Search This Blog

Tuesday, September 26, 2023

Two Large Medical Groups Shun Medicare Advantage Plans

 Signaling what may be an emerging national trend, two influential medical groups with San Diego-based Scripps Health are cancelling their Medicare Advantage contracts for 2024 because of low reimbursement and prior authorization hassles, leaving 30,000 enrolled seniors to look for new doctors, or different coverage.

"Negotiations with the payers for MA with our medical foundation groups and Scripps Health were unsuccessful and we have been forced to withdraw from those plans due to annual losses that exceeded $75 million," Scripps CEO Chris Van Gorder told MedPage Today in an early morning email.

He said the losses are due to "low reimbursement, denials, and administrative costs to manage high utilization and out of network care."

Van Gorder emphasized that about 30,000 enrollees will have to make a change in their coverage or pick another doctor. About 1,000 physicians and advanced practitioners such as physician assistants are members of the two groups.

"We certainly regret any inconvenience to them," he said, but "that kind and size of loss is unsustainable by Scripps. We will remain in MA with our IPAs [independent physician associations] as those contracts are structured differently and of course, traditional Medicare."

The two medical groups affected are Scripps Coastal and Scripps Clinic Medical Group. Five other Scripps medical groups will continue to take MA plans, he said. Affected beneficiaries should receive a notice directly from the plans.

Enrollees "can continue to see Scripps through traditional Medicare at all our hospitals and affiliated medical groups or can switch to an independent medical group (IPA) that still maintains a MA contract at Scripps Mercy, Scripps La Jolla, and Scripps Encinitas hospitals," he added.

Patients can also switch to Kaiser Medicare Advantage during re-enrollment starting Oct. 15, or to another hospital system whose physicians still take MA plans.

However, switching to traditional Medicare without a supplemental plan -- also called a Medigap plan -- means patients incur 20% of all physician, lab, imaging, and emergency room costs, along with a $1,600 deductible per hospitalization episode this year. In California and in 44 other states, supplemental plans can reject applicants with common health conditions such as cancer, high blood pressure, a prior hospitalization, or joint replacement. In addition, these plans are expensive, with increasing monthly premiums as one gets older.

But Van Gorder said he had no choice. "We are a patient care organization and not a patient denial organization and, in many ways, the model of managed care has always been about denying or delaying care – at least economically. That is why denials, [prior] authorizations and administrative processes have become a very big issue for physicians and hospitals – not to mention that the reimbursement is insufficient in most government programs as we all know."

"Now with intermediaries taking their profit and offering insurance to beneficiaries for free in many cases [the extra benefits like trips to the doctor], the end of the economic food chain is once again the hospitals and physicians."

Van Gorder said patients should ask themselves, "'Am I receiving the care I need if my hospital and physician are not even covering their costs? How long is that sustainable?'"

Where these patients will go is an open question.

More than half of all Medicare beneficiaries are now enrolled in MA plans nationally. In San Diego County, the fifth largest in the country, that percentage is 54%opens in a new tab or window of those eligible.

Nate Kaufman, a San Diego-based health system consultant, wasn't surprised at Scripps' news.

"I advise all hospitals to terminate their Medicare Advantage plans with anybody unless they're getting over 115% of Medicare," Kaufman told MedPage Today.

The problem is complicated, but in a nutshell the issue is a lack of funds to go around to pay hospitals and doctors the cost of care.

Medicare's contracts with Medicare Advantage plans pay less than what Medicare pays for traditional Medicare enrollees on the expectation that the plans will save money," Kaufman said. "Then, the MA plan takes a piece off the top. The remaining funds go into two buckets. One for MA plan pharmacy benefits and the other for hospital and physicians. And that requires a major reduction in utilization to maintain profitability."

Kaufman said all of this is made worse by the issue of prior authorization, which is now under Congressional scrutinyopens in a new tab or window.

"It creates hassles for everybody and cost," he said. "The foundation upon which Medicare Advantage was built, which was that there's excess money somewhere, has disappeared after the insurance company takes their cut off the top and captures the pharmacy rebates."

