Minority Report

1

Table of Contents SECTION 1: Introduction ...................................................... …………2 SECTION 2: Overview........................................................................... 4 SECTION 3: A Detailed Critique of Commission Recommendations 6 SECTION 4: Assessing Commission Cost Projections .......................20 SECTION 5: An Overview of the Commission’s Operations .............26 SECTION 6: Further Recommendations.............................................35 Appendix A: Key Health Care Reform Questions ..................................37

2

Minority Report
Linda Gorman R. Allan Jensen January 26, 2008 SECTION 1: Introduction This document offers an alternative to the recommendations approved by the Colorado Blue Ribbon Commission for Health Care Reform (the 208 Commission or Commission) at its meeting on November 19, 2007. Its authors are among the Commissioners who voted against that set of recommendations. On November 7, 2007, the Commission passed a rule requiring any commissioner who wished to submit a minority report to vote against the entire package of recommendations. Although the authors voted against the entire package, they do agree with some of the recommendations contained in the set. In general, the authors believe that the Commission recommendations view the private sector as the source of U.S. health care woes and an expansion of government control as the solution. The authors of this report have an opposite view. They believe that ill designed government interference has done positive harm to the development of the U.S. health care system over the last 80 years. As government programs have grown, they have begun to stress U.S. health care to the breaking point. In short, government is the problem, not the solution. Significant health care reform requires a transformation of government policy, with the goal of lowering costs through deregulation and of aligning incentives by ensuring, to the largest extent possible, that individuals buying health care are not spending someone else’s money. One of the biggest problems in health care reform is that parties with different viewpoints do not agree on basic facts. Simple logic dictates that it is nearly impossible to agree upon a reform plan without agreement on what needs to be reformed, and a basic weakness in the overall 208 Commission process was the failure to establish an agreed upon body of basic facts. Without this factual basis, the Commission members often could not even agree upon the problems that needed to be addressed let alone on sensible solutions to them. There are three main areas in which the Commission recommendations fall short. The first is that a vibrant free market for private health care and health insurance, one that offers responsible people a wide choice of health plans, physicians, and treatments, with a variety of ways to pay for each, should be the central part of any serious health care reform plan. Though some Commission recommendations mention consumer choice and market reform, other recommendations make such reforms impossible, too many of its recommendations would destroy or severely damage private health plans, private health insurance, and private medical care. An unacceptable number of Commission recommendations simply mimic the salient control points in the 2006 Massachusetts health care reform legislation, along with the disastrous Massachusetts decision imposing guaranteed issue and community rating on the individual insurance market in the early 1990s. Those regulations destroyed the Massachusetts individual insurance market and ultimately led to the adoption of the 2006 statute.

3 Similar regulations had similar effects in other states, effectively destroying the individual insurance markets in New Jersey, Maine, Tennessee, Kentucky, New York, and Vermont. Their imposition in Colorado will cripple its individual market, increase health insurance costs for large numbers of people, expand dependence on government programs, and retard innovation in health care delivery and coverage. The second major area of disagreement is the Commission’s neglect of promising developments in accountbased consumer-directed health care initiatives and the decision to instead favor various mandated insurance programs directed or controlled by government. While there is considerable evidence that account-based consumer-directed programs reduce costs, there is no evidence that the Commission recommendations for government expansion programs decrease costs. There is, in fact, some evidence that such programs actually increase them. The third area of disagreement is that the Commission recommendations substantially extend government control of medical practice without addressing compelling evidence that this has the potential to degrade care and increase costs. Though the Commission frequently asserts that its recommendations will lower costs, improve care, extend medical care to more people, or foster useful innovation, it does not provide adequate evidence to support its case. Cost estimates for the reform plans are likely understated because the model used to estimate costs was subject to a number of known problems. They are discussed further in Section 4. It is the authors’ view that any successful health care reform policy needs to address: 1) substantive reform of government programs, 2) incentives to reduce waste, and 3) the reduction of costly and unneeded administrative and regulatory burdens. These are the foci of the largest cost problems in the current health care delivery system. When the cost of health care drops, health insurance premiums drop and paying cash for care becomes possible. Paying cash further reduces costs by reducing third party payer overhead, with the result that more people can receive better health care for the same money. The authors also believe that the organizational processes adopted by the Commission likely caused its decision making to suffer from moderate to severe anchoring, framing, and availability biases. The lack of structured fact finding, discussed in detail in Section 5.4, was an important contributor to these problems. Report Format The second section of this document summarizes the areas of agreement and dissent. The third section provides detailed explanations of the authors’ reasons for dissenting. The fourth section discusses the organizational imperatives that impelled the Commission to produce recommendations with such a narrow view of health care reform and includes alternative recommendations for the operational structure of future Commissions. The fifth section provides recommendations for health care reform not endorsed by the Commission. The authors would like to thank their colleagues on the Commission for the time they spent on Commission activities, their principled participation, and the education that they provided. The Commission consisted of twenty seven citizens with different backgrounds, experiences, and areas of expertise that provided a valuable resource. The Commission staff was notable for its efficiency, knowledge, and good work in keeping Commission deliberations on track.

4 This report does not discuss all of the recommendations made by the Commission. It addresses the draft Commission recommendations adopted in the November 19, 2007 meeting. The final draft of all Commission recommendations was not available until 24 hours before the deadline for the submission of this minority report. Silence about a recommendation does not signify the authors’ agreement with it. In the following discussion, comments about various numeric quantities do not refer to the estimates provided by the Lewin Group unless the fact that a given quantity is a Lewin estimate is specifically noted. SECTION 2: Overview 2.1 Evaluative principles The Commission itself promulgated a set of principles for guiding its discussions. They are discussed in Section 5. The authors used the following set of principles in assessing the package of Commission recommendations: • • • • • • • Provide the most medical care to the largest number of people at the lowest possible cost. People should pay for their own medical care. If people cannot pay for their own medical care, taxpayers subsidize certain kinds of care. Market competition in the provision and financing of health care lowers costs. Lower costs are essential to ensuring that more people can pay for their own medical care. Choice is an American cultural imperative. An efficient health care system will respect this and ensure that people can choose their own health care and means of financing. Regulation increases costs and retards innovation. New regulations should be adopted only if they have been conclusively demonstrated to improve efficiency and further meaningful consumer choice. In general, a necessary but not sufficient requirement for meaningful consumer choice is that individuals control the funds used to pay for their health care. The failure of the government, at all levels, to properly fund Medicare and Medicare is unethical, and should be remedied before those programs are expanded or any additional government programs are adopted.

•

2.2. Areas of Agreement with Commission Recommendations Areas of agreement with Commission recommendations are listed below. In general, these areas of agreement overlap individual Commission recommendations. They do not reflect the specific language adopted by the Commission and may be broader, or more narrow, than the specific recommendation adopted by the Commission. Where applicable, references to specific Commission recommendations are in parentheses. They refer to the document entitled “Summary of Approved Recommendations” dated November 19, 2007 that was downloaded from the Commission website on December 2, 2007. In some cases these recommendations were subsets of larger recommendations. Inclusion of a subset of a larger recommendation does not signify agreement with the rest of the recommendations in a particular category. The same is true of recommendations that refer to a major category heading. Agreement with a major category heading does not imply agreement with the way that subsequent related recommendations interpret the major category. 2.2.1 Reform Medicaid ! Build on the success of the Consumer Directed Attendant support program in the Colorado Medicaid program.

5 ! ! Study the Commission recommendations for specific Medicaid reforms contained in Appendix X of the Commission recommendations. Seriously study extensive Medicaid reform, including the possibility that Medicaid regulatory burdens are such that the state of Colorado could save enough by opting out of Medicaid to provide better health care for those dependent on it.1

2.2.2 Reduce overall health system administrative and regulatory burdens and costs (2.D.3) 2.2.3 Devolve the health care delivery system to consumer control. (2.E) 2.2.4 Eliminate uncompensated care by reforming existing government programs (2B, 2D2, 3W)) 2.2.5 Eliminate laws preventing health insurance premiums from being adjusted to reward healthy behaviors 2.2.6 Minimize licensing barriers for medical practitioners 2.2.7 Support the appropriate use of information technology in health care (2.J) 2.2.8 Look for opportunities to duplicate successful local efforts that improve health care and lower cost. (2.O.) 2.3: Areas of Disagreement with Commission Recommendations This section summarizes areas of disagreement with Commission recommendations. As noted above, these areas of disagreement overlap individual Commission recommendations. They do not reflect the specific language adopted by the Commission and may be broader, or more narrow, than the specific recommendation adopted by the Commission. Extended discussions of the reasons for dissent are located in Section 3. 2.3.1 The Commission recommendations mandate that people to buy health insurance before anything else. This is an unethical policy recommendation. 2.3.2 The Commission recommendations promote an increase in the bureaucratic control of health care. This will increase the cost of health care, denigrate individual health, and is contrary to the Commission’s charge. 2.3.3 The Commission recommendations put care of the vulnerable at risk by altering the state’s focus from caring for the vulnerable to requiring insurance for all, implementing wellness programs of dubious value, and controlling medical practice. 2.3.4 The Commission recommendations do not implement effective consumer directed account-based reforms. Innovative private and public programs have shown that consumer-directed account-based reforms significantly reduce cost, improve health, and motivate consumer education.

1

Readers should note that while the Commission discussed reforming Medicaid, it never discussed studying whether Colorado could improve care for current Medicaid recipients by dropping out of the program. The savings from not having to abide by Medicaid rules were roughly estimated for one Western state a decade or so ago. The estimate suggested that this option might be worthy of more careful study.

6 2.3.5 The Commission recommendations claim cost savings for programs that have not performed well when tested in the real world, or for which convincing evidence of performance is lacking. 2.3.6 The Commission recommendations would destroy the market for individual insurance in Colorado, eliminating low cost coverage that encourages thousands of satisfied families to take control of their own health care. It will also raise the costs for these Coloradans and severely restrict consumer choice. 2.3.7 The Commission recommendations treat different residents of Colorado differently. For example, not all residents would be subject to the individual insurance mandate. 2.3.8 The cost estimates for Commission recommendations are unreliable due to the assumptions behind the model used to estimate them and to the fact that the Commission did not take public choice theory into account when in developing recommendations aimed at minimizing cost. 2.3.9 The Commission recommendations would have government needlessly duplicate a number of functions already available in the private sector, thus increasing health care costs. 2.3.10 Because preferred health care arrangements will vary from person to person, and over time as innovation occurs, government should be neutral with respect to the choice between alternative methods of health care delivery and financing. The Commission recommendations support particular health care delivery systems and discriminate against cash payment. 2.3.11 The recommendations include payments by taxpayers to particular groups without adequate supporting evidence that such payments are the most valuable use of taxpayer funds.2

SECTION 3: A Detailed Critique of Certain Commission Recommendations 3.1 The Commission recommendations force people to buy health insurance before they buy anything else. This is unethical. A lot of things can be more important than having health insurance, including buying food, paying for housing, having a job, and having reliable transportation to get to that job. The individual insurance mandate recommended by the Commission ignores this. It specifies that all legal Colorado residents will be required to purchase health insurance regardless of cost, and before meeting any of their other needs. Illegal residents of Colorado will be exempt from this requirement. The recommendations contemplate requiring proof of coverage when registering for school, and getting or renewing a driver’s license or car registration. People who cannot show proof of coverage will be fined a year’s worth of premiums when they file their income taxes. By recommending that the power of the Colorado Department of Revenue and the Department of Motor Vehicles be used to enforce the requirement that people
For example, the Commission recommends increasing payments to safety net providers, increasing payments to providers in health service shortage areas, and increasing funding for the public health system. Although health service shortage area sounds like an important designation, the HRSA website suggests that in addition to the rural parts of the state where one would think that health professionals might be in short supply, the criteria used to designate a HRSA are so broad that Colorado shortage areas include parts of Commerce City, Littleton, the Fort Lyon Correctional Facility, Denver, Colorado Springs, Arvada, and Englewood. http://hpsafind.hrsa.gov/HPSASearch.aspx as of December 11, 2007.
2

7 have health insurance, the Commission implicitly supports a significant increase in government control over individual decisions in normal household matters. The philosophical issue of what constitutes minimal health insurance is likely to be a much larger problem than is commonly recognized. In Commission discussions, and in the reform plans presented to the Commission, there was a distinct split between those who favored the imposition of policies with low deductibles and low plan limits, policies like those that had $100 deductibles with a plan maximum of $50,000, and those who felt that individual mandates should be limited to policies designed to cover catastrophic events, generally those with deductibles of thousands of dollars and plan limits in the millions of dollars. Individual insurance needs vary with things like age, location, health status, wealth, income, medical care preferences, and propensity to travel. Commission recommendations do not take these individual differences into account. Under Commission recommendations, an unelected, unaccountable, panel periodically reviews what the government will accept as a minimum health insurance policy. Benefits will be adjusted as needed. Legal Colorado residents will be required to pay for those minimum benefits whether or not they are a good value relative to other household needs. For example, the unelected panel might arbitrarily decide that the minimum benefit package should include dental care. At present, many people use cash to pay for their dental care. Paying for dental care via insurance is more expensive than paying cash― in addition to the dentist’s charges, people who buy dental insurance must also pay for insurance company profit and overhead. Should the unaccountable panel decide that the minimum policy includes dental care, following the Commission recommendations would have increased the overall expenditure on dental care. The Commission recommends that the minimum mandated health insurance policy have an average monthly premium of approximately $200. As a current scan of the available health insurance policies on ehealthinsurance.com for the Denver shows, policies offered in Colorado that meet that standard with a low deductible are temporary health coverage plans with varying plan maximums and deductibles. Long-term major medical coverage at those prices has deductibles that rise with age, hovering at about $2,000 for a 47 year old man. In view of the content of Commission discussions and of the progress of the debate over minimum coverage in the failing Massachusetts health reform plan, the imposition of an individual mandate will likely mean that thousands of Coloradans who currently have health insurance will find that their policies do not meet the minimum coverage requirement because their deductibles are deemed to be “too high.” This decision will be made by a panel that has no idea of the incomes, assets, health status, or values of the policyholders. From an individual’s point of view, the Commission is also recommending a new tax. By forcing people to buy a product that they may not want at a price that they cannot control, the individual mandate functions as an unlimited tax for health insurance. It should at least be subject to a statewide vote for approval. The individual mandate may also violate the Commission’s charge. It will likely increase the cost of health care for Colorado residents. People who currently get health care but have no insurance will be required to purchase insurance, thus increasing their costs. People who cannot purchase insurance because they are said to be unable to afford it will have to be subsidized to a larger extent than they are at present. Funding those subsidies will require tax increases that will raise costs for all Coloradans whether they be taxes on insurance premiums, provider taxes, sales taxes, taxes on food, or increases in the income tax.3
3

