STRANGE BEDFELLOWS

Governor Rendell unveiled Prescription for Pennsylvania amidst much national and local fanfare on January 17, 2007 in the spacious and historically decorated conference room adjacent to his Capitol office in The Harrisburg, Pa. State Capitol Building. "We can no longer stand by while health care costs spiral out of control,” he said at the time, promising to expand “access” to affordable health care to most of the “767,000 adult Pennsylvanians” who lacked health insurance. That is the number of uninsured Pennsylvania adults the Governor has cited ever since (though the number mysteriously morphed down to 747,000 in the PABC initiative launched earlier this week). It is a number the media has accepted virtually on faith and seemingly never challenged, or questioned, but more about that later.

There was something of a carnival atmosphere in that large, yet surprisingly packed chamber that day. The media was there en masse, but they were clearly outnumbered. The room was virtually swarming with men and women in business attire whose round colored pins identified them as lobbyists, mostly from the health insurance and related healthcare industries. The Governor also chose to have someone by his side on the podium that day to share this historic moment with him. It wasn’t another politician or some trusted aide, nor his wife, nor some other loved one. It was Anita M. Smith, CEO of Capitol Blue Cross, a major Pennsylvania Health Insurance company with headquarters in Harrisburg. Although the Governor introduced Ms. Smith to the assembled crowd very early in his opening remarks few subsequent media accounts of the press conference mentioned her prominent role at the Governor’s side. The media also took little interest in the generous campaign finance support Governor Rendell had received from the health care industry including well over $1 million in contributions to underwrite the $2.5 million cost of his inaugural ceremonies held the same week he unveiled his health care initiative. Such donations are not regulated by the State Election Bureau because they are not considered campaign contributions. Capitol Blue Cross, Independence Blue Cross, Keystone Health Plan, United Health Group, Highmark Blue Shield and the University of Pittsburgh Medical Center each contributed $50,000 toward the inaugural festivities while Blue Cross of Northeastern Pa,, GlaxoSmith Kline and Shire Pharmaceuticals contributed a more modest $25,000. One might be forgiven for wondering whether the Governor’s efforts to bring so-called “universal” health care to Pennsylvania would inflict much pain on the health insurance or pharmaceuticals industries, both of which have contributed toward making American Healthcare by far and away the most expensive in the world while leaving 47 million uninsured and millions more underinsured.

HOW MANY UNINSURED? GOVERNOR RENDELL’S FUZZY MATH

To hear Governor Rendell tell it Pennsylvania is pretty well off when it comes to health care, having one of the lowest uninsured rates in the country. The fact that his estimate of 767,000 uninsured adults is based upon a highly dubious 2004 survey and is contradicted by other surveys that utilized far more orthodox methodology does not seem to bother him, nor the media, which has apparently has never heard of the Census Bureau or of Google.

The uninsured statistics Rendell regularly cites come from a survey commissioned by the Pennsylvania Insurance Bureau in 2004. The methodology utilized in that survey borders on the bizarre if the true goal was to get accurate and reliable estimates. This writer feels compelled to wonder if the true objective in commissioning that survey was not a quest for accurate data, but rather an attempt to come up with more conservative numbers than had been arrived at by the Census Bureau, numbers that would make the problem appear less severe than it is, and hence make the Governor’s proposed reforms appear to be more wide-reaching than they really are.

For starters, the survey estimated that 8% of the Pennsylvania population, 900,000 people, was uninsured. The Governor’s lower number of 767,000 represents that number less the 133,000 estimated uninsured non-adults. It is unclear why Governor Rendell chooses to include only the number of adults when citing the numbers of uninsured. Those non-adults are, after all, uninsured. The survey itself was conducted by the research group Market Decisions, utilizing a telephone sample frame and targeting 100 households in each of Pennsylvania’s 67 counties (except for Philadelphia whose sample was 173 households). Thus Forest County with a population of 4,946 was targeted for 100 interviews (1 in every 49 of the households in the county); Alleghenny County (Pittsburgh), with a population of 1,281,666 was also targeted for 100 interviews (1 in every 12,817 households); and Philadelphia County with a population of 1,517,550 accounted for 173 interviews (one in every 8,772 households). The sample frame itself poses a problem because non-telephone households tend to be poorer than the general population with a larger percentage of ethnic minorities – the very demographics most likely to lack health insurance. A study that specifically examined this issue back in 1990 found that non-telephone households were three times as likely to be uninsured than telephone households and concluded that “there are marked differences between the telephone and nontelephone groups, and adjusting the former is unlikely to result in reasonable overall population estimates or to lead to increased understanding of being uninsured. In this case, telephone ownership appears to introduce a bias unreconcilable by recourse to social demographic and health status measures.” Translation: Any survey attempting to determine the extent to which a population is uninsured that utilizes a telephone sample frame will understate the extent of the problem.