Additionally, providers are seeing delays in getting paid, which carries its own cost. And because enrollees pay very low or no premiums, there are less funds for most of the providers, he said.

The issue is likely to keep many independent insurance agents busy. Christopher Westfall of Senior Savings Network, who is licensed to write Medicare contracts in 47 states, also sees providers ending their MA relationships as a national trend.

He said it can be extremely frustrating for his agents when seniors either don't check their plan or choose the wrong plan thinking their provider is in network, only to find out after Jan. 1 that their doctors are in different plans, or have dropped out.

Many health systems have announced that they're terminating their MA contracts, or are strongly considering it.

The Mayo Clinic in Jacksonville, Florida, and Scottsdale, Arizona, told beneficiariesopens in a new tab or window last October that it would no longer take most MA plans. If those patients sought care, it would be considered out-of-network, leaving them with a higher share of the costs.

Samaritan Health Services in Corvallis, Oregon, endedopens in a new tab or window its MA contracts with UnitedHealthcare, one of the largest Medicare Advantage contractors in the country.

Regional Medical Center in Cameron, Missouri terminatedopens in a new tab or window contracts with Cigna's MA plans in 2023, and planned to drop Aetna and Humana MA contracts in 2024. Cameron's Regional CEO Joe Abrutz blamed the plans' practice of "delaying any action on reimbursement."

Stillwater Medical Center, a 117-bed hospital in Oklahoma, called it quitsopens in a new tab or window last year with all of its in-network MA plans, blaming rising operating costs and a 22% prior authorization denial rate, compared with a 1% denial rate for traditional Medicare.

Brookings Health System, a 49-bed hospital in South Dakota, won't be in networkopens in a new tab or window with any MA plan starting in January to preserve its financial sustainability.

St. Charles Health System in Oregon encouragedopens in a new tab or window its seniors not to enroll in MA this year as it re-evaluates its participation in Medicare Advantage contracts.

And Baptist Health Medical Group in Louisville, Kentucky failed to agree on termsopens in a new tab or window by its deadline with Humana's Medicare Advantage plan and alerted their patients to seek other options.

Officials for the Medicare Advantage industry had not returned requests for comment as of press time.

https://www.medpagetoday.com/special-reports/exclusives/106483

White House Cancer Moonshot Is a Dud

 The White House Cancer Moonshot

opens in a new tab or window recently announced new efforts to "End Cancer as We Know It." Yet, the announcementopens in a new tab or window includes no mention of the FDA, which is not only directed at protecting and promoting the public health but also has existing powers that could reduce cancer deaths and harms more rapidly than any existing or planned Cancer Moonshot initiatives.

The Moonshot announcement stresses the fact that smoking is the "biggest single driver of cancer deaths in this country." But its newly announced anti-smoking measures are remarkably weak and indirect, focusing largely on increasing cessation services and related smoker awareness. In particular, one new initiative would provide $15 million over 5 years to help increase implementation and enforcement of policies to increase awareness of smoking cessation services among disadvantaged smokers or prohibit the sale of menthol or other flavored tobacco products. But $15 million over 5 years is a drop in the bucket compared to the tobacco industry's expenditures of more than $7.5 billionopens in a new tab or window per year to promote cigarettes.

Moreover, any new efforts to support state and local policies to prohibit flavored tobacco products would be largely unnecessary if FDA implemented its long-overdue final rulesopens in a new tab or window to ban menthol cigarettes (the only added-flavor cigarettes currently allowed under federal law) and prohibit cigars with added flavors. By reducing youth initiation of smoking and prompting many existing smokers to quit, those two rules could sharply reduce cancer and cancer deaths. A recent University of Michigan studyopens in a new tab or window, for example, estimated that banning menthol cigarettes, by itself, would prevent 650,000 people from dying from smoking over the next 40 years -- and roughly one-thirdopens in a new tab or window of those prevented deaths would have been caused by smoking-caused cancers. But FDA has not received the support it needs from the White House, either through the Cancer Moonshot or otherwise, to get those final rules issued.

