexpected charges. It fails to discuss the unaffordable deductibles created by the Affordable Care Act policy requirements. It ignores the fact that the Affordable Care Act has made it impossible for most people buying individual policies in Colorado to have health savings accounts or buy a plan with a nationwide commercial PPO network comparable to those offered by many employer plans. It also fails to address how the Affordable Care Act Medical Loss Ratio requirements limit innovation and new entry into the market for coverage.
While the Report does mention that most of the coverage increase is due to Colorado’s Medicaid expansion, it does not discuss whether the increased coverage is a worthwhile policy goal if covered people still cannot get adequate medical care. Even though people have coverage on paper, the Commission heard public testimony that secondary care for Medicaid patients is virtually unobtainable in some parts of the state.
Before the Affordable Care Act, CoverColorado guaranteed health insurance to everyone in Colorado regardless of health status. The state indigent care program helped those who still choose to be uninsured manage the cost of unexpected acute care. People without major medical coverage still had coverage through auto policies, workers’ compensation, and EMTALA. They also had access to free primary care through federally qualified health clinics. Even with a few necessary modifications, the cost of these programs may have been less than the cost of the ACA Medicaid expansion, the ACA premium and tax increases, and the cost of running Connect for Health Colorado. Quality may have been better as well. The Report does not examine these possibilities.
Claiming that the failure to “honor risk corridor payments payable to insurance companies…meant that affected companies had to increase rates to make up their losses” misleads the reader by ignoring the fact that individual insurers did not have such losses prior to the passage of the Affordable Care Act.
Even more important omissions occur in the recommendations on pharmaceuticals. There is a large literature on how pharmaceutical companies ration global drug supplies in response to price. There is significant research on the response to government price controls of the sort recommended in the Reports, and on how monopsony negotiations with Medicare are likely to affect innovation. The Reports fail to mention these results.
There is a growing literature on the extent of drug counterfeiting and the danger it poses for those who re-import drugs. The Reports fail to mention it. The Reports also fail to discuss widely accepted facts about how FDA delay in drug approvals has created an opening for opportunistic pricing of generic drugs. They blame “opportunistic pricing behaviors by pharmaceutical companies or PHMs” instead.
The Reports suffer from sloppy methodology. Consider how they define health care costs. On page 12 of the 2017 Report, health care cost is defined as “the resources it takes for a health care supplier to produce goods or services, including labor, equipment, facilities and administration.” Costs that are important but not included in this definition include patient costs for specific types of health care delivery incurred in time, money, and the individual risk created by various treatment options forgone. Omitting patient costs also compromises the recommendations about payment structures. Emerging evidence suggests that seriously ill people may receive better, higher quality, care when payments are made on a fee-for-service basis. The Reports fail to mention this.
The data that are provided in the Reports often lack appropriate context. In the 2017 Report, a bar graph compares “Health and Social Care Spending” as a percentage of GDP for the US and other countries. It does not inform the reader that different countries use different accounting systems. While the US counts assisted living and home care as health spending, other countries include the same expenses under social spending. Despite the larger share of US health spending in overall spending, the Report confidently informs people that “the US devotes a relatively small share of its economy to social services designed to address social determinants.”
Recommendations to expand public programs also neglect the costs that state program expansions impose on citizens in the form of higher taxes and forgone program options. Even as no credible review of health care costs can neglect patient costs, no credible set of policy recommendations can neglect to examine the effect that specific recommendations may have on patients and, where applicable, taxpayers.
The Reports do not analyze the extent to which rent seeking may affect the outcomes of the recommendations they make. Rent seeking occurs when people lobby the government to pass rules or regulations that benefit them. Benefits may take the form of increased payments, protection from competition, a more influential position, or more power over others. As an example, the testimony on the effects of requiring patients to watch videos on end of life care was from a group of academic physicians well positioned to benefit from payments from hospitals required to show their videos by states seeking to reduce health care costs.
Though the Reports recommend a variety of policies to reduce spending in the last year of life, they do not discuss recent research suggestions that the fraction of health care spending on end of life care has been stable for decades. Recent estimates suggest it is roughly 13 percent of total health care spending and that a focus on reducing other areas of spending might be more productive [Aldridge and Kelley, 2015]. The Reports fail to address the fact that letting someone die is generally less expensive than treating him and that governments overly concerned with controlling cost have ended up killing patients. Britain’s Liverpool Pathway scandal provides an illustrative recent example of the dangers in this area.
Report claims that primary care reduces health care spending are based on limited evidence from one set of studies. The economics literature questions whether those studies properly address underlying population differences in health, wealth, and education.
Given that the Reports recommend determining such things as “value-based payment” and “value-based models” some consideration should have been given whether value-based experiments have been successful. This is important as numerous pay-for-performance efforts have done little to reduce costs or improve care. There is little evidence that value-based purchasing has improved hospital quality. [Ryan et al, New England Journal of Medicine, 2017]
In contrast, the Reports implicitly assume that health care value can be objectively determined. This assumption is probably incorrect. Economic research has demonstrated that an individual may value the same type of health care differently depending upon his stage of life and health status. It has also shown that many people enrolled in Medicaid probably place a lower value on it than the cost of the program would indicate should be the case.
The assumption of objectively definable value may lead the creation of economic rent if measures of value are constructed to control payments. The drive to produce quality metrics has produced large flows of funds to entities seeking the rents that flow from government adoption of their metrics. The possibility of rent seeking by entities seeking to determine value should not be overlooked.
The Reports exhibit a superficial understanding of what markets are, how markets operate, and the conditions required for reasonable market performance. Even though it is well known that market power is seldom determined solely by the number of competitors in a market, the 2017 Report states that fewer insurance carriers “could reduce the level of market competition.” After observing that Summit County has three carriers and high premiums while several Front Range counties have lower premiums, it concludes that “competition among carriers is not the only factor associated with high insurance premiums.”
The Market Advisory Committee was a separate group of people convened by the Commission to “discuss the important role that both market forces (and competition) and regulations play in controlling the cost of health care and identifying the role that market forces and regulations have on the principal drivers of health care costs.” Rather than describe how market competition has made sophisticated care generally available on a timely basis throughout the US and created physicians and hospitals that are as productive as any in the world, the Market Advisory Committee operated from the presumption that regulation comes first. A major topic of discussion was “when should the market be allowed to work unfettered by regulation.”
The Committee produced a series of recommendations that would extend the medical loss ratios that have harmed insurers to pharmaceutical drug manufacturers, put price and profits controls on pharmaceutical companies, prohibit balance billing by physicians and hospitals, and inject government into decisions about where a physician decides to practice. The Reports make little effort to provide evidence that this kind of political control will reduce health care costs or improve quality.
The Reports do not contain a discussion on the costs and distortions likely to be produced by the Market Committee recommendations or any description of the known consequences that those recommendations have had in Britain, Canada, and other countries that have already adopted similar policies. It did not discuss how treatment is provided in cash markets or how market competition makes those markets relatively inexpensive. These results are well known and should have been provided along with the recommendations.
Finally, it is unfortunate the Reports did not see fit to address the extent of existing state health programs. Several other states have produced detailed lists of state health programs that show significant overlap and unnecessary duplication of effort. Without knowing what the state currently spends, how can people be certain that Report recommendations for more spending are not simply duplicating existing state and private sector efforts?
Linda Gorman is health care policy center director at the Independence Institute, a free market think tank in Denver. She was a member of the Colorado Commission on Affordable Health Care.