Moreover, the spreading out of the sample equally across the state’s 67 counties regardless of their population virtually guaranteed a radical under-sampling of the inner city neighborhoods in Philadelphia and Pittsburgh (as well as other smaller cities throughout the state) where residents were most likely to want for health insurance. Finally, Market Decisions interviewed only one individual per household, but collected responses relating to every member of that household from that individual. Market Decisions’ own web site cautions that this methodology has its drawbacks: “Respondents may not be familiar with coverage of all members in extended household;” and “systematic error – larger households will have a greater error and are also associated with specific demographics.” (Read: lower income minority households). The Prescription for Pennsylvania web site acknowledges, but is dismissive of the fact that Market Decisions’ uninsured estimates are lower than those of the Census Bureau. It states that “the Census Survey contacted far fewer Pennsylvania households than the Insurance Department’s survey did,” suggesting that this in itself renders the Market Decisions data more reliable than the Census estimates. In fact, the Census Bureau survey consists of 6,000 Pennsylvania households per year which, from a purely statistical viewpoint, would generate numbers nearly as reliable as a sample of 6,700. But the Census Bureau also samples all households, not just telephone households, and their cooperation rate is enhanced by personal visits by Census Bureau employees when all else fails. Moreover, the Census Bureau encourages use of three-year rolling averages with a sample of roughly 18,000 – far more than the 6,700 in the Market Decisions survey. Only the homeless are excluded from the Census Bureau sample frame, and it is likely that if the homeless were included the estimates for the number of uninsured would increase.

Susan Korbel, PhD., owner of Core Research in San Antonio, Texas writes: “Researchers have long known that new arrivals (read: migrants) are less likely to participate in telephone research, especially if it sounds official due to concerns about other governmental institutions. In order to overcome these obstacles, it is necessary to OVERsample urban areas, as well as those in neighborhoods more likely to have English as a second language.” She adds: “This is a very curious sampling procedure… it does nothing for a state-wide aggregate estimate, and weighting geographically would create very large swings in data, especially if they also were weighting by demographics as well.” Ms. Korbel also dismisses as a “fallacy” the notion that sample size alone “regardless of distribution” is better, citing as an example a football stadium holding 100,000 people filled on one occasion with people who came to see the Pope speak, and on another with people who came to see the Rolling Stones. “Would you say that the responses of the 100,000 sample would be similar for all questions because they were both the same size? Probably the Pope goers would be able to represent Catholics in the region, but if you were asking about abortion, they wouldn't be representative of all people living in the area, any more than the Rolling Stones fans would be representative respondents.”

So just how many Pennsylvanians are actually uninsured? Estimates vary, but there is a consensus among all except the Governor’s office that the number is considerably higher than 900,000 (or 767,000 adults). The most recent Census Bureau estimate is 1,255,000, a number that includes only people who are chronically uninsured. A study released last year by the Center for Disease Control estimated that 10% of Pennsylvania’s population was uninsured, which would project to 1,235,000, virtually the same number arrived at by the Census Bureau. An October 2005 Keystone Research Center press release concluded that 494,000 Pennsylvanians had lost their employer-sponsored health insurance since 2000, implying that the number of uninsured might be as high as 1.8 million. It should be noted that Pennsylvania clearly has a lower uninsured rate than the rest of the nation, part of which is explained by its large elderly population resulting in a higher per capita enrollment in Medicare. Still, any effort to bring true healthcare reform to Pennsylvania is ill-served by the Governor’s attempts to minimize the true scope of the problem by low-balling the estimates regarding the number of uninsured.