The absence of any mention of FDA or any support for the FDA rules to banopens in a new tab or window menthol cigarettes and prohibitopens in a new tab or window added-flavor cigars in the new Moonshot announcement is odd given that FDA has otherwise been mentioned as a participant since the Cancer Moonshot was reignitedopens in a new tab or window in February 2022. A March 2022 fact sheetopens in a new tab or window even mentioned the planned FDA rules to ban menthol cigarettes and added-flavor cigars as a major part of the Moonshot. In addition, the federal Unified Agendaopens in a new tab or window of upcoming government regulatory action has included references to the two rules for years, and the most recent Unified Agenda said that the menthol cigarette and flavored cigar final rules would be issued by August 2023.

However, FDA missed that August deadline and has subsequently stated only that it expects to issue the final rules "in the coming monthsopens in a new tab or window." The Cancer Moonshot says nothing about this inexplicable delay in effective federal anti-cancer measures. Nor has the White House, FDA, or the Office of Management and Budget (which oversees federal regulatory efforts) provided any new deadlines or other assurances that the final rules will actually be implemented soon or at all.

If the White House and its Cancer Moonshot were truly serious about preventing and reducing cancer, they would have not only ensured that these two FDA rules were already fully implemented but would have also supported additional FDA anti-smoking rule-making that would reduce cancer deaths and harms even more quickly and sharply. Most notably, an FDA-coordinated studyopens in a new tab or window found that a new rule to reduce nicotine levels to non-addictive levels in cigarettes would, at a minimum, save millions of lives through reducing smoking initiation and increasing quitting. In addition, since fall 2017 the Unified Agenda has repeatedly said that FDA planned to issue a nicotine-reduction ruleopens in a new tab or window. But its announced date for issuing just the preliminary proposed rule has been repeatedly postponed, with the nicotine rule falling completely off the Agenda between spring 2019opens in a new tab or window and spring 2022opens in a new tab or window; and implementing such a rule has never been part of the Cancer Moonshot. The most recent Unified Agendaopens in a new tab or window says the proposed nicotine reduction rule will be issued before the end of this year. But if that were a real deadline that FDA was actually going to meet -- or if FDA were actually going to issue its final rules on menthol cigarettes and flavored cigars soon -- one would think the Cancer Moonshot would have at least mentioned that as a major upcoming success in federal efforts to reduce cancer deaths and harms.

Unfortunately, the White House Cancer Moonshot's failure to support these FDA anti-smoking rules is nothing new. Since 2009, when the Tobacco Control Actopens in a new tab or window first provided FDA with extensive powers and resources to regulate cigarettes and other tobacco products and their labeling, marketing, distribution, and sale, no U.S. president, vice president, or major White House official has publicly expressed any support for any FDA rulemaking to prevent or reduce smoking or overall tobacco use deaths and harms.

Thanks in large part to this absence of White House supportopens in a new tab or window, FDA has not yet been able to implement any substantive rule that would significantly reduce smoking or overall tobacco-caused deaths and harms. Without such support (or a statutory or court-ordered deadline), it is extremely difficult for FDA to get a draft proposed or final rule successfully through the federal bureaucracy's behind-the-scenes prior review and clearance process. The Unified Agenda listings stating that FDA would issue final rules to ban menthol cigarettes and added-flavor cigars in August 2023 suggested that FDA was finally getting the White House support it needs to take effective action. But those missed deadlines and the absence of any reference to those or any other FDA anti-smoking rules in the subsequent Cancer Moonshot announcement suggests otherwise. It is also especially troubling that the White House Moonshot highlights a new initiative to support state and local policies to ban flavored tobacco products when the White House could simply ensure that FDA quickly did that nationwide instead.

This 14-year and counting absence of any major FDA tobacco control rule is a national tragedy. It is difficult to imagine any public health or policy justification for FDA not using (or not being allowed to use) its extensive tobacco control powers and resources, not only to reduce cancer deaths and harms more sharply and quickly than any other Moonshot initiative but also to prevent and reduce enormous amounts of other unnecessary smoking-caused death, disability, disease, and other harms and costs.

