By Dr. William Harvey, M.D.
Editor’s Note – Dr. Harvey submitted this piece to me on July 22nd, but due to the the move and some technical complications (Blogger does not like tables) I was unable to post it before today. This article is another incredibly valuable piece to the healthcare puzzle, and don’t for a second believe that recent setbacks in the House mean healthcare “reform” is dead. It’s not dead, and Dr. Harvey’s excellent research is as crucial as ever.
Second Editor’s Note – Dr. Harvey corrected some numerical errors in the table, and I uploaded a corrected version of the table today, 7/31/2009.
My original plan after describing the Democrat’s proposed healthcare legislation (H.R. 3200) was to look at Divisions A and B in detail, but I’ve decided to postpone that discussion so that I can incorporate responses to some recent comments from President Obama’s press conference. Instead, I will focus on concerns regarding other parts of the legislation: particularly Division C.
Division C addresses the healthcare workforce. You might think this seems like a technical and low-priority part of the bill for the public. Think again. At best this part of the bill is designed to address a key component of healthcare quality and efficiency; at worst, this part of the bill is designed to allow the Federal government to take over the entire healthcare infrastructure and workforce and in the process irrevocably create a medical establishment devoted to a single-payor, single-provider healthcare system. In any case, it will represent a very expensive intrusion of the Federal government into medical education, training and medical research.
This will be a two-part posting. Here, I’ll frame the nature of the problem. In the next post, we’ll look at the proposed solution.
Verizon Wireless ads frequently show the Verizon Team checking to make sure their coverage is all encompassing. Someone runs to all kinds of distant places, makes a call, and asks, “Can you hear me?” Today we’re going to talk about a major implication of increasing healthcare coverage to more Americans: what are the practicalities of accomplishing that task? To do so, we’ll need to dig into some of the numbers: numbers of patients (I refuse to “genericize” this conversation by calling them/us “healthcare consumers”) and number of doctors, nurses, other healthcare workers. I’m going to include sources, because numbers and statistics, as Mark Twain told us, can be made to say such about anything we want them to:
Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.’” – Mark Twain in “Chapters in My Autobiography”
So, let’s do what our Congressional representatives haven’t done this session: let’s think through strategically the implications of healthcare reform.
According to the US Census Bureau (Table 1 below, adapted from Table C-1, p. 69, in Walt, Carmen, Bernadette D.Proctor, and Jessica C. Smith, U.S. Census Bureau, Current Population Reports, P60-235, Income, Poverty, and Health Insurance Coverage in the United States: 2007, U.S. Government Printing Office, Washington, DC, 2008.), the 2007 population of the US was 299.1 million. Of these, 253.4 million (84.7%) were covered by some form of health insurance – 202 million (67.5%) by non-governmental health care plans and 83 million (27.8%) by government-based healthcare plans (Medicare, Medicaid, active-duty military). This left 45.8 million (15.3%) of the population uninsured. The percentages of insured and uninsured hasn’t changed significantly over the last 20 years (see Figure 1); some of the changes that did occur were because of other factors (e.g., changes in the way the census counted people or calculated values; or the one time change last year because of Massachusetts implementing universal care in that state).
There’s a problem here: the Census is conducted every 10-years, but the Census Bureau refines its information with smaller surveys every year. If they don’t sample the population correctly, the information will be distorted. In particular, the Census Bureau knows that certain populations are under-sampled, notably the poor, the homeless, and unauthorized immigrants (the new, politically correct term for “illegal aliens”). In the civilian version of “don’t ask, don’t tell”, the Census Bureau doesn’t ask “are you an unauthorized immigrant?” and the individual being interviewed certainly doesn’t volunteer “oh, by the way, I’m an unauthorized immigrant.”
So the likelihood is that the uninsured number quoted above doesn’t represent the unauthorized immigrant population. Fortunately, there is another source for that information. According to the Pew Hispanic Center’s latest report on Unauthorized Immigration (Jeffrey S. Passel and D’Vera Cohn. A Portrait of Unauthorized Immigrants in the United States. Washington, DC: Pew Hispanic Center, April 2009), there are about 11.9 million unauthorized immigrants in the US, about 4% of the US population. Interestingly, most of us assume that almost all unauthorized immigrants are uninsured; in fact, in 2007, 59% of undocumented adults were uninsured (Figure 25), meaning that 41% had health coverage during at least a part of 2007. This compares to 24% of legal immigrant adults and 14% of US born adults. The situation is even more complicated (and the data is therefore muddled) for children of unauthorized immigrants, since many such children are born in the US and are thus US citizens, while their parents may be legal immigrants or unauthorized immigrants. According to the Pew report (Figure C1) there are about 4 million US born children and 1.5 million unauthorized immigrant children of unauthorized immigrant parents. Also (Figure 26), 45% of immigrant children whose parents were unauthorized immigrants didn’t have health insurance, compared to 25% of US citizen children of unauthorized immigrant parents. For legal immigrant parents, the comparable figures were 22% for their immigrant children compared to 14% of their US citizen children.