A 2003 study by the Congressional Budget Office found that Census uninsured estimates are actually significantly under-stated because they include only people who are uninsured for a full year or more. They concluded that the real number, including people who were uninsured for only part of the year, was as much as 44% higher than the published Census figures. This is a shortcoming that researchers at the Census bureau readily acknowledged in discussions with this writer.

“UNDERINSURED,” A WORD GOVERNOR RENDELL AVOIDS

A group often left out of the debate over health care is the underinsured who, according to a recent study by Consumer Reports, account for 24 % of the U.S. population. If these numbers are accurate there are actually 50% more people in this country who are underinsured than are uninsured. According the Consumer Reports survey, the underinsured have a median household income of “$58,950, well above the U.S. median. Twenty-two percent live in households making more than $100,000. Still, many of the “underinsured” don’t have the resources to keep up with the rising costs of deductibles and co-pays, so much so that 43% reported that they postponed going to the doctor because they couldn’t afford it.” According to a recent study by Families USA, 2.2 million Pennsylvanians are in families that will spend more than 10% of their pre-tax income on healthcare costs in 2008. 87.1% of them are insured. These people live with the constant threat of being overwhelmed by their healthcare costs, including health insurance, and when their numbers are combined with the number of uninsured Pennsylvanians the numbers of uninsured or underinsured Pennsylvanians starts to look more like 4 million or more (about a third of Pennsylvania’s population), rather than the Governor’s 767,000. The budget the Governor submitted to the Legislature last year, but never drew attention to publicly, and which also went unnoticed by the media, aspired to insure an additional 359,101 uninsured Pennsylvanians by year five if his plan passed. In his recently submitted 2008/2009 budget (which has also largely escaped notice by the public or the media) that number had declined by more than 40% to 210,214, a number that probably represents only about 12% of Pennsylvania’s uninsured and only about 5% of the combined uninsured and underinsured. One wonders whatever inspired Governor Rendell to dub this plan “Cover All Pennsylvanians.” Under PABC the year five goal has been further down-sized to 202,466 and even that woefully small number is contingent upon the Governor getting everything he asks for from the Legislature, assuming it even passes. Neither the original nor the new plan does anything to help the millions of Pennsylvanians who are insured but still face financial ruin because of healthcare bills. Little wonder that more than 46% of personal bankruptcies in this country are directly caused by medical expenses.

THE SINGLE-PAYER ALTERNATIVE

While the media lavishes praise on Governor Rendell’s efforts to reform healthcare in Pennsylvania, and unquestioningly parrots his estimate of 767,000 uninsured adults, a better and cheaper alternative is out there virtually pleading for media coverage. The Pennsylvania Family and Business Healthcare Security Act would bring true comprehensive universal healthcare to every man, woman and child in the Commonwealth of Pennsylvania. With little fanfare the act, being promoted by HealthcareforALLPA.org (a group of unpaid citizens from around the State which, in the interests of full disclosure, this writer is a member of) is co-sponsored by thirty-eight members of the Pennsylvania House (HB 1660) and six members of the Pennsylvania Senate (SB 300). The single-payer bill would preserve privately delivered health care, but would remove health insurance companies from their role as gatekeepers of the healthcare system. A new Government agency would be created to administer the system, seeing to it that both those in need of healthcare and those providing it are properly attended to. Under this system co-pays and deductibles would become a thing of the past, and prescription drugs (which would be bulk purchased by the new agency in a manner similar to what the Veterans Administration has employed for years) would also be covered. Virtually the entire industrialized world operates under a single-payer universal healthcare system or something akin to it except for the United States, whose per capita healthcare costs are more than double those of the rest of the world while depriving millions of adequate care. The Pennsylvania plan would be financed with a 10% payroll tax (far less than most businesses contribute to the health coverage of their employees); a 3% personal income tax surcharge (again, far less than the average person pays for insurance, co-pays, deductibles, and prescription drugs); Federal funding that is already available; and proceeds from the tobacco settlement. Proponents of the plan maintain that it will save billions of dollars a year, and while that figure is difficult to precisely predict in the absence of an Economic Impact Study, they have long urged Governor Rendell to commission one. Similar studies in other states have found that a single-payer system would save hundreds of millions of dollars, and in the case of California (which actually passed single-payer legislation in 2006 only to see it vetoed by Governor Schwarznegger), $8 billion.

 1  |  2