Eric N. Lindblom, JD,opens in a new tab or window is a former official at FDA's Center for Tobacco Products and a long-time tobacco control policy analyst. He currently serves as an independent consultant to researchers, lawyers, and investors on tobacco-related issues and is a senior scholar at Georgetown Law's O'Neill Institute for National and Global Health Law.

https://www.medpagetoday.com/opinion/second-opinions/106455

GLP-1 Agonist Plateau No One's Talking About

 The declines in body weight that patients experience with injectables like semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro) are no exception to the concept that nothing lasts forever.

Eventually, everybody reaches a "plateau," even on newer GLP-1 receptor agonists. It's a phase at which the body reaches a new "settling point," specialists said, and weight, along with other metabolic markers like blood pressure and HBA1c stabilize, or fluctuate only slightly. For some, this may mean a gradual increase in appetite or "food noise"; others may be able to maintain their current state.

Studies have shown that, on average, this plateau happens at a little over a year with semaglutide. Even so, physicians say some patients are surprised to learn that there's a limit to what these medications can do.

"Everyone will plateau, of course. No one on my watch has disappeared. No one has vanished," Jody Dushay, MD, an endocrinologist at Beth Israel Deaconess Medical Center in Boston, told MedPage Today. "It's alarming to me that people find that surprising, but everyone will reach a plateau and there's no way to know when you start the medication what that will be, what percent weight loss that will be, and how quickly they will reach it."

Gitanjali Srivastava, MD, an obesity medicine specialist at Vanderbilt University Medical Center in Nashville, Tennessee, noted that "we see that often and it's a question that gets asked frequently. There's going to be a new homeostatic balance that's achieved, and we see that with any other disease phenomenon."

For example, Srivastava told MedPage Today that a patient will not become hypotensive after a certain amount of time on a blood pressure medication, or have blood glucose levels decrease indefinitely with diabetes medications. "Evolutionarily, we need to be able to do that, so we can protect against the extremes," she added. "Because the alternative is that you continue to wither away, and that can be dangerous."

It's still unclear what may predispose patients to longer or shorter responses to GLP-1 agonists, but Dushay said that, typically, early responses tend to predict later ones. If a patient experiences steep weight loss on lower doses of semaglutide, for example, they can stay on a lower dose for longer, with more time to uptitrate if necessary. Patients on semaglutide for type 2 diabetes also tend to experience less weight loss overall, she said.

Karl Nadolsky, DO, an endocrinologist and obesity medicine specialist at Holland Hospital in Michigan, told MedPage Today in an email that "a history of childhood obesity with any suspicion [of] specific genetic or syndromic etiology" might also hint at a predisposition to hyporesponse

Specialists have said that before prescribing this class of medication, they thoroughly discuss what to expect, from side effects to plateauing and the possibility of non-response. Dushay noted that it's important that providers set these expectations with patients, and for patients to seek out doctors who have the time and clinical experience to do so.

In clinical trials known as STEP1opens in a new tab or window and STEP2opens in a new tab or window that looked at semaglutide 2.4 mg per week, participants' weight loss tapered off around week 60, with about 10% to 15% of body weight lost. Semaglutide's effect on blood pressure and HBA1c appeared to plateau even earlier. In STEP 5opens in a new tab or window, spanning 2 years, patients hit a weight plateau once again around 60 weeks, and were able to maintain that weight for the remainder of the study. In the SURMOUNT trialsopens in a new tab or window, which looked at various doses of tirzepatide for 72 weeks, participants on the 5-mg dose had reached a plateau within 60-72 weeks, but this was not the case on the higher doses. A 2-year trialopens in a new tab or window is expected to offer more insight.

However, clinicians are more interested in what such averages hide. It's nearly impossible to know how well a patient will respond to semaglutide or tirzepatide, and individuals may have wildly different medical histories, medications, and comorbidities that all affect how well, and for how long, a given drug may work.