The Lewin Group has presented a number of charts showing average family health spending by income group under various reform proposals. Usually the groups shown are incomplete and the numbers shown do not provide a picture of overall spending as they do not show how many families are in each group. They also do not include the economy-wide effects of various tax increases on jobs, business formation, and incomes. In slide 18 of a November 15th presentation, the effect on people with incomes below $50,000 was

8 Because individual mandates are always coupled with subsidies to people judged to have too low an income to comply, enforcement of this provision, and the administration of the subsidies that make it possible, will require the collection of substantially more income data than is presently the case. Although the Massachusetts plan has been in operation only since April 1, 2006, it has already generated a 13 page Certificate of Exemption application that allows people to ask for permission to not purchase health insurance if they can demonstrate sufficient hardship. Among other things, hardship is defined as a notice of eviction, notice of utility shutoff, natural disaster or human caused event that causes large damage to you, your home, or your possessions, or you can establish that purchasing health insurance would have “caused you to experience serious deprivation of food, shelter, clothing or other necessities.” In effect, the individual mandate in Massachusetts requires Massachusetts citizens to petition the government for relief whenever they suffer a serious financial reversal. 3.2 The Commission recommendations substitute bureaucratically ponderous government health programs for individual choice in health care arrangements. This will likely endanger health by limiting access and choice. The Commission recommends that an unaccountable group study the “best scientific evidence to foster clinically, ethically, and culturally appropriate end-of-life care.” It also recommends the establishment of the Orwellian-sounding “Improving Value in Health Care Authority” to “fundamentally realign incentives” in the Colorado health care system by regulating provider payment and determining acceptable treatment. These recommendations endanger Colorado citizens. Although the Commission recommendations state that the Authority should simply study end-of-life care and best practices, based on what has happened in other cases when these recommendations have been put into practice, the Authority will end up transferring control over medical care and practice from individual citizens and their doctors to unaccountable, unelected, regulatory authorities. The potential for harm is made clear by the Commission recommendation to ‘Pay providers based on quality, such as use of care guidelines, performance on quality measures, coordination of patient care, and use of health information technology.’ While the Commission never defines quality, it is confident that the Authority will know it when it sees it. Physicians who do not do what the Authority demands will face financial harm. If what the Authority wants differs from what patients want, the Commission recommendations will give physicians an incentive not to act in the best interests of their patients The Commission has not discussed the possibility for harm in these regulations. Neither has it explained why it believes that an Authority will do a better job of aligning incentives than a program of deregulation that puts smart shoppers using their own money in charge of health care decisions, a system that has amply proven its ability to successfully align the incentives that producers face. The pay for quality recommendation means that the Improving Value in Health Care Authority would end up using evidence-based measures to regulate physician behavior and, ultimately, medical practice. At present, physicians are free to disregard evidence-based recommendations that conflict with their experience in clinical practice or with their patients’ wishes. Physician freedom of action is crucial to good medical care because it protects physicians and patients from researchers and regulators with an agenda, and from those whose values conflict with those of patients and doctors.

given for each $10,000 in income. For amounts above $50,000 the increments increased to $24,999 and then to $49,999. In 20042006, the Census Bureau put median household income in Colorado at $54,039. As the chart is restricted to averages between income groups, it is impossible to know what will happen to overall average spending. In the November 1 interim report by the Lewin Group reported that “About 70.4 percent of all Colorado families would see a net increase in health spending of $20 or more.”

9 The U.S. National Heart Lung and Blood Institute’s JNC 7 clinical guidelines for treating hypertension provide a recent example of how agenda driven research that can create seriously flawed evidence-based national guidelines with the potential to increase patient morbidity and mortality. The JNC 7 guidelines recommend starting all patients with high blood pressure on thiazide-type diuretics. In support of this, the express JNC 7 guideline for primary care physicians flatly states that “Thiazide-type diuretics have been the basis of antihypertensive therapy in most outcome trials. In these trials, including the recently published Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), diuretics have been virtually unsurpassed in preventing the cardiovascular complications of hypertension.”4 This statement is grossly misleading. In fact, the ALLHAT study has been subjected to withering criticism and the JNC 7 guidelines are not widely agreed upon. In Britain, the National Institute for Health and Clinical Excellence recommends ACE inhibitors as the first choice for initial therapy in patients younger than 55. The JNC 7 guidelines also ignore evidence suggesting that treatment with diuretics may increase the risk of developing new-onset type 2 diabetes. Newer antihypertensive drugs appear to have a beneficial or neutral effect on the glucose and lipid metabolism.5 By combining evidence-based standards with the pay-for-performance rules as advocated by the Commission, the Improving Value in Health Care Authority might use the results from poorly designed clinical trials to pressure physicians to use less expensive, older, and less effective therapies regardless of their relatively poor side effect profiles or of their effect on individual patients. The Commission recommendations also set the stage for various methods of provider profiling, including hospital and physician report cards, two currently fashionable quality initiatives which have been shown to have serious technical problems. They also give physicians an incentive to deny care to people who are very ill. Seriously ill people pose higher risk for a poor outcome. The physician may decide that he is better off not risking the poor report card grade produced by treating riskier patients.6 When such power is concentrated in the hands of an unaccountable group that has no personal contact with those affected by its decisions, patients become mere costs. The usual results of such policies are severely restricted access to advanced therapies for those who have complex medical needs, or are disabled, chronically ill, or in need of advanced medical care. In the Netherlands, physician caused deaths are increasingly commonplace. The utilitarian ethic adopted by the Royal Dutch Medical Society has virtually eliminated any prosecution for physicians who kill the elderly, the mentally ill, or the disabled.7

JNC 7 Express, The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. National Heart, Lung and Blood Institute, National Institutes of Health, U.S. Department of Health and Human Services, p. 7. Online version accessed November 28, 2007. http://www.nhlbi.nih.gov/guidelines/hypertension/express.pdf 5 There is a growing literature exploring possible relationships between treatment for high blood pressure and the onset of type 2 diabetes. For an example, see Kuti et al. April 23, 2007, “The development of new-onset type 2 diabetes association with choosing a calcium channel blocker compared to a diuretic or beta-blocker,” Curr Med Res Opin; Stas et al. October 2006. “Metabolic safety of antihypertensive drugs: myth versus reality,” Curr Hypertens Rep. 6 For sample literature on the topic of problems with quality measurement and report cards see Hofer TP et al. June 9, 1999. “The unreliability of individual physician ‘report cards’ for assessing the costs and quality of care of a chronic disease,” JAMA, 281(22): 2098-05; Sorokin R. 2000. “Alternative explanations for poor report card performance,” Eff Clin Pract. 3(3): 156; Shahian DM et al. March 27, 2007. “Comparison of clinical and administrative data sources for hospital coronary artery bypass graft surgery report cards,” Circulation, 115(12):1508-10; Normand SL et al. Jan-Feb 2007. “Assessing the accuracy of hospital clinical performance measures,” Med Decis Making, 27(1): 9-20; Epstein, AJ. August 2006. “Do cardiac surgery report cards reduce mortality? Assessing the evidence,” Med Care Res Rev. 63(4); Krumhlolz, HM. March 13, 2002. “Evaluation of a consumer-oriented internet health care report card: the risk of quality ratings based on mortality data,” JAMA, 287(10), 1277-87. 7 For a general discussion of the Dutch experience see Ezekiel Emanuel. March 1997. “Whose Right to Die,” Atlantic Monthly.

4

10 The importance of private sector benchmarking for government health programs was also ignored in Commission discussions. 3.3 The Commission recommendations will likely increase the cost to taxpayers of providing health care for people who cannot pay for their own health care. 3.3.1 The governing assumption of U.S. health care policy is that people who can afford health care should subsidize essential health care for those who cannot. In fairness, those who must pay for people who cannot pay for themselves deserve to have an efficient system for providing subsidized care, one that minimizes their costs. It should be noted that by international standards, virtually all competent people in the United States, regardless of their ability to pay, do get health care. Because the Commission recommendations include individual health insurance mandates and those mandates require providing large subsidies to people judged unable to afford health insurance, it likely that they will end up increasing the cost of already existing government programs designed to ensure that those who cannot pay get essential health. It will do this, in part, by encouraging people who presently pay for their own health care to stop doing so. Commission recommendations include subsidies for people with incomes up to 400% of the federal poverty level, $40,810 for a single person and $82,600 for a family of 4. This is higher than Colorado median incomes, which in 2006 were roughly $30,000 for households with no earners, $41,700 for households with 1 earner, and $77,000 for households with two earners. These policies have the potential to increase taxes on families who have employer health insurance but make $40,000 a year in order to subsidize families who make almost twice as much. 3.3.2 Even if users of Veterans Administration hospitals are counted as uninsured, studies suggest that the uninsured also pay for at least half of their own health care.8 Expanding public programs to cover people who are already paying for their care will eliminate those payments. As health insurance is an expensive way to buy health care, it is possible that it may actually be less expensive to provide care under the existing mixture of public subsidies to providers and private charity than to provide care under the system that would be created by the Commission’s recommendations. The Commission did not study alternatives to insurance coverage. It has not provided a compelling case that mandated health insurance buttressed by a large new bureaucracy dedicated to the control of insurance markets, medical practice, and extensive income redistribution is the lowest cost method of providing health care to those who need it but cannot pay. 3.3.3 Contrary to popular belief, the uninsured use emergency rooms at about the same rate as the insured.9 Generous estimates of the uncompensated care that the uninsured generate suggest that it is about 3 to 5 percent
8

For an example of a case in which care provided by the Veterans Administration is counted as care for the uninsured, see Jack Hadley and John Holahan. February 12, 2003. “How Much Medical Care Do the Uninsured Use, And Who Pays For It?” Health Affairs, Web exclusive. The problem, of course, is that the Veterans Administration is not supposed to serve those who are not Veterans. The second problem is that people meeting the criteria for lifetime health care from the Veterans Administration might rationally consider themselves insured and would not purchase private policies or enroll in other public ones. 9 For examples see Pines, JM and Buford, K. April 2006. “Predictors of frequent emergency department utilization in Southeastern Pennsylvania. J Asthma, 43, 3, pp 219-23; Sun, BC, Burstin, HR, and Brennan, TA. April 2003. “Predictors and outcomes of frequent emergency department users.” Acad Emerg Med. 10, 4, 320-8; Hunt, KA et al. July 2006. “Characteristics of frequent users of emergency departments,” Ann Emerg Med. 48, 1, 1-8; Fulda, KK and Immekus, R. July 2006. “Frequent users of Massachusetts emergency departments: a statewide analysis.” Ann Emerg Med. 48,1,6-16; Cunningham PJ. 2006. “What accounts for differences in the use of hospital emergency departments across U.S. communities?”, Health Aff, Sep-Oct; Zuckerman and Shen. 2004. “Characteristics of Occasional and Frequent Emergency Department Users: Do Insurance Coverage and Access to Care Matter?” Med

11 of private insurance premiums, which may be less than the revenue collection that has been proposed to finance the Commission recommendations. The 30 percent increase in private premiums that is widely bruited about as the additional premium cost for those who are insured includes the overcharges caused by Medicare and Medicaid. The Commission says that
In addition to reducing the number of uninsured Coloradans, an individual mandate would also reduce the premiums paid by those who are currently insured. This is due to the “cost shift” created when hospitals and other providers increase their rates to private insurance companies in order to cover the cost of care provided free or at reduced rates to the uninsured. Colorado health care providers gave $777 million in uncompensated care in 2007.10

