Why Doctors Oppose AB 890
Last weekend, I flew from San Diego to Phoenix to catch up with a friend from medical school and residency. We’re both New Jersey transplants who attended Robert Wood Johnson Medical School and graduated in 2013, prior to matching at residencies at Cleveland Clinic in our respective specialties. At the time, I was blissfully unaware of the recent LA Times piece by David Lazarus that painted the California Medical Association opposition to AB 890, the bill that would give nurse practitioners increased independence in California, as that of an organization of physician obstructionists. Lazarus argues that if nurse practitioners are freed from the shackles of physician oversight, the shortage of physicians in California will magically be solved as nurse practitioners fill the gaps and provide equally good care to patients. Yet his view is extremely myopic and ignores the inevitable downstream consequences of such an arrangement. Years after finishing my family medicine residency, my questions for my neurooncologist friend revolved around her experience with so-called “physician extenders” or “mid-level providers”: nurse practitioners (NPs), physician assistants (PAs), and pharmacists. We both value our colleagues and know their assistance is of paramount importance in the ever worsening physician shortage as Baby Boomers age and we fight epidemics in obesity, type 2 diabetes mellitus, and opioid use disorder. But both of us worried most about the increasing independence of nurse practitioners specifically.
In the LA Times piece, the key piece of criticism levied against doctors’ opposition to AB 890 is that doctors could only provide anecdotal evidence of the issues with such an arrangement. I realize that some of what I will provide is anecdotal evidence, and doctors are well aware that case studies are low on the totem pole of strength of evidence. Yet anecdotal evidence is often all we have, at least to start. The ideal of well-designed, high-powered, randomized, double-blinded controlled trials is the exception rather than the rule. Especially when what you are trying to study is not amenable to such a study anyway. With patient lives, every single one counts, and the errors that occur deserve to be studied in the granular detail a case study provides.
In physician-only Facebook groups, doctors are incredulous at the suggestion that increasing the independence of nurse practitioners will lead to lower malpractice costs. It’s not hard to see why, when on a daily basis I see physicians asking their colleagues for advice on whether to take jobs in which they’d be the medical director signing off on the charts of 4-5 PAs and NPs who are miles away. Sandwiched between posts like this, I recently came across a case in which an NP in Oklahoma initially tried to provide the appropriate care that she clearly shouldn’t have been attempting to provide in the first place… only to make an error that leads to the patient tragically dying. According to trial testimony, this was her second-to-last shift, and she had already been terminated by the hospital 27 days prior to the episode due to quality/safety concerns. “She was only a family nurse practitioner. However, 8 months earlier Mercy granted her privileges to provide care and treatment to acutely and critically ill patients in the ER at Mercy El Reno. In fact, according to trial testimony, she was often the only medical provider in the emergency room in El Reno.”
A subsequent malpractice suit naming the supervising physician rewarded the patient’s family over $6 million. A nurse practitioner trained in Family Medicine was virtually unsupervised in an Emergency Department setting. Initially, she correctly attempts to order a CT to rule out a pulmonary embolism in a young woman on oral contraceptive pills. A urine drug screen is performed, the results of which derailed the plan and delayed the diagnosis of the CT for 8 hours. While it’s unethical to order a urine drug screen (UDS) without notifying the patient, it is commonly done at many hospitals by everyone from nurses to physicians. The UDS was positive for methamphetamine, but negative for amphetamines, which calls into question the accuracy of the result. Subsequently, the focus shifted to the fact that she used an illicit substance, and thus the CT was canceled as her chest pain suddenly was presumably attributed to her being a dishonest drug addict. It’s a red herring that ultimately leads to the patient dying. A repeat UDS and autopsy would confirm she had not used drugs. 8 hours later, the CT is finally ordered, but it’s not ordered with the appropriate urgency, and it’s likely already too late: the patient is dying from untreated pulmonary emboli that should have been diagnosed hours before. Who gets sued? The supervising physician, who never saw the patient. The nurse practitioner faced no disciplinary action and limited negative publicity. She’s still practicing today.