Fatima Cody Stanford, MD, MPH, MPA, MBA, an obesity medicine specialist at Massachusetts General Hospital in Boston, pointed out that a patient's expectation may not match their ultimate response to a GLP-1 agonist. "Everybody comes in and they're like, 'I want to do what this person did' -- they have their family, their friend, their sister, their brother, cousin, aunt, and I'm like, 'well, we don't know,'" she said. "The only people that I expect to respond almost identically are identical twins."

Stanford said she would like to see studies on GLP-1 agonists in the future that stratify patient response by various characteristics like genetics. Right now, she noted, weight-loss drugs require a lot of trial and error. Predicting the likelihood of success for individual patients would save time -- and money. "When a new cancer drug comes out, not everyone's like, 'ooh, there's a new cancer drug, let's just start everyone on that.' Right? Nobody does that," she said. "I want to know who the drug is right for."

"If I know that, on average, this person is going to be a really poor responder to a GLP-1 agonist, I won't [prescribe it]," she added. "It is a really arduous pathway for a lot of reasons -- access, coverage, prior authorization. It's burdensome. It's burdensome on the patient, it's burdensome on the system, it's burdensome on me."

Experts said it's common for patients to want more. For example, they may bring their blood glucose within a normal range, go off of blood pressure medications, and maintain overall positive health outcomes with a GLP-1 agonist, but hit a "plateau" and still want to lose weight.

Ultimately, this is where psychological and societal ambitions collide with clinical ones.

Patients are "having to negotiate with what society tells them, so they're still 200 pounds and society says 'for your height and weight you should be 125,' even though their health looks amazing," Stanford said. "Not always, but often, they still want to be whatever this number is."

Dushay said it's rare for patients to achieve the weight loss goal they arrive at an initial consult with. Often, Dushay's noticed, this goal is the weight they were on their wedding day. "I think that some of it is literally [that] they want to weigh that," she said. "But I think there is a big component of 'I want to rewind time.'"

"I've almost never had someone hit a plateau when they were like, 'okay, I'm good.'"

In the meantime, clinicians have strategies to move past a plateau if a patient hasn't yet met important clinical goals. Generally, some said they might increase the dose if possible, if the patient can tolerate it well. Failing that, they can supplement with a second drug that targets a different neuronal or hormonal pathway, like phentermine (Lomaira). Dushay said that in practice she's noticed that "drug holidays" or stopping and restarting a GLP-1 agonist, have typically not affected plateaus.

https://www.medpagetoday.com/special-reports/exclusives/106464

Inmates Escape From Hospitals in Recent Spree

 There have been a bevy of recent reports of inmates escaping from hospitals -- by stealing a vehicle, rappelling down the side of a building, or simply walking out the front door.

Just this week, police in Marlborough, Massachusetts announced that they had recaptured 24-year-old Isaac Rivera, who escaped custody while undergoing a medical procedure over the weekend at UMass Marlborough Hospital, NBC10 Boston reportedopens in a new tab or window. At the time of his escape, Rivera was in custody for charges including domestic assault and battery, strangulation, assault and battery on a police officer, and illegal possession of a firearm.

Last Friday, authorities said that a man convicted of child sex crimes was back in custody after walking away from Mercy Hospital South in St. Louis, Missouri, where he had been taken for medical treatment, AP reportedopens in a new tab or window. Tommy Wayne Boyd, 45, is currently serving a 30-year sentence that was handed down in 2007, according to online court records.

Earlier this month, police said 44-year-old Yenchun Chen, who tied bed sheets together to rappel down the side of a New York City hospital, was back in custody, as reported by CBS Newsopens in a new tab or window. Officials said Chen was arrested on July 31 for possession of a controlled substance, and had been taken to Mount Sinai Beth Israel in Gramercy Park to receive medical care prior to his escape.

Finally, new details shared by Oregon State Hospital provided insight into how 39-year-old Christopher Lee Pray, who had been taken to the hospital for medical care after a fight with another patient, seized on unattended keys to a transport van to enact his late August escape, Portland's KOIN 6 News reportedopens in a new tab or window. Pray, who was recaptured after being found in the van stuck in mud, was originally in custody for attempted aggravated murder.