The Lewin Group estimated that total Colorado health spending was about $30 billion. This means that the estimated cost of uncompensated hospital care for the uninsured is less than 3 percent of overall spending. In another context, the Lewin Group estimated that about 40 percent of the Colorado hospital shortfall is passed along to private payers. If correct, this would suggest that hospital care for the uninsured is about 1 percent of total spending. The reform proposal created by the Commission would increase health spending in Colorado by $2.7 billion, $854 million of which would come from an increase in personal income taxes. It is not clear to what extent these estimates of the premium increase caused by a “cost shift” include the amounts of uncompensated care generated by care for those nominally insured under government programs like Medicare and Medicaid.11 In Washington state, Milliman, Inc. estimated that the cost shift from Medicare and Medicaid to private payers was 14.3 percent of commercial hospital cost or about 4.8 percent of commercial premiums. With typical commercial health insurance premiums of $850 a month per family contract in 2004, the government program cost shift was about $490 a year. Physician underpayment by government programs was higher.12 By expanding Medicaid and Medicaid-like programs, the Commission recommendations run the risk of expanding uncompensated care, even with the higher Medicaid reimbursement rates recommended by the Commission. The Medicaid expansion will likely increase utilization and encourages people to substitute government payments for health care for their own payments for health care. TennCare, the Tennessee Medicaid expansion designed to insure everyone, also promised to reduce uncompensated care. Some years later, uncompensated care costs had increased. Tennessee abandoned the program after its costs threatened to bankrupt the state. 3.3.4 The Commission also failed to address losses that might arise from the disincentive to work created by the high marginal tax rate that people who receive subsidies will face as their incomes rise. People are free to make the choice between leisure and labor, between part-time work and full time work, between high paying jobs and jobs that pay less but are more congenial. By recommending such rich subsidies for health insurance the Commission may promote policies that increase the number of people who choose lower incomes in order to qualify for taxpayer supported programs. This appears to have been a particular problem with SCHIP/CHP+
Care, Feb 42(2). Urban Institute researchers Zuckerman and Shen concluded that “The uninsured do not use more [ER] visits than the insured population as is sometimes argued,” they write. In fact “the publicly insured are overrepresented among [ER] users.” 10 The Lewin Group. Technical Assessment of Health Care Reform Proposals. Appendix B: Health Spending in Colorado. p B-30. 11 In a personal communication with Commission staff, the Lewin Group cited a paper on physician pricing by Thomas Rice et al. as a source for its assumption that shortfalls in reimbursement were passed along to private payers in the form of higher hospital charges. As that paper discussed the effect of changes in Medicaid compensation changes on the volume of services provided this reference was apparently provided in error. The remainder of the communication simply said that “Our [Lewin’s] own analysis of hospital data indicates that about 40 percent of the increase in hospital payment shortfalls (i.e., revenues minus costs) in public programs were passed on to private payers in the form of the cost-shift during the years studied. Based upon this research, we estimate that 40 percent of increases in reimbursement would be passed back to payers in the form of reduced charges.” 12 Will Fox and John Pickering. May 2006. Payment Level Comparison Between Public Programs and Commercial Health Plans for Washington State Hospitals and Physicians. Milliman, Inc. https://www.premera.com/stellent/groups/public/documents/pdfs/dynwat%3B5724_15009501_3580.pdf

12 where an estimated 6 out of 10 new enrollees drop private insurance to participate in the subsidized public program. The rate is higher as more of the family is made eligible for insurance coverage. And, contrary to Commission assertions that waiting periods can control crowd-out, Gruber finds that “the anti-crowd out efforts that have accompanied the SCHIP program have probably raised crowd-out more than lowering it.”13 3.3.5 Finally, the Commission recommends that subsidies be provided to any household between 300 and 400 percent of the federal poverty level that cannot buy employer group insurance and spends more than 9% of its income on health insurance.14 This means that any family of 4 with an income of $61,950 that spends more than $5,576 on health insurance, and any family of 4 with an income of $82,600 that pays more than $7,434 on health insurance, is eligible for subsidies. According to the 2006 Consumer Expenditure Survey, families in this income bracket spent 5.3 percent of their household incomes on entertainment, 4.3 percent of their incomes on cash contributions, 3.8 percent of their incomes on household furnishings and equipment, and 5.5 percent of their incomes on food away from home. In view of this spending pattern, meeting a 9 percent premium burden would not seem to be beyond the range of possibility for these households, and subsidizing them would seem to place an unfair burden on other taxpayers. Focusing on the payments for health insurance discriminates against people who substitute cash savings for insurance, and only purchase health insurance that covers very large expenses. People who purchase a health insurance policy with a $10,000 deductible may never pay more than 9 percent of their incomes for health insurance but may very well have total health expenses that exceed 9 percent of income in one or two years every so often. Policies that encourage the purchase of health care using third party payment and discriminate against cash payments increase the cost of health care by increasing administrative overhead. Subsidies are likely to be more expensive than currently envisioned. One reason for the individual mandate is to limit the need for subsidies by requiring that everyone spend money on government defined health insurance. The fact that people who have low medical expenditures are exceptionally resistant to purchasing standard insurance policies should be an indicator that simply expanding the insurance model is a mistake unless regulators act to also lower economic costs. A substantial fraction of the uninsured simply feels that health insurance is a bad deal at current prices. Shifting that bad deal to taxpayers does little to change the cost/benefit situation. Estimates of the price elasticity of demand for individual insurance, the percentage change in policies bought divided by the percentage change in price, range from -1.0 to -0.3, suggesting that a 10% increase in insurance premiums results in a 3 to 10 percent decline in the number of policies purchased. Those in poor families without access to group coverage and not eligible for public plans are least likely to purchase individual insurance regardless of subsidy. Married couples tend to be less affected by price increases, single people are more sensitive. Marquis et al. find that even substantial subsidies for individual insurance would “have modest effects on the number of uninsured.”15 The Lewin Group uses an average price elasticity of -0.34 percent to estimate the price elasticity of the demand for health insurance to develop its estimates of coverage. Its estimate is derived from data from the Current Population Survey for 1987 to 1997. However, it goes on to say that it varies the elasticities that it uses by
Jonathan Gruber and Kosali Simon. January 2007. Crowd-Out Ten Years Later:Have Recent Public Insurance Expansions Crowded Out Private Health Insurance? Working Paper No. 12858, National Bureau of Economic Research, Cambridge, Massachusetts. p. 28. 14 9 % was chosen because research suggests that 75% of people with incomes in the subsidy range considered spend 9 percent or less on health care. 15 M. Susan Marquis, et al. October 2004. “Subsidies and the Demand for Individual Health Insurance in California,” Health Services Research, 39(5), p. 1564.
13

13 income. For those with incomes of $10,000 the income elasticity is assumed to be -0.55. For those with incomes of $100,000 the price elasticity is assumed to be -0.09.16 While the Lewin assumptions may be among the most reasonable available, how accurately this application of elasticities mirrors actions in the real world is unknown. In Wisconsin, a 2004 evaluation of the BadgerCare program speculated that “the mere perception of the premium [one equal to 3 percent of income for those with incomes over 150 percent of the federal poverty level] could be holding back applicants who would not be required to pay it.”17 Bundorf et al. find that the likelihood of purchasing health insurance increases with expected health expenditures, and that this effect is more likely to be observed in the large group market than in the individual market.18 Bundorf reported that in 2002 the average employee payment for single coverage was $450—about the average expected health expenditure for a man 25-29 years old. “If the wage difference between jobs without [sic] and without health insurance reflects the average premium for coverage ($3,060 for single coverage in 2002), the reduction in wages associated with coverage may generate income effects for low income workers that make jobs with coverage unattractive relative to those without coverage.” If people with the largest expected health costs are already insured, estimates of the savings from insuring the uninsured may be overstated. For employer provided insurance, Gruber and Washington use results from the transition of U.S. federal employees to pre-tax health insurance premiums from 1991 to 2002 to estimate the effect of after tax price on insurance takeup and plan choice. They find that lower premium shares led people to choose more expensive plans but had little effect on overall plan choice. The authors point out that targeting people who are already offered employer subsidized insurance but refuse it is very costly because the fact that these people have already turned down a highly subsidized produce means that they are exceptionally price sensitive or already have insurance from another source. They estimated that the federal government spent between $31,000 and $83,000 per person who was newly insured.19 This conclusion is roughly in line with the results in Maine, where the DirigoChoice program spends almost $16,000 in taxpayer funds to insure one additional uninsured person.20 Economic theory predicts that people with smaller medical expenditures would be more sensitive to the price of health insurance than those with larger, and less discretionary, medical expenditures. Empirical research produces some support for this supposition, suggesting that individuals self-select into insured or uninsured status depending on their knowledge of their individual health. The result is that the uninsured are not an isolated population subgroup, and to make sure that the majority of people have health insurance it is necessary to change behavior at relatively high levels of the income distribution. There is little evidence that insurers act to “cherry pick” and sort across plans, suggesting that worries about adverse selection in insurance markets are likely exaggerated. This means that the regulatory schemes proposed

The Lewin Group. August 20, 2007. Technical Assessment of Health Care Reform Proposals, Proof Report. Prepared for the Colorado Blue Ribbon Commission for Health Care Reform, p. D 12. 17 Chris Swart, Nina Troia, and Dorothy Ellegaard. July 2004. BadgerCare Evaluation. Evaluation Section, Office of Strategic Finance, Wisconsin Department of Health and Family Services, p. 54 18 M. Kate Bundorf, Bradley Herring, Mark Pauly. September 2005. Health Risk, Income, and the Purchase of Private Health Insurance, National Bureau of Economics Working Paper No. 11677, Cambridge, Massachusetts. 19 Jonathan Gruber and Ebonya Washington. March 2003. Subsidies to Employee Health Insurance Premiums and the Health Insurance Market, Working paper No. 9567, National Bureau of Economic Research, Cambridge, Massachusetts. 20 Tarren Bragdon. January 30, 2006. “Eight Challenges for Dirigo Health in 2006.” DirigoWatch, 3(1), Maine Heritage Policy Center, Portland, Maine.

16

14 to correct the problem, mainly guaranteed issue and community rating, are unnecessary and likely do more harm than good.21 The other problem with subsidies is that means tested subsidies have the potential to be extremely unfair and create a disincentive to act responsibly. For example, if the Commission recommendations are followed, the State of Colorado could end up taxing a young married couple with employer provided health insurance, a baby, and an income of $25,000 to provide health insurance subsidies to an older married couple with substantial home equity and retirement savings, with three children, an annual income of $68,000, and a business that does not provide health insurance. In order to provide more health care for all at a lower cost, other options than the comprehensive insurance model—subsidized clinics, designated hospitals to which those who cannot pay can be transferred, removing the regulations that discourage physicians from participating in charitable activities and charitable organizations from operating them, insurance plans that provide small benefits for low cost and insurance plans that offer catastrophic benefits—need to be explored and evaluated. 3.4 The Commission recommendations divert focus from the efficient use of Medicaid dollars in caring for the vulnerable. Because tax supported health programs essentially consist of telling law abiding citizens that they should “either hand over the money for someone else’s health care or face punishment,” government has a responsibility to ensure that tax dollars flow to those who are most in need of other people’s support. Historically, tax supported health care programs have focused on two areas: public health programs to limit environmental health hazards, the spread of infection, and communicable diseases, and public programs providing individual care for people who were unable to provide it for themselves, primarily children, the frail and impoverished elderly, people with grievous injuries or diseases, and people with severe birth defects or developmental disabilities. There is a finite amount of tax money available for subsidizing health care. In our view, the Commission recommendations would divert substantial resources to areas in which state government has little prior experience and, in some cases, a poor record of success. These include extensive record keeping on large numbers of complex transactions for every individual in the state, developing and deploying effective information technology architectures that are new and untested, developing new regulations for every area of medical practice, developing and promoting wellness initiatives of dubious merit, vastly expanding means testing for subsidies, enforcing the health insurance mandate, and doing extensive systems design research. Tax money spent on these initiatives is tax money not available for projects to ameliorate the conditions of those with serious disease or disability. This is of particular concern in view of the fact that many of the people who testified before the Commission were concerned about the inadequate care being given by existing public subsidy systems. 3.5 The Commission recommendations virtually ignore consumer-directed account-based reforms. These have been shown to significantly reduce costs, improve wellness, and motivate education.
Patrick Bajari, Han Hone, and Ahmed Khwaja. August 2006. Moral Hazard, Adverse Selection and health expenditures: A Semiparametric Analysis. Working Paper No. 12445, National Bureau of Economic Research, Cambridge, Massachusetts; M. Susan Marquis and Melinda Beeuwkes Buntin. October 2006. “How Much Risk Pooling Is There in the Individual Insurance Market?,” Health Services Research, 41(5), 1782-1800.
21

15 For the purposes of this discussion, consumer-directed accounts should be understood to be sums of money that people control, benefit from, and can spend at will on certain broadly designated categories. In health care, they are usually combined with health insurance policies that have lower premiums and deductibles of at least $1,100. If people save money on health care, savings accumulate in their health savings account. Health Savings Account (HSA) balances belong absolutely to the person who owns them, accrue interest tax free, and can be spent on any medical expense recognized by the Internal Revenue Service. HSA balances can be willed to beneficiaries. After age sixty-five, funds can be used for other purposes. Health Reimbursement Accounts are not owned by individuals and amounts in them may be lost when an individual changes employers. Health care reforms that arbitrarily limit financing choices to a few, governmentally approved, options are not consumerdirected. With the exception of the Medicaid reform recommendation that would create consumer-directed account-based program for purchasing supplies like adult diapers, the Commission recommendations do not promote any of the consumer-directed private or public sector initiatives that have been reducing costs and improving health in both the public and private sectors since the late 1990s. The use of consumer-directed health savings accounts (HSAs) coupled with high deductible health insurance policies (HSA/HDHP) has grown rapidly since their inception in December 2003. There were 1 million HSA/HDHP accounts open by March 2005. The number had risen to 4.5 million by January 2007. Projections recently released by AHIP forecast that the use of HSAs will double in the coming year, and that the use of all consumer-directed products will more than triple. In employer sponsored plans, the Mercer National Survey of Employer-Sponsored Health Plans suggests that enrollment in consumer-direct health plans has risen to 5% of all employees. The 2007 average cost per employee for HSA plans is $5,679, roughly $700 less than the average $6,644 cost for PPOs with deductibles of at least $1,000. Mercer comments that this “lends support to the theory that the account feature encourages more careful health spending.”22 Private insurers have already begun to increase coverage and lower costs using consumer-directed accountbased products. The lower premiums associated with the HSA/HDHP policies have helped to reduce the number of insured: an estimated twenty-seven percent of the 1 million people covered by individual HSA/HDHP policies in force by January 2007 previously had no health insurance. They are particularly appealing to the young: 39 percent of the people covered by HSA/HDHP policies are under age 29. Disease management is offered with over 80 percent of the policies, covering conditions such as diabetes, coronary artery disease, congestive heart failure, asthma, and chronic obstructive pulmonary disease. Over 85 percent of the companies writing HSA policies offered health education information, information on physicians, hospitalspecific quality data, and health care cost information. Seventy-two percent offered online personal health records. Policies that are owned by individuals are portable from job to job and, if purchased from a national company, are often portable when someone moves within the United States. Consumer-directed account-based reforms also have reduced costs and improved health for disabled Medicaid participants. The Colorado Consumer Directed Attendant Support Program has improved health while saving 20 percent or more on attendants for the disabled simply by freeing those in the program from Medicaid regulation. The flexibility that the Robert Wood Johnson Foundation’s Cash & Counseling accounts have brought to Medicaid spending have also unambiguously increased access to needed services and reduced unmet needs.