Moreover, in San Diego, there is good evidence that malpractice costs are rising as a result of the increase in medical care provided by nurse practitioners. One colleague summed it up with the following quote from MedPro Group’s report (prior to his post mysteriously getting deleted):
Indemnity Cost: Indemnity cost is the amount paid on behalf of insureds to plaintiffs to resolve a claim. The costs associated with malpractice payments for nurse practitioners has more than doubled over the last ten years. As shown below, payments made in 2007 and 2017 were $27 million and $68 million respectively. This increase can’t be credited entirely to the growth of nurse practitioners, since the growth of indemnity costs was 30% higher than the growth of the industry during this time.
MedPro Group, a subsidiary of Berkshire Hathaway, is among the top 4 malpractice providers in California according to Gallagher Healthcare, the largest malpractice broker for physicians in the nation. I’m not cherry picking statistics to prove my point.
Before delving into more anecdotal evidence, let’s take the plunge into the two studies Lazarus cites regarding malpractice trends. Both studies were published in Medical Care Research and Review. In assessing the prestige and peer review process of a medical journal, an Impact Factor is calculated, which compares the number of citations in a year to the number of published articles for a journal in an attempt to quantify its impact. For this journal, it was 2.315 in 2017, and according to its site, it ranks 39/94 in Health Care Sciences & Services and 21/79 in Health Policy & Services that year. By way of comparison, The New England Journal of Medicine (NEJM) had an Impact Factor of 79.258 that year. We are dealing with different tiers of journals here, and I find the absence of a study from NEJM or another top tier journal to be a striking omission in Lazarus’ article. While NEJM has published a perspective piece that echoes some of the points Lazarus mentions, a quick search of the Health Policy section of their site also reveals that opinions on this topic are not so universal.
The first study Lazarus cites is from 2017, and it is authored by the President of the PA Experts Network (Jeffrey G. Nicholson) and an independent health policy consultant (Roderick S. Hooker), in conjunction with the corresponding author, an Associate Professor (now emeritus) who holds a PhD in Social and Personality Psychology. But despite the apparent bias of having not a single physician as an author of the study - or as editor of the journal, for that matter - one of the findings was that diagnosis-related malpractice allegations against physicians were significantly LESS than those of PAs or NPs. This was despite the fact that 94.8% of the providers examined were physicians. Per capita, the authors note, claims were more frequent for NPs and PAs in the categories of diagnosis- and treatment-related errors, and this may be at least in part due to NPs and PAs being at a greater risk of making an error in these two areas. It’s testament to the fact that trying to extrapolate the results from a single study on malpractice claims is problematic, and I wonder if Lazarus read the study. Or, for that matter, the dentist who authored AB 890, Assemblyman Jim Wood. There are too many variables at play, and simply being named in a suit does not indicate that malpractice actually occurred. Indeed, I was named in a malpractice suit in 2017 regarding possible lack of communication by a specialist on a diagnosis of H pylori infection years before. I was the patient’s primary care provider, but at the time of said incident, I was an intern in Cleveland, not a primary care physician in San Diego. The case was subsequently dismissed, though not before it made it difficult for me to find a job despite the fact that I was not involved at all in the diagnosis. If doctors were purely concerned about their liability, we’d happily let nurse practitioners practice independently. But we’re not. We’re concerned about the care patients receive. Lazarus’ characterization hurts.
The second study, from 2018, is from the same journal. The corresponding author Benjamin J. McMichael, JD/PhD, was a postdoctoral fellow at Owen Graduate School of Management at Vanderbilt University prior to assuming his current position as Assistant Professor at University of Alabama. The other 2 authors are Barbara J. Safriet of Lewis & Clark Law School and Peter I. Buerhaus from Montana State University. As with the first study, my beef is less with the article itself and more with the way that Lazarus misreports its findings:
Unfortunately for Lazarus, a close reading of the articles reveals that the authors would not agree with his spin on their findings.