Risk Factors

"Inmates get transferred to hospitals out of jails and prisons for treatment all the time," Jeffrey Keller, MD, who has long worked in correctional medicine and serves as president of the American College of Correctional Physicians, explained to MedPage Today.

Inmates may need to have surgery for a condition such as appendicitis or an injury from a fall, he said, or they may need specialized treatment for a chronic disease like heart disease, cancer, or lupus.

"It's been happening forever, and it's something that all jails and correctional facilities plan for, but just by its nature, since inmates are going to a less secure place, the opportunities to escape are increased," Keller noted.

Accordingly, inmates are never supposed to know exactly when their outside appointments are, he said. Inmates may be told they're scheduled to see a cardiologist, but not when.

"There will be a correctional officer at the door to the hospital room," he added, "but that of course is not as secure as being in a prison cell with multiple layers."

Keller pointed to an incident in 2007, in which an inmate transferred to a Salt Lake City orthopedic clinic for an MRI stole a correctional officer's gun, and shot and killed the guard before being recapturedopens in a new tab or window.

The more recent spate of cases has also included the July escape and recaptureopens in a new tab or window of 35-year-old Eric Abril, a suspect in a Northern California hostage-taking homicide and shootout with police, from a medical facility in Roseville, a suburb of Sacramento, as well as other escapees who are still at large.

Earlier this month, the Metropolitan Police Department in Washington, D.C. announced that a reward for information leading to the arrest of 30-year-old Christopher Haynes had increased to $30,000opens in a new tab or window. Haynes, who had been in custody related to an August shooting and murder, physically assaulted an officer and escaped after being transported to George Washington Hospital for treatment of a prior ankle injury, the police department said.

In addition, 21-year-old Naseem Roulack, who had been serving a 13-year sentence for a string of crimes, hasn't been seenopens in a new tab or window since walking out of Bon Secours St. Mary's Hospital in Henrico, Virginia last month.

Potential Remedies

To help reduce the number of transports of inmates for medical care, jails or prisons may turn to telemedicine, Keller said. However, nearly all jails and prisons are "underfunded and overcrowded," he noted. The challenges become where to put a telemedicine room, for instance, and how to pay for it.

In other instances, people who are in jail for a very limited time for a less serious reason and who need medical care may be bonded out or simply released, he said. In certain cases, those who are serving lengthy prison sentences and who are suffering from a terminal illness may be compassionately released, though this process is more difficult.

Furthermore, all inmates are assigned a security or risk level, Keller said, which is taken into account for every type of transport to a hospital or otherwise, such as to appearances in court. In addition to hand and feet shackles, high-risk inmates may be monitored by a pair of officers, one inside their hospital room and one outside.

"From a security standpoint, it would be better if the specialist came to the jail or prison, but very few are willing to do that," he explained. "Anything that can't be done inside the walls of a jail or prison has to go outside."

However, there are a few exceptions.

"If a prison system is large enough, they can have their own hospital ... Some jails that are big enough have jail wards within a hospital," Keller said. However, "most jails and prisons don't have anything like that."

https://www.medpagetoday.com/special-reports/features/106505

AI Not Ready to Replace Radiologists Interpreting Chest X-Rays

 Commercially available artificial intelligence (AI) tools were accurate to varying degrees in flagging chest x-ray abnormalities but turned up more false-positives than radiology reports, a Danish study found.