Mercer Group. November 19, 2007. U.S. employers’ health benefit cost continues to rise at twice inflation rate, Mercer survey finds. http://www.mercer.com/referencecontent.jhtml?idContent=1287790

22

16 Results from private sector employers like Wendy’s, John Deere, and Whole Foods suggest that in addition to reducing costs, account-based consumer-directed health insurance increases the use of preventive care. Reports from other employers indicate that people covered by consumer-directed accounts are more compliant in their use of recommended medications and more active in disease management programs. A recent paper by Greg Scandlen reviews the evidence on consumer-directed account-based health care reform and considers whether consumer-directed insurance has lived up to initial predictions. It concludes that initial indications suggest that account-based insurance is changing patient behavior by reducing the demand for unnecessary services, encouraging higher compliance with treatment recommendations, and increasing the use of preventive care. The rate of increase in costs for users of the account-based plans has decreased substantially. There are early indicators that account users are fueling a transformation of service delivery.23 It should be noted that discussions at Commission meetings suggested that a number of Commissioners are hostile to the continued use of HSA/HDHP policies. Enacting a number of the Commission recommendations, including those on guaranteed issue, community rating, a ban on medically underwritten policies, and the specification of a minimum benefit policy for all Coloradans, would likely eliminate or severely restrict the availability, and benefits from, HSA/HDHP policies currently in force in Colorado. The Commission’s animus towards consumer-directed solutions is reflected by the fact that the Commission voted to recommend that the legislature study the possibility of imposing complete government control over all Colorado health care via the imposition of a single payer system. On the same day, it defeated a motion to recommend that the legislature also study the possibility of using account-based consumer-directed health care reforms in health care reform. If the experience in New York, Maine, New Jersey, Tennessee, and Massachusetts are any guide, the enactment of Commission recommendations to impose guaranteed issue and community rating will severely damage the individual insurance market, stop the consumer-directed insurance market in its tracks, stifle important health care innovations, and lessen competition among insurers. This could expand the number of people who are either uninsured or dependent on government for health coverage. Expanding the number of people on government programs could be costly if they are removed from innovative private programs designed to manage chronic health conditions. In Colorado, Aetna’s migraine headache management program reduced MRI use, increased the use of appropriate medications, and improved the quality of life for migraine suffers. Great-West’s oncology management program reduced the rate of hospital readmissions by 17 percent by hiring nurse managers to help patients cope with treatment. Rocky Mountain Health Plans developed a diabetes management program that combined pay-for-performance measures with case management fees. It improved the percentage of diabetic members with good blood pressure control, and increased the number of members with acceptable LDL-cholesterol levels. The Kaiser Permanente ALL program has pharmacists call all members with diabetes or coronary artery disease to ensure that they are taking all of the medication that they should. Humana analyzes monthly data on claims to ascertain whether people might benefit from its personal nurse coaching service. 24 Judging from the extent to which Commission recommendations would require government to develop programs already in development in the private sector, it is not clear that the Commission was aware of the
Greg Scandlen. November 2007. Working as Intended: What We Have Learned About Consumer Driven Health Care. Consumers for health care Choices, Hagerstown, MD. 24 America’s Health Insurance Plans. March 2007. Innovations in Chronic Care: A New Generation of Initiatives to Improve America’s Health.
23

17 innovations occurring in private insurance. It is also not clear that the Commission fully appreciated the extent to which the internet, coupled with the consumer-directed health insurance revolution, has simplified comparing, pricing, and purchasing a health insurance policy. 3.6 The Commission recommendations assume cost savings for programs that have not been tested, and that have been tested but have not performed well in the real world. Although the Commission asserts that centralized electronic medical records will cut costs, evidence supporting this assertion is lacking. In general, it did not systematically study the costs and benefits of its information technology mandates or assess the existing evidence on whether a statewide reporting system would be a sensible use of health care dollars. Had it done so, it would likely have been compelled to note that such data systems have yet to prove themselves in practice. 25 The evidence to date suggests that electronic medical records will increase the risk of misuse of individual health information. Identity theft is already common. New criminal uses of individual health information include using someone else’s name to get expensive health care services and attempting to extort money from employers by threatening to publish patient records, a breach that could lead to serious penalties for being in violation of HIPAA. Electronic records also increase the risk to state taxpayers, who could be liable for damages caused by stolen or misused records. The Veterans Administration, long praised for its electronic records, has repeatedly lost sensitive data on millions of patients and has spent tens of millions of dollars repairing the damage caused by such thefts.26,27 Although popular mythology assumes that electronic records will reduce cost, the evidence from hospital based systems is mixed. Hospital based computerized order entry systems for prescription drugs do appear to reduce medication prescribing errors at the possible cost of increased workloads and decreases in human vigilance against error. Experts fear that this combination may harm patients in situations when rapid treatment is essential. There are scant data on whether electronic prescribing records improve health outcomes, and a small but growing literature on the new kinds of errors that they facilitate.28 Other problems with electronic records that the Commission did not address before issuing its recommendations include how to control error propagation, and the lack of correspondence between clinical and administrative records. Medical records contain errors, and those errors are neither reduced nor corrected by computerizing them. A November 21, 2007 article from the AP described the errors that physicians found in their own medical records.29 Under HIPAA, there is no requirement that those who maintain health records be required to correct them. There are important questions about who should have the authority to alter electronic patient records. Data system robustness is a concern. There are also studies that have found that the records themselves change behavior. In the Veteran’s Administration system, a significant number of patient records have case notes that are electronically copied from one record to another in order to save time.
Evidence on the effectiveness of electronic health records in improving care differs. Crosson et al. 2007. “Electronic medical records and diabetes quality of care: results from a sample of family medicine practices,” Annals of Family Medicine, found that practices not using electronic medical records were more likely to meet their standards for high quality care. 26 FierceHealthIT. June 17, 2007, “VA could spend $20M on data breach response,” http://www.fiercehealthit.com/story/va-couldspend-20m-on-data-breach-response/2007-06-18 27 http://www.aao.org/publications/eyenet/200402/news.cfm 28 For an example, see Koppel et al. 2005. “Role of computerized physician order entry systems in facilitating medication errors,” JAMA, 293(10). 29 http://ap.google.com/article/ALeqM5jophmSjTNwlCyvhfctYIJbcrht0AD8T288Q07
25

18 A final problem is that the drive to use patient records for billing and monitoring may degrade their usefulness in patient care. Patient records were originally developed to help clinicians provide care. If administrators insist on standardizing them in order to use them for process control and provider evaluation it is likely that clinicians will respond by not keeping notes that can be used against them. In Britain, hospital trusts have “adjusted” patient records in order to suggest that patients had been treated on time.30 In the U.S., physicians already keep multiple sets of records. One is in the format demanded by payers like Medicare. The other may be private notes that suit a physician’s personal style and helps him facilitate patient care. 3.7 The Commission recommends extending guaranteed issue and community rating to the individual insurance market. These recommendations have increased costs and reduced coverage in the other states in which they have been applied. 3.7.1 As the Commission deliberated, all manner of claims were made about the cost of individual health insurance in Colorado. Many of them are incorrect. As there are a number of websites that give basic information on this, there is little reason for not knowing basic facts. One of those facts is that the Massachusetts Connector Authority is now quoting individual insurance prices that are higher than those currently prevailing in Colorado. This is illustrated by the representative premiums given in the table below. Massachusetts is the real world outcome of the Commission recommendations for an individual mandate and guaranteed issue and community rating in all insurance markets. The Commission recommends guaranteed issue for all health insurers operating in Colorado. It apparently believes that by requiring applicants with specified pre-existing conditions to enroll in Cover Colorado, the state’s guaranteed issue plan for individuals, a “cushion” will be created that will offset the increased cost of making everyone who already has insurance pay the bills for any person who knows he will have high medical expenses and wishes to buy an insurance policy to pay his bills. In fact, individual insurance underwriters currently have three options: accept an individual application as applied for, decline the application, or accept it with conditions, such as waivers and ratings. They already charge higher prices for higher risks. People with specified pre-existing conditions are already covered by Cover Colorado. This means that there is no particular reason to believe that the Commission plan will create premium stability. And every reason, based on experience in other states, to believe that premiums will rise significantly.

A Comparison of Massachusetts and Colorado Monthly Health Insurance Premiums (as quoted to insurance brokers the week of December 3, 2007) Colorado individual market (zip: 80222) 10 year old child 40 year old man $102.00 $172.00 Cover Colorado guaranteed issue $125.30 $250.18 Massachusetts (zip: 02101) Guaranteed issue $193.81 $246.10 $2,000 deductible $2,000 deductible

30

BBC, December 19, 2001. “Crackdown on waiting-list ‘fiddles,’” http://opinionjournal.com/editorial/feature.html?id=95000390

19 3.7.2 The Massachusetts plan penalizes the parents of the healthy children, and people who purchase health insurance before they get sick. This is inefficient and unfair. It raises the rates paid by the vast majority of people. In doing so, it discourages the purchase of private insurance coverage, especially in lower income brackets where people are especially sensitive to premium price increases. Colorado achieves guaranteed issue for all, but does it by directly subsidizing insurance for the uninsurable and letting those who act responsibly save by purchasing medically underwritten insurance in a market with far more flexible pricing. Rather than supporting a bureaucracy to control all health insurers, Colorado efficiently uses taxes to support a far more limited bureaucracy that focuses on providing coverage to thousands of Colorado residents who cannot purchase health insurance due to pre-existing conditions. In short, it achieves the Massachusetts result of making insurance available for everyone, and does so at a lower cost. 3.7.3 Recommending the extension of community rating and guaranteed issue to Colorado’s individual insurance market is also irresponsible in view of the agreement that it will increase cost and reduce access to insurance. The academic literature on this is clear: guaranteed issue and community rating increase costs and decrease coverage. Examples of statements from the literature include: “States limiting risk rating in individual insurance display lower premiums for high risks than other states, but such rate regulation leads to an increase in the total number of uninsured people;”31 “only small changes in risk pooling because the extent of pooling in the absence of regulation is substantial;”32 such regulations “have succeeded only in making individual health insurance coverage more expensive and less available than it otherwise would have been;”33 “individual health insurance markets deteriorated after the introduction of GI and CR reforms…premium rates tended to increase, sometimes dramatically. We did not observe any significant decreases in the level of uninsured persons.”34 Recommendations known to increase costs and reduce coverage are not in accord with the Commission’s legislative charge. 3.8 The Commission recommends replicating a Massachusetts Style Connector in Colorado. This is unnecessary as it duplicates existing private systems and is very costly. A good deal has been written about Connectors ever since Massachusetts incorporated the concept into its reform measures. A recent academic analysis by Schneider, et al., of the Health Economics Consulting Group (Aug 2007) allows that the “connector” concept, as a proposed mechanism of moving commercial health insurance market away from an employer-sponsored to an individually based environment holds certain “intuitive appeal.” However, they note that there are numerous trade offs and consequences, which prevent such programs from accomplishing their goals. The information centralization function envisioned by these programs is an unnecessary and costly addition to the administrative costs of the health care coverage system. In fact, well developed mechanisms already exist to offer consumers the opportunity to search for and compare various health coverages.Various internet and insurance carrier websites provide complete coverage descriptions and allow coverage comparisons. Government publications in various forms explain health insurance. As is the case in other markets selling complex financial products, agents provide significant amounts of consumer education, act as ombudsmen to
Mark V. Pauly and Bradley Herring. May/June 2007. “Risk Pooling and Regulation: Policy and Reality in Today’s Individual Health Insurance Market,” Health Affairs, 26(3), 770-779. 32 Bradley Herring and Mark V. Pauly. August 2006. The Effect of State Community Rating Regulations on Premiums and Coverage in the Individual Health Insurance Market, Working Paper No. 12504, National Bureau of Economic Research. 33 Conrad F. Meier. 2005. Destroying Insurance Markets. The Heartland Institute, Chicago, Illinois. 34 Leigh Wachenheim and Hans Leida. August 2007. The Impact of Guaranteed Issue and Community Rating Reforms on Individual Insurance Markets. Milliman, Inc. Brookline, Wisconsin.
31

20 intervene with insurers on behalf of clients, and provide a check on insurer quality by refusing to market plans that treat customers poorly. Other problems with connectors include: ! ! ! ! ! Displacement of existing coverage which may stress remaining risk pools Legal issues including ERISA, the implications of using the Section 125 provisions of the IRS code, HIPAA and COBRA 1985, list billing and guaranteed issue Loss of product innovation and choice Disproportionate risk and premium cost increases due to adverse selection whether risk pools are voluntary or mandatory Fairness in the allocation of risk and financial burden