My understanding is that this issue should no longer be an issue, because of passage of the S-CHIP bill by Congress during the last months of the Bush administration. Still, it seems that unauthorized immigrants often don’t apply for benefits even when they are legally permitted to.
Thus, the Census figure of 45.8 million uninsured should have 7 million unauthorized immigrant adults (59% of 11.9 million) and 2.5 million children (45% of [4 + 1.5 =] 5.5 million children of unauthorized immigrants) added, for a final estimate of 55.3 million potential uninsured people in America. Clearly, you may believe that unauthorized immigrants should receive no healthcare, or should only receive care for life-threatening conditions, or should be covered as a matter of “good samaritanship”. Whatever your belief, we have now tried to carefully define the size of the problem.
Now let’s ask a simple but devastating question: how many healthcare workers does the US have to take care of the 87.1 million currently insured people, and how many more healthcare workers will we need to cover an additional 45.8 (or 55.3) million newly insured people? How much capacity does the current system have or is it “maxed out”?
That secondary question is easy to answer. Under present conditions and the current system of care, the existing healthcare workforce has no excess capacity. The current healthcare infrastructure (hospitals; clinics; medical practices; staffing by doctors, nurses, other healthcare providers, public health workers, etc.) is fully occupied taking care of the covered population.
To answer the primary question (how many healthcare workers in the US today?), we turn to the Dartmouth Atlas of Healthcare, a wonderful resource for researchers in healthcare policy. Real people also find it useful because it allows you to compare the hospitals around your area to see how they’re staffed, how many surgeries they perform, what the relative cost of care is, what the frequency of mortality (death) and morbidity (complications and side effects) are, etc. These are things that are otherwise hard for the typical patient to find easily. There are other resources, put together by quality organizations or (more recently) by insurance companies, but the Dartmouth Atlas has been collecting data for 20 years and is committed to making this information publicly available.
The Dartmouth Atlas lists non-MD staffing as (number of staff/1,000 population) and MD staffing as (number of MDs/100,000 population). I’ll save you the spreadsheet gymnastics; the results are in the Table below. I’ll also save you the trouble of sending comments that the Atlas figure for the US population in 2006 is 298 million, while the US Census number is 296.8 million. This qualifies as a Mark Twain “statistic”: depending on how the counting was done, 1.2 million people got lost in the shuffle.
There are two assumptions in this Table (remember that to “assume” is to make an “ass” out of “u” and “me”). The first is that there is no efficiency or inefficiency in adding the new patients, so the number of new staff needed is the ratio of new patients to current patients. The second is that the number of acute hospital beds increases by the same ratio; this is likely to be an incorrect assumption, but I have no good way to justify the size of adjustment to this number. Here’s the bad news (I’ve put the key summary lines in bold):
As anyone in healthcare will tell you, adding roughly 132,000 physicians and 242,000 nurses in “a couple of years” is impossible. Nurse training is 4 years of undergraduate training, followed by at least 1-2 years of post-graduate training. Similarly, physician training is 4 years of undergraduate training, 4 years of medical school, and at least 3 years of post-graduate training and possibly as many as 7 years for certain surgical specialties. So, if we add 40-50 million new patients in the next 12 months … it will be at least 7 years before we have properly trained medical staff to care for them. In the meantime … imagine your physician and his/her office with a 50% increase in patient load and you’ll get some idea of what ObamaCare may look like in the short-term.
So … can you hear me NOW?
Adapted from Table C-1 (p. 69) of DeNavas-Walt, Carmen, Bernadette D.Proctor, and Jessica C. Smith, U.S. Census Bureau, Current Population Reports, P60-235, Income, Poverty, and Health Insurance Coverage in the United States: 2007, U.S. Government Printing Office, Washington, DC, 2008.
Figure 1. Percent of Population insured and uninsured, by year, 1987-2007.
From the data in Table C-1 (p. 69) of DeNavas-Walt, Carmen, Bernadette D.Proctor, and Jessica C. Smith, U.S. Census Bureau, Current Population Reports, P60-235, Income, Poverty, and Health Insurance Coverage in the United States: 2007, U.S. Government Printing Office, Washington, DC, 2008.
William Harvey is a physician with extensive experience in drug research and development. He began as an academic researcher but has been a pharmaceutical executive in the global development arena for almost two decades. His current position involves the strategic use of comparative effectiveness research to speed drug development and to educate healthcare stakeholders: government, payors, prescribers, and patients. He lives in the greater Philadelphia area.