Testing four CE-marked AI tools on real-world radiographs from the Copenhagen region, investigators reported areas under the receiver operating characteristic curves ranging from 0.83-0.88 for airspace disease, 0.89-0.97 for pneumothorax, and 0.94-0.97 for pleural effusion using radiology reports as reference

Louis Plesner, MD, of the University of Copenhagen, Denmark, and coauthors found a wide range of sensitivity and specificity values for the AI tools:

  • Annalise Enterprise CXR (version 2.2): sensitivity 72% and specificity 86% for airspace disease; 90% and 98% for pneumothorax; 95% and 83% for pleural effusion
  • SmartUrgences (version 1.24 with high sensitivity threshold): sensitivity 91% and specificity 62% for airspace disease; 73% and 99% for pneumothorax; 78% and 92% for pleural effusion
  • ChestEye (version 2.6): sensitivity 80% and specificity 76% for airspace disease; 78% and 98% for pneumothorax; 68% and 97% for pleural effusion
  • AI-Rad Companion (version 10): sensitivity 79% and specificity 72% for airspace disease; 71% and 98% for pneumothorax; 80% and 92% for pleural effusion

"Among the AI tools examined in this study, we observed an acknowledgeable difference in the balance between sensitivity and specificity for the individual tools, which seems unpredictable. Therefore, when implementing an AI tool, it seems crucial to understand the disease prevalence and severity of the site and that changing the AI tool threshold after implementation may be needed for the system to have the desired diagnostic ability," the group wrote in Radiology

"Furthermore, the low sensitivity observed for several AI tools in our study suggests that, like clinical radiologists, the performance of AI tools decreases for more subtle findings on chest radiographs," the study authors noted.

According to the American College of Radiology, nearly 400 FDA-approved AI tools are applicable for useopens in a new tab or window in radiology.

"While AI tools are increasingly being approved for use in radiological departments, there is an unmet need to further test them in real-life clinical scenarios. AI tools can assist radiologists in interpreting chest x-rays, but their real-life diagnostic accuracy remains unclear," said Plesner in a press release.

In their study, the authors reported that in chest radiographs with four or more findings, AI's specificity dipped to the 27-69% range for airspace disease, 96-99% for pneumothorax, and 65-92% for pleural effusion.

Ultimately, Plesner stated that while the tools are useful, potentially providing a confidence boost for radiologists, they should not be autonomous in regards to making a diagnosis for patients.

Masahiro Yanagawa, MD, PhD, and Noriyuki Tomiyama, MD, PhD, both of the Osaka University Graduate School of Medicine in Japan, agreed, emphasizing the limits of AI in this setting.

"Given that anteroposterior chest radiographs and chest radiographs with multiple findings reduced the specificity of AI tools, radiologists should be aware of the limitations of the tools with respect to both sensitivity and specificity. Care must be taken not to overestimate the results of AI tools in such challenging cases," Yanagawa and Tomiyama wrote in an invited commentaryopens in a new tab or window.

For their retrospective study, Plesner and colleagues invited AI vendors to test their algorithms on real-world chest x-rays from four hospitals within the Copenhagen region. The radiographs had come from 2,040 consecutive adult patients (50.6% women; average age 72 years).

Four out of seven invited AI vendors agreed to participate and have their AI tools compared to radiologist-made clinical radiology reports for reference.

All four AI tools produced significantly more false positives. For example, in identifying airspace disease, false positives ranged from 13.7% with the Annalise algorithm to 36.9% with the SmartUrgences. For comparison, radiologists had a false positive rate of 11.6%.

Only the SmartUrgences algorithm, when tuned to high specificity, did not produce more false positives than radiologists in flagging pneumothorax and pleural effusion.

False negative rates varied widely depending on the finding and the AI tool.

One limitation was that radiologists had access to clinical information, lateral chest radiographs, and prior imaging that the AI tools did not, potentially giving them an "unfair advantage," the authors said. Other possible limitations of the study included the lack of AI evaluation of lateral chest radiographs and that the findings may not be applicable to non-hospital settings.

Disclosures

This study was supported by research grants from the Danish government.

Plesner reported a relationship with Siemens Healthineers. Coauthors disclosed relationships from Siemens Healthineers, Innovation Fund Denmark, Roche, Orion, Pharmacosmos, Novartis, Bavarian Nordic, Merck, Philips Healthcare, and Boehringer Ingelheim, and one coauthor is employed by Novo Nordisk.

Yanagawa disclosed grant support from the Japan Society for the Promotion of Science and the Japan Agency for Medical Research and Development. He is also associate editor for Radiology: Artificial Intelligence. Tomiyama had no disclosures.