Connectors are not inexpensive. In 2006, the Massachusetts Connector Authority estimated that its expenses of $24 million in 2007 would be rising to $36 million in 2009.35 It is not clear that the Commission was aware of these costs, and of the other issues cited, when it made its recommendation to create a Connector agency. SECTION 4: Assessing Commission Cost Projections 4.1 Modeling Overview The Commission hired the Lewin Group to model its health reform proposals. The Lewin Group has developed a specialty model that looks at health spending with a 1990s perspective and can be adopted to various front end conditions with the creation of a synthetic population. In its June meeting, the Commission listened to a presentation of the baseline estimates developed using the Lewin model. Commissioners immediately began to view Lewin as an expert source of policy information. This effect was so pronounced that it can be fairly reported that the choices of which proposals to send forward for modeling were based on the briefing received from Lewin. The numbers developed by Lewin are those against which health care reform proposal modeling forecasts will be compared. At the outset, the Lewin group warned the Commission that it could not model the broader, longterm, economic impacts of various health reforms on such things as earnings or job losses. It also warned that it could not forecast the health care shortages or waiting lists that might be created by various reforms. Since relatively little is actually known about health insurance and health spending in the Colorado population, the Lewin model creates a synthetic population based on some known facts about the Colorado population and puts that population through its paces based on its general assumptions on how various population segments will change their behavior in response to different policy requirements. If the synthetic population is accurate, and assumptions about costs and individual behavior in response to change are accurate, then modeling results may be accurate predictions of future results. It should be noted that the baseline estimates made for Colorado appear to make a number of choices that may overstate the population of the uninsured. (see 4.2.2)
Commonwealth Health Insurance Connector Authority. December 14, 2006. Plan of Operations: Three Year Financial Plan/Budget.
35

21 For a sense of the errors that can be introduced in modeling, and why models must always be thoroughly checked against common sense and the real world, the following discussion outlines some of the issues raised by Lewin documents describing the development of the Commission’s baseline estimates, the modeling estimates against which estimated outcomes for the selected reform plans will be compared. Like any other model, the one used by the Lewin Group has strengths and weaknesses and the cost and benefit estimates produced by it must be placed in proper perspective. For policy purposes, the Lewin estimates share three important limitations with almost all other models. Although they provide valuable information about how policy changes might interact, their applicability to the real world is limited for the following reasons: 1. Model estimates reflect conditions at a specific time and generally assume instant adjustment to new conditions. The long-term effects of some proposed actions cannot be considered, nor can cost estimates account for changes that might occur over time. These might include significant price changes that increase or decrease the costs of specific policies. This makes the models much less reliable for cost predictions in the mid or long-term, than the short term. 2. The cost projections assume that every aspect of a proposal is put into place at the same time. The cost estimates do not apply to policies that are implemented a piece at a time. 3. Lewin informed the Commission at the outset that it could not model shortages created by inappropriate pricing. It also does not model broader economic effects such as the effect of tax increases on employment and earnings. 4.2 Specific Shortcomings 4.2.1 The administrative costs assumed for private insurers are too high. In one case, a Lewin presenter said that the administrative costs for individual insurance products were as high as 44 percent. The administrative costs more commonly cited in Lewin written materials were in the range of 34 to 35 percent based on data from the Department of Insurance. Administrative data typically view administrative costs as the difference between revenues received and benefits paid. In this formulation everything that is not a benefit payment is an administrative cost. This includes profits, programs that generate savings on health care (and therefore reduce benefit payments), and fraud control that reduces benefits payments. For individual policies, all administrative functions are included in the premiums. For employer group policies, the difference between premiums and benefits payments would not include all of the additional human resources costs that companies incur to run their insurance plans. The authors of this report contacted 12 insurance carriers during the preparation of this minority report. They said their administrative costs for individual policies ranged from 15 to 23 percent.36 4.2.2 Data limitations make the model rely on small samples that may or may not represent Colorado The detailed estimates of Colorado health spending and insurance coverage depend on 2004 Medical Expenditure Panel Survey (MEPS) data. When the MEPS sample was redesigned in 2004, MEPS expected roughly 560 responses from Coloradans in the private sector. No one knows how those 560 people would self36

Allan Jensen, personal communication.

22 select for participation. Perhaps they are more likely to have time on their hands because they are in poor health and miss more days of work or work less, perhaps not. As the small number of Colorado MEPS survey participants provides severely limited information about conditions in Colorado, the Commission model creates a baseline Colorado population using MEPS estimates for the western United States. These MEPS estimates are used to estimate household spending, spending by military personnel and veterans, out-of-pocket spending, the cost of employer sponsored insurance, retiree premiums for employer provided insurance, and spending on state and local government employees. The Western United States includes California. Although Lewin used generally accepted methods to account for Colorado characteristics by weighting the Western United States results, the weights chosen may not accurately reflect differences in the use of health care. For example, Lewin weighted the MEPS results to account for the fact that the California population is 25% Hispanic while the Colorado is only 18% Hispanic. But simply being Hispanic is a crude measure of health care utilization. For example, for those over 35 years old, the diabetes related mortality rate varies from 251 per 100,000 Mexican Americans, to 204 per 100,000 for Puerto Ricans, and 101 per 100,000 for Cuban Americans,37 asthma prevalence varies from 13.2 percent for Mexican Americans to 23.0 percent for Cuban Americans,38 birth rates range from 105.1 births per 1,000 women of Mexican origin to 49.3 per 1,000 for Cuban women,39 and Hispanic immunization rates vary by area of origin.40 Absent other controls, it would also not account for the large difference in public program participation between those who have recently moved to the United States and those who have been in the U.S. for several generations.41 4.2.3 The method chosen to estimate the number of uninsured likely produces overestimates. Lewin bases its estimates for the number of uninsured in Colorado on the March Current Population Survey (CPS), pooled from 2004 to 2006 to provide a bigger sample size. For a variety of technical reasons, it is generally agreed that the CPS overestimates the number of uninsured, in part because respondents appear to report their insurance status at the moment they are questioned rather than for the whole year.42 The differences between CPS estimates and those of other surveys can be large. A 2004 comparison of CPS estimates with those from the Survey of Income and Program Participation (SIPP) found that the CPS estimate of the uninsured was 8 percent higher than the SIPP estimate.43 In 1998, the Wisconsin Family Health Survey estimated that 4 percent of Wisconsin residents were uninsured for an entire

Smith CA, Barnett E. Dec 2005. “Diabetes-related mortality among Mexican Americans, Puerto Ricans, and Cuban Americans in the United States,” Rev Panam Salud Publica, 18(6); 38 Davis AM et al. August 2006. “Asthma prevalence in Hispanic and Asian American ethnic subgroups: Results from the California Healthy Kids Survey,” Pediatrics, 118(2). 39 Sutton PD and Mathews TJ. May 2006. “Birth and fertility rates for states by Hispanic origin subgroups: United States, 1990 and 2000.” Vital Health Stat 21. 40 Herrera GA, et al. May 2001. “Variation in vaccination coverage among children of Hispanic ancestry.” Am J. Prev Med. 41 For example, see Borjas GJ and Hilton L. May 1996. “Immigration and the Welfare State: Immigrant Participation in Means-Tested Entitlement Programs,” Q J Economics, 111(2). 42 For an example of the literature on this topic see Cathi M. Callahan and James W. Mays. March 31, 2005. Estimating the Number of Individuals in the United States Without Health Insurance. Working paper for the Department of Health and Human Services, online version accessed December 10, 2006, http://aspe.hhs.gov/health/Reports/05/est-uninsured/report.pdf. 43 Shailesh Bhandari. June 8, 2004. People with Health Insurance: A Comparison of Estimates from Two Surveys, Report No. 243, The Survey of Income and Program Participation, U.S. Census Bureau. Online version, assessed July 10, 2007. http://www.sipp.census.gov/sipp/workpapr/wp243.pdf

37

23 year. The CPS estimate for that year was 11.8 percent.44 Wisconsin believes that the Family Health Survey is more accurate. To estimate the baseline estimate of Colorado uninsured, the Lewin model took the CPS estimates and combined them with Colorado Medicaid enrollment data to estimate the “real” number of uninsured. It then reduced the number of Colorado uninsured from 758,800 to 562,800 people to account for the Medicaid undercount associated with the CPS. Saying that it “estimates that there were another 506,800 people who were uninsured for part of year [sic],” Lewin added this to its estimate of 562,800 to arrive at a figure of 1,069,000 people uninsured at any point in the year. Further calculations provide an estimate of 785,200 people uninsured in any given month, 17.2 percent of the Colorado population. The effort to number the uninsured on a monthly basis is a result of the fact that Medicaid enrollment is reported a month at a time. But someone who is uninsured for a month or two in a year does not face the same difficulties as someone who is uninsured due to chronic illness. Judging policy by the elimination of everyone who was ever uninsured may distort policy, leading people to favor expensive measures to take care of large groups when in fact a much smaller group is in real need of assistance. Obviously more people will be uninsured in any given month than over an entire year. It also makes comparisions subject to error introduced by various adjustments―other data used in the baseline estimates, like that from MEPS and the employer surveys, are on an annual footing. As a check, note that the 2005 CPS estimate for the uninsured in Colorado for 2003-2005 was 16.6 percent ± 1 percent.45 4.2.3 Odd assumptions about administrative costs almost certainly overstate the administrative cost savings from centralizing control. The Lewin model makes various assumptions about cost allocations. These depend heavily on assumptions about administrative costs. For example, 13 percent of total hospital costs are attributed to administration, a number arrived at based on conversations with the hospital industry. To allocate facilities costs to administrative and non-administrative functions, the baseline numbers allocate 13 percent of all expenses for plant maintenance, housekeeping, depreciation, and leasing and rental expense to administrative costs. The physician administrative costs used in the Lewin model are based on a survey of 335 physician practices conducted by the Medical Group Management Association. The survey is voluntary. Most of those who fill it out are members of the MGMA. Typical MGMA members are specialty groups of 3 or more physicians. The MGMA provided one of the speakers invited to address the Commission. It sells software and a variety of other services. 4.2.3.1 Lewin documents suggest that it used a subset of the 335 responses in developing baseline numbers for the Commission. Its June 15 report to the Commission says “We used the distribution of operating costs for non-hospital or IDS (Integrated Direct Service) multi-specialty practices” in the Western region. Based on whatever those cost numbers were, Lewin assumed that 10 percent of nurses’ time was administrative. Physician administrative expense was arrived at by “allocating costs to expense categories not directly attributable to providing patient care.” The document also says that “Building and furniture expenditures were attributed to administrative functions in proportion to the allocation of other physician costs to administrative functions (approximately 35 percent).” So, if a physician rents an office to see patients, 35 percent of his rent is
Catherine A. Frey. March 2000. “Wisconsin’s Uninsured Population:How Low—4% or 11.8%?” Population Health Institute, University of Wisconsin, 1,1. Online edition, assessed July 11, 2007. http://www.pophealth.wisc.edu/UWPHI/publications/briefs/march00brief.htm 45 U.S. Census Bureau, http://pubdb3.census.gov/macro/032006/health/h06_000.htm
44

24 allocated to administrative costs. Other examples of physician costs attributed to administration in developing the Commission baseline data include medical record costs, employee staff benefits, general administration, information technology expenses, expenses for furniture, housekeeping, and insurance premiums. 4.2.3.2 Other difficulties include getting an idea of hospital cost ratios. The Commission model gets its hospital information from a databank maintained by the Colorado Hospital Association. The data set includes “general, financial and utilization information at the facility level for 62 Colorado hospitals in fiscal year 2004.” The problem is that hospital payment to cost levels are calculated using hospital charges. Hospital charges often bear little relation to the price actually paid, one of the major problems in health care that has been exacerbated by government programs and other forms of third party payment. As the Colorado Hospital Association points out in its reference guide,
Charity care, bad debt, Medicare and Medicaid underfunding are defined below in terms of charges. Charges reflect expenses for providing care, plus an amount for underfunded and unpaid care and a margin for capital replacement, principal payments on long term debt, and other financial needs. Charges within a hospital must, by federal law, be the same for all patients for the same service. What a hospital actually collects can be quite different.46

Interpreting reported hospital charges is also difficult given that hospitals actively manage their financial reports. For example, there is evidence that non-profit hospitals adjust discretionary spending and accounting accruals to manage their earnings to a range just above zero. This makes many hospital financial reports difficult to interpret.47 4.2.4 Employer behavior and workforce data are limited and date as far back as 1991. The Lewin model uses the survey of employers conducted by the Kaiser Family Foundation and the Health Research and Education Trust (HRET) to model employer behavior. In 2005, the survey telephoned 2,013 firms nationwide. The response rate was roughly 50 percent. Its estimates say that 60 percent of the firms surveyed firms employ 3 to 9 workers. Firms with fewer employees are not represented because the survey sample was drawn from a Dun & Bradstreet list of employers with three or more workers.48 In 2004, the Statistics of U.S. Business from the Census Bureau reported that 79 percent of Colorado firms had 9 or fewer workers, including 20,183, or 16 percent, which had no employees at all. 49 These numbers suggest that the data used in the Lewin model may not accurately reflect the Colorado business climate. To create workforce statistics which are ultimately used to predict the amount that employers will save on health spending in the various reform proposals, the Lewin model statistically matches each MEPS household worker to the HRET firms. But some detailed information that affects health insurance coverage, things like age, sex, coverage status, policy type, and wage level, is not covered in the HRET survey. To create it, the
Colorado Hospital Association. March 2005. Reference Guide to Colorado Hospital Financial & Utilization Data 2002 & 2003. Colorado Health & Hospital Association, Greenwood Village, Colorado. p. 12. Online version accessed July 10, 2007, http://www.chha.com/download/referenceguidenp.pdf 47 See for example Andrew J Leone and R. Lawrence Van Horm. 2005. “How do nonprofit hospitals manage earnings? Journal of Health Economics, 24, p. 815-837. 48 The Kaiser Family Foundation and Health Research and Educational Trust. Undated. Survey Design and Methods, Employer Health Benefits 2005 Annual Survey, online edition accessed July 11, 2007. http://www.kff.org/insurance/7315/sections/upload/7315DesignandMethods.pdf 49 U.S. Census Bureau. Statistics of U.S. Businesses: 2004. Online version, accessed July 11, 2007. http://www.census.gov/epcd/susb/2004/co/CO--.HTM
46