The blurred line between physicians and other members of the healthcare team is a source of angst among physicians. The parlance that is now routine lumps us together with those of different educational backgrounds as “providers,” and Lazarus’ article indicates that a belief in the equivalency between physicians and nurse practitioners has seeped outside the medical community to those who are relatively uninformed about the distinction in our occupations. The term bulldozers over our differences. For many, it hearkens back to “Behandler” - the German word for “provider” that was used to demonize Jewish physicians during the Holocaust. The American Academy of Family Physicians recently took the unusual step of releasing a position paper on the use of the term. While some may scoff that we are splitting hairs over terminology, words have meaning, and precise language exists for a reason. This should be evident to a journalist for a major newspaper.
As for the conversation with my friend over more anecdotal evidence? We both agreed that nurse practitioners assist doctors and save lives. Yet as doctors, there are times we just want to talk to a doctor and not go through a middleman with regard to the necessary workup or the specific clinical question for which we place a consult. We rely on the 15,000+ hours of training that only MDs and DOs have in assisting with diagnosis. In residency, the nurse practitioner would tee up the patient who really needed that endogastroduodenoscopy or endoscopic retrograde cholangiopancreatography. Yet it was redundant work; if the gastroenterologist had the time to read my notes, review the patient’s labs, and see the patient, he wouldn’t need a nurse practitioner to write his consult notes and tell him that he needed to perform an endogastroduodenoscopy and colonoscopy ASAP for an 80 year-old patient on anticoagulants with a hemoglobin of 7 who has not had either of these procedures before and has required blood transfusions. And inevitably, when faced with a challenging patient with suspected Sphincter of Oddi dysfunction, the gastroenterologist, not the nurse practitioner, is the person whose opinion the patient and the physician treating him deserves. Such inefficiencies are now commonplace in our medical system due to the administrative burden created by electronic medical records and insurance companies.
Moreover, my friend and I both agreed that in cases like that of Keith Davis, Jr., an evaluation by a nurse practitioner ought not to take the place of an evaluation by a trauma surgeon or intensivist. I’ve previously discussed Davis’ case on this blog (in fact, more than once!). The Cliffnotes version is that he was shot 3 times by the Baltimore Police Department and woke up in a hospital after needing surgery for damage caused by a bullet that went through his left carotid artery, maxillary sinus, and mandible before settling in the subcutaneous tissue of his neck. His neck wound subsequently developed an abscess the size of a golfball, and it wasn’t until months later that a judge noticed the abscess on his neck and mandated that the State of Maryland have the surgeon see him within 30 days. (They failed to comply, by the way.) I don’t fault the surgeon; with the amount of penetrating wounds Baltimore hospitals see, he was likely in surgery constantly and the hospital understaffed, though I have no way to be sure. He rearranged his schedule to try to evaluate Davis in a timely manner after discharge from the hospital. While I’m generally reticent to play Monday morning quarterback of cases I haven’t personally seen, the care Davis received was clearly substandard. The innumerable physicians I’ve talked to agree that he should’ve had a surgeon evaluate him prior to discharge given the extent of his injuries and his girlfriend-turned-wife’s concerns for a developing infection prior to discharge. My colleagues of various specialties agree that AB 890 would cause cases like this to become all the more common. “The solution is not less oversight,” opined one of my best friends, a critical care physician. You remove the oversight that is theoretically there, but in actuality often is not, and you have even more nurse practitioners out of their scope of practice forced to make decisions they shouldn’t be making.
I don’t pretend to have all the answers to solving the physician shortage. I realize how difficult it is to get bipartisan Congressional support for anything these days, especially when it relates to increasing Medicare funding to allow more residency spots. Moreover, those spots need to be filled by medical students, and unless there are more medical schools and larger class sizes, the physician shortage will only worsen. But that’s why the California Assembly needs to listen to the concerns of the California Medical Association and try to understand the problem fully, rather than paint doctors as self-concerned obstructionists.
About me: I’m a Family Medicine-trained physician based out of San Diego, California. I’m a member of the California Medical Association, though the views expressed are my own. If you like what you read, please subscribe. The LA Times failed to respond in a timely manner to my criticism of Lazarus’ article, so I have posted my personal rebuttal here instead.