Primary Source

Radiology

Source Reference: opens in a new tab or windowPlesner LL, et al "Commercially available chest radiograph AI tools for detecting airspace disease, pneumothorax, and pleural effusion" Radiology 2023; DOI: 10.1148/radiol.231236.

Secondary Source

Radiology

Source Reference: opens in a new tab or windowYanagawa M, Tomiyama N "Clinical performance of current-generation AI tools for chest radiographs" Radiology 2023; DOI: 10.1148/radiol.232139.


https://www.medpagetoday.com/radiology/diagnosticradiology/106508

opens in a new tab or window.

Wrath Of Khan Wrecks Tech As Downbeat Dimon Batters Bonds, Bullion, & Banks

 Jamie Dimon dares to question goldilocks/soft-landing narratives overnight, warning that The Fed may be forced by stickier inflation to hike rates further (to 7%) fearing a stagflationary scenario that no one is prepared for... and (away from the main headlines), Dimon warned of more bank failures as rate rise (and that didn't help) as the KBW Bank Index fell below the initial SVB collapse levels...

Source: Bloomberg


Then the macro data started with Dallas Fed Services and Manufacturing ugly (prices up, activity down), new home sales finally hitting the wall (as homebuilders folded on filling the affordability gap and their stock prices are starting to realize that)...

Source: Bloomberg

...then consumer confidence crumbled (driven by a slump in hope and worsening labor market conditions). Overall, 'hard' data is near 5-month lows, as 'soft' survey data hits its highest since Jan 2022, sending 'hope' about as high it goes (before the soft surveys collapse)...

Source: Bloomberg

Add to that the wrath of Lina Khan as the FTC sued Amazon (again) and that was enough - with corporate buybacks blacked out still - to take the equity market lower, dollar higher, gold lower, and yields higher (rising after early declines).

Khan to Amazon/Bezos: "I've done far worse than kill you. I've hurt you. And I wish to go on hurting you... I mean to avenge myself upon you, Amazon."

AMZN at 3-mo lows...

Early on we saw the Megacap Tech names getting hit hard (while unprofitable tech was not), which Goldman suggested was driven by derisking overall. But as the selling continued, everything was dragged lower...

Source: Bloomberg

The drag of the Magnificent 7 weighed everything down with Nasdaq and S&P the big underperformers on the day, but everything was ugly...

0-DTE traders countered the initial thrust lower after the cash open (as the S&P broke below the Put-Wall at 4300), but as Khan unleashed her FTC ear-bugs,

Source: SpotGamma

The last month has seen both Value and Growth stocks hammered as the broad market weakens, but most notably, the Russell 2000 Value index is now in the red YTD (down around 3%)...

Source: Bloomberg

VIX soared up near 20 today, its highest since May and VVIX surged up into the danger-zone above 100...

Source: Bloomberg

The long-end of the yield curve was sold again today with 30Y yields up another 4bps to fresh cycle highs (and the 2Y auctioned at 16 year highs)...

Source: Bloomberg

And that sent the yield curve (2s30s) steeper still (to its least inverted since May)...

Source: Bloomberg

Stocks and Bond (Prices) are moving more and more in sync (with correlation between yields and stock prices nearing their lower limit)...

Source: Bloomberg

The dollar rallied for the 5th day in a row to its strongest since Dec 2022...

Source: Bloomberg

And as the dollar rallied, so gold was clubbed like a baby seal with spot prices testing back below $1900..

Oil ended the day higher, back above $90 (despite the strong dollar) bouncing back hard from an overnight plunge...

Finally, the Magnificent 7 stocks have lost over $1 trillion in market cap from their July highs falling back to near 4-month lows.

Source: Bloomberg

From Greed to Fear... fast!

That escalated quickly... as suddenly the markets' blinkers on consumer strength (remember Bidenomics) are ripped off.

https://www.zerohedge.com/markets/wrath-khan-wrecks-tech-downbeat-dimon-batters-bonds-bullion-banks