25 Lewin model uses data from the 1991 Health Insurance Association of America employer survey data for “detailed” information on employer workforce.50 This adjustment is a reach. Using these data are equivalent to assuming that the Colorado population and economy have remained fairly static in the 16 years since 1991. 4.2.5 Data specifying individual insurance choice predate the introduction of HSA/HDHP policies, relying on 1987 to 1997 CPS data. In evaluating reform plans, the Lewin model estimates things like the number of people who drop private coverage to take up state coverage, the number of people eligible for Medicaid who actually enroll, how employers and employees respond to changes in the cost of insurance, and how employers decide to provide coverage using an unspecified “multivariate analyses.” Some of the parameters controlling this analysis are based on the 1997 Robert Wood Johnson Survey of Employer Characteristics and 1996 MEPS data on people offered coverage through an employer. Individual decisions to purchase individual coverage are modeled using an unspecified multivariate analysis of the likelihood that an individual will purchase coverage from the 19871997 CPS data. Premiums are imputed based on employer survey data. Needless to say, there is no room in these estimates for the effect of individual high deductible plans on either spending behavior or coverage take up. The new HSA qualified plans were not available in 1997. 4.2.6 Health spending projections have been projected from 2004 estimates and adjusted using 1998 Medicare data. Health spending projections by payer and type of service comes from 2004 estimates from the State Health Expenditure Accounts developed by CMS. These estimate spending on provider using Census surveys of service establishments and state tax data. The amounts were projected to FY 2007-2008 based on estimates from FY 2004, adjusted to eliminate double counting, adjusted to exclude “non-health items that are included in national health spending estimates,” and partly based on hospital financial reports. The projections were based on past ratios of the growth of Colorado health spending and U.S. health spending. Since state health expenditure accounts include spending by people from other states and exclude spending by Colorado residents outside of Colorado, the data were adjusted. The adjustment used is based on 1998 Medicare data. The bulk of the Medicare population is over 65. Another example of an adjustment that is difficult to follow in the Lewin model is the apparent use of CPS survey data on average and marginal tax rates for the households used from the MEPS survey. Why this is done is unclear as MEPS contains significant income data in its own right. Those income data match well with the CPS results. About 22 percent of MEPS income supplement data is missing wages, which are imputed from the employment section. 51 4.2.7 The estimates of the illegal population in Colorado are significantly less than those used by other sources. The original Lewin baseline estimates found that 167,000 of an estimated 785,200 Colorado uninsured are noncitizens. Using the 2000 Census data and CPS data from 2002, the Urban Institute estimated that there were
The Lewin Group. January 29, 2007. Summary Description of the Health Benefits Simulation Model (HBSM), Attachment B, p. 5. This is discussed in Attachment B, Summary Description of the Health Benefits Simulation Model (HBSM) dated January 29, 2007 and distributed to the Colorado Health Care Reform Commission. The comparison of CPS and MEPS results is outlined in Jessica S. Banthin and Thomas M. Selden. July 2006. Income Measurement in the Medical Expenditure Panel Survey, Working Paper No. 06005, Agency for Healthcare Research and Quality. Online version, accessed July 11, 2007, http://207.188.212.220/mepsweb/data_files/publications/workingpapers/wp_06005.pdf.
51 50

26 175,000 to 200,000 illegal aliens in Colorado.52 The Center for Immigration studies used the 2005 CPS to estimate that the illegal population in Colorado is 220,000, of whom an estimated 152,000 were uninsured. Including illegal aliens and their foreign and U.S. born children under age 18, an estimated 183,000 are uninsured.53 In all, the Center estimates that a fifth of the uninsured in Colorado are illegal aliens. One fifth of the Lewin estimate of 785,200 is 157,040 people, 94 percent of the estimate of Lewin’s estimate of the noncitizen uninsured. Omitting estimates of illegal aliens in the uninsured has implications for spending if a reform plan contemplates using federal Medicaid matching funds to defray Colorado health expenses. The reason is that Colorado cannot legally claim matching federal funds for non-emergency Medicaid health services provided to illegal aliens. Reform plans that use federal disproportionate share funds to pay for coverage extensions may also create significant problems for hospitals. The hospitals will still be required to provide services to illegal aliens under EMTALA, but the disproportionate share funds intended to compensate them for this will have been diverted to other uses. SECTION 5: An Overview of the Commission’s Operations In evaluating the Commission’s work, it is important to understand the depth of disagreement that persisted throughout the Commission’s term. Opinions on health care reform range from a belief that government programs like Medicaid and Medicare are a major cause of the problems in U.S. health care to those who believe that government control of health care is the cure for all that ails health care delivery. Virtually the entire spectrum of health care reform beliefs was represented on the panel. 5.1 The 208 Commission and Its Charge It is the opinion of the authors that limitations and vagaries kept the Commission from being as useful as it might have been. In fact multiple opportunities were missed that would have presented a more substantive range of solutions for legislators to consider. The following section provides background on how this happened. SB 208 required that the Commission: 1. Examine health care coverage and reform models “designed to ensure access to affordable coverage for all Colorado residents” with technical assistance and guidance from the project administrator 2. Select three to five specific health care reform proposals to meet the needs of the residents of Colorado. 3. Solicit reform concept papers and detailed proposals from interested parties. 4. Select the top proposals for detailed “technical analysis” by an independent consultant. 5. Hold at least one meeting in each Congressional district for the purpose of receiving public comment 6. Present a final report to the General Assembly. The Act also stipulated that the Commission was to be administered by a project administrator. That administrator was not appointed by the Commission; in fact the commission staff was hired before the commission first met.
Jeffrey S. Passel et al. January 12, 2004. Undocumented Immigrants: Facts and Figures, Immigration Studies Program, Urban Institute. http://www.urban.org/UploadedPDF/1000587_undoc_immigrants_facts.pdf, 53 Steven A. Camarota. December 2005. Immigrants at Mid-Decade: A Snapshot of America’s Foreign-born population in 2005. Center for Immigration Studies. Washington, DC. http://www.cis.org/articles/2005/back1405.pdf
52

27 The administrator was to “submit acceptable proposals as determined by the administrator to the Commission for discussion and the ultimate selection of three to five favorable proposals.” The process to identify “insurance reform proposals” was to “include, but not be limited to” an invitation for interested parties to submit proposals that followed the content proposals developed by the administrator, and any proposals that the administrator found acceptable.” What in fact occurred was that the Commission assumed the powers given to the project administrator. This made it less effective. The Commission focused on the characteristics of external reform plans submitted by groups that had an interest in, and the ability to outline, full blown reform plans. After that, during the summer of 2007, it focused on building its own reform plan. With a single exception the plans submitted uniformly proposed decreasing individual choice in health care and health insurance and increasing government control over individual health decisions. The proposal developed by the Commission followed the same pattern. 5.2 The Ambiguities in the Commission Charge In addition to the results created by limiting reform discussions to the plans submitted, other ambiguities in the statute created problems for the Commission. In Section 2, the SB208 charges the Commission with: 1. “studying and establishing health care reform models to expand health care coverage and to decrease health care costs for Colorado residents.” 2. examining “options for expanding affordable health coverage for all Colorado residents in both the public and private sector markets” 3. paying “special attention” to the “uninsured, underinsured, and those at risk of financial hardship due to medical expenses.” This section refers to “health coverage.” Some Commission members made an assumption that health coverage referred to health insurance. Others pointed out that health insurance is not the same as health coverage because having health insurance is no guarantee that someone can get medical care. Matters were further confused by the references to health care reform, health care spending, and health care in Section 1 and the reference to identifying “insurance reform proposals” in section 2, paragraph (4)(a). A Commission member who had attended the original committee hearings on the bill reported that legislators were not clear on the coverage versus insurance problem at the hearings, either. In practice the Commission operated as if its charter was to extend health insurance to all and that it was appropriate to assume that the mere extension of third party payment to all guaranteed appropriate medical care. As a result, the Commission operated as if its charge was to create a totally new mechanism for delivering health care and health insurance to everyone without regard for any upset caused by increasing government control of Colorado’s health care system. Commissioners who believed that this assumption was incorrect were unsuccessful in altering this operating mindset although they frequently pointed out that many people eligible for Colorado Medicaid, and on government plans in other countries, were unable to access appropriate medical care. 5.3 Operating Methods

28 The first few months of the Commission’s efforts were given over to what was described as “operational and procedural” discussions. The time spent on this frustrated some Commissioners because there was little to show for the time spent. Bylaws created by the Commission’s Operating Committee were brought forward and voted on. They required consensus decisions without explicitly defining what that meant. In retrospect, this blurred how decisions were reached. A phrase often heard at commission meetings was “When was that decided?” Decisions made by the Operations Committee affected agendas, timeframes, and invited speakers. As a supermajority was needed to change agendas put forth by the Operations Committee, it effectively controlled the items presented for Commission consideration. Speakers invited to address the Commission were, with one or two exceptions, from organizations devoted to increasing government control of health care, people affiliated with organizations that collected the data used in the Lewin model, or people presenting the results of polls exploring people’s attitudes towards health care reform. On one memorable day, a speaker representing the Colorado Progressives was invited to talk for almost ½ hour on the results of a poll that surveyed people who attended Colorado Progressives meetings on the kind of health care reform they preferred. The sample consisted of 200 self-selected people who, to no one’s surprise, wanted a single payer health care system. The presentation was so patently political that more than one Commissioner noted that this kind of presentation was completely inappropriate. The compressed time frame under which the Commission operated severely limited thoughtful discussions. There were significant delays between commission meetings and the publication of that meeting’s minutes. This made it difficult to ascertain whether the minutes were full and accurate transcriptions of all matters. Subcommittee meetings did not, in most cases, produce minutes. 5.4 Lack of Agreement Concerning Basic Health Care Facts The most serious flaw in Commission procedure were produced by its failure adopt any of the standard academic or legal methodologies for establishing agreement about basic facts. As a result, its activities were illsuited to building the consensus that its bylaws envisioned. Quite often, statements and assertions were made during Commission deliberations that reflected nothing more than anecdotes or “popular wisdom,” a problem which likely resulted in availability bias in Commission recommendations. For example, a fundamental assumption in the 208 legislation is that the uninsured pose a problem for the health care delivery system, and that this problem will be mitigated if all of those people have a third party payer for their health care bills. As discussed above, it is not clear that a third party payer will reduce expenditures because coverage increases utilization and reduces self-payment. Furthermore, there is considerable evidence suggesting that the uninsured are less of a burden to the health care system than government programs, the main source of uncompensated care. The Commission did not consider questions like this. Nor did it consider who needs care and is not getting it, how many of those people there are, and whether they are the same as the uninsured. If Commission recommendations are poorly targeted at helping the needy, it is because the Commission did not see this as its charge. The authors of this report believe that Commission discussions would have benefited from a well organized fact finding period. Its deliberations were frequently marred by popular claims that have little or no support when subjected to close examination, or in available academic studies. A few examples of these claims are given in Table 5.4.1.

29
Table 5.4.1: Popular claims and alternative evidence
Claim People without health insurance have no access to care Other viewpoints Among those with comparable incomes, the uninsured get about the same amount of health care as those with insurance. The largest amounts of uncompensated care are generated by Medicare and Medicaid patients. This occurs because Medicaid and Medicare pay providers less than cost. Eliminating the uninsured by putting them on Medicaid may actually increase the amount of uncompensated care by eliminating the payments that the uninsured make for their own care, increasing utilization, and increasing administrative overhead. In the 1990s, Tennessee insured everyone in the state under the TennCare program. The program was supposed to eliminate uncompensated care. By the late 1990s, uncompensated care had increased. 54 In Colorado, a 40 year old woman can choose from a number of comprehensive health insurance policies that cost less than $100 a month. Adding two children adds about $50 to $100 a month. The most that woman would have to pay for health insurance, regardless of health status, would be $425 a month under Cover Colorado, the state’s insurance plan for the uninsurable. Recent papers55 suggest that Medicare administrative costs are similar to those in the private sector even without including the administrative costs of Medicare supplemental policies or the fact that Medicare is not solvent. Overhead is not necessarily bad. It includes case management for patients with chronic conditions, health education expenses, fraud detection, and customer service, areas in which Medicare is notoriously weak.56 In 2002, the Washington State Office of the Insurance Commissioner determined that administrative expenses for companies filing annual statements with the state averaged 12.6 percent of overall revenues. This was the definition of the estimate of the uninsured provided by the Lewin Group, the modeler hired by the Commission. Obviously there are a number of different estimates of the uninsured including those who are uninsured for a year or more, those who are covered by Medicaid but have simply not enrolled, and those who are uninsured by choice because they believe that they do not need health insurance. Evidence from existing system suggests that results are mixed and that there are significant concerns with record availability, accuracy, and security. Where evidence-based medicine decision processes have been implemented, there is evidence that they are used to control costs by denying access to effective therapies.

Insuring people will eliminate uncompensated care

Health insurance is unaffordable for individuals.

Medicare has lower administrative costs than private insurance plans; private insurers have administrative costs of 30 percent.

People who are uninsured for even one month should be counted as uninsured.

Mandating electronic health records will lower costs and improve quality Mandating evidence-based medicine will lower costs and improve quality

Data from the RAND Health Insurance Experiment suggest that “we estimate that with no insurance at all, people would have spent about half the cost of free care.” Emmett B. Keeler. Summer 1992. “Effects of Cost Sharing on Use of Medical Services and Health, Medical Practice Management, P.318. Online version accessed December 10, 2007. http://www.rand.org/pubs/reprints/2005/RP1114.pdf 55 See Merrill Matthews. January 10, 2006. , Council for Affordable Health Insurance, Alexandria, Virginia. http://www.cahi.org/cahi_contents/resources/pdf/CAHI_Medicare_Admin_Final_Publication.pdf; Banjamin Zycher. October 2007. Comparing Public and Private Health Insurance: Would A Single-Payer System Save Enough to Cover the Uninsured?, Medical Progress Report No. 5, Center for Medical Progress, Manhattan Institute, New York, New York. http://www.cahi.org/cahi_contents/resources/pdf/CAHI_Medicare_Admin_Final_Publication.pdf 56 JP Wieske.May 2007. High Loss Ratios Undermine Affordable Health Insurance. Council for Affordable Health Insurance, Alexandria, Virginia. http://www.cahi.org/cahi_contents/resources/pdf/n141lossratio.pdf

54

30
The State of Colorado has experimented with disease management. It lowers costs in some cases but not in others. Case management for expensive events like trauma is already routine for private insurers. Ongoing experiments concentrate on managing some chronic conditions known to generate avoidable costs. Not all higher spending is waste. Wealthier people spend more on health to improve functioning just as they spend more on housing, transportation, and entertainment. Countries with lower levels of health care spending have worse health outcomes than the U.S. along a variety of measures. Within the U.S., vacationers admitted to the emergency room in high spending areas have lower mortality rates than similar visitors in lower-spending areas. The medical literature shows the opposite. Disparities between health care access for the rich and poor are lower in the U.S. than in other countries. A few examples of comparative outcomes include: lower infant mortality rates in the U.S., higher cancer survival rates, better population blood pressure control, lower mortality and morbidity from cardiac disease, better diabetic treatment, more preventive care, and better health and quality of life for spinal cord injury patients. Compared to the NHS, US medical care provides more services for roughly the same expenditures. Spending on the indigent has risen significantly and there is little evidence of positive effects. It may be time to study how money is spent rather than simply spending more. Integrated health care systems have raised costs in states such as Wisconsin, where hospital networks use primary care practices as feeders to their higher margin hospital services and as barriers to competition. The State of Colorado has determined that Medicaid managed care costs more than its current fee for service system,57 possibly due to higher overhead costs. Buying insurance for expected expenses is the most expensive way to purchase them. Lower deductibles come with higher premiums. Someone spending $10,000 on health insurance with a $500 deductible might be able to buy a policy with a $5,000 deductible for $5,000 a year and save the remaining $5,000 in a tax free health savings account. How does the higher deductible make him worse off? A recent look at a census of all frequent users of Massachusetts emergency rooms suggests that ER use by the uninsured is roughly the same as for the privately insured. The Urban Institute has concluded that the uninsured do not use emergency rooms at a higher rate than the insured.58 If this were true, the Soviet Union would have had the lowest costs in the world. In fact, smaller systems tend to have lower administrative costs. Counter evidence for the superiority of competitive systems includes a comparison of the Northern California Kaiser Health Plan with the British National Health Service. The researchers found that costs were comparable but that Kaiser provided more for the money.

Care coordination and case management will lower costs

Because the U.S. has the highest per capita health care spending, it “spends too much on health care.”

The U.S. health system spends more money and has poorer outcomes than health systems in other countries.

More spending on the indigent will improve health outcomes

Integrated health care systems will lower costs.

People are better off if their health insurance policies have lower deductibles and pay for routine care.

The uninsured get their care at the emergency room driving up costs for everyone.

Centralizing administration will lower costs.

In its December 2006 Joint Budget Committee hearings, the Colorado Department of Health Care Policy and Financing wrote that “Although managed care organizations should experience savings over fee-for-service due to their improved ability to reduce unnecessary hospitalizations, emergency room visits, and other overutilization, there are also extensive administrative costs for care management, utilization management, providing networking to ensure access, and other processes such as bill paying and risk management.” Colorado Department of Health Care Policy and Financing. FY 07-08 Joint Budget Committee Hearing. page 55. Online version, accessed October 16, 2007. http://www.chcpf.state.co.us/HCPF/Budget/jbc%2007-08%20hearing/FY%200708%20HCPF%20Hearing%20Agenda%20and%20Response_new.pdf 58 Stephen Zuckerman, Yu-Chu Shen. February 1, 2004. Do Insurance Coverage and Access to Care Matter? Medical Care. 42(2), 176-82.

57

31
There is a great deal of evidence showing that for-profit entities minimize cost better than non-profit entities. Competitive markets generally make price increases difficult. When that happens, only cutting costs will generate certain profits. In some cases, the efficiencies created by the drive to minimize costs allow for-profit firms to provide services that are better and less expensive than their non profit competitors even though the for profit entities must pay higher taxes and shareholder dividends. There is no evidence that health insurers are making abnormal profits. With the exception of some childhood immunizations, preventive care increases expenditures. This is because most preventive care consists of screening for early detection of diseases that are less expensive to treat if caught early. While screening lowers individual risk, it increases overall expenditures because the savings from the relatively small number of early cases detected are smaller than the total costs of screening the population. This is why there is more preventive care screening in the U.S. than in government run health care systems. Individuals are more likely to pay more to lower their own risks, government accountants are more likely to be concerned with total expenditures.

Insurance company profits increase the cost of care

More preventive care will save money.

Consensus was elusive on such basic facts such as how one should measure the uninsured, whether state taxpayers should pay for insurance policies for illegal aliens, how private insurance markets operate, and the current cost of health insurance in Colorado. In the end, virtually all disagreements ended up being settled by a vote or by a ruling by the Chairman. 5.5 The Commission’s Guiding Principles Early on, the Commission voted to adopt a number of principles for health care reform. The principles were notable for the fact that they contain undefined terms that are central to any discussion of health care reform. The Commission did not grapple with the fact that the value placed on many of its reform principles will vary with individual preferences. It is therefore likely that any reform satisfying them would be required to increase system flexibility so as to maximize innovation and individual choice. If this were the case, adherence to the principles adopted by the Commission would preclude most of the reform proposals recommended by it.

Colorado Health Care Reform Commission Principles for Health Care Reform 1. 2. 3. Protect and improve the health status of all Coloradans. Expand coverage of essential health care services for all Coloradans, with an emphasis on the uninsured and underinsured. Align incentives to provide high-quality, costeffective and coordinated care.

Undefined concepts in Principles for Health Care Reform Health status Coverage Essential health care services Underinsured 59 Uninsured High-quality Cost-effective

Although uninsured would seem to be a fairly clear term, some people define uninsured as someone without health insurance for even a day in a given year. The term is often used to include people who are in fact eligible for government health care programs should they need them, but who have not signed up. They do have someone else to pay for their health care, but are technically uninsured. Uninsured also refers to people who have significant assets that they can use for care and so have no need to purchase insurance from a third party.

59

32
4. 5. 6. Support a system that is financially viable, sustainable and fair. Provide opportunities for meaningful choice and encourage personal responsibility. Emphasize wellness, prevention, health education and consumer empowerment. Coordinated care Financial viability Sustainable Fair Meaningful choice Personal responsibility Wellness Consumer empowerment

Without knowledge of what individuals value, is difficult to determine how one measures health status, what defines essential health care services, how one defines the underinsured, how one measures high quality care, how one defines cost-effective, how one determines what is fair, how one measures financial viability, what constitutes meaningful choice, what is encompassed in personal responsibility, and what constitutes consumer empowerment. The Commission did not define coordinated care, consider what initiatives constitute wellness, or outline what was meant by a “sustainable” system. It also failed to ask whether there is evidence to support the effectiveness on additional spending on health education in view of the fact that people with chronic conditions do get health education and the results suggesting that many health education programs have at best a weak effect on the outcomes at which they are targeted.60 That Commission deliberations would have benefited from more substantive discussions of these issues is evident. For example, those in favor of extending community rating and guaranteed issue to Colorado’s individual insurance market also favored a mandate requiring that everyone purchase insurance. The reason given for this was that since community rating would raise the cost of insurance for young people, they would drop policies and increase the number of uninsured unless they were forced to purchase the higher cost policies. Those who are older typically have higher incomes and greater assets. Whether it is fair to overcharge the young in order to benefit those who are older and wealthier was never systematically examined. The question of individual mandates and minimum acceptable coverage are examples of other generally accepted reform principles that violated Commission principles. Many Commissioners supported an individual mandate to require people to purchase health insurance. At various times, Commissioners gave various reasons for their enthusiasm for this idea. The Commission never discussed whether this was fair or whether it diminished individual choice and consumer empowerment, two principles that were said to be important. The Commission also appears to have violated its principles by ignoring most financing questions about the health care reforms that it did examine. The lack of agreement on the core principles was to bedevil discussions throughout the Commission’s tenure. It was a particular problem because a number of Commissioners were simply unwilling to seriously consider any reforms intended to increase the diversity of available insurance plans, place more emphasis on cash payment, deregulate the medical sector, or systemically reform Medicaid. Only two changes in Medicaid were seriously considered: its expansion and the possibility that some provider reimbursements would have to be raised. This unwillingness was particularly unfortunate in view of recent evidence documenting early successes of higher deductible, lower premium plans in increasing insurance coverage, reducing cost, and improving health. It prevented the Commission from exploring the notable success that cash practices are having in reducing costs in other states. It prevented the Commission from exploring the success that some specialty hospitals are having
The evaluation of the outcomes of health education programs finds that many of the studies on which the enthusiasm for health education are based are relatively weak. A 2002 Cochrane Database Systematic Review of school-based smoking prevention programs by Thomas and Perara, for example, found that roughly half of good quality studies show that smoking education reduces smoking.
60

33 in lowering costs and improving outcomes by focusing on a particular type of medical procedure. This will likely be a particular problem in Colorado. Its legislature recently followed the poor example of Medicare in outlawing specialty hospitals in which physicians have a financial interest. What was not recognized in these principles, and what is central to any sensible discussion of health care reform, is that the definition of such things as fairness, quality, essential services, and meaningful choice varies from individual to individual and depends on whether or not the individual in question has to pay for the service. As the RAND health insurance experiment clearly showed, the average person significantly alters behavior when he must use his own money to pay for health care, and these alterations have no discernable effect on health. Having passed principles that were not well defined, the Commission then proceeded to a set of discussions on what were said to be “Key Questions” for health reform. These questions, and the answers to them, were supposed to provide a basis for choosing among various health care reform options. On Friday, June 15, 2007 an email was sent to Commissioners and other parties asking that ideas for key questions be mailed in by Monday, June 18th for consideration at the meeting on Tuesday June 19. Staff summarized “key threads” in the questions mailed back. The key questions were chosen from those threads using an exceptionally fuzzy process at the unusually confusing June 19,th 2007 meeting. The key questions which were developed by Commission staff are presented in the following table. Note that none of the Key Questions deal with substantive system reform, cost minimization, or the reform of government programs. Although the answer to question 1 includes the comment individuals can use “resources” to pay for health care, by the answer to question 2, those resources are discounted and individuals must have insurance because they have “responsibilities to the larger community.” No philosophical justification for this statement is given. Given that the FY 05-06 Colorado budget included almost $2 billion in spending on health care subsidies for roughly 400,000 people in Medicaid and SCHIP programs, question 7, “Will there be subsidies to assure affordable coverage,” indicates the lack of a factual basis. To further illustrate the limited nature of the Commission’s “Key Questions,” an alternative list of key questions is provided in Appendix 1. This list, which has been distributed nationally, was submitted to the Commission by one of the authors of this report when the Commission asked Commissioners to submit suggestions for key reform questions.

Tentative Answers to Key Questions th July 18 , 2007 Commission Meeting 1. What will be the role of the individual? Individuals have responsibility for being able to pay for care, through resources or insurance. Individuals have the responsibility for taking care of themselves (e.g., through healthy lifestyles), and for using the health care system appropriately. Individuals have responsibilities toward the larger community. 2. Will there be an individual mandate? Yes, although other decisions will affect this decision, such as the minimum benefit package and subsidies for low-income individuals. 3. What will be the role of employers? Employers will not be required to provide coverage to employees, but they

34
may be required to contribute toward a health care reform plan. Possible roles include arranging 125 plans and payroll deductions for their employees, or contributing revenue for subsidies or uncompensated care. 4. What will be the role of government? Government will provide subsidies for purchase of private coverage or to reform/ expand Medicaid and CHP+. Government will continue to provide services such as public health services and support for safety net providers. Will there be an expansion and/or reform of the Medicaid and CHP+ programs? There may be an expansion and reform of the Medicaid and CHP+ programs. 6. Will portability or continuity of coverage be assured? The Commission did not have the time to fully consider this question. th (Portability will be considered at the August 13 Proposal Committee meeting.) 7. Will there be subsidies to assure affordable coverage? Government will provide subsidies to assure affordable coverage 8. What will the minimum benefits be? There will be a minimum benefit package for all of those who are subject to the mandate. The minimum benefit package may be an actuarial value for a package, allowing for flexibility in plan design. A separate minimum benefit standard may be created for those who qualify for a subsidy, to assure that these individuals have affordable out-of-pocket costs. 5.

5.6 Missed Opportunities The work of the Commission was limited by its self-imposed charge. If a proposal for a particular method for health care reform was not submitted, the Commission did not consider it. It was unwilling, in the development of its 5th proposal, to consider regulatory and Medicaid reform initiatives being experimented with in other states, ideas for funding and reimbursement reform, and current efforts to eliminate waste. In May 2007 when the Commission first selected 4 proposals to model, it was driven by the desire the present a range of options. The options that were selected proposed the following plans: 1. Expanding Medicaid/CHP+; no other market alterations except offering a limited benefit medical plan like those already available, with subsidies to cover the working poor 2. Impose an individual mandate, expand Medicaid/CHP+, subsidize a limited benefit medical plan for the working poor, use Medicare reimbursement rates for provider reimbursement. 3. Individual mandate, employer mandate, state managed guaranteed issue insurance products, Medicaid/CHP+ expansion. 4. Have state government take over and pay for all Colorado health care. The authors believe that the most appropriate use of the 5th proposal position would have been to expand this range of proposals to investigate the possible effects of deregulation and government program reform. A market oriented reform proposal was included in the original submissions. It was not selected for a variety of reasons,

35 some related to the proposal itself. At the time, some commissioners hoped that the 5th proposal option would be used to explore market based reforms. Once the 5th proposal committee began to meet, market oriented reforms were never seriously considered. In November 2007, the Commission even voted against recommending that the legislature study market oriented reforms while voting for legislative study of a single payer plan. SECTION 6: Further Recommendations Changes in the delivery system for health care must, of simple necessity, be considered comprehensively and the implementation of changes phased into place. These are some of the recommendations that we believe should have been considered. 6.1 As government assistance forms the basis for most fee schedules, including private insurance, submit Medicaid and CHP+ to true reform. Use the Section 1115 and HIFA waiver system to institute greater flexibility and innovation within the delivery system of these programs so that existing dollars can be optimized. Use these dollars to purchase private health care coverage, except in the cases of the most sick and vulnerable for which dedicated funds (medically needy) be established. Revise the benefit schedule to more realistic coverage levels. Permit and expand CDAS types of programs which incent people to use their health care dollars wisely; 6.2 Attack waste, especially that caused by government regulation. The area of duplicative services offers an excellent opportunity for collaboration as one point of departure. 6.3 As the above processes are underway, the reform of the current regulatory and administrative system that oversees the private health insurance be completed to achieve lowest possible costs of compliance, and to ensure that available premium dollars are optimized for health care delivery. A list, without discussion or any particular order, is provided below. It primarily concentrates on regulatory reforms that the state can affect that will lower premium costs by reducing administrative overhead. There are a few suggestions that will also increase the competitiveness of the health care market. 1. Reconsider regulations that require that insurers must charge the same premium for policies covering 1 or 12 children. Right now policies are priced at 2.8 to 3.2 children per family. People with fewer children, particularly single mothers with one or two children, pay more per child than others. 2. Maternity coverage should not be mandated for people who do not need it 3. Colorado should make sure that waiting periods in its laws harmonize with Federal law. Harmonization will decrease administrative overhead. 4. Non-network physicians, (generally these are specialists like radiologists and anesthesiologists) get paid whatever they charge under the network adequacy laws in order to protect the individual policy holder. As a consequence, overall rates for non-network care are passed along in the overall rate structures (millions of dollars per year) and premiums go up. Colorado law should be changed to allow negotiated networks in network facilities — that is, any provider giving service in that facility should only be paid reasonable and customary charges, which would allow lower insurance policy premiums, thus favoring policyholders. 5. Hospitals are protected from the “general contractor” rule by allowing physicians “privileges” to practice in their facilities (related to #4). Changing this rule should be studied. 6. Recent statute requires that private policies pay for court ordered mental health treatment. Courts are not accountable for the costs that they impose on others. The legislature should revisit this.

36 7. Recent statute requires that private policies pay for self-inflicted injuries due to the influence of alcohol and controlled substances. This drives up costs for responsible policy holders, and should be revisited by the legislators. 8. Insurance mandates should be revisited. In 2007 Colorado had 46 mandates. Arizona had 29, Indiana had 34, Kansas had 37. 9. Reconsider the mandate requiring that private policies pay for early childhood disability evaluation. 10. Statute provides that small groups under 15 may self-insure for maternity, however Colorado’s Department of Insurance interprets a 1980s era lawsuit (Budde) in a way that effectively prohibits small groups from doing so. A close reading of the decision makes this regulatory stance questionable, but the current effect is higher premium pricing for all small groups 11. Consider eliminating the statute that prohibits list billing; 12. Reconsider the statute rescinded rating flexibility for small groups. This increases the effect of the community rating straitjacket and will increase premiums; 13. Examine Department of Insurance regulations that deviate from the National Association of Insurance Commissioner standards, which are intended to decrease administrative costs by increasing uniformity across states; 14. Reconsider regulatory restrictions on solving the retro term problem; 15. Reconsider state continuation regulations. They are not the same as COBRA, which causes significant administrative problems; 16. Reconsider regulations requiring that men and women be charged the same amounts regardless of their use of health care. Men cost less from roughly age 18 to age 50. After age 50 they cost more. Premiums should reflect this to effectively communicate differences in health care usage to policy holders. 17. State law prohibits short term medical plans in excess of 6 months, with a limit of 2 per 12 month period. Reconsider those limits. 18. Common ownership restrictions create excess administrative costs, and loss of opportunities to create larger risk pools. 19. Consider reforms of regulations that create fragmented risk pools 20. Reconsider state statutes making it illegal for physicians to have financial stakes in specialty hospitals. As physicians are highly qualified to invent a better delivery model, prohibiting them from doing so limits competition and innovation. Concerns about conflict of interest can be handled with disclosure requirements. 21. Consider allowing insurers in the individual market to offer mandate-free or mandate-light insurance policies, perhaps on an experimental basis of a few hundred or few thousand policies per year, or, alternatively, to those policyholders with HDHL/HSA policies. 22. Consider allowing Colorado residents to purchase health insurance from any insurer authorized to do business in any state, not just those licensed in Colorado. Once government assistance programs and payments are brought into line, once waste within the system is properly addressed, we will find that the private coverage system will adapt to the new, more efficient realities of the overall health care delivery system and we will find stabilized and relatively lower costs. In addition, the private coverage system will further adapt and innovate, bringing products to consumers that are more sensitive and useful, as well as being more cost effective.

37 Appendix A: Key Health Care Reform Questions That Should Be Asked 1. Does the proposal organize the health care system so as to provide maximum value to those who use its services, with value defined from their point of view? 1.1. Pricing 1.1.1. Does the proposal further market pricing for medical services? Does it rely on price controls of any kind, including administrative price setting? 1.1.2. Does the proposal ensure that any physician or health provider, and any facility, is free to treat any patient in exchange for direct payment of a mutually agreeable fee? 1.2. Outcomes 1.2.1. Does the proposal ensure that patients can determine the treatments they will receive and physicians the treatments they will provide subject to their own consciences? 1.2.2. Does the proposal include organizational provisions that ensure that firms, industries, professions, and subsidy recipients will not be able to use the reform plan to their financial advantage? 1.3. Consumer protection 1.3.1. Does the proposal ensure that participation in government programs is voluntary? 1.3.2. Does the proposal encourage people to accumulate assets that may be used for future health care expenses in lieu of third party insurance? 1.3.3. Does the proposal allow people to modify the amount of financial risk they are willing to bear by choosing among different third party insurance policies as their circumstances change? 1.3.4. Does the proposal remain neutral with respect to the form that third party insurance should take as long as insurers can meet their contractual obligations? 1.3.5. Does the proposal remain neutral with respect to paying for health care with cash or with third party insurance? 1.3.6. Does the proposal subject businesses operating in health care to the same rules as businesses operating in other sectors of the economy with respect to anti-trust, ownership, pricing, contracting, and reporting requirements? 1.3.7. Does the proposal protect people from involuntary participation in any non-governmental insurance program? 1.3.8. Does the proposal allow the purchase of health insurance that is not associated with an employer? 1.3.9. Does the proposal ensure that people can buy health insurance from any insurance company approved by a state government?

38 1.3.10. Does the proposal allow for the fact that people purchase health care from a variety of sources, some of which are both outside of Colorado and outside of the United States? 1.3.11. Does the proposal protect consumers from arbitrary restrictions on their ability to access medical therapies? 1.4. Government obligations 1.4.1. Does the proposal include mechanisms to ensure that government programs do not use government power to compel unpaid services from providers? 1.4.2. Does the proposal have mechanisms to ensure that government treats all providers fairly and does not discriminate between providers via different payments for the same service or regulatory structures that favor some providers over others? 2. Does the proposal contain adequate structures for reducing costs? 2.1. Does the proposal ensure that all providers and third party payers in the health care systems are subject to credible competitive threats? 2.2. Does the proposal expose existing providers, including government and quasigovernment entities, to competitive pressures? 2.3. Does the proposal ensure that all entities using or providing health care are free to contract with others as they see fit? 2.4. Does the proposal ensure that participation in any health care program under the control of Colorado state government, or any entity created by statute, is voluntary? 2.5. Does the proposal ensure that any physician or health provider, and any facility, is free to treat any patient in exchange for direct payment of a mutually agreeable fee? 2.6. Does the proposal ensure that for profit and non-profit providers are treated equally? 3. Regulatory reform 3.1. How does the proposal plan to determine which health care regulations produce a net benefit and which produce a net cost? 3.2. Does the proposal embrace legal reforms that protect participants in the Colorado health care system from unreasonable torts and contradictory regulations? 3.3. Does the proposal require that businesses operating in health care are subject to the same rules as businesses operating in other sectors of the economy with respect to things like anti-trust, ownership structure, pricing, contracting, payment, purchasing, taxation, and reporting requirements? 3.4. Does the proposal protect consumers from unreasonable charges? 3.5. Does the proposal contemplate legal reforms that would encourage all participants to exercise good judgment?

39 3.6. How does the proposal plan to determine whether current licensing, inspection, and reporting requirements produce net benefits? 3.7. Does the proposal contemplate legal structures that will protect providers from arbitrary and capricious peer reviews? 3.8. Does the proposal reduce legal barriers to entry affecting hospitals, specialty hospitals, long-term care providers, in-store medical practices, insurers of all kinds, providers of professional services, drug and device manufacturers, and suppliers of drugs and medical equipment? 3.9. Does the proposal contemplate the legal reforms that would be necessary to encourage people who wish to create charity care clinics can do so without risking their personal assets? 4. Does the proposal promote the use of economically efficient subsidies designed to maximize the general welfare? 4.1. Does the proposal reform Medicaid? 4.1.1. Do Medicaid subsidies accrue to individual patients rather than to providers? 4.1.2. Can individual Medicaid patients spend the money that they receive at the provider of their choice? Can they purchase necessary supplies and services from the supplier of their choice? 4.1.3. Does the proposal contemplate regulatory reform that allows the program to develop regulations and programs that treat different Medicaid populations according to their needs? 4.1.4. Does the proposal contemplate Medicaid reforms that encourage Medicaid clients to use their Medicaid benefits wisely? 4.1.5. Does the proposal include public access to Medicaid financial data so that amounts paid to providers, vendors, consultants, administrators, contractors, overseers, investigators, tax collectors, auditors and so on, as well as the purpose of the expenditures, can be clearly discerned? 4.1.6. Does the proposal provide ways to discriminate between—and effectively manage—financial arrangements for people in legitimate need and those who take unfair advantage of subsidized and safety net programs? 4.1.7. Does the proposal ensure that taxpayer-funded services will be provided only to eligible persons for eligible services? 4.1.7.1.How will the proposal ensure that taxpayer-funded services are not provided to deceased persons, persons with fraudulent identification, nonresidents, persons not meeting financial requirements, illegal aliens, and so on? 4.1.7.2.What penalties will be assessed for those who try to defraud the system by faking evidence of eligibility? 4.1.7.3. What mechanisms in the proposal are designed to ensure that payment for taxpayer-funded services is actually rendered?

40 4.2. How does the proposal contemplate providing medical care for people who, by reason of incapacity or simple cussedness, do not comply with administrative requirements? 4.3. Will the subsidies contemplated by the proposal encourage or crowd-out private mechanisms for financing medical services? 4.4. Does the way subsidies are distributed in the proposal deepen Colorado’s “lowwage trap” by imposing effective marginal tax rates on low-income people trying to work their way out of dependency? 4.5. How does the proposal plan to distinguish between essential and non-essential health care services? 4.6. How does the proposal contemplate ensuring that taxpayer-funded programs provide good value for the money spent? 4.7. Given that funds for taxpayer-funded programs are limited, how will the proposal manage the tradeoffs that are necessary in a resource constrained subsidy program? 4.8. How does the proposal propose to measure the effectiveness of taxpayer-funded subsidy programs? 4.9. How does the proposal plan to determine the type and level of subsidies? 5. Programmatic considerations 5.1. Does the proposal have a sunset provision? 5.2. How does the proposal plan to measure whether it is a success? 5.3. What trigger mechanisms automatically sunset the proposal in the event of budget excesses or